1V4ESI.J RETURNING MATERIALS: PIace in book drop to uaamuas remove this checkout from .‘IIIKSIIIL. your record. FINES wil] be charged if book is returned after the date stamped beIow. HYPERTENSIVE PATIENTS' PERCEPTIONS OF SOCIAL SUPPORT IN FOLLOWING A THERAPEUTIC REGIMEN: A DESCRIPTIVE STUDY By Sandra Ann AItenritter A THESIS Submitted to Michigan State University in partiaI fquiIIment of the requirements for the degree of MASTER OF SCIENCE IN NURSING CoIIege of Nursing 1985 To my husband. Edward, our chiIdren. David and Shannon, and my parents. BeverIy and WiIIiam WorIey. ACKNOWLEDGMENTS I wish to thank Barbara and 8111 Given for the use of data from their research project "Patient Contributions to Care--Link to Process and Outcome)’ SpeciaI thanks to Barbara Given. my academic advisor and thesis committee chairman. Her guidance and support have been greatIy appreciated throughout my graduate studies. Thanks to my thesis committee members. Patty Peek. Rita GaIIin. and 8111 Given. for their advice and support throughout the thesis process. I aIso wish to thank Bryan Coer for his assistance with data anaIysis for this study. My warmest thanks are reserved for my family. who have demon- strated the true meaning and vaIue of sociaI support. To my parents. NiIIiam and BeverIy WorIey. thank you for your Iove and assistance as I strived to reach my educationaI goaIs. To my chiIdren. Shannon and David. I appreciate your understanding and heIp when I had to be away from the famin. To my husband. Edward--whose patience. encouragement. and love have earned him the titIe "A-#I Support Person"-—Love and Thanks! TABLE OF CONTENTS LIST OF TMLES O O O O O O O O O O 0 LIST OF FIGURES Chapter 1. INTRODUCTION TO THE STUDY . Introduction . . . . . . . Purpose . . . . . . . . . Statement of the Question Definition of Concepts . Limitations Assumptions Overview of the Thesis . 2. CONCEPTUAL FRAMEWORK . . . . overv16w O O I O O O O O 0 Definition of Hypertension NormaT ReguIation of BIood Pressure EssentiaT Hypertension . . . . . . . The Therapeutic Regimen--Recommendations Management of Hypertension .. . .. . .. . SociaT Support . . . . . . . . . . . . . . . . . Genera] Systems Theory and the Nursing Theory of Imogene King . . . . . . . . . . . . . . . . . ImpIications for Nursing . . . . . . . . . . . . 3. LITERATURE REVIEW for t (D o o o o h IntrOduction O O 0 O O O O O O O O I O O O O O O FoIIowing a Therapeutic Regimen for Hypertension Overview--Socia1 Support . Definitions of SociaI Support Types of Support . . . . Measurement of SociaI Support Page vii d O‘DNOxlflO‘a SociaI Support and FoTIowing a Therapeutic Regimen for Hypertension . . . . . . . . . . . summa ry O O O O O O O O O O O 0 O O O O O 4. METHODOLOGY AND PROCEDURE . . . . . . . . . Overview . . . . . . . . . CoIIection Sites . . . . . PopuTation .. . .. . .. SampIe . . . . . . . . . . Human Rights Protection . Interview Procedures--Intake . ExperimentaI and ControT Groups . . . . . . Interview Procedures Post-Intervention and FoITow-Up . . . . . . . . . . . . . . . Interviewers . . . . . . . . . . . . . . . OperationaIization of the Study VariabIes Coding and Scoring of the Questionnaire . ReTiabiIity and VaIidity of the Instrument Research Questions . . . . . . . . . . . . StatisticaT AnaIysis of the Data . . . . . Summary . . . . . . . . . . . . . . . . . 5. DATA PRESENTATION AND ANALYSIS . . . . . . . Introduction . . . . . . . . . . . . Description of the Study SampTe . Sociodemographic Characteristics Sex . . . . . . . . . . . . . Age . . . . . . . Race/Ethnicity . . MaritaT Status . . Number of ChiIdren Income . . . . . . Work Status . . . Occupation . . . . Education . . . . Living Arrangements . Number of ChiIdren Living at Size of Househon . . . . . . . . Duration of Hypertension . . . . . . Summary of Sociodemographic Findings Data Presentation for Research Questions Research Question I . . . . . . . . . Research Question 2 . . . . . . . . . Research Question 3 . . . . . . . . . is: Page 111 112 112 112 113 113 114 114 116 116 117 117 117 118 118 122 124 6. AdditionaI Findings SociaI Support by a Nurse Summary SUMMARY. INTERPRETATIONS. AND RECOMMENDATIONS overv16w I O O O O O O O O O O O O O O O O O O Sociodemographic Characteristics Sex . . . . . Age . . . . . Race . . . . . MaritaI Status Income . . . . Education . . . . . . Duration of Hypertension Work Status. Occupation. Research Questions . . . . Support Person . . . . Type of Support . . . Quantity of Support . Limitations . ImpIications . . . . . . . 3 oQoooooo omoooooo ImpTications for Nursing Practice 000.020.0000 of the SampIe o o o o o a o o o o o o a U' o o (D o o o o ImpIications for Nursing Education ImpIications for Research . Summary APPENDICES O O O O O O O O O O O O 0 O O A. LETTER FROM PROVIDER . . . . . . CONSENT FORM 0 O O O O O O O O O INTAKE SOCIODEMOGRAPHIC INSTRUMENT SOCIAL INTERACTION QUESTIONNAIRE . . NINE-MONTH SOCIODEMOGRAPHIC INSTRUMENT REFERENCES 0 O O O O O O O O O O O O O I O O 0 vi .5 O O O O C O O O O O 0 II“. 0 O O O O o o o o o o o o o o O. o o o o o 3' O O O O O O O O O u-Io. O O O O O u—l o o o o o o o o o 0 Q. o o o 0 fl 0 o o (D o o o o :3 Page 141 141 142 143 143 143 144 144 144 145 145 146 146 146 148 148 153 157 162 163 163 171 173 182 183 184 186 188 192 204 206 Tab1e 2.1 2.2 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.11 5.12 5.13 5.14 5.15 LIST OF TABLES C1assification of B/P . . . . . . . . . . . . . . . . . Stepped-Care Approach Number and Percentage Number and Percentage Number and Percentage Income Number and Percentage Number and Percentage Number and Percentage Number and Percentage Arrangements . . . Number of Subjects by to Drug Therapy . . . . . . . . . of Subjects of Subjects of Subjects of Subjects of Subjects of Subjects of Subjects Duration of by Age 0 0 0 O O O O by Race by Tota1 Famin by Work Status . . . by Occupation . . . . by Education . . . . by Living Hypertension Number and Percentage of Supportive Person by ModaIity Number of Subjects and Means of Quantity of Support by Mada] 1ty O O O O I O I O O O O O O O I O O O O O 0 Number and Percentage of Subjects by Type of Support . Number and Percentage of Patients Who Identified Support Person by Condition . . . . . . . . . . . . . Number and Percentage of Subjects by Condition and Type of Support . . . . . . . . . . . . . . . . . . . Number and Percentage of Subjects Who Identified Support by Sex Number and Percentage of Subjects by Sex and Type of Support vii Page 14 19 111 112 113 114 115 115 116 118 119 121 123 125 126 127 129 Number and Percentage of Subjects. Marita1 Status by Type Of support 0 O O O O O O I O I O O O O O O 0 Number and Percentage of Subjects by Income and Type of Support . . . . . . . . . . . . . . . . . . . Number and Percentage of Subjects by Occupation and AvaiTabiIity of Support . . . . . . . . . . . . . . . Number and Percentage of Subjects by Education and Type of Support . . . . . . . . . . . . . . . . . . . Number and Percentage of Subjects by Living Arrangement and Type of Support . . . . . . . . . . . . . . . . . viii Page 131 133 135 137 139 Figure 2.1 2.2 2.3 6.] LIST OF FIGURES Open Living System . . . . . . . . . . . . . . . . Open Living System Reiated to Hypertension . . . . Socia1 Support Interaction . . . . . . . . . . . . A Proposed Mode1 for Socia1 Support and Adaptation to Stress Page 26 28 29 176 ABSTRACT HYPERTENSIVE PATIENTS' PERCEPTIONS OF SOCIAL SUPPORT IN FOLLOWING A THERAPEUTIC REGIMEN: A DESCRIPTIVE STUDY By Sandra Ann A1tenritter A descriptive study was conducted to identify hypertensive patients' perceptions Of socia1 support in fo1IOwing a therapeutic regimen. Differences in perceptions of socia1 support (support person. type Of support. and quantity of support) among the moda1ities and subgroups of the samp1e were aISO examined. Data were c011ected from 102 hypertensive patients by means Of a se1f-administered questionnaire. Data were ana1yzed using descriptive statistics. chi-square. and tetests. Hypertensive patients were most often abTe to identify someone ab1e to support them for the moda1ity Of diet. Spouse was the most frequentTy identified support person for a11 moda1ities. Patients perceived high 1eve1$ of psychoiogicaI support for medication and sTight1y more tangibie than psycho1ogica1 support for diet and exercise. There were significant differences in perceptions among the moda1ities. experimentaI and contr01 groups. and sociodemographic characteristics of sex. income. occupation. and Iiving arrangements. CHAPTER 1 INTRODUCTION TO THE STUDY Intmdusiien Hypertension is a chronic. debi1itating. and Often fataI condition so widespread in the American popu1ation that many refer to it as epidemic (NationaT Institute of Heaith. 1981). ApproximateTy 60 mi]- 1ion Americans have an increased risk of heart attack. coronary artery disease. stroke. and kidney disease as a resu1t Of high bTOOd pressure (Kanne1. 1976; Nationai High BIOOd Pressure Information Center. 1978; Hypertension Detection and FO11ow-Up Program Group. 1979. 1982; Joint NationaI Committee on Detection. Eva1uation. 8. Treatment of High B/P. 1984). Estimates of the cost Of hypertension and its comp1ications range from $16 to $30 mii1ion per year. 'This cost inc1udes direct medicai expenditures and 1ost income through i11ness. disabiTity. pre- mature 1055 of productivity. and death (Kochar. 1981) but not the cost in changes in the qua1ity of Tife for those individua1s and their famiIies that are impossibTe to measure. Morta1ity from cardiovascu1ar disease has dec1ined since 1950. especia11y after 1970. with deaths from hypertension—re1ated diseases deCIining at a much sharper rate than those categories not re1ated to hypertension. From 1968 to 1978. the age-adjusted death rate for hypertensive-re1ated diseases fe11 28.5%. Sti11. in 1982 there were 729.000 deaths from hypertension and re1ated diseases (U[§--A Statisti- cal_EOntLa1t_O£;the_Amsnican_Esnnle. 1983). 'This dec1ine in mortaIity and morbidity is be1ieved to be due in part to the treatment of hyper- tension (NHLBI Hypertension Detection and Fo11ow-Up Program. 1972-79; V.A. Study Group on Antihypertensive Agents. 1963-1970) and the estab- 1ishment of the Nationa1 High B1ood Pressure Education Program with its concurrent growth Of community programs in 1972 (Hypertension Detection & F011ow-Up Program Group. 1982). Yet. despite the avai1abi1ity of effective treatment. the majority Of hypertensive Americans do not have effective1y contr011ed bIOOd pressure. These uncontroTIed hypertensive individua1s inc1ude persons who are not identified as hypertensives. persons who are identified and not treated. and persons who are identified but receive ineffective treatment (Kochar. 1981; Mason. 1982L A review of surveys that meas- ured pub1ic know1edge and attitudes toward hypertension from 1973 to 1979 (Haines & Ward. 1981) reveaied that a1though the pub1ic had an increasing awareness of the seriousness and risks of hypertension. the etioTogy of hypertension. the avai1abiIity and Iength of treatment. and the roIe of diet in treatment. there was no substantia1 change in the percentage of hypertensives who stayed in treatment. In 1975 the Nationa1 High BIood Pressure Education Program in cooperation with the American Nurses Association and the Nationa1 League for Nursing sponsored a Task Force on the ROIe of the Nurse in High B1ood Pressure Contro1. 'The goa1 of the task force was to review the nurse's roTe in the treatment of high b100d pressure and to deIineate the education necessary to fqui11 that rOTe (Grim & Grim. 1981). To accompIish these goa1s. the task force made the fo110wing recommendations: 1. high b1ood pressure be given greater emphasis in nursing curricu1a and continuing education programs; 2. more nurses become prepared to provide primary care for patients with uncomp1icated hypertension; and 3. nurses participate in and conduct research re1ated to the care of hypertensive patients.(Guide1ines for Educating Nurses. 1980) These objectives are being met. Co11eges and universities have made information regarding the etio1ogy. detection. and management of hypertension avai1ab1e to a11 nurses in either their basic or continu- ing educationa1 programs. Graduate programs are preparing nurses to provide primary care to hypertensive patients and to conduct the research necessary to bui1d nursinghs know1edge base regarding hyper- tension and the roIe nurses p1ay in its management. Reports from nurses invo1ved in the ciinicaI management of and research on hyperten- sive patients support the beTief that the most cha11enging aspect of practice with hypertensive patients continues to be assisting the patient in 1ong-term adherence to a therapeutic regimen (Brown & BIoom. 1978; Danie1s & Kochar. 1979; Earp et a1.. 1982; Giinn. 1978; Given et 3L" 1979. 1982; Grancio et a1" 1980; Grim & Grim. 1981; Havi1and. 1983; Heine. 1981; Mason. 1982; McCombs et a1.. 1980; Rogus. 1981; Stecke1 et a1" 1977; Swain & SteckeI. 1981L The treatment or therapeutic regimen for hypertension consists of severa1 moda1ities that may invo1ve 1ife-sty1e changes. ‘The use of medication. dietary regu1ation for weight 1055 or sodium restriction. inc1usion of exercise. exc1usion of smoking. and more effective manage- ment of stress may be inc1uded in a hypertensive patientfls therapeutic regimen. The difficuities of reIinquishing 01d habits and modifying 1ife sters to fo110w a therapeutic regimen are we11 documented (Cap1an et a1.. 1976; Sackett & Haynes. 1976; Swain 8. Stecke1. 1981). Zisook (1980). in a review of the 1iterature. estimated that as many as 93% of patients do not f011ow recommended therapeutic regimens. The hyperten- sive patient may be more 1ike1y not to fo11ow the recommended regimen because he/she does not fee1 111 or experience symptoms before starting a therapeutic regimen and therefore may fee1 no "improvement" whi1e on the regimen. Those interested in hea1th care have studied many strategies that may infiuence a patientfls abiiity to foI1ow a therapeutic regimen. among which an important one has been socia1 support. 'The ear1iest and most commom focus Of research in the fie1d of socia1 support was on the re1ationship of an individuai's support system to his/her abiiity to cope with stressfu1 circumstances (Atch1y. 1979; Bi11ings & Moos. 1980; Cobb. 1976; Dean & Lin. 1977; Gop1erod. 1978; Gore. 1978; Kap1an et a1.. 1977; Murawski et a1.. 1978; NuckoTTs et a1.. 1972; PearTin et a1.. 1981). More recentTy. research has documented the va1ue of socia1 support not onIy as an enhancer of se1f-esteem. but a1so as a faci1i- tator in fo11owing a therapeutic regimen. The major prob1em of this type of research has been the inconsistency in the definition and measurement of socia1 support. As summarized by Kap1an et a1. (1977). There is TittTe strong empirica1 evidence to confirm the ro1e it may p1ay in hea1th and iI1ness. This is not surprising: attempts at conceptuaIization and measurement have been inadequate. disci- p1ine bound (or study bound) and usua11y formuIated for post-hoc interpretation of unexpected. but striking resu1ts. St111. more studies have reported positive re1ationships between socia1 support and fOI1owing a therapeutic regimen (Berkman & Syme. 1979; Davis & Eichorn. 1963; DeAraujo et a1.. 1972; Earp. 1982; Haynes. 1976. 1979; King. 1982; Kirscht et a1.. 1981; McKenny et a1.. 1973) than have reported no corre1ation (Hershay et a1.. 1980; Ne1son et a1.. 1978). In empirica1 research. socia1 support is measured in two ways: from the perspective of an outside observer (the objective approach) or from the frame Of reference of the target person (the subjective approach). A1though supporters of the objective approach note that Objective assessment is not prone to the se1f-reporting biases of the subjective approach. the subjective approach remains va1uab1e. Dona1d et a1. (1978) noted. In favor of the more subjective approach is the argument that individua1s have different needs and tastes; therefore. the nature and number of interpersona1 contacts with friends. reiatives. and others necessary to achieve socia1 hea1th may vary great1y. The differences may not be adequate1y reercted in measures of objec- tive socia1 hea1th constructs. Using the subjective approach to measure socia1 support may aISO he1p determine the person's perception of support avai1ab1e and received. For it is the perception of support rather than the objec- tive measures of support that determines how peop1e w111 respond in a situation. If peop1e perceive they are supported. regardTess of actua1 feeTings or intentions toward them. they wi11 respond. ids. cope. effective1y (Given et a1.. 1979). A major goa1 of hea1th-care providers dea1ing with hypertensive patients is to assist them in f011owing a therapeutic regimen. As noted earIier. research has documented positive re1ationships between socia1 support and fo110wing a therapeutic regimen. It is therefore important for hea1th-care providers to be ab1e to determine what socia1 support is avai1ab1e to and received by the patient. Since the percep- tion of socia1 support determines the response in a situation. it is necessary to examine patients' perceptions Of socia1 support (subjec- tive approach) rather than tota11y re1y on the provider's assessment of support avai1ab1e and received (objective approach). Thus. the purpose of this study is to investigate hypertensive patients! perceptions of socia1 support in foIIowing a therapeutic regimen. Bums The purpose of this study is to examine and describe hypertensive patients' perceptions of socia1 support in fo110wing a therapeutic regimen. The data were c011ected for the research project "Patient Contributions to Care--Link to Process and Outcome." by B. Given and C. W. Given. co-principa1 investigators. The project was a contrOIIed fie1d experiment in which the effects of a nursing intervention. over a 6-month period. on c1inica1 parameters and other indicators of manage- ment and controI of hypertension were expTOred. In this thesis I wi11 attempt to identify hypertensive patients' perceptions about who was supportive. what type of support they received. and how much support they received. An attempt wiTI be made to identify any differences in perceptions of socia1 support among the various modaTities Of the therapeutic regimen (medication. diet. and exercise) and among standard sociodemographic characteristics of the hypertensive patients studied. With this information the hea1th-care provider may be in a much better position to assist the patient in the use of socia1 support to make Iife-sty1e changes necessary to fo11ow his/her therapeutic regimen. Sliifim9n1_QI_Ih§_QUQEIIQn The study is descriptive in nature. and ana1ysis of the findings wi11 be used to answer the fo110wing questions: 1. What are hypertensive patients! perceptions of socia1 support in fOT1owing a therapeutic regimen? 2. Are there differences in perceptions of socia1 support among the various moda1ities of the therapeutic regimen? 3. Are there differences in perceptions of socia1 support among the various subgroups of the samp1e studied. based on standard socio- demographic characteristics? W The f011owing are definitions Of concepts introduced in the research questions and used throughout the study. Hypertensixe_patients are defined in this study as femaIes and ma1es. ages 21 to 65 inc1usive. with an estabiished diagnosis of essen- tia1 hypertension whose therapeutic regimen inc1udes one or more of the foTIowing: medication. therapeutic diet (weight reduction. sodium restriction. or both). or prescribed exercise. For inc1usion in this study. patients must (a) have no other chronic i11ness; (b) have no evidence of stroke. cancer. b1indness. end-stage rena1 disease. psycho- sis. or active pregnancy or Tactation; (c) be 1iterate; (d) speak and read Engiish; and (e) have two b1OOd-pressure readings separated over time indicating a systoIic pressure above 160 mm Hg and a diasto‘lic pressure Of 90 mm Hg or above. WW refers to the expressed beIief of the individua1 as to the existence or nonexistence of socia1 support. §gg151_§uppgni is defined as "information 1eading the subject to be1ieve that he is cared for. Toved. esteemed. and va1ued. and a member of a network of mutua1 ob1igationsfl (Cobb. 1976). This support may be task oriented or tangib1e "behavior directed toward providing the person with tangib1e resources that are hypothesized to benefit his/her menta1 or physica1 we11 being" (Cap1an. 1979). It may a1so be psycho- 1ogica1 support or "behaviors directed toward providing the person with cognitions (va1ues. attitudes. be1iefs. and perceptions) and toward inducing affective states that are hypothesized to promote we11 being" (Cap1an.1979). Ihegapeu119_negjnuuh This inc1udes strategies documented or beIieved to reduce hypertension which are prescribed or recommended by a hea1th-care provider to a hypertensive patient. These may inc1ude both pharmaco1ogic and nonpharmaco1ogic modaIities. 1J%. medication. dietary restrictions for weight Toss or sodium reduction. exercise. stress reduction. reduction or cessation of a1cohOI consumption and smoking. and restoration of normaI sIeep patterns. For the purpose of this study. the moda1ities of the therapeutic regimen wi11 be 1imited to medication. diet. and exercise. Limitations The foTTowing 1imitations were identified in this study: 1. The members of the samp1e were vo1unteers and may be different from those in the popu1ation who refused to participate. Therefore. the samp1e may not be representative of a11 hypertensive patients. and the resu1ts are not generaIizab1e to a11 hypertensive patients. 2. Portions Of the instrument used to coITect the data studied were not tested before use and have 1imitations Of vaIidity and re1ia- bi1ity. 3. The hypertensive patients! perceptions Of socia1 support may change over time or with circumstances. ‘Therefore. the findings of this study may not ref1ect the hypertensive patients! perceptions at any other point in time. 4. This study re1ied on se1f-reported data and subjects may have responded in a socia11y desirab1e manner. thus posing a threat to the va1idity of the resu1ts. Mullen: For the purpose of this research. the foIIowing assumptions were made: 1. Fo1IOwing a therapeutic regimen for hypertension wi11 invoTve modification of habits and possiny Tife sty1e. 10 2. 'The hypertensive patient has perceptions regarding socia1 support in f011owing a therapeutic regimen. 3. The hypertensive patient is ab1e to answer questions consistent with his/her perceptions. 4. The hypertensive patient is ab1e to read and comprehend the questions in the instrument. 5. The instrument is sensitive enough to document perceptions of socia1 support in fo110wing a therapeutic regimen. W The thesis is organized into six chapters. Inc1uded in Chapter 1 are the introduction. statement of the prob1em. research questions. conceptua1 definitions. Timitations. and assumptions Of the study. In Chapter 2. the concepts and theories reIevant to this study are outIined. integrated into a conceptua1 framework. and presented in graphic form. A review of the Titerature re1ated to major concepts of the study is presented in Chapter 3. A discussion of the methodo1ogy and procedures used in conducting the research is presented in Chapter 4. A description of the c011ec- tion sites. popu1ation. samp1e. human rights protection. data- c011ection procedures. instruments. and scoring procedures is inc1uded in this chapter. The data c011ected are presented and ana1yzed in Chapter 5. 11 Research findings are summarized and interpreted in Chapter 6. Conc1usions and recommendations for nursing practice. education.iand research are a1so presented. CHAPTER 2 CONCEPTUAL FRAMEWORK Qxerxjew The purpose of this chapter is first to discuss brief1y the definition of hypertension and the regu1ation of b100d pressure in both norma1 tension and hypertension. Next. the therapeutic regimen (recom- mendations or protocoTs) for treatment of hypertension wi11 be pre- sented. The possib1e ro1e of socia1 support in faci1itating Tong-term management of hypertension then wiII be considered. Fina11y. a concep- tua1 framework integrating systems theory. the nursing theory of Imogene King. and the major concept of the study. perceptions of socia1 support. w111 be described. A schematic representation of the concep- tua1 framework provides a base from which the research questions may be studied. WM There continues to be some disagreement over what b1ood pressure reading denotes hypertension and deserves treatment. 'The Wor1d Hea1th Organization has defined hypertension as the presence of casua1 b100d pressure greater than 160/95 mm Hg. As a resu1t of new research data on risks of inghtIy e1evated biood pressure. the norms were refined by 12 13 the Nationa1 Center for Hea1th Statistics and the fo11owing guide1ines estainshed (Mason. 1982). Definite High--B1ood Pressure Not Adequate1y Treated 160 mm Hg or over Systo1ic pressure 95 mm Hg or over DiastoIic pressure BorderTine High--BIOOd Pressure Not Under Medication Pressure beIOw 160 mm Hg systOIic and Be1ow 95 mm Hg diasto1ic. but not simuItaneousTy be1ow both 140 mm Hg systo1ic and 90 mm Hg diasto1ic Hypertensives on Adequate Treatment Previousiy diagnosed hypertensives with bIOod pressure beIOw 160 mm Hg systo1ic and 95 mm Hg diasto1ic and on a therapeutic regimen for high b100d pressure Norma] B100d Pressure Pressure beIOw both 140 mm Hg systo1ic and 90 mm Hg diastoIic Notwithstanding these guide1ines. the most wider fo1IOwed defini- tion appears in the recommendation Of the Joint Nationa1 Committee on Detection. Eva1uation. and Treatment of High B1OOd Pressure (1984). which stated: F011owing screening. the diagnosis of hypertension in adu1ts is confirmed when the average of mu1tip1e diastOIic B/Ps on at 1east two subsequent visits is 90 mnan or higher; or when the average of mu1tip1e systOIic B/Ps on two or more subsequent visits is consis- tent1y greater than 140 mm Hg. Hypertension is then c1assified by the Committee as shown in Tab1e 2.1. W Under norma1 circumstances. the b100d pressure 1eve1 is maintained by the interpiay of various physioTogica1 mechanisms. B1ood pressure is primarin a function of cardiac output and periphera1 resistance. This re1ationship is summarized by the formu1a: BTOOd pressure = cardiac output x periphera1 resistance Cardiac output is the vqume of b1ood ejected into the aorta per minute. It is the major determinant of the systoIic b100d pressure. Diasto1ic bIOOd pressure is primariTy determined by the resistance in the arterio1es. Cardiac output and periphera1 resistance are directIy and indirect1y affected by factors such as b1ood vo1ume. biood viscos- ity. sympathetic nervous activity. renin angiotensin a1dosterone sys- tem. and autacoids (vasoactive substances) such as prostag1andins and bradykinin (Kochar. 1981). TabTe 2.1: C1assification of B/P Range. mm Hg Category DiastOTic < 85 Norma1 B/P 85-59 High norma1 B/P 90-104 Mi1d hypertension 105-114 Moderate hypertension 115 Severe hypertension Systo1ic. when diastoIic is 90 < 140 140-159 160 Norma1 B/P BorderIine iso1ated systoTic hypertension Iso1ated systo1ic hypertension Source: Joint Nationa1 Committee on Detection. Eva1uation and Treat- ment of High B100d Pressure (1984). 15 W More than 95% of patients with e1evated b1ood pressure have essentiaT (idiopathic. primary) hypertension. These patients do not have an identifiab1e cause for their hypertension. but they have a disease of b1ood pressure regu1ation. In ear1ier stages of mi1d hyper- tension. the cardiac output is e1evated. As the pressure rises fur- ther. the cardiac output fa11s and the e1evated b1ood pressure ref1ects increased periphera1 resistance. Vasoconstriction is maintained by excess sodium content of the arterio1ar smooth musc1e ce115. increased sympathetic activity. imba1ance between vasoconstrictor angiotensin and vasodi1ator prostagTandins and kinins. and other unknown mechanisms. Other factors associated with hypertension inc1ude Obesity. increased heart rate. heredity. physica1 activity. and increasing age (Kochar. 1981). 'The purpose Of the therapeutic regimen. then. is tO reduce or e1iminate physioTogica1 and psychoTogica1 phenomena which a1ter norma1 b100d pressure regu1ation. Mmpeutmaedimenflmmendammne WWW Findings of the Hypertension Detection and F011ow-Up Program (1982) suggest that 1ong-term reduction of b1ood pressure decreases overa11 morta1ity at a11 Teve1s of hypertension. In patients with miId hypertension it a1so reduces the risk of cardiovascu1ar comp1ications such as stroke. congestive heart faiIure. 1eft ventricu1ar hypertrophy. and progressive e1evations in b100d pressure. In those patients with moderate or severe hypertension. reducing b1ood pressure decreases 16 overa11 mortaTity and cardiovascu1ar morbidity OLA. Administration Cooperative Study Group. 1969. 1970). It is therefore the recommendation of the Joint Committee of Detection. Eva1uation. and Treatment Of High BTood Pressure (1984) that even mi1d hypertension be treated. Benefits outweigh risks of pharma- coTogic therapy for those with a diastOTic B/P persistentTy e1evated above 95 mm Hg and for those who are at high risk (e.g.. patients with target organ damage. diabetes meTTitus. or other major risk factors for coronary heart disease). Aggressive nonpharmacoTogic therapy is sug- gested for those with diastOTic B/Ps Of 90 to 94 mm Hg who are otherwise at Tow risk. WhiTe nonpharmacoTogic therapy has been used as an adjunct to drug therapy for some time. evidence has grown in the past five years in favor of using these approaches as definitive interventions for patients with mi1d uncompTicated hypertension. As described in Chapter 1. nonpharmacoTogic approaches may inc1ude dietary restrictions (weight Toss and sodium reduction). reduction or cessation of aTcohoT consump- tion and smoking. exercise. stress reduction. and return to norma1 sTeep patterns. In the Report of the Hypertension Task Force (1979). a strong corre1ation between body weight and B/P was noted. This was particuTarTy true among chderen and young to middTe—age adu1ts. Reisin et a1. (1978) found that weight reduction often resu1ted in substantia1 decrease in B/P even if ideaT body weight was not achieved. This B/P reduction is in addition to any effect of restricted sodium intake. Based on these findings. the 1984 Joint Nationa1 Committee on l7 Detection. Eva1uation. and Treatment of High BTood Pressure recommends that weight reduction shoqu be an integraT part Of therapy for a11 Obese persons (> 115% of ideaT body weight) with hypertension. The CommitteezaTSO recommends counseTing hypertensive patients for moderate sodium restriction. MacGregor et a1. (1982) and Parijs et a1. (1973) have documented reductions in e1evated B/P with moderate dietary sodium restriction to a 1eve1 of 70 to 90 mEq/day. A1though not a11 patients with hypertension respond to moderate sodium restriction. there is no hazard and it may aTso decrease the amount of potassium Tost with diuretic therapy. In a study of the re1ationship of aTcohOT and the cardiovascu1ar system. KTatsky (1982) found that heavy aTcohOT consumption (> 56.8 g/day of a1cohoT) may e1evate arteriaT B/P. One ounce (28 g) of ethanoT is contained in 2 oz. of 100 proof whiskey. 8 oz. of wine. or 24 oz. Of beer. Therefore. the recommendation for controTTing hypertension is for those who drink to drink moderateTy (< 56.8 g/day). As with a1cohOT. nicotine wiTT raise arteriaT B/P for a short time. StiTT. there is no definitive evidence that proTonged nicotine use wiTT increase the risk of hypertension. Persons who smoke. however. do increase their risk of cardiovascu1ar disease (StamTer et a1u» T975) and for this reason. avoidance of smoking is encouraged in hypertensive patients. A reguTar isotonic or aerobic exercise program can aid in weight controT and stress reduction and may be heTpfuT in reducing B/P through these mechanisms. Exercise programs shoqu be initiated by 18 hypertensive patients graduaTTy. after consuTtation with a hea1th-care providen. These exercise programs are recommended for patients with uncomp1icated essentiaT hypertension. Stress reduction and return Of norma1 sTeep patterns are accomp- Tished by various re1axation or biofeedback therapies. Studies by PateT et a1. (1981) and EngeT et a1. (1983) have demonstrated modest. but substantia1. B/P reduction with the use Of re1axation and biofeed- back techniques in seTected groups outside the Taboratory. The B/P reductions have been maintained for as Tong as 1 year. As with other nonpharmacoTogic approaches. the techniques of re1axation and biofeed- back are most reTevant for the treatment of mde hypertension but are aTso usefuT as an adjunct to pharmacoTogic therapy in more severe hypertension. When e1evated B/P cannot be reduced by the use of nonpharmacoTogic therapies or when diastoTic B/Ps are persistentTy 95 mm Hg or above. pharmacoTogic therapy is recommended in addition to nonpharmacoTogic measures. The most common approach to pharmacoTogic therapy is "stepped care." Stepped care is suggested as a guide for treating hypertensive patients when drug therapy is indicated. This approach 1eaves room for indi- viduaTization and fTexibiTity in management. and it has been used effective1y in major c1inica1 triaTs demonstrating reduction of morbidity and mortaTity. In numerous studies. normotensive 1eveTs have been achieved in more than 80% of the patients using this simpTe and reTativeTy inexpensive approach to treatment. 'The SC program suggests initiating therapy with a smaTT dosage of an anti-hypertensive drug. increasing the dose of that drug. and then adding or substituting one drug after another in graduaTTy increasing doses as needed untiT goa1 B/P is achieved. side effects become intOTerabTe. or the maximum dose Of each drug has been reached. 19 See Tab1e 2.2 for an out1ine Of the stepped-care approach. Tab1e 2.2: Stepped-Care Approach to Drug Therapy Step Drug Regimen 1 Begin with Tess than a fuTT dose of either a thiazide—type diuretic or a B-bTocker; proceed to fuTT dose if necessary and desirab1e 2 If B/P controT is not achieved. either add a smaTT dose of an adrenergic-inhibiting agent or a smaTT dose Of thiazide-type diuretic; proceed to fu11 dose if necessary and desirab1e; addi- tionaT substitutions may be made at this point 3 If B/P controT is not achieved. add a vasodi- Tator. hydraTazine hydrochToride. or minoxidiT for resistant cases 4 If B/P controT is not achieved. add quanethidine mcnosquate Source: Joint Nationa1 Committee on Detection. DevaTuation. and Treat- ment Of High BTood Pressure (1984). COTTaboration among hea1th professionaTs is a further component of successfuT B/P controT effort.(Grim a Grim. 1981; HiTT et.aT.. 1984; Kochar. 1981; Mason. 1982). The nurse's rOTe in this coTTaborative effort is expanding but remains based on standards of nursing practice. CardiovascuTar nursing practice is defined in Standards Of Cardiovascu- Tar Nursing Practice (1975) as the nursing care of individuaTS who have known or predicted a1tera— tion in cardiovascu1ar physioTogic function. The scope Of cardio- vascu1ar nursing practice encompasses those nursing activities which assist the individua1 to modify his Tife styTe and environ- ment so that he can attain optimum cardiac function and acceptabTe qua1ity Of Tife in congruence with his Tife goaTS. (p. 7) 20 Inc1uded in the standards—-which are based on the nursing process of assessment. deveTOpment Of nursing diagnosis. deveTOpment of goaTS and a nursing-care pTan to achieve the goaTS. and evaTuation of the pTan of care--are factors particuTarTy re1ated to this study. Assess- ment factors inc1ude hea1th data on the individuaT's perceptions and expectations which re1ate to her/his hea1th-111ness state and hea1th- care services. the psychosociaT behavior of the individua1 and her/his responses or patterns Of coping/adaptation. and information as it re1ates to the individuaT's habits and socia1 and work roTes. The foTTowing patient outcomes are Tisted under goaTS for nursing care: (a) the individua1 participates in pTanning the modification of Tife sty1e and accepts the modifications. (b) the individua1 demonstrates effective coping mechanisms to adapt to her/his aTtered Tife sty1e. and (c) the individua1 maintains a dietary intake. activity pattern. and pharmacoTogic regimen that is compatibTe with therapeutic and personaT goaTs (Standards Of CardiovascuTar Nursing Practice. 1975). £EEJ£J.§HDRQ£I As earTy as 1951. researchers proposed that socia1 support was somehow positiveTy re1ated to hea1th (Durkheim. 1951). ‘The exact nature Of the rOTe of socia1 support in hea1th. however. has been eTusive. EmpiricaT findings in the studies of Brown et a1. (1975). LowenthaT and Haven (1968). and MiTTer et a1. (1976) have shown that emotionaTTy intimate reTationships are consistentTy found to be 21 protective Of hea1th and that one intimate re1ationship is more protec- tive than severa1 more casua1 re1ationships. CasseTT (1974). Cobb (T976). and Dean and Lin (1977) supported the notion that socia1 support acts as a protective buffer against major stress and Tife changes that may increase an individuaT's susceptibiTity to physica1 or menta1 iTTness. More recentTy. PearTin et a1. (1981) pTaced socia1 support aTOng with coping as interventions that do not act directTy to buffer iTTness but rather minimize the eTevation of iTTness by dampening the antecedent process. Bi11ings and Moos (1980) aTso paired socia1 supports and coping as "attenuating" the re1ation- ships between undesirabTe Tife events and personaT functioning (which inc1udes hea1th). A review of 22 research articTes in which support-reTevant vari- ab1es were measured in re1ation to foTTowing a therapeutic regimen (Haynes & Sackett. 1979) reveaTed 15 studies that reported positive re1ationships. six studies that found no association. and one study that reported evidence contrary to the hypothesis that socia1 support is positiveTy associated with foTTowing a therapeutic regimen. CapTan (T976. 1979) and Kirscht et a1. (1981) both reported positive reTation- ships between socia1 support and seTf-reported adherence to a therapeu- tic regimen among Targe groups of hypertensive patients. CapTan (1979). however. concTuded that compTiance was highest when both socia1 support and motivation are present and that compTiance appeared to determine support as weTT as be determined by it. 22 In T982. MinkTer set forth three major hypotheses regarding the mechanism of socia1 support. The first hypothesis suggests that the existence of a supportive network may mean that an individua1 is being encouraged by his/her socia1 contacts to take preventive action or seek needed hea1th care. She cited resu1ts Of Berkman and Syme (1979) from a study in which the mortaTity experience of 7.000 ATameda County residents over 9 years was foTTowed. The most significant finding of the study was that those study subjects with few ties to other peop1e had mortaTity rates two to five times higher than those with more ties. These differences were independent of seTf-reported hea1th status and traditionaT risk factors. The second hypothesis proposes that support provided by members of an individuaTLS socia1 networks increases coping abiTity and hence heTps "short-circuit" the iTTness response to stress. Studies by AtchTey (1979) and Gore (1978) provided support for this hypothesis with findings that job Toss. retirement. and bereavement appear to be Tess traumatic in terms of morbidity and mortaTity when support net- works are avaiTabTe and used. The third hypothesis budes in part on the first two and suggests that. over time. peopTe's perceived support from others may Tead them to a more generaTized sense Of controT. In view of Antomorsky (T979) and Satariano (1981). it is perhaps this more gTobaT feeTing of controT and confidence that serves as the missing Tink expTaining why socia1 support woqu be so criticaTTy re1ated to hea1th. 23 It is not the intention of this research to unraveT compTeteTy the compTex and iTT-defined concept of socia1 support. Rather. an attempt wiTT be made to add to the body Of know1edge regarding socia1 support by examining an important aspect of the concept. the patient's percep- tion of socia1 support. 'This wiTT be accompTished using a conceptua1 framework that combines generaT systems theory and the nursing theory of Imogene King. WWW Macaw—13.1119 In 1933. Ludwig von BertaTanffy pubTished "Modern Theories of DeveTopment: An Introduction to TheoreticaT BioTogyJ‘ This paper marked the beginning of a dynamic new era in the sciences which stressed whOTeness and organization. The purpose was to deveTop theoreticaT modeTs. principTes. and Taws that promote the unity Of science and eTiminate the dupTication Of theoreticaT efforts through improved communications between the discipTines. The generaT systems theory appTies universaT principTes to systems or their subcTasses irrespective Of the particuTar kind Of system. the nature of the component eTements. or the re1ationship or force between the eTements. The systems modeT provides a framework by which otherwise unconnected parts are integrated and many different pieces faTT into pTace (Hazzard. T971). BertaTanffy (1955) referred to a system as a compTex of eTements in interaction; HaTT and Fagan (T968) spoke of a system as a set of parts of components together with reTationships between the parts and 24 between the properties Of the parts; whi1e Hart and Herriott (1977) defined a system as a set of components interacting with each other and a boundary that possesses the property of fiTtering both kind and rate Of fTOw of inputs to the system and outputs from the system. In a 1iving system. the modeT used for this thesis. the boundaries are semi- permeabTe to aTTow for an exchange of matter. energy. or information between the system and the environment. The boundary can be thought of as a hoteT door that connects two adjoining rooms and Tocks or unTocks from either side. Systems on either side of the boundary may choose whether or not to "unTock the door" and how much to»"open the door" to exchange matter. energy. or information. This function of the system is known as boundary maintenance. The maintenance Of the boundary between the system and the environment is one of the system's main functions. Bredemeir (1962) suggested there are four ways in which a system maintains its boundary whi1e interacting with the environment. First. the system retains within itseTf matter. energy. and information necessary for the system to function. Second. the system obtains from the environment needed matter. energy. or information. Third. the system cgntains within the environment matter. energy. or information not wanted or required. Fourth. the system disposes of unwanted matter. energy. or information to the environment. The matter. energy. and information Obtained by the system are input: to the system. Disposed-Of matter. energy. and information are outputs from the system. When there is some constancy in the input-output exchanges of an open system. it is said to be in a 25 steady state. This does not impTy a static state. for continuaT dynamic processes are taking pTace whi1e the reTationships between the system parts remain at or near a particuTar baTance (HaTT. T977). HaTT (1977) noted. "It is the purpose and goa1 Of the open system to determine the [steady] state desired. It is the system's work to make decisions which wiTT maintain its continuum of baTance and pur- posefuT direction in the environment" (p. 23). Decisions necessary to maintain a desired state in a system are based on information input to the system as feedback. Feedback as defined by MeTcher (1981) is the property Of an open system to use aTT or part Of an output as an input to effect future outputs and/or behaviors. Both positive and negative feedback are necessary for a system's survivaT. Positive feedback is system output information that wiTT cause a system to change its pre- vious activity. Negative feedback is output information that wiTT cause the system to maintain its current activity. Vickers (1959) proposed four ways the system may adjust through feedback when usuaT responses faiT. The system (a) a1ters itseTf. (b) a1ters the environ- ment. (c) withdraws from the environment. or (d) a1ters its desired state or goa1. In the modeT Of an open Tiving system used for this thesis (Figure 2.1). input. in the form of demands and supports. is taken into the system and through a compTex dynamic interaction of processes and internaT feedback. decisions. or actions may be outputs to the environment. In this modeT the goa1 or desired state of the system is normotension. The demand of the system is foTTowing a therapeutic 26 A._mm_ .ccmNNm: EOcuv .Emum>m mc_>__ coco “_.N Ocammu goo; xumnuoou .mccOuxm :o_uumcmuc_ .l . . _ . . _ _ . . " u_Emc>o " . . _ . _ WIIIIIIIJ . ( mzo_eu< mo " u + + Pampas ..i.mzo_m_uwo m r. IIIIIII .2 m " xumnuaac u " co_umEc0mc_ _mcc0uc_ " u u _ . h " TIT: mhmommam Aflllllllll moz; Ou Beam—OE Emum>m mc_>m_ coao "~.~ Oc:m_u —l I l I I | | | I I I I I I I I I I I I I I I I I | I I I l l I I I I I l I I l l I I I I I I l _J camcoa coaaaman codaam u u A hmoamzm coemmOc _m_u0m mo :o_uaoucom 4<_u0m 03:89.05 1 IIIIII J < + + Sagohll Lo 3033.... _I IIIIII L .582. _m_OOm Om: Cu mco_uum co mco_m_uoo ucm_umansucoQa:m _m_OOm mo comuaoucom co_chucOa>; .0cucou cu cOE_mOc u_uaoa umcozu mc_30__0m A11: l'"""'"""""""""I _ _ _ _ _ _ _ _ _ _ _ _ _ .e. m 30__0m _ _ _ _ _ _ _ _ _ _ _ 29 A._wm_ .mc_x 50cm neuamv_mc0ucma>: 30 demands made upon them. the supports avaiTabTe to them. and responses appropriate for themseTves. King (T981) defined socia1 systems as groups of individuaTs who join together in a network or system of socia1 reTationships to achieve common goaTS deveToped around a system of vaT ues. Nurses are 1ike1y to work with one or more of the three 1eveTs Of soci a1 systems proposed by King--individuaT. group. or society. The individua1 or personaT socia1 system is the focus Of the thesis. In it. both the hypertensive patient and the nurse or other support person are invoTved in a dyadic re1ationship in which both individuaTs are perceiving the other simuT- taneousTy. making judgments. and formuT ating in this menta1 process some kind of action. (See Figure 2.3.) This is a continuous dynamic process rather than separate incidents. in which the verbaT and nonver- baT responses of one person infTuence the perception and response of the other person and vice versa. This attention to the importance of the individuaT's perceptions and its rOTe in an interaction make King's theory unique among nursing theorists and Of va1ue in examining the data in this study. According t0 King (1981). the cTients' perceptions serve as a basis for nurses to gather and interrupt information regarding cTients. as weTT as expTain- 1'19 their behaviors and interactions with others. King defined percep- t‘Ions as "each individuaT's representation or image of reaT ity; an awareness of objects. persons and events" (p. 20). Further. she stated that perception is "a process of organizing. interpreting. and trans- f0r‘ming sensory data and memory; it is a process of human transactions 31 with the environment and gives meaning to one's experience and infTu- ences one's behavior" (p. 24). Was The concept of perception is aTso a key point in King%s(1981) definition of nursing. which states that "nursing is a process of human interactions between nurse and cTient whereby each perceives the other and the situation; and through communication they set goa1 s. expTore means and agree on means to achieve goaTs" (p. T44). The goa1 of nursing practice is the attainment and maintenance Of hea1th. Most cTients aTso have hea1th as a Tife goa1. For the particuTar group of patients being studied. a portion of their hea1th is controT of their hypertension through foTTowing a therapeutic regimen. The concept of hea1th as seen by King is very appropriate for these patients. She described hea1th as a dynamic state in the Tife cycTe of an organism which impTies continuous adaptation to stresses in the internaT and externaT environment through the optimum use of one's resources to achieve maximum potentiaT for daiTy 1iving. In this instance. the internaT stress may be the physioTogicaT effects of hypertension as weTT as the psychosociaT stresses of having a chronic iTTness whose controT reTIes heaviTy on aTterations in current TifestyTe habits. ExternaT stresses may be the expense and time invoTved in contrOTTing hypertension and pressure by famiTy. friends. and hea1th-care providers to gain or maintain controT Of the hypertension. By communicating with the cTient regarding specific short-term goaTs in foTTowing the therapeutic regimen. the nurse and hypertensive 32 cTient can set goaTs that are reaTistic and achievabTe. An exampTe might be foTTowing an exercise program of waTking 20 minutes three times a week. The nurse and cTient can then expTore means and resources avaiTabTe to he1p the cTient achieve the goa1. The cTient may have a famiTy member. friend. or pet to wa1k with; he/she may have someone who coqu remind him/her to waTk; theecTient might write him/ herseTf notes and put them in conspicuous pTaces or set a wristwatch aTarm for the time of the waTk. The nurse and cTient next agree on which Of these means the cTient wiTT use in achieving her/his goa1. WhiTe the nurse may identify resources she/he sees as avaiTabTe to the cTient. the finaT decision as to what resources are used wiTT be based on what the cTient perceives as avai1ab1e and appropriate. By care- fuTTy examining the cTientfls perception of the resources. the nurse can faciTitate the cTientfls adaptation and improve her/his potentiaT for meeting both short-term goaTs and the Tong-term goa1 of hea1th. In this study I wi11 attempt to identify the hypertensive patientfls perceptions of the resource of socia1 support in foTTowing a therapeutic regimen. An attempt is aTso made to identify differences in perceptions among the three mOdaTities of the therapeutic regimen (medication. diet. and exercise) and among various subgroups of the samp1e based on sociodemographic characteristics. With this informa- tion. the nurse can assist the cTient in identifying sources and types of socia1 support in their environment that other hypertensive patients have perceived as heTpfuT in foTTowing a therapeutic regimen. 33 In summary. systems theory can be weTT used by nurses to examine many areas Of nursing practice. Itlnay be most heTpfuT when concepts are not we11 defined or when reTationships among variabTes have been uncTear. as with socia1 support. Systems theory can be particuTarTy usefuT to nursing as a profession. as it tries to estabTish further its body of know1edge. Nursing care is so varied and compTex that it needs a theoreticaT base to organize. simpTify. and describe exactTy what it is that nurses do that is vitaT to the hea1th care of our cTients. CHAPTER 3 LITERATURE REVIEW IDILQQUQIIQD The purpose of this study is to identify hypertensive patients' perceptions Of socia1 support in foTTowing a therapeutic regimen. ‘The Titerature review wiTT inc1ude a brief review of the Titerature on foTTowing a therapeutic regimen for hypertension and an in-depth review of Titerature on the major concept of the study. socia1 support. 'Two areas Of socia1 support wiTT be addressed: definition and measurement of the concept and works that examine the re1ationship between socia1 support; and foTTowing a therapeutic regimen for hypertension. MW Reports Of an enormous number of studies exist in the Titerature regarding compTiance/adherence. Researchers examined foTTowing a therapeutic regimen for a wide range of iTTnesses fran minor acute prob1ems such as otitis media to major injuries or iTTnesses such as severe burns. stroke. or cancer. This review of the Titerature wiTT consist of schoTarTy papers and empirica1 research which address foT- Towing a therapeutic regimen for hypertension. Researchers have agreed that the greatest cha11enge in the treat- ment and management of hypertension is getting the hypertensive patient 3h 35 to foTTow a therapeutic regimen (BaiTe & Gross. 1979; CapTan et a1.. 1976; DanieTs & Kochar. 1980; Dracup & MeTes. 1982; Heine. 1981; Kirscht et a1.. 1981; Sackett & Haynes. 1976; WOTTam & Gifford. 1978). The maintenance or foTTowing of a therapeutic regimen is noted in the Titerature as compTiance or adherence to the regimen. ConverseTy. noncompTiance is not foTTowing the regimen. Not foTTowing the regimen may be expressed through a variety of behaviors. such as deTay in acquiring hea1th care. appointment faiTure. not taking prescribed medi- cation. or not foTTowing recommended Tife-styTe changes in diet. exer- cise. stress reduction. smoking. or aTcohOT consumption. In reviewing the Titerature on foTTowing a therapeutic regimen for hypertension. three major prob1ems were discovered. First. operationaT definitions of adherence/compTiance are inconsistent. Marson (1970) noted that even the most objective measures of medication compTiance. urine drug excretion and b100d tests. have presented difficuTties in an accurate measurement of patient compTiance. For a1though the tech- niques of the procedures may be the same. the operationaTized defini- tions of compTiance are varied. WhiTe one researcher may base compTiance on one urine test at an outpatient cTinic. another may use a series Of tests as the basis for their estimate. Less objective measures Of compTiance such as piTT counts or se1f- report on compTiance with medication. diet. or exercise have even more chance for error in the estimation of compTiance. It shoqu be noted that seTf-report methods (questionnaire or interview). whiTe Tess accurate. may ref1ect the patient's behavior with regard to the totaT 36 regimen rather than a portion of it. StiTT. the wide variation in the operationaTization of these compTiance measures makes it difficuTt to compare resu1ts of studies. The second prob1em invoTves inconsistent description of the way in which adherence/compTiance was operationaTized. This was especia11y true in earTier works. but there continues to be a discrepancy. Some investigators have given a brief. vague description which woqu be difficuTt to repTicate without a good dea1 of further information (Widmer et a1.. 1983). Others have written page after page of extremeTy detaiTed descriptions of the measure (Cap1an et ah" 1979). The finaT existent prob1em is the overTap or confounding Of vari- abTes which may affect adherence/compTiance. VariabTes such as patient characteristics (sociodemographic characteristics. personaTity traits. and desire for active participation). disease characteristics (dura- tion. severity. and symptoms). treatment characteristics (compTexity of the regimen. side effects. cost). provider-patient re1ationship charac- teristics (consistent provider. convenience. demonstration of warmth and concern). hea1th beTiefs (perceived susceptibiTity. severity. bene- fits. barriers). knowTedge. and socia1 support have been studied as factors that may affect adherence/compTiance. ResuTts of the studies have shown interactions of these variabTes not onTy with adherence/ compTiance but aTso with each other. In many instances it is extremeTy difficuTt. if not impossibTe. to unraveT these interactions with even the most compT ex anaTytic techniques. To further compound the diTemma. 37 the majority of the variabTes noted above aTso have had wideTy varied and inconsistent operationaT definitions. Sackett (T980) cited the resu1ts of five studies that examined compTiance based on taking anti-hypertensive medication. SampTes varied from 49 bTack hypertensive patients from Detroit to 240 white steeT workers in Canada. The measure of compTiance in three Of the studies was piTT count. but a11 three were operationaTized differentTy. One study measured bTood 1eveTs of PropranodOT (Briggs et a1.. 1975). and the Tast measured Thiazide 1eveTs in urine (LowenthaT et a1” 1976). In addition. definitions of compTiance ranged from 80% to 95% of prescribed medication; each study used a different percentage. LeveTs of compTiance ranged from 47% for those taking 95% of the medi- cation (BTackweTT. 1976) to 53% for those taking 80% Of the medication (Sackett et a1.. 1975). CompTiance for the two studies was defined as detectabTe 1eveTs of medication in b100d samp1e (51%) or urine samp1e (65%). This brief review by Sackett iTTustrates the difficuTty in com- paring resuTts from studies up to 1980. The question that needs to be answered. then. is: Have investigators heeded the advice of their predecessors and adopted a standard measure to operationaTize compTi- ance with a hypertensive regimen? To answer this question. the remainder of this section. on foTTowing a therapeutic regimen for hypertension. wiTT focus on studies conducted since 1980. Two studies have examined the effect Of interventions that increase know1edge Of hypertension on bTood pressure controT. Tanner 38 and Noury (T981) expTored the effect of structured teaching about essentiaT hypertension on controT of diastoTic bTOOd pressure in hyper- tensive patients. The samp1e consisted of 30 bTack (n = 14) and white (3 =16). maTes (n = 11) and femaTes (n =19) between the ages of 18 and 65 who had been diagnosed as having essentiaT hypertension. A11 subjects had diastoTic bTOOd pressure readings of 90 mm Hg or above on their Tast two visits to the hea1th center before entry into the study. The outcome variabTe Of mean diastoTic B/P was obtained from diastoTic bTOOd pressures recorded in the sitting position. on the right arm. at eight scheduTed appointments over a 4-month period. ‘The same institutionaT mercury sphygmomanometer was used to measure a11 bTOOd pressures. KnowTedge was measured on a 15-item instrument deveToped by the investigators and re1ated to eight specific areas: hypertension. medication. diet. stress. exercise. smoking. TifestyTe. and bTOOd pressure monitoring technique. .ATT subjects compTeted the TS-question instrument pre- and post-intervention. Participants in the experimentaT group were given a brochure. "Guide to EssentiaT Hypertension." which was deveToped by the investi- gators. Information contained in the guide was ref1ected in the know1- edge instrument and was discussed with the experimentaT participants at each of the eight visits to the hea1th center. ExperimentaT subjects were encouraged to ask questions and discuss any prob1ems they were experiencing. Subjects in the controT group were given the usuaT care provided at the hea1th center. 14%. appointments. at various 39 intervaTs. with physicians for the purpose of evaTuation Of their hypertension. The resu1ts of the study showed no significant difference between experimentaT and controT groups on mean diastoTic bTOOd pressure (91.26 mm Hg and 92.33 mm Hg. respectiveTy). The experimentaT group did have higher posttest than pretest scores (1; = —2.31; .p, < .05) and higher posttest scores than the controT group (1 = -3.40; p_< .05). supporting previous research in which instruction was found to increase know1edge but not increase compTiance (Sackett et a1.. 1975). A1though the authors cited 1imitations (sma11 samp1e size. seTec- tion from one agency. and nonincTusion of dropouts from therapy). the study was concise and resu1ts were easiTy understood. OperationaT definitions of compTiance and know1edge were c1ear and described in detaiT. The Timited number of variabTes eTiminated interaction or confounding. In the second study. Powers and WOOTridge (1982) examined the effect of four aspects of an educationa1 program on the outcome variabTes of patient know1edge. attainment of identified goaTs. and mean arteriaT bTOOd pressure. The 160 subjects who participated were from five types of cTinicaT settings: an inner-city hea1th faciTity. a community hea1th center. private hospitaT c1inics. university hospitaT cTinic. and private physicians's offices. A11 subjects were under age 75. on prescribed medication for essentiaT hypertension. not under psychiatric care. and had one recorded B/P greater than 140/90 in the 6 months prior to the study. A high portion of the subjects were bTack 4O (72%) and women (70%). There was aTso a high representation of patients of Tow socioeconomic status. 14%. unempToyed (34%). unskiTTed workers (29%). and those with an eighth-grade education or Tess (46%). A factoriaT design was used to test the reTative effectiveness of four variations in the nurse's hea1th-teaching approach. These variations were (a) the directiveness of the nurse's interaction sty1e. (b) the degree of emphasis of seTf-responsibiTity and active hea1th-care participation by the patient. (c) the degree of emphasis on negative consequences of uncontroTTed hypertension. and (d) the number of meetings with the project nurse. ‘The crossing of the four dichoto- mous approach variabTes resu1ted in a factoriaT design with 16 dif- ferent treatment combinations. The intervention invoTved patients receiving hypertension instruction from three sources: a 2-1/2 minute tape recording. a foTTow-up discussion with the project nurse. and a 12-page iTTustrated brochure presenting facts about essentiaT hypertension. The project nurse and the patient jointTy identified goaTs re1ated to the patientfis specific condition and/or generaT hea1th and Tife circumstances. toward which the patient was to work. Each patient was then scheduTed for a repeat visit 2 weeks Tater (Time 2). At Time 2. the project nurse and patient discussed the information in the brochure. reviewed the patientfis medication. and discussed progress toward identified goaTs. At Time E. approximateTy 3 months after Time 1. the project nurse reviewed the educationa1 information and re-emphasized the importance 41 of continuing treatment. ‘The degree of progress made toward achieving goaTs was aTso discussed. The variations in educationa1 approach were manipuTated and opera- tionaTized in the foTTowing ways: directiveness Of the nurseksinter- vention was manipuTated by changing the characteristics in the interac- tion according to specific guideTines. The Tow direct approach informed the patient about what was taking pTace in a factuaT. imper- sonaT manner; The high indirect approach pTaced emphasis on accepting feeTings. eTiciting comments. and discussing the patients' ideas. Patients assigned to the high responsibiTity/participation group were presented with audio-visuaT and written materiaT that stated it was up to the patient to decide and act to reduce high bTOOd pressure. The physician and nurse were presented as potentiaT sources of he1p. Patients assigned Tow responsibiTity/participation were presented with materiaTs that ordered them to controT their bTOOd pressure. with emphasis pTaced on foTTowing the physician's orders. The emphasis on negative consequences was operationaTized by a1tering audio-visuaT and written materiaT presented. Subjects in the high emphasis group heard and saw arousing terminoTogy and graphics (emy. "siTent kiTTerfl' "stroke." "death." "crippTe." bright red H on yeTTow brochure. 5 Of 40 tombstones coTored to denote deaths re1ated to hypertension). Those subjects in the Tow emphasis group received Tess detaiTed instruction. and a brochure in which Tess arousing terminoTogy and graphics were used (e.g.. "harm." "concern." a famiTy standing under an umbreTTa TabeTTed "BTOOd Pressure ControT." smaTT bTack H on tan brochure). The 42 additionaT meeting variabTe was operationaTized by whether or not three additionaT meetings were scheduTed 2 weeks apart between Time 2 and Time E. At these meetings emphasis was pTaced on ways to achieve goaTs identified. The styTe of the educationa1 approach was maintained according to the originaT treatment condition assigned. Outcome variabTes inc1uded measures of patient know1edge. assess- ment by the nurse of patient attainment Of goaTs. and reduction of the patientfls mean arteriaT bTOOd pressure. These variabTes were opera- tionaTized in the foTTowing ways. Three scaTes. deveToped by the investigators. measured know1edge of etiOTogy and symptoms. prescribed medications. and bTOOd pressure eTevation. ‘The measurements were made during three meetings with the patient: Time 1 (first meeting).‘Time 2 (repeat meeting 2 weeks Tater). and Time E (exit meeting). Questionnaires were administered by a research assistant who was unaware of treatment assignment. Degree Of goa1 attainment (measured at Time 2 and Time E) was scored by asking the nurse to identify for each patient the specific goaTs decided on at Time T and then to evaTuate the degree of progress toward attaining these goaTs. TotaT goa1 scores and subscaTe score were caTcuTated using the foTTowing rating system: regressed = -1. unchanged = 0. improved = +1. fuTTy achieved = +2. Types of goaTs inc1uded generaT medication behavior. diet. exercise. weight controT. controT of smoking or drinking. stress reduction. improvement of rOTe reTationships. unempToyment. and finances. Data Of B/P were obtained from the research nurse's measures and the patient's medicaT record. A mean arteriaT B/P 43 of 106 mnan is equivaTent to 140/90. the usuaT criterion for cTini- ca11y significant bTOOd pressure eTevation. The main goa1 appTicabTe to a11 patients was B/P reduction. About three-fourths of the samp1e (74.4%) exhibited e1evated bTOOd pressure at Time 1 (average 114 mm HgL. The mean arteriaT bTOOd pressure at Time E averaged 108.2 mm Hg. a mean drop of 6 mm Hg (1 = -4.27; p, < .001). The mean arteriaT bTOOd pressure aTso remained Tower for a 6- month period foTTowing Time E (average 108.5 mm Hg). Despite these significant overa11 decreases in bTOOd pressure. a regression ana1ysis found no significant regressions of B/P changes on the approach variabTes or interaction. This suggests that the variations in approach had TittTe or no effect on the reduction of bTOOd pressure. There were some statisticaTTy significant interactions of the other variabTes studied. but they wiTT not be discussed in this review since the focus is on compTiance measures. A detaiTed discussion may be found in Powers and WOOTridge (1982). This study was much more compTex than the first. WhiTe subjects did experience decreases in bTOOd pressure. the reason for this is not c1ear. The compTexity of the study added TittTe information to how or if know1edge affects compTiance. A1though both studies (Powers 8. WOOTridge. T982; Tanner & Noury. 1981) examined know1edge as it affects compTiance. operationaTized as contrOTTed B/P. both know1edge variabTes and compTiance variabTes were operationaTized in very different ways. It woqu appear that inconsistent operationaTization is a continuing prob1em even in simiTar research areas. 44 Examining another definition Of compTiance. Meyers et a1. (1983) compared four strategies designed to improve appointment compTiance of hypertensive individuaTs. One hundred forty-eight subjects identified as hypertensive and missing a c1inic appointment at a community hea1th center were recruited through a city-wide hypertension screening program. Subjects ranged in age from 15 to 87 (mean = 46.7) and were from a predominantTy Tow-income. b1 ack community. Of the subjects. 44.6% were maTe and 55.4% femaTe; 80 were defined as borderTine hyper- tensive (diastoTic B/P between 90 and 99 mm Hg) and 68 as cTinicaT hypertensive (diastoTic B/P 3 100 nun Hg). Appointment compTiance was operationaTized as keeping a scheduTed appointment. To test the four strategies. appointments were made for borderTine hypertensives at the hea1th center approximateTy 2 months after the screening and for cTinicaT hypertensives 2 weeks after initiaT screening. IndividuaTs who reported to the hea1th center did not participate in the study. whi1e those who missed appointments were referred for inc1usion in the study. BorderTine hypertensive and c1inica1 hypertensive individuaTs who missed appointments were randomTy assigned to a postcard. phone ca11. home visit. or rotating contact group and contacted within a week of the missed appointment. Subjects in the postcard. phone ca11. and home visit groups received up to three presentations of a compTiance message in the appropriate modaTity. one foTTowing each Of three consecutiveTy missed appointments. Subjects in the rotating contact group aTso received up tO three presentations. one postcard. phone caTT. and home 45 visit each foTTowing one of three consecutiveTy missed appointments. The order of the rotating presentation was baTanced across the six possib1e combinations. The message content was identicaT among the modaTities. It brief1y stated that high bTOOd pressure coqu be harm- fuT if uncontroTTed. but that it was easiTy treated. It tOTd subjects that when their bTOOd pressure was Tast checked it was high and since they had missed their Tast appointment they needed to come in and have it rechecked. The hea1th center address and phone number were inc1uded. .A11 contacts were made by trained upper-1eve1 undergraduate or graduate students. Chi-square ana1ysis was used to evaTuate appointment compTiance. Significant differences were found among the modes in eTiciting appointment compTiance (X2 [3] = 21.87; p, < .01). The rotating contact (81.1% compTiance). home visit (67.7%). and phone caTT (54.1%) modes were a11 significantTy more effective than postcard (29.7%). CompT i- ance rates across aTT contact conditions were higher for cTinicaT hypertensives than borderTine hypertensives. The second contact pro- duced the highest rates Of appointment compTiance across a11 message modes. When specific comparisons were made between compTiance rates for the baseTine hea1th center samp1e and the study's rotating condi- tion. the rotating condition was superior for the totaT samp1e (X2 = 29.87; p_< .01). The cost Of the contacts was caTcuTated for each message mode. and those figures were used to determine the cost for each subject in each condition. For a11 subjects. home visit was the most expensive 46 mode at $13.92 per compTiant subject. Post card. rotating. and phone caTT foTTowed at $6.49. $4.92. and $2.20 per compTiant subject. respectiveTy. This study examined a different definition of compTiance. keeping scheduTed appointments. The independent variabTes. four contact modes. were operationaTized in a precise manner. as was the measure of compTiance. Information Obtained from the study is usefuT to other researchers and hea1th-care providers in c1inica1 practice. It documents not onTy an empiricaTTy effective but aTso a cost effective method of increasing appointment compTiance in hypertensive patients. A unique or. as the authors proposed. "state of the art" study was conducted by Widmer et a1. (1983). The investigators in this study stated that previous research on compTiance among hypertensive cTients was severeTy biased because it had focused on patients from Targe urban c1inics and teaching hospitaTs. To rectify this shortcoming. they examined compTiance characteristics of 291 hypertensive patients from a ruraT midwestern area. The T-year prospective study invoTved patients of seven famiTy physicians in towns with popu1ations Of 700 to 8.500. Physicians were not asked to foTTow any speciaT criteria for diagnosis. treatment pTan. B/P recording technique. or record-keeping protocoT. If a pharmacy was not present in the town. physicians dispensed medica- tion. Patients were unaware they were invoTved in a research study. The 29T-patient samp1e had a mean age Of 63.2 years. Of the subjects. 37.5% were maTe and 63.5% femaTe. Farm dweTTers comprised 47 53% of the group and those 1iving in towns 47%. The raciaT background Of the samp1e was not given. The OperationaT definition of compTiance was based on two assump- tions: (a) drugs bought and paid for by semi-ruraT and ruraT midwest- erners are taken. and (b) refiTTs purchased on scheduTe indicate adher- ence to the prescription. Therefore. a compTiance percentage was determined by dividing the number of piTTs purchased by the patient by the number of piTTs he or she required during the study period. If a patient was on mu1tip1e medications. the medication with the Towest compTiance was used for ana1ysis. The mean number of days in the study was 272.3. IndividuaT compTiance percentages of the 291 subjects were used to caTcuTate a group mean of 86.6%. A person from each office recorded demographic information. date Of visits for hypertension. name and number of prescribed medications. directions for use. and B/P readings. The independent variabTes were sex. how medications were dispensed. median weight. compTications. how often medications were taken. and smoking status. Specific operationaT methods were not given. Data were taken from information recorded by Office personneT. For purposes of ana1ysis. the independent variabTes were dichoto- mized. that is. sex--maTe/fema1e. dispensing Of medication--by doctor/ by prescription. median weight--above median/beTow median. comp1ica- tions--no compTications/one or more comp1ications. how Often medication taken--one or two times daiTy/three or four times daiTy. and smoking status--smoker/nonsmoker. .ATthough not in the operationaT definition 48 of compTiance. diastoTic B/P was cited as a measure of compTiance in the resu1ts. Taking piTTs three or four times a day was the onTy measured factor that significantTy affected compTiance (p < .05). Of those subjects who took medication once or twice a day. 78% had diastoTic B/P oi’90 mnIHg or Tess; Of those who took medication three or four times daiTy. 63% had a diastoTic B/P of 90 mm Hg or Tess. A1though not statisticaTTy significant. considerabTy more femaTes than maTes had diastoTic B/P Tess than 90 MHTIEL The authors suggested their resu1ts supported the contention of Sackett et a1. (1975). who stated comp1i- ance of at 1east 80% shoqu achieve a positive outcome from an anti- hypertensive medication treatment regimen. Seventy-five percent of subjects had compTiance percentages of 80% or more (X2 = 3.822; g < .0506L The researchers aTso cited that a decrease in mean diastoTic B/P (83.7 mm Hg) from pretreatment 1eveTs of 90 mm Hg or more indicated good controT at the end Of the study. This finaT research was very unusuaT when compared to the first three studies. The samp1e had a mean age higher than any this researcher has seen in any other study. A1though the samp1e was Targe. the resu1ts of the study are questionabTe due tO a Tack of standardiza- tion on definition Of hypertension. treatment goaTs and pTans. B/P measurement. and record keeping. OperationaT definitions were vague if present. with the exception of the compTiance percentage of medica- tions. As noted ear1ier. mean diastoTic pressure was not noted as a measure of compTiance untiT it was cited in the resu1ts section. WhiTe 49 the investigators noted that a Tower percentage of subjects taking medication three to four times a day had diastoTic B/P of 90 mm Hg or Tess. they did not re1ate that this may have been due to the severity of these patients' hypertension. The fact that their B/P's were higher may have been due to severity and not Tack of compTiance. It is not possib1e to make this determination since no data were c011ected re1ated to severity or duration Of hypertension. A smaTTer samp1e with more precise methods woqu probabTy yier more usefuT information. With the exception of the Tast study reviewed (Widmer et a1.. 1983). it appears that investigators in the area of compTiance for hypertensive patients have been foTTowing the suggestion of earTier researchers and c1ear1y and preciseTy OperationaTTy defining both compTiance measures and independent variabTes. The few studies that have appeared in the Titerature since 1980. however. are stiTT not consistent in the way they operationaTize compTiance. and this con- tinues to make comparisons of resu1ts difficuTt. It is tempting and Often necessary to expTore a Targe number of variabTes that may affect compTiance when a Targe samp1e is avaiTabTe; resu1ts. however. are often confusing. ResuTts are Of TittTe va1ue to future investigators who wish to repTicate or expand on the research and of no practicaT va1ue to the practicing hea1th-care provider. The review above focused on Titerature re1ated to foTTowing a therapeutic regimen for hypertension. In the next section of this chapter. Titerature re1ated to the concept Of socia1 support and works 50 that examine the re1ationship between socia1 support and foTTowing a therapeutic regimen wiTT be reviewed. Sbgumdsm::§Ocial_SunnOLt Authors and researchers have demonstrated a great deaT of interest in the concept of socia1 support over the Tast decade. This interest evoTved in the 19705 when severa1 major papers presented discussions and Titerature reviews on socia1 support.(CapTan. 1974; CasseT. 1976; Cobb. 1976; Dean & Lin. T977L These authors found mounting evidence that socia1 support had both a direct positive effect on hea1th status and served as a buffer or modifier Of the effects of stress (psycho- sociaT and physica1) on the menta1 and physica1 hea1th of an indi- viduaT. Investigators have been enthusiastic about socia1 support because it is thought that it may be easier to change the socia1 reTationships of an individua1 than the exposure to stress or the other mediators such as personaTity traits or coping styTe (Thiots. 1982). For this reason. socia1 support is currentTy regarded as a centraT psychosociaT issue in hea1th research. In a review of Titerature on socia1 support. Wortman (1984) re1ated that in the Tast 10 years hundreds of studies and numerous review articTes and books have addressed socia1 support. She noted that the research is increasing at an astounding rate. with more citations on socia1 support in the Tast 2 years than in the entire previous decade. The recent outpouring Of research has caused some investigators to criticaTTy re-examine ear1ier cTaims about socia1 support. As stated 51 ear1ier in this thesis. the greatest weakness of socia1—support research is the Tack of consensus on the definition and operationaTiza- tion of socia1 support. Numerous authors (Barrera & AinTay. T983; Broadhead et a1.. 1983; Dimond & Jones. 1983; E11. 1984; GottTieb. 1981; Henderson. 1984; Wortman. 1984) have cited variabTes as diverse as the presence of Tittermates. the wantedness of pregnancy. invoTve- ment in seTf-heTp groups. and care deTivered by nurses and physicians as evidence of the effectiveness of socia1 support in mitigating stress. Other questionabTy re1ated constructs such as socia1 cTass. job satisfaction. and insufficient financiaT resources have been used by some researchers as operationaT definitions for socia1 support. The research design typicaTTy used to study socia1 support repre- sents another commonTy identified weakness. Most Of the concTusions about socia1 support have been based on corre1ationa1 data c011ected at a singTe time (Wortman. 1984K Three TongitudinaT studies were cited as "adequateTy designed" by Broadhead and associates (1983). The first. by PTess and Satterwhite (1973). was a randomized controTTed cTinicaT triaT of Tay famiTy counseTors and famiTies of chderen with chronic diseases. ResuTts showed improved psychoTogicaT status was greater in chderen in the treatment group (60%) than in the controT group (41%). The second was a randomized controT design to evaTuate the effect of physician-Ted support groups for parents having their first chde. by GottTieb (1981). He was ab1e to document an increase in amount of support received by patients when they were outside their group but coqu not demonstrate a reduction in subjective ratings of 52 stress or an improvement in sense of we11 being. Sosa et a1. (1980) used supportive Tay companions for women during Tabor. aTso a random- ized contrOTTed design in the third study. ContrOTs had higher compTi- cation rates (Caesarean section. meconium staining. eth during Tabor and deTivery. Even comparing onTy uncompTIcated deTiveries. the experimentaT group had markedTy shorter Tabors than controTs (8&3 versus 19.3 hours). were more often awake after deTivery. and stroked. smiTed at. and taTked to their babies more than did controT mothers. A review by Broadhead et a1. (1983) noted that numerous cross- sectionaT. retrospective. and case-controT studies have shown a direct effect of socia1 supports in various hea1th/disease states. Yet Broadhead and associates feTt these studies were fTawed by their inabiTity to address causaTity. ATternative expTanations for the findings such as the confounding of Tife events and socia1 supports. as proposed by Thiots (1982). are highTy pTausibTe. Thiots noted that many important Tife events are aTso interpretabTe as Tosses or gains in socia1-support reTationships (i.eu. death of a spouse. divorce. marriage. famiTy member 1eaving or joining a househonL Another expTanation is that though the variabTes are causaTTy re1ated. oneks prognosis. coping. or prior adjustment infTuences the amount of support avaiTabTe (Wortman. 1984). ‘Those individuaTs with a poor prognosis or numerous hea1th prObTems may make more demands for support and receive it. Those individuaTs with poor coping skiTTs or adjustment may Tack socia1 competence and drive others away with socia11y inappropriate behaviors. 53 Most recentTy. critics have noted a Tack of information regarding the process through which socia1 support may affect hea1th outcomes (Barrera & AinTay. 1983; Dimond & Jones. 1983; PearTin. T982; Thiots. 1982; Wortman. 1984). WhiTe a majority of the avaiTabTe research has documented a re1ationship between socia1 support and hea1th. the question of why or how socia1 support infTuences hea1th has onTy recentTy been addressed. ModeTS for socia1 support and adaptation to stress have been presented by Dimond and Jones (1983). PearTin et a1. (1981). and Thiots (T982). Barrera and AinTay (1983) have deveToped and tested a typoTogy of socia1 support. The above works are an attempt to give a framework from which to answer the question of how and why socia1 support affects hea1th. If there is one area of widespread agreement in the 1iterature. it is the recognition of the compTex and mu1tifaceted nature of socia1 support. There is a great need for more systematic and precise conceptuaTization and measurement of the construct (Wortman. 1984). W HistoricaTTy. the concept Of socia1 support has been vagueTy defined. As House (1984) and Thiots (1982) remarked. many of the earTy investigators in the socia1-support area faiTed to provide any defini- tion of the socia1-support construct. Some of the earTy studies referred to Durkheinfls (1951) notion of socia1 integration. the strength or weakness Of ties binding the individua1 to the group. The investigators then stated their operationaT definition of socia1 54 support (Andrews et a1.. 1978; Myers et a1.. 1975; NuckoTTs et a1.. 1972). Others used definitions termed circuTar or vague by House (1984) and Thiots (1982). In a criticism of the definition of socia1 support by Kap1an and associates (1975). "support is defined by the reTative presence or absence of psychoTogicaT support resources from significant others)‘ ‘Thiots (1982) noted the term "support resources"- is too imprecise to be theoreticaTTy usefuT. In a simiTar vein. House (1984) found the definition of socia1 support by Lin and coTTeagues (T979)--"support accessibTe to an individua1 through socia1 ties to other individuaTs. groups. and the Targer comnuuflty"--as essentiaTTy defining socia1 support as support that is socia1. The importance of how socia1 support is defined shoqu not be underestimated. for the definition of the concept has far-reaching impTications for its measurement. .As House (1984) discussed. an impre- cision in the definition Of socia1 support is Often mirrored in the operationaTization of the construct. As noted earTier. investigators have used operationaT definitions of socia1 support that are wideTy varied. 1.3”) financiaT resources. seTf-esteem. and job satisfaction. Wortman (1984) discussed the even more prob1ematic tendency of some researchers tO use operationaT definitions of support that may overTap with the outcome they are assessing. VariabTes such as adaptabiTity or crying have been used as operationaT definitions of socia1 support. There are authors who have defined socia1 support in a more precise manner. Cobb (1976) defined socia1 support as information 1eading a person to be1ieve that he/she is (a) cared for and Toved. 55 (b) esteemed and va1ued. and (c) beTOngs to a network of mutua1 obTigations. (Cobb's definition was used for this study.) Kahn (1979) defined socia1 support as interpersona1 transactions that invoTve the expression Of positive affect. the affirmation or endorsement of the person's beTiefs or va1ues. and/or the provision Of aid or assistance. WaTker and co—workers (1977) defined socia1 support as behavior that assures peop1e that their feeTings are understood by others and con- sidered norma1 in the situation. WhiTe these definitions may differ in focus. House (1984) and Wortman (1984) emphasized that there is some agreement about what aspects of reTationships faTT within the generaT definition Of socia1 support. Each of these definitions coqu permit a c1ear operationaT definition of socia1 support. W As investigators discovered the compTexity of socia1 support. they attempted to identify its components. Taxonomies or Tists of compo- nents have been continuaTTy deveToped. In a review of socia1-support 1iterature. Dimond and Jones (1983) noted four areas of agreement in the taxonomies of socia1 support. First. socia1 support invoTves the communication of W A second area of agreement centers on social_1nteg£1119n. that is. to have membership in groups and exchange with others. The third component was not agreed upon entireTy; that is the provision of mate:1a1_gn_tangible_nid. Fina11y. reciprocity was noted as necessary for the continuance Of mutuaTTy satisfying and supportive interactions. Citing a simiTar. but sTightTy expanded. version. Wortman (T984) compiTed a Tist of six distinct types 56 Of support based on the works of severa1 authors (Cap1an. T979. 1974; Cobb. 1976; House. 1984; Weiss. 1974). They are as foTTows: (a) expression of positive affect. (b) agreement with acknowTedging the appropriateness of a personhs beTiefs or feeTings. (c) encouraging the open expression Of beTiefs and feeTings. (d) offering advice or infor- mation. (e) provision of materiaT aid. and (f) providing information that the person is part of a network. One empirica1 study of types of socia1 support was conducted by Barrera and AinTay (1983). They deveToped a taxonomy Of socia1 support based on a review Of Titerature and determined six initiaT categories: materiaT aid. behavioraT assistance. intimate interaction. guidance. feedback. and positive socia1 interaction. To test this taxonomy. a 40-item scaTe (the Inventory Of SociaTTy Supportive Behavior) was compTeted by 370 introductory psychoTogy students. Through factor ana1ysis and varimax rotation. four factors were extracted. The first was TabeTTed directive guidance. It featured activities that were inc1uded in the conceptua1 categories of guidance and feedback (941' teaching skiTTs. providing advice. giving feedback). These actions provided support of a practicaT nature and accounted for 76.3% Of the variance in the rotated pattern matrix. The second factor contained activities typicaTTy associated with nondirective counseTing (intimacy. unconditionaT avaiTabiTity. esteem. trust. physica1 affection). This factor was TabeTTed nondirective support. It expTained 11.6% of the variance. and its content corresponded to the theoreticaT category of intimate interaction. The third factor was TabeTTed positive socia1 57 interaction based on the content of three items with the highest Toadings (joking and kidding. taTking about interests. and engaging in diversionary activitiesL There was some overTap of four additionaT items that aTso had high Toadings on Factors I and II. Factor III accounted for 7% Of the variance in the rotated factor matrix. The fourth factor had a11 highTy Toaded items representing tangib1e assistance (physica1 assistance. providing sheTter. money. or physica1 objects of va1ue. This factor accounted for 5% Of the variance. WhiTe this study had Timitations. such as the use Of coTTege- student sampTe and items representing a distinct samp1e of action variabTes. the findings may be used to conceptuaTize socia1 support as behavioraT transactions and organize them into empirica1 categories usefuT in further research. This type of empirica1 testing is cruciaT but not Often found in socia1-support research. PracticaT Timitations may make it impossibTe to investigate many different types of support in a given popu1ation. Therefore. it is important to consider which types Of support may be most important and heTpfuT to that popu1ation. ‘This thesis wiTT provide information about hypertensive patients' perceptions of the type of socia1 support they received. This information can then be used by other researchers in determining the type/types of support to investigate for other hyper- tensive patients. 58 WW As with a11 other aspects of socia1 support. there is no c1ear agreement on how the concept shoqu be measured. In generaT. the measurement Of socia1 support has been approached in four ways.(Hender- son. 1984). In the first instance. coarse sociodemographic variabTes have been used as indices of isoTation or not having an intimate re1a- tionship. These findings are Timited in that they ref1ect onTy indirect1y the socia1 environment of the individua1. A second method has been to determine socia1 support from a few questionnaire items regarding maritaT status. Tiving arrangements. socia1 functioning. and invoTvement in group activities. WhiTe some have praised this approach used by Berkman and Syme (1979) because they consistentTy examine effects of socia1 reTationships on "hard" outcomes such as mortaTity (House. 1984). others have criticized its use because it provides TittTe information about the process through which socia1 ties affect hea1th. There is no evidence to determine if it is socia1 support or some other aspect of the re1ationship that is beneficiaT. The third method invoTves examining specific eTements within a person's socia1 network. UsuaTTy the structure of the network such as size (the number of peopTe with whom there is contact. density (extent to which members are in contact). accessibiTity. stabiTity over time. and reciprocity is examined. WhiTe such measures may be more Objective than measures Of perceived support. House (1984) advised against examining socia1 support exc1usive1y in terms of structuraT variabTes. The existence of a re1ationship provides TittTe information about its 59 nature. quaTity. or content. In some cases socia1 interaction may be stressfu1 and have detrimentaT effects. .Speaking of patients with chronic or serious iTTness. DiMatteo and Hayes (1981) cited severa1 ways socia1 support may have a negative impact. The "famiTy equiTib- rium" may be severeTy strained if one individua1 receives too much attention and support due to iTTness or injury. Because Of the pos- siDTe negative effect on the famiTy. the support person/persons aTso need some kind of support. In working with hypertensive patients. providers need to remember that the patientfls abiTity tO foTTow a therapeutic regimen may aTso be negativeTy infTuenced by socia1 support. If the treatment is accept- abTe to them. famiTy and friends wiTT encourage and assist the patient in foTTowing the regimen. If however. the treatment regimen is con- trary to the va1ues. beTiefs. and usuaT patterns Of conduct in the famiTy. they may sabotage the therapeutic regimen or ignore it; There- fore. both the positive and negative infTuence of the patientfls socia1- support system shoqu be examined. InvoTvement of famiTy and friends in pTanning and impTementation of the regimen can aTTeviate some of the negative effects possib1e with socia1 support. Socia1 support may aTso undermine the patient's seT f-esteem in one Of two ways (DiMatteo & Hays. 1981L. First. patients may fee1 they are a "burden" to their famiTy or friends as a resu1t Of their iTTness. EmotionaT. financiaT. or physica1 support may be viewed as an intrusion on Toved ones"time and energy. Second. acceptance of socia1 support from others requires the patient to recognize his or her status as an 6O impaired person. Some patients may find the costs of discTosing information regarding their iTTness to their support persons to be greater than the benefits. To protect others from their iTTness and maintain seTf-esteem. they withdraw from the support system. This action may aTienate the patient from the support system. straining further interaction and eroding trust. Since there are few definitions of socia1 support that inc1ude possib1e negative effects. few instruments contain items that Opera- tionaTize its possib1e negative infTuence or effects. Yet there is evidence that prob1ems resuTting from socia1 reTationships represent an important share of the stresses peop1e experience in their daiTy Tives (Schaefer et a1.. 1984). For this reason it is as important to iden- tify reTationships that are predominantTy negative and nonsupportive as those that are supportive. The negative aspect Of socia1 support may be especia11y important to evaTuate for hypertensive patients because of the chronic nature of their i11ness. ReTationships that are sup- portive initiaTTy may become nonsupportive over a period of time. Behaviors such as reminding someone to take a piTT or foTTow a diet may be perceived as "nagging" or "pressure" and no Tonger have a positive effect on foTTowing the therapeutic regimen. Identification of these reTationships may aTTow the hypertensive patient to change the char- acter Of the reTationship or find a new re1ationship that is more supportive. Henderson (1984) noted the Tast approach as purpose-buiTt instru- ments such as the PsychOTogicaT Assets Questionnaire (NuckoTTs et a1.. 61 1972). the Interview ScheduTe for Socia1 Interaction (Henderson et a1.. 1981). the PersonaT Resource Questionnaire (Brandt & Weinert. 1981). the Norbeck Socia1 Support Questionnaire (Norbeck et a1.. 1981). and the Socia1 Support Questionnaire (Sarason et a1.. 1983). Most of these instruments have indices that differentiate between instrumentaT (tan- gib1e) and affectionaT support. between intimate and more diffuse reTationships. and between avaiTabiTity and perceived adequacy. The reTiabiTity Of such scaTes has been found to be satisfactory by Hender- son (1984). There are. however. technicaT prob1ems in vaTidation. WhiTe vaTidation Of avaiTabiTity measures can be achieved by Obtaining information from a coTTateraT source. this approach faiTs for adequacy indices which ref1ect an individuaT's perceptions of support. WOW WW EmpiricaT research regarding socia1 support and its effect on foTTowing a therapeutic regimen for hypertension wiTT be reviewed in a simiTar fashion to that on compTiance/adherence and hypertension. Studies conducted before 1980 wi11 be reviewed and comparisons made among the studies regarding operationaT definitions of compTiance/ adherence. operationaT definitions of socia1 support. other independent variabTes studied. and the generaT research design. Next. studies conducted from 1980 through 1984 wiTT be reviewed and comparisons made in the same manner as for research before 1980. Fina11y. a discussion wiTT be presented on whether or not progress has been made in the way 62 in which research on socia1 support and foTTowing a therapeutic regimen for hypertension has been conducted. .Beseangn_pnign_tg_1980. In 1973. McKenney et a1. examined the effect of cTinicaT pharmacy services on patients with essentiaT hypertension. Fifty hypertensive patients receiving care in a Mode1 Neighborhood Hea1th Program were inc1uded in the study. A11 subjects had met the foTTowing criteria: an average diastoTic B/P greater than 90 mnan on three consecutive visits prior to the study. receiving pharmacy services from one Of three pharmacies participating in the study. not bedridden or debiTitated. and no other hea1th prob1ems that made hypertension a secondary concern. 0f the subjects. 11 were maTe and 38 femaTe. The subjects' average age was 60 years; 38 were bTack and 11 white. Adherence/compTiance was operationaTTy defined for taking medica- tion (taking more or Tess than 10% Of prescribed anti-hypertensive medication) and normotension (two or more consecutive monthTy diastoTic B/P readings averaging Tess than 90 mm Hg). Taking medication was confirmed by piTT counts. whiTe B/P readings were obtained from records of physicians at the hea1th centen. KnowTedge for a11 subjects was aTso measured pre- and post-intervention on a 21-item true-faTse test designed to evaTuate the patientfls generaT know1edge of hypertension and its drug and dietary management. The intervention for the experimentaT group (n = 24) invoTved each subject's being seen monthTy by appointment for a period of 5 months by the pharmacist investigator. .At the initiaT visit the pharmacist 63 Obtained a medicaT and pharmaceuticaT history and B/P recording. ques- tioned the patient about drug use and how cTOseTy he/she foTTowed prescribed therapy. and discussed compTaints. reactions. or prob1ems re1ated to hypertension. At subsequent visits the pharmacist investi- gator evaTuated responses to drug and dietary management. identified and managed other prob1ems (14;. adjusted therapy on physician orders and referred for speciaTized care). and provided and evaTuated educa- tiona1 materiaTs. Both experimentaT and controT patients were seen by regu1ar physician providers at the hea1th center. (kxnmerciaT hyperten- sion Titerature was avaiTabTe to aTT cTients and physicians throughout the study. Test resu1ts regarding know1edge found a significant difference in scores between the experimentaT and controT groups post-intervention (E [1.32] = 23.47; p < .001). WhiTe both groups of patients were noncompTiant with prescribed therapy for a 7-month period prior to the study. those in the experimentaT group cTOseTy foTTowed the prescribed regimen during the study period. ControT patients remained noncom- pTiant prior to. during. and after the study period. There was a significant difference between groups re1ated to medication compTiance (x2 = 14.487; p < .001). A record audit reveaTed that after the study period. the experimentaT group returned to the prestudy 1eve1 of com- pTiance. As with medication compTiance. the B/P of most experimentaT patients (79%) dropped to norma1 1eveTs during the interaction with the pharmacist. After the study they rose to prestudy 1eveTs. Most 64 controT patients (80%) remained hypertensive before. during. and after the study period. This difference was statisticaTTy significant (E [1.46] = 21.988; p < .001). An interesting additionaT finding was that the experimentaT patients kept 92% of the 100 scheduTed appoint- ments with the pharmacist. whiTe controT patients kept 82% of the 44 scheduTed physician visits. ResuTts of this study suggest that support from a pharmacist can increase compTiance as measured by taking medica- tion and SIP controT. whiTe patients"1eve1 of know1edge can be increased as a resu1t of an interaction. In their study. McKenney et a1. (1973) documented fairTy c1ear OperationaT definitions of compTiance and socia1 support in the form of the provision of cTinicaT pharmacy services. The know1edge variabTe was not discussed in great detaiT in the articTe which described the study. Intervention methods were we11 defined by the investigators. but TittTe information was given on how piTT counts took pTace and if there were any standards for B/P measurement. ResuTts were easiTy understood and were usefuT for future research or c1inica1 interven- tion. A samp1e of 38 maTe Canadian steeT workers who were neither com- pTiant with medication or a diastoTic B/P goa1 after 6 months of treat- ment were studied by Haynes et a1. (1976). ‘This sampTe was seTected from that of an earTier study (Sackett et a1.. 1975) in which increased convenience of medicaT care and know1edge had faiTed to affect medica- tion compTiance or diastoTic B/P controT. CompTiance with medication was operationaTized in this second study as 80% of medications taken. 65 as measured by piTT counts. urine 1eveTs. and bTOOd 1eveTs. ContrOT Of bTOOd pressure was defined as fifth-phase diastoTic reading Tess than 90 mm Hg. Subjects were assigned to experimentaT (n = 20) and controT (n = 18) groups. The intervention for the experimentaT group was conducted by a high schooT graduate with no hea1th-profession education. Each subject met with the intervenor an average of 10 times over 6 months. At the initiaT visit. subjects were Toaned an aneroid sphygmomano- meter and stethoscope and instructed in their use. They were aTso given daiTy p111 and B/P charts and asked to record fifth-phase B/P and piTTs taken and missed each day. DaiTy habits were identified and when possib1e Tinked to piTT taking. On subsequent visits. at 2-week inter- vaTs. piTT taking and B/P charts were reviewed. If B/P was beTow 90 man Hg or«4 mm HG or more beTow initiaT readings. the patient was praised and received a $4 credit toward ownership of the B/P cuff and stetho- scope. Praise was aTso given for perfect medication compTiance. and reasons for noncompTiance were discussed. At the end of 6 months. patients were examined at home and at work. Examiners were "b1ind" as to which patients were in which group. Home visits consisted Of B/P measurement. urine specimen (unannounced) for drug 1eve1. and an unobtrusive piTT count. Within a few days of the home visit. subjects were examined at the miTT whereeB/P measurement and bTOOd samp1es for drug 1eve1 were taken. Criteria for success of the intervention were set in advance: (a) exceed the previous 6 months' compTiance 1eve1 by 20%. (b) the 66 difference (6 to 12 month) of compTiance had to be statisticaTTy sig- nificant. and (c) compTiance in the experimentaT group had to exceed those of the controT group by 20%. The intervention was successfuT. and a11 three criteria were met. ExperimentaT subjects exceeded 6- month compTiance by more than 20%. and the difference was significant at.p <.OOT. The compTiance rate of éxperimentaT subjects exceeded the controT subjects by more than 20%. and the difference was significant at p < .025. There aTso was a decrease in diastoTic B/P which averaged 5.4 mm Hg for the experimentaT group versus 1.9 mm Hg for the controT group. It shoqu be noted that the investigators feTt a confounding variabTe may have been the amount of "attention" given the experimentaT group during the two studies. A1though the intervention in the second study was not intended as socia1 support per se. it may be equated to socia1 support in that the Tay project coordinator offered praise and materiaT incentives for medication compTiance and improved B/P controT. Attention in the first study by physicians providing routine care for hypertension and project staff providing information regarding hyper- tension and its treatment did not affect an increase in medication compTiance or B/P controT. OperationaT definitions of compTiance were given as noted above but were not detaiTed (i.e.. type of bTOOd or urine test. 1eve1 of test which denoted 80% compTiance. and who did end-of-study examination). The overa11 study design inc1uded the first study (phase I) and the study which was reviewed (phase II). The authors did not eTaborate on 67 when the particuTar intervention strategies in the second study were chosen. It is not c1ear if they were pTanned at the same time as the interventions for the first study or as a resu1t of the outcome of the first study. NeTson et a1. (1978) expTored the re1ationship of 18 hea1th-beTief variabTes with compTiance for hypertension. The samp1e consisted Of 142 hypertensive subjects. 69% femaTes and 31% maT es. BTacks comprised 54% Of the group and whites 46%. Twenty-eight percent Of the patients were 49 years of age or Tess. 41% were between 50 and 64 years. and 31% were 65 years or OTder. CompTiance was operationaTTy defined in three ways: B/P controT (diastoTic average 90 mm HG or Tess). seT f-reported medication taking (doses of medication missed in the past 28 daysO. and appointment keeping (80% or more = high. Tess than 80% = Tow). MedicaT records for the TZ-month study period were audited to Obtain B/P readings and appointment data. Data on medications were c011ected during interview by asking the patient to identify anti-hypertensive medications and state how many doses were missed in the past 28 days. Independent variabTes from the Hea1th BeTief Mode1 were from three categories: core perceptions (controT over hea1th matters. priority Of hea1th in Tife. and perceived severity Of hypertension); modifying factors--genera1 (sociodemographics four items. quaTity of Tife without hypertension. time costs Of doctor visits. anxiety when hypertension first diagnosed. intention to compTy when diagnosed. perceived hypertension symptoms when diagnosed. ever perceived hypertension 68 symptoms. ever perceived side effects of medication. and impact of hypertension on Tife styTe); and modifying factors--cues to action (support given by famiTy and other medications prescribed). These variabTes were measured with from one to six questions during the interview. Five questions asked the patient about support given by the famiTy. The authors did not give the content or an exampTe of these questions. Patients were offered $7 to cover the expense of coming to an interview. Trained interviewers her the interview at the physicianus Office. hospitaT. and for the few who found the first two inconvenient. the patients' homes. Average Tength of the interview was 1-1/4 hours. For the purpose of ana1ysis. data on taking medication were dichotomized into "taking a11 medication" and "missing one or more doses." The authors found a positive correT ati on between seTf-reported medication taking and B/P contrOL. The statisticaT difference was significant (51 = 25%. p = .02) between patients who reported "taking a11 medication" (74%) and patients who reported "missing one or more doses" (49%). No statisticaTTy significant associations were detected among medication taking. appointment keeping. and B/P controT. Five independent variabTes significantTy corre1ated with se1f- reported medication taking (12 < .05). Age. perceived severity of hypertension. and having a medication prescribed for another chronic iTTness corre1ated positiveTy with medication compTiance. whiTe side effects from anti-hypertensive medication and experiencing anxiety when 69 hypertension was diagnosed were negativeTy correTated with medication compTiance. There were aTso five variabTes corre1ated with B/P controT (p < .05). Two were the same that associated with medications: medication prescribed for another chronic iTTness and anxiety at the time of diagnosis. ‘The other three were perception that regimen was effective. TittTe impact on Tife sty1e. and Tess education. OnTy two demographic characteristics were associated with appointment keeping: age and empToyment status. OTder and empToyed patients were more 1ikeTy to keep appointments than younger unempToyed patients. NO re1ationship was noted between socia1 support by famiTy and either Of the compTiance variabTes. A1though a Targe number of independent variabTes were examined. ana1ysis and resu1ts were straightforward and easiTy understood. The Targe number of variabTes did Timit the number of items used to examine each variabTe. and therefore onTy five questions re1ated to socia1 support. 'The dependent variabTes were we11 defined. but there is a question regarding the consistency with which B/P was recorded and how many readings were used to determine the average diastoTic pressure. The design was simiTar to others using an interview and record audit to coTTect data. The finaT pre-1980 study to be reviewed was conducted by CapTan et a1. (1979). UnTike earTier research in which socia1 support was a minor variabTe. this study was designed to test the theory that socia1 support shoqu Operate to increase compTiance in hypertension. To do 70 this. an experimentaT intervention was deveToped in which differences in the amount of objective socia1 support were introduced. The samp1e consisted of 438 hypertensive patients. from five hea1th cTinics. who met the foTTowing criteria: (a) diagnosis of essentiaT hypertension or Tast two B/P prior to study entry 3 165/105. (b) ab1e to compTete a seTf—administered questionnaire at the high schooT reading 1eve1. (c) have medication prescribed for hypertension. (d) absence of psychosis or other disabTing iTTness. (e) voTunteer and consent. and (f) be ambuTatory. ApproximateTy 64% of the subjects were femaTe and 46% maTe. whiTe 52% were bTack and 48% white. The mean age of the samp1e was 54 years. CompTiance was operationaTized both ObjectiveTy and subjectiveTy. The first objective measure. bTOOd pressure controT. was defined as both diastoTic pressure 96 mm Hg and systOTic pressure 160 mm Hg. BTOOd pressure was recorded in the Teft arm. in the sitting position. with a mercury sphygmomanometer before and after each questionnaire administration. Mean systoTic and diastoTic pressures for each visit were caTcuTated and used as the visit B/P score. The second Objective compTiance measure. adherence. was defined as adherence to scheduTed appointments. CTerks at each site kept a record of appointments kept. canceTTed. and missed. Subjective adherence was operationaTized in two ways. First. a pair of "adherence vignettes" in which one member of a pair described an adherent patient and the other described a nonadherent patient were presented. The subjects were asked to indicate the extent to which one 71 or the other vignette described them. An index of adherence behavior was the second measure. Subjects were asked "how Often" they had missed adhering to various aspects Of the regimen. and the Tast time they had forgotten to take their medicine. Objective socia1 support was operationaTized by the three conditions of the intervention: socia1 support. partner. and controT. In the socia1-support condition. socia1 support was provided to the patient by a project nurse. The nurse was to expTain the purpose Of the meetings as educationa1 and heTpfuT. expTain the regimen. provide praise and encouragement. aTTow the patient to discuss concerns. and be warm and friendTy. At the first meeting (T1). the cTient received a bOOkTet of factuaT information with space for specific information on individua1 B/P readings and regimen. ‘Then the nurse set up a time in 2 weeks for the patient to teTephone the nurse and discuss how he/she was doing. The nurse provided support in the forms Tisted above. according to a protocoT. during the conversation. At the end of the meeting. both the patient and nurse compTeted a sheet that inc1uded the ques- tion. "OveraTT. how much nclp and reaT ccnccrn about you and your hea1th has been shown by the nurse at your c1inic or hospitaT?" The rating scaTe ranged from "very TittTe" to "a great deaLfl Other mu1ti- item indices were used throughout the study and had their content changed. but this question corre1ated weTT with the mUTtI-item indices (r = .86 -.90) and was therefore used for ana1ysis. The nurse met with the patient at each visit scheduTed by the norma1 provider and provided support as noted above. At T2 (3 months 72 Tater). each patient was provided with a pTastic piTT box with compart- ments for the 7 days of the week. The nurse aTso set up one more phone caTT appointment; thereafter. phone caTTs were onTy made if a change in regimen required foTTow-up. The partner condition was identicaT to the socia1-support condi- tion with the addition of (a) the nurse asking the patient to bring a partner at T2. (b) a reminder Tetter to bring a partner was sent prior to 12. (c) materiaTs reviewed at T2 on the type of support the partner shoqu provide and rationaTe for providing support; and (d) partners were asked to compTete questionnaires and support sheets at T2 and 13. The nurse foTTowed the same principTes of support in interacting with the partner as with the patient. Subjects in the controT condition did not meet with a nurse. They. as a11 subjects. on arriving at T1 and T3 (10 months after T1) had their B/P taken whiTe seated in the waiting room. The patients were then handed a questionnaire which took about 35 minutes to compTete. After the questionnaire was compTeted. a second B/P was taken. Patients in the controT condition saw their routine provider. Patients in the socia1-support or partner condition met with the project nurse and then saw their routine providen. The experimentaT intervention ended at T4 (12 months after T1). and patients and partners were maiTed question- naires to be compTeted and returned. Subjective socia1 support of the partners was Obtained on the questionnaire from a11 patients in the study and was obtained on the rating sheets from patients in the experimentaT conditions after the 73 meetings of the patient. partner. and nurse from T2 through T3. Items ref1ected perceptions of both tangib1e and psychoTogicaT support. Items re1ating to the variabTes of know1edge Of the regimen. compTexity Of regimen. somatic compTaints. depression. and motivation to compTy were presented in questionnaires administered at T1. T3. and T4. Descriptions of these items are Tengthy and may be found in CapTan et a1. (1979. pp. 53-58). They wiTT not be discussed in this review. AnaTysis showed the onTy statisticaTTy significant finding was to be a positive correTation between socia1 support from a nurse in the socia1-support experimentaT group and subjective (seTf-reported) adher- ence (r = .30; p < .05). There were neither significant changes in mean diastoTic or systOTic bTOOd pressure due to the intervention. nor associations between Objective adherence (appointment keeping). and the intervention. One of the Timitations of the study was the unreTiabiTity of record keeping used for objective compTiance. In many instances the c1inic staff simpTy did not record data regarding appointments. If there had been any significant findings in this area. they woqu have been questionabTe. A second Timitation of the study was the rather high 1eve1 set for B/P controT (systoTic < 166 mm Hg and diastoTic < 96 mm Hg). Investi- gators in the three previousTy reviewed studies and other research regarding hypertension rareTy cite this high a reading unTess it is age adjusted. Again. had there been significant findings they woqu have been questionabTe and difficuTt to compare to other studies. 74 One item was used to ana1yze data for each variabTe regarding amount Of socia1 support received from the nurse. amount of tangib1e support received from the partner. and amount of psychoTogicaT support received from the partner. Even if they were highTy corre1ated with mu1tip1e questions. using one question from each variabTe to document the major concept Of the study is unusuaT. Perhaps the most severe Timitation of this study is the difficuTty in interpreting resu1ts. ‘There were a Targe number of interreTation- ships. severaT Of which were reciprocaT. This confounding restricts mu1tivariate ana1ysis and makes it difficuTt to determine the precise meaning of any association. (These reTationships wi11 not be discussed here but may be examined in CapTan et a1.. 1979.) It is this type Of confounding of variabTes that has made the findings of this and other studies with a Targe number of variabTes difficuTt to interpret without further compTex ana1ysis. In some cases. no amount of ana1ysis wiTT reveaT the reTationships. W. A comparison of the four studies prior to 1980 reveaTs that a11 the researchers used some form of B/P controT and medication taking as operationaT definitions of compTiance. BTOOd pressure controT was operationaTized simiTarTy in three of the four studies. diastoTic < 90 mm Hg (Haynes et a1.. 1976; McKenney et a1.. 1973; NeTson et a1.. 1978). but the number Of record- ings. method of recording. and where the B/P was taken varied. As noted earTier. the B/P controT 1eve1 used in the CapTan (1979) 75 research (systoTic < 166 mnIHg and diastoTic < 96 mm Hg) was not consistent with other 1eveTs cited in the Titerature. Both McKenney and associates (1973) and Haynes and coTTeagues (1976) used piTT counts as definitions of medication compTiance. CompTiance percentages. however. were different. NeTson et a1. (1978) and CapTan et a1. (1979) used seTf-support measures that were simiTar. noting doses of medication not taken. OnTy the study by Haynes et a1. (1976) determined compTiance with urine and bTOOd drug 1eveTs. Appointment keeping was an operationaTized definition of comp1i- ance in the research Of NeTson et a1. (1978) and CapTan et a1. (1979). Both studies recorded appointments kept. The NeTson study (1978) defined compTiance as keeping 80% Of scheduTed appointments. whiTe the 1eve1 necessary to be compTiant was not given by CapTan et a1. (1979L ExperimentaT interventions in the two earTier studies (Haynes et a1.. 1976; McKenney et a1.. 1973) were not viewed as socia1 support by the authors. but rather as a series Of strategies to increase comp1i- ance. The fact that another person was assisting or "supporting" the subject with these strategies was incidentaT. In a Tater study by NeTson et a1. (1978). socia1 support was examined as a cue to action from the Hea1th BeTief Mode1 and was operationaTized by five question- naire items in an interview design. No intervention was invoTved. The research by CapTan and associates (1979) used an experimentaT interven- tion to test a theory re1ated to socia1 support and compTiance. There appeared to be an increasing emphasis in the research on the importance Of socia1 support as a major factor infTuencing compTiance with a 76 regimen for hypertension. UnfortunateTy. the significance of the re1a- tionship was not we11 documented in the CapTan study resu1ts due to confounding of socia1 support and other variabTes. W. Investigators in two empirica1 studies exam- ined socia1 support as an educationa1 intervention aimed at increasing compTiance with a therapeutic regimen for hypertension. In the first study (Kirscht et a1.. 1981). a factoriaT design was empToyed to deTiver four sequentiaT educationa1 interventions about 4 months apart to randomTy seTected subgroups of the samp1e. Subjects were nearTy a11 white. most (60%) had a high schooT education. and most were currentTy empToyed in bTue-COTTar jobs. Median famiTy income was $12.000 per year. The age of most subjects (78%) was over 50 years. Four hundred thirty-two patients from six private practices participated in the study. "High bTOOd pressure" was defined by each physician. and no bTOOd pressure "cut-off" 1eve1 was used for inc1usion in the study. The researchers focused on adherence to three areas of the regi- men: medication. dietary restrictions. and controT of weight. Impact Of the intervention on B/P 1eveTs was not examined. Medication adher- ence. dietary adherence.and weight controT were operationaTized through seTf-report. In an interview. the patient was asked. "Have you been ab1e to foTTow your doctor's advice? For exampTe. how many piTTs did you take yesterday? The Tast time you didn‘t‘take the medication what was the reason? How often does this happen?" SimiTar items were used for other aspects of the regimen. Answers were coded on an ordinaT scaTe from Tow to high. and each patient was assigned an index Of 77 adherence for each eTement of the regimen. In addition. medication adherence was operationaTized as a pharmacy score. a ratio of piTTs avaiTabTe to the patient. To caTcuTate the ratio. the number Of daiTy doses of each prescribed medication was estimated from pharmacy records. This number was divided by the number of days the prescrip- tion was in force. Three interventions were provided in the foTTowing sequence. First a printed message was given to each patient. Three groups received one of three messages-~a threatening message. a positive message. or no message. Second. the experimentaT group received a Tetter and phone caTT to discuss the regimen and reinforce positive behaviors. The controT group received no intervention. Third. two experimentaT groups were taught seTf-monitoring techniques. One group kept daiTy charts on behaviors invoTved in the regimen; the other recorded B/P readings. Both interventions Tasted 2 weeks. The controT group received no intervention. Socia1 support was the fourth of the sequentiaT interventions. It was operationaTized as an intervention in which a nurse made an appointment by phone to meet with patients assigned to the experimentaT group and a person seTected by the patients to he1p them with prob1ems in foTTowing the regimen. ‘The nurse visited the subject and support person. discussed the purpose of the intervention. Obtained a commit- ment that the pair woqu work on particuTar aspects Of the regimen. and staged a roTe-pTaying exampTe. Two weeks Tater. the nurse caTTed both the patient and support person to give reinforcement and answer 78 questions. ControT subjects received a Tetter thanking them for their continued participation but were not otherwise contacted. For the whoTe study. 36 groups were uTtimateTy formed (3 x 2 x 3 x 2% The sequence of interventions was the same for aTT patients. whether or not they received one or more experimentaT treatments. ‘The progression Of interventions was from impersonaT to personaT and corresponded to an increase in patient invoTvement. The first intervention--printed materia1--did not significantTy increase compTiance. The controT subjects had higher scores for diet that were significantTy different from the experimentaT group (2'= .837; c < .05). The second intervention--nurse phone caTTs--increased medication compTiance (p, < .05). and the third intervention-~se1f- monitoring of medication and B/P--Ted to better weight controT. The resu1ts documented the socia1-support intervention signifi- cantTy increased medication adherence (p'<.OS) on both the pharmacy score and seTf-report medication score. There was aTso a tendency for the intervention to have a positive effect on weight controT. Dietary- adherence scores were not affected by socia1 support. There was no cumuTative impact of the interventions when tested by factoriaT ana1ysis of variance. Different aspects of the regimens were not significantTy reTated to one another. The authors feTt these resu1ts were consistent with the concTusion that adherence is a mu1ti- dimensionaT concept which requires different interventions for differ- ent components of the therapeutic regimen. 79 OperationaTization for both compTiance and socia1 support was c1ear1y defined. A1though the design of the study was somewhat compTex with 36 subgroups. the data ana1ysis and resu1ts were presented in a manner that was easiTy understood. ResuTts denoting different types Of interventions that affect different aspects of the regimen are usefuT to both researchers and hea1th-care practitioners. and this researcher coqu find no major Timitations in the study. There was no definition of hypertension for entry into the study. but since B/P controT was not an outcome variabTe this created no prob1ems. The authors Tisted two Timitations in their summary: (a) repeated data coTTection through patient interviews may constitute potentiaT interventions to an unknown degree. and (b) the interventions a11 invoTved peop1e outside the group of the patient's norma1 hea1th-care providers. In the second of the studies using an educationa1 intervention. Morisky et a1. (1982) used a factoriaT design to examine educationa1 approaches toward increasing adherence and patient activation in hypertensive patients. The samp1e of 200 bTack. inner-city hyperten- sive patients was a subgroup from an ear1ier study (Levine et a1" 1979). Subjects in the samp1e of this study were equaTTy divided between the experimentaT and controT groups. FemaTes comprised 70% of the samp1e. which had a median age of 54 years. a median income of $4.250. and a median Of 7 years of education. CompTiance was operationaTized by two methods. medication taking and B/P controT. Medication-taking compTiance was operationaTTy defined as a seTf-report measure based on patients' nesponses to four 80 questions about their usuaT patterns of medication taking. An age- adjusted set of criteria defined B/P controT: patients under age 40. < 140/90 mm Hg; patients age 40 to 59. < 150/95 mm Hg; and patients aged 60 years or OTder. < T60/100 mm Hg. A record of systoTic and diastoTic B/P readings was kept throughout the study period and then averaged for ana1ysis. Socia1 support was operationaTTy defined for three sequentiaT interventions. The first intervention was an exit interview conducted by a hea1th-education student immediateTy after the cTinic visit. During the 5- to TO-minute session. the regimen was reviewed and instruction reinforced. prob1ems were discussed. and an attempt was made to taiTor the regimen to the patient's daiTy routine. The second intervention invoTved two home visits: one with the patient regarding know1edge. attitudes. and behaviors concerning hypertension; and a second at the same time or within severa1 days. was with a patient- chosen significant other. The Objectives Of the visit with the identified support person were to (a) identify ways famiTy members can assist the patient in hypertension treatment and reinforce seTf-care behaviors and (b) have the support person make a commitment to he1p the patient remember to take medications and keep appointments. Nurses made the home visits. The third intervention consisted of three weekTy 2-hour sessions of smaTT groups co—Ted by a socia1 worker and a nurse. Three main steps characterized the sessions: (a) a generaT discussion of hea1th behaviors and compTiance issues. (b) teaching and inducing rehearsaT of 81 coping skiTTs and prob1em-soTving activity. and (c) encouraging patients to practice and appTy newTy acquired skiTTs to stressfu1 situations. The focus of these steps was to increase the patients! sense of inter- naT controT over stressfu1 situations. Issues raised by the subjects frequentTy centered on grief and Toss. Four other variabTes were examined in this study. KnowTedge was operationaTized as 10 items regarding symptoms. causes. effects. and treatment of hypertension. BeTief in seriousness was operationaTized as four items re1ated to a patient's be1ief that hypertension can cause bodiTy harm or target-organ damage. BeTief in benefits was operation- aTized as three items about effects of drug therapy on controTTing B/P. Fina11y. Tocus of controT was operationaTized by six items from a scaTe deveToped by WaTTston and WaTTston to examine change in the expectancy dimension of adherence. The interview to measure these variabTes and compTiance was administered to patients at the beginning Of the famiTy- support intervention and again in the finaT foTTow-up interview. 3 to 6 months foTTowing the compTetion of the smaTT-group sessions. For purposes of ana1ysis. the authors divided the study groups into two categories: those assigned to famiTy-support interventions but not smaTT-group sessions (Group I) and those assigned to both the famiTy-support interventions and smaTT-group sessions (Group II). Significant differences were found between the groups on finaT B/P controT (x2 = 4.22; n < .05) with 62% (n = 87) of the members in Group II demonstrating controT versus onTy 46% (n = 86) of those in Group I. None of the attitudinaT or the medication-compTiance variabTes 82 demonstrated statisticaTTy significant differences at thei.05 1eve1 between the two study groups. ResuTts of a 3-year foTTow-up study of the same popu1ation supported no decay factor with respect to B/P controT. It was noted that the most significant main effect for sustained B/P controT was the famiTy-support intervention. ‘This finding reinforces documentation in the socia1 science Titerature of the importance of budeing in support of new beTiefs and behaviors within an individuaT's naturaT environment. The operationaT definition for B/P controT was the onTy age- adjusted definition among the studies reviewed. ‘This age-re1ated adjustment increased the vaTidity of the study findings. Both the exit intervention and famiTy-support interventions were cTearTy defined. The third intervention. smaTT-group sessions. was outTined but the exact process changed from group to group depending on issues raised by group members. Since patients inc1uded in the smaTT-group sessions and famiTy support had significantTy better B/P controT. it woqu be inter- esting to examine what happened in the smaTT-group sessions that made a difference. This type Of ana1ysis was not possib1e with the data c011ected from this study. In the Tast Of the studies to be reviewed. the effectiveness of two socia1-support strategies designed to Tower patients' bTOOd pressure was examined by Earp et a1. (1982). Patients were compared to each other and to members of a controT group receiving routine care in a randomized cTinicaT triaT extending over a period of 2 years. The samp1e (n = 218) was predominantTy bTack (77%). married (60%). and 83 femaTe (59%). OnTy 28% of the subjects were high schooT graduates. Participants resided in urban. semi-ruraT. and ruraT areas. DiastoTic B/P controT was the dependent variabTe for compTiance. with < 95 mm Hg used to operationaTTy define B/P controT. The entry B/P was an average Of a11 readings from 811 c1inic visits patients made in the year prior to and incTUding the first visit in the study. BTOOd pressure for the end of the first year was the average of B/Ps recorded at a11 visits between 7 and 12 months; end-Of-the-second-year pressure represented the average of readings charted between T9 and 24 months after entry into the study. Socia1 support was operationaTized by means of two experimentaT interventions. Patients were randomTy assigned to one of three groups. Group 3 (n,= 63) inc1uded patients who received routine medicaT care at a university or community hospitaT c1inic and served as a controT group. Patients in Group 2 (n = 56) received standard medi- caT care pTus home visits to motivate and reinforce positive hea1th behavior. The home visits were over an T8-month period and were car- ried out by either pubTic-heaTth nurses or speciaTTy trained pharma- cists. The totaT number. frequency. duration. and content of each visit were at the discretion of the hea1th professionaT. Group 1 (n = 99) patients received standard medicaT care. home visits. and chose a "significant other" to activeTy participate in both the home visits and home bTOOd pressure monitoring. A four-item index measuring the amount of socia1 support received from famiTy and friends for continued medi- cation and appointment compTiance was aTso used. 84 Two other variabTes were expTored. A three-item index measuring perceived worry over possib1e impact of hypertension was inc1uded with the four-item sociaT-support index. Questions about barriers assessed the affordabiTity and accessibiTity of care. Findings at the end of the first year Of the study showed no group had a statisticaTTy significant advantage in reducing diastoTic B/P. A11 three groups experienced a dec1ine in the number of patients with uncontroTTed B/P. At the end Of the second year. however. Group 3 (controT) had regressed somewhat whi1e both experimentaT groups demon- strated continued improvement in B/P controT. This trend was not statisticaTTy significant (p = .07). The investigators reported that whiTe each of the two supportive interventions seemed cTearTy superior to the controT group. over time they were unabTe to determine which Of the two was more effective. The operationaT definition for compTiance in this study was simpTe. invoTving one outcome variabTe. diastoTic B/P controT. The OperationaT definition for socia1 support was more vague. Main Objec- tives for the interventions were documented. but there was no consis- tency or standardization of how the home visits were conducted. Each nurse or pharmacist determined the content. format. frequency. and duration of the visits with the patient or patient and significant other. The Tack of consistency severeTy Timits others' abiTity to repTicate this study. There aTso was no type of ana1ysis of possib1e differences in resu1ts based on the type Of hea1th professionaT who carried the home visits. For both future research and use in c1inica1 85 practice. it woqu be heTprT to know if the nurse and pharmacist were equaTTy effective in the home visits or if one type of provider offered an advantage in assisting patients and significant others in B/P con- troT. .Qcmpcr1scn_cI_stnd1c§_128Q_tc_prcscnt. In comparing the dependent variabTes used to measure compTiance. one group of investigators used both B/P controT and medication taking (Morisky et a1. 1982); one expTored medication taking. dietary restrictions. and weight controT (Kirscht et a1.. 1981); and one examined onTy B/P controT (Earp et a1.. 1982). Kirscht and Earp both noted that their decision to use onTy adherence to the regimen or B/P controT was based on resu1ts of earTier research which documented a significant re1ationship between B/P con- troT and adherence measures (Haynes et a1.. 1976; Levine et a1.. 1979; McKenney et a1.. 1973; NeTson et a1.. 1978). Morisky and associates used data from the Levine et a1. (1979) study. Adherence to the regimen was operationaTized by seTf-report meas- ures by both Morisky et a1. (1982) and Kirscht et a1. (1981). Research by Kirscht and coTTeagues examined three aspects of the therapeutic regimen. whiTe medication taking was the onTy variabTe expTored by Morisky and associates. The operationaTization Of B/P controT was distinctTy different in the two studies that examined it. Morisky et a1. empToyed an age-adjusted mean systoTic and diastoTic B/P reading. The research by Earp et a1. (1982) defined a mean diastoTic B/P < 95 mm Hg as a 1eve1 of controT. As noted previousTy. the 86 age-adjusted controT 1eve1 is unique in the hypertension and compTiance Titerature reviewed for this thesis. Socia1-support variabTes were c1ear1y defined as such in the Titerature since 1980. In two studies (Earp et aL. 1982; Morisky et a1.. 1982). sociaT support was provided through a number Of sequentiaT support interventions. whiTe in the third study (Kirscht et 81“» 1981) socia1 support was the Tast of four sequentiaT interventions. Each Of the studies had as one of the interventions a home visit with the hypertensive patient and a support person Of the patientfis choice. The focus of the home visit varied sTightTy but basicaTTy invoTved an expTanation Of the concept of socia1 support. provision of information regarding the patientfls regimen. and enTisting the aid Of the support person in assisting the patient to foTTow the therapeutic regimen. WhiTe the project staff in the Kirscht and Morisky studies each made one home visit. those invoTved in the Earp research made from T to 12 visits over 18 months for an average of 5.5 visits per subject. AdditionaT support interventions by Morisky et a1. (1982) and Earp et a1. (1982) were in the form of patient-onTy interventions. Patients in the Earp study were visited in the home. whereas those in the Morisky study were seen after the first c1inic visit. SmaTT-group sessions of hypertensive subjects were the third intervention of the Morisky research. The type of hea1th-care professionaT used in the support interventions varied between the studies: Kirscht empToyed nurses; Morisky used hea1th educators. nurSes. and socia1 workers; and Earp used nurses and pharmacists. 87 A11 of the investigators used a factoriaT experimentaT design for their research. The number of independent variabTes. other than socia1 support. was much smaTTer than noted in ear1ier studies. ResuTts of the studies were at times difficuTt to interpret because the effects of more than one sociaT-support intervention coqu not be separated and examined aTone. Summary A review of the Titerature on socia1 support and foTTowing a therapeutic regimen for hypertension has shown changes taking pTace in the manner in which research is being conducted. First. socia1 support is being identified as a major concept and studies are addressing it as a primary variabTe. In a recent study (Earp et a1". 1982). the entire experimentaT intervention revoTved around socia1 support. In the past. the trend was to examine socia1 support as one of a Targe number of variabTes re1ated to compTiance with a therapeutic regimen. A second trend in research has been to use seTf-report measures as OperationaT definitions Of compTiance. This is a resu1t of practicaT necessity when dea1ing with a Targe number of subjects and has been documented in the Titerature (NeTson et a1.. 1978) to be positiveTy corre1ated with B/P control. SeTf-report measures can be especia11y usefuT in examining areas Of the regimen other than medication (iJa. diet and exercise) and have been found to be reTiabTe when compared to Objective measures of compTiance (Gierszewski. 1983; King & Fredrick- son.1984). 88 As noted ear1ier. investigators appear to be Timiting the number of independent variabTes examined aTong with socia1 support. In 1978. NeTson examined socia1 support as one of T8 variabTes re1ated to compTiance with a therapeutic regimen for hypertension. In recent studies in which socia1 support has been a major variabTe (Earp et a1.. 1982; Kirscht et BL» 1981; Morisky et ah» 1982). fewer variabTes (one to four) were examined. Fina11y. the factoriaT design has become popu1ar with investi- gators examining the effects of socia1 support through experimentaT interventions. This design has the advantage of being abTe to admin- ister mu1tip1e interventions to subgroups Of a sampTe. The findings are then examined for differences in effectiveness of the interven- tions. UnfortunateTy. this change in the research has become a Timita- tton in one recent study (Earp et a1.. 1982). The investigators used sequentiaT sociaT-support interventions and at the end of the study were unabTe to determine which Of the interventions were responsibTe for changes in compTiance measures. This was not the case when socia1 support was onTy one of four interventions examined (Kirscht et a1" 1981). With a11 the progress made. OTd prob1ems stiTT remain in research regarding socia1 support and foTTowing a therapeutic regimen for hyper- tension. OperationaT definitions of compTiance/adherence for hyperten- sion have become more consistent. whi1e operationaT definitions of socia1 support remain wideTy varied. ‘The Tack of consistency in opera- tionaTizing socia1 support is the resu1t of a stiTT—inconsistent 89 conceptua1 definition. ‘There is more agreement among schoTars and researchers now than 10 years ago about the definition of socia1 sup- port. but the agreement is far Tess than totaT. This fact is ref1ected in the research. There continues to be an overTapping or confounding of socia1 support with other variabTes that may affect foTTowing a therapeutic regimen. This prob1em has improved as fewer variabTes are studied simuTtaneousTy with socia1 support. but extraneous variabTes remain a prObTem. The area in which the 1east amount Of progress has been made is increasing information regarding the process of socia1 support. Researchers are continuing to document that socia1 support affects adherence/compTiance in hypertension. Yet no research reviewed for this thesis has given any information about how or why the effect occurred. A cTearer understanding Of the process Of socia1 support wiTT Tead to more agreement about its conceptua1 definition and more consistency in its operationaTization. In an effort to add to the know1edge regarding socia1 support and foTTowing a therapeutic regimen. data regarding supportive person. type of support. and quantity Of support wiTT be examined in this thesis. CHAPTER 4 METHODOLOGY AND PROCEDURE 9111119.! This study was designed to identify and describe hypertensive patients' perceptions Of socia1 support in foTTowing a therapeutic regimen. More specificaTTy. the focus of the study was the identifica- tion and descriptions of socia1 support among seTected subgroups Of hypertensive patients as weTT as the identification and description of socia1 support for each modaTity of the therapeutic regimen (TJL. medication. diet. and exercise). Differences among subgroups and modaTities were aTso examined. Data for this study were c011ected in 1980-1981 as part of a federaTTy funded research project. "Patient Contributions to Care--Link to Process and Outcome." grant #5R01NU00662. Co-principaT investi- gators were B. Given and C. W. Given. The agency granting funding was the PubTic Hea1th Service. Division of Nursing. For this reason. the description Of sites. samp1e and popu1ation characteristics. and data- coTTection procedures wi11 refer to those used in the research project. The study design for the research project was a controTTed fie1d experiment in which the effects Of a 6-month nursing intervention on cTinicaT parameters and other indicators of management and controT Of hypertension were expTored. .A more detaiTed discussion Of the 90 91 methOdOTogy and procedures used for this thesis is aTso presented in this chapter. EDll§§119n_511§§ The study. "Patient Contributions to Care--Link to Process and Outcome." was conducted in four sites Tocated in midwestern urban areas. ‘Three of the sites were ambuTatory care centers staffed by famiTy practice residents. These famiTy practice residents were comparabTe in terms of their training 1eve1 (18 and 24 first. second. and third year) and in the range Of patient visits handTed per month (900 to 1.200 patients). The distributions Of the number of visits for the top 20 diseases/prob1ems were virtuaTTy identicaT. with hyperten- sion among the top 10 most frequentTy occurring prob1ems in aTT sites. The fourth site consisted of two private Offices staffed by four generaT internists. These physicians were a11 board-eTigibTe or board- certified in generaT internaT medicine. Information about the number Of patient visits per month and distribution of visits based on disease/prob1em was not Obtained from this site. ‘The popu1ations of the three cities in which the famiTy practice programs were Tocated ranged from 92.000 to 198.000. Two of the three cities were TightTy to moderateTy industriaTized. whereas the third was heaviTy industriaT- ized. ‘The city where the private practices were Tocated had a popu1a- tion of approximateTy 85.500 and was TightTy industriaTized and moder- ateT y com merci a1. 92 mm A popu1ation of hypertensive patients was identified at these four sites by two methods. In the three famiTy practice centers. data contained in a computerized hea1th information system were empToyed. In the private practices. patient Tists were drawn up by the four physicians (in two practices) who participated in the study. PopuTation criteria specified that the patients had to (a) be between the ages of T8 and 65; (b) have an estabTished diagnosis of essentiaT hypertension; (c) be Titerate; (d) show no evidence Of can- cer. end-stage rena1 disease. stroke. bTindness. or psychosis; (e) show no evidence of active pregnancy or Tactation; and (f) be on a pre- scribed dietary and/or medication regimen for hypertension. Four hundred thirty-three patient records were originaTTy screened at a11 sites for hypertension. Of these. 177 were exc1uded during a second screening by the co-principaT investigators. The majority of these 177 patients. 124. were exc1uded because their bTOOd pressure fe11 beTOw the Timits set to define hypertension as out of controT: a systoTic BP of 140 mm Hg or a diastoTic B/P of 95 mm Hg. Saans SampTe seTection occurred in two stages. In the first stage. trained auditors (graduate students in the FamiTy Nurse CTinicaT SpeciaTist program at Michigan State University) screened the medicaT records of the patient popu1ation and abstracted data from the records Of those patients who met the criteria for inc1usion in the study: two bTOOd pressure readings taken on two separate occasions indicating a 93 diastoTic pressure Of 90 mnan or above. name of medication and dosage and/or type of diet prescribed. and two weights taken on two separate occasions. In the second stage. the principaT investigators screened the data abstracted on each patient to determine eTigibiTity for inc1usion in the finaT samp1e. An additionaT criterion for inc1usion in the finaT samp1e was that the subjects'inedicaT records indicate two bTOOd pressure readings taken on separate occasions with cjthcr a systoTic pressure of 140 mm Hg or above 9r a diastoTic pressure of 95 mm Hg or above. Two hundred fifty-six hypertensive patients who met the criteria for inc1usion in the study were Sent Tetters requesting their partici- pation in the research project. One hundred fifty-eight patients consented to participate and be interviewed. Of this number. 102 hypertensive patients continued in the study and compTeted the portions of the Socia1 Interaction Questionnaire examined in this thesis. HumaLBightijmtsctjm The rights of the respondents were protected through adherence to standard criteria set forth by the Michigan State University Committee on Research InvoTving Human Subjects. A11 patients were sent a Tetter before being contacted by an interviewer. (See Appendix AJ The Tetter. signed by either the medicaT director Of the hea1th—care center where the patient sought care or by the patient's private physician. 94 described the study and its benefits. assured the patient of anonymity and confidentiaTity. and requested his or her participation in the study. An interviewer initiated teTephone contact with patients who returned a postcard indicating a wiTTingness to participate in the study. patients who requested more information about the study. and patients who did not return a postcard. During the teTephone conversa- tion. the study was again described to the potentiaT subject. questions were answered. and if the person indicated a wiTTingness to partici- pate. an appointment time was arranged to meet with the interviewer at the site. At the time the interviewer met with a patient. she or he described the study and ton the subjects they had a right to refuse to participate. that refusaT to participate woqu in no way affect their hea1th care. and that they coqu withdraw at any time during the study. The patient was then asked to sign a consent form before proceeding with the interview. The consent form provided an expTanation of the research. the purpose Of the study. use Of resu1ts. and assurances of anonymity and confidentiaTity. (See Appendix BJ WWW Upon obtaining the patientfls agreement to participate in the study. the interviewer expTained the content Of the five seTf- administered questionnaires. The individua1 was pTaced aTOne in a room by an interviewer to compTete the five seTf-administered question- naires. Inc1uded among these forms was the sociodemographic 95 questionnaire from which data for this thesis were drawn. The sub- ject's progress was checked periodicaTTy by the interviewer as he or she compTeted the questionnaires. Upon compTetion Of the se1f- administered forms. the interviewer c011ected the instruments and reviewed them for omissions. Next the interviewer administered two finaT questionnaires. one re1ated to the patient's current therapeutic regimen and the other to current symptomOTogy. The instruments were then returned to the research staff for coding. After compTetion Of the questionnaire. subjects were assigned to the experimentaT or controT groups (see method beTOw). Subjects in the experimentaT group were asked to meet eight times over the foTTowing 6 months with a nurse intervener. The interviewers gave the patients the name of the nurse intervener who woqu be contacting them and asked the best time Of day for the nurse intervener to caTT. Subjects assigned to the controT group were thanked for their assistance and ton they woqu be contacted again in 6 months for another interview. (See Appendix C--Intake Sociodemographic InstrumentJ MW At the concTusion of the intake interview. interviewers assigned patients to the experimentaT or controT group using the foTTowing aTgorithm for randomization. It was anticipated that no more than 50 patients woqu be seTected per site for the study. so an array of 54 numbers was randomTy ordered. Assignment was achieved by consuTting a tabTe of random numbers from T to 100 and randomTy seTecting 54 96 numbers. Once the random array was compTete. the numbers were assigned to experimentaT and controT group patients on the basis of a two~thirdl one-third spTit favoring assignment to the experimentaT condition. Since three persons were empToyed as interviewers at each site. the 54 numbers were divided into three groups: 1 through 20 to the first interviewer. 21 through 40 to the second. and 41 through 54 to the third. Subjects in the experimentaT group were asked to meet with a nurse intervener eight times over a 6-month period. The goa1 of the inter- vention was that the nurse together with the patient identify a pTan and specific strategies to invoTve the patient in carrying out the hea1th behaviors needed to foTTow the therapeutic regimen. Progress was measured by the responsibiTity that patients took for impTementing strategies designed to address their priority hea1th prob1ems. This process recognized each patientfls unique prob1ems in attempting to achieve the behaviors that Tead to better controT and management of their hypertension. A11 materiaTs for the intervention phase were standardized and systemized to faciTitate use by nurse interveners at each site. as weTT as to ensure patient invoTvement. Extensive documentation by the nurse took pTace at each intervention visit. The detaiTed recording enabTed the staff to ana1yze the context and focus of each visit and ensured consistency among the nurse interveners. Assignment to the controT group did not invoTve participation in a specific nursing intervention. These subjects received standard hea1th 97 care by their usuaT provider. Providers for both the experimentaT and controT patients were asked to compTete a form with information regard- ing hea1th status and current therapeutic regimen at each visit to the provider for the duration of the intervention period. 1W Mom-Up FoTTowing the intervention phase of the project. 811 subjects were asked to compTete the same five seTf-administered and two interviewer- administered questionnaires. This interview occurred approximateTy 6 months after the intake interview for the controT group of patients. The experimentaT group compTeted the interview upon compTetion of the nursing intervention. The time was generaTTy sTightTy more than 6 months after the intake interview. Most of the patients had difficuTty compTeting eight visits in 6 months due to scheduTe confTicts. hoTi- days. or iTTness. The foTTow-up interview was conducted 3 months after the post- intervention interview (approximateTy 9 months after the intake interview). ‘This finaT interview was done by maiT. A11 subjects were again asked to compTete the same five seTf-administered questionnaires as we11 as seTf-report versions Of the two questionnaires previousTy administered by the interviewer. In addition. the subjects were aTso asked to compTete a newTy constructed questionnaire on socia1 interac- tion. (See Appendix D.) The Socia1 Interaction Questionnaire was designed after the research staff noted a good deaT of documentation re1ated to sociaT support in the nurse intervenor records. A decision 98 was made to more thoroughTy expTore the concept of socia1 support among study subjects. CompTeted questionnaires were returned to the research staff in stamped. pro-addressed enveTopes. Patients who did not return the questionnaire in 3 weeks were contacted by teTephone for foTTow-up. We With the exception of one site. none of the interviewers were hea1th-care providers. In the one exception. three graduate students from the nursing program at Michigan State University interviewed patients. (This researcher was one of the interviewers.) The non-hea1th-care-provider interviewers were recruited by per- sonneT at the centers and were interviewed by the research staff before being hired. The interviewers received 2 days of training. which inc1uded an overview of the research. ethics of interviewing. and the responsibiTities and techniques of interviewing. Interviewing skiTTs were sharpened through roTe p1ay with simuTated patients. Each interviewer was assigned a Tist Of patients to contact and was responsibTe for an accounting of each patient inc1uded on her or his Tist. InitiaTTy the interviewer met with the subject at the site and expTained the nature and purpose of the study. Interviewers were responsibTe for obtaining written consent to participate or. if the subject chose not to participate. to record the reason. They then recorded patient assignment. expTained future participation. and pro- vided the nurse intervener with a Tist of patients to contact. The interviewers were aTso responsibTe for contacting a11 subjects to be interviewed at the 6-month or post-intervention period. Again 99 they conducted the interviews as previousTy detaiTed. recorded data pertinent to questionnaires compTeted. prob1ems encountered. and returned the instruments to the research staff. Fier supervisors for each site debriefed the interviewers and spot checked their work on a weekTy basis. ‘This heTped to faciTitate the interview process and to ensure qua1ity and consistency of their work. WW Information to answer the research question "What are hypertensive patients' perceptions of socia1 support in foTTowing a therapeutic regimen?" was c011ected in the foTTow-up phase Of the experiment (approximateTy 9 months after intake and 3 months post-intervention). The instrument used was the SociaT Interaction Questionnaire. The Socia1 Interaction Questionnaire was deveToped by the research staff of the project "Patient Contributions to Care--Link to Process and Outcome)‘ The questionnaire was structured so the patient woqu focus on one moda1ity of the therapeutic regimen at a time as he or she answered questions regarding socia1 support. The format (both Open- and cTOsed-ended questions) and the content (who gives support. type Of support provided. and how much support provided) were based on a review of the Titerature and the staff's previous experience with the concept of socia1 support. The words "socia1 support" were avoided on the questionnaire as it was discovered during the nursing intervention that many patients were unfamiTiar or uncomfortabTe with the terminoTogy. 100 Patients' perceptions of socia1 support were operationaTized into three sets of variabTes: (a) who did or did not provide support. (b) what type Of support was provided. and (c) how much support was provided. TO determine whom patients perceived as abTe to support them. they were asked to Tist the first name and re1ationship of peop1e who coqu he1p them take on new habits or soTve prob1ems in foTTowing their treatment pTan. (See Appendix D. Question TA (PTease note that a11 question numbers refer to Appendix D. the Socia1 Interaction Question- naire. untiT further notedJ Beginning with the modaTity Of medica- tion. patients were asked if any peop1e they Tisted had been ab1e to support them in foTTowing the moda1ity of the therapeutic regimen (Question 2). Those answering "yes" were asked to identify who had been supportive and who had been mcst supportive (Questions 2a and 2b) in foTTowing the modaTity Of the therapeutic regimen. An identicaT format was used for the moda1ities Of diet and exercise. The identification of peop1e abTe to be supportive was taken from the Tist in Question 1. The "yes" or "no" responses to anyone ab1e to he1p for diet and exercise were Questions 4 and 6. respec— tiveTy. Questions regarding who had been supportive and who had been mcc: supportive were 4a and 4b for diet and 6a and 6b for exercise. Patients were aTso asked if they thought a nurse coqu support them in foTTowing a therapeutic regimen (Question 16L The operationaTization of hypertensive patients' perceptions of type of socia1 support was based on CapTan%s(1979) definitions for subjective tangib1e and subjective psychoTogicaT support. Both types 101 Of support are anaTogous to their objective counterparts but are deter- mined by the perception that the supportive condition exists. TangibTe support is "behavior directed toward providing the person with tangib1e resources that are hypothesized to benefit his/her menta1 or physica1 we11 being" (Cap1an. 179. p. 85). PsychoTogicaT support is "behavior directed toward providing the person with cognitions (va1ues. atti— tudes. beTiefs. and perceptions) and toward inducing affective states that are hypothesized to promote we11 being" (Cap1an. 179. p. 85). In this instance. patients were asked to describe how the most supportive person had heTped or been supportive in foTTowing the therapeutic regimen in each of the modaTities. ‘The description of support given for taking medication was Question 2c. for diet 4c. and for exercise 6c. Those who responded "yes" to Question 16 (coqu a nurse offer support) were asked to identify how a nurse coqu be supportive (16a). No respondents were asked why they did not think a nurse woqu be supportive (TObL This group of questions (16. 16a. 16b) was inc1uded to identify specificaTTy patients' perceptions Of the roTe Of the nurse in providing socia1 support in foTTowing a therapeu- tic regimen. To measure the Tast of the three variabTes of sociaT support--How much support was provided?--patients were asked to answer 17 cTosed- ended questions. ETeven of these addressed positive support or concern for foTTowing a moda1ity Of the therapeutic regimen (Questions 2d. 2e. 2f--medication; 4d. 49. 4f. 4g--diet; and 6d. 66. 6f. 69--exercise). An exampTe is given beTOw: 102 2d. How much does this person he1p you remember to take your piTTs? (CHECK ONE) A great deaT A TittTe ______A Tot ______Not at a11 Some Questions 11 through 13 on the Socia1 Interaction Questionnaire attempted to quantify possib1e negative support the patient perceived as "pressure" to foTTow each of the three modaTities (medication. diet. and exercise). A third group of questions (8. 9. and TO) attempted to quantify the patients' perceptions of support in more generaT cate- gories. such as seTf-worth. concern for hypertension controT. and se1f- reTiance in hypertension controT. MW Patients' responses to questions regarding who was or was not supportive (Questions 1. 2a. 2b. 4a. 4b. 6a. 6b. and 14a) were categorized as: spouse. chde. other reTative. friend/workmate. physician. nurse. no one/myseTf. none. don’t need he1p. counseTor/ psychoTogist/sociaT worker. or no answer; ‘These categories were coded with numbers 01-12. with 99 representing no answer. Perceptions of how a support person was heTpfuT (Questions 2c. 4c. 6c. and 14b) were categorized as foTTows: 1. TangibTe support (active or task oriented). e.g.. pTaces piTT on pTate. fixes Tow-sodium foods. waTks with me 103 2. PsychoTogicaT support-~e4}. encourages me when IHn down. taTks with me if I have a prob1em. expTains importance of taking medication 3. Both tangib1e and psychoTogicaT support TangibTe support originaTTy appeared as 2 in the code book. Psy- choTogicaT support is a combination of categories originaTTy coded 1 (education). 3 (verbaT support). and 6 (education and verbaT supportL Both tangib1e and psychoTogicaT support were originaTTy coded 4 (verbaT and tangib1e) and 5 (education. verbaT. and tangib1e). Perceptions of how much support was provided (Questions 2d. 2e. 2f. 4d. 49. 4f. 49. 6d. 6e. 6f. 69. 8. 9. 10. 11. 12. and 13) were measured on one Of two S-point Likert scaTes (see samp1es beTow). How much does this person he1p you to remember to foTTow your diet? _____ A great deaT _____ A TittTe ______ A Tot Not at a11 Some How much does this person remind you to exercise? Far tOO much _____ Somewhat too TittTe ______Somewhat too much ______Far tOO TittTe _____ About the right amount "A great deaT" and "Far too much." though coded 1. were given a score Of 5. with each response in order receiving the next Towest whoTe number to T for "Not at a11" and "Far too TittTe." Questions in the format Of the first exampTe demonstrate the highest degree of support with a score Of 5 and the Towest with a score of T. In the second exampTe. a score Of 3 denotes the patients' perceptions of adequate support. with other answers indicating too much or too TittTe support 104 perceived. The scoring of the second exampTe is aTso used for ques- tions that examine possib1e negative support or "pressure" to foTTow modaTities in the therapeutic regimen. Research Question 2. "Are there differences in perceptions of socia1 support among the various moda1ities of the therapeutic regimen?" was examined by ana1ysis of the responses that Took specificaTTy at each modaTity. These responses are a11 on the Socia1 Interaction Questionnaire. as noted earTier. and are as foTTows for each modaTity (See Appendix 0): Medication: Questions 2. 2a. 2b. 2c. 2d. 2e. 11. and 14b Diet: Questions 4. 4a. 4b. 4c. 4d. 4e. 4f. 49. 12. and 14c Exercise: Questions 6. 6a. 6b. 6c. 6d. 69. 6f. 69. 13. and 14d The examination Of "differences in perceptions Of socia1 support among subgroups of the samp1e studied based on standard sociodemo- graphic characteristics." as stated in Research Question 3. was again done by ana1yzing responses given in the Socia1 Interaction Question- naire. In this case. however. responses were grouped according to the sociodemographic characteristics that were c011ected with data upon entry into the study (Intake--see Appendix A) and at the time the Socia1 Interaction Questionnaire was administered (foTTow-up--see Appendix E). Data were c011ected on the foTTowing characteristics: sex. age. race/ethnicity. maritaT status. number of Tiving chderen. income. work status. occupation. education. Tiving arrangements. number of chderen Tiving at home. size of househon. and duration of hyper- tension. OperationaTization and coding of these sociodemographic 105 characteristics can be found in Appendices A and E (intake and 9-month sociodemographic instrument). WWW ReTiabiTity and vaTidity of the Socia1 Interaction Questionnaire have not been estabTished due to the one-time administration and the abstract nature of the concept of socia1 support. ReTiabiTity. as defined by PoTit and HungTer (1978). is the consistency with which the instrument measures the attribute it is intended to measure. The statisticaT measurement of reTiabiTity is normaTTy accompTished in one of two ways: repeated administration of the instrument (test-retest) and/or by methods that examine internaT consistency of the instrument. such as spTit-han reTiabiTity or the Kuder-Richardson FormuTa 20. Neither of these methods was possib1e with the Socia1 Interaction Questionnaire. The questionnaire was administered one time so the test-retest was not possib1e. ‘The format of the questionnaire did not Tend itseTf to methods that measured internaT consistency. The major- ity of the questions on the instrument were open ended. with patients responding "yes" to a fiTter question compTeting one portion of the questionnaire and those answering "n0"‘t0 a fiTter question compTeting another portion of the questionnaire. There was a fiTter question for each modaTity of the therapeutic regimen (medication. diet. and exer- cise). Therefore. the patient coqu have answered the questionnaire in six different ways. making it very difficuTt to use internaT- consistency methods. 106 A1though not quantified. an attempt to ensure reTiabiTity of the Socia1 Interaction Questionnaire was made by using consistent activi- ties in the coding process. Answers to open-ended questions were examined by the supervisory staff and coders. Together the members of this group discussed and determined how the categories for coding each question woqu be constructed. This process increased each staff Inemberis understanding of the coding categories and ensured consistent pTacement of answers in the correct categories. ‘To further enhance the intercoder reTiabiTity. spot checks of the coding were made weekTy or at more frequent intervaTs by the supervisory staff. If discrepancies were noted. the matter was discussed with the coder and data were recoded in the correct category. VaTidity refers to the degree to which the instrument measures the trait or concept it is desired to measure. As noted by King (1982). it is impossibTe to deveTop objective criteria against which to compare abstract concepts such as socia1 support. "Therefore it becomes neces- sary to depend on subjective criteria upon which to base evaTuation Of the vaTidity of the instrumentJ' VaTidity. then. must rest with the investigator and his/her estimation of the degree to which the instru- ment wi11 measure the concept. The vaTidity of the Socia1 Interaction Questionnaire was enhanced by the research stafT"s previous experience with research in which socia1 support was a major concept. The content and format of the questionnaire were based on a review of current Titerature by the research staff. Most important among the works were those that examined the perception or subjective aspects of socia1 107 support (Cap1an. 1979). The Titerature review. with a focus on research of perceptions of socia1 support. aTso heTped to ensure that the questionnaire woqu indeed measure perceptions of socia1 support. W 1. What are hypertensive patients' perceptions of socia1 support in foTTowing a therapeutic regimen? 2. Are there differences in perceptions of socia1 support among the various modaTities of the therapeutic regimen? 3. Are there differences in perceptions of socia1 support among the various subgroups of the samp1e studied. based on standard socio- demographic characteristics? WW3 Descriptive and inferentiaT statisticaT methods were used to ana1yze data and attempt to answer the research questions. Data re1ated to Research Question 1. "What are hypertensive patients' per- ceptions of socia1 support in foTTowing a therapeutic regimen?" wiTT be described using frequencies of responses in different categories. Who was supportive (Questions 1. 2a. 4a. 6a. and 14a). type of support (Questions 2c. 4c. 6c. and 16a). and quantity Of support (Questions 2d. Ze. 2f. 4d. 4e. 4f. 49. 6d. 6e. 6f. 69. 8. 9. TO. 11. 12. and 13) were data taken from the Socia1 Interaction Questionnaire and ana1yzed in this manner; Contingency tabTes and bar graphs wiTT be used to dispTay these descriptive statistics in a bivariate form. ids. supportive person x type of support and supportive person x quantity of support. 108 To answer Research Question 2. "Are there differences in percep- tions Of socia1 support among the various modaTities of the therapeutic regimen?" frequencies in contingency tabTes and bar graphs wi11 be used to present the variabTes of socia1 support (support person. type of support. and quantity of support) by the three modaTities of the thera- peutic regimen in this study (medication. diet. and exercise). As with the two previous questions. contingency tabTes and bar graphs wiTT be used to dispTay data pertinent to Research Question 3. "Are there differences in perceptions of socia1 support among the various subgroups Of the samp1e based on standard sociodemographic characteristics?" Each Of the three variabTes re1ated to socia1 sup- port (support person. type of support. quantity Of support) was exam- ined with the sociodemographic characteristics noted earTier: sex. age. race/ethnicity. maritaT status. number Of Tiving chderen. income. work status. occupation. education. Tiving arrangements. number Of chderen 1iving at home. size of househOTd. and duration of hyperten- sion. The differences between experimentaT and controT groups were aTso examined as subgroups of the samp1e. Chi-square and 1 tests were used to ana1yze data for Questions 2 and 3. which address differences; ida. are the frequencies Of this supporter. modaTity. or subgroups equaT or not? and is the 1eve1 of support given equaT? As the N permits. extension of the 1 test may be needed in the form of ana1ysis of variance for mu1tip1e independent variabTes and/or mu1tivariate _t's for mu1tip1e dependent variabTes. 109 This was necessary since the dependent variabTes were 1ike1y to be intercorreTated. Summarx The methodoTogy and procedures used in the thesis were presented in Chapter 4. The sites. popu1ation and samp1e. and data-coTTection techniques were described and discussed. Human rights protection was outTined. and procedures for statisticaT ana1ysis were presented. A detaiTed ana1ysis of the data is presented in Chapter 5. CHAPTER 5 DATA PRESENTATION AND ANALYSIS IDILQQHQIIQH In this chapter the sampTe wiTT be described by its sociodemo- graphic characteristics. Data obtained from the Socia1 Interaction Questionnaire wiTT be presented and ana1yzed aTong with the sociodemo- graphic data in an attempt to answer the foTTowing research questions: 1. What are hypertensive patients' perceptions of socia1 support in foTTowing a therapeutic regimen? 2. Are there differences in perceptions Of socia1 support among the various moda1ities of the therapeutic regimen? 3. Are there differences in perceptions of socia1 support among subgroups of the samp1e based on standard sociodemographic characteristics? Wale The study sampTe was part of a Targer samp1e of 158 hypertensive patients who participated in the research project "Patient Contribu- tions to Care--Link to Process and Outcome»" 8. Given and C. W. Given. co-principaT investigators. The present subsampTe inc1uded 102 maTe and femaTe hypertensive patients ranging in age from 24 to 65 years. 110 111 The onTy criterion for inc1usion in this study was compTetion of the Socia1 Interaction Questionnaire. MW The sociodemographic characteristics examined in this study were sex. age. race/ethnicity. maritaT status. number of Tiving chderen. income. work status. occupation. education. Tiving arrangements. number of chderen Tiving at home. size of househon. and duration Of hyper- tension. Sex A nearTy equaT number of maTe and femaTe hypertensive patients participated in this study. 52 maTes (51%) and 50 femaTes (49%). Ass Study participants ranged in age from 24 to 65 Years. with a mean age_ Of 47598 years. The distribution and percentages of subjects by age are dispTayed in Tab1e 5.1. Tab1e 5.1: Number and Percentage Of Subjects by Age (n = 102) Age Number of Subjects Percentage 24-39 27 26 40-49 27 26 50-59 28 31 60-66 16 16 112 W Each participant was asked to record his/her raciaT or ethnic background. ‘The majority of the subjects in this study were white in = 90; 88%). Data on the number and percentage Of subjects by race are iTTustrated in TabTe 5.2. Tab1e 5.2: Number and Percentage of Subjects by Race (n,= 102) Race Number of Subjects Percentage White 90 88 BTack TO 10 Mexican-American T 1 Other 1 T .Marital_§tatus Three-quarters (83%) of the hypertensive patients in this study were married. The remaining subjects were aTmost equaTTy divided among the categories of singTe (7%). divorced (5%). and widowed (5%). MW Subjects reported having from 0 to 12 1iving chderen. ‘The mean number of chderen was 3. Thirteen percent of the participants (n = 13) had no 1iving chderen. A smaTTer percentage (n = 8; 8%) had one chde. ApproximateTy one-quarter of the subjects (n = 28; 28%) had two chderen. An aTmost equaT number (n = 33; 33%) of the hypertensive patients had three or four chderen. Eighteen participants (18%) reported having five or more chderen. 113 Income Data on yearTy totaT famiTy income were Obtained from 100 subjects. Sixty-three percent of the subjects (n = 63) reported incomes above $20.000. Tab1e 5.3 inc1udes the distribution and per- centage Of subjects by income. Tab1e 3: Number and Percentage of Subjects by TotaT FamiTy Income (3 = 100) Income Number of Subjects Percentage Less than $5.000 3 3 5 5.000- 6.999 2 2 5 7.000- 8.999 4 4 S 9.000-10.999 5 5 $11.000-12.999 4 4 $139000-149999 1 1 $15.000-16.999 7 7 $17.000-19.999 11 11 $20.000-24.999 16 16 $25.000 or more 47 47 We Current work status was recorded for each participant. Distribu- tion and percentage of participants by work status are presented in Tab1e 5.4. ApproximateTy two-thirds (n = 68; 67%) of a11 subjects were working outside the home for pay. 114 Tab1e 5.4: Number and Percentage Of Subjects by Work Status (n,= 101) Work Status Number of Subjects Percentage Working 68 67 UnempToyed or Taid Off 4 4 Retired 8 8 DisabTed 4 4 Housewife 18 18 Occupation Sixty-six participants responded to a question asking them to describe their work. This number is two Tess than the 68 subjects who stated they were working outside the home. ‘The number and percentage Of subjects according to occupation are iTTustrated in TabTe 5.5. The HoTTingshead OccupationaT ScaTe was used to code the occupationaT variabTes. The category of cTericaT/saTes was the most frequentTy reported in = 14; 21%). Least reported was the category of unskiTTed Tabor (n = 4; 6%). The remaining subjects reported occupations fairTy evenTy distributed within the other occupationaT categories. Education ApproximateTy one-third (n = 33; 32%) of the samp1e were high schooT graduates. NearTy one-han (n = 47; 47%) had attended or graduated from coTTege. whereas 20% (n = 20) of the subjects had Tess than a high schooT education. The number and percentage Of hyperten- sive patients by education are presented in Tab1e 5.6. 115 Tab1e 5.5: Number and Percentage of Subjects by Occupation (n = 66) Occupation Number of Subjects Percentage Higher executive. major professionaT 12 18 Business manager. Tesser professionaT 12 18 Administrator. minor professionaT 7 11 CTericaT. saTes T4 21 SkiTTed. manuaT 9 T4 Semi-skiTTed 8 12 UnskiTTed 4 6 Tab1e 5.6: Number and Percentage of Subjects by Education (n_= 102) Education Number of Subjects Percentage None or some schooT (Tess than 7 years) 5 5 Junior high (compTeted 9 grades) 4 4 Some high schooT TT 11 High schooT graduate 33 32 TechnicaT. business. or trade schooT 2 2 Some coTTege (Tess than 4 years) 20 20 COTTege graduate 12 12 Postgraduate or profes- sionaT 15 15 116 MW Information regarding whom the subject Tived with was Obtained. Tab1e 5J7 represents the distribution and percentage of participants by Tiving arrangements. ATmost one-han (n = 48; 48%) of a11 subjects were married and 1iving with a spouse and chderen. Tab1e 5.7: Number and Percentage of Subjects by Living Arrangements (n,= 100) Living Arrangements Number of Subjects Percentage Unmarried; Tiving aTone 9 9 Unmarried; Tiving with reTative or unreTated person 3 3 SingTe; Tiving with chderen 6 6 Married; Tiving with spouse and chderen 48 48 Married; Tiving aTone with spouse 28 28 Married; Tiving with spouse. chderen. and other 4 4 _ reTatives Married; 1iving with spouse and other reTatives 2 2 NumOoLoLQhJJdLoLLinnuLflome The number of chderen Tiving at home was obtained from each participant. The numbers ranged from zero to seven. with 42% (n_= 41) reporting no chderen 1iving at home. ApproximateTy one-fourth (n = 24; 24%) had one chde Tiving at home. Subjects with two chderen in the home represented 17% (n = 17) of the totaT samp1e. whi1e those with 117 three chderen represented 9% (n = 9). those with four chderen 5% (n = 5). and those with five or seven chderen 1% each (n =1. 11 =1). Sizs_9i_fl9u§£hold The size of househon was reported by 95 subjects. One-third of the samp1e (n = 32; 33%) had two other peop1e Tiving in the househOTd in addition to themseTves. and one-fourth (n = 23; 24%) had three other peop1e in the househon. Subjects having one. four. or five others in the househon were aTmost equaTTy distributed in = 11. 11%; n = 13. 13%; and n = 10. 10%. respectiveTyL Five subjects (5%) reported six others in the househOTd. and two subjects reported seven others (n = T; 1%) and nine others (n =1; 1%) in the househOT d. DuLaIJOLoLUxooLtension The distribution and percentage of subjects by duration of hyper- tension are inc1uded in Tab1er5.8. NearTy one-quarter of the partici- pants (n = 24; 24%) had had hypertension for 3 to 5 years. Sixteen subjects (16%) had had hypertension for 15 years or more. WW Descriptive findings re1ated to sociodemographic characteristics of the samp1e were presented in the previous section. Examination of the descriptive statistics showed a nearTy equaT number of maTes and femaTes in the sampTe. The majority of participants were found to be white. married. middTe income. empToyed. and graduates of high schooT (Of these. most had attended or graduated from coTTegeL 118 TabTe 5.8: Number of Subjects by Duration of Hypertension (n,= TOO) Duration of Hypertension Number of Subjects Less than 1 year 14 1-2 years 17 3-5 years 24 6-8 years 17 9-11 years 10 12-14 years 2 15 or more years 16 QatLEcssoniatioLtoLBesoaLoOMostions The research questions wiTT be presented in this section aTong with pertinent data. Data for each research question wi11 be presented by the three variabTes Of socia1 support examined (support person. type of support. and quantity of support). The statisticaT techniques Of chi-square and I test wiTT be used to ana1yze differences in percep- tions of socia1 support among the moda1ities of the therapeutic regimen and subgroups of the samp1e. Descriptive statistics (frequencies. percentages. and means) wiTT be used to describe hypertensive patients' perceptions Of socia1 support in foTTowing a therapeutic regimen. BosearoLQuostionJ What are hypertensive patients' perceptions Of socia1 support in foTTowing a therapeutic regimen? Suppcrt_pcrscn. Participants responded to a question asking if someone was ab1e to he1p (support) them with each modaTity of the therapeutic regimen. Hypertensive patients were most often ab1e to identify someone ab1e to he1p with the moda1ity of diet (n = 53; 71%). 119 foTTowed by medication (3 = 60; 67%) and finaTTy exercise (n = 44; 48%). Ninety subjects responded to the question for medication. 75 responded regarding diet. and 91 responded for exercise. Subjects were aTso asked to identify the most heTprT (supportive) person for each moda1ity of the therapeutic regimen. Fifty-nine Of 60 subjects who stated someone was abTe to heTp with medication identi- fied the most he1pfu1 person. A11 53 subjects who re1ated someone was ab1e to he1p with diet identified a support person. whi1e 40 of the 44 participants did the same for exercise. Spouse was the most frequentTy identified support person for a11 moda1ities: diet in = 35; 66%). medication (n = 37; 63%). and exercise (n = 5; 9%). The distributions and percentages of support person for each modaTity are dispTayed in Tab1e 5.9. Tab1e 5.9: Number and Percentage of Supportive Person by ModaTity Support Person Medication Diet Exercise 2' % .3 % fl- % Spouse 37 63 35 66 18 45 Chde 3 5 5 9 3 7.5 Other reTative 2 3 T 2 0 O Friend/workmate 5 9 7 13 T4 35 Doctor 1 2 0 O 0 0 Nurse 3 5 3 6 T 2.5 NO one/myseTf 3 S 1 2 O 0 Dietitian O 0 1 2 0 0 Pet 0 0 0 O 3 7.5 More than T of above 4 7 0 0 1 2.5 120 mm. Subjects answered a question asking them to describe the he1p (support) they received from the most heTpfuT person for each moda1ity Of the therapeutic regimen. Fifty-three subjects responded to the question for medication. 54 for diet. and 44 for exercise. Hypertensive patients identified much higher 1eveTs of psy- choTogicaT support for medications (n = 40; 75%) than for diet (11 = 19; 35%) or for exercise (11 = 16; 36%). ConverseTy. subjects were more TikeTy to identify tangib1e support for exercise (11 = 23; 52%) and diet (3 = 26; 48%) than for medication (11 = 7; 13%). A fairTy evenTy distributed percentage of participants identified receiving both psy- choTogicaT and tangib1e support: medication 11% (n = 6). diet 17% (n = 9). and exercise 11% (n = 5). W. Means were computed for 17 questions that attempted to quantify support for foTTowing a therapeutic regimen. Subjects recorded responses on two different 5-point Likert scaTes. The first scaTe was used to measure "how much he1p" was given for each of the modaT ities. Answers ranged from "A great deaT" to "Not at a11." The second scaTe was used to quantify "how much concern" was shown. "how much the support person reminded." and "how much pressure" was pTaced on the subject to foTTow each moda1ity. The responses for each ranged from "Far too much" to "Far too TittTe." The number of subjects and mean for each question by moda1ity are presented in Tab1e 5.10. 121 Tab1e 5.10: Number of Subjects and Means of Quantity of Support by ModaTity Quantity of Support Medicatign Diet_' Exercise EL >_<. o i o X How much he1p 59 3.66 53 3.79 42 3.55 How much concern (foTTow moda1ity) 61 3.02 53 3.02 43 3.02 How much concern (weight) -- -- 52 2.96 -- -- How much concern (estabTish exercise) -- -- -- -- 43 3.00 How much remind 59 3.02 52 2.98 43 3.14 How much pressure 92 2.73 96 2.78 94 2.68 The Tast three questions attempted to quantify more generaT fee1- ings of support: How much did famiTy and friends make the subject fee1 worthwhiTe (n = 95; X= 3.91)? How much concern did famiTy and friends show about their hypertension controT (n = 97; X= 3.54)? and How much did the subject have to re1y on himseTf/herseTf to take care of their hypertension (11 = 96; -.)_(_ = 3.15)? Subjects perceived the quantity of "he1p" to be between "some" and "a Tot." Other aspects of support (i.e.. concern. reminding. pressure) were perceived as "about the right amount." In summary. hypertensive patients most often perceived he1p avai1ab1e for diet. foTTowed by medication and exercise. Spouse was perceived as the most supportive person. foTTowed by friend. for a11 moda1ities. Subjects perceived much higher 1eveTs Of psychoTogicaT than tangib1e support for medication. STightTy more tangib1e than psychoTogicaT support was perceived for diet. whiTe nearTy equaT 122 amounts of tangib1e and psychoTogicaT support were perceived for exercise. The quantity of concern. reminding. and pressure perceived were "just about right." whi1e he1p perceived ranged between "some" and "8 Tot." BeseanoLQueinoLZ Are there differences in perceptions Of socia1 support among the various modaTities of the therapeutic regimen? ‘Suppcrtrpcrscn. There was a significant difference in perceptions (x2 = 10.79. df = 2. p < .01) among the moda1ities regarding the hypertensive patients' abiTity to identify someone abTe tO he1p with each moda1ity of the therapeutic regimen. There was no statisti- caTTy significant difference in perceptions among the modaTities regarding the person identified as most ab1e to he1p. Data were dichotomized into categories of "spouse" and "other" for ana1ysis. This eTiminated a Targe number Of ceTTs with frequencies of zero. W. For purposes of ana1ysis. subjects who recorded perceptions of both tangib1e and psychoTogicaT support were added to both the tangib1e and the psychoTogicaT categories. A signifi- cant difference in perceptions (X2 = 1332: df = 2' .12 < .001) Of type of support was noted among the moda1ities Of the therapeutic regimen. The number and percentages Of subjects by type Of support are presented in Tab1e 5.11. .ngntlty_c£_suppcr1. Paired 1 tests were used to ana1yze differences in quantity of socia1 support among the moda1ities. This test is appropriate when two measures are obtained from the same 123 subject (PoTit & HungTer. 1981). In this case. the mean scores for quantity of support for medication. diet. and exercise were compared and ana1yzed. Since the scores are paired. the n of the test ref1ects the n of the modaTity with the fewest subjects. usuaTTy exercise. Tab1e 5.11: Number and Percentage of Subjects by Type of Support* ModaTity TangibTe PsychoTogicaT Both r, % g_ % n_ % Medication 7 13 40 75 6 11 Diet 26 48 TO 35 9 17 Exercise 23 52 19 35 5 11 *Significant at.p < .001. The paired t test was performed on a11 data re1ated to "how much he1p." "how much concern." "how much reminding." and "how much pres- sure" the subject perceived for foTTowing each moda1ity Of the regimen. StatisticaTTy significant differences were found among the moda1ities re1ated to "how much heTp" subjects perceived. Subjects perceived more "he1p" for medications than exercise (1; = -2.05. df = 27. ,p = .05. n = 28) and more "heT p" for diet than exercise (1; = -2.74. df = 27. c = .01. n_= 28). There was no significant difference of "heTp" between medication and diet. nor were there significant differences re1ated to "concernfl'"reminding." or "pressureJ' In summary. there were statisticaTTy significant differences found among the moda1ities of the therapeutic regimen regarding patients' 124 perceptions of "someone ab1e to he1p." type of support. and "how much he1p" they received. BeseaLobJTuostionJ Are there differences in perceptions Of socia1 support among various subgroups Of the samp1e based on standard sociodemo- graphic characteristics? Data to answer this question wiTT be presented by condition or sociodemographic characteristic in re1ation to the variabTes of socia1 support (support person. type of support. and quantity of support). .Ccndjjjcn_by_suppcrt_pcrscn. ApproximateTy two-thirds (E: n = 44; 68% vs. C:.n = 16; 64%) of both the experimentaT and controT groups were ab1e to identify someone ab1e to he1p (support) them for the moda1ity of medication. ApproximateTy three-quarters Of both the experimentaT and controT groups identified someone ab1e to heTp (sup- port) them for the modaTity of diet. There was aTso a simiTarity between the groups for the moda1ity of exercise. with one-han Of both groups identifying someone ab1e to he1p. ‘The number and percentage of patients who identified a person ab1e to he1p are dispTayed in Tab1e 5.12. NO statisticaTTy significant differences were found between conditions for any of the moda1ities. Spouse was the most frequentTy identified as "person most he1p" for both experimentaT (n = 22; 52%) and controT (n = 15; 88%) groups for the modaTity of medication. ‘The onTy other person identified by the controT group was friend (n = 1; 6%). Subjects in the experimentaT group identified chde (11 = 3; 7%). re1ative(n = 2; 5%). friend (11 = 4; 10%). and others. e.g.. doctor. nurse (n = 11; 26%). When data were 125 categorized into "spouse" and "other" for ana1ysis. there was a sig- nificant difference in perceptions between the experimentaT and controT groups (x2 = 6.63. df =1. 12 < .01) as to who was identified as the most supportive person for the moda1ity of medication. Tab1e 5.12: Number and Percentage of Patients Who Identified Support Person by Condition ModaTity ExperimentaT ControT n_ % n_ % Medication 44 68 16 64 Diet 40 70 T3 72 Exercise 31 46 T3 54 Spouse was again the most Often identified support person for experimentaT (n = 25; 66%) and controT (n = 10; 67%) groups for the moda1ity Of diet. Both groups identified friend (E: n = 4; 11% vs. C: n = 3; 20%) and other (E: n = 4; 11% vs. C: n = T; 7%). ResuTts were very simiTar between groups for the moda1ity. and no statisticaTTy significant differences were found (c f .05). For the moda1ity of exercise. the experimentaT group identified spouse and friend equaTTy (n = 10; 36%) as the most supportive person. They aTso identified chde (11 = 3; 11%). pet (n = 3; 11%). and other (11 = 2; 8%). Subjects in the controT group identified on1y spouse (n = 8; 67%) and friend in = 4; 33%). These data were aTso ana1yzed in two categories. "spouse" and "other." ResuTts showed significant 126 differences in perceptions of "who was most supportive" for the moda1- ity of exercise (X2 = 6.04. df = 1. p < .02). WW. A1though some differences were noted between the experimentaT and controT groups on the type of sup- port identified for the modaTities of diet and exercise. these were not statisticaTTy significant (9 g .05). The number and percentage of subjects by condition and type of support are presented in Tab1e SIEL Tab1e 5.13: Number and Percentage Of Subjects by Condition and Type of Support Type of Support ExperimentaT ControT n_ % n % TangibTe--medication 9 21 4 24 PsychoTogicaT--medication 33 79 13 76 TangibTe--diet 22 49 23 51 PsychoTogicaT--diet 11 68 5 31 TangibTe-~exercise 22 63 6 42 PsychoTogicaT--exercise 13 37 8 57 ‘Qcnd1I1cn_by_guanrity_cr_suppcrt. Differences in quantity Of support were again ana1yzed using 1 tests. No statisticaTTy signifi- cant differences were found between the experimentaT and controT groups for any of the moda1ities. jgnanrjuuuuujLJuugguL MaTes identified more peop1e ab1e to he1p for medication in = 34; 71%) and diet (n = 32; 78%) than for exercise (3 = 18; 36%). FemaTes identified peop1e ab1e to he1p (support) equaTTy for a11 moda1ities at 62% (medication 3 = 26. diet n = 21. 127 exercise n = 26L The difference between the number Of maTes and femaTes who were ab1e to identify a support person for the modaTity Of exercise was statisticaTTy significant (X2 = 4.77. df = T. p,< .05). Differences in perceptions for moda1ities of medication and diet were not significant. Tab1e 5.14 inc1udes these data. Tab1e 5.14: Number and Percentage of Subjects Who Identified Support by Sex ModaTity MaTe FemaTe fl. % n_ % Medication 34 71 26 62 Diet 32 78 21 62 Exercise 18 36 26 62* *Significant at.p < .05. MaTes were much more TikeTy to identify a spouse as the most heTpfuT (supportive) person--medications (n = 26; 79%). diet (n = 27; 84%). and exercise (n = 9; 53%)--than femaTes--medications (n = 11; 42%). diet (11 = 8; 38%). and exercise (11 = 9; 39%). FemaTes identified friends as equaTTy supportive as a spouse for the modaTity of exercise (n = 9; 39%) and the next person most 1ike1y to he1p with the modaTi- ties Of medication (n = 4; 15%) and diet (n = 5; 24%). MaTes perceived friends as most heTpfuT after spouse for the moda1ity of exercise (n = 5; 29%). 128 When data were ana1yzed in two categories Of "spouse" and "other." significant differences in perceptions of "who was most he1pfu1" were noted between maTes and femaTes for the modaTities Of medication and diet. For the moda1ity of medication. X2 = 6.04. df =1. 12 < .02. For the modaT ity of diet. X2 = 12.09. df = T. ,p < .001. There was no statisticaTTy significant difference found for the moda1ity of exer- cise. WM. Both maTes and femaT es perceived and identified higher 1eveTs of psychoTogicaT than tangib1e support for the modaTity of medication. FemaTes. however. identified onTy 5% (n = 1) tangibTe support whiTe maTes identified 19% (n = 6). This difference was found to be significant (X2 = 6.24. df = T. p < .02). MaTes identified more tangib1e support than femaTes for the moda1- ity of diet; maTes (n = 23; 62%). femaTes (n = 10; 42%). FemaTes identified more tangib1e support than maTes for the modaTity of exer- cise; femaTes (n =18;66%). maTes (n = 10; 41%). Neither of these differences was significant. Data for this section are dispTayed in Tab1e 5.15. Wm. One statisticaTTy significant difference was noted between men and women in perceptions Of quantity of support. Women perceived Tess "pressure" to take medication than men (J; = -2.98. df = 60. p = .004). 129 Tab1e 5.15: Number and Percentage of Subjects by Sex and Type Of Support Type of Support MaTes FemaTes g_ % g_ % TangibTe--medication 6 19 1 5* PsychOTogicaT--medication 25 81 21 95 TangibTe--diet 23 62 TO 42 PsychoTogicaT--diet T4 38 14 58 TangibTe--exercise TO 41 18 66 PsychoTogicaT--exercise 12 59 9 -33 *Significant at.p < .02. .Bacc. As noted ear1ier. a Targe majority of the samp1e was white in = 90; 88%). ‘The smaTT number Of subjects from other raciaT back- grounds did not aTTow for ana1ysis of data that woqu show either statisticaT significance or practicaT differences. This is due to having Targe numbers of ceTTs in a chi-square test with zero entries. A1though the statistic may prove significant difference between whites and Mexican-Americans. it is not sensibTe to base that concTusion on 90 responses from whites and 1 response from Mexican-Americans. .Mcr11n1_stctus. Data were dichotomized into categories of married and unmarried for ana1ysis due to the smaTT number of subjects in each Of the unmarried categories (singTe. divorced. and widowed). A higher percentage of married subjects were ab1e to identify a support person for the moda1ities of medication (n = 52; 70%) and diet in = 46; 73%) than those who were not married: medication (n = 8; 50%) and diet (a = 130 7; 58%). For the moda1ity of exercise. both groups equaTTy identified peop1e abTe to he1p: married (11 = 37; 49%). unmarried (n = 7; 47%). None of these findings was statisticaTTy significant (p < .05). It was difficuTt to ana1yze differences between married and unmar- ried subjects as to "who was most heTprLfl Consistent with other data. married subjects most Often identified spouse as most he1pfu1: medications (n = 37; 72%). diet (11 = 35; 76%). and exercise (11 =18; 51%h. Friend was the person most often identified by unmarried sub- jects: medication (11 = 4; 50%). diet (11 = 3; 43%). and exercise (n = 5; 100%). The chi-square technique was not used because of the zeros present in the spouse ceTT for unmarried subjects. WWW. Tab1e 5.16 portrays the number and percentage of subjects by maritaT status and type of support. Both married and unmarried subjects perceived more psychoTogicaT support for medications. Both groups aTso identified more tangibTe than psycho- TogicaT support for exercise. For the moda1ity of diet. however. married subjects identified sTightTy more tangib1e support. whiTe unmarried subjects identified more psychoTogicaT support. 'This differ- ence was not statisticaTTy significant at p = .05. .Mar11a1_sratu5_by_gunntity_cr_§upccrr. Data regarding quantity of support and maritaT status were not ana1yzed. This decision was based on the fact that onTy one significant finding emerged from data on the two samp1e characteristics which previousTy had the most differences. condition and sex. Therefore. no further data ana1ysis on quantity of 131 support re1ated to other sociodemographic characteristics wi11 be presented. Tab1e 5.16: Number and Percentage of Subjects. Marita1 Status by Type of Support Type of Support Married Unmarried .g % g_ % TangibTe--medication 11 22 2 25 PsychoTogicaT--medication 4O 78 6 75 TangibTe--diet 31 57 2 33 PsychoTogicaT--diet 23 42 4 66 TangibTe--exercise 23 56 5 63 PsychoTogicaT--exercise T8 44 3 37 W. The Targe range in the number Of Tiving chderen (0 to 12) made the data difficuTt to ana1yze with chi- square. The Targe number of ceTTs with zero may produce a faTse statisticaT significance. Therefore. differences in perception were not caTcuTated for this sociodemographic characteristic. W. To ana1yze the income data more easiTy. data were dichotomized into categories of "Tess than $25.000" and "$25.000 or more." Both groups were nearTy equaT in identifying someone ab1e to he1p for each moda1ity: medication (< $25.000 n = 28; 62%; > 525.000 11 = 30; 69%). diet (< $25.000 n = 22; 69%; > $25.000 n = 29; 72%). and exercise (< $25.000 n = 21; 48%; > $25.000 n = 23; 51%). 132 Subjects in the $25.000 or more category identified spouse more often than those with incomes under $25.000 for the moda1ities of medication and diet: medication (< $25.000 n = 22. 76%; > $25.000 n = 14. 54%). diet (< $25.000 n = 23. 79%; > $25.000 n = 11. 52%). Those in the Tess than $25.000 category identified chde. reTative. friend. and others more often than subjects with incomes of $25.000 or more. The support variabTe of "who was most he1pfu1" was aTso dichotomized for the purpose of ana1ysis to "spouse" and "other." There was a stati sticaTTy significant difference in perceptions Of "who was most he1pfu1" for the modaTity of diet (x2 = 4.06. df = 1. o < .05). A significant difference was aTso noted for the modaT ity of exercise (X2 = 3.92. df =1. p < .05). In the exercise modaTity. subjects with income Tess than $25.000 were more 1ike1y to identify friend (n = 10; 56%) than spouse (n = 5; 28%). Those with incomes of $25.000 or more identified spouse most often (11 = 13; 59%). foTTowed by friend (11 = 4; 18%). W. Perceptions of type of support were simiTar between the two income categories. Both groups identified higher 1eveTs of psychoTogicaT support for the modaTity of medication. Both groups perceived nearTy equaT amounts of tangibTe and psychoTogi- caT support for diet. Both groups identified sTightTy more tangib1e support for the modaTity of exercise. Data representing number and percentage of subjects by income and type of support are found in Tab1e 5.17. 133 Tab1e 5.17: Number and Percentage of Subjects by Income and Type of Support Type Of Support < $25.000 > $25.000 n. % n_ % TangibTe--medication 4 T7 8 25 PsychOTogicaT--medication 20 83 23 75 TangibTe--diet 12 50 20 55 PsychoTogicaT--diet 12 50 16 4S TangibTe--exercise 15 63 13 52 PsychoTogicaT--exercise 9 37 12 48 .Wcrk_§tctu§. Data for work status were categorized as "empToyed" and "unempToyed" for ana1ysis. None of the findings was significant at p f .05. EmpToyed and unempToyed subjects identified a person most ab1e to he1p for the modaTity of diet. with the unempToyed group per- ceiving more support (n 18; 82%) than those empToyed (n = 35; 67%). Both groups had a simiTar percentage of subjects who perceived someone ab1e to he1p for medication (empToyed 68%. unempToyed 63%) and exercise (empToyed 45%. unempToyed 54%). W. The person identified as most he1pfu1 for a11 modaTities was spouse. The empToyed group perceived spouse as most heTpfuT: 68% medication. 68% diet. and 50% exercise. UnempToyed subjects identified spouse: 63% medication. 71% diet. and 39% exercise. Friend was the next most frequent response for both groups. with the moda1ity of exercise dispTaying the Targest 134 percentage of support by friends (empToyed n = 10. 39%; unempToyed n = 4. 31%). .flQLh_5InIu§_h¥_I¥n§_o£_§uooorI. Both empToyed (61%) and unem- pToyed (78%) subjects perceived more psychoTogicaT than tangib1e sup- port for the modaTity Of medication. NearTy equaT amounts of tangib1e (55%. 53%) and psychoTogicaT (45%. 47%) support were reported for the modaTity of diet. For the moda1ity of exercise. both empToyed (56%) and unempToyed (62%) subjects identified sTightTy more tangib1e than psychoTogicaT support. .Qccuccticn. For ana1ysis. data which were coded in seven categories on the HoTTingshead OccupationaT ScaTe were regrouped into three new categories. The first (I) contained occupations from the three upper categories Of the HoTTingshead ScaTe--executives. business managers. and a11 professionaTs. The second (II) restructured category consisted Of saTes. cTericaT. and skiTTed manuaT workers. The finaT group (III) was composed of semi- and unskiTTed Taborers. Significant differences were found among subjects in these categories in their perception of someone being ab1e to he1p for each moda1ity. Tab1e 5.18 dispTays these data. Qccupc119n_by_suppcrt_pcrscn. For ana1ysis. data regarding "who was most he1pfu1" were categorized as "spouse" or "other." WhiTe spouse was identified most Often for a11 modaTities by groups I and II. subjects in group III identified spouse onTy 25% for medication. 0% for diet. and 33% for exercise. This difference in perceptions was sig- nificant for the moda1ity Of diet (x2 = 8.61. df = 2. o < -02). where 135 group III identified chderen (n = 2; 100%) as the most supportive person. Tab1e 5.18: Number and Percentage of Subjects by Occupation and AvaiTabiTity of Support I II III ModaT ity _n_ % g % fl % Medication 20 69 16 80 4 36* Diet 20 8O 12 60 2 29** Exercise 15 50 8 40 4 40 *x2 = 6.21. df = 2. o < .05. "x2: 6.72. df = 2. o < .05. Executives. business managers. and professionaTs. SaTes. cTericaT. and skiTTed manuaT workers. SemiskiTTed and unskiTTed Taborers. Note: Group I Group II Group III .QQQuna119n_h¥_I¥n§_QI_§uooort. Consistent with previous data. a11 groups I. II. and III identified more psychoTogicaT than tangib1e support for the modaTity of medication (65%. 83%. 75%). TangibTe support was perceived as greater or equaT for diet: group I (60%). group II (62%). and group III (50%). The onTy subjects to perceive Tower 1eveTs of tangib1e support for exercise were group II at 42%. with group I at 62% and group III at 50%. .Educnticn. The data for education were regrouped into three categories: I--1ess than a high schooT education. II--high schooT graduate or technicaT schooT. and III--some coTTege to postgraduate. Two-thirds of a11 groups were abTe to identify someone ab1e to he1p 136 with the modaTity of medication. Group I was Tess often ab1e to iden- ti fy someone ab1e to he1p with diet (I n = 9. 60%; II 3 =18. 72%; III 11 = 26. 74%) and exercise (I n = 5. 31%; II 11 =16. 52%; III _n = 23. 52%). These differences were not statisticaTTy significant. .EdnQa1190_h¥_§flRRQEI_REL§on. ApproximateTy one-han of groups I (55%) and III (57%) identified spouse as the most he1pfu1 person with medication. Three-quarters of group II (75%) did the same. For the moda1ity of diet (group I 78% and group III 70%) and exercise (group I 40% and group III 43%). groups I and III had a simiTar percentage of subjects who identified spouse as most he1pfu1. Group II represented 8 Tower percentage for diet (53%) and a higher percentage for exercise (50%) in their perceptions of spouse as most he1pfu1. None of these differences. however. was statisticaTTy significant. Worm Tab1e 5.19 denotes the number and percentage of subjects by education and type of support. The most striking difference among the groups was found in the modaTity of diet. where 100% (n_= 9) Of the subjects in group I identified tangib1e support. Group II identified tangib1e support 33% (n.= 7) and group III 54% (n = 17) for the moda1ity of diet. This difference was sig- nificant (X2 = 11.5. df = 2. o < .01). 137 Tab1e 5.19: Number and Percentage of Subjects by Education and Type of Support I II III Type of Support 3_ % g_ % n_ % TangibTe--medication 3 33 2 10 8 28 PsychoTogicaT--medication 6 66 T9 90 21 72 TangibTe--diet 9 100 7 33 T7 57* PsychoTogica1--diet 0 0 T4 66 13 43 TangibTe--exercise 2 33 11 6T 15 58 PsychoTogicaT--exercise 4 66 7 39 11 42 *Significant at.p < .01. Less than a high schooT education. High schooT graduate or technicaT schooT. Some coTTege to postgraduate. Note: Group I Group II Group III Ljyjng_crrnngcmcnts. The eight originaT options for 1iving arrangements were condensed to four categories for ana1ysis: group I-- unmarried. Tiving aTone (n = 9); group II--unmarried. Tiving with other in = 9); group III--married. Tiving with spouse and other (n = 54); and group IV--married. Tiving with spouse aTone (n = 28). Of the four groups. group II recorded the Towest percentage of peop1e ab1e to he1p for each modaTity: medication 38% (n = 5). diet 60% (n = 3). and exercise 25% (n = 2). Group III reported the highest percentages for medication (78%) and diet (74%). whi1e group I was found to have the highest percentage of peop1e ab1e to he1p for exercise (63%). WW. Data were again ana1yzed as "spouse" and "othenfl’ As with the characteristic of maritaT status. 138 the two groups in which subjects were married (III and IV) most Often identified spouse as most he1pfu1 for a11 moda1ities of the therapeutic regimen. ‘Those in group I most often identified friend for a11 moda1ities. whiTe those in group II identified friend most Often for medication and exercise and chde most often for diet. Chi-square woqu give a faTse representation of the significance Of these resu1ts because of the zeros in ceTTs where spouses were not present in the 1iving arrangement; therefore. it was not used for this portion of the characteristic. .L1y1ng_arrangcment_by_tyoc_c£_suoocrn. ResuTts were simiTar among a11 four groups except for the moda1ity of medication. ApproximateTy two-thirds of groups I. II. and III identified psychoTogicaT support. whiTe 100% of group IV did so. TangibTe and psychoTogicaT support were aTmost equaT among a11 groups for diet and exercise. Tab1e 5.20 dis- pTays these data. WWW. Data for this characteristic were dichotomized into "no chderen at home" and "chderen at homeJ' Chderen at home represented 57 subjects with from one to seven chderen Tiving at home. Differences between the two groups were not statisti- caTTy significant at.pi<.05. Subjects with chderen in the home identified a sTightTy higher percentage of someone ab1e to he1p for medication (70%) and diet (76%) than those without chderen—-medication (62%) and diet (66%). Perceptions were nearTy equaT for exercise (50% and 48%). 139 Tab1e 5.20: Number and Percentage Of Subjects by Living Arrangement and Type of Support Type of Support I II III IV g_ % g_ % fl_ % g_ % TangibTe--medication 2 33 1 33 10 29 0 0 PsychOTogicaT--medication 4 66 2 66 24 71 15 100 TangibTe--diet 2 50 1 33 22 59 8 50 PsychoTogicaT--diet 2 50 2 66 16 41 8 50 TangibTe--exercise 3 50 T 50 14 54 9 64 PsychoTogicaT--exercise 3 50 1 SO 12 46 5 36 Note: Group I Unmarried. Tiving aTone. Group II = Unmarried. Tiving with other. Group III = Married. Tiving with spouse. Group IV = Married. Tiving with spouse aTone. WWW. Both groups identified spouse as most he1pfu1 person for a11 moda1ities. ‘The percentage was simiTar for medication (63% and 66%h. Those with chiT- dren identified spouse more for diet (72%) and Tess for exerciset(39%) than those without chderen: diet (55%) and exercise (56%). NumooLoLoMJdLoLaLhomethooALsuooort- WhiTe both groups identified higher 1eveTs of psychoTogicaT than tangib1e support for medications. those without chderen in the home were higher (86%) than those with chderen (67%). For the moda1ity of diet. those with chderen identified more tangib1e support (55%) than those without chderen (40%). The opposite was true for exercise. Those without chderen perceived more tangib1e support (58%) than those with chderen (46%). 140 .Size_oi_hou§ehold. Data were categorized into two groups: "1 or 2" or "3 or more" peopTe in a househOTd. ResuTts were very simiTar to those for number of chderen at home and wiTT not be presented here. None of the differences was statisticaTTy significant. Duratlcn_cr_hypcrtcnsicn. Data for the characteristic were divided into four categories for ana1ysis: group I--Tess than 1 year duration (n = 14). group II--T to 5 years duration (n = 41). group III--6 to 14 years duration (3 = 29). and group IV--15 or more years duration (n = 16). For the moda1ity of medication. subjects in group IV most often identified someone ab1e to he1p (81%). Those in group III identified those ab1e to he1p 1east Often for medication (52%). Both group I and group IV reported 100% of subjects having someone to he1p with diet. This was a significant difference in perceptions (X2 = 10.86. df = 3. p, < .02) among the groups. as group II perceived 65% and group III 59% of subjects had someone ab1e to he1p with diet. Inter- estingTy. the two groups with 100% for diet were the two Towest groups abTe to identify someone ab1e to he1p with exercise: group I (31%) and group IV (27%). DurcI1cn_cI_hypcr1cusicn_by_supccrt_ocrscn. Spouse was the most frequentTy identified person "most ab1e to he1p" by aTT groups for aTT modaTities with one exception. Group I. those with hypertension Tess than 1 year. onTy identified friend in = 4) as being the most he1p with the moda1ity of exercise. In a11 other groups. friend was the second most Often identified support person for a11 moda1ities. 141 WWW. A11 groups Perceived more psychoTogicaT than tangib1e support for the modaTity of medica- tion. Those with hypertension the shortest time. group I. perceived the highest tangib1e support for medication (43%). whi1e those with hypertension the Tongest time. group IV. perceived the Towest tangib1e support (9%). A11 groups identified sTightTy more tangib1e than psy- choTogicaT support for the modaTity Of diet. Groups I and IV. which were 1east ab1e to identify someone ab1e to he1p for exercise. aTso perceived Tower 1eveTs of tangib1e support for exercise (group I. 25%; group IV. 33%) than the other subjects (group II. 56%; group III. 75%). In summary. significant differences were found among the foTTowing sociodemographic characteristics: condition by support person for medication. condition by support person for exercise. sex by person ab1e to he1p for exercise. sex by support person for medication. sex by support person for diet. sex by type of support for medication. income by support person for diet. income by support person for exercise. occupation by person ab1e to he1p for medication. occupation by person ab1e to he1p for diet. occupation by support person for diet. education by type of support for diet. and duration of hypertension by someone ab1e to he1p for diet. W W A11 subjects were asked if they thought a nurse coqu he1p them foTTow their therapeutic regimen. Forty-three (47.8%) responded with yes. whiTe 47 (52.2%) responded no. Those who answered yes described 142 the type of support the nurse coqu provide. The majority stated the nurse coqu he1p with psychoTogicaT support. ids. moraT support. caring. guidance. and know1edge. Subjects who feTt the nurse coqu not heTp were asked why the nurse coqu not be he1pfu1. The Targest number (n = 19) stated that foTTowing the therapeutic regimen was their responsibiTity. The next Targest group feTt they needed no he1p. ‘Two subjects stated it was not the nurse's roTe. Summarx In Chapter 5 the samp1e was described by its sociodemographic characteristics. Descriptive and inferentiaT statistics were used to present data reTevant to the research questions. StatisticaTTy sig- nificant differences were found among the modaTities Of the therapeutic regimen and subgroups of the samp1e based on sociodemographic charac- teristics. The data described in Chapter 5 wi11 be interpreted in Chapter 6. Conc1usions and impTications for nursing education. research. and practice wiTT aTso be discussed. CHAPTER 6 SUMMARY. INTERPRETATIONS. AND RECOMMENDATIONS Dietitian A summary and interpretation of findings are presented in Chapter 6. Inc1uded in the summary and interpretation are a discussion of the sociodemographic characteristics of the study samp1e. findings re1ated to the research questions. and the Timitations of the study. ImpTications of the findings for nursing practice. education. and research. as they re1ate to the conceptua1 framework. wiTT aTso be presented. WWW Sociodemographic characteristics which were commonTy reported in the reviewed research wiTT be discussed first in this section (14%. sex. age. race. maritaT status. income. education. and duration of hypertension). Next. sociodemographic characteristics not commonTy found in the Titerature wiTT be brief1y discussed. and the samp1e wi11 be compared with the generaT United States popu1ation. Fina11y. a summary of sampTe sociodemographic characteristics wi11 be presented and compared with characteristics of the samp1es Of other research. 1113 144 ApproximateTy equaT numbers of maTes (n = 52) and femaTes (n = 50) participated in this study. This sex distribution is sTightTy differ- ent from that of previous studies in which a higher percentage of femaTes was reported (e.g.. Earp et a1..1982. 59%; CapTan et a1.. 1979. 64%; NeTson et 81.. 1978. 69%; Morisky et 81.. 1982. 70%; McKenney et a1.. 1973. 77%). OnTy one study. that of Haynes et a1. (1976). used a monosexuaT samp1e. that is. 38 maTe steeT workers. Age Study participants ranged in age from 24 to 65 years. with a mean age of 48 years. Study participants appear sTightTy younger than samp1es in other studies. It must be noted that 65 was the upper age Timit for acceptance into this study. whi1e other investigators set no upper age Timit for inc1usion. Thirty-one percent Of the samp1e in NeTson and coTTeagues' study (1978). for exampTe. were 65 years or OTder. Subjects in two studies (Cap1an et a1.. 1979; Morisky et a1.. 1982) had a mean age of 54 years. whiTe participants in the research of McKenney et a1. (1973) averaged 60 years of age. Base More than four-fifths (88%; n = 90) of the participants in the study were white. a portion simiTar to the study samp1e used by Kirscht et a1. (1981) in which 91% of the participants were white. These were the onTy two studies reviewed in which more whites than bTacks partici- pated. BTacks comprised a Targe portion of the samp1es in studies 145 conducted in Targe inner-city c1inics (McKenney et a1.. 1973. 77%; Earp et a1..1982. 77%; Morisky et a1..1982. 91%) for two reasons. First. bTacks represented the Targest raciaT group in the generaT popu1ation where the studies were conducted. Second. there is a higher percentage of hypertensives in the bTack popu1ation than in the white popu1ation. Macital.§tatus WeTT over three-quarters (88%) Of hypertensive patients in this study were married. The remaining subjects were aTmost equaTTy divided among the categories of singTe. divorced. and widowed. The 88% is higher than percentages cited in other studies which documented this characteristics. Both CapTan et a1. (1979) and Earp et a1. (1982) reported approximateTy two-thirds (68% and 60%. respectiveTy) of par- ticipants as married. Income Sixty-three percent of subjects (n_= 63) reported annuaT totaT famiTy incomes above $20.000. A1though data were c011ected 1 to 2 years earTier. Kirscht et a1. (1981) reported a median famiTy income of $12.000 per year. whi1e Morisky et a1. (1982) cited a median income Of $4.250 for study participants. It shOUTd be noted that the Morisky samp1e was composed TargeTy Of bTack femaTes. with Tess than a high schooT education. Sti11. the median famiTy income for study partici- pants was markedTy higher than for other hypertensive subjects studied. 146 Education Eighty percent of the subjects had at 1east a high schooT education. with 47% of the high schooT graduates aTso having attended or graduated from coTTege. Again. the participants of the Kirscht et aT.(1981) study were most simiTar when compared on this characteris- tic. with 60% Of participants being high schooT graduates. The thesis findings are vastTy different from those cited by Earp et a1. (T982)-- 28% high schooT graduates--and Morisky et a1. (1981)--median of 7 years of education. It woqu appear that the participants in this study were more highTy educated than other study samp1es for which this character- istic was reported. QULatJoLoLflxothensjon Fourteen participants in the study had hypertension Tess than 1 year. 41 from 3 to 5 years. 29 from 6 to 14 years. and T6 for 5 years or more. OnTy two studies reviewed reported information regarding duration of hypertension. Subjects in the CapTan et a1. (1979) study had received treatment for hypertension for an average of 5.4 years. whi1e those in the study by Morisky et a1. (1981) averaged 6 years. The findings regarding the samp1e for this study are consistent with previous research. WWII NumbeLoLQbiLdLon No information was found in the Titerature reviewed regarding work status. occupation. or number of chderen. ApproximateTy two-thirds (n_= 68; 67%) of a11 subjects studied were working outside the home for 147 pay. During the same year. 1981. 64.6% Of the U.S. popu1ation was empToyed (StatisticaT Abstract of the U.S.. 1982). The study popu1a- tion had a sTightTy higher percentage Of peop1e empToyed than the generaT popu1ation Of the U.S. Of subjects working outside the home. 47% were in some type of professionaT/manageriaT position. 21% indicated they were in cTericaT/ saTes positions. 14% were in skiTTed trades. and 18% were in semi- or unskiTTed Tabor. Considering that 63% of the samp1e had famiTy incomes over $20.000. the Targe percentage of professionaT/manageriaT positions is not unexpected. Subjects in the skiTTed trades aTso contributed to the high income 1eveTs noted. Subjects reported having from O to 12 Tiving chderen. with 3 chderen being the mean. The average number Of chderen per white famiTy was 1.72. whiTe the average number per bTack famiTy was 2.16 in the U.S. (StatisticaT Abstract Of the U.S.. 1981). Thus the average number Of chderen for participants in this study was higher than the nationaT average. Yet. the numbers of chderen 1iving in each home-- one chde. 24%; two or more chderen. 33%--were nearTy equaT to those Of the generaT U.S. popu1ation. It woqu appear. then. that a number of the chderen of study participants were oner and not Tiving in the home. In summary. the majority of subjects studied for this thesis were found to be white. middTe income. empToyed. with chderen. and graduates of high schooT (of these. most had attended or graduated from coTTege). A nearTy equaT number Of maTes and femaTes participated. 148 Comparing the subjects from this study with those from other research studies reviewed. it was found there were more maTes. more whites. more married. higher income. more chderen. and higher educationa1 1eveTs among subjects in this study. Boscacchmgstjons Study findings. as they re1ate to the research questions. wi11 be summarized and interpreted in this section. The discussion wiTT be presented by the three variabTes of socia1 support examined: support person. type of support. and quantity Of support. Findings re1ated to Research Question T--What are hypertensive patients' perceptions of socia1 support in foTTowing a therapeutic regimen?--wiTT be presented first in each section. ‘This wiTT be foTTowed by findings regarding Research Question 2--Are there differences in perceptions Of socia1 support among the various moda1ities of the therapeutic regimen? Fina11y. findings from data Of Research Question 3--Are there differences in perceptions of socia1 support among subgroups Of the samp1e based on standard sociodemographic characteristics?--wi11 be discussed. W No previous research has asked hypertensive patients if they have someone ab1e to support them for various aspects of the therapeutic regimen. Hypertensive subjects examined in this study were most Often ab1e to identify someone ab1e to he1p with the modaTity of diet (71%; n = 53) foTTowed by medication (67%; n = 60) and finaTTy exercise (48%; n 149 = 44% The difference in perceptions between the modaTities was statisticaTTy significant (X2 = 10.79. df = 2. c < .01) and was most 1ike1y due to the Tower perception of support for exercise than for medication or diet. It is known from an audit of subjects! medicaT records that exercise was the 1east routineTy prescribed of the three moda1ities. It is reasonabTe to assume that fewer hypertensive patients were engaged in an exercise program than were taking medication or foTTowing a diet for controT of hypertension. Subjects not participating in a treatment modaTity may be Tess ab1e to identify support for the modaTity. Such woqu be the case for subjects in the controT group. One might expect the experimentaT subjects to be more abTe to identify support for exercise since the modaTity was introduced and goaTs and strategies for exercise were deveToped as part of the nursing interven- tion. In fact. subjects in the experimentaT group identified Tess support for exercise than did subjects in the controT group. A pos- sib1e expTanation is that experimentaT subjects asked for support for exercise during the intervention and did not receive it. ExperimentaT subjects woqu then perceive Tower 1eveTs of support than controT subjects and both woqu contribute to the significant difference noted. MaTe subjects in the study may aTso have strongTy infTuenced the difference between exercise and the other moda1ities. MaTes identified a much Tower percentage of peop1e ab1e to he1p with exercise (36%) than femaTes (62%) whi1e identifying higher percentages of peop1e ab1e to heTp for medication and diet than femaTes. It may be that men fee1 150 they need Tess he1p or no heTp to exercise and therefore identify no one as ab1e to he1p. Or perhaps men did not participate in exercise programs as Often as women and thus coqu not identify someone as ab1e to he1p. This may not be the case for diet if they depend on a support person to purchase and prepare foods or for medication. where a reminder to take medication may be viewed as he1pfu1. Neither socioeconomic characteristics (work status. income. occupation. education) nor famiTy structure characteristics (maritaT status. Tiving arrangements. number of chderen) appeared to infTuence the perception of Tower support for the moda1ity of exercise compared to medication and diet. No significant difference among the modaTities was noted for any of these characteristics. Spouse was the most frequentTy identified support person for a11 the modaTities: diet (66%; n = 35). medication (63%; n = 37). and exercise (45%; n,= 12). The next most Often identified support person was friend. being the highest for exercise (33%; n = 14). The differences among the modaTities were not statisticaTTy significant. SeveraT investigators (Cap1an et a1.. 1979; Kirscht et a1.. 1981; Morisky et a1.. 1982) noted that the person most often chosen to par- ticipate in the intervention was a famiTy member. OnTy Earp et a1. (1982) described the support person chosen. spouse (50%). chde (25%). and nonreTative (7%). but no differentiation was made for various aspects of the regimen. Findings in this thesis indicate higher 1eveTs of spouse support for medication and diet and higher 1eveTs Of nonreTa- tive support for exercise than reported in Earp et a1. (1982). 151 Higher 1eveTs of spouse support may be the resu1t of the high percentage of married subjects in the samp1e (88%) compared to Earp and coTTeagues (T982) (60%). Spouse support may be higher for diet and medication than exercise because eating and medication taking may more Often take pTace in the home or in the presence of the spouse than exercise. EXercise coqu take pTace in the home with a spouse present or outside the home with a spouse. friend. or other. MaTe subjects were found to have a significantTy higher percentage of spouse support than femaTes for the moda1ities of medication and diet. The reason for these differences is not c1ear. A nearTy equaT number of maTes (n = 44; 84%) and femaTes (n = 41; 82%) were married. One can hypothesize that women fee1 free to ask for he1p from a variety Of sources. whiTe men may fee1 they can onTy ask a spouse to assist them with areas Of the regimen which are re1ated to "iTTness." i.e.. medications and diet. With the recent emphasis on physica1 fitness. both men and women coqu ask for assistance from friends in foTTowing an exercise program without re1ating it to their hypertension. Women may feeT that as the nurturer in the famiTy it is not as acceptabTe for them to ask for support within the famiTy and therefore seek he1p from friends. StatisticaTTy significant differences were 8150 noted between the experimentaT and controT groups with regard to whom they identified as a support person. WhiTe both groups identified spouse most Often for medications. the difference. 52% experimentaT versus 88% controT. was substantia1. The onTy other person identified by the controT group was 152 friend. whiTe subjects in the experimentaT group identified chde. reTative. friend. and others (int. doctor. nurse). SimiTar findings were noted for exercise. with the experimentaT group identifying friend and spouse equaTTy (36%). as we11 as chde. pet. and other. ControT subjects identified onTy spouse (67%) and friends. The most TikeTy expTanation for the above findings is that subjects in the experimentaT group were ab1e to identify a broader base Of support as a resu1t of the nursing interventions. During the nursing interventions experimentaT subjects were exposed to the concept of socia1 support. expTored various sources of socia1 support. and deveToped strategies using socia1 support to foTTow the therapeutic regimen. A broader base of support woqu aTTOw a hypertensive cTient to choose a support person they perceive as most appropriate for a particuTar strategy (ixs. a friend to waTk withL If the first-choice support person was unavaiTabTe. the cTient coqu identify an a1terna- tive source of support for the strategy (ins. waTk the dog). Using a variety of support peop1e aTso Tessens the burden of the most fre- quentTy identified person. The amount of spouse support was significantTy different in onTy one of the socioeconomic characteristics. income. Subjects with incomes more than $25.000 perceived more spouse support for the moda1i- ties Of diet and exercise than those with incomes Tess than $25.000. A possib1e expTanation is that there were more singTe persons in the Tess than $25.000 group to whom spouse support was not avaiTabTe. 153 W Hypertensive patients identified much higher 1eveTs of psychoTogi- caT support for medications (81; n = 40) than for diet (54; n = 19) or exercise (42; n_= 16). ConverseTy. subjects identified more tangib1e support for exercise and diet than taking medications. These differ- ences were statisticaTTy significant (X2 = 18.32. df = 2. c < .001). The onTy study in the Titerature that tapped hypertensive patients' perceptions of the type of support they received was CapTan et a1. (1979). CapTan's measure. however. re1ated to how much Of each type of support was received and cannot be compared to these findings. In the present study. psychoTogicaT support was defined as "behav- iors directed toward providing the person with cognitions (va1ues. attitudes. beTiefs and perceptions) and toward inducing affective states that are hypothesized to promote we11-being (Cap1an et a1.. 1979. p. 14% That this type of support was perceived as forthcoming most often for the modaTity of medications is TogicaT. Taking medica- tions invoTves the Teast amount of change in TifestyTe of the three modaTities. Often it is accompTished by the patient with TittTe more than encouragement from famiTy and friends and a shared beTief that the modaTity is beneficiaT. FemaTe subjects perceived significantTy more psychoTogicaT support for the moda1ity of medications (95%) than maTes (81%). This finding may infTuence the higher amount of psychoTogicaT support noted for medication than diet or exercise. FemaTes may have perceived such high 1eveTs Of psychoTogicaT support because it is the easier of the two 154 types of support to obtain or because it was the type of support they desired for the moda1ity of medication. TangibTe support. defined as "behavior directed toward providing the person with tangib1e resources hypothesized to benefit his/her menta1 or physica1 we11 being" (Cap1an et a1.. 1979. p. 14) is more appropriate when changes in TifestyTe are required. Modifying dietary intake and foTTowing an exercise program are two such TifestyTe changes. There was one statisticaTTy significant difference in perceptions of type Of support for the moda1ity of diet. Nine subjects (100%) with Tess than a high schooT education perceived receiving tangib1e support. whi1e 33% of those with a high schooT education and 57% of those with post-high-schooT education perceived tangib1e support for diet. Subjects with Tess than a high schooT education were more TikeTy to aTso be in the Tow-income groups. TangibTe support in the form of food. money or transportation to purchase food. and money or transportation for weight-Toss support groups may be viewed as more he1pfu1 than encouragement or other forms Of psychoTogicaT support for the moda1ity of diet. TangibTe support may aTso be the onTy type of "he1p" this group Of subjects coqu identify. Hypertensive patients in generaT may find it easier to change dietary and exercise patterns with tangib1e support. If famiTy members offer support and are invoTved in changes (such as purchasing and preparing Tow-sodium foods or estabTishing a famiTy exercise program of swinuning or jogging). the TifestyTe change is incorporated in the famiTy patterns and the patient has TittTe difficuTty maintaining the 155 new famiTy pattern. Friends and other nonreTatives can aTso Offer tangib1e support as noted above. If. however. the dietary or exercise habits of the famiTy are in confTict with desired patterns Of diet and exercise. the hypertensive patient may require increased tangib1e support from friends and nonreTatives to counterbaTance the Tack Of famiTy support. Hypertension is a chronic i11ness. Therefore. TifestyTe changes re1ated to the therapeutic regimen (medication. diet. and exercise) are Tong term and require continued socia1 support. It was found that there were differences in subjects! perceptions of type of sociaT support re1ated to duration of hypertension. A11 groups perceived more psychoTogicaT than tangib1e support for the moda1ity of medication. Subjects with hypertension the shortest period of time. Tess than 1 year. perceived the highest tangib1e support for medication (43%). whi1e those with hypertension the Tongest time. 15 years or more. perceived the Towest support (9%). It woqu be TogicaT to assume that those with hypertension Tess than 1 year were stiTT trying to estabTish a daiTy pattern or routine for taking medica- tion. TangibTe support in the form of purchasing the medication. putting medication in a conspicuous pTace. or actuaTTy handing the subject the piTT woqu be appropriate for recentTy diagnosed hyperten- sive patients. ConverseTy. subjects with hypertension for 15 years or more most TikeTy have we11-estabTished patterns of medication taking. good or bad. and perceive TittTe need for the type of tangib1e support used above. 156 A11 groups perceived sTightTy more tangib1e than psychoTogicaT support for diet. Subjects who were the 1east ab1e to identify someone abTe to he1p for exercise. duration Tess than 1 year and 15 years or more. aTso perceived Tower 1eveTs of tangib1e support than other subjects for the modaTity of exercise. It may be that subjects with hypertension for Tess than 1 year were focusing on strategies to controT hypertension for the areas of the regimen most commonTy prescribed. medication and diet. They may. therefore. have asked support persons for tangib1e support for medica- tion and diet rather than exercise. If the hypertensive subject asked for tangib1e support for a11 three moda1ities. the support person may have chosen to provide support for medication and diet rather than exercise. Less "energy" is generaTTy needed to provide tangib1e sup- port for medication and diet than for exercise. This may expTain why subjects with hypertension Tess than 1 year identified friend most Often as the support person for the moda1ity of exercise. ‘They may have aTready used a11 the spouse or famiTy support for the modaTities Of medication and diet. Subjects with hypertension for 15 years or more may have perceived Tower 1eveTs of tangib1e support than other subjects for exercise for severa1 reasons. First. if their therapeutic regimen was Tong stand- ing. it may not have inc1uded an exercise program and subjects may have perceived TittTe need for tangib1e support for a moda1ity they did not participate in. Second. subjects may have asked for tangib1e support for exercise in the past and have not received it. The Tonger duration 157 of hypertension for these subjects provided more opportunity for support persons to refuse or tire Of providing tangib1e support for exercise. Due to the Tong-standing nature of the regimen. it may no Tonger be in the forefront Of the patients' awareness. Fina11y. subjects with hypertension 15 years or more may have required more support than other subjects due to starting an exercise program Tater in Tife. QuantituLSuoooLt Subjects responded to 17 questions which attempted to quantify support for foTTowing a therapeutic regimen. The questions addressed how much he1p. concern. reminding. and pressure was received from the support person for foTTowing each moda1ity of the regimen. Mean scores for a11 of the questions were caTcuTated (Tab1e 55F” and found to be simiTar to the midpoint score on the Likert scaTe (3). with one exception. Mean scores for the questions asking how much he1p was received for each modaTity ranged from 3.55 (exercise) to 3.79 (diet). with statisticaTTy significant differences noted between medication and exercise (9, = .05) and diet and exercise (.9 = .01). The reason for these differences is uncTear. OnTy subjects who identified a support person for that particuTar moda1ity answered the question regarding how much he1p was received. It is possib1e that Tess support was perceived for exercise because it is the 1east prescribed of the modaTities. Therefore. the patient and support person may be Tess famiTiar with the moda1ity of exercise and with strategies to impTement and maintain an exercise program. or the patient may be unsure of which activities woqu be supportive. 158 The means noted above are simiTar to the mean score reported by CapTan et a1. (1979) for quantity of tangib1e support by a significant other (3.7). The same Likert scaTe was used in both studies. A common prob1em with the use of Likert-type scaTes is that some respondents characteristicaTTy choose the middTe-range answer (PoTit & HungTer. 1978). With this type Of scaTe. subjects may aTso respond with an answer they fee1 is most socia11y desirab1e rather than ref1ective Of their attitude. With few other research findings tO confirm or refute these mean scores. it cannot be ruTed out that they were affected by the response biases noted above. These biases may aTso expTain why no statisticaTTy significant differences in perceptions were found among the moda1ities regarding "concern." "reminding." or "pressure" from the support person to foTTow the therapeutic regimen. There was. however. a significant difference in perceptions between maTes and femaTes. FemaTes perceived Tess "pressure" to take medication than maTes (j = -2.98. df = 90. p = .004). WhiTe the initiaT impression is that the finding is better for femaTes than maTes. the reverse is true. The mean score of the maTe response was cTosest to the quantity "just about right" on the Likert scaTe. whi1e the mean score Of the femaTes was cTosest to "some to TittTe." This finding may be expTained by the support person each group identified as heTping with medication. MaTes identified spouse much more Often (79%) than femaTes (42%). It is 1ike1y that contact with a spouse woqu be more consistent than that of a friend. unTess the friend was Tiving in the same househOTd. This consistent contact may 159 make the difference in perception that "pressure" to take medication is "just about right." An aTternative expTanation is that femaTes simpTy perceived Tess support as avaiTabTe (62% vs. 71%) than maTes for the moda1ity of medication. Pressure. though intended as a possibTy negative aspect Of support. was evidentTy aTso perceived as Tess avai1ab1e by femaTes than maTes. There were no statisticaTTy significant differences among the experimentaT and controT groups regarding quantity Of support. This is not unexpected since there were no significant differences between the groups in perception of someone ab1e to he1p and type of support. Differences noted regarding support person may not affect the quantity of support perceived. Due to the few significant differences noted among the modaTities and characteristics of condition and sex. no further ana1ysis of data for other samp1e characteristics and quantity of support was done. There wi11. however. be a brief discussion of findings re1ated to two groups Of samp1e characteristics: socio- economic characteristics (iJa. work status. income. occupation. educa- tion) and famiTy structure characteristics (i.e.. maritaT status. 1iving arrangements. number Of chderen. number Of chderen in the home). Findings among the socioeconomic characteristics were simiTar regarding the abiTity to identify someone ab1e to heTp for each modaTity. with one exception. Subjects in the semi-lunskiTTed occupation group perceived significantTy Tess avai1ab1e support than 160 other subjects for the moda1ities Of medication (12 = .01) and diet (2 = .05). This finding may not be as significant as the statistic woqu suggest because the _n's were very smaTT (medication. 11 = 4; diet. 11 = 2). Two socioeconomic characteristic groups perceived significantTy Tess spouse support. Those with incomes under $25.000 perceived Tower TeveTs of spouse support for the moda1ities Of diet and exercise. As noted previousTy. this may be a resu1t Of subjects with Tower incomes being unmarried and having nO spouse support avai1ab1e. Subjects in the semi-lunskiTTed occupation group perceived no spouse support. onTy support by chderen. for the moda1ity of diet. Again the significance is questionabTe because the n = 2. There was one significant difference among the socioeconomic characteristics re1ated to type Of support. Subjects with Tess than a high schooT education perceived onTy tangib1e support (n = 9) for the moda1ity of diet. As discussed in an earTier section. this may be a resu1t of onTy perceiving "heTp" as something tangib1e (ids. purchas- ing or preparing appropriate foods). A11 four of the significant differences in the socioeconomic characteristics invoTved subjects with a Tow socioeconomic status (i.e.. Tow educationa1 TeveT. Tow occupation status. and Tow income). The corre1ation of these findings might be expected since educationa1 TeveT affects the subject's occupation and occupation determines income. It shoqu be remembered. however. that two of the differences invoTved a very smaTT group of subjects. 161 It was aTso interesting to find that three of the four differences were for the moda1ity of diet. The reason for this is not c1ear but may be re1ated to the fact that for subjects in Tow socioeconomic groups basic. much Tess prescribed. dietary needs may not be as easiTy met as for subjects in higher socioeconomic groups. Lack Of money. transportation. and know1edge regarding seTection and preparation of food may Timit the subjects'.abiTity to meet dietary needs. This fact woqu affect the avaiTabiTity Of support. support person. and type of support desired and perceived by subjects with Tow socioeconomic status. A1though there were no statisticaTTy significant differences noted in any of the famiTy structure characteristics (ids. maritaT status. Tiving arrangements. chderen. and chderen Tiving at home). findings re1ated to one group shoqu be noted. Unmarried subjects 1iving with a chde (singTe parents) perceived the 1east avai1ab1e support for each modaTity of the therapeutic regimen. SingTe parents aTso perceived the Towest TeveTs of tangib1e support for diet and exercise. This is TogicaT because spouse was previousTy noted as the support person most Often identified for a11 moda1ities. The absence of this support person in the home woqu decrease avai1abi1ity Of support and may make tangib1e support Tess 1ike1y. Parents with young chderen woqu receive TittTe support from the chderen and may be unabTe to have contact with or have TittTe contact with their most-Often-identified support person (friend) because of caring for the chderen. 162 Limitations Perhaps the greatest Timitation of this thesis was the Tack Of estimated reTiabiTity and va1idity Of the socia1 support instrument. This can be rectified in the future by administration Of the question- naire to other subjects with hypertension or other chronic diseases. AnaTysis Of the findings and refinement of the instrument may increase its reTiabiTity and va1idity. The fact that the socia1 support questionnaire was onTy adminis- tered after the nursing intervention was aTso a Timitation. Adminis- tration of the questionnaire pre- and postintervention to a11 subjects woqu aTTow the researcher to examine any changes in the experimentaT group's perceptions as a resu1t Of the nursing intervention. The threat Of bias due to socia1 desirabiTity which existed in the questions measuring perceptions Of quantity of socia1 support shoqu aTso be viewed as a Timitation. Findings suggested that hypertensive patients perceived "about the right amount" of socia1 support. These findings may be infTuenced by the subject responding with the socia11y desirab1e answer rather than their true perception Of the quantity of the support they receive. Fina11y. the characteristics of the samp1e may be viewed as a Timitation. The samp1e for this thesis varied a great deaT from samp1es of other studies which examined socia1 support and hyperten- sion. This Timits the reader's abiTity to compare resu1ts between this thesis and other studies on a re1ated topic. Findings may have been 163 infTuenced by the smaTT number of unmarried subjects. subjects with Tow socioeconomic status. and nonwhite subjects. Implications ImoJicationfloLNuLsincicactics This study has identified hypertensive patients' perceptions Of socia1 support in foTTowing a therapeutic regimen. When study findings are considered in re1ation to Kingus(1981) concepts concerning nurs- ing. a number Of practicaT impTications for nursing practice emerge. According to King (1981). "nursing is a process of human interactions between nurse and cTient whereby each perceives the other and the situation; and through communication they set goa1 s. expTore means and agree on means to achieve goaTs" (p. 144). The discussion wi11 focus on primary care because the majority of hypertensive patients are diagnosed and treated in primary-care settings. Often. hypertension is discovered whiTe a patient is receiving routine hea1th maintenance or treatment for another hea1th- care prob1em by the primary-care provider. UnTess the patient is in a hypertensive crisis. the primary-care provider is in the best position to monitor and treat the patient. if necessary. The patient is TikeTy to be famiTiar with both the provider and the setting in primary care. and this faciTitates the patient returning for foTTow-up visits. The primary-care provider is famiTiar with the patient's generaT hea1th status and possibTy with some areas Of the patient's psychosociaT status which may affect current B/P TeveTs. If the patient shoqu require referraT to another provider or an 164 acute-care setting. the primary-care provider can coordinate generaT hea1th care and foTTow-Up. The nurse in advanced practice provides hea1th care to hyperten- sive patients in primary-care settings as noted above. She/he func- tions as an assessor. practitioner. counseTor. educator. evaTuator. advocate. and coordinator of care in the roTe of primary provider. Nurses educated at both the basic and graduate TeveTs may find the study findings usefuT as they interact with hypertensive patients. A1though both groups of nurses wi11 use the nursing process in these interactions. the assessment. goaTs. strategies for impTementation. and evaTuation Of outcomes of the nurse in advanced practice wi11 be more compTex and sophisticated than those Of baccaTaureate—prepared nurses. The increased compTexity and sophistication is a resu1t of (a) appTica- tion of theories (ids. systems. famiTy. roTe. educationa1) and con- cepts (ids. socia1 support. grief. Toss) which broaden the theoreticaT know1edge base and (b) didactic and practicum courses re1ated to a cTinicaT area which increase the TeveT of cTinicaT judgment. In the first step Of the nursing process. assessment. nurses at both TeveTs of practice shoqu expTore the patients' perceptions Of avaiTabiTity and desirabiTity of socia1 support for each aspect Of the therapeutic regimen. It is important to assess each aspect Of the regimen individuaTTy. ResuTts of the study have shown there were significant differences in the avaiTabiTity of support among the moda1ities. Hypertensive patients were more ab1e to identify support for the modaTities of diet and medication than exercise. Therefore. 165 the nurse shoqu determine if exercise is a part of the patientfls therapeutic regimen. Patients may not consider the physician teTTing them to go out and waTk prescribed exercise. If the hypertensive patient perceives support for a moda1ity. the nurse shoqu then assess who the support person/persons are. what type of support they provide. and whether or not the patient is satisfied with the support. The patientfls perceptions are essentiaT to the deveTopment Of appropriate goaTs and strategies for managing hyperten- sion. King (1981) re1ated that perceptions are "each individuaTks representation or image of reaTity; an awareness of Objects. persons and events" (p. 20). Assessment of socia1 support by the nurse in basic practice coqu be brief. focusing on the most supportive person for each aspect of the regimen. The assessment by the nurse in advanced practice coqu be expanded to examine the totaT support net- work reTated to type. quaTity. and quantity of support. The identified support person shoqu aTso be assessed to determine her/his know1edge of socia1 support and wiTTingness to provide support. The nurse shoqu reaTize that factors in the initiaT assessment may change over time and that the infT uence of other stressors may override the patient's con- cern regarding hypertension. GoaTs. set in the pTanning stage Of the nursing process. wiTT be based on information from the totaT assessment and may re1ate to an aspect of the regimen di rectTy (i.e.. Tose 3 pounds in 1 week) or indirect1y (i.e.. identify support person for diet). Therefore. the goaTs determined by the baccaTaureate nurse and hypertensive patient. 166 re1ated to socia1 support. may invoTve choosing or obtaining a support person. MutuaT goa1 setting between the nurse in advanced practice and the cTient woqu ref1ect the more detaiTed assessment and coqu invoTve goaTs re1ated to obtaining support. modifying the type Of support. or expanding the existing support network. Study findings may aTso be he1pfu1 to nurses as they deveTop strategies with hypertensive patients and impTement the pTan to meet the goa1. Awareness by the nurse that spouse and friend were most often identified as a support person for a11 moda1ities wiTT guide the nurse in suggesting possib1e support persons for patients who did not perceive support. The baccaTaureate nurse coqu use this information tO assist the patient in determining whom to ask for support. The nurse in advanced practice may go beyond the above strategy and assist the patient in deciding how they coqu ask for support. Nurses aTso need to reaTize that perceptions of who can be supportive may be broadened. as with the experimentaT group. to inc1ude chderen. other reTatives. physicians. nurses. other hea1th professionaTs. and even pets. ‘This may be accompTished through the introduction of aTternative sources Of support when the patient perceives a spouse or friend as nonsupportive. ResuTts of this thesis support the findings Of Kirscht et a1. (1981) that different strategies or interventions are required for foTTowing different components of the therapeutic regimen. Study findings reveaTed that there were significant differences among the modaTities for the type Of support and quantity Of support perceived. 167 Hypertensive patients perceived much more psychoTogicaT support for the modaTity Of medication than for diet or exercise. ConverseTy. they perceived more tangib1e support for diet and exercise than for medication. KnowTedge of types of support perceived by patients for each aspect of the regimen wiTT aTTOw the nurse and cTient to expTore strategies which use the type of support most often identified by other hypertensive patients for each moda1ity. ‘This may increase the chance for success in meeting the goa1 set by the patient and nurse and encourage the patient to attempt other strategies invoTving socia1 support. There were aTso significant differences in perception of socia1 support based on sociodemographic characteristics of sex. income. occupation. education. and duration of hypertension. FemaTes identi- fied Tess support avaiTabTe than men for medication and diet and more support avaiTabTe than men for exercise. Men were much more TikeTy to identify spouse as a support person for aTT moda1ities than women. Women perceived high TeveTs of psychoTogicaT support for medication and high TeveTs Of tangib1e support for exercise. WhiTe men perceived the same type of support as femaTes. the TeveTs were much Tower and were significantTy different for medication. Subjects with a Tow socioeconomic status (ids. Tow income. Tow occupationaT status. and Tow educationa1 1eve1) perceived Tess avai1ab1e support for the modaTity of diet than other subjects. Newly diagnosed hypertensive patients (Tess than 1 year duration) perceived 168 Tess support avai1ab1e for exercise than medication or diet and higher 1eveTs of tangib1e support for medication than other subjects. Patients with hypertension of 15 years or more duration perceived the Towest 1eve1 of tangib1e support for medication. A1though not statis- ticaTTy significant. a practicaTTy significant finding re1ated to singTe parents was found. SingTe parents perceived Tess avaiTabTe support for each modaTity of the regimen than any other group Of sub- jects. The nurse must be aware of the differences in patients! percep- tions as she/he records the patients' perceptions and deveTOps her/his own perceptions of the patients' support. Findings noted above can guide the nurse in deveToping goaTs and strategies with the patient that are most appropriate for each modaTity and the characteristics of the patient. For exampTe. the nurse in advanced practice may have a femaTe patient who cannot identify support for the moda1ity of exercise. In assessing socia1 support for this patient. the nurse discovers that the patient was ton to "waTk" as part of her therapeutic regimen. The nurse and cTient decide on a goa1: the patient wiTT waTk 20 minutes a day. three times a week. ‘The nurse reTates to the patient that other femaTe hypertensive patients have found spouse and friend equaTTy supportive for the modaTity of exercise. The nurse and patient aTso discuss that tangib1e support is sometimes more he1pfu1 than psycho- TogicaT support for exercise. especia11y among women. Based on this information. the patient chooses to ask a friend to waTk with her as a strategy to meet her goa1. The above strategy coqu be further 169 adjusted for the femaTe patient if she was a singTe parent. recentTy diagnosed as hypertensive. or from a Tow socioeconomic group. The baccaTaureate nurse coqu aTso use the know1edge of differ- ences re1ated to the moda1ities and sociodemographic characteristics as a guide in deveToping goaTs and strategies with the hypertensive patient. The strategies. as were the assessment and goaTs. woqu be Tess compTex than those used by the nurse in advanced practice. For exampTe. the baccaTaureate nurse may present information regarding differences of support person for women and exercise as a strategy to assist the patient in choosing a support person for exercise. The chance of patient success in meeting the goa1 is increased by using strategies that have been successfuT for other hypertensive patients with simiTar characteristics in simiTar situations. A positive inter- action with a support person wiTT aTso encourage the patient to con- tinue to use socia1 support in strategies to foTTow the therapeutic regimen. The nurse in advanced practice is often responsibTe for the c1ini- ca1 management and evaTuation of treatment Of newTy diagnosed hyperten- sive patients. In many instances. the nurse and cTient focus on non- pharmacoTogic aspects Of the regimen before trying medication. Study findings re1ated to type Of support and support persons for the modaTi- ties Of diet and exercise can direct the nurse and patient in choosing strategies which wi11 be successfuT. Success with dietary restrictions and exercise coqu e1iminate the need for medication for some hyperten- sive patients. The nurse in advanced practice is abTe to make such 170 decisions regarding the therapeutic regimen based on her/his advanced c1inica1 judgment. Using study findings. the nurse in advanced prac- tice coqu aTso recommend or organize sUpport groups for hypertensive patients with simiTar characteristics who face simiTar difficuTties in identifying and using sociaT support to foTTow their therapeutic regimen. The interventions discussed above are most appropriate for primary-care settings. However. nurses in acute-care settings coqu aTso use study findings whi1e conducting initiaT assessments of newTy diagnosed or uncontroTTed hypertensive patients. The assessment reTated to socia1 support shoqu inc1ude perceptions of avaiTabiTity of support and identification of a support person for each moda1ity of the regimen. The information from the assessment coqu then be given to the nurse educator or the person responsibTe for in-patient and out- patient education regarding hypertension and strategies for its management. Eva1uation of the success of the strategies is based on the degree to which short-term (ids. patient wiTT Tose 3 pounds. patient wiTT identify a support person for exercise) and Tong-term goaTs (1J%. patient's individua1 B/P goa1. patient's weight goa1) are met. If the goa1 was accompTished. the strategy was successfuT and a new goa1 may be set and strategies deveToped. If the goa1 was tota11y or partiaTTy unmet. the appropriateness of both the goa1 and strategies shoqu be evaTuated and adjusted. Determination of aTternative goaTs and strategies may inc1ude evaTuation of success by the support person. 171 To return to the exampTe of the nurse in advanced practice and the femaTe hypertensive patient. suppose the patient did not meet the goa1 of waTking 20 minutes three times per week. First the patient and nurse woqu evaTuate the goa1 to determine if it was reaTistic. In this case. the patient and nurse decide the goa1 is reaTistic. Next. they examine the strategy. a friend to waTk with the patient; The patient reTates that the friend was onTy ab1e to go waTking with her once. The other two times the patient caTTed her friend the friend was busy and the patient did not Tike to waTk aTone. Since the patient wOUTd stiTT Tike to waTk with the friend. the nurse suggests setting a specific time aside on specific days that are acceptabTe to both the patient and her friend. The patient decides that further specification of the strategy wiTT be he1pfu1 and agrees to contact her friend to set up times to waTk. The modified strategy woqu be evaTuated in the same manner at the next visit. ImincationsioLNuLsinLEducation WhiTe most nursing education programs address the pathophysioTogy and recommended treatment of hypertension. few deaT with the major prob1em Of hypertension management--assisting the patient to foTTow his/her therapeutic regimen. Yet as hea1th-care practitioners. many nurses deaT with this prob1em on a daiTy basis. Previous research has documented a positive re1ationship between socia1 support and foTTowing a therapeutic regimen for hypertension (Cap1an et a1.. 1979; Earp et a1.. 1982; Haynes et a1.. 1976; Kirscht et a1.. 1981; McKenney et 81.. 172 1973). Findings of these investigators woqu suggest that the concept of socia1 support shoqu be inc1uded in the curricuTums of basic. graduate. continuing. and inservice nursing education. ResuTts of this study impTy that it is important for nurses to assess the hypertensive patientis perception of socia1 support if they are to assist the patient in using socia1 support to foTTow the thera- peutic regimen. The abiTity to do so depends on the nurse's awareness of the importance of perceptions in socia1 support. Nurses must under- stand that a patient's perception of socia1 support may be very dif- ferent from those of the nurse or support person. and their decisions to foTTow the therapeutic regimen are based on that perception. Nurses must aTso reaTize that patients' perceptions Of socia1 support are constantTy changing through interaction with the nurse and other sup- port persons. KnowTedge of the socia1 support interaction can be obtained if systems theory and Kingks(1981)‘theory Of nursing are inc1uded in the curricuTum. The resu1ts of this study found differences in hypertensive patients' perceptions of socia1 support among the modaTities of the therapeutic regimen and subgroups of the samp1e. The findings woqu impTy that further research is needed at the graduate 1eve1 of nursing education to expand and add to nursing's know1edge base regarding socia1 support in generaT and as it reTates more specificaTTy to patients with hypertension. Courses in research methods. theory of socia1 support. and c1inica1 management and evaTuation of treatment of hypertension woqu be appropriate at the graduate 1eve1. 173 ImplicationsJoLBeseaLob WhiTe investigators have expTored the effect of socia1 support on foTTowing a therapeutic regimen for hypertension (Cap1an et a1.. 1979; Earp et a1.. 1982; Haynes et a1..1976;Kirscht et a1..1981;McKenney et a1.. 1973; Morisky et a1.. 1982; NeTson et a1.. 1978). no other study has described patients' perceptions of socia1 support in foTTowing a therapeutic regimen for hypertension. ‘The present study may therefore be viewed as adding another dimension to the present know1edge regarding socia1 support. Findings of the study may aTso be used as a base for future research. ResuTts re1ated to differences between maTes and femaTes. patients with hypertension Of various durations. patients with Tow socioeconomic status. and unmarried patients (especia11y those with chderen at home) suggest that further research is needed of these subgroups of the samp1e. In future research. the 1imitations of the present study shoqu be considered. For exampTe. administration Of the socia1 support ques- tionnaire before and after a socia1 support intervention woqu aTTow examination of the differences or changes in perceptions of socia1 support of the experimentaT group. VaTidity and reTiabiTity of the questionnaire coqu be estimated by administration Of the questionnaire to severa1 groups of hyperten- sive patients or patients with other chronic diseases. The question- naire coqu aTso be administered at the same time as other instruments which measure socia1 support. CorreTation Of the resu1ts of this study 174 with other questionnaires woqu indicate the Socia1 Interaction Ques- tionnaire was indeed tapping the concept of socia1 support. The study coqu be repTicated using a samp1e with different sociodemographic characteristics: races other than white. Tower income 1eveTs. Tower educationa1 TeveTs. unmarried. no upper age Timit. or a11 of one sex. ResuTts coqu be compared to the findings of this study and document the differences noted. SimiTar resu1ts with different samp1es woqu make the findings more generaTizabTe to the Targer popu1ation of hypertensive patients. Fina11y. the threat of response bias due to socia1 desirabiTity coqu be decreased by vaTidating the present questions with simiTar questions worded in a more suthe. indirect manner. Both negative and positive statements regarding quantity of support coqu be presented (14%. "You can aTways depend on someone to he1p you take your medicine)‘ "No one wiTT remind you when rvs time to take your medicineJU. The patient woqu then be asked to record the extent to which he/she agrees or disagrees with the statement. If a patient recorded perceptions of socia1 support for reminding as "just about right? to take medications and disagreed with the statement that "no one woqu remind them to take medication." the chance of bias is smaTT. If. however. a patient perceived support as "just about right" but agreed with the statement that no one wOUTd remind them to take medication. the chance of bias is great and the initiaT questions shoqu be revised. 175 As discussed in ear1ier chapters. understanding the process Of socia1 support is centraT to the c1ear conceptua1 and operationaT definitions of the concept. SeveraT authors have deveToped modeTs of how socia1 support affects adaptation to stress (Dimond & Jones. 1973; PearTin et a1.. 1981; Thoits. 1982). ‘This researcher beTieves the most promising modeT is that proposed by Dimond and Jones (1983) because the modeT expTains both main and interactive effects of the components Of sociaT support. (See Figure 6.1. A proposed modeT for socia1 support and adaptation to stress. Dimond & Jones. 1983J The key eTements in the abovementioned modeT aree(a) the nature of the stressor. (b) characteristics of the support network. (c) the nature Of support Offered. (d) the perceived adequacy of support. (e) adaptive responses to stress. and (f) environmentaT resources. The nature of the stressor denotes the type of stress (iat. crisis. tran- sition. or deficit) as we11 as whether the stressor was expected or unexpected. voTuntary or invoTuntary. temporary or permanent. As viewed by the hea1th-care provider. hypertension is an unexpected. invoTuntary. stressor which may be a permanent hea1th deficit. ‘The patient's perception of hypertension as a stressor is di rectTy infTu- enced by environmentaT resources. EnvironmentaT resources are noted in four areas by Dimond and Jones (1983). The first are resources of the individua1 (i.e.. age. sex. deveTopmentaT stage. personaTity. and hea1th). Next are resources of the physica1 environment (int. 1iving situation. sanitation. Tocation of dweTTing. number of peop1e in dweTTing). ‘The third area. 176 ENVIRONMENTAL RESOURCES (Individual, Physical, Personal, Sociocultural) NATURE OF STRESSOR Figure 6.1: (GO TO QUES. l6) l5. How many cigarettes do you smoke in a day? (CHECK ONE) 1. Less than five cigarettes a day ___ 2. Six to nine cigarettes a day ___ 3. Ten to nineteen cigarettes a day ____ 4. Twenty to twenty-nine cigarettes a day ___ 5. Thirty or more cigarettes a day ___ l6. Do you drink alcoholic beverages? (CHECK ONE) 1. st __ 2. No __->(GO T0 QUES. l8) l7. How often do you drink alcoholic beverages? (CHECK ONE) l. Occasionally ___ 2. Heekends only ___ 3. Several times a week ___ 4. One to two drinks a day ___ 5. More than two drinks a day ___ l8. Do you have diabetes? (CHECK ONE) 1. tfs __ 2. No __¢—>(GO TO END OF QUESTIONNAIRE) l9. How long have you had diabetes? (CHECK ONE) l. Less than one year ___ 2. One to two years-___ 3. Three to five years ___ 4. Six to eight years ___ 5. Nine to eleven years ___ 6. Twelve to fourteen years ___ 7. Fifteen years or more ____ END: You have completed this part of the questionnaire. Please begin answering the next section. (43 (49 (5 ) END OF CARD 02 APPENDIX D SOCIAL INTERACTION QUESTIONNAIRE I92 193 I.D. # Packet é SOCIAL INTERACTION THIS IS A QUESTIONNAIRE YOU HAVE NOT COMPLETED AT PREVIOUS INTERVIEWS. THE QUESTIONS RELATE T0 PEOPLE AROUND YOU WHO COULD HELP 0R ENCOURAGE YOU TO TAKE ON NEW HABITS 0R SOLVE PROBLEMS RELATED TO YOUR HIGH BLOOD PRESSURE. PLEASE ANSWER THE QUESTIONS TO THE BEST OF YOUR ABILITY. . PLEASE READ THE DIRECTIONS CAREFULLY. YOU WILL N91 BE ANSWERING ALL OF THE QUESTIONS. FOR EXAMPLE, IF YOU ANSWER QUESTION NO. 2 YES. YOU WILL ANSWER ALL THE QUESTIONS ON PAGE I AND THEN SKIP TO PAGE 3. QUESTION 4. IF YOU ANSWER QUESTION N0. 2 NO, YOU HILL SKIP TO PAGE 2 AND ANSWER QUESTION 3. THE DIRECTIONS AT THE END OF EACH QUESTION TELL YOU WHICH PAGE AND QUESTION TO GO TO NEXT. THOUGH IT MAY SOUND CONFUSING, THE QUESTIONNAIRE IS QUICK AND SIMPLE T0 COMPLETE IF YOU READ AND FOLLOW THE DIRECTIONS CAREFULLY. 19A 1. Many people find it easier to take on new habits or solve problems when others help them. Please write in the name and relationship (for example. husband, friend) of people who could help you to take on new habits or solve problems in following your treatment plan. NAME RELATIONSHIP 2. Have any of the peOple you just listed been able to help or support you in taking your pills? (CHECK ONE) _ Yes _ N0 + e : [fiOTOP.2.Q.3I 2a. who has been able to help or support you in taking your pills? (LIST THE PERSONS‘ NAME AND RELATIONSHIP TO YOU.) i NAME RELATIONSHIP i ! 26. Who has been the most supportive in helping you take your pills? (HRITE IN) h.) to M ‘4. . How does he/she help you? (HRITE IN) ' . How much does this person help yOu to remember to take your pills? (CHECK ONE) ___ A great deal ____A little ___ A lot ___ Not at all ____Some . Overall, how much concern has this person shown about whether or not you take y0ur pills? (CHECK ONE) Far too much Somewhat too little SOmeha: too much Far too little About the right amount . How much does this person remind you to take your pills? (CHECK ONE) Far too much Somewhat too little Somewhat too much Far too little Aaeut the right amount F50 TO P. 3. 0. fl ——--—-— _— — —--——u¢——H —.~-... I95 3. Would you like to have someone who is close to you help you (for example, remind or encourage you) to take your pills? (CHECK ONE) _ Yes _ No o + + 3a. Have you ever asked or indicated to anyone that you would like them + to help (remind or encourage) you? (CHECK ONE) + _ Yes _ No I + + ‘r + 3b. What happened when you asked for help? (WRITE IN) i I i I ‘ 6 mo TO 9. 3. q. 4i + i I + 3c. Why haven't you? (WRITE IN) : i I 4 IGO TO p. 3, 0.41 i -..: 3d. Why wouldn't you? (WRITE IN) (GO TO P. 4, 0. ST 196 4. Now think about the people whom you identified as being able to help or support you. Have any of these people been able to help you with your diet? (CHECK ONE) _ Yes _ No 4 + ‘ FED—Tm + . , . 4 4a. Who has been able to help you with your diet? (LIST NAMES AND THE PERSONS' RELATIONSHIP TO YOU.) NAME RELATIONSHIP 4b. Who has been the most supportive in helping you follow your diet? (WRITE IN) 4c. How does he/she help you? (WRITE IN) 4d. How much does this person help you to remember to follow your diet? (CHECK ONE) ___ A great deal ____A little _A lot _Not at all __ Sone 4e. Overall, how much concern has this person shown about whether or not you follow yOur diet? (CHECK ONE) ___ Far too much ___ Somewhat too little ____Somewhat too much ____Far too little ____About the right amount 4f. And how much concern has this person shown about whether or not you achieve weight control? (CHECK ONE) ____Far too much ____Somewhat too littl- ___ Scmewhat too much ___ Far too little _ About the right ambunt 4;. How much does this person remind you to follow your diet? (CHECK ONE) ___ Far too much ____Somewhat too little Somewhat too much ____Far too little About the right amount FCC TO 3. 5. O. 5 I 197 Z O 5. WOuld you like to have someone who is close to you to help you (for example, remind or encourage you) to follow your diet? (CHECK ONE) ___IYes 4 5a. Have you ever asked or indicated to someone that you w0uld like them to help (remind or encourage) you? (CHECK ONE) _ Yes _ No i + I + I Sb. What happened when you asked for help? (WRITE IN) + . + I i ! i I60 TO P. 5,40. 6 I I 6 + 5c. Why haven't you? (WRITE IN) 160 TO P. 5, Q. 6| +++++++++++++++++4-4-o—4- “'I . Why wOuldn't yOU? (WRITE IN) 160 TO P. 5, O. SCI I98 6. Again, think about the people whom you previously identified as being able to help or support you. Have any of these people been able to help you with exercise? (CHECK ONE) _Yes __No 4 + : r0070p.5,rL7) i 6a. Who has been able to help you with your exercise? (LIST NAMES AND PERSONS' RELATIONSHIP TO YOU.) NAME 6b. Who has been most supportive in helping you with exercise? (WRITE IN) 6c. How does he/she help you? (WRITE IN) 6d. How much does this person help you to remember to follow your exercise program? (CHECK ONE) _ A great deal _ A little 2 ___ A l°t __y Not at all i ____Some 6e. Overall. how much concern has this person shown about whether or not you follow yOur exercise program? (CHECK ONE) ____Far too much ___ Somewhat too little ___ Somewhat too much __ Far too little ___ About the right amount ' l 6f. And how much concern has this person shown about whether or not you I establisn a regular exercise pattern? (CHECK ONE) g I '___ Far too much .___ Somewhat too little I I ___ Somewhat too much ___ Far too little ' ; ___ AbOut tne right amount I . | 69. How much does this person remind you to follow your exercise program? i I (CHECK ONE) l Ear too much ____Somewhat too little : . Somewnat too much ___ Far too little About the right amount I .L . .‘. 199 7. Would you like to have someone who is close to y0u help you (for example, remind or encourage yOu) with exercise? (CHECK ONE) _ Yes N0 + i 7a. Have you ever asked or indicated to someone that you would like I them to help (remind or encourage) you? (CHECK ONE) 6 6 ___ Yes ____No ‘ I I 7b. What happened when you asked for help? (WRITE IN) + i f I I + 150m mfg). 8L . f I + 7:. Why haven't you? (WRITE IN) : + + IGOTOP.7,Q.81 : + 7d. Why w0uldn‘t you? (WRITE IN) TIGO TC P. 7, Q. 8,1 200 8. How much do y0ur family and/or friends make you feel worthwhile and good about y0urself? (CHECK ONE) ___ A great deal ____A little _ A lot _ Not at all ____Some 9. How much do your family and/or friends show concern about your hypertensive control? (CHECK ONE) ___ A great deal ___ A little _ A lot _ Not at all '___ Some 10. How much do ygg_have to rely on yourself to take care of your hypertension? (CHECK ONE) ___ Far too much .___ Somewhat too little _ Somewhat too much __ Far too little ____About the right amount 11. How much do you feel pressured by others to take your pills? (CHECK ONE) ____Far too much ____Somewhat too little ___ Somewhat too much .___ Far too little ___ About the right amount 12. How much do you feel preSSured by others to follow your diet? (CHECK ONE) ___ Far too much .___ Somewhat too little ___ Somewhat too much ‘___ Far too little ___.About the right amount 13. How much do you feel pressured by others to exercise? (CHECK ONE) ____Far too much -—-— ____Somewhat too little ___ Somewhat too much ___:Far too little ____About the right amount 14. Are there people. for example, family members or friends, who have not helped you with your treatment program §UI_whom you would like to help you with it?, (CHECK ONE) "' " Y S No cu..- J A a 9 o 1: a-a—c-o—ID I l 14 “‘ 3r a. Who are these peop+e? (LIST NAME AND THE PERSONS' RELATIONSHIP TO YOU.) 1 NAME RELATIONSHIP [GO TO U} 3. 0. Idol 20] QUESTION 14 CONTINUED 14b. How would you plan with these people a way that they can best help you take your medicines for hypertension? (WRITE IN) 14c. How would you plan with them a way that they can best help you follow yOur diet for hypertension? (WRITE IN) 14d. How would you plan with them a way that they can best help you to exercise? (WRITE IN) Ide. Will it be hard for you to ask for help? (CHECK ONE) ___ Yes N0 + + l -o it . 15 v I + d 14f. Why will it be hard for you to ask for help? i I H I 15. Besides people, what things-ohobbies, pastimes. habits, or personal activi- ties-~help yOu in dealing with your treatment program for hypertension? (WRITE IN) 16. Do you think a nurse could support you to Successfully follow your treatment program for high blooo preSSure?- _Yes __No ‘ u i‘ . , . C‘ 6 16a. How so you think sne could do this? 202 I l I I I 166. Why don't you think a nurse could support you? QUESTION 16 CONTINUED (WRITE IN) END: You have completed this part of the questionnaire. the UNDERSTANDING HIGH BLOOD PRESSURE section. Please begin answering APPENDIX E NINE-MONTH SOCIODEMOGRAPHIC INSTRUMENT 203 201i SOCID-DENOGRAPHIC The following questions describe general things all the questions to the best of your ability. l. How old were you on your last birthday? (WRITE IN) 2. What is your marital status? l. Married 2. Single, figver married __ 3. Separated __ 4. Divorced __ 5. Widowed __ (CHECK ONE) Penin Card A9 3. How many living children do y0u have, includino adopted and stepchildren? No living children (CHECK) Number of living children (HHITE IN) 4. Taking all scurces of money into consideration, what was yOur family's total income before taxes and other deductions for the past )2 months? (CHECK ONE) 00. Below $5,000 OI. $5.000-$6,999 02. S7,DCO-SB,999 03. $9,000-SIO,999 O4. SlI,DOO-SIZ,999 OS. 06. O7. 08. 09. Site TT) Pt. 1.0. ___—(TU_—_ -1 ‘— Fomifl Type (12-l3) (la) Card No. A 9 Date __ __ ___ (TS-TE) (l7-27) about you. Please answer years __ (E 4) (ES) (EC-'7) Sl3,000-Sl4,999 ‘___ Sl5,000-Sl6.999 ___ Sl7,000-SIP,999 ____ 520,000-524,999 ___ $25,000 or over ___ __ __ ( 8- 9) 5. Are you working now at a regular job, unemployed, retired, a housewife, or what? l. Working now at regular job I [GO TO PAGE 2, ms. 6] (CHECK ONE) I I I I | I I I I I I I I I I I I I I I I U I I I I I I I I I I I I J Unemployed or laid off __ Retired ___ Disabled __ Housewife __ Other (Specify) 205 What is the main occupation you work at? (What type of work do you do?) (WRITE IN) What kind of business or industry is that in? (what do they make or do?) Is it your own business (NRITE 1N) Nho lives in your household, besides y0urself? (CHECK AS MANY AS APPLY) No one else __ Husband/wife Children (Write in number livino at home) __ Other relatives (write in relationships: example, mother-in law, niece) e. Non-related persons (write in: example, 2 friends; l boarder) QOU'W Do you smoke cigarettes? (CHECK ONE) l. Yes 2. No -)(GO TO QUES. ll) . how many cigarettes do you smoke in a day? (CHECK 03E) 1 Less than five cigarettes a day 2 Six to nine cigarettes a day ___ 3. Ten to nineteen cigarettes a day ___ 4 Twenty to twenty-nine cigarettes a day ___ 5 Thirty or more cigarettes a day . Do you drink alcoholic beverages? (CHECK ONE) l. Yes 2. No __ -—)(GO TO END OF QUESTIOHOAIRE) . How often do you drink alconolic beverages? (CHECK ONE) Occasionally __ Weekends only ___ Several times a week __ One to two drinks a day __ More than two drinks a day U'Ibwmd - o o o 0 You have completed this part of the Questionnaire. Please begin answering the BELIEFS Ayoul HXGw £1033 PRESSURE sectiOn. A wl \J b) b) w I V Ihl V A U) ml n ml V A b) \J v A Lul (.0 v E30 OF CARD A9 REFERENCES 206 REFERENCES Andrews. 6.. Tennant. C.. Hewson. D. M. et al. (1978). Life event. stress. social support. coping style. and risk of psychological impairment. lhe Jeumal .ef. Memes end Mental Diseases. M. 307-316. Antonovsky. A. (1979). Health. st:ess..and.egping. San Francisco: Jossey Bass Pub. Co. Atchley;. R. (1979)..HidQHhQQfi.and.netinemenl.ei.£eleiixel¥.Einixielfl ljfe.eyents. Presentation at the Institute of Human Development. University of California. Berkeley. Baile. H.. & Gross. R. (1979). Hypertension: Psychosomatic and behav- ioral aspects. Erlmerx.£e££. 5. 267-283. Barrera. Mr. Jr.. & Ainlay. S. (1983). The structure of social support: A conceptual and empirical analysis. Journal 91 Community Egy- W: .11. 133-144. Berkman. A.. & Syme. B. (1979). Social networks. host resistance. and mortality: A follow-up study of Alameda County residents. American Jeumal ef EpidemJeJegx. 109. 2. 186-204. Billings. A.. & Moos. R. (198l). The role of coping responses and social resources in attenuating the impact of stressful life events. leumel ef. Bebexlenal Medicine. A. 139-157. Blackwell. B. (1976). Treatment adherence in hypertension. Amenican laymal 9i Bhazmaex. JAB. 75-85. Brandt. P.. & Weinert. C. (1981). The PRO: A social support measure. NHLSJLDQ Beseaeeb. 30. 5.. 277-280. Bredemeir. H.. & Stephenson. R. (1962). Ihe analysis of seelal §¥§I§m§. New York: Holt. Rinehart & Winston. Briggs. H.. Lowenthal. D.. Cirksena. W. et al. (1975). Propranolol in hypertensive dialysis patients: Efficacy and compliance. Clinical Eharmaeeleeieal IheLapy. .15. 606-612. 207 208 Broadhead. W.. Kap1an. B.. James. S. et al. (1983). The epidemiologic evidence for a relationship between social support and health. Ameriean ldutnal et Epjeeminlegy. 1.11. 521-537. Brown. E.. & Bloom. J. (1978). The nurse practitioner in hypertension control program..Nnnse.E£e£I1I19ne£. 3. 10. Brown. 6.. Bhrolchain. M.. & Harris. T. (1975). Social class and psy- chiatric disturbance among women in an urban population. 53.19.1931. 2: 225-254 . Caldwell. J.. Cobb. 3.. Dowling. M. et al. (1970). The drop-out problems in anti-hypertensive treatment: A pilot study of social and emotional factors influencing a patient's ability to follow anti-hypertensive treatment..leunnel.QI.then1§.Dlseese. 22. 579-592. Caplan. R.. Robinson. E.. French. J. et al. (1976). Afiheming.te medical Legimenszfluetezseeeimentsjneatientedueatienandsmialsup: .RQLI. Ann Arbor. MI: Research Center for Group Dynamics. Institute for Social Research. Caplan. R.. Van Harrison. R.. Wellons. R. et al. (1979). 599131 sup- eectandeatientadneceneetfizeeeimentalandsumxflndings. Ann Arbor: University of Michigan Press. Cassel. J. (1974). An epidemiological perspective of psychosocial fac- tors in disease etiology. American lenmal ei Eublie Health. 64. ll. 1040-1043. Cobb. S. (1976). Social support as a moderator of life stress. Esychg; somatic Medicine. 3.8. 300-312. Daniels. L.. & Kochar. M. (1979). What influences adherence to hypertension therapy..Munsing.EQ£um..lB. 231. Daniels. L.. & Kochar. M. (1980). Monitoring and facilitating adherence to hypertension therapeutic regimens. Candigzlasenlan Nursing. 16. 2' 7-120 Davis. M.. & Eichorn. R. (1963). Compliance with medical regimens: A Panel study. Jeucnal e1 Health and Human Benetton. A. 240-249. Dean. A.. & Lin. N. (1977). The stress-buffering role of social support. lhe Jeutnal ef. Memes and Mental Disease. 1.65. 403-417. DeAraujo. G.. Dudley. 0.. & VanArsdal. P. (1972). Psychosocial assets and severity of chronic asthma. Ihe.leunnel.QI.Alle£Q¥.end.Qlin1: 9.1]. We 50: 5.: 259-263. 209 DiMatteo. M.. & Hays. R. (1981). Social support and serious illness. In B. Gottlieb (Ed.). Seeialnetuenks andseeialsepeent. Beverly Hills: Sage Publications. Dimond. M.. & Jones. S. (1983). Social support: A review and theoreti- cal integration. In P. Chinn (Ed.). AdiLanees in naming theecx fieyelgpment. Rockville. MD: Aspen Systems Corp. Durkheim. E. (1951). Suicide. Glencoe. IL: Free Press. Earp. J.. Ory. M.. & Strogatz. D. (1982). The effects of family involvement and practitioner home visits on the control of hyper- tension. Ametiean.leu£nal.ei.Eublie.flealth. 72. 10. 1146-1154. Ell. K. (1984. March). Social networks. social support. and health status: A review. Seeial.§e£xlse.fiex1eu. pp. 133-149. Enget. 8.. Glasgow. M.. & Garader. K. (l983). Behavioral treatment of high blood pressure: III. Follow-up results and treatment recom- mendations. Eseenesematic Medicine. as. 23-29. Giblin. E. (1978). Controlling high blood pressure. American.leutnal.ei Musing. 15. 824-828. Gierszewski. S. (1983). The relationship of weight loss. locus of control. and social support. Munsing.8esee£9h. 32. 1. 43-47. Given. 8.. & Given. C. W. (1982). Eatient contributions 19 ea:e--Link 19.2Leeess,and,gutegme. Unpublished research. Michigan State Uni- versity. Given. C. W.. Given. 8.. Gallin. R. et al. (1979). Emotignal stress and xague templaints .in pennant eaLe. Unpublished research pro- posal. Michigan State University. Goplerud. E. (1980). Social support and stress during the first year of graduate school. Enetessienal.Esxeheleg¥. 11. 283-290. Gore. S. (1978). The effect of social support in moderating the health consequences of unemployment. Journal 91 Health and Social Benayiec. 19.. 157-165. Gottlieb. B. (1981). Social networks and social support in community health. In Socia1.netwgnks and social support. Beverly Hills: Sage Publications. Grancio. S. (1980). Components of community control programs. In fixpectensjentlneseepeeitheemblemandthemleettbenunse. ,Massaghusetts.Nurse/M.N.A. Bulletin. 210 Grim. C. M.. & Grim. C. E. (1981). The nurse's role in hypertension control. Eamily.and.Communit¥.Health. 14. 1. 29-40. Guidelinesionedueatingnursesinhiehhleedenessuceeenttol. (1980). NIH Publ. no. 80-1241. Bethesda. MD: National High Blood Pressure Education Program. National Heart. Lung. & Blood Institute. Haines. C.. & Ward. G. (1981). Recent trends in public knowledge. atti- tudes. and reported behavior with respect to high blood pressure. m m BM! 96: Q: 514-522. Hall. A.. & Fagan. R. (1968). Definition of a system. In W. Buckley (Ed.).M9_deLns1stemsteseatchtecthebehafleLalse1entist. Chicago: Aldine Publishing Co. Hall. J.. & Weaver. B. (1977). DJstflhuthemcsingoLaetieetAsxstem approach.to.eommun1ty health. Philadelphia: J. B. Lippincott Corp. Hart. 5.. & Herriot. P. (1977). Defining systems theory for nursing practice. In J. Hall & B. Weaver (Eds.). Distnihutiye muzsing - A 5151311) Web to eommunitx health. Philadelphia: J. B. Lippincott Corp. Haviland. E. (1983). lheLeJationshiobetueeneeneeJiedseoJalstLes: sonsandhaeeiecstodietarxeemeuaneeaAstudxeimideezaoed hypertensiye women. Unpublished master's thesis. Michigan State University. Haynes. R.. Gibson. E.. Hackett. B. et a1. (1976). Improvement of medi- cation compliance in uncontrolled hypertension. Lancet. 1265-1268. Hazzard. M. (1971). An overview of system theory. Nursing Clinics,ef m mm: 69 3.9 385-393. Heine. A. (1981. 3rd quarter). Helping hypertensive clients help them- selves: The nurse's role. Eatient Counseling and Health Education. pp. 108-112. Henderson. A. (1984). Interpreting the evidence on social support. Social.Es¥§hiatLy..12. 49-52. Hershey. J. et a1. (1980). Patient compliance with antihypertensive medication. American Manual .of. Euhlio Health. 70. .10. 1081-1089. House. J.. & Kahn. E. (in press). Measures and concepts of social sup- port. In S. Cohen & L. Syme. Eds.)..finciel.sunnent.and health. New York: Academic Press. 211 Hypertension Detection and Follow-Up Program Co-operative Group. (1979). Five-year findings of the Hypertension Detection and Follow-up Program: I. Reduction of mortality of persons with high blood pressure. lournal at the American Medical Association. 24.2. 2562-2571. Hypertension Detection and Follow-Up Program Co-operative Group. (1982). Five-year findings of Hypertension Detection and Follow-Up Program: III. Reduction in stroke incidence among persons with high blood pressure. lounnal.ot.the.American.Medieal.Associa: .Iien. 252. 633-638. Joint National Committee on Detection. Evaluation. and Treatment of High Blood Pressure. The 1984 report. Archixes gi.lnternal.Medi: 5.1.339 15.4.: 1045-1057 . Kahn. R. (1979). Aging and social support. In M. W. Riley (Ed.). Birth to death: Interdisciplinary pecspectjies. (American Association for the Advancement of Science Selected Symposium no. 30) Boulder. CO: Westview Press. Kannel. W. (1978). National history and epidemiology of hypertension. In Hypertensign Huddle. New York: Pfizer. Kap1an. B.. Cassel. J.. & Gore. S. (1977). Social support and health. Medical .QaLe. .15. 471-482. King. A.. A Frederikson. L. (1984). Low-cost strategies for increasing exercise behavior: Relapse training and social support. Behayigr MM: 8: .1: 3-21. King. I. (1981). A theory for nursing. New York: Wiley & Sons. King. 5. (1982). lnteccelatienshieseiknouleddeeidiseaseLAneuJedae eiteeatmentlsecialsudnemderceiuedimeactetdiseaseand stated compliance in a group of. patients thh COED. Unpublished thesis. Michigan State University. Kirscht. J.. Kirscht. J.. & Roenstock. I. (1981). A test of interven- tions to increase adherence to hypertensive medical regimens. Health Education Quartefli. 8. 3.. 261-272. Klatsky. A. (1982). The relationship of alcohol and the cardiovascular system. Annual Befleu 21 Nutrition. 2. 5l-7l. Kochar. M. (1981). Hypertension--An overview. Eamily and Community Health. 49 1: 1-20. 212 Levine. D.. Green. L.. A Deeds. G. (1979). Behavioral and clinical effects of health education in randomized trials with hypertensive patients. Journal at the American Medical Asseciation. 241. 1700-1703. Lin. N.. Simeone. R.. A Ensel. W. (1978). Social support. stressfu1 life events. and illness: A model and an empirical test. Journal otHealthandSocialBehaademZQ. 108. Lowenthal. 0.. Briggs. W.. Mutterperl. R. et al. (1976). Patient com- pliance for antihypertensive medication: The usefulness of urine essays. patient Iberapeutic Regimens. l2. 405-409. Lowenthal. M.. A Haven. C. (1968). Interaction and adaptation in inti- macy: A critical variable. American 593.191.99.131 39.1191. 5.3.. 20. MacGregor. 6.. Best. F.. Cam. J. et a1. (n.d.). Double-blind randomized crossover trial of moderate sodium restriction in essential hyper- tension. Lancet. 1. 351-355. Marston. M. (1970). Compliance with medical regimens: A review of the literature. Nursing Research. 19. A. 312-323. Mason. 5- (1982). HyoectensjehcaeeaAguldeioreatienteducatien. Norwalk. CN: Appleton-Century-Crofts. McCombs. J.. Fink. J.. A Bandy. P. (1980). Critical patient behaviors in high blood pressure control. Cardio:1ascular.flursing..16. l9. McKenney. J.. Slining. J.. Henderson. H. et al. (1973). The effect of clinical pharmacy services on patients with essential hyperten- sion..91£cula119n. 155. 1104-1111. Melcher. 0. (1981). From course notes. General systems theory--Some definitions and concepts. Michigan State University. Meyers. A.. Thackwray. 0.. Johnson. C. et a1. (n.d.) A comparison of prompting strategies for improving appointment compliance of hypertensive individuals. Reflexion Iherapy. 14. 267-274. Miller. P. Ingham. J.. A Davidson. 5. (1976). Life events. symptoms. and social support..Journal.oI.Esychosomatic.Research. 2Q. 515- 522. Minium. E. (1970). Statisticalieaseninginasuchelegyandeducatlen. New York: John Wiley A Sons. 213 Minkler. M. (1981). Applications of social support theory to health education: Implications for work with the elderly. Health Educa: tion Quartenlx. 8. 2. 147-164. Morisky. 0.. Bowler. M.. A Finlay. J. (1982). An educational and behav- ioral approach toward increasing patient activation in hyperten- sion. loumal at Community Health. 7. 3. l7l-l81. Murawski. 8.. Penman. 0.. A Schmitt. M. (1978). Social support in health and illness: The concept and its measurement. Cancer Hutsing. .1. 365-371. Myers. J. Lindenthal. J.. Pepper. M. et a1. (1972). Life events and mental status: A longitudinal study. Journal of Health and Social Behavior. .13. 298-306. National High Blood Pressure Information Center (1978). New hyperten: sixeeceualencedataandcecmnendedeublicstatements. Report approved by High Blood Pressure Co-ordinating Committee. Nelson. E.. Stason. W.. Neutra. R. et al. (1978). Impact of patient perceptions on compliance with treatment for hypertension. Medical .Qane. 56. 11. 893-906. Norbeck. J. 5.. Lindsey. A.. A Carrieri. V. (1981). Nursing Research. 30. 5. 264-269. Nuckolls. K.. Cassel. J.. A Kap1an. B. (1972). Psychosocial assets. life crises. and the prognosis of pregnancy. American Journal of Enldemielegy. 25. 431-441. Patel. 0.. Marmot. M.. A Terry. D. (1981). Controlled trial of biofeedback-aided behavioral methods in reducing mild hyperten- sion. British Medical .lcurnal. 2&2. 2005-2008. Pearlin. L.. Liberman. M.. Menaghan. E. et a1. (1981). The stress pro- cess. mm .91 Heelih and 5.99.131 Behaflgr. 22. 337-356. Pless. 1.. A Satterwhite. B. (1973). A measure of family functioning and its application..§ec1al.Sclence.1n.Medicine. 1. 613-621. Polit. D.. A Hungler. B. (l978)..Nunsing.researcha.Erinciples.and ,methods. Philadelphia: J. B. Lippincott. Powers. M.. A Wooldridge. P. (1982). Factors influencing knowledge. attitudes. and compliance of hypertensive patients. Research in Nucsing and Health. 5. 171-182. 214 Reisen. E.. Abel. R.. Modan. M. et a1. (1978). Effect of weight loss without salt restriction on the reduction of blood pressure in overweight hypertensive patients. Men England.lournal.ot.Medicdne. 2.28.9 1'6. Report of the Hypertension Task Force. (1979). Current research and recommendatiensicomthetashiecceceaoctsenthecameneg: nancy. and obesity. U.S. Dept. of Health. Education. A Welfare Sackett. 0. L. (1976). The magnitude of compliance and non-compliance. In D. Sackett A R. Haynes (Eds.)..Comoliance Hiih therapeutic regimens. Baltimore: Johns Hopkins University Press. Sackett. 0. (1980). Editorial: Hypertension v. compliance with anti- hypertensive therapy. Qanadian lournal cf Euhlic Health. 1.1. 153-156. Sarason. I.. Levine. H. Basham. R. et al. (1983). Assessing social support: The social support questionnaire. Journal of Bensgnality m Midi 3512091931. 44: .1: 127-139. Satariano. W.. A Syme. S. (1981). Lifechange and disease in elderly popu1ations: Coping with change. In J. McGough et al. (Eds.). .Biglogy; hehayior. and aging. New York: Academic Press. Schaefer. C.. Coyne. J.. A Lazarus. R. (in press). The health related functions of social support. Journal of Benayioral Medicine. Sosa. R.. Kennell. J.. Klaus. M. et a1. (1980). The effect of a supportive companion on perinatal prob1ems. length of labor. and mother-infant interaction. New.England.ldu£nal.QI.Medlcine. m, 5 97-600 0 Stamler. J.. Rhomberg. P.. Schoenberger. J. et a1. (1975). Multivariate analysis of the relationship of seven variables to blood pressure: Findings of the Chicago Heart Assoc. Detection Project in Industry. lournal .ot Chaonic Disease. 2d. 527-548. Standacdsgtcacdioiascularnutsingccactice. (1975). ANA publi- cation code: MS-4 20M 2/75. Kansas City. MO: American Nurses Association. Stecke1. 3.. A Swain. M. (1977). Contracting with patients to improve compliance..HQ§21Idls. 51. 81. Swain. M.. A Stecke1. S. (1981). Influencing adherence among hyperten- sives. Research in NuLsing and Health. A. 173-226. 215 Tanner. G.. A Noury. D. (1981). The effect of instruction on control of blood pressure in individuals with essential hypertension. Journal of Aduanced Nutsing. 6. 99-106. Thoits. P. (1982). Conceptual. methodological and theoretical problems in studying social support as a buffer against life stress. Journal of Health and Sociological Behayior. 23. l4S-159. HLS--A statistical portrait of. the American people. (1973). A. Hacker (Ed.). New York: Viking Press. U.S. Bureau of Census. (1982). Statistical abstract 91 the United .Stateet,1281. (102nd edition). Washington. D.C. Vickers. G. (1959). Is adaptability enough? Behaxional.§clence. A. 219-234. vonBertalanffy. L. (1955). General systems theory. Main Cunrents in Modem lbought. ll. 75-83. Walker. K.. MacBride. A.. A Vachon. M. (1977). Social support networks and the crisis of bereavement. Social.Science.and Medicine. 11. 35-410 Weiss. R. (1974). The provision of social relationships. In 2. Rubin (Ed.). Doing unto.othens. Englewood Cliffs. NJ: Prentice-Hall. Weissman. M. et a1. (1981). The assessment of social adjustment: An Update. Archives of .Genecal Estoniatry. 3.8. 1250-1258. Widmer. R.. Cadoret. R.. A Troughten. E. (1983). Compliance charac- teristics of 291 hypertensive patients from a rural midwestern area. Journal.oi.Eamily.Eractice..11. 619-625. Wollam. G.. A Gifford. R. (1978. August). Four basic problems in con- trolling hypertension..Qonfiultant. pp. 25-30. Wortman. C. (1984). Social support and the cancer patient: Conceptual and methodologic issues..£anco£. 53. 10. 2339-2359. Zisook. 5.. A Gammon. E. (1980-81). Medical noncompliance. Inter; national Journal of Esychiatry in Medicine. l0. 4. 291-301.