{I 1 .‘ ‘5‘. "' .3 l .i"':\ ”I“ ‘31. 7 , (u “::| I "' '3'):- If. D.- 9111'” .HI' ‘ 1 ' '1 ‘ ' 1 WWI ‘ 7 ‘17.. "\ 7. \I N " " N u - - ' '10“ ' - 3 " > " ‘ ' 1" . 1 L '. ‘ "' ‘, . I 7 Q I, ' ' .,l 7 ,' t n , ' ' 3 ,‘1' ‘11:, _ 17.41.”: 3 ., ;. .' in“ 13%.”: 3,11,.“ , l1¢r , c" . . H J. - v 1 1- . ‘1‘: ' , .1 3' ‘1‘. .1 ’-,‘ I . 3' _ '1. , ' '7'I' , . ‘7 .‘... ;_ " .u ' ’ I ‘_~ .7.'."'~7‘ .7 “r" .‘ 73'7"“ 371,7 “ " ‘ '7' " 7. 37'} 7' 7- 33 .37 "1‘ ‘- M." fit-31,7; 7-1 §7 7“- ; 7 7 - 7:. 7 7 . ‘ {\J..-..-+~»,._7.? , 3,, 3-,»,‘7 ’7 . '\ W c.5377. i113 ' 7: " ‘7 ' '7. 79 7' 3.7.7,. ,3 737 37,-- .. 77". “I: v ?.\;:' - :11” 3,. ’5‘. 3,057 ‘ 13‘ 31,1, 7 'X I. " ‘- 33", . 11,1“, .,7 . ,1,“ ., l I ,3??? , ‘73,,- 1,313.! 30;,“ S :H' 1:. ..f.‘1\‘ ,. V. ,1“. : '7 73 N1, "Y”L'Pfi'k’” 5:31).“ "'"l’ .'4 13,7, ' $1., 3‘79 =7: 3‘ ‘77 3131.717 7‘! I I," ‘g' ~,7~.V1II0.‘., :: “\1141H'LII'31},IL.1-Lflr;AJar|'V'-Itnm 9:”. 'QI"- ”11"‘1'Ew‘dfl ‘1 k3,.” w' ‘1'"1'3‘ . I In " I 41 Qcfilfi '1‘ 'Iite'g' 3. J Slug]: 1d,; 11%;: ”1‘7 fif'ul V'EI. "1" J,‘ ”131?, 7." ‘. L‘l‘Ir‘VNI . I-r7 .‘V’T'. 7‘ : . A 3‘1'7, ‘fimK-fi fig . 7. I , , ,.1‘5'H\ ifs-:yrfig . t ." ’ ': ":'37"‘LI:'IVI(DI:.' ‘1‘1' "gm'y 4' ‘ it‘d . ‘ ,1 "fin-N J, I - .3, 3 25717 0' . . . . . 3 - I 7.7 \‘V1‘ 3". ' ‘ v. 7377.33}? ’5'“ ‘ufl IL'M (Avg! 331,541”, ,r "','| 3' 1|" [3.1'.r,.,l I 3 '1'11II' . _ j 7 . 7'11‘1'3'1'J 77.3373. '. 7.3.3.733- 331377 u. 3.7.7.3 {31:43.37 1 ' 7‘3" ,1 5' “15“,, .3? .. “L 4.7 .713 7 . ‘ 3:. ."m. k“ I, | m-.-“, , LIBRARY“ ' Michigan State University This is to certify that the thesis entitled THE RELATIONSHIP OF PERCEIVED SEVERITY 0F MYOCARDIAL INFARCTION AND COMPLIANCE WITH THE POST-MI THERAPEUTIC REGIMEN presented by Ellen M. Potter has been accepted towards fulfillment of the requirements for Master of Sciencgegree in Nursing / Major professor Date /, //° /?(0 / / 0-7539 MS U in an Afi‘irmau'vc Action/Equal Opportunity Institution MSU LIBRARIES ‘3..- REIORNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE RELATIONSHIP OF PERCEIVED SEVERITY OF MYOCARDIAL INFARCTION AND COMPLIANCE WITH THE POST-MI THERAPEUTIC REGIMEN By Ellen M. Potter A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN NURSING College of Nursing 1986 ABSTRACT THE RELATIONSHIP OF PERCEIVED SEVERITY OF MYOCARDIAL INFARCTION AND COMPLIANCE WITH THE POST-Ml THERAPEUTIC REGIMEN By Ellen M. Potter Myocardial infarction is a leading cause of death in the United States yearly. A descriptive correlational design was utilized to determine the relationship between perceived severity of myocardial linfarction and compliance with the therapeutic regimen. Data were taken from an earlier research study (Kline 5 Warren, 1981). The sample used in this study consisted of 70 males who had experienced a myocardial infarction within the past 12 months and who were given diet, medication and exercise prescriptions. Subjects completed a sociodemographic questionnaire which included the measurement of perceived severity of myocardial infarction and a compliance measurement scale. The Pearson product-moment correlation coefficient was employed to analyze the strength of relationships between compliance and the sociodemographic items. No relationship was found between perceived severity of myocardial infarction and compliance with the therapeutic regimen. Slight relationships (r - .20 - .35) were found between various compliance subscales and age, work status, length of time since the myocardial infarction occurred, past participation in an in-hospital teaching program, current participation in a cardiac Ellen M. Potter teaching program and the presence of additional chronic diseases. More research is needed to develop scales to measure perceived severity and compliance. Until these scales are developed, nursing implications cannot be specified. ACKNOWLEDGEMENTS I wish to thank Brigid Warren and Nancy Kline for the use of their data and Bryan Coyle for his assistance in statistical analysis. I would also like to thank the following faculty members for their service on my thesis committee: Barbara Given, Brigid Warren, Clare Collins, Sandy Hayes and Mary Nugent-Polk. TABLE OF CONTENTS Page List of Tables . . . . . . . . . . . . . . . ..... . . . . vii List of Figures . . . . . . . . . . . . . . . . . . . . . . . . viii List of Appendices . . . . ....... . . . . . . . . . . . ix CHAPTER I: Introduction to the Study Introduction . . . . . . . . . . . . . . . . . . . . . . . 1 Background . . . . . . . . . . . . . . ......... . 2 Compliance . . . . . . . . . . . . . . . . . . . . . . 2 Compliance with Post-MI Therapeutic Regimen ..... h Perceived Severity . . . . . . . . . . . . . . . . . . A Purpose of the Study . ................ . . 7 Research Questions . . . . . . . . . ..... . . . . . . 7 Definitions of Concepts .............. . . . . 7 Demographic Variables . . ........ . . . . . . 10 Structural Variables . ................ 10 Assumptions . . . . . . . . ................ 11 Limitations . . . ............... . ..... 11 Outline of Chapters . . . . . . . . . . . . . . . . . . . . 13 CHAPTER II: The Conceptual Framework Overview . . . . . . . . . ............ . . . . 14 Introduction . . . . . . . . . . . . . . . . . . . . . . . 1h Pathophysiology of Myocardial Infarction . . . . . . . 14 Treatment and Recovery Period . . . . ...... . . 16 The Health Belief Model . . . . . . . . . . . . . . . . . . 20 King's Nursing Theory . . . . . . . . . . . . . . . . . . . 3h Interaction of King's Theory with the Health Belief Model . . . . . . . . . . . . . . . . . . . . . . . #1 Summary . . . . . . . . . . . . . ...... . . . . . . . #5 CHAPTER III: Review of the Literature Introduction . . . . . . . . . . ...... . . . . . . . Compliance Behavior in Health Care . ........... The Measurement of Compliance . . ............. Compliance with the Post-MI . . . . . . ..... . . . . . The Post-MI Medication Regimen . . . . ........ The Compliance with the Post-MI Medication Regimen The Post-MI Dietary Regimen . . . . . . . . . . . Compliance with the Post-MI Dietary Regimen ..... The Post-Ml Exercise Regimen . . . . . . . . . . . Compliance with the Post-MI Exercise Regimen ..... The Health Belief Model/Perceived Severity . . . . . . sumary O O I O O O O O O O O O O O O O O O O I O O O O 0 CHAPTER IV: Methodology Overview ......................... Research Question . . . . . . . . ...... . . . . . . . Secondary Research Questions ............. Operationalization of the Study Variables ......... Perceived Severity of MI . . . . . . . . . . . . . . Stated Compliance with the Therapeutic Regimen . . Modifying Factors ................ Sample . . . . . . . . . . . . . . . . . . . . . . . . Data Collection Procedure ................. Rel‘ability and vaIidity O O O O O O O O O O O O ..... Statistical Analysis . . ................. Human Subjects Protection ............ . . . sumary O O O O O O O O O O O O O O O O O O O O O O O O 46 A6 #8 55 56 59 71 73 80 82 89 98 101 101 102 102 102 103 103 10k 106 108 109 110 110 CHAPTER V: Data Analysis Overview . . . . .................... Descriptive Findings of the Study Sample . . . . . . . . Characteristics of the Study Sample . . . . . . . . . . . Factor Analysis of Compliance Subscales . ........ Reliability of the Compliance Instrument ........ Intercorrelatlonal Matrix for Subscales . . . . . . . . . Presentation of the Data Related to the Research Question Presentation of the Data Related to Modifying Factors . . . Demographic Variables . . . . . . . . . . . . . . . Structural Variables . . . . . . . . . . . . . . . . Other Findings . . . . . . . ..... . . . . . . . . . Summary . . ....................... CHAPTER VI: Summary and Conclusions Overview . . . . . . . . . . . . . . . . . . . . . . . . Summary and Interpretation of the Findings ...... Characteristics of the Study Sample . ...... Research Questions ................. othér Findings 0 O O O I O O O O I O I O O O O I O 0 Summary of the Findings and Limitations of the Study Nursing Implications ................ . . Implications for Nursing Research ...... . . . Implications for the Study Model . . . . ...... Implications for Measurement . . . . . ..... . . Implications for Future Research ........ . . . . Implications for Nursing Practice . . . . ..... Implications for Nursing Education . . . . . . . . . sumary O O C O O O O I O O O O O O O O O O O O O O O O 0 vi 111 111 112 119 121 123 123 127 128 128 129 131 133 I33 133 IHD 1h5 150 152 152 152 156 158 159 167 170 Table 1: Table 2: Table 3: Table A: Table 5: Table 6: Table 7: Table 8: Table 9: Table 10: Table 11: Table 12: Table 13: Table 1A: LIST OF TABLES Modifying Factors, Item Number on Questionnaire #5 and Scoring Used in Correlations with Perceived Severity . . . . . . . . . . . . . . . . Age Range Distribution of Subjects ......... Education Levels of Subjects . . . . . . . . . . . . Frequency Distribution by Occupation . . . . . . . . Frequency Distribution of Combined Annual Income . . Distribution for Length of Time in Months Since MI . Distribution of Other Reported Chronic Diseases . . . Distribution of perceptions of Severity of MI . . . . Compliance Subscales, Questionnaire Items Included in Subscales and Alpha Coefficients for Each SUbscale O O O O O O O O O O O O O O C O O 0 O O O Intercorrelation Matrix for Compliance Subscales Correlations of Perceived Severity of Myocardial Infarction and Compliance with the Therapeutic Regimn O I O O O O O O O O O O O O O O O O O O O 0 Correlations of Perceived Severity of MI and Demographic Variables Using Pearson r . . . . . Correlations of Perceived Severity of MI and Structural Variables Using Pearson r . . . . . . . The Relationship Between Medication, Diet, Exercise Activity, and Total Compliance and Modifying Factors using Pearson Product Moment Correlations . vii Page 105 112 113 11k 115 117 118 119 122 12k 125 127 128 130 Figure 1: Figure 2: Figure 3: Figure A: Figure 5: Figure 6: Figure 7: LIST OF FIGURES Page Schematic Diagram of the Original Health Belief Mdel O O O ...... O O O O O ......... 22 Health Belief Model for Chronic Disease . . . . . . . 23 Selected Components from the Health Belief Model as Used in this Study . . . . . ........... 33 Dynamic Interacting Systems ............. 35 Process of Human Interaction . . . . . . . . . . . . 37 Integration of Concepts from the Health Belief Model and Kings' Model as used in this Study ..... . #2 Final Modification of the Model used in this Study . 155 viii Appendix Appendix Appendix Appendix Appendix Appendix Appendix LIST OF APPENDICES Letter of Introduction to the Project Physician Letter to Potential Study Participants Instruction Letter to Study Participants Instruments Consent Form and Investigators' Statement Project Approval Letters from UCRIHS Table 1 ix CHAPTER I Introduction to the Study Introduction Coronary heart disease, a major clinical manifestation of which is the myocardial infarction (MI), is the leading cause of death in the United States yearly (National Heart, Lung 5 Blood Institute, 1982). As many as one and one-half million Americans suffer myocardial infarctions yearly, with about 550,000 survivors (U.S. Department of Health 5 Human Services, 1983). The American Heart Association (1985) estimates that there are h,600,000 Americans still living who have experienced a MI. There is general agreement among health care providers that compliance with the post-MI therapeutic regimen will improve prognosis and decrease morbidity and mortality among these survivors (Croog 5 Levine, 1982). Compliance is a complex and important issue which health care providers must address (Dimatteo 8 DiNicola, 1982; Haynes, 1979; Sackett 8 Snow, 1979). Many variables can be found in the literature that are believed to influence compliance behavior, including perceived severity of the disease, specific patient beliefs about health and illness, patient motivation, previous socialization personality factors of both the patient and the health care provider, complexity of the regimen, and barriers to compliance such as cost (Becker, 197k; DiMatteo 5 DiNicola, 1982). Kline and Warren (1983) found perceived severity of myocardial infarction by a husband to be a significant predictor of the husband's perception of marital functioning of the couple and recommended further research focusing on the concept of perceived severity. In this study the focus will be on the relationship between the post-MI patient's perceived severity of his MI and his compliance with the prescribed therapeutic regimen. Background Compliance Compliance has been defined as the extent to which the patient carries out the therapeutic recommendations of health care providers concerning prescribed medications, diet and behavior modifications (Given 8 Given, 1982). The term "compliance" is problematic because it presents a picture of the patient as a passive follower of orders. In reality, the patient is far from passive. He makes choices, some conscious and some unconscious, about taking medications, changing diet and exercise patterns and following general therapeutic suggestions. These choices are based on his perceptions, beliefs and attitudes about his health status and the prescribed regimen (Loustau, 1979). Davis (1968) noted the difference between compliance behavior and compliance attitudes or orientation. As an attitude, compliance consists of a willingness or orientation toward doing what the provider advises. In this behavioral aspect, however, compliance can be said to exist only when the patient actually carries out his providers orders (Davis, 1968). In other studies (Kegels, Kirscht, Haefner, 8 Rosenstock, 1965; Sackett 8 Snow, 1979), a positive relationship between favorable orientation toward compliance and actual performance has been found. Kegeles, et al (1965) found that the best indicator of subsequent follow-through in obtaining a cytological examination for cervical cancer (although not a perfect predictor) was stated intention to do so. It was estimated that #02 of all medical recommendations given in the 0.5. are not followed (Davis, 1982). The regular taking of prescribed medications is avoided at an alarming degree (DiMatteo 5 DiNicola, 1982). Patients fail to follow short-term medication regimens at rates of 202-302 when the regimen is curative and 302-h02 when the regimen is preventive (Sackett 8 Snow, 1979). With lifetime medication regimens and with long-term behavioral scheduling, such as eating a restricted diet, noncompliance is usually about 502 initially and increases as therapy continues (Haynes, 1979; Sackett 8 Snow, 1979). Clearly, noncompliance can reach major proportions in patients with chronic disease for whom a lifetime therapeutic regimen has been recommended, such as the post-MI patient. It is not the purpose of this study to validate current treatment of the post-MI therapeutic regimen It is assumed that the treatments used (dietary changes and medication and exercise prescriptions) are based on sound scientific research and that compliance to these treatments will improve the health and well-being of the patient through risk factor reduction. For the purposes of this study, compliance is viewed as a positive action on the part of the patient toward the achievement of his maximum health potential. Compliance with Post-MI Therapeutic Regimen Croog and Levine (1982), in a longitudinal study of heart attack victims across eight years, analyzed compliance attitudes/orientation with regard to ten principle areas of medical advice (medications, diet, smoking, work, weight control, resting, alcohol consumption, avoiding quarrels, exercise and sexual activity). Patients were questioned at seven weeks, one year and eight years post-MI regarding medical recommendations and compliance in these ten areas. While total compliance with medications remained between 88.52 and 908, compliance in all other areas was initially lower and decreased by as much as 262 to 282 by the eighth year follow-up interview. Clearly, Croog and Levin (1982) found compliance with the therapeutic regimen for these post-MI patients to be an issue of concern to health care providers that warrants further investigation. Categories of medical advice can vary greatly between health care practitioners. In addition, information given in each of the categories may also vary. Three categories of medical advice, medications, dietary regimens and exercise regimens, were chosen for this study because these three treatment areas appear to be well standardized in practice (Andreoli, Fowkes, Zipes, 5 Wallace, 1983; McGurn, 1981; Wenter 8 Hellerstein, 1979). Perceived Severity A differentiation is made in this study between severity of myocardial infarction and perceived severity of myocardial infarction. Severity of myocardial infarction refers to objective findings the physician uses to determine the extent of damage to the cardiac tissue. These objective measures include elevations of cardiac enzymes and electrocardiograph changes (Karlinger 5 Gregoratos, 1981; Petersdorf, Adams, Braunwald, Isselbacher, Martin 8 Wilson, 1983). Perceived severity refers to the myocardial infarction patients subjective feelings about the severity of his MI. Perceived severity has not generally been found to be related to physicians estimates of the seriousness of the disease (Becker 8 Maiman, 1975; Bonnar, Goldberg 5 Smith, 1969; Charney, 1972; Davis, 1968). In this study perceived severity of MI, rather than objective measures of severity, is the focus of research. The dimension of perceived severity as it relates to compliance behavior appears complex. Various perceptual factors, including severity, have been placed in a formulation called the Health Belief Model (Becker, 197“). This model has been used to study the relationship between perceptions and compliance. Although findings relating perceived severity and acceptance of preventive health recommendations are mixed, patient estimates of the seriousness of the present illness (or of some other attack) are consistently predictive of compliance with the prescribed regimen (Becker, 197%). Patients with extremely high or low levels of anxiety have been found to comply less well with both preventive and treatment-oriented recommendations (Becker, 197k; Ley, 1965). Other factors may influence perceptions of severity as well, including severity of symptoms, chronicity of disease and length of time since treatment Was initiated (Becker, 197k; Kasl, 197k). There are a number of literature reviews available that summarize research on the Health Belief Model and report significance of the major variables in the model, including perceived severity (Cummings, Jette 8 Rosenstock, 1978; Janz 8 Becker, 198k; Loustau, 1979; Mikhail, 1981; Stunkard, 1981). Janz and Becker (198A) provide the most recent review. They found perceived severity to be significantly related to health behavior in 652 of the total studies published through 198k and in 882 of the sick role behavior studies. Severity appears to have increased importance In chronic disease populations. No studies have been found in the literature relating perceived severity of a myocardial infarction to compliance with the post-MI therapeutic regimen. Hijeck (198“), however, has presented a theoretical framework for cardiac nurses using the Health Belief Model as a guide to patient care. Hijeck describes perceived severity as the extent to which the patient believes that myocardial infarctions generally are dangerous and that the patient's own myocardial infarction is serious. If the myocardial infarction or its ramifications are not perceived as serious for him or her, then, theoretically, the probability that the individual will engage in preventive or rehabilitative care, including compliance with the therapeutic regimen, lessens. More research is needed to determine the relationship of perceived severity to compliance with the therapeutic regimen in the post-MI patient. Purpose of the Study The purpose of this study is to determine the relationship between the post-Ml patients' perceived severity of his myocardial infarction and his compliance with the prescribed therapeutic regimen. Research Questions Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed therapeutic regimen? 1) Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed medication regimen? 2) Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed dietary regimen? 3) Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed exercise regimen? Definitions of Concepts For the purposes of this study, the concepts contained in the research questions are defined as follows: 1) 2) 3) Post-myocardial infarction patient: A male patient between the ages of 35 and 69, who has been diagnosed and hospitalized with a heart attack within the past 12 months as identified by physicians in an office setting. Perception: The unique, subjective way in which each individual views reality; the process by which information is organized, interpreted and transformed from sensory data and memory to give meaning to one's experiences (King, 1981). In this study, perception refers to the way the patient perceives and describes the MI experience. Stated compliance with the therapeuticgregimen: The extent to which the patient carries out the therapeutic recommendations of health care providers concerning prescribed medications, diet and exercise. Therapeutic regimen refers to the elements of the discharge treatment plan prescribed by health professionals as perceived and described individually by the post-MI patient. According to the literature and confirmed through practice, the regimen generally includes diet modification, pharmacologic interventions, reduction or elimination of risk factors defined by the American Heart Association, intervals of rest and prescribed physical activity (Barnell, 1979; Dehn, 1980; Goldberg, 1973; Kallio, 1978; Niccoli 8 Brannel, 1976; Owens, McCann, 8 Hutelmyer, 1978; Payne 8 Boineau, 1980; Wenger, 1978). A) In this study, stated compliance with the therapeutic regimen refers to the patient's report of the degree to which he implements the prescribed regimen. Each of the statements in this instrument is followed by a six-point Likert scale, e.g., I eat foods which are prepared according to my dietary restrictions: 1) all the time; 2) more than half the time; 3) about half the time; A) less than half the time; 5) never; or 6) not applicable to my treatment plan (see Appendix D, Questionnaire 2). Client perception of severity: the extent to which the patient believes that myocardial infarctions generally are dangerous and that the patient's own myocardial infarction is serious (Hijeck, 198A). This definition of severity has two components, one relating to patients beliefs about myocardial infarctions in general and the other relating to patients beliefs about their own myocardial infarction. Only the later component of this definition will be used in this study. Since the Random House Dictionary defines severe as "serious or grave” (Stein, 1980), the two terms severe and serious are considered to be synonymous. For the purposes of this study, perceived severity will be defined as the response on the instrument to the question "How severe was your last heart attack”? where subjects may indicate 1) very severe, a lot of heart damage; 2) moderately severe, some heart damage but not a lot; 3) mild, only a little heart damage; and A) no heart damage. 10 5) Modifying_factors: Factors believed to have some influence on a study variable or variables. In the Health Belief Model, the researchers postulate that certain modifying factors may influence individual perceptions (Becker, 197A). The Health Belief Model includes three categories of modifying factors; demographic variables, sociopsychological variables and structural variables. Certain modifying factors from the categories of demographic variables and structural variables are used in this research. These variables will be analyzed for relationships with the study variables perceived severity and compliance with the therapeutic regimen. Demographic Variables Demographic variables include such factors as age, sex, ethnicity, income and others (Becker, 197A). The demographic variables that will be included in this study are age, ethnic background, educational level, occupation and annual income. Structural Variables Structural variables include such factors as prior knowledge about the disease and prior contact with the disease (Becker, 197A). The authors of the Health Belief Model do not offer a conceptual definition of structural variables. For the purposes of this study, structural variables are viewed as the unchangeable characteristics of each patients' past history and experience. The structural variables that will be included in this study are number of heart attacks (one or more than one), length of time in months since last ll heart attack, past participation in in-hospital post-MI teaching programs (yes or no), present enrollment In special heart programs (yes or no), and the presence of chronic health problems (yes or no). Assumptions 1) Compliance with the recommended post-MI therapeutic regimen is a health behavior that will improve present and future health through risk factor reduction. The therapeutic treatment measures prescribed to the study sample are based on sound scientific research. 2) Health-related perceptions affect compliance and other health-related behaviors. 3) The concepts of perceived illness severity and stated compliance as defined in this study are real and measurable phenomena. A) Measurement of stated compliance is a reliable method of measuring compliance with a therapeutic regimen. 5) The instruments used in this study are sensitive enough to measure the concepts of perceived severity and stated compliance accurately. 6) The study participants have responded accurately to the instruments measuring compliance. Limitations 1) The subjects who agreed to participate in this study may be different from those who refused and, therefore, not representative of all post-Ml patients. 2) 3) A) 5) 6) 7) 12 In this study, compliance measurement is limited to medications, dietary changes and exercise. Since compliance with other recommendations may also be part of a therapeutic regimen, this measurement may not reflect actual compliance with the entire regimen accurately. The definition of perceived severity in this study is limited. Hijeck (198A) has defined perceived severity of MI as the extent to which the patient believes that myocardial infarctions generally are dangerous and that the patients own myocardial infarction is serious. Only perceptions of the patients own myocardial infarction are used in this study. In addition, the measurement of perceived severity is limited to one item on the questionnaire. The measurement of compliance and severity is subjective and variation may occur at different points in time. This study does not control for other variables which could influence compliance (such as social support, barriers to compliance and complexity of the therapeutic regimen). Subjects included in this study varied from zero to 12 months post-Ml. Perceptions of severity and of compliance probably vary at different points in time from occurrence to one year post-MI. Therapeutic regimens will not be validated with the patient record or the physician. 13 ' 8) Clinical documentation of MI is not available for study participants as subjects were referred by their physicians. Subjects may have been admitted to the study who do not meet the clinical criteria for diagnosis of myocardial infarction. 9) Secondary data was used in this research. Data collected for another purpose may not be the most accurate for secondary I purposes. Outline of Chapters This study contains six chapters. Chapter I includes the background and purpose of the study, research questions, definition of the concepts, and the assumptions and limitations of the study. The conceptual framework of the study is described in Chapter II and includes the concepts of severity and compliance with the therapeutic regimen as they relate to the post-MI patient and to King's nursing theory. In Chapter III, a review of the literature pertinent to the research problem is presented. Chapter IV contains a description of the research methodology and design utilized in the study. Study findings and interpretations are presented in Chapter V. Finally, Chapter VI includes a summary of the study, conclusions, and implications for nursing practice. CHAPTER II The Conceptual Framework Overview The purpose of this study is to explore the relationship between perceived severity of MI and stated compliance with the therapeutic regimen for the post-MI patient. In this chapter, the conceptual framework on which the study is based is presented. Included is a brief description of pertinent details in pathophysiology, diagnosis, treatment and recovery period for the MI patient. Next, the Health Belief Model will be presented as a framework with which to analyze perceived severity and compliance behavior. Finally, the theory of Imogene King will be presented as a framework from which to view nursing intervention with the post-MI patient. Introduction Pathophysiologyiof Myocardial Infarction Myocardial infarction, technically speaking, is an irreversible death of cardiac tissue. The primary cause of MI in the United States is coronary heart disease (CHD), an occlusion of coronary arteries. In most persons the underlying process leading to M1 involves the formation of fatty plaques in the lining of the vessel walls or atherosclerosis (Dawber, 1980). Atherosclerotic heart disease involves the progressive narrowing of the lumen of coronary heart vessels due to an accumulation of lipids, calcium, fibrous tissue, complex carbohydrates, and blood products in the lining of the arterial wall (Jones, Dunbar, 8 JirOvec, 1978). 15 Under normal conditions, the production of atherosclerotic plaque is accompanied by the development of collateral circulation to meet the hearts' nutritional and oxygen demands. Atherosclerotic heart disease may be completely asymptomatic and identifiable at autopsy, or it may be symptomatic, resulting in chest pain (angina pectoris) or in cases of severe occlusion, myocardial infarction (Jones, et al., 1978). The diagnosis of MI includes consideration of the clinical presentation of the disease and physical findings, laboratory serum enzyme studies, and electrocardiogram (EKG) findings (Petersdorf, et al., 1983). There are no clearcut clinical presentations or physical findings that are completely diagnostic of MI. Chest pain is, of course, the most common clinical presentation but there are many variations. Serum enzyme studies are very useful in the determination of the diagnosis because necrosis of the myocardial cells releases certain enzymes into the circulation. The amount of enzyme released correlates with the size of the infarct and the temporal pattern of enzyme release is of diagnostic importance (Petersdorf, Adams, Braunwald, Isselbacher, Martin 8 Wilson, 1983). There is no single EKG pattern associated with MI and there may be a wide variety of QRS and ST-T changes (Karllnger 8 Gregoratos, 1981). The sin qua non of MI is the presence of Q waves not previously seen (Dawber, 1980). 16 Treatment and Recovery Period As stressed previously, myocardial infarction involves the irreversible death of cardiac tissue (Dawber, 1980). Healing of the damaged tissue occurs by replacement of necrotic cells with non-functional connective tissue within six to ten days following completion of the infarct and continues for up to six weeks post-infarction (Jones, et al., 1978). Thus, the post-MI state can be seen as a state of chronic disease in that return to the previous state is not possible and the individual will need to incorporate this understanding and all its implications into his life-style. Rehabilitation of the cardiac patient occurs in four phases as identified by Wenger and Gilbert (1978). Phase I is the acute illness phase, which lasts three to five days from the time of the infarct. The Phase I patient is usually observed in a coronary care unit. During this phase, the pain associated with acute infarct is relieved and the patient is stabilized from its life threatening complications. Patient education technically begins at admission and a formal cardiac rehabilitation program is usually Initiated when the patient is stable. Such programs typically include a graded activity program, diet and exercise, and a discussion of risk factors (Devney, 1980). Phase II refers to the remainder of the hospital stay after the patient is moved from the coronary care unit (Wenger 8 Gilbert, 1978). Post-MI hospitalization generally lasts ten days to three weeks for uncomplicated cases. Rehabilitation should include 17 gradually progressive physical activity, basic teaching regarding pathophysiology of MI and its relationship to therapy, risk factors and specific treatment modalities such as medications and dietary changes. Patients and their families should be counseled regarding sexual activity, recreational activity, returning to work and family and individual adjustment (Wenger 8 Gilbert, 1978). Phase III is termed "convalescence" (Wenger 8 Gilbert, 1978) and generally lasts from three to eight weeks after the MI occurred. Convalescence takes place in the home after hospital discharge. A gradual increase in activity should allow most patients who are able to return to work by the eighth to twelfth week post-MI. The central focus of rehabilitation in Phase III is usually aerobic training (in a group setting if possible), optimization of drug therapy and reassessment of limitations on life-style. An important feature of the immediate post-discharge period should be an evaluation of the patients' functional status and reinforcement of inpatient education (Andreoli, Fowkes, Zipes, 8 Wallace, 1983). Phase IV is termed ”recovery-maintenance" and begins at approximately eight weeks post-Ml (Wenger 5 Gilbert, 1978). The patients are evalauted based on these criteria (Andreoli, et al., 1983): 1) return to gainful employment if this is medically feasible; 2) achievement of an acceptable level of functional capacity that allows the patient to perform desired work or leisure activity; 3) optimal control of all risk factors for recurrent cardiac events; A) successful patient education regarding necessary 18 restrictions on activity, proper medication usage and continued risk factor control. Some patients may meet these criteria within a few months after infarction and some will never reach Phase IV (Andreoli, et al., 1983). Consideration of these phases of cardiac rehabilitation is necessary in order to understand the context within which the post-MI patients exist. Permanent life-style changes will be required of these patients. Some of these life-style changes will be discussed in greater detail in Chapter III, the literature review. Sackett and Snow (1979) report that compliance with lifetime medication regimens and with long-term behavioral scheduling, such as eating a restricted diet, is usually about 502 initially and decreases as therapy continues. Compliance to the post-MI therapeutic regimen is necessary for passage through these rehabilitative stages and optimum rehabilitative potential. Risk factor reduction, as indicated in Phase IV of cardiac rehabilitation, plays a major role in cardiac rehabilitation in the post-MI patient. The Framingham Study (Dawber, 1980) was undertaken to determine the risk factors for coronary heart disease and other atherosclerotic disorders. The epidemiologic study was started in 1950 and ended 2A years later with a total of 5,127 subjects included. The study took place in Framingham, Massachusetts. Risk factors that were significant at the pig .05 level were being male, hypertension, high blood cholesterol, inactivity or low activity level, obesity, cigarette smoking and diabetes. Risk was found to 19 increase with age. Combined risk factors were found to increase risk. With the presence of only moderate elevation but with a number of factors involved, the total risk was still found to be considerable (Dawber, 1980). Of those risk factors mentioned in the previous paragraph, only sex and age cannot be altered. Diabetes and hypertension can be controlled to varying degrees with diet and medication and the remainder of the risk factors (high blood cholesterol, inactivity, smoking, obesity) can all be eliminated or reduced by life-style changes (Dawber, 1980; McGurn, 1981; Wenger 5 Hellerstein, 1978). It is clear in reviewing these risk factors that life-style changes will be a major part of the treatment for the post-MI patient. The degree of compliance with the therapeutic regimen can have a major impact on recovery and long-term survival. Compliance with lifetime medication regimens and life-style changes, such as dietary changes, is notoriously low (502) and decreases over time as therapy continues (Sackett 8 Snow, 1979). Assistance with life-style changes and risk factor reduction is an integral part of nursing care for the post-MI patient (Dracup, Meleis, Baker 5 Edlefsen, 198A). Clearly, nursing intervention in the area of compliance with the therapeutic regimen should play a major role in nursing care of the post-MI patient after discharge from the hospital. The next section of this chapter will provide a framework within which compliance with the post-MI therapeutic regimen can be analyzed based on concepts from the Health Belief Model. 20 The Health Belief Model The Health Belief Model has been useful as a theory base from which to analyze compliance behavior. The Health Belief Model was developed in the early 1950's by Rosenstock, Hochbaum, Kegeles and Leventhal to explain preventive health behaviors (Rosenstock, 197A). The model was derived from the psychological theory of Lewin (19A8) which postulated that an individual existed in a life space composed of regions, some of which were positively valued (positive valence), others of which were negatively valued (negative valence), and still others of which were neutral. Diseases, if they were represented In the life space at all, would be regions of negative valence which would be expected to exert a force moving a person away from that region, unless doing so would require him to enter a region of even greater negative valence. One's activities were thus conceived of as a process of being pulled by positive forces and repelled by negative forces (Lewin, 19A8). The earliest characteristics of the Health Belief Model postulated that in order for a person to take action to avoid a disease he would need to believe (1) that he was personally susceptible to it, (2) that the occurrence of the disease would have at least moderate severity on some component of his life, and (3) that taking a particular action could in fact be beneficial by reducing his susceptibility to the condition or, if the disease occurred, by reducing its severity, and that it would not entail overcoming important psychological barriers such as cost, convenience, pain or embarrassment (Rosenstock, 197A). With respect 21 to taking a test for early detection of a disease, the same factors were deemed necessary. In addition, there was the requirement that the individual believe he could have the disease even in the absence of symptoms (Becker, 197A). See Figure 1 for a schematic representation of the model. Later additions to the model were made when Rosenstock (197A) postulated that In order for an individual to take action to prevent disease or detect it, he/she must first be motivated to do so. ng§_ to action and general health motivation were added at that time (Rosenstock, 197A). The Health Belief Model has been utilized in examining preventive health behaviors, including screening tests for tuberculosis, cervical cancer and for dental disease and rheumatic fever (Rosenstock, 197A). Although it was originally used to predict preventive health behavior, there are a number of studies which have used one or more of these variables to predict patient compliance with the therapeutic regimen for chronic illnesses (Becker, Orachman, Haefner, Kirscht 8 Maiman, 1977; Cummings, Jette, 5 Rosenstock, 1982; Given 8 Given, 1982; Morisky, Bowler, 5 Plnlay, 1982). With some modification, the Health Belief Model can be used as a framework from which to examine compliance with the therapeutic regimen in the post-MI patient. See Figure 2 for a modification of the Health Belief Model for the chronically ill patient. According to Kasl (197A), the same variables of perceived severity, perceived susceptibility, benefits minus barriers, cues to action, motivating INDIVIDUAL PERCEPTIONS 22 MODIFYING FACTORS (personality, social Demoggaphic variables (age, sex, ethnicity, etc.) LIKELIHOOD OF ACTION Socigpsychologjcal Variables class, peer group pressure, etc.) Structural Variables (knowledge about the disease, prior contact with the disease, Perceived Benefits of preventive action minus Perceived Barriers to preventive action etc.) to disease ”X“ Perceived Seriousness of disease ”X" Perceived Susceptibility W V Perceived Threat of disease "X" Likelihood of taking recomnended preventive health action Cues to Action (mass media campaigns, advice, illness of family member or friend, newspaper article, etc.) Motivation (intent to comply, positive health activities, etc.) ‘fjjyire 1: (Becker, et al., 197A) Schematic Diagram of the Original Health Belief Model INDIVIDUAL PERCEPTIONS 23 MODIFYING FACTORS Waphlc Variables (age, sex, ethnicity, etc.) Sociopsycholggjggl Variables (personality, social class, peer group pressure, etc.) Structural variables (knowledge about the disease, prior contact with the disease, etc.) Perceived Susceptibil (conpl icat ions, another v Ml, etc.) Perceived Seriousness of chronic disease Ity LIKELIHOOD OF ACTION Perceived Benefits of treatment regimen minus Perceived Barriers to treatment regimen V _) Perceived Inpact of chronic disease Likelihood of complyipg with recommended treat- _ utentggimen I Cues to Action (mass media campaigns, advice, illness of fanily menber or friend, newspaper article, etc.) Motivation (intent to comply, positive health activities, etc.) Figure 2: Becker, et al., 197A and Kasl, 197A) Health Belief Model for Chronic Disease (adapted from 2A factors and modifying factors are applied to chronic illness. These factors together influence the patients' perception of the impact of the chronic disease which, in turn, influences compliance with the therapeutic regimen. In this study, compliance in relation to the post-MI therapeutic regimen is the health action of concern. Susceptibility and severity are perceived as threat components of the model (Kasl, 197A). Even if individuals recognize personal threat, they will not take action unless the course of action is believed to be beneficial in reducing the threat. Perceived susceptibility is defined by Rosenstock (197A) as the subjective risks of contracting a condition. For the post-MI patient, perceived susceptibility may be viewed as the patients belief in his susceptibility to another M1 or complicating condition thereof. Denial has been found to play a role in the psychological coping strategies of some post-MI patients. Thus, denial could alter perceived susceptibility in these patients (Thomas, Sappington, Gross, Noctor, Friedmann, 8 Lynch, 1983). Convictions concerning the seriousness of a given health problem vary from person to person. The degree of seriousness may be judged both by the degree of emotional arousal created by the thought of a disease as well as by the kinds of difficulties the individual believes a given health problem will create for him (Becker, 197A). A person who sees his health problem in terms of its medical significance would be concerned about death, reduced physical or mental functioning and disability. The perceived seriousness of a 25 condition may, however, for a given individual, include such broader and more complex implications as the effects of the disease on his job, on his family life, and on his social relations. Laustau (1979) has described the use of the Health Belief Model in predicting patient compliance and has defined perceived severity as (1) the individuals recognition of the seriousness of his health problem, (2) how he rates the health problem in relation to other illnesses, (3) the possibility that the problem could result in future illnesses and (A) the impact of his illness on significant others in his environment. Laustau emphasizes that perceived severity is not related to professional estimates of how serious the illness may be. The method of defining and measuring perceived severity often varies from researcher to researcher. Becker and Maiman (1975) define perceived severity as the extent to which the person believes that becoming ill would bring serious organic and/or social repercussions. Becker and Maiman note that the definition and measurement of perceived severity (and other Health Belief Model components) necessitates appropriate refinement and specification according to the condition being studied. This poses a problem in relation to the study of perceived severity of myocardial infarction in the post-MI patient. No studies could be located by this researcher in which the Health Belief Model or perceived severity was applied to the study of compliance (or other health behaviors) in the 26 post-MI patient. Research is needed linking the Health Belief Model components to myocardial infarction, a major health threat in the U.S. today. Hijeck (198A) has presented a theoretical framework for cardiac nurses using the Health Belief Model as a guide to patient care. Hijeck defines perceived severity as the extent to which the patient believes that myocardial infarctions generally are dangerous and that the patients' own myocardial Infarction is serious. If the MI or its ramifications are not perceived as serious, the probability that the individual will comply with the therapeutic regimen, theoretically, lessens. Hijeck is the only theorist that could be identified by this researcher who defined perceived severity in relationship to a myocardial infarction. Hijeck's definition of perceived severity was, therefore, chosen for use In this study. In research utilizing the Health Belief Model, perceived severity has produced the lowest overall significance ratios, producing significant results in only one-third of the studies published between 197A and 198A (Janz 5 Becker, 198A). Speculation about these findings seems to Indicate that the low significance of "perceived severity" may be due in part to difficulties that study respondents have in conceptualizing this dimension (1) when they are asymptomatic, (2) for health threats that are usually thought to be long-term, (3) concerning medical conditions with which they have had little or no personal experience, and (A) when the subjects tend to view the condition as very serious and, consequently, there is little 27 variability with which to distinguish compliers from noncompliers. While "perceived severity" was the least significant of the health Belief Model components in these studies, it was significant in 882 of the studies dealing with sick role behavior. Most likely this is due to the more important role of perceived severity in individuals with diagnosed illnesses. For chronic disease, benefits and barriers refer specifically to the treatment regimen. A persons beliefs about the availability and effectiveness of various courses of action (or perceived benefits), and not the objective facts about the effectiveness, determine his course of action. An alternative is likely to be seen as beneficial If it relates subjectively to the reduction of one's susceptibility to or seriousness of the disease. For the MI patient, perceived benefits would be those actions that would reduce one's susceptibility to complications or another MI or would reduce the perceived seriousness of the disease in terms of disability, possibility of death, or disruption of social relationships. For example, the post-MI patient may perceive the post-MI therapeutic regimen of exercise as a benefit if he believes it will increase exercise tolerance (and decrease physical disability) and increase collateral circulation (and decrease the possibility of another MI). Benefits must be examined with barriers in mind because even the patient who perceives an action as very beneficial may not comply due to the overriding strength of the barriers. 28 The barriers to taking action (or complying to the therapeutic regimen for the post-MI patient) can take many forms and be perceived or real. Cost, inconvenience, unpleasantness or extent of life change required are only a few barriers. For the post-MI patient, side effects of medications may be a major barrier. In Michigan, a major barrier to exercise may be winter weather. A cue to action is a factor that serves as a cue or trigger to appropriate action (Becker, 197A). The combined levels of susceptibility and severity provide the energy or force to act and the perception of benefits (less barriers) provides a preferred path of action. The combination of these could, however, reach quite considerable levels of intensity without resulting in overt action unless some instigation event occurs to get the process in motion. In the health area, such events or cues might be internal (e.g., perception of bodily states) or external (e.g., interpersonal interactions, the impact of the media, or receiving an appointment reminder card from the health care provider). The required intensity of a cue that is deemed sufficient to trigger behavior presumably varies with differences in the levels of susceptibility and severity (Becker, 197A). With relatively little acceptance of susceptibility to or severity of a disease, rather intense stimuli may be needed to trigger a response. On the other hand, with relatively high levels of perceived susceptibility and severity, even slight stimuli may be adequate. 29 Cues to action for the post-MI patient may include anginal pain (cue to take prescribed nitroglycerine), running low on medications (cue to get a refill), or a mailed reminder to have a medical check-up (cue to make appointment). Motives selectively determine an individuals perceptions of the environment. Different individuals have different degrees of readiness to undertake health actions. Motives are dispositions within the individual to approach certain classes of positive incentives, and it is postulated by Becker et al., (197A) that the desire to attain or maintain a positive state of health and to avoid a state of illness is a dimension of health motivation. Thus, the post-MI patient may have a high or low degree of general health motivation (or somewhere between high and low on a continuum). There are a variety of modifying factors which may influence an individuals perception of the threat of disease. The Health Belief Model includes three categories of modifying factors: demographic variables, sociopsychological variables, and structural variables (Becker, 197A). These factors are believed to condition both individual perceptions and the perceived benefits of preventive or treatment health actions. Demographic variables Include age, sex ethnicity, income and others. Sociopsychological variables include such factors as personality, social class, and peer and reference group pressure. Structural variables include prior knowledge about the disease and prior contact with the disease. 30 For the purposes of this study, selected demographic and structural variables were chosen to be examined for possible relationships with perceived severity of MI and compliance with the therapeutic regimen. Demographic variables chosen are as follows: age, ethnic background, educational level, occupation and annual income. Structural variables chosen are: number of heart attacks, length of time since the heart attack occurred, past participation in in-hospital teaching programs, present participation in a special heart program, and the presence of additional chronic health problems. It is difficult to determine the relationship between perceived severity and the modifying factors. Most literature dealing with perceived severity utilizes the Health Belief Model but does not report correlations between the modifying factors and the components of the model (susceptibility, severity, benefits and barriers). This Is due to the focus of the Health Belief Model, which concentrates on relationships between susceptibility, severity, benefits and barriers and the resulting likelihood of taking the recommended health action. No studies could be found by this researcher reporting the relationships between perceived severity and age, ethnic background, occupation, educational level and income. Some studies have found no relationship between individuals' perceived severity of illness and the physicians estimate of seriousness (Becker 8 Maiman, 1975; Bonnar, Goldberg 8 Smith, 1969; Charney, 1972; Davis, 1968) but severity was not specifically correlated with number of heart 31 attacks, or additional chronic health problems. Haefner 8 Kirscht (1970) found that education about heart disease, cancer and tuberculosis presented in a film did have a slight positive relationship to perceived severity and that age was positively related to perceived seriousness of cancer and tuberculosis in those who received the education. Since a well population was used in this study, it is not appropriate to generalize these findings to education received by the post-MI patient in or out of the hospital. If modifying factors influence individual perceptions of severity and if perceived severity influences patient compliance, then it logically follows that those modifying factors should be related, indirectly, to compliance. In the literature, sociodemographic and illness variables tend to have no significant relationship to compliance (Becker 8 Maiman, 1975). Davis (1968) reported no significant relationship between compliance and patient age, sex, race, marital status, religion, occupation, or education. Ruffalo, Garabedian-Ruffalo and Pawlson (1985) and Peck and King (1982) reported that increased knowledge about treatment is not significantly related to compliance. Yoos (1981) summarized the compliance literature In regard to sociodemographic variables and concluded that age, sex, socioeconomic status and patient education generally had no significant association with compliance. A consistent relationship has not been established between the resulting disability of a disease and compliance (Gillum, 197A; Marston, 1970). In Hilbert's (1985) study of spouse support and 32 compliance in the myocardial infarction patient, compliance was not related to spouse support or demographic or illness variables but was related to present enrollment in a cardiac rehabilitation program. In conclusion, there is very little literature available describing the relationship between perceived severity and modifying variables. Additionally, modifying variables generally tend to have no relationship to compliance in the literature. The modifying factors included in this study will be examined for relationships with perceived severity of MI and compliance with the therapeutic regimen. In summary, the Health Belief Model variables previously described influence an individuals' perceived impact of chronic disease. It is these perceptions of the impact of the MI that influence an Individuals likelihood of complying with the therapeutic regimen. Selected components of the Health Belief Model are used in this research as a theoretical base with which to visualize the relationship between perceived severity of the MI and the likelihood of taking the recommended action or compliance. This model is Illustrated in Figure 3. The modifying factors (selected demographic and structural variables) depicted to the left of the model influence perceptions of severity of MI in the middle of the figure. Perceptions of severity, in turn, influence the level of compliance. An arched line extends from the modifying factors over the top of the 33 110111 vac FACTORS \ age \\ education \ work outside the home ~\ annual income PERCEIVED SEVERITY number of MI's OF MI \ time since MI 7 \ in-hospital teaching current participation in \ teaching program \ additional chronic diseases v LEVEL OF COMPLIANCE (transaction) figure 3: Selected Components from the Health Belief Model as Used in this Study 3A model to the level of compliance. This arched line indicates that the modifying factors may directly or indirectly (through other unknown factors) influence the level of compliance. The Health Belief Model is useful in analyzing how individual perceptions Influence health behavior. It does not, however, provide a framework within which the nurse may interact with the client in a therapeutic intervention. King's nursing model is an ideal complement to the Health Belief Model in working with patients in whom compliance may be an issue of importance, such as the post-MI patient. King's Nurgipg Theory For King (1981), nursing is seen as a process of human interaction between nurse and client, whereby each perceives the other and the situation; and through communication, they set goals, explore means, and agree on means to achieve goals. King has developed a nursing theory of goal attainment that incorporates a systems perspective (see Figure A). Three dynamic interacting systems have been presented as an open systems framework for nursing. Individuals comprise one type of system in the environment called personal systems (King, 1981). Individuals interact to form dyads, triads, and small and large groups, which comprise another type of system called the interpersonal system (King, 1981). Groups with special interests and needs form organizations, which make up communities and societies and are called social systems (King, 1981). Figure A: //"" ————————————— \ // SOCIAL SYSTEMS I (Society) I, ,— —- —————————— \\ I INTE RPERSONAL SYSTEMS . | (Groups) I / '— ----- ‘x I C I PERSONAL SYSTEMS I I i (Individuals) 9 I I I I I 1 L_-_- I r — - -' '- I I I I 1 \ I I \ _ a I I I I 1 I I \ I I \ \ ________ // \ ‘\ \ \ / 35 Dynamic Interacting Systems (King; 1981) 36 Concepts from the theory of goal attainment are: erce tion, judgement, action, reaction, interaction, and transaction (King, 1981). See Figure A for a diagram of Kings' process of human interaction. Perception is defined as each persons representation of reality. It is an awareness of persons, objects, and events. One's perceptions are related to past experiences, concept of self, socioeconomic groups, biological inheritance and educational background. Perception is seen as universal in that all persons perceive other individuals and objectives in the environment and these experiences provide the person with information. Perceptions are unique to each Individual as each person permits certain stimuli to enter from the environment. Perception is action-oriented in the present because each person views the world from information that is available to him. Perception is a process of human transactions with environment. It gives meaning to one's experience, represents one's image of reality, and influences one's behavior (King, 1981). For the purposes of this study, perceptions are defined as the unique, subjective way in which each individual views reality; the process by which information is organized, interpreted, and transformed from sensory data and memory to give meaning to one's experiences (King, 1981). Judgement refers to the value that an individual places on objects or events, based on perceptions. 37 , Feedback --- - - j 1 .,___L_. Perception I I I 1 Nurse ___,. Judgnent , I Action I I Reaction ———) Interaction —> Transaction Action I 1 I \/ Patient -———al Judgment ‘ 1 1 I 1 Perception L- — - — Feedback --- J Figure 5: Process of Human Interaction (Kingyl981) 38 Actions are based on previous perceptions and judgements. Action is a sequence of behaviors of interacting persons that includes: 1. mental action or recognition of presenting conditions 2. physical action or initiation of operations or activities related to the condition or situation 3. mental action to exert some control over the events and physical action to move to achieve goals (King, 1981). Individuals 5522; as total organisms responding to their unique perception of the environment and experience (King, 1981). According to King, reactions are not directly observable. Interactions are the acts of two or more persons in mutual presence (King, 1981). Interaction is defined as a process of perception and communication between person and person, represented by verbal and nonverbal behaviors that are goal-directed. In person-to-person interactions, each individual brings different knowledge, needs, goals, past experiences and perceptions, which all influence the interactions. Transactions are defined as goal attainment. Transaction is viewed by King (1981) as an observable behavior of human beings in achievement of valued goals. It involves bargaining, negotiation, and social exchange. It is viewed as growth and development. Transaction for the nurse involves communication with the patient to help identify goals and solve problems. 39 King's model is useful in analyzing communication between the post-MI patient and the nurse. First, the nurse perceives the post-MI patient, using past experience and knowledge. The patient also has perceptions of his illness and his situation, based on his knowledge and past experience. Both nurse and patient perceive each other within the context of their own perceptual realities. Information sharing and collection both Influence perceptions. The post-MI patient may alter perceptions about his disease or health status based on information given by the nurse during patient education and the nurse may change perceptions based on information the patient has offered concerning health beliefs, symptomatology and other concerns. With information collected and perceptions defined, the nurse and the post-MI patient make individual value judgements about the situation. The nurse may decide ”this patient is not taking his disease seriously and may be in psychological denial," for example, or the patient may decide ”I've had a heart attack and I am going to die." Actions, physical or mental, are based on perceptions and judgements. Denial (mental action) or noncompliance (physical action) may be the post-MI patients action response to his perceptions and judgements. The nurse may take specific action in the form of additional patient education, collaboration with other professionals, counseling or other nursing actions. A0 Both nurse and post-MI patient £3255 to each other as total organisms responding to their unique perceptions of the environment and experience. How they react will determine whether or not mutual goals will be established. In the Interaction phase, mutual goal setting takes place. The nurse and the post-MI patient use perception and communication, represented by verbal and nonverbal behaviors that are goal-directed. For the nurse, one goal of choice is usually compliance with the therapeutic regimen. The interaction phase is a planning phase where nurse and patient determine goals and negotiate to establish a mutually acceptable plan towards meeting the goals. Transaction is goal attainment. For the post-MI patient, this may be compliance with the treatment regimen or getting back to work. Transaction is observable behavior of human beings interacting with their environment and is the valuation component of human interactions. Nurses may evaluate their interactions with patients based on the degree of goal attainment or transaction that has taken place. Transaction may involve short-term and long-term goal attainment. As in all systems, feedback is a constant in this process of human interaction. Feedback, in the form of information, is constantly altering perceptions, judgements, actions, reactions, interactions and transactions. Feedback may start at any point in the system. For example, feedback at the interaction phase concerning the unacceptability of a certain plan (e.g., patient has Al foot drop and can only walk slowly and for short distances) feeds back into the system and alters perceptions. The process of interaction continues, altered by feedback (e.g., the nurse and patient negotiate other possibilities for a therapeutic exercise regimen, such as substituting a stationary bicycle). A new plan will be formed, based on new information. Interaction of Kipgip Theory with the Health Belief Model The Health Belief Model is helpful in assisting the nurse to assess the post-MI patients' perceptions and attempt to identify those components and levels of perception that are necessary in achieving the goal of compliance. King's model assists the nurse in the actual therapeutic communication process that is necessary in achieving the goal of compliance together. Components of both of these models are used to create the conceptual framework used in this research. See Figure 6 for a diagram of the conceptual framework used in this study. The following is a description of the model and its' use in this study. The post-MI patient and the nurse are unique human beings. Each of them reacts to, and interacts with individuals and the environment based on their perceptions of the world around them. Their perceptions are influenced by their unique knowledge gained from past experiences. A2 MODIFYING FACTORS \ age \\ education \ work outside the home annual income PERCEIVED SEVERITY \\ number of MI's '———2 OF MI \ time since MI 7 \ in-hospital teaching /I ' current participation in \ teaching program Nurse-Patient Interaction \ additional chronic diseases I L LEVEL OF COMPLIANCE (transaction) NURSING INTERVENTION Figure 6: Integration of Concepts from the Health Belief Model and Kinggi Model as Used in this Study A3 The perceived severity of a myocardial infarction is a subjective state resulting from the interaction of individual perceptions and the influence of modifying factors. This perceived severity may be high or low and, according to the Health Belief Model, is operant in deciding whether the threat to self is sufficient to take action. While a theoretical body of knowledge exists regarding the relationships between perceived severity of myocardial infarction and compliance with the therapeutic regimen, these relationships have not been confirmed through research. The clinical nurse specialist must begin to collect information to test these relationships in the primary care setting. Using the nursing process, data collected regarding perceptions of severity, modifying factors, and other health beliefs can assist the nurse in predicting levels of compliance as well as intervening to alter these beliefs to increase compliance rates. Individual perceptions are formed prior to the nurse-patient interaction (Figure 6). The nurse must understand the post-MI patients entering this phase bring with them the totality of their own individual realities. It is within the process of interaction that nurse and patient develop a common, shared understanding for working toward a goal. During interaction, the nurse and post-MI patient explore and agree upon the means by which compliance with the therapeutic regimen might be obtained (mutual goal-setting) by exploring alternatives and making decisions concerning the means for achieving compliance. With 114 the goal of compliance with the therapeutic regimen identified, the nurse and the patient move toward the transaction phase where goal attainment takes place. A smooth transition between interaction and transaction frequently does not occur. Nurse and patient may continue to share Information, negotiate, plan and revise plans in the Interaction phase for a length of time before transaction ever occurs. The outcome of the model is appropriate behaviors to reduce risks for the post-MI patient. At any point in the model, new information or observations may alter perceptions and the process of nurse-patient Interaction continues, based on new perceptions. A plan or the means to achieve it may thus be changed. The Health Belief Model, then, is a tool by which the nurse, at the point of interaction, is able to assess the post-MI patients perceptions. By knowing if perceived severity is below or above a level theoretically presumed necessary for behavior to occur, the nurse can assess the likelihood that the patient will comply with the post-Ml therapeutic regimen. Thus, interventions and strategies can be appropriately modified. Both Kings' theory and the Health Belief Model assume that change in behavior is possible. As perceptions are identified, steps can be taken to change them and as perceptions are alterable, so too are behaviors. A5 Summary In this chapter, basic background in relation to pathophysiology, diagnosis and treatment for the post-MI patient was presented. The major study concepts of perceived severity of MI and stated compliance with the therapeutic regimen were drawn from the Health Belief Model as a theoretical framework with which to view the post-MI patient. Concepts from King's nursing theory of goal attainment were incorporated into the framework as a means for nursing intervention to increase compliance with the post-Ml therapeutic regimen. In Chapter III, a review of the literature pertinent to the major study variables will be presented. CHAPTER III Review of the Literature Introduction The research question posed in this study is "What is the relationship between perceived severity of myocardial infarction and compliance with the post-MI therapeutic regimen"? In this chapter, relevant literature pertaining to the study variables of perceived illness severity and compliance with the therapeutic regimen will be reviewed. The review will include compliance with the post-MI therapeutic regimen as it relates to medication, diet and exercise, the use of the Health Belief Model, and perceived severity. Compliance Behavior in Health Care Historically, compliance has been a concern of health care practitioners for many years. Hippocrates remarked, ”The physician should keep aware of the fact that patients often lie when they state that they have taken certain medicines" (Yoos, 1981, p. 27). Numerous theories have evolved since that time in an attempt to understand compliance behavior and to improve it. A rather large body of literature is dedicated to this subject. Most authors define compliance as the extent to which the patient follows therapeutic recommendations (Hayes, 1979; Ruffalo, et al., 1985). The term does not Imply judgement or fault by either the patient or health provider but simply indicates that client behavior coincides with health advice. Dracup (1982) has defined compliance as the extent to which an individual chooses behaviors that coincide with a clinical prescription, emphasizing the issue of free will. A7 Some investigators of compliance have stated that the term "compliance" has undeniably authoritarian undertones (DiMatteo 8 Friedman, 1982; Dracup, 1982; Stanitis 8 Ryan, 1982; Yoos, 1981). Yoos (1981) suggested the term implies acceptability of one person having power over another and DiMatteo and Friedman (1982) state that the term implies an obligation on the part of the patient to blindly follow orders. Stanitis and Ryan (1982) state that the term Is unacceptable because it implies guilt on the part of the patient. They fear labeling the patient as "noncompliant" may negatively affect the patient-provider relationship and discourage Independence and decision making on the part of the patient. Stanitis and Ryan (1982) and Edel (1985) have challenged the use of the term ”noncompliance” as a nursing diagnosis. They questioned the appropriateness of a term that they believe views the nurse- patient relationship as one in which power is used by superiors to control or direct subordinates. While the definition was accepted at the Fourth National Conference on Nurse Diagnosis, the Fifth National Conference recommended that less value laden terms or a different approach to the terms be developed (Edel, 1985). "Adherence," "cooperation" and other terms have been suggested as substitutes for the word compliance (DiMatteo 8 DiNicola, 1982; Dracup, 1982). To date, ”compliance" is the most accepted and utilized term in the literature (DiMatteo 8 DiNicola, 1982) and has been chosen for use in this study. The term is not meant to imply A8 judgement or fault by either the patient or health care provider but simply to indicate the extent to which patient behavior coincides with health advice. Researchers and practitioners have provided a number of competing theories to explain the phenomenon of compliance. DiMatteo and DiNicola (1982) place them into three broad categories: intrapsychic factors, environmental factors, and the practitioner-patient relationship. In intrapsychic theories, thoughts, feelings or attitudes are believed to favorably or unfavorably dispose patients toward specific treatments or practitioners. Motivational theories fall into this category. In environmental theories, compliance is believed to be Influenced by environmental factors such as time, money and sociocultural needs. The theories that involve practitioner-patient relationships emphasize practitioner behavior toward the patient. These theories focus on communication patterns, resistance and reciprocation. The problem of patient non-compliance does not lend itself to a single solution by a single technique. Compliance is a multifactorial concept and is undeniably complex (DiMatteo 8 DiNicola, 1982). The Measurement of Compliance There are few valid instruments available for measuring compliance with a therapeutic regimen (DiMatteo 8 DiNicola, 1982; Haynes, 1979; Rudd, 1979). Furthermore, it is difficult to classify A9 levels of actual compliance, particularly when dealing with areas of advice which are varied and of differing levels of specificity (Croog 8 Levine, 1982). Gordis (1979) reviewed measures of medication compliance in the literature and identified direct and indirect measures. Direct measures included methods that detect the drug or its metabolites in blood or urine. Indirect measures included therapeutic outcome, patient interview, pill counts and physician assessment. In Gordis' review, each method was found to have drawbacks. In the use of direct methods, the researcher encounters bioavailability variations. Differences among individuals in absorption, distribution, metabolism and excretion of drugs can reduce the accuracy of compliance measurement. In addition, direct methods of measurement do not detect consistency of administration over time (Dunbar, 1980; Gordis, 1979). Due to costs and difficulties in obtaining specimens for direct measures, indirect measures appear to be more widespread in compliance literature, although no perfect method of indirect measurement has been Identified (DiMatteo 8 DiNicola, 1981; Gordis, 1979, Rudd, 1979). The use of therapeutic outcome (e.g., blood pressure or the presence of angina) poses problems as well. The researcher must keep in mind that compliance is only one variable that may influence physiologic response in individuals (Dunbar, I980; Gordis, 1979). Therapeutic outcomes may be the result of concurrent 50 therapies. For example, medication and exercise programs may reduce episodes of angina or environmental effects (e.g., marital discord) may increase anginal episodes despite compliance with therapies. Glanz (1980) reviewed measures of compliance with dietary. regimens and stressed the importance of attempting to measure eating behavior rather than therapeutic outcome. For example, while serum cholesterol levels may correlate roughly with dietary intake for a given patient, the behavior change is the important factor to concentrate on to get an accurate picture of true compliance. Indeed, for some individuals, cholesterol level may be a poor indicator of dietary compliance regardless of its' importance as a risk factor due to lipid pathophysiology (Brown, Ginsberg, 8 Karmally, 198A; Grundy, 198A). In Gordis' review of medication compliance, physician assessment of patient compliance was reported to be the least accurate method of compliance measurement and was found to be little better than a chance estimate (Grodis, 1979). Undoubtedly, the reliability of physician or provider assessment is influenced by exactly what the physician or provider assesses. Assessments that include patient interview supplemented by therapeutic outcomes and blood tests would probably be considerably more accurate in terms of compliance measurement than just physician impressions, for example. The inaccuracy of physician assessments suggests that physicians tend to rely less on more objective measures of compliance and perhaps more 51 on impressions based on personality characteristics or other characteristics of the patient. More research is needed to determine how physicians make these decisions. Pill counts have several disadvantages in terms of compliance measurement. They do not identify consistency of administration (Dunbar, 1980). In addition, some researchers have had difficulty in obtaining usable pill counts due to incomplete records or difficulty having patients remember to bring their prescriptions to the appointment (Fletcher, Papius 8 Harper, 1979; Rudd, 1979). Gordis (1979) reported that a review of pill count research indicated serious overestimation of compliance as compared to urine or blood tests. Patient interview, in Goris' review, was found to be more accurate in identifying noncompliers and was less valid in identifying compliers. Generally, patients were found to overreport compliance and underreport noncompliance (Gordis, 1979). Haynes, Taylor, Sackett, Gibson, Berhzolz and Mukherjee (1980) found self-report by interview to be generally a more reliable measure of compliance than blood pressure, urine tests or blood tests when compared to pill counts for 135 newly treated male hypertensive patients. A limitation of this study is the assumption that pill counts are an accurate measure of true compliance. In summary, all indirect methods of measurement appear to have drawbacks. In general, self-report by interview is a more reliable measure of compliance, probably because a trained interviewer can elicit more specific information from the patient. 52 Fletcher, Papplus and Harper (1979) studied the feasibility of compliance measurements in a clinical setting In patients with congestive heart failure. Specifically, medication compliance measurement for digoxin therapy was attempted by interview, pill count and serum digoxin concentration (SOC). Selected patients (N - 173) were telephoned the day before their regular clinic appointments and asked to bring all their medications and to refrain from taking medication the morning of the appointment. Prior to seeing their physician, patients were interviewed nonjudgementally about the medication regimen and asked to rate their compliance level as 02, 252, 502, 752 or 1002 of the time. In addition, patients were asked reasons for noncompliance. Following the interview, the number of pills in the digoxin bottle, the date on the bottle, and the number of pills dispensed at that time were noted. Consumption was calculated as the percentage of the number of pills that should have been taken since the data on the bottle. Venous blood was drawn for a determination of SDC. Only 11A patients brought medication bottles to the clinic despite phone call reminders. Pill counts were possible for only 68 patients (392). A steady state SDC could be obtained for 1A7 patients who refrained from taking digoxin that morning and blood samples for SDC were obtained from 1A3 of these patients. The mean SDC was found to correlate with the level of stated compliance but no relation was found between the pill count and the SDC. Actual correlations and levels of significance were not reported. 53 In this study by Fletcher, Papplus and Harper (1979), the researchers concluded that patient Interview IS the most useful method of measuring medication compliance in a clinical setting among chronically Ill patients prescribed digoxin. Interview was determined to be more feasible, easier and more timely that either A pill count or measurement of SOC. Furthermore, the Interview provided additional information in terms of reasons for possible noncompliance. In addition, compliance measurement by interview correlated with SDC values were not reported. Two major problems with the use of pill counts were the Inability to obtain the bottles and the inability to determine the actual date the patients began a new bottle. The findings in relation to pill counts are questionable due to the small number of patients (N - 68, 392 of the sample) for which pill counts were possible. Glanz (1980) reports that a valid assessment of compliance by interview depends on good interview skills and requires careful phrasing of questions to make it socially acceptable to report errors. It is also necessary during the interview to reinforce good accuracy In recall as poor memory will affect the accuracy of compliance measurement (Dunbar, 1980; Glanz, 1980). Researchers have suggested that impressions of family members used In combination with other methods may be helpful in estimating patient compliance (Brownell, 1981; Hilbert, 1985; Lichtenstein, 1979; Miller, 1981). Hilbert (1985) measured compliance with ten aspects of the post-MI therapeutic regimen (medication, diet, weight 5A reduction, stress reduction, exercise, physical activity, smoking cessation, alcohol and caffeine use and work). The data was collected in interview format and scored according to the patients' self-reported compliance from zero (not at all) to four (all the time). Wives were asked similar questions about their husbands' compliance in a separate interview. All interviews were recorded and judges reviewed the recordings to rate compliance on the same zero to four scale. Husbands' scores were found to correlate highly with wives' scores (p<<.001) with exception of alcohol use and physical activity and stress reduction, which were not significantly correlated. The judged ratings and the self-reported ratings were correlated significantly for all ten aspects of the therapeutic regimen (p<:.001). More research is needed to determine whether a family members' stated impression of the patients' compliance is a more useful measure of compliance than the patients' Stated compliance. Dunbar (1980) reported the increasing use of self-monitoring in compliance measurement. Self-monitoring is the observing and recording of one's own behavior. While self-monitoring does overcome the problem of poor memory and may more accurately measure consistency or inconsistency over time, it tends to alter natural compliance patterns. Dunbar suggests that self-monitoring is a better intervention than research method (Dunbar, 1980). This sensitization of the patient to the monitoring of compliance behaviors occurs with most methods of compliance measurement (Rudd, 1979). 55 In summary, compliance is a complex, multifactorial behavior that is very difficult for health care providers to measure accurately. A review of the literature on compliance measurement reveals no ideal method of measurement and none that are significantly more accurate than others. The methods reviewed were direct methods (e.g., blood tests), therapeutic outcome, physician assessment, pill counts, patient self-report, family members impressions and self-monitoring. In general, authors have found self-report to be one of the most satisfactory, particularly self-report by interview, and found physician assessment to be one of the least satisfactory methods. In the next section of this chapter, literature dealing with compliance with the post-MI therapeutic regimen will be critiqued. Compliance with the post-MI medication, diet and exercise regimens will be addressed. Compliance with the Post-MI Treatment Regimen The goal of rehabilitation therapy for the post-MI patient is aimed directly or indirectly at reducing the risk of recurrent MI and sudden death (Wenger 8 Hellerstein, 1978). As previously stated, it is not the intention of this study to explore in detail the scientific validity of commonly accepted modes of therapy prescribed for the post-MI patient. The three general categories of treatments examined in this study are medication therapy, diet therapy and exercise therapy. All three categories have widely accepted modes of therapy that are commonly prescribed for the post-MI patient. In 56 addition, these three categories have been highly standardized in practice by medication protocols, exercise protocols and specific dietary guidelines (Andreoli, et al., 1982; McGurn, 1981). This standardization ensures a greater possibility of uniformity of advice. Each of these three treatment areas will be briefly described In the following sections of this chapter in order to establish the general validity of the given treatments in risk reduction for the post-MI patient. These descriptions will each be followed by a critique of literature Pertaining to compliance with that part of the treatment regimen. The Post-MI Medication Regimen Medication therapy for the post-MI patient is widely variable and dependent upon medical findings and pathophysiologic consequences of the MI. The most common pathophysiologic sequelae following MI are congestive heart failure resulting from decreased contractility of the myocardium and myocardial hypoxia, causing angina or recurrent MI (Wenger and Hellerstein, 1978). In addition, dangerous arrhythmias can cause sudden death, especially in the early post-MI period (McGurn, 1981). For convenience, common post-MI medications can be divided into the following categories: 1) antiarrhythmic agents; 2) agents for ischemic heart disease; and 3) agents for treatment of congestive heart failure. In addition, the post-MI patient may be on antihypertensive agents (McGurn, 1981) or drugs to treat other conditions. 57 A discussion of cardiac arrhythmias and risk reduction specific to drug treatment for each type of arrhythmia is beyond the scope of this study. The effectiveness of drug treatment may vary considerably with the type of arrhythmia present, type of drug prescribed and patient characteristics. The presence of some types of arrhythmias may greatly increase the risk of myocardial infarction and sudden death. Therefore, successful treatment of these arrhythmias may reduce risks considerably (McGurn, 1981). Agents used to treat Ischemic heart disease have the potential to reduce risks of a recurrent MI by increasing myocardial blood flow. The use of beta-blocking agents such as propanolol has been confirmed to reduce mortality in the post-MI patient (Hjalmarson, Herlitz, Malek, Ryden, Vedin, Waldenstrom, Wesel, Elmfeldt, Holmberg, Swedberg, Waagstein, Waldenstrom, 8 Wilhelmsen, 1981; National Heart, Lung 8 Blood Institute, 1981). Calcium channel blockers are the most recent advance in the treatment of Ischemic heart disease and are effective in the treatment of angina pectoris (Schneck, 1985). Calcium channel blockers, however, have not been found to decrease morbidity and mortality for the post-MI patient or reduce the incidence of Infarction in threatened MI compared with placebo. More long-term studies are needed to determine any long range risk reduction benefits (Schneck, 1985). The presence of congestive heart failure, which is often caused by Hi, increases considerably the risk of recurrent MI (Hurst, 1982; Wenger 8 Hellerstein, 1978). The efficacy of digitalis and diuretics 58 in treating and controlling congestive heart failure (CHF) has been well documented (Hurst, I982; McGurn, I981; Wenger 8 Hellerstein, 1978). Vasodilator drugs have also been increasingly used in the treatment of CHF (Goroll, 1981). Antihypertensive agents can reduce the risk of another infarct (and/or congestive heart failure or other health problems) in the hypertensive post-MI patient (Dawber, 1980). The Framingham study has demonstrated the efficacy of treatment for hypertension. The participants in the study (N- 5,18A) were followed bi-anually for more than two decades. Uncontrolled hypertensives were found to have three times as much coronary heart disease and four times as much congestive heart failure as normotensive participants while rates for morbidity and mortality for hypertensives In control were reduced to approximately the rates for normotensives (Dawber, 1980). In other studies, researchers have demonstrated the efficacy of antihypertensive therapy in preventing congestive heart failure, myocardial infarction and other complications (Borhani, 1981; Hypertension Detection and Follow-up Program Cooperative Group, 198A). In the Hypertension Detection and Follow-up Program (HDFP) study pharmacotherapy led to reduction in both the morbidity and mortality of hypertensives across the range of mild, moderate and severe hypertension. The HDFP was a longitudinal seven year study of 10,9A0 hypertensives aged 30-69 years with a diastolic blood pressure greater than 90mm Hg. Subjects were randomized into two groups. Morbidity and mortality outcome were compared for patients treated 59 with optimum antihypertensive drug therapy (the stepped care group) versus patients referred to community physicians for customary care (referred care group). Morbidity and mortality surveys, conducted yearly on all subjects, included interviews and blood pressure measurements. Clinical assessments were performed initially and repeated at two, five and seven years. An overall 16.92 reduction in all causes of mortality was found in the stepped-care group compared to the referred-care group. In addition, mild hypertensives (diastolic blood pressure 90 to 10A mm Hg) who received the stepped-care approach demonstrated a 20.32 reduction in overall mortality (Hypertension Detection and Follow-up Cooperative Group, 198A). In summary, medications commonly prescribed for the post-MI patient have the potential to reduce considerably the risk of recurrent MI and/or other complications. Compliance with these medication prescriptions is, therefore, an important part of risk factor reduction for the post-MI patient. In the next section of this chapter, compliance with the post-MI medication regimen will be reviewed. Compliance with the Post-MI Medication Regimen Although considerable attention has been given to etiology, epidemiology and treatment of heart disease during the acute phases, there has been relatively little research on the long-term, chronic aspects of this major illness. Croog and Levine (1982) conducted a study of the history of men following a first heart attack. They 6O began the study in 1965 at Harvard University School of Public Health. The study was intended to be primarily sociological and psychological in focus and data was collected concerning various aspects of this focus. In this review, the focus will be on the compliance aspects of the study. Croog and Levine (1982) examined compliance with regard to ten principle areas of physician advice: medications, diet, smoking, work, weight control, resting, alcohol consumption, avoiding quarrels, exercise and sexual activity. Because this study is concerned with compliance in the areas of medication, diet and exercise, only those areas will be reviewed. The background of the study and compliance with medications will be addressed in this section of the literature review. Diet and exercise compliance will be reported in the sections of the literature review dealing with those areas. Croog and Levine (1982) defined ”compliance orientation" as the degree of expressed willingness of the patient to comply with an item of advice as indicated by his report of past and/or intended compliance behavior. They based their rationale for using patient reporting upon difficulties measuring and reporting the actual behaviors and past research which shows a positive compliance orientation is more strongly associated with actual compliance than Is an ambivalent or negative intention. 61 The study population used in this longitudinal, prospective survey-type design was intlally 3A5 males between the ages of 3D and 60. Subjects were limited to those who experienced a first MI and had no history of previous illness. Interviews were carried out by specially trained interviewers, each of whom held a master's degree In social work. The interviews occurred at week seven post-MI (N - 3A5), year one post-MI (N - 293) and year eight post-MI (N - 205). In the interviews, patients were given a list of areas on which physicians commonly give advice to persons with heart disease. The list included such areas as medication, diet, exercise, smoking, and the like. The men were asked to specify the items on which they had received advice. For each of these, they were then asked, "How well have you been able to follow the doctors advice"? Responses were coded: "completely, "for the most part," "somewhat," or ”not at all." The data were reclassified in terms of three categories for analysis: complete compliance, partial compliance and noncompliance. "For the most part" and ”somewhat" were treated as partial compliance. For some analytic purposes, the researchers further combined responses into positive compiler and noncomplier. The positive compiler was defined as the patient who reports complete compliance with the advice. Compliance was highest In the category of medication advice. The percentage of patients classified as positive compliers were 88.3% at week seven, 90.22 at year one and 88.52 at year eight for those who 62 indicated that they had received advice concerning medications. Compliance remained stable over the eight years for those patients on medication therapy while it generally declined over the eight years in other categories of advice. Compliance orientation with regard to medications was cross-tabulated against the Individual independent variables of age, occupation, education level, personality characteristics, illness severity and previous compliance orientation. Because nearly 902 of the patients were reported positive compliers, the number of noncompliers was insufficient for most analyses and the analysis produced relatively few statistically significant associations. No significant levels were reported. Older men reported significantly higher positive compliance orientation at year eight, which is generally not consistent with other studies. The respective gamma measures of association between age and compliance orientation were reported as weight control (.3A), dietary practices (.AA) and the use of medications (.A8). It should be noted that age was not found to be related to measures of health and functional capacity. In general, compliance was found in all categories to be unrelated to educational level, occupation, knowledge of etiology of heart attacks (e.g., smoking, obesity), current health status, psychological self-rating and number of items of advice given by the physician. In addition to age, the only other variable found to be related to compliance in this study was previous compliance 63 orientation. The findings present suggest evidence about the persistence of particular attitudinal orientations over extended periods of time. The authors cite possible lack of adequate sensitivity of the compliance instrument as a limitation of the study that may have resulted in the lack of variables relating to compliance. The authors, however, also note that with the exception of the findings with regard to age, findings of this study seem to be congruent with other compliance research. This study has significant importance as the first long-term longitudinal study on compliance of post-MI patients with therapeutic regimens. Miller, Wikoff, McMahan, Garrett and Johnson (1982) developed a tool to assess attitudes of patients with cardiac disease toward performing prescribed behaviors of their medical regimen. Background of the study is described here as well as findings related to medication compliance. Compliance with other aspects of the treatment regimen will be discussed in the sections of the literature review dealing with those aspects. The Miller Attitude Scale was developed in several stages. First, an open-ended questionnaire was used in which 27 patients with heart problems were asked the advantages and disadvantages of performing the actions of the medical regimen. This data was used to develop a semantic differential with biopolar adjectives. Content validity was established by experts in the disciplines of nursing, medicine and social psychology. 6A Two groups of patients with heart disease were used to examine validity and reliability of the Miller Attitude Scale. One group included A80 members of Mended Hearts, Inc., and the second group consisted of 35 patients diagnosed with a first myocardial infarction who were ready for home discharge. The second group repeated the attitude scale six months post-hospitalization. Performance of the medical regimen by this group was verified at the six-month follow-up period. Degree of adherence behaviors for all actions of the medical regimen was measured using a Health Behavior Scale at the six-month follow-up. The Health Behavior Scale, a five-point Likert scale from ”unlikely follows" (1) to I'likely follows” (5), was developed by the investigators to assess actual behavior of the subject in performing or not performing prescribed actions of the medical regimen. The actions polled in the questionnaire for the situations of home, work, social and sports were: 1) following diet; 2) stopping smoking; 3) performing activity; A) taking medications; and 5) altering stress response. The family member or significant other designated by the patient provided a measure of the accuracy of the patients responses by completing a reworded version of the same scale. Alpha reliabilities for the scales in the Health Behavior tool ranged from .81 to .99. Attitude and adherence behavior scores for Group Two (N - 35 post-MI patients) were correlated using Spearman rank correlations. Attitude and behavior were not significantly related for medication 65 (r - .12). Significant relationships, however, were found (r - .81, ;><:.001) between patient adherence responses and significant other responses for medications. Initial testing of the Miller Attitude Scale indicated it to be a valid and reliable tool for eliciting personal and situational factors that affect patient compliance. The authors suggested further studies to test the tool and used it in a later study published in 1985. The Miller Attitude Scale (MAS) and three other scales were used by Miller, Wikoff, McMahan, Garrett and Ringel (1985) to investigate the relationships between demographic and medical variables, attitudes, perceived beliefs of others, and intentions toward medical adherence and actual post-hospitalization regimen adherence in 112 patients recovering from a first MI. The 112 subjects (87 men and 25 women) were selected from five institutions that provided cardiac rehabilitation programs. All programs had similar formats and information but varied in the sequence of content and program administration. In all five institutions, multidisciplinary teams instructed patients on dietary restrictions, medications, cessation of smoking, activity progression and reduction of stress. Subjects were all first time MI patients without complications, and were 30 to 70 years of age, literate and able to walk. Four instruments were administered to the study sample: The Miller Attitude Scale (MAS), Health Intention Scale (HIS), Health Behavior Scale (H85), and a demographic and medical data form. The 66 MAS is a 12-ltem, seven-point semantic differential scale that measures attitudes toward aspects of the therapeutic regimen (medication, stress, activity, smoking, diet). Cronbach alpha reliabilities of the five subscales for this sample were: medication .90; activity .85; stress .83; and diet .7A. The HIS is a five-point Likert scale ranging from one (unlikely) to five (likely). In Part A, subjects indicate their intentions to perform the medical regimen. In Part 8, subjects indicate their beliefs about which actions of the medical regimen significant others think they should perform. Cronbach alpha reliabilities of the five subscales for Part A and B ranged from .65 to .98. The HBS, a modification of the HIS, Is a five-point Likert scale ranging from one (unlikely) to five (likely). Intention statements in the HIS group were changed to behavior statements, e.g., ”if I am at home, I follow my diet" was changed to "if I am at home, I will follow my diet." A similar HBS scale was also included for completion by the significant other. Part I of the demographic and medical data forms contained sociodemographic data as well as blood pressure, medical regimen prescriptions, smoking history, history of heart disease and similar data. Part II contained information on repeat hospitalization, repeat heart attack, frequency of angina, weight and vital signs. Form III was completed if a patient died and listed cause of death, health problems and similar information. 67 Immediately prior to dismissal from the hospital, the MAS and HIS were administered by a trained investigator. Demographic and medical data was collected. Six to nine months later, the MAS, H85 and demographic and medical data Part II were administered to patients in their homes by a trained investigator. The Health Behavior Scale-Family (HBSF) was also completed by a family member at that time. Correlations between patient responses and responses of family members were moderate to high (diet: t - 1.10, r - .76; activities: '6 . -.77, A - .60; medications: t - -.33, r - .66). The data were submitted to multiple regression analysis. In summary, the researchers found that: 1) during hospitalization, attitudes and perceived beliefs of others were strong indicators of intentions to adhere to the medical regimen, but they were pp; indicators of actual adherence post-hospitalization; 2) post- hospitalization, attitudes and perceived beliefs-of others were strong indicators of actual regimen adherence; 3) all five areas of the treatment regimen correlated significantly with the hospitalized patients attitudes and perceptions of the beliefs of the significant other concerning the subjects intentions; and A) no significant relationships were found between adherence and the sociodemographic and medical variables (sex, age, education, occupation, blood pressure, weight, and amount smoked). In regard to findings related specifically to the medication regimen: 1) post-hospitalization adherence was not significantly related to intentions (self or others) or attitude during 68 hospitalization; 2) attitudes at home and during hospitalization were significantly related but correlations were small (r - 1.87, ;><:.02); 3) correlations between perceived beliefs of others during hospitalization and at home were significant (r - .212, p‘<:.01); A) no correlations were found between adherence to the medication regimen and sociodemographic and medical variables (sex, age, education, occupation, blood pressure, weight and amount smoked). Two findings from the Miller et al (1985) study have particular significance to the present study. First, Miller et al (1985) confirmed the findings of past research with regard to sociodemographic variables (no relationship was found between sociodemographic variables and compliance). Second, Miller et al (1985) illustrated changes in intent to comply while in the hospital and after discharge to the home. Changes in intent are probably related to changes in perceptions, addressed in the present study. Greene, Weinberger, Jerin and Mamlin (1982) investigated the relationship between medication compliance and 1) the perceived support by social and familial systems, 2) the perceived susceptibility and severity of an illness, 3) knowledge about the general nature of the illness, A) knowledge regarding the specific components of the regimen, 5) the complexity of the regimen, 6) patient satisfaction, and 7) demographic characteristics. The data were collected from patients receiving care at an outpatient department of a municipal teaching hospital. Since the goal was to survey patients with chronic medical conditions, only those who had 69 had three appointments in the last two years and a fourth scheduled in the future were considered. Selection criteria included only those over age 18 who did not reside in a nursing home and were not being treated for psychiatric disorders. A computerized medical record system generated the population of patients who met these criteria. There were 327 subjects selected at random for the study. Data were collected from the patient's medical record and an in-home interview. In the interview, lasting approximately A5 minutes, information was collected regarding the patients' social support systems, health beliefs, opinions about the physician and the most recent medical encounters, perceptions of the facility in which the care was obtained, knowledge about and attitudes toward their medical conditions, and knowledge of the specific components of their regimens. Only the data dealing with medications were considered in the current report. The data were examined through correlational and path analysis. The researchers found significant relationships between medication compliance and general knowledge of illness, social and familial support, knowledge of the regimen, complexity of the regimen (Inversely related to compliance), and overall satisfaction. Relationships that were not significantly related to medication compliance were perceived susceptibility and severity of illness, having relatives with similar problems and demographic characteristics (age, sex, income, family size, education, years of 70 illness). These findings cannot infer causality. They do, however, suggest relationships that may be scrutinized further in prospective cohort studies. Mayou, Foster and Williamson (1979) conducted interviews with post-MI patients to investigate medical advice received, compliance and psychological and social problems in convalescence. One hundred patients and their spouses were interviewed during hospital admission, two months after discharge and one year after discharge. Except for the first patient interview, which took place in the hospital, all interviews were conducted in the patients' home and were taperecorded. The researchers stated that uncertainty as to exactly what advice had been given limited the assessment of compliance. Compliance was reported as most satisfactory for medication, least satisfactory for diet and activity. Compliance was poor if it involved persistent effort by the patient (e.g., diet) or required an understanding of principles. Compliance in one area was reported to be unrelated to compliance in all other areas. Limitations of the study include a lack of description of the type of questions asked, method of data collection during the interviews and statistical techniques used. The findings regarding better compliance with medications than with diet and activity are consistent with other literature (Croog 8 Levine, 1982). 71 In summary, researchers suggest that compliance with the post-MI medication regimen can reduce morbidity and mortality for these patients through risk factor reduction. Compliance with medications is usually better than compliance with other aspects of the treatment regimen and tends to remain stable over time while compliance with other regimens decreases. Compliance with medications is not generally related to sociodemographic variables, although one study found older subjects tended to comply more (Croog 8 Levine, 1982). Conflicting findings were found In studies relating medication compliance with knowledge of disease, social support, complexity of the regimen and previous compliance behavior. In the next section of this chapter, the typical post-MI dietary regimen is described. Literature concerning compliance with the post-MI dietary regimen is then reviewed. The Post-MI Dietary Regimen Diet therapy for the post-MI patient may involve the treatment of other health problems. Examples are the ADA diet for diabetics, low salt diet for hypertensive patients and reduction in fats and calories for the obese patient. Since diabetes, obesity and hypertension are risk factors for myocardial infarction (Dawber, 1980), such a diet may be part of risk-reduction management for post-MI patients. For the MI of atherosclerotic etiology, a diet low in fats, particularly cholesterol, is frequently recommended to reduce the level of fats in the blood serum and, thus, reduce the risk of further development of atherosclerosis and recurrent MI. 72 Considerable attention has been given to the status of blood cholesterol as a risk factor for coronary atherosclerosis and Hi. It is widely accepted that a high level of cholesterol In the blood (greater than 220 mg percent) is a risk factor for NI (Brown, Ginsberg, 8 Karmally, 198A; Dawber, 1980; Grundy, 198A; Levy, 198A; Rifkind, 198A; Tyroler, I98A). In the Framingham Study, young men, ages 30-39, with cholesterol levels of 260 mg percent or more had four times the risk of heart attack compared with those whose levels were less than 200 mg percent. This ratio dropped to about double in the 50-59 age group (Dawber, 1980). The "Lipid Research Clinics Coronary Primary Prevention Trial (LRC-CPPT)" tested the efficacy of lowering cholesterol levels in reducing the risk of coronary artery disease (CHD) in 3,806 asymptomatic middle-aged men with primary hypercholesterolemia (Rifkind, 198A). In this randomized, double-blind clinical trial, the group treated with cholestyramlne (a cholesterol-lowering drug) had 8.52 to 12.62 greater reductions in total and low-density lipoprotein levels than the placebo group and a 192 reduction in risk of myocardial infarction (Rifkind, I98A). The American Heart Association is now recommending that everyone eat a diet low in saturated fats and cholesterol and keep the serum cholesterol level below 200 mg percent. Fat intake should not exceed 302 of total calories per day. For patients with serum cholesterol level near 200 mg percent, dietary changes should be strongly emphasized and for those with levels that consistently run from 2A0 73 to 260 mg percent or greater despite dietary therapy, consideration should be given to the use of cholesterol-lowering drugs such as cholestyramlne (Levy, 198A). The American Heart Association's current recommendation for cholesterol consumption is 300 mg/day for men and 225 mg/day for women (Levy, 198A). In summary, high blood cholesterol has been determined to be a risk factor for myocardial infarction. Increasing levels of serum cholesterol are positively correlated with increasing risk. Compliance with a low cholesterol diet can reduce the risk of recurrent MI by reducing serum cholesterol. In the following section of this literature review, studies investigating compliance with the post-MI dietary regimen are reviewed. Compliance with the Post-MI Dietary Regimen Very little research is available investigating compliance with the post-MI dietary regimen. Therefore, studies of dietary compliance in other patient populations were included in this literature review. Croog and Levine's (1982) longitudinal prospective, survey-type design described earlier had very different results In the area of diet compliance when compared to those reported for medication compliance. At week seven (N - 3A5), 63.A2 of those who reported they had received dietary advice complied with that advice. At year on (N - 293), that percentage dropped to A8.82 and at year eight (N - 205), It was 3A.82. A pattern of reduction in compliance over time was evident. The researchers stated that these data suggest 7A differences in the level of concern or anxiety about the illness over time. There was, however, a high consistency between positive diet compliance orientations reported at week seven and at year eight, suggesting the persistence of attitudinal orientations over extended periods of time. As mentioned earlier, age was the only variable found to be related to compliance, with older individuals reporting higher positive compliance orientation. The Miller Attitude Scale (MAS) developed by Miller et al (1982) was described earlier. The MAS (a 12-item, seven-point semantic differential scale that measures attitude toward aspects of the therapeutic regimen) was investigated for correlations with the Health Behavior Scale (HBS, a five-point Likert scale that assesses actual adherence behavior). Miller et al administered the scale to 35 subjects at six months post-MI. A family member also completed a reworded version of the H85 in order to provide a measure of the accuracy of patient responses. Spearman rank correlations indicated that attitude and behavior were significantly related for the diet subscale (r - .31, p <1.05). In addition, a significant relationship was found between patient responses and significant other responses for the diet subscale (r . .85, p‘<:.001). Replication with a larger sample would allow the use of multivariate statistical techniques such as multiple regression analysis, which would be necessary to verify that attitudes, as measured by the MAS, are predictive of dietary compliance. 75 Miller et al (1985), used a larger sample (N - 112) in their second published study using the Miller Attitude Scale. The study was described in detail earlier. The researcher used the MAS and H85 and also included a Health Intention Scale (HIS) and a sociodemographic and medical questionnaire. Data were collected before hospital discharge and six to nine months later. The data were submitted to multiple regression analysis. The researcher found that: 1) post-hospitalization adherence to diet was significantly related to patients' own intentions to comply (r - .33, p‘<:.001); 2) attitudes toward the dietary regimen during hospitalization and at home were significantly related but the correlation was small (r - .236, p <1.01); 3) correlations between perceived beliefs of others during hospitalization and at home were not significant for the diet subscale; and A) there was no significant relationship between dietary adherence (or intent to adhere) and sociodemographic and medical variables (sex, age, education, occupation, blood pressure, weight and amount smoked). Unfortunately, cholesterol levels were not reported in this study. Glanz, Kirscht and Rosenstock (1981) conducted a longitudinal study of A32 hypertensive patients to determine the effect of written messages, nurses phone calls, self-monitoring and social support on patient compliance. Data were collected through patient interviews and supplemented by information obtained from medical and pharmacy records. The patients were then randomly assigned to each of four sequential intervention experiments. 76 Measures of adherence were derived principally from interview self-report items. For each element in the regimen of a given respondent, such as medication or diet, a series of questions were asked. From the responses to these items, indexes of aherence were constructed, reflecting the extent to which the respondent claimed to follow the regimen exactly. At the time of the initial interview, only 262 of the subjects reported faithful adherence to dietary recommendations regarding sodium restriction and only 122 had lost the weight recommended by the physician. Intervention one (written messages) consisted of a control group, a threat message group and a positive message group. The messages were informational tabloids that presented organized material on hypertension and its treatment. Intervention two (nurses phone call) consisted of a control group and a group that received a phone call from a nurse who asked them about their treatment regimen and gave verbal reinforcement. Intervention three (self-monitoring) consisted of a control group and a treatment group. The treatment group kept a diary of treatment regimen behaviors. Intervention four (social support) included a control and treatment group. In the treatment group, subjects met with a nurse and the support person of their ehoice to outline an action plan to enlist the social support person's help with some aspect of the treatment regimen. Glanz et al (1981) stated that while subjects often reported that the interventions were Interesting and helpful and that they took increased health action of some sort based on the interaction, actual 77 adherence did not increase with any of the interventions. In relation to nurse phone calls, it was found that the number of reinforcements given was negatively correlated with dietary adherence (p‘<:.05), suggesting that more reinforcements might negatively affect adherence. In addition, one-third of the subjects did not recall the nurse reminders. A major limitation of this study regarding compliance with a dietary regimen is the reported low number of subjects (203 subjects, A72 of the sample) who had dietary restrictions. More research of this type is needed for subjects on low cholesterol diets. A study by Hyman, Insull, Palmer, O'Brien, Gorden and Levine (1982) assessed the relative validity of laboratory and non-laboratory measures of compliance with a fat-controlled diet by hypercholesterolemic males. Non-laboratory measures of compliance included three day food records, structured interviews that utilized three day food recall, and compliance ratings by nutritionists and interviewers. Laboratory measures of dietary compliance included saturated fat, linoleic acid, polyunsaturated fat, polyunsaturated fat/saturated fat ratio and cholesterol. The sample consisted of 16A subjects between the ages of 21 and 59 years who had a fasting total serum cholesterol exceeding the age specific value for the approximate 95th percentile of the population screened by a New York City laboratory. Baseline serum lipid values were obtained while subjects were on their habitual diets. Then subjects were instructed on a fat controlled diet which included the 78 following: total fat intake 30-3A2 of total calories, saturated fat 5-72, ratio of polyunsaturated to saturated fat greater than or equal to two, carbohydrates 372-5A2 of calories, protein 182-262, and total cholesterol less than 300 mg/day. Patients were instructed in keeping the food records. After approximately 13 weeks on the fat-controlled diet, lab measurements were repeated. Patients were interviewed about ten days after they had filled out their food records. Compliance to diet was assessed by analyzing the three day food record, three day food recall obtained by interview, and summary ratings of compliance on a five-point scale (done before calculations of the food record). Laboratory and non-laboratory methods of compliance measurement were investigated for significant correlations. There were statistically significant (at p <:.05 level) correlations between changes in serum cholesterol and changes in nine of the ten non-laboratory measures of compliance. Compliance measures were less valid for less-educated subjects and were more valid for younger subjects. Younger subjects may have been better able to fill out food records and respond accurately to the interviewer or may have had lipid levels more responsive to dietary changes. An Important conclusion was drawn from this study: for patients on a fat-controlled diet, non-laboratory measures of compliance may be as effective as laboratory measures and combinations of both may be a more valid measurement of compliance. A limitation of this research in reference to the study of dietary compliance is the lack 79 of reporting for the compliance rates using the non-laboratory methods. Only correlations with blood values were presented. An experimental study utilizing a control group and various experimental treatment groups (e.g., three day recall, three day diary, social support, etc.) would yield more information about individual contributions of treatment to compliance. In summary, compliance with post-MI dietary recommendations is reportedly lower than compliance with medications and decreases over time. Croog and Levine (1982) reported a 632 self-report dietary compliance rate at seven weeks post-MI which decreased to A92 at one year post-MI and to 352 at eight years post-MI. Post-hospitalization, dietary compliance has been reported to be significantly related to patients own intentions to comply. While in the hospital, patients intentions were reported to be related to dietary compliance but the correlations were small. In addition, family members estimates of patient compliance have been found to be significantly correlated with the patients' dietary compliance. Sociodemographic variables have been found to be unrelated to dietary compliance with the exception of age in one study. Croog and Levine (1928) reported that older subjects tended to comply more. Interventions such as written messages, self-monitoring and social support were reported by Glanz (1981) to be unrelated to dietary compliance for hypertensive patients. In addition, Glanz found a negative relationship between nurse phone call reminders and compliance among subjects. For hypercholesterolemic males, 80 laboratory measures (serum cholesterol) and non-laboratory measures (food records, interviews utilizing food recall and compliance ratings) of dietary compliance were reported by Hyman et al (1982) to be significantly correlated. In the next section of this chapter, the typical post-MI exercise regimen Is described. Literature concerning compliance with the post-MI dietary regimen is then reviewed. The Post-MI Exercise Regimen Today, early ambulatlon Is an accepted practice for patients with uncomplicated myocardial infarction (Winslow 8 Weber, 1980). Exercise testing is frequently used to evaluate cardiovascular function objectively, and exercise training programs are instituted to improve the coronary patient's functional capacity (Wenger, 1979). After physical conditioning, myocardial oxygen requirements are lowered due to the reduction in heart rate and systolic blood pressure at submaximal workloads. Physical conditioning thus permits coronary patients to engage in more vigorous physical activities before reaching their ischemic threshold. The reduction In heart rate also enhances the myocardial oxygen supply. Since coronary blood flow is interrupted during systole and diastole is disproportionately shortened during tachycardia, a decreased heart rate will extend coronary perfusion time and, thus, enhance myocardial oxygen supply (Dehn, 1980). 81 Although exercise training does improve cardiovascular fitness and may enhance the quality of life for some people, there is not sufficient evidence avaiable yet to prove that it will prevent coronary mortality and morbidity (Dehn, 198D). Retrospective studies have shown, however, that there is an association of lower cardiovascular mortality and morbidity and mild to moderate activity, as compared to sedentary life-styles (Bruce, 197A; Dawber, 1980; Hirsch, 198A; Oberman, 1985). Shepherd, Corey, and Kavanagh, (1981) reported that they found three exercise sessions per week to have a favorable effect upon post-MI prognosis. The 610 post- coronary patients who participated were followed for an average of 36.5 months. Sustained physical activity was found to be associated with a five-fold improvement in the odds ration for both fatal and non-fatal recurrences of the infarct. Central to a cardiac conditioning program is the concept of prescribed, individualized exercise, with prior multilevel exercise stress testing required for the safety and accuracy of the exercise prescription (Wenger, 1979; Wenger 8 Hellerstein, 1978). Through research, a range of safe and effective cardiovascular stress has been identified as 602 to 802 of an individual's functional aerobic capacity as determined by exercise stress testing (American Heart Association, 1985). Since there is a direct relationship between oxygen uptake and heart rate, this is the most convenient and practical way to prescribe work intensity. Maximal heart rate can be determined by exercise stress testing and it is a simple matter to 82 compute the 602 to 802 of that rate. Exercise sufficient to maintain the heart rate between those limits correspondingly generates the desired aerobic performance (Comoss, Burke, 8 Swails, 1979; Dehn, 1980; Wenger 8 Hellerstein, 1978). To achieve the desired training effect, patients should exercise at least two to three times weekly for approximately 20 to 30 minute sessions with include a warm-up and a cool-down period (Dehn, 1980; Hellerstein 8 Franklin, 1978). In summary, sustained physical activity has been shown to be associated with decreased risk of recurrent MI. Compliance with the prescribed exercise regimen may, therefore, be a factor in reducing these risks for the post-MI patient. In the following section of this chapter, studies investigating compliance with the post-MI exercise regimen are reviewed. Compliance with the Post-MI Exercise Regimen In Croog and Levine's (1982) longitudinal prospective survey-type design described earlier, reported compliance with exercise regimens among those patients who reported they had received an exercise prescription was as follows: At week seven (N - 3A5), reported exercised compliance was 75.62. At year one (N - 293), reported compliance with exercise regimens dropped to 51.62. At year eight (N - 205), reported compliance was A92. Those patients who were exercise compliers at year one were compliers at year eight, suggesting the persistence of attitudinal orientations over time. Older subjects reported higher positive compliance orientation. 83 Croog and Levine's (1982) study is valuable as an indication of post-MI exercise compliance over time. The researchers did not compare data on previous compliance rates at week seven and year one for those subjects who had died by year eight with those subjects who survived through year eight. They did report, however, that exercise compliance, as well as compliance with other areas of medical advice, was not associated with current health status as measured by reported symptoms, physician ratings of functional capacity, recency of visits to a physician, and hospitalization for heart disease during the two years prior to the year eight interview. Shepherd, et al (1981) investigated the relationship between exercise compliance and the prevention of a recurrence of MI. The subjects were 610 middle-aged men, recruited to an exercise program an average of 8.2 months (range of two to 109) following a MI. Observation of these men was continued for an average of 36.5 months, keeping note of non-fatal recurrences (total of 21); fatal recurrences and other "cardiac deaths" (total of 23); and deaths from other causes (total of 12). The emphasis of the exercise program was on progressive, long, and slow distance training. Each subject was given a personal prescription specifying a daily distance, a time in which to cover this distance, and a description of limiting symptoms where appropriate. The aim was to bring subjects to a level of fitness where they could cover A.8 km in 30 minutes if under A5 years of age, or in 35 minutes if over A5 years of age. 8A Adherence to the prescribed exercise was monitored by use of an exercise log noting distances, times, symptoms and pulse rates obtained. Exercise was also observed at the rehabilitation center weekly for two years, and, thereafter, every eight weeks. Laboratory exercise tests were repeated semi-annually or as indicated by symptoms or failure to progress. A total of 505 patients (82.82 of the sample) continued to fulfill their prescription at least three times per week. Of the remaining 105 men, 78 claimed to be exercising at least twice per week, while the remaining 27 were taking little or no exercise. The researchers considered all 105 men who were taking two sessions or less per week as "non-compliant." Shepherd et al (1981) analyzed the data in separate categories for smoking behavior and disease severity (as indicated by angina, exercise-induced ST-semental depression, cardiac enlargement or ventricular aneurysm). Stratification of the data indicated that in almost all categories of subject, sustained physical activity was associated with a five-fold improvement in the odds ration of both fatal and non-fatal recurrences of the infarction. The risk of a recurrence of M1 for an individual who did not exercise, relative to one who did exercise, was measured by the odds ratio. Therefore, a conservative approach must be taken to estimates of statistical significance. In addition, the researcher stressed that the association between poor compliance with a prescribed exercise regimen and an adverse prognosis following myocardial infarction has 85 only been demonstrated for those referred to a rehabilitation center, and does not necessarily apply to all patients who have sustained myocardial infarction. Mulder (1981) investigated compliance with home exercise programs. The purpose of Mulder's study was to determine a method for family physicians to predict home exercise compliance by four weeks of patient participation. The subjects were randomly selected from individuals who presented for outpatient exercise stress testing. Twenty-two men and seven women agreed to participate. Five of the males were classified as "cardiac patients” and the rest of the subjects were basically healthy. An exercise stress test was performed and aerobic exercise was prescribed to raise the pulse to 782 to 852 of the maximum rate for 30 to 60 minutes, three times per week. Telephone contact was made with each of the subjects every four to six weeks. During this phone interview, the time spent exercising, types of exercise activity used, and pulse rates attained were assessed. Problems that arose were also discussed and modifications in the exercise prescription were made as needed. Each participant also maintained a log of exercise activity and pulse rates achieved. This was used to confirm information received during the telephone interviews. Based on data obtained from follow-up interviews and logs, determinations were made as to the extent to which participants were exercising. Compliance was defined as the percentage of the prescribed exercise goal that was actually achieved. Those 86 accomplishing less than 752 of the prescribed exercise goal were classified as non-compliant. The researchers had hypothesized that non-compliance to the exercise program could be predicted by lack of motivation, lack of understanding of disease, alcohol abuse, poor exercise compliance at four weeks, lack of regular scheduling of exercise and other specific reasons. For the purposes of this study, it was predicted that any patient with two or more of these factors at four weeks would be non-compliant and patients with none or one of these factors would be compliant with a long-term exercise program. Compliance was evaluated at 32 weeks of participation in the program. Twenty-nine subjects (55.22) achieved greater than 752 of their exercise goal at 32 weeks. All compliant and non-compliant individuals were accurately predicted. A negative correlation was found between compliance and length of time in the program. The only other factor found to have a significant correlation (at the p <1.05 level) with compliance was sex; while only ten of 22 male participants in the program were compliant, six of seven females demonstrated compliant behavior. Mulder's (1981) study has significance as one of the few studies on exercise compliance that used home exercising subjects rather than subjects enrolled in a rehabilitation program. The study also has practical significance for use in predicting compliance in a primary care setting. A major limitation of the study is the small number of subjects included (N - 29). Replication with a larger sample of 87 recovering MI subjects would yield useful information about home compliance in the post-MI population. The Miller Attitude Scale (MAS) developed by Miller et al (1982) was described earlier. The MAS (a IZ-item, seven-point semantic differential scale that measures attitude toward aspects of the therapeutic regimen) was correlated with the Health Behavior Scale (HBS, a five-point Likert scale that assesses actual adherence behavior). Miller et al administered the scale to 35 subjects at six months post-MI. A family member also completed a reworded version of the HBS in order to provide a measure of the accuracy of patients responses. Spearman rank correlations indicated that attitude and behavior were significantly correlated for the subscales of activity (r - .A6, p <1.01). Patient responses and significant other responses were also significantly correlated with the activity scale (r - .25, p <1.05). As identified earlier, small sample size was a limitation of the study. The MAS was used by Miller et al (1985) with a larger study sample (N - 112). .The study was described in detail earlier. The researcher used the MAS and H85 scales and also a Health Intention Scale (HIS) and a sociodemographic and medical questionnaire. Data were collected before hospital discharge and six to nine months later. The data was submitted to multiple regression analysis. The researcher found that: 1) post-hospitalization, adherence to activity regimens was significantly related to the patients own intention to comply (r - .26, p <:.OOI); 2) attitudes at home and 88 during hospitalization were significantly related (r = .335, p'<:.Ol); 3) correlations between perceived beliefs of others during hospitalization and at home were significant for the activity subscale (r - .3OA, p‘<:.OOl); and A) there were no significant relationships between compliance with the activity regimen and the sociodemographic and medical variables of sex, age, education, occupation, blood pressure, weight, and amount smoked. A limitation of the Miller et a1 (1982, 1985) studies is that the researcher did not define activity as including a specific exercise regimen. Since the sample included only first time MI patients without complications who could walk, an assumption could be made that a typical post-MI exercise regimen was probably prescribed for the majority of the subjects. In addition, the authors did not report the results of correlations between exercise compliance and angina in subjects. In summary, compliance with post-MI exercise regimens is reported to be approximately 752 and decreases to half by the first year post-MI. Compilers at year one tend to comply at year eight post-MI. Older subjects have been reported to comply more in one study (Croog 8 Levin, 1982), but other sociodemographic variables are unrelated to exercise compliance in the post-MI subject. Compliance with the post-Ml exercise regimen is generally not related to actual functional capacity and symptomatology. Attitude towards compliance was found to be significantly related to actual compliance with activity regimens and patient and significant other responses to compliance questionnaires were correlated as well. 89 In the next section of this chapter, literature concerning the Health Belief Model is reviewed. The variable of perceived severity of MI is included in this review. The Health Belief Model/Perceived Severity The original Health Belief Model, according to Rosenstock (197A), proposed that the likelihood a person will take health action is determined by l) psychological readiness to take action, 2) belief that the individual is personally susceptible to a particular problem or disease, 3) belief that the problem or disease has at least moderate severity on some component of his life, and A) perceived benefit of action weighted against perceived barriers to performing that action. The Health Belief Model was modified by Kasl (197A) for explaining behavior related to chronic disease. In the modified model, perceived susceptibility refers to complications (including death) of the disease and benefits and barriers refer specifically to the treatment regimen. These factors, along with cues to action, motivation and sociodemographic variables, determine the perceived impact of the chronic disease, which influences an individuals' likelihood of complying with the recommended treatment regimen (Kasl, 197A). A number of literature reviews are available which review research on the Health Belief Model, report significance of the variables of susceptibility, severity, benefits and barriers, and make recommendations for further research (Janz 8 Becker, 198A; Jette 90 8 Rosenstock, I978; Loustau, 1979; Mikhail, 1981; Stunkard, 1981). These reviews provide substantial evidence of the utility of the Health Belief Model in predicting health actions. Diversity of measures, however, makes it difficult to establish validity of measures and comparability among studies. Janz and Becker (198A) provide the most recent review of the Health Belief Model studies. They summarize that for the majority of cases each Health Belief Model dimension was found to be significantly associated with the health-related behaviors under study. The significance ratios for all studies using the Health Belief Model dimensions prior to 198A were reported as: barriers (892), susceptibility (812), benefits (782) and severity (652). While severity was significant in only 652 of the total studies, it jumped to 882 for sick role behavior studies. Janz and Becker (198A) concluded that perceived severity had increased importance in understanding sick role and chronic disease behaviors when compared to preventive health behaviors because well subjects probably have more difficulty conceptualizing a severity dimension when 1) they are asymptomatic, 2) the health threats are long-term (getting cancer from smoking, for example) and 3) the subjects have no personal experience with the medical condition. For patients with a diagnosed chronic disease, severity becomes an important concept because they are coping with the Illness at the present time with knowledge that the illness will continue into the future. Janz and Becker's (198A) review of the Health Belief Model studies is, therefore, helpful in 91 consideration of the present study. The post-MI patient must cope with a "post-MI status" that will continue for the rest of his life. Janz and Becker postulate that perceived severity is a much more important concept for such patients. In these studies reviewed by Janz and Becker (198A), increases in perceived severity were associated with increases in compliance, in accordance with the Health Belief Model. Some authors (Becker, et al., 1977; Leventhal, 197A; Mikhail, 198A) have suggested that extremely high levels of perceived severity may actually have the opposite effect, reducing compliance or the health behavior of concern. These suggestions are consistent with theoretical literature concerning the use of fear or threat communications. According to Leventhal (1973), the emotional response of fear normally functions as a drive that mediates belief and behavior change. The performance of health recommendations is, therefore, expected to reduce that fear. With extremely high levels of fear arousal or if ways to reduce the threat are not available or are seen as ineffective, the threat can theoretically be counterproductive. Avoidance behavior or denial may follow (Leventhal, 1973). Denial has been documented In victims of myocardial infarction (Billings , Lindell, Sederholm, 8 Theorell, I980; Gentry, Baider, Dude-Heme, Musch, Gary, 1983; Gentry 8 Williams, 1979; Shapiro 8 Levine, 1971), but research is needed concerning the use of fear arousal and its effects on compliance and to determine what intensity of fear communication is most effective in producing the desired change and 92 what level of fear becomes counterproductive. In addition, questions remain concerning how personal factors interact with individual perceptions of the threat (Mikhail, 1981). No studies could be found in the literature by this researcher that use the Health Belief Model with post-Ml patients. Hijeck (198A) presented a theoretical framework for cardiac nurses using the Health Belief Model as a guide to patient care and developed a questionnaire for predicting patient entrance to a cardiac rehabilitation program based on Health Belief Model dimensions. The predictive ability of the tool has not been tested. Research is needed linking the Health Belief Model dimensions to compliance with post-MI treatment regimens in order to validate the usefulness of the model in therapeutic interventions with this population. While the Health Belief Model has not been used in research with post-MI subjects, studies are available using the model to examine compliance with antihypertensive regimens. Andreoli (1981) used the Health Belief Model as a conceptual framework in investigating compliance with antihypertensive regimens. Compliance was measured by blood pressure readings and "nurse interpretations.” Questionnaires measured the Health Belief Model dimensions and self-concept. A two-tailed t-test was used to compare self-concept and health beliefs for compliant and non-compliant subjects. There were no significant differences between the two groups of subjects in self-concept or health beliefs for the 112 male subjects, including the dimension of perceived severity. 93 In studies investigating the relationship between perceived severity and compliance with antihypertensive regimens, results are conflicting. Two other studies with hypertensive subjects found positive relationships between perceived severity and compliance (Cummings, et al., 1982; Taylor, 1979). Another study with hypertensive subjects found no relationship between perceived severity and compliance (Moriski, 1982). In general, more studies have found positive relationships between perceived severity and compliance with hypertensive regimens. Greene, Weinberger, Jerin, and Mamlin (1982) studied levels of compliance in a group of 190 patients who were under treatment for chronic medical Illnesses. Data collected in interviews with subjects measured perceived severity in relation to worries about health, recent changes in health status and life-style changes. Subjects who saw their illness as more severe were less likely to follow the recommended treatment regimen. Variation in the significance of perceived severity in compliance studies may be due in part to difficulties in the measurement of perceived severity. Janz and Becker (198A) report that, for the most part, every investigator has developed a unique approach to operationallzing each variable of the Health Belief Model. This variability renders interpretation of results and comparison of findings across studies problematic (Janz 8 Becker, 198A; Mikhail, 1981). In addition, many researchers do not report how they measured the dimensions specifically. When the measurement of severity is 9A specified, it is generally in the format of a statement where patients respond to a Likert scale on the extent to which they agree or do not agree with the statement. Andreoli (1981) used the Health Belief Model dimensions in a study attempting to determine If there were differences in self-concept and health beliefs in compliant and non-compliant hypertensives. This study was reviewed previously. Andreoli developed the instrument used to measure Health Belief Model dimensions from previously used questionnaires, the investigators previous experience with hypertensive patients, literature review and from interviews with nurses and physicians involved in the care of hypertensive patients. A Likert scale was developed that consisted of 15 perceptual statements on which a person was asked to rate himself in three categories of health beliefs; susceptibility, severity and benefits. The reliability of the instrument was measured through a test-retest with seven hypertensive adults. For the total health beliefs, the coefficient of correlation was .71. Andreoli (1981) did not give further information on the types of statements used in the three categories. Becker, Radius, Rosenstock, Orachman, Schuberth, and Teets (1978) were more specific about the type of information used to measure the severity construct. Interviews were conducted to determine the effects of mother's health beliefs on compliance with their chlldrens' asthmatic treatment regimens. Questions were developed to measure severity along two dimensions: 1) the severity of the 95 condition itself; and 2) interference with the childs' or mothers' social functioning. Mothers (N - 111) were asked to respond to statements about the seriousness of the childs' condition, were asked to rank the seriousness of asthma with three other childhood diseases and were asked to respond to statements reflecting asthma as painful, uncomfortable, harmful and powerful. In addition, statements concerning interference with the childs' schooling, interference with the mothers' activities, and asthma as a cause of embarrassment were included. Perceptions of the overall severity of the childs' asthma in absolute terms was reported to surpass the predictive usefulness of all items contrasting asthma to other illnesses (p <:.01). In addition, interferences with schooling, mothers' activities and asthma as a source of embarrassment were all predictive of compliance to a lesser degree (p< .05). While Becker et al (1978) used a very different population (mothers of asthmatic children) than that used in the present research, their study is one of the few studies using the Health Belief Model components that includes individual items asked in the interview for the readers review and reports the specific dimensions of severity included in the study. No reliability testing was reported and it is not clear how the researcher rated the responses obtained in the interviews. Champion (198A) conducted research on Health Belief Model scale development. The dependent variable chosen for scale development was frequency of breast self-examination. The independent variables were 96 susceptibility, seriousness, benefits, barriers and health motivation. Scales were developed to measure the Health Belief Model dimensions based on concept measurements from past research, using Likert scales. Initially, 20 to 2A items were written for each independent variable and 10 to 12 were selected by judges for content validity. A questionnaire was then developed from the content valid items. Subjects (all female) were recruited by mailed and personally delivered questionnaires requesting participation. Of 6A0 initial requests, 190 subjects indicated a willingness to participate and 60 of these were chosen at random and mailed the questionnaires. There were 57 questionnaires returned (952). Cronbach alpha was used for reliability testing of the scales. Items demonstrating low correlations with their respective scales were deleted. The severity scale had a reliability coefficient of .78. The revised scales were then administered to 57 persons to calculate test-retest reliabilities. All test-retest reliabilities were found to be above .7 and significant (p <:.OOI), with the severity correlation reported as .76. The final form of the severity scale contained 12 items. Six of these Items were related to physical or emotional signs of fear at the thought of breast cancer (nausea, rapid heart beat, feelings of hopelessness, worry). In two items, career and financial effects were reflected (e.g., if I had breast cancer, my career would be endangered). In one item, effects on relationships were reflected (e.g., breast cancer would endanger 97 my marriage or significant relationship). In one item, breast cancer seriousness was compared to "other diseases." The two remaining items were "Problems I would experience from breast cancer would last a long time" and "if I had breast cancer, my whole life would change." The scales were submitted to factor analysis. Susceptibility, severity, benefits, barriers and health motivation were found to be mutually exclusive. The seriousness construct resulted in three factors relating to 1) physician symptoms of fear, 2) long-term effects of breast cancer, and 3) financial or career problems. In this study by Champion (198A), the results are suggestive that seriousness may not be a unidimensional construct. More scale development studies are needed using subjects who have had myocardial Infarction and other chronic diseases to determine severity dimensions in these populations. Hijecks' (198A) theoretical use of the Health Belief Model to predict patient entrance into a cardiac rehabilitation program was described earlier.. Hijeck suggested the use of three statements with seven-point Likert responses to measure the seriousness construct: 1) How serious do you belief heart attacks are?; 2) How would you compare heart attacks to other illnesses?; and 3) How will your present illness change your present life-style? A major drawback to these questions Is that the individual is not asked ”How serious was YOUR heart attack"? which is the focus of the concept. As stated previously, the predictive ability of Hijeck's (198A) tool has not been tested. 98 In summary, there are studies in which positive and negative relationships have been reported between perceived severity and compliance with chronic disease regimens. The studies that show positive relationships between perceived severity and compliance with chronic disease regimens support the Health Belief Model postulate that a certain level of perceived severity is necessary for action to occur and that increased perceptions of severity make action more likely. The studies that show negative relationships between perceived severity and chronic disease regimens support the postulate by Leventhal (1973) that higher levels of perceived severity increase fear that inhibits action through avoidance behavior and denial. Most studies of perceived severity use differing methods of measurement, making comparison difficult. Perceived severity has not been investigated for the post-MI patient population. Summary The post-MI treatment regimen frequently consists of daily medications, dietary reduction of cholesterol, saturated fats, and total fats, and initiation and maintenance of an aerobic exercise training program. Scientific studies suggest that compliance to these treatment regimens will reduce morbidity and mortality for the post-MI patient. In the compliance literature, few studies were found by this researcher that dealt with the post-MI treatment regimen. Compliance has long been known as a difficult behavior to measure and no perfect 99 method of compliance measurement exists. For the post-MI patient, compliance with medications is usually better than compliance with diet and exercise regimens. Compliance often decreases over time but those who are compliers are more likely to stay compliant than to become noncompliant. Compliance with medications, diet and exercise is generally not related to sociodemograhic variables or current health status. This researcher found conflicting findings in studies investigating the relationship between compliance and knowledge of the disease, social support and complexity of the regimen. No studies were found using the Health Belief Model to investigate compliance with the post-MI therapeutic regimen. The Health Belief Model dimensions of perceived susceptibility, severity, benefits and barriers have all been shown to be useful in predicting health behavior. Severity has been least predictive of compliance in total studies using the Health Belief Model but severity was significantly predictive of compliance in 882 of studies dealing with sick role behavior. While the Health Belief Model is useful in the prediction of compliance, it makes no contribution to intervention for improving compliance. In addition, different methods of measurement of compliance and of the Health Belief Model dimensions make comparison between the studies difficult. In this chapter, a review of the literature pertaining to the major study variables of perceived severity and compliance with the post-MI therapeutic regimen was presented. In summary, more research is needed to develop measurement standards for compliance and for the 100 Health Belief Model dimensions, to identify factors that influence compliance for the post-MI patient, and to identify components of perceived severity for the post-MI patient. In Chapter IV, the methodology used in this study is presented. CHAPTER IV Methodology Overview The data used in this study were collected between 1980 and 1983 by Kline and Warren for the purpose of describing the relationship between perceptual differences and level of function of a marital couple following myocardial infarction. The study included five questionnaires dealing with the measurement of the health regimen, adherence to the health regimen, perception of marital functioning and sociodemographic data. This researcher will use parts of the instruments to investigate compliance with the therapeutic regimen. A descriptive correlational analysis will be used in this study to explore the relationship between perceived severity of myocardial Infarction and compliance with the post-MI therapeutic regimen. In this chapter, the research methodology and design used in the study will be discussed. Operational definitions of the variables and the instruments used to measure them will be described. Sample selection, procedures for data collection, statistical analysis and protection of human subjects will also be included. The data collection procedures were prepared with the assistance of Rogers (1985). Research Question Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed therapeutic regimen? 102 Secondarnyesearch Questions 1. Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed medication regimen? 2. Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed dietary regimen? 3. Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed exercise regimen? Operationalization of the Study Variables Perceived Severity of MI Perceived severity of MI is defined as the extent to which the patient believes that myocardial infarctions generally are dangerous and that the patient's own myocardial infarction is serious (Hijeck, 198A). For the purposes of this study, only the patient's perceptions of severity regarding his own myocardial infarction will be addressed. The format used to measure the perception of severity will be an item with Likert-type responses to the question of ”How severe was your last heart attack"? where subjects may answer: 1) very severe, a lot of heart damage; 2) moderately severe, some heart damage but not a lot; 3) mild, only a little heart damage; and A) no heart damage. 103 Stated Compliance with the Therapeutic Regimen Stated compliance with the therapeutic regimen is defined as the extent to which the patients indicate they carry out the therapeutic recommendations of health care providers concerning prescribed medications, diet and exercise. Compliance will be measured by the use of a total of 15 questions (see Appendix D, questionnaire #2) which measure the patient's stated relative frequency of compliance with the use of medication (items 8-10), diet (items 3-7) and exercise (items 11-16, 18, 27). The format of the questions is a statement of action followed by a five-point Likert scale In which the subject Is able to indicate the extent to which he carried out the action. A numerical score will be assigned to each of the possible responses such that a low score will be indicative of a high degree of compliance. The score ranges for each of the compliance subscales are: diet, 5-35; medications, 3-15; and exercise, 8-6A. Mean scale scores will be used to correlate compliance with perceived severity. The compliance scale will be submitted to factor analysis. Results will be reported in Chapter V. Modifyieg_Factors The modifying factors used in this study include the demographic variables of age, ethnic background, educational level, occupation and annual income (categorical). These variables will be measured using items 1, 2, 6, 8, and 10 on questionnaire number five (see Appendix D). 10A In addition, modifying factors used In this study include the following structural variables: number of heart attacks (one or more than one), length of time in months since last heart attack, past participation in an in-hospital post-MI teaching program, present enrollment in special heart programs and the presence of chronic health problems. These variables will be measured using Items 1A, 15, 17, 21 and 22 on questionnaire number five (see Appendix 0). Descriptive statistics will be used to characterize the study sample. Frequency and percentage tables will be computed for the modifying variables. Using Pearson product moment correlations, the modifying variables appropriate for correlation will be evaluated for potential moderating effects on the study variables of perceived severity of MI and compliance with therapeutic regimen. See Table 1 for a summary of the modifying factors that will be correlated with the major study variables, the item number on questionnaire number five that measures each factor and the method of scoring used in the correlation. Sample The subjects for Kline and Warren's study were a convenience sample of 98 individuals living in Lansing and the surrounding area, Flint and Grand Rapids. A formal cardiac rehabilitation program at Lansing Community College and six physicians' offices provided names of potential subjects. Criteria for inclusion in the study were: 1) males between the ages of 35 and 69 years of age; 2) diagnosed and 105 Table 1 Modifying Factors, Item Number on Questionnaire #5 and Scoring:Used in Correlations with Perceived Severity Modifying Factor Item I Scoring__ Age in Years 1 low to high Education Level 6 1 (grades 1-6) to 7 (beyond A years college) Annual Income 10 1 (0-9,999) to 9 (over 80,000) Work Outside the Home 8 1 (yes) to 2 (no) Number of His 1A 1 (more than 1) to 2 (I only) Time in Months since MI 15 I to 12 Past participation in an in- patient teaching program 17 1 (yes) to 2 (no) Current participation in a teaching program 21 1 (yes) to 2 (no) Presence of additional chronic health problems 22 1 (yes) to 2 (no) 106 hospitalized with a myocardial infarction within the past twelve months; 3) married and living with spouse at the time of the study; and A) able to read, write and speak the English language. No attempt was made to randomize the subjects or to limit the sample based on severity of myocardial Infarction, history of cardiac bypass surgery or previous MI, or the presence of residual cardiac problems such as angina or congestive heart failure. From Kline and Warren's sample of 98, a sample of 70 individuals was selected who indicated that since the heart attack, they were: 1) following a special diet; 2) taking medications on a daily basis; 3) following a program of regular exercise at least three times a week. The 70 patients who answered yes to these three items were selected for Inclusion in this study. Data Collection Procedures The data utilized by this researcher were collected between 1980 and 1983 for a study conducted by Kline and Warren, two nursing faculty members at Michigan State University. The data collection procedures followed by Kline and Warren are described below. To secure an adequate subject pool, physicians and agencies with potential contacts with MI patients were identified by the Investigators. A letter introducing the research project was sent to those identified requesting their participation (see Appendix A). Six physician offices (twelve participating physicians) and a formal 107 cardiac rehabilitation program offered through the local community college responded to the request and agreed to provide names of potential subjects for the study. In each setting, the investigators met with the registered nurse designated by the physician as having primary responsibility for identifying patients appropriate for the study. The sample selection criteria were discussed with the nurse at this time and a written copy of the criteria left at each office for reference. Ongoing contact was maintained with each office by the Investigators at regular monthly intervals to ensure continuation of the established selection procedure and provide a channel for feedback with the participating sites. The nurse at each site selected male patients who met the study criteria. These men were then mailed a letter signed by the patient's physician inviting their participation in the study. A postcard was enclosed in each letter for the potential subjects to return to the investigators if they were interested in participating and/or receiving more information about the study. Potential subjects who agreed to be contacted were telephoned by project members and the study was explained. At this time, they were advised of the type of information asked in the questionnaires and the amount of time required to complete them. In addition, subjects were informed that participation in the program was voluntary, would in no way interfere with the care they received and that they were free to withdraw at any time without jeopardizing their future care. 108 Assurances of subject anonymity and the confidentiality of questionnaire responses were also made to all potential subjects contacted. These subjects who agreed to participate were then formally enrolled in the study and mailed consent forms, instruction letters and questionnaires. Subjects were instructed to answer each question as honestly as possible. Consent forms and completed questionnaires were returned to the investigators by mail in an envelope provided for this purpose. The data were coded on Fortran coding sheets, keypunched and logged into the computer to prepare for data analysis.. The data were made available to nursing graduate students for research purposes. Students were unable to identify study participants. Reliability and Validity Reliability is the degree of consistency or accuracy with which an instrument measures an attribute. In this study, reliability of the compliance subscales will be measured by the use of coefficient alpha to determine the internal consistency among the items in each subscale of medication, dietary and exercise compliance. Validity refers to the degree to which an instrument measures what it is supposed to be measuring. There are many different types of validity and various methods of analyzing validity of an instrument. Content validity is based on judgement regarding the sampling adequacy of the content area being measured (Polit and 109 Hungler, 1983). Content validity of the compliance instrument was concluded to be adequate for the purposes of this study by the researcher and thesis committee members. Statistical Analysis Descriptive and inferential statistical techniques will be used to describe the sample and test the research question. The modifying factors will be summarized with percentages and frequencies. Pearson product moment correlations will be utilized to describe the relationship between the major study variables of perceived severity of MI and reported compliance with the therapeutic regimen. Correlations will be performed on each of the compliance subscales and on the instrument as a whole. Contingency tables will be used to describe the relationship between perceived severity and pertinent modifying factors of a nominal nature and those modifying factors suitable for correlation will be analyzed for relationships with perceived severity using Pearson r. Items on the compliance. instrument will be scored into content subscales (medications, diet and exercise) if they are verified in the reliability analysis. Items that do not cluster will be individually correlated or deleted if necessary. Items marked "not applicable" will be treated as skipped items. Pairwise deletion will be used for missing data. A p <:.05 level of significance will be used to determine statistically significant correlations. 110 Human Subjects Protection Specific procedures were employed by Kline and Warren (1983) to assure the rights of the study participants. The rights of the participants were protected using the standards from the University Committee on Research Involving Human Subjects (UCRIHS). Refer to Appendix F for the letter of acceptance from UCRIHS. Subjects were informed of the purpose of the study, length of time to complete the Instruments and requirements for participation. Assurance of anonymity and confidentiality were provided. In addition, code numbers were used on the questionnaires for identifying data as it was collected. This researcher had access to the data only after it was logged into the computer and there were no means to identify subjects. Application was made to UCRIHS for exemption in the use of secondary data. See Appendix F for the approval letter frOm UCRIHS. Summary In this chapter the methodology utilized for this study was presented. Specific topics discussed were the study question, Operationalization of the study variables, sample, instruments, data collection procedures, human subjects protection, and statistical analysis procedures. In Chapter V the analysis of data and a discussion of the results relevant to the research question are presented. CHAPTER V Data Analysis Overview In this chapter the study sample is described and data are presented to delineate the relationships between perceived severity of myocardial infarction and compliance with medications, diet and exercise regimens. Reliability testing for the compliance instrument is discussed. Additional findings are discussed In regard to the modifying factors. Data from the 70 post-MI patients was used to answer the following research question: Is there a relationship between pgtients' perceptions of MI severity_and compliance with the therapeutic regimen? 1) Is there a relationship between patients' perceptions of MI severity and compliance with the medication regimen? 2) Is there a relationship between patients' perceptions of MI severity and compliance with the dietary regimen? 3) Is there a relationship between patients' perceptions of MI severity and compliance with the exercise regimen? Descriptive Findings of the Study Sample Data from questionnaire number five (see Appendix D), the sociodemographic questionnaire, were used to describe the characteristics of the sample. Findings in each of these areas are discussed in the next section of this chapter. 112 Characteristics of the Study Sample The sample consisted of 70 English-speaking males who had experienced a myocardial infarction within the past year and indicated that they were presently following a post-MI treatment regimen that involved medications, diet and exercise. With the exception of one subject who reported his background as Indian, subjects were white. Subjects ranged in age from A0 to 7A years, with a mean age of 56.5 years. Approximately 512 of the subjects were in the 55-6A age group. The age distributions and percentages for the subjects are presented in Table 2. Table 2 Age Range Distribution of Subjects (N-70) Age Range Number Percent AO-AA A 5.7 A5-A9 . 9 12.9 50-5A 13 18.6 55-59 21 30.0 60-6A 15 21.A 65-69 6 8.6 70-7A _2 __2;§ TOTAL 70 100.0 113 One of the subjects did not respond to the question regarding educational level. Levels of education ranged from "junior high school" to "beyond four years of college." had completed high school or had partial college preparation. Over half of the subjects Distribution according to level of education is summarized In Table 3. Table 3 Education Levels of Subjects (u-ég) Level of Education Number Percent Less than seven years school 0 -- Junior high school 2 2.9 Partial high school 8 11.6 High school 29 A2.0 Partial college 16 23.2 College graduate 6 8.7 Beyond four years college 8 11.6 TOTAL 70 100.0 11A There were 37 subjects (532) who reported they were currently working outside the home. Of these, 29 subjects (782) were working full-time and eight (222) were working part-time. The remaining 33 subjects (A72) indicated that they were not working. There were 67 subjects who responded to the question regarding current occupation. This is a larger number than those reporting that they were presently working outside the home. One possible reason for this discrepancy is that some of the subjects had not returned to work yet but planned to do so in the future. In addition, 1A subjects reported their occupation as retired. A summary of the distributions of occupations if found in Table A. Table A Frequency Distribution py Occupation (N-6Z) Occupation Number Percent Clerical 1 1.5 Professional 10 15.0 Executive 9 13.A Skilled worker 15 22.A Semi-skilled/unskilled worker A 6.0 Own business 5 7.A Retired 1A 20.9 Unemployed, but looking for work I 1.5 Other _8 _ll;9 TOTAL 67 100.0 115 Two of the subjects failed to respond to the question regarding their combined annual Income. Subjects' incomes ranged from under $9,999 to over $80,000 yearly. Almost half of the subjects reported incomes of $20,000 to $39,999 or more yearly. A summary of distributions and percentages for combined annual income is presented in Table 5. Table 5 Frequency Distribution of Combined Annual Income (N-68) Income Number Percent $ 0 - $ 9,999 A 5.9 $10,000 - $19,999 8 11.8 $20,000 - $29,999 19 28.0 $30,000 - $39,999 14' 20.6 $40,000 - 5A9.999 1A 20.6 $50,000 - $59,999 2 2.9 $60,000 - $69,999 2 2.9 $70,000 - $79,999 2 2.9 $80,000 and over _3 __£;fi TOTAL 68 100.0 116 To further describe the study sample, subjects were asked to respond to the following items: 1) number of months since the MI; 2) previous hospitalization for a MI (yes or no); 3) presence of other chronic health problems (yes or no); A) participation in teaching programs before hospital discharge (yes or no); and 5) current participation in an organized heart education program (yes or no). Subjects responses are presented below. Four of the subjects failed to respond to the item regarding length of time since the MI. For the remainder of the sample, answers ranged from less than one month to 12 months, which a mean of 5.5 months. The frequency distribution for number of months since the MI is presented in Table 6. A total of 23 subjects (332) reported they had been hospitalized more than one time for a heart attack. The remaining subjects (672) denied being previously hospitalized for a MI. A total of 33 subjects (A72) reported having additional chronic health problems, while 37 subjects (532) denied having such problems. A summary of distributions for specific chronic disease reported by the subjects is presented in Table 7. In regard to participation in a cardiac education program during hospitalization for the MI, 5A of the subjects (772) indicated that they had participated in such a program while 16 subjects (232) stated that they had not. Regarding current participation in an organized cardiac program, 21 subjects (302) reported such participation and A9 subjects (702) reported that they were not currently participating in this type of program. 117 Table 6 Distribution for Length of Time in Months Since MI (N'66) Time in Months Number Percent 1 3 11.5 2 3 4.5 3 12 18.2 1. 7 10.6 5 1A 21.2 6 6 9.1 7 2 3.0 8 8 12.2 9 A 6.1 10 A 6.1 11 2 3.0 12 _1 42 TOTAL 66 100.0 118 Table 7 Distribution of Other Reported Chronic Diseases (N-ZO) Disease Number Arthritis 8 Cancer 0 Hypertension 1A Lung disease 8 Diabetes Mellitus 6 Other ‘12 TOTAL 48 While perceived severity of MI is a major study variable, distributions are included in this section for the purpose of description of the study sample. Three of the subjects did not respond to the Item regarding perceived severity of their MI. Responses ranged from "very severe; I have a lot of heart damage" to "I have no heart damage.” The majority of subjects (652) indicated that their heart attack was very severe or moderately severe. A summary of distributions and percentages for severity of the MI is presented in Table 8. 119 Table 8 Distribution of Perceptions of Severity of MI (N-67) Severity Number Percent Very severe; a lot of heart damage 20 29.8 Moderately severe; some heart damage 26 38.8 Mild; only a little heart damage 17 25.A No heart damage _A __§;Q TOTAL 67 100.0 Factor Analysis of Compliance Subscales The compliance instrument used in this study was submitted to factor analysis. The factor analytic procedure is described below and the results of factor analysis of the compliance instrument are described. Factor analysis reduces a large set of variables into a smaller, more manageable set of measures and also assists the researcher in identifying theoretical constructs. Factor analysis disentangles complex Interrelationships among variables and Identifies which variables "go together" as unified concepts. The underlying dimensions thus identified are called factors (Polit 8 Hungler, 1983). The first phase of factor analysis, called factor extraction, involves condensing the variables into a smaller number of factors. The factors are usually derived from the intercorrelations among the 120 variables in the correlation matrix. The first factor to be extracted accounts for the maximum amount of variance. For the following factor, the highest possible amount of variance Is again extracted from what remains after the first factor has been taken into account. Factor extraction continues until there is no meaningful variation left. The factors thus delineated represent independent scores of variation in the data matrix. The second phase of factor analysis Is called factor rotation. In factor rotation, clusters of variables in a factor matrix are rotated in such a way that these clusters are distinctly associated with a factor, enhancing interpretability. The rotated factor matrix is used to interpret the factor analysis with factor loadings (or weights on a factor) interpreted in the same way as a correlation coefficient. Factor loadings express the correlations between individual variables and factors. The compliance instrument (see Appendix D, Questionnaire 2, items 3-7, 8-10, 11-16, 18, 27) was submitted to factor analysis. Using the items formulated for each of the three therapeutic regimen categories of medications, diet and exercise, and Interitem correlation matrix was computed. Oblique multiple group factor analysis was then applied to the matrix. Various factor solutions were attempted, using from one to four scales. Items were examined to ensure that the contents of item clusters were indicators of the underlying construct. 121 Items in the medication subscale were not found to be unidimensional and were, therefore, used individually rather than as a subscale in correlations with other variables. One item (#5) was deleted from the diet subscale. The deleted Item questioned whether subjects read food labels before eating foods and may have been an Inappropriate question for middle-aged married males if wives carried out food purchases and preparation. The exercise items "fell into" two subscales. Content analysis revealed two of the items specified exercise while the remaining five items could more accurately be labeled "activity.” The two items in the exercise subscale were unidimensional with the diet subscale but were kept separate due to content differences. Based on content and cluster analyses, the following compliance subscales were identified for use in this study: 1. Medication compliance: no subscale, questionnaire Items #8, #9, #10 used separately. 2. Diet compliance: questionnaire items #3, #A, #6, #7. 3. Exercise compliance: questionnaire items #11, #12. A. Activity compliance: questionnaire items #13, #IA, #15, #16, #27. Reliability of the Compliance Instrument The reliability of the compliance instrument was measured by computing coefficient alpha for each of the subscales. Coefficient alpha is an indication of homogeneity or internal consistency and 122 estimates the extent to which different subparts of an instrument are equivalent in terms of measuring the critical attributes (Polit 8 Hungler, 1983). The original subscale of three items used to measure medication compliance had an alpha coefficient of .A6. It was decided that these items would be used separately. The diet subscale had an alpha coefficient of .80 with item #5 deleted. The exercise subscale had an alpha coefficient of .88. The activity subscale had an alpha coefficient of .71. See Table 9 for a summary of alpha coefficients for the compliance subscales used in this study and the questionnaire items included in each subscale. Table 9 Compliance Subscales, Questionnaire Items Included in Subscales and Alpha Coefficients for Each Subscale Compliance Subscale Items Alpha Medication (items to be used separately) 8, 9, 10 .A6 Diet 3, A, 6, 7 .80 Exercise 11, 12 .88 Activity 13, 1A, 15, 16, 27 .71 123 Intercorrelational Matrix for Subscales All of the subscales and medication items had a moderate or strong relationship to the entire compliance instrument, as expected. Medication items tended to have slight relationships or no relationships to other subscales. Diet and exercise subscales had a strong relationship (r - .6A6, pf; .001). This strong correlation is consistent with the findings of factor analysis, which indicated that the diet and exercise subscales were unidimensional. The strong correlation between diet and exercise compliance indicated a strong tendency for diet compliers to be exercise compliers and vice versa. See Table 10 for the intercorrelation Matrix for the compliance subscales. Presentation of the Data Related to the Research Question In this section the major research question is addressed. Pearson product moment correlations were utilized to calculate relationships among the study variables of compliance with the therapeutic regimen and perceived severity. A summary of the findings related to the research question is presented in Table 11. Research Question: Is there a relationship between post-Ml patients' perceptions of MI severity and compliance with the prescribed therapeutic regimen? Table 10 12A Intercorrelation Matrix for Compliance Subscales Medication 8 9 10 Diet Exercise Activity Meds #8 Meds #9 .336** Meds #10 .2A7* .076 Diet .272* .273* .2A9* Exercise .178 .280* .1A5 .6A6*** Activity .082 -.O71 .166 .115 .OA3 TOTAL .381** .A67*** .A67*** .80A*** .607*** .6A2*** * IDS-05 ** pf; .01 Wt pg .001 A correlation coefficient of -.001 was computed between subjects' scores of perceptions of MI severity and total stated compliance with the therapeutic regimen. There was no significant relationship between perceptions of MI severity of subjects and their total stated compliance with the therapeutic regimen. Subguestion #1: Is there a relationship between post-MI patients' perceptions of MI severity and compliance with the prescribed medication regimen? 125 Table 11 Correlations of Perceived Severity of Myocardial Infarction and Compliance with the Therapeutic Regimen Compliance Item Correlation (r) Significance (p) Medication #8 -.07A .277 Medication #9 (N-69) -.193 .060 Medication #10 (N-66) -.202 .056 Diet .079 .271 Exercise .150 .118 Activity .113 .195 A correlation coefficient of -.07A was computed between subjects' scores of perceptions of MI severity and compliance with medication item #8. There was no significant relationship between perceptions of MI severity of subjects and their stated compliance with medication item #8. A correlation coefficient of -.193 was computed between subjects' scores of perceptions of MI severity and compliance with medication item #9. There was no significant relationship between perceptions of MI severity of subjects and their stated compliance with medication item #9. A correlation coefficient of -.202 (p - .056) was computed between subjects' scores of perceptions of MI severity and compliance with medication item #10. The level of significance was 126 unacceptable, therefore, there was no significant relationship between perceptions of MI severity of subjects and their stated compliance with medication item #10. Subguestion #2: Is there a relationship between post-MI patients' perceptions of MI severity and compliance with the prescribed dietary regimen? A correlation coefficient of .079 was computed between subjects' scores of perceptions of MI severity and compliance with the dietary regimen. There was no significant relationship between perceptions of MI severity of subjects and their stated compliance with the prescribed dietary regimen. Subquestiogl3: Is there a relationship between post-MI patients' perceptions of MI severity and compliance with the prescribed exercise regimen? A correlation coefficient of .150 was computed between subjects' scores of perceptions of MI severity and their stated compliance with the prescribed dietary regimen. There was no significant relationship between perceptions of MI severity of subjects and their stated compliance with the prescribed exercise regimen. While this study did not include a research question based on compliance with an "activity” regimen, a question was formulated based on factor analysis that indicated "exercise" and "activity" were separate constructs. Activity Subguestion: Is there a relationship between post-MI patients' perceptions of MI severity and compliance with the activity regimen? 127 A correlation coefficient of .1132 was computed between subjects' perceptions of MI severity and compliance with the activity regimen. There was no significant relationship between perceptions of MI severity of subjects and their stated compliance with the activity regimen. It must be noted that sample selection criteria included only those subjects who had received medication, diet and exercise prescriptions from their health care providers and did not include those reporting "activity" prescriptions. The activity subquestion may, therefore, not be relevant for this study population. Presentation of the Data Related to Modifyingefactors Pearson product moment correlations were utilized to calculate the relationships between post-MI patients' perceptions of MI severity and the modifying factors. Tables 12 and 13 summarize these relationships. Table 12 Correlations of Penceived Severity of MI and Demeggaphic Variables UsingyPearson r (N-70) Variable Pearson r Correlation Age .039 Education (N-69) .18A Work outside the home (yes/no) -.177 Annual income (N-68) .172 128 Table I3 Correlations of Perceived Severity of MI and Structural Variables UsingAPearson r (N-70) Variable Pearson r Correlation Number of MI's (one or more than one) -.117 Time in months since MI (N-66) -.05A In-hospital teaching (yes/no) .OAO Current teaching program (yes/no) -.056 Additional chronic diseases -.158 ** pf; .01 Demographic Variables None of the demographic variables of age, education, work outside the home (yes or no) and annual income was significantly correlated with subjects' perceived severity of MI. Ethnicity and occupation are nominal data used only to describe the population. See Table 12 for a summary of correlation coefficients for perceived severity of MI and the demographic variables. Structural Variables The structural variables used in this study are number of MI's (one or more than one), length of time on months since the last MI, past participation in an in-hospital teaching program (yes or no), current participation in an organized cardiac teaching program (yes 129 or no), and the presence of additional chronic diseases (yes or no). None of these structural variables were significantly related to perceived severity of MI. See Table 13 for a summary of correlation coefficients for perceived severity of MI and the structural variables. Other Findingp Using Pearson product moment correlations, modifying factors were evaluated for relationships with compliance. Significant findings are discussed in the following section of this chapter. See Table 1A for a summary of correlation coefficients for modifying factors and compliance subscales (medication items used separately). Age was significantly related to the diet subscale (r - -.208, pf .05), the activity subscale (r - -.2A7, pf .05) and the total compliance score. (r =- -.262, pf .05), indicating that younger subjects were slightly more likely to comply with diet and activity prescriptions and with treatment recommendations in general. 3 Working outside the home was significantly related to the activity subscale (r - -.218, p‘<:.05) This findings indicates that those who did work outside the home complied slightly more with the activity regimen. Time in months since the MI occurred was significantly related to compliance with medication item #8 (r - .217, p<(.05), the exercise subscale (r - .280, pf .05) and the total compliance score (r - .2A9, pf .05). These findings indicate that as length of time 130 Table 1A ty, and 1vl The Relationship Between Medication, Diet1 Exercise Act Total Compliance and Modifying Factors Using Pearson Product Moment Correlations _oo.uV.Q««« _o.HUVQ«« mo. V A? m:o.: wmp.1 :mc., on..- «om~.1 mm..- «m:~. mac. —m—. ch.1 :No.1 FRO. om..- «m_~.- m~_.- w:_. «~w~.1 «m:~.1 «kmom.n mmo.1 Nuo.1 sop.1 m_o.1 «mmN. .mo. mm_.1 mmo.1 @mo. :mo.1 «o:~.1 ~m_.1 m__.1 m:_.1 «omw. ~m_. wmo. mwo.1 «N_~. mw_.1 Nmo.1 mo_. coo. omo. moo.1 ooo. m:_.1 w__.1 om_.1 mmo. m~_. Nqo. qo_. mmo.1 .mo.1 .oo.1 mm_.1 mmo.1 mwo.1 :mo.1 «mo~.1 ~ao.1 mmo.1 om..- ommom_o o_coczo _eco_u_ou< AOC\mo>V EmLmoLa mc_;omou ucoccau AOC\mu>V mc_comou .mu_amo; c. Amzucoe c_v .2 ooc_m oE_k Aocoe Lo ocov m_z mo Lan:z . oEOuc_ .m:cc< AOC\mo>V use; one oo_muao xcoB co_umo:cm om< a_au< am_umeu Sa_o o_ m m mco_umo_uoz cocoon mc_xe_coz 131 since the MI occurred increased from zero to 12 months, compliance with medication and exercise regimens, and with the therapeutic regimen in general, increased slightly. Particlpgtion in an in-hospital teaching program was significantly related to the diet subscale (r a -.2A6, pf; .05) and the total compliance score (r . -.270, pf; .05). These findings indicate that those who participated in an in-hospital teaching program were slightly more likely to comply with dietary regimens and with the therapeutic regimen In general. Current participation in a cardiac teachlgg progppm was significantly related to the exercise subscale (r - .258, pf§..05), indicating that those who were enrolled in such a program complied slightly less with the exercise regimen. The presence of additional chronic diseases was significantly related to the exercise subscale (r - -.305, pf; .01). This finding indicates that those who had additional chronic diseases tended to comply more with the exercise regimen. Summary In conclusion, the purpose of this study was to determine whether or not a relationship exists between perceived severity of MI and compliance with the therapeutic regimen. In Chapter V, data were presented that described the Study sample in relation to demographic and structural characteristics. Factor analysis of the compliance instrument was summarized and reliabilities of the scales were 132 reported using coefficient alpha. Using Pearson product moment correlations, perceived severity of MI was analyzed for relationships with medication, diet and exercise compliance. No significant relationships were found between perceived severity and compliance scores. No other modifying factors were significantly related to perceived severity of MI. Slight relationships (r <:.35) were found between various compliance subscales and age, work status, length of time since the MI occurred, past participation in an in-hospital teaching program, current participation in a cardiac teaching program and the presence of additional chronic diseases. In Chapter VI, the research and data described in Chapter V will be interpreted and summarized. In addition, the implications of the findings for nurse practice, nursing education and nursing research are delineated. CHAPTER VI Summary and Conclusions Overview In Chapter VI, a summary and interpretation of the research findings will be presented. First, findings regarding the study sample characteristics will be presented. Conclusions drawn from study findings in relation to the major study question and three subquestions will be discussed. Finally, implications of the findings relevant to nursing practice, education and research will be delineated. Summary and Interpretation of the Finding; Characteristics of the Study Sample The sample characteristics are compared with other research findings in the following section of this chapter. Due to the lack of available research regarding compliance with post-MI regimens, compliance studies using other types of populations were reviewed. Some of the sample charactersitics may not lend themselves to comparison with some of these study populations. The mean age of the study sample was 56.5 years. This is consistent with the Croog and Levine (1982) study of post-MI patients in which 5A2 of the study sample were reported to be between 50 and 59 years, and with the studies by Miller (1982) and Miller et al (1985) in which the mean ages were 58 and 56 years, respectively. In contrast, Shephard, Corey and Kavanagh (1981) studied a sample of younger post-Ml victims, with an average of A8.l years. 13A This study included subjects ranging from A0 to 7A years. This is in contrast to Croog and Levine's (1982) study, in which subjects were limited to those who were 60 years of age or less. Miller (1982) reported the mean age of subjects to be 58 years and did not report the age ranges, which were not limited. Miller et al (1985) reported an age range of 32 to 70 with a mean age of 56. Limiting the age of subjects in the present study to 60 years and under may have secured a more homogeneous sample by reducing the presence of additional chronic diseases and myocardial infarctions. The subjects in this study were limited to males. Croog and Levine (1982) and Shephard, Corey and Kavanagh (1981) also limited their samples to male subjects. Both the Miller (1982) and the Miller et al (1985) studies and the Mulder (1981) study contained samples that were three-quarters male. Human et al (1982) included only male hypercholesterolemics in their study sample. Educational level of the subjects in this study ranged from junior high school to beyond four years of college, with 88.62 having completed high school and A2.92 of the sample having reported college attendance. Distributions for levels of education in this study were higher than those reported in most other studies reviewed. Croog and Levine (1982) reported that A3.82 of the subjects had three or less years of high school and 2A.12 had one year or more of college. attendance. Miller (1928) and Miller et al (1985) reported average educational level for subjects to be 11 to 12 years. Glanz et al (1981) reported that 602 of their sample of hypertensives had graduated from high school. 135 With the exception of one subject who reported his background as Indian, subjects in this study were white. This is consistent with Croog and Levine's (1982) sample, which was all white, and with the Miller (1982) study sample, which was 98.52 white. Miller et al (1985) did not report ethnicity of their subjects. Greene et al (1982) had a study sample comprised mostly of black subjects. In the present study, 532 of the subjects reported that they were currently working outside the home. Croog and Levine (1982) reported that 27.52 of their sample returned to work less than two months after the Infarct but 83.62 were working at one year post-MI. In other studies reviewed, researcher did not report employment status with the exception of the Haefner and Kirscht (1970) study, which was 862 female and all subjects were employed. Employment status is difficult to compare between studies because the subjects used in this study were all at different points in time up to one year post-Infarction. In addition, pre-infarction work status is not taken into account and some of the subjects may have been retired or unemployed at that time. In this study, the most frequently reported occupational categories were as follows: 2A2 skilled workers, 202 retired, 1A.32 professional, 12.92 executive and 7.22 own business. While it is difficult to compare occupations across the studies reviewed due to differing categories of occupation and lack of complete reporting, the present study appears to be fairly comparable to the Miller (1982), Miller et al (1935), and Croog and Levine (1982) studies with 136 two exceptions: the present study had more retired persons and approximately 102 less semi-skilled or unskilled workers. These two differences are possibly reflection of differences in age and education level. In the study of hypertensives by Glanz et al (1981) most subjects were reported to be blue collar workers. Combined annual income In this study ranged from no income to over $80,000/year. Approximately A72 of the subjects reported incomes between $20,000 and $A9,000. Only 122 of the sample reported $0 to $19,000/year. Glanz et al (1981) reported a median income of $12,000 per year, considerably lower than yearly incomes in this study sample. Other studies did not report income with the exception of Croog and Levine (1982), who reported salaries for the period from 1965 to 1967, making comparisons invalid. Since the average yearly income for white families in the U.S. was reported to be $21,117 in 1982, (U.S. Bureau of the Census, 198A), over half the present study sample is above the national average. This is congruent with the higher level of education reported in the sample. Subjects in the study had a mean of 5.5 months since the MI at the time of participation in the study, with a range up to 12 months post-MI. Most studies did not include subjects at all points across the trajectory of rehabilitation. In other studies data was collected at specific time intervals after the MI. Croog and Levine (1982) studied subjects at three weeks, one month, one year and eight years post-MI. Mayou, Foster and Williamson (1978) studied subjects at hospital admission time and two months and one year post-diagnosis. 137 Miller (1982) studied subjects while hospitalized and at two and twelve months post-discharge. Miller et al (1985) studied patients between six and nine months post-MI. Shephard, Corey and Kavanagh (1981) studied exercise compliance in subjects between two and 109 months after the heart attack, a considerably longer than any other study reviewed. Comparability between studies could have been enhanced had the present study limited data collection to specific time periods post-MI, as the majority of studies reviewed had done. In the present study, 23 subjects (332) reported that they had been hospitalized more than one time for a heart attack. In constrast, other studies limited subjects to those with a first MI (Croog 8 Levine, 1982; Miller, 1982, Miller, et al., 1985). Shephard, Corey and Kavanagh (1981) and Mayou, Foster and Williamson (1979) did not document this information and other studies reviewed involved mixed or non-MI populations. The perceptions of severity of M1 for the 23 subjects who had more than one MI may differ from those who had only one. The majority of subjects in this study reported their MI as moderately severe (26 subjects, 37.12) or very severe (20 subjects, 28.62). Another 2A.32 (17 subjects) reported a mild heart attack and there were four subjects (5.72) who reported no heart damage. No other studies could be found in which perceived severity of MI was measured. 138 Over half of the subjects (37 subjects, 532) denied having additional chronic health problems. The remaining 33 subjects (A72) reported having at least one chronic health problem. The most frequently reported cases were hypertension (1A cases), arthritis (eight cases) and lung disease (eight cases). None of the studies reviewed reported additional chronic diseases for M1 subjects. Croog and Levine (1982) included only people without a history of other major illnesses. Miller et al (1985) excluded those with cerebral, renal or pulmonary complications. Mulder (1981) excluded those who had any of a long list of diseases, including hypertension, diabetes, kidney or liver problems and marked obesity. Subjects were classified as “healthy” or "cardiac." The Greene et al (1982) study reported the inclusion of a variety of patients with chronic disease but, unfortunately, did not report them. The Hyman et al (1982) study included hypercholesterolemic males but sampling requirements excluded those with hypertension or cardiovascular disease. As with additional previous Mls, the presence of other chronic diseases may influence perceptions of the subjects and is, therefore, a confounding variable. In the present sample, 772 reported that they had participated in an in-patient heart teaching program and 232 reported that they were currently enrolled in an organized heart program. This is difficult to compare with other studies. Miller (1982) obtained an original study sample in a cardiac rehabilitation program and used a smaller sample of hospitalized patients for instrument testing purposes but 139 in-patient teaching is unclear. In the Miller et al (1985) study, the sample was required to have been hospitalized in an institution that provided a cardiac rehabilitation program that met certain specifications, including instruction on dietary restrictions, medications, and activity progression. Miller et al (1985) do not make reference to out-patient program participation. Shephard, Corey and Kavanagh (1981) used patients recruited to an organized post-MI exercise program but were unclear about in-patient programs. Other studies are difficult to compare because they involved non-MI or mixed populations. Most studies were unclear about previous educational programs. Croog and Levine (1982) reported only the ”percent of patients reporting receipt of advice from their physician." Mayou, Foster and Williamson (1978) reported similar information. In summary, this study sample was basically composed of white males with a mean age of 56.5 years, which is congruent with the literature regarding post-MI subjects. Beyond this, many dissimilarities were found and are summarized as follows: 1) One third of this sample was over age 60. One major study (Croog 8 Levine, 1982) included only those under age 60. Another study (Miller, et al., 1985) included subjects up to age 70 years but did not report age ranges. Congruent with age findings, a large portion (202) of this sample was retired. Other studies did not report percent retired. 2) Subjects In this study were more educated and had higher annual incomes than the national average. 1A0 3) In this study, subjects were included across the trajectory in time up to 12 months post-MI, with a mean of 5.3 months since the MI. All other studies except one (Shephard, Corey 8 Kavanagh, 1981) chose specific time intervals post-MI. A) One-third of the subjects had more than one MI. Most studies have limited subjects to those who had a first MI. 5) One-half of the subjects had additional chronic diseases. Most studies excluded those with additional chronic diseases. 6) Perceived severity of MI, activity level, participation in in-patient teaching programs and current participation in an organized heart program are all variables that are difficult to compare to other literature due to sampling techniques or lack of reporting of the variable. Because a convenience sample was used in this study, some of these variations mentioned above may reflect a sampling bias. For example, those with higher education and income or those in a rehabilitation program may have been more likely to participate in the study. In conclusion, generalizations from this sample to the larger population must be made with caution. Research Questions The purpose of this study was to determine the relationship between the post-MI patients' perception of the severity of his myocardial infarction and his compliance with the prescribed therapeutic regimen. A major research question and three subquestions were addressed in this study. For the purposes of data 1A1 interpretation, the following r values will be designated as weak, moderate or strong: 1) 0.0 -.20, no relationship; 2) .20 -.35, very slight relationship; 3) .35 -.85, moderate to fair relationship; and A) .85 - 1.00, high to perfect positive relationship (Borg 8 Gail, 19799 513-51“). Research Question: Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed therapeutic regimen? No relationship was found between perceived illness severity scores and total stated compliance scores. No studies could be found that investigated this problem for the post-MI patient but these results support the findings of Andreoli (1981) and Greene et al (1982) who reported that perceived severity of hypertension had no significant relationship to compliance. Other similar studies with hypertensive subjects found statistically significant relationships between perceived severity and compliance (Cummings, et al., 1982; Taylor, 1979). Lack of agreement in the literature is probably due in part to differences in the measurement of perceived severity as well as lack of reporting by researchers as to how they measured severity. The lack of significance between perceived severity and compliance in the present study may be due to the unidimensionality of the severity measurement. While Becker et al (1978) found that 1A2 the perception of overall severity in absolute terms (as used in this study) was the most predictive of compliance in their study (Champion, 198A) presented evidence suggestive of severity as a construct containing more than one dimension. Subguestion #1: Is there a relationship between the post-MI patients' perception of the severity of his MI and his compliance with the prescribed medication regimen? No relationship was found between perceived MI severity scores and compliance with medication items #8, #9, and #10. Medication items #9 and #10, however, approached significance (p - .06 and p - .056, respectively). An interesting finding is the negative direction of the relationship. Increasing perceptions of severity are associated with decreasing medication compliance, a finding in congruence with Greene et al (1982) but generally in conflict with most Health Belief Model literature. Theoretical literature regarding fear or threat communications suggests that a very high level of perceived severity may actually lower compliance levels (Becker, et al., 1977; Leventhal, 1973; Mikhail, 1981). Since the majority of the subjects reported MI severity as moderately severe or very severe, this is one possible explanation for the negative relationship. Another is the unidimensionality of the severity scale, as previously stated. 1A3 Responses to the medication items reveal that subjects generally reported a high degree of compliance to medications (see Appendix G), congruent with post-MI literature. Since subjects tend to overreport compliance, more noncompliers may have been identified by using statements such as "I forget to take my medication" as well as statements more specifically related to the actual behavior. In addition, more information regarding what medications are taken, side effects, and time schedules may have allowed more specific questions to be answered in regard to compliance with medications. Subguestion #2: Is there a relationship between the post-MI patients' perceptions of MI severity and his compliance with the prescribed dietary regimen? No relationship was found between perceived MI severity and dietary compliance scores. No other studies could be found that investigated this relationship between perceived severity of myocardial infarction and dietary compliance. Lack of significant findings may be attributed to the unidimensionality of the severity scale or to the lack of specificity of the dietary compliance items on the questionnaire. In analysis of mean responses to the dietary compliance items (see Appendix G), "more than half the time" is the average response. This is consistent with Croog and Levine (1982), who report 63.A2 and A8.82 dietary compliance rates at week seven and year one post-MI, respectively. It is possible that subjects perceived ”half the time," for example, differently. At any rate, accuracy in measurement is difficult to achieve in reference to a complex behavior such as eating. 1AA Subguestion #3: Is there a relationship between the post-MI patients' perceptions of MI severity and his compliance with the prescribed exercise regimen? No relationship was found between perceived MI severity scores and exercise compliance scores. No other studies could be found that Investigated the relationship between perceived MI severity and compliance with the post-MI exercise regimen. The mean exercise compliance scores (see Appendix G) approach medication compliance levels, indicating subjects reported that they comply most of the time. This is slightly more than Croog and Levines (1982) sample, in which three-quarters of the subjects compiled at week seven and one-half compiled at year one. Again, insignificant findings could be due to unidimensionality of the severity scale or compliance measurement difficulties. In addition, the use of only two items to measure exercise compliance may have affected the study results as more items may have more accurately measured exercise compliance. Activity Subguestion: Is there a relationship between the post-MI patients' perceptions of MI severity and his compliance with the prescribed activity regimen? No relationship was found between perceived MI severity scores and activity regimen compliance scores. These findings may be related to the unidimensionality of the severity measure but are certainly related to compliance measurement difficulties. In factor analysis of the compliance instrument, ”exercise“ and "activity" were 11.5 found to be two different constructs, indicating that subjects viewed their exercise regimen as separate from activity items. While it was known that all subjects reported having daily medication and dietary regimens and at least three times weekly exercise regimens prescribed by a physician, no such information was obtained regarding the ”activity" regimen. Without the knowledge of exactly what advice was given in these areas, activity compliance scores have limited usefulness. In addition, the researcher would need to know how often the activity statements are applicable to the patients daily routine. For example, the subjects who report "never" to the statement “I rest whenever I feel tired" may "feel tired“ many times each day or once a month. More information is needed before actual "activity" compliance can be measured with an acceptable degree of accuracy. Other Findingp In accordance with the Health Belief Model, modifying factors influence perceptions of severity. The modifying factors included in this study were investigated for relationships with perceived severity. Modifying factors were not found to be significantly related to perceived severity. Some of the modifying factors, however, were significantly related to compliance, which confirms that other factors influence compliance as well. A summary and interpretations of these findings are presented in the next section of this chapter. 1A6 Working outside the home was significantly related to the activity subscale, indicating that those who did work outside the home complied slightly more with the activity regimen. Perhaps those who worked outside the home had more occasion to be aware of compliance with activity items because the items were more apt to interfere with the daily work routine (e.g., resting when tired, avoiding stair climbing). As previously stated, the validity of the activity scale as a measure of compliance with an activity regimen is questionable. Time in months since the MI occurred was significantly related to medication compliance item #8, the exercise subscale, and the total compliance score, indicating that compliance in these areas increased slightly as length of time since the MI occurred increased. Croog and Levine (1982) found that compliance with post-MI therapeutic regimens generally decreased over time exception medication compliance, which remained stable over time. The average time in months since the MI occurred was 5.5 months. Perhaps subjects with more recent infarctions had not had a chance to incorporate all the behavioral changes required for compliance in general. In addition, other factors such as outpatient interventions with health care providers, participation in outpatient cardiac programs, and involvement with other supportive groups may have increased compliance in these areas over time. Finally, patients more familiar with a regimen may be more likely to overreport compliance with it. 1A7 In regard to medication item #8, the medication dosages would more commonly be adjusted in the earlier post-MI period to alter side effects or therapeutic outcomes. In particular, digitalis and antihypertensive medication dosages are commonly adjusted during this period. Compliance may, therefore, be lower in the earlier period due to more frequent changes in the medication dosages. Once dosages are established, compliance with the right amount of medication may then increase. Exercise regimen compliance may increase with time for the same reasons. The typical post-MI exercise regimen requires changes in pulse rates and time exercised until a certain prescribed time and pulse rate is reached. When the prescribed final level of exercise is achieved, subjects may find compliance easier. Time in months since the MI occurred is a confounding variable in this study that makes interpretation of the findings more difficult. Using subjects at the same point in time since the MI occurred (for example, at six weeks, six months and one year post-MI) would have provided more useful information regarding compliance over time. Participation in an in-hospital teaching program was significantly related to the diet subscale and the total compliance score, indicating that those who participated in an in-hospital teaching program were slightly more likely to comply with dietary regimens and with the therapeutic regimen in general. Literature that would most closely parallel these findings concerns the effects of education and/or knowledge on compliance. Study results are 1A8 conflicting but generally additional knowledge or education about a regimen or disease has not been found to alter compliance behavior (Peck 8 King, 1982; Ruffalo, Garabedian-Ruffalo, 8 Pawlson, 1985; Yoos, 1981). Unfortunately, content of the teaching programs is not known except with references to general categories such as "diet" and "exercise." It is not known what type of teaching or other interventions were used and a combination was probably used in most cases. Since patient education includes much more than the simple imparting of information, much more information regarding content, methods and other interventions is needed to determine the significance of these relationships. In addition, patients who are more likely to participate in education programs may have larger or stronger support systems or be more motivated to comply. Why the dietary compliance subscale was significantly correlated with participation in an in-hospital teaching prOgram and other subscales were not is a difficult question to try to answer. Medication compliance may not be related because patients usually receive one-to-one medication teaching from the physician and nurse rather than in a special teaching program. Patients may differentiate between this informal instruction and formal teaching programs offered in many hospitals. It would seem that exercise regimen compliance should be related to participation in an in-hospital teaching program because considerable time is usually spent on this topic in such classes. This is not the case. Perhaps unknown factors related to the actual exercise behaviors exert more influence. 1A9 The findings in relationship to participation in an in-hospital teaching program are difficult to interpret for two reasons. First, while general categories of discussion are reported (e.g., medication, diet, exercise), specific content of the program is unknown. In addition, the knowledge base of the subjects regarding the regimen is unknown. Current participation in a cardiac teaching program was significantly related to the exercise subscale, indicting that those who were enrolled in such a program complied less with the exercise regimen. This findings is in conflict with a study by Hilbert (1985), who reported a positive relationship between compliance and enrollment in a cardiac rehabilitation program. It is difficult to postulate why this relationship is negative in the present study. In a study by Glanz et al (1981), dietary compliance was found to be reduced by "nurse phone call" intervention. Perhaps subjects in this study acted similarly by reducing exercise compliance in response to cardiac teaching program interventions. The presence of additional chronic diseases was significantly related to the exercise subscale, indicating that those who had additional chronic diseases tended to comply more with the exercise regimen. One possible reason for this finding is that an exercise regimen may have been prescribed for the treatment of the additional chronic disease/diseases as well as the MI. Diabetic, hypertensive and COPD treatment regimens commonly include a prescribed exercise 150 regimen, for example. If a subject had been compliant with (or at least introduced to) a prescribed exercise program prior to the MI, this familiarity might enhance compliance post-MI. Missing data was not a problem in this study. No questionnaire item had less than 65 respondents and most items had 67 to 70 respondents. Summary of the Findings and Limitations of the Study In summary, perceived severity of MI was not found to be related to compliance with the post-MI therapeutic regimen in this study, although compliance with medications had a negative relationship with perceived severity that approached significance. Slight relationships were found between various compliance measures and the following modifying factors: working outside the home, time in months since the MI occurred, participation in an in-hospital teaching program, current participation in a cardiac teaching program, and the presence of additional chronic diseases. A number of limitations are evident from the summary and interpretations of this study. They are summarized as follows: 1. Perceived severity measurement is limited to one item on the questionnaire. Scale development studies are needed to identify other possible dimensions,such as functional status, role changes and financial career changes, for example. 151 Compliance measurement is limited to medication, diet, exercise and activity prescriptions. In reality, a more accurate compliance measurement might include other areas of the therapeutic regimen. The type of questions used to measure compliance were general in scope. More specific questions regarding actual behaviors may increase measurement accuracy. In addition, actual validation of the regimens and specific prescriptions given by the health care provider are not known. Subjects knowledge of the regimens is not known. This makes compliance measurement more difficult as well as the relationship of compliance to knowledge. The convenience sample used in this study may not be generalizable to other populations due to various population characteristics reviewed earlier. In particular, differences in age, race, education, and income may influence generalizability. One-third of the subjects had had more than one MI and almost half of the subjects had additional chronic diseases. These are confounding variables which may affect study results. A more homogeneous sample would include only those with a first-time MI who had no additional chronic diseases. Separation of the study sample into two groups (first-time MI subjects and those with an additional MI or chronic disease) is another alternative. Subjects were included in this study at all points in time from zero through 12 months post-MI. Perceptions of severity and levels of compliance may differ across this trajectory. 152 In the next section of this chapter, implications of the study are discussed. Specifically, implications for nursing research, nursing practice and nursing education are presented. Nursingflmplications Nursing implications are derived from the findings of this study in three major areas: practice, education and research. Because the data are more significant from a research perspective, implications for nursing research are presented first. Implications for Nursing Research Implications for nursing research are addressed here in three major areas: 1) implications for the model used In this Study; 2) implications for measurement; and 3) recommendations for future research. These areas are addressed in the following section of this chapter. Implications for the Study Model In this study, no relationship was found between perceived severity of MI and compliance with the therapeutic regimen. In addition, modifying factors used in this study (age, education, work outside the home, annual income, number of MI's, time since MI, in-hospital teaching, current participation in a cardiac teaching program and additional chronic diseases) were not related to perceived severity. However, various modifying factors were found to be significantly related (at the p w4 zo_ezu>¢mez_ uz_m¢:z m _ 1 machue_¢m>mm om>_mu¢wd _ A/ /\ zo_kzm>¢mkz_ oz_mmaz mommom_u o_cocco .mco_u_ocm Emcmoca mc_zomou c_ co_uma_o_ucme ocecunu mc_cumou .mu_amozuc_ .2 ouc_m oE_u m._: mo Lopez: oeooc_ _m:ccm ueoz one uc_mu:o xcoz co_umo:ou pom m¢0pum_ooz in This Study. Used Modification of the Model Final ure 7 Fl 156 therapeutic outcomes for patients with given chronic diseases. For the purposes of some studies, researchers may be more interested in therapeutic outcome than In patient compliance. The study of perceptions is a very difficult area in which to conduct research. Conceptualizing and operationallzing perceptual variables is difficult, as evidenced by the wide variety of methods of measurement used in past studies. Implications for Measurement In the literature review presented In Chapter III, this researcher presented other authors views regarding the measurement of perceived severity. A wide variety of measures exists and researchers attest to the need to tailor measurement to specific health states or diseases. One conclusion can be drawn regarding the measurement of perceived severity; most researchers who measure perceived severity believe that there are multiple dimensions to the concept. In addition to a generalized disease severity belief, other researchers have identified such areas as role changes, physical capabilities, financial or career dimensions and physical symptoms of anxiety. Inclusion of a scale for each of the above dimensions identified might greatly increase the accuracy of measurement. Scale development is needed for the concept of perceived severity of MI. The use of only one Item to measure severity in this study has undoubtedly influenced the findings. Since two medication compliance Items came close to significant relationships with perceived severity, further severity scale development may have produced significant findings due to more accurate measurement. 157 Measurement problems are also evident in the compliance items. All the time, more than half the time, about half the time, less than half the time and never (and not applicable) were the compliance choices offered to subjects in the present study. These choices are probably adequate to show direction and strength of relationship for the purposes of this study. In studies where researchers must decide what Is and what is not compliant behavior, these choices would be too ambiguous to operationallze the concept. In measuring compliance, it is important that the researcher know exactly what the subject was told to comply to in terms of the therapeutic regimen. An ideal study would provide all subjects with the same information about specific regimens before collecting data regarding compliance. In the present study, no such information is available. Subjects could have interpreted some items in different ways. One patient may be more compliance than another to dietary restrictions, for example, but may indicate less compliance on the questionnaire due to individual interpretation. More specific questionnaire items would leave less room for individual interpretation. This is why specific information is more valuable to compliance measurement. Asking the patient how many eggs per week or servings of red meat per week that he eats would yield more useful information, for example, than a general question about frequency of dietary adherence. 158 Implications for Future Research In summary, implications of the present study for nursing research have been identified. Recommendations for future research are outlined below: 1. Random sampling and the limitations of patients to those less than 60 years old and those with a first MI is recommended to limit confounding variables. Women and other minority groups should be used in further studies to test the research questions in these populations. The limitation of data collection at specific time intervals post-MI would allow the study of perceptual differences in severity at these different points in time. Further instrument development is needed to measure perceived severity across other possible dimensions. It is suggested that the following dimensions should be investigated: functional or life-style change, physical symptoms of worry/anxiety, financial or career dimensions, comparison of severity across different illnesses. Further instrument development is needed to measure compliance. specific behaviors, such as substituting low fat margarine for butter, should be measured. Other aspects of the regimen should be included in further studies. The entire Health Belief Model should be used in future studies to test the models' use in the post-MI population. 159 7. All subjects should receive the same patient education information regarding the regimens before compliance is measured. Ideally, a specially trained nurse should educate all subjects. 8. Patient knowledge and specific information about the regimen should be recorded in further studies dealing with compiance to those regimens. 9. Future research should be directed at specific intervention strategies which might modify perceptions and behavior of those with known low compliance levels. Subjects should all receive similar patient education and should be randomly divided into intervention groups and control groups. 10. More studies are needed to investigate the role of the nurse in facilitating compliance with chronic disease regimens. lmplications for Nursing Practice Implications for advanced nursing practice derived from this study include implications specific to the study findings as well as more generalized implications for nursing practice. Both are addressed below. Specific Implications In this study, no relationship was found between perceived severity of MI and compliance with the therapeutic regimen. This researcher has not concluded that no such relationship exists. Therefore, nurses in advanced practice must consider the possibility that perceived severity of MI might effect patient compliance as 16D postulated in the Health Belief Model and act accordingly. Perceptions of severity should be estimated based on patient interview and observation. The CNS should be aware that a low level of perceived severity may indicate denial or may be too low to motivate the person to act. Denial requires specific psychosocial interventions. Low perceptions of severity may require more information exchange to increase the patients knowledge of the disease. Very high levels of anxiety resulting from high levels of perceived severity may interfere with compliance. Counseling to reduce anxiety Is then indicated. In this study, certain modifying factors were found to be related in some way to compliance. These factors are working outside the home (Increased activity regimen compliance) time in months since the MI occurred (increased medication, exercise and general compliance), past participation in an in-patient teaching program (increased dietary and general compliance), current participation in a cardiac teaching program (decreased exercise compliance) and the presence of additional chronic diseases (Increased exercise compliance). Nurses in advanced practice can use these findings to predict patient compliance. Since the modifying factors used in this study cannot be altered in the post-MI patient (with the exception of current participation in a cardiac teaching program, which could be encouraged by the CNS), the CNS can do little to intervene based on these findings. The findings regarding decreased exercise compliance with current participation in a teaching program is not supported by the literature. 161 In summary, the specific implications for advanced nursing practice are based on the researchers skepticism regarding the validity of the present findings. These findings are in conflict with most current literature regarding perceptions and compliance. General Implications General implications for advanced nursing practice in the primary care setting are discussed within King's framework of goal attainment. In Kings' theory, individual perceptions, beliefs and values are formed prior to entering the interaction phase between nurse and patient. Therefore, an individual patient has perceptions regarding the severity of his Ml that he brings with him to the therapeutic interaction with the nurse. In turn, the nurse brings expert knowledge as well as perceptions, beliefs and values to the interaction. In interacting with the post-MI patient, the Clinical Nurse Specialist (CNS) must be aware of the importance of individual perceptions and the possible factors that may influence these perceptions. The nurse begins each interaction at the assessment phase in the nursing process where Information regarding these perceptions is collected. While adequate research findings are not available to confirm relationships between perceived MI severity and compliance with the therapeutic regimen, the nurse must attempt to answer these questions in practice by assessing levels of perceived severity of MI and levels of compliance. 162 In assessing the post-MI patients level of perceived severity, the CNS must attempt to identify the factors that influence these perceptions of severity. While the literature generally does not contain evidence that demographic and illness variables are related to perceived severity, all possibilities must be considered until a body of knowledge developed though research has identified the most fruitful areas of intervention. The assessment of compliance is difficult and often inaccurate. The CNS must develop competent assessment skills related to compliance and should rely on a variety of methods for maximum accuracy in measurement. In the assessment of compliance, the CNS must be aware of the limitations of present methods of measurement. In addition, the communication of trust and respect for the patient and his decisions is necessary in order to foster an appropriate environment in which the patient can confide noncompliance to the practitioner. Good interviewing skills are essential. In view of difficulties defining the measuring patient compliance as well as assessing it, considerations should be given to the use of therapeutic outcomes as a major goal for nurses in advanced practice. Compliance would then be seen as one of the means to achieve the goal of a particular outcome (e.g., decreased blood pressure) and the nursing focus would be placed on the outcome goal rather than compliance. Therapeutic outcome goals are easier to define, measure and document and may facilitate the delivery of primary care. Some therapeutic outcomes, however, depend strongly upon compliance to a regimen. 163 Following assessment, data gathered provides the framework for planning, implementation and evaluation of nursing interventions. In the interaction phase between nurse and patient, both strive to agree upon mutual goals. It is important that the nurse give the patient all the information he needs to begin to form goals and the plans to achieve them. King (1981) states that individuals have the right to knowledge about themselves, to participate in their own health care decisions and that nurses have the responsibility to share information that helps individuals make informed decisions. The nurse must also be aware that Individuals may need to take time to synthesize information and may not be ready to begin working on anything short of brief, short-term goals at the beginning of the interaction. In the present study, compliance was related to time in months since the MI occurred and increased as time passed. Mutual goals involving complex behavior changes may take more time than initially expected to develop, especially with frequent changes in treatment regimens. In some instances, it may be difficult to determine whether or not the goals developed are truly mutual. Patients know that it is socially acceptable to follow recommendations and may appear to participate in forming goals but never reach the transaction phase of goal attainment. The nurse must assist the post-MI patient toward truly mutual goals by assisting him to develop insight into his motivations In a nonjudgmental way. Nurses must seek clarification of the patients needs and goals to avoid erroneous assumptions 16A regarding them. Questions like ”How can I help you?” and "What would you like to accomplish from our visits"? can assist In this clarification. Ethical issues arise when nurses encounter patients with goals that conflict with health needs. Values clarification, patient education and assisting patients in seeking insight into these motivations are necessary first steps in nursing intervention. Ultimately, the patient has the right to make his own decisions regarding health behaviors but this does not change nursing responsibilities. In addition to providing patient education, the CNS may need to assist the post-MI patient in altering inaccurate perceptions. The nurse must be aware, as the negative direction of the relationship between perceived severity and medication compliance in this study may indicate, that high levels of perceived seriousness may interfere with mutual goal setting and compliance by increasing anxiety beyond a level at which the individual is presently prepared to cope. In this case, the nurse must assess present coping abilities and resources and assist the post-MI patient in coping and anxiety reduction. The nurse must also be aware that denial is a common manifestation of this anxiety and must be dealt with before the patient is able to be active in his plan of care. Anxiety reduction can be achieved through counseling and the activation of individual, family and community resources. Persistent denial and high levels of anxiety may require referral to mental health specialists. 165 According to King (1981), individuals are part of larger systems of family and community. The CNS must be aware that the post-MI patient is a part of these systems and that utilization of family and community resources is an integral part of the nursing care plan. There are many specific interventions in nursing literature that can be used to assist the patient in meeting compliance goals. The nurse and patient should decide together which of these are most appropriate for the individual patient. Initially, nursing interventions can be directed at increasing the patients' knowledge of risk factors and risk factor reduction. With this knowledge, patients become aware that they can influence future health and are not just passive sufferers of disease. In addition to basic teaching interventions, written prescriptions, patient self-monitoring (e.g., food and exercise diaries), behavior modification techniques, participation in community heart programs, patient contracting and other nursing interventions may be used to assist the post-MI patient in achieving treatments goals. Whichever interventions are chosen, continuity of care is essential to facilitate compliance. The nurse in advanced practice is in an ideal position to monitor chronic illness, evaluate the prescribed treatment plan and the patients compliance with it, and provide nursing interventions that may influence the long-term well-being of the post-MI patient through risk factor reduction. Longitudinal follow-up and coordination with other health care disciplines ensures continuity of care. 166 Nursing assessment at each health care visit will yield feedback (in the form of new information regarding perceptions and compliance) that should be incorporated into further nurse-patient Interactions. Results of a continual evaluation process indicate the effectiveness of the nurses interactions. If the interventions are effective, outcome evaluation should indicate the presence of appropriate behaviors to reduce the risks of an additional MI. The CNS must be aware that compliance with regimens involving life-style changes is extremely difficult for many patients. Some victims of chronic disease struggle a lifetime to comply with treatment regimens. Others may give up in resignation. Some patients readily admit defeat to health care providers. These patients provide challenges to nurses in advanced practice. Kings nursing theory of goal attainment can be used as a framework in promoting quality of life through facilitation of decision making and reduction of risk factors for post-MI patients. In summary, Kings theory of goal attainment can be applied to the facilitation of compliance with the post-Ml treatment regimen, leading to a reduction of risk factors for recurrent MI. The nurse interacts with the post-MI patient to make an assessment of the patients perceptions of MI severity and compliance with the post-MI therapeutic regimen. Data collected in the assessment phase regarding perceptions can provide a basis for nursing interventions that facilitate a mutual and effective plan of care for the post-MI patient that will lead to compliance and risk factor reduction. 167 Although the use of King's model to increase compliance in the post-Ml patient appears promising, the model was not tested in this research. Much more information is needed about other patient perceptions, modifying factors and the quality of nurse-patient interaction before King's model can be confirmed. Implications for Nursing Education Implications drawn from these study findings are pertinent to both graduate and undergraduate nursing programs. Nurses have more contact with patients than any other health care provider and, therefore, have more opportunities to influence health perceptions and health behavior. The Health Belief Model should be presented to nursing students as a theoretical framework for conceptualizing the relationship of perceptions to health care decisions. The Health Belief Model can be used as a health assessment tool to test its theory and its applicability for clinical practice. Nursing students and health educators must be aware, however, that health behavior is complex and is influenced by many factors, some of which may be unknown at the present time. Curriculae should, therefore, include diverse theoretical models and research literature related to health care decision making. The importance of individual perceptions in influencing health behaviors should be stressed in all programs preparing health educators. Nurses must learn to accurately assess patient perceptions regarding chronic disease, therapeutic regimens and 168 health in general. While more information is needed to clarify the role various levels and types of perceptions play in influencing certain behaviors, nurses can begin to incorporate known knowledge and test current theories regarding these perceptions. For example, findings from this research suggest that very high levels of perceived severity of MI may reduce medication compliance in the post-MI patient. In addition to the assessment of perceptions, nursing education must include interventions that facilitate the change or modification of perceptions when research suggests a benefit to the patient. For example, modifying patient perceptions of a very severe MI through patient education and counseling may increase compliance. Nursing curriculae should include education dealing with self-evalution and insight into the students' own perceptions because these perceptions influence the therapeutic interaction of nurse and patient. Effective communication is not possible without this insight (King, 1981). Nurses at all levels need to increase their knowledge of patient compliance. Nurses need to know that compliance is a complex behavior that can be influenced by many factors. Nursing education should include techniques useful in assessing compliance and interventions useful in facilitating compliance. In addition, primary care providers need to be educated in ways to develop effective programs within organizations and communities which facilitate compliance with healthy life-styles as well as 169 treatment regimens. Compliance with therapeutic regimens for patients with chronic disease must be examined in terms of life-style issues. Nurses must be prepared to view compliance with therapeutic regimens as a life-style change and be innovative in planning nursing Interventions to facilitate behavior change for a lifetime. Nursing education programs need to encourage students to investigate ethical issues related to patient compliance. Nurses must be fully aware of their responsibilities to patients labeled ”noncompliant." Because health education of patients is an integral part of nursing care, nursing education must continually strive to upgrade the teaching skills of its' students. Theories of learning and educational techniques must be utilized in instructing patients about treatment regimens, risk factor reduction, and healthy life-styles. Clear understanding of the regimen and its implications is necessary for optimal patient compliance. In summary, nursing education implications from this study were presented. Implications are applicable to undergraduate and graduate nursing programs. Nursing education must prepare nurses with a knowledge of the importance of perceptions in influencing health behaviors and the skills to facilitate these behaviors to promote optimum health and well-being for the post-MI patient as well as others under the care of nurses. 170 Summary In Chapter VI, a summary and interpretation of the research findings was presented. King's nursing theory of goal attainment was used as a conceptual framework with which to view nursing intervention to increase compliance in the post-MI patient. Implications were drawn from the study findings that are relevant to nursing research, nursing practice and nursing education. BIBLIOGRAPHY American Heart Association. (1985). Heart Facts 1985. Dallas, TX: Author. Andreoli, K. (1981). Self-concept and health beliefs in compliant and non-compliant hypertensive patients. NursingeResearch, 32(6). 323-328. Andreoli, K., Fowkes, V., Zipes, 0., 8 Wallace, A. (Eds.). (1983). Comprehensive cardiac care. St. Louis: C.V. Mosby Company. Barnwell, W. (1979). Cardiac rehabilitation. The Journal of the South Carolina Medical Association, 15(A), A20-A22. Becker, M. (Ed.). (197A). The health belief model and personal health behavior. New Jersey: Charles Slack, Inc. Becker, M., Orachman, R., Haefner, D., Kirscht, J., 8 Maiman, L. (1977). The health belief model and prediction of dietary compiance is a field experiment. Journal of Health and Social Behavior, 18(A), 3A8-366. Becker, M., 8 Maiman, L. (1975). Sociobehavioral determinants of compliance with health and medical care recommendations. Medical 9553,,13I1), 10-2A. Becker, M., Radius, 5., Rosenstock, I., Orachman, R., Schuberth, K., 8 Teets, K. (1978). Compliance with a medical regimen for asthma: A test of the health belief model. Public Health Re orts, 93(3), 268-277. Billings, E., Lindell, B., Sederholm, M., 8 Theorell, T. (1980). Denial, anxiety and depression following myocardial infarction. ngchosomatic , 21(8), 639-6A5. 172 Bonnar, J., Goldberg, A., 8 Smith, J. (1969). Do pregnant women take their iron? Leneep,‘l(1), A57-A58. Borhani, N. (1981). Epidemiology of hypertension as a guide to treatment and control. Heart 8 Lumg,‘lg(2), 2A5-25A. Brown, W., Ginsberg, H., 8 Karmally, W. (198A). Diet and the decrease of coronary heart disease. American Journal of Cardiolo , 53(A), 27C-29C. Bruce, R. (197A). The benefits of physical training for patients with coronary heart disease. In F.J. Ingelfinger (Ed.), Controversy in Internal Medicine, Vol 2 (PP. 1A5-161). Philadelphia: W.B. Saunders Company. Champion, V. (198A). Instrument development for health belief model constructs. Advances in Nursing Science, §(3), 73-85. Comoss, P., Burke, E., 8 Swails, S. (1979). Cardiac rehabilitation: A comprehensive nursing approach. New York: J.B. Lippincott Company. Croog, S., 8 Levine, S. (1982). Life after a heart attack. Boston: Human Sciences Press, Inc. Croog, 5., Shapiro, D., 8 Levine, S. (1971). Denial among male heart patients. Psychosomatic Medicine, 33(5), 385-397. Cummings, K., Jette, A., 8 Rosenstock, I. (1978). Construct validation of the health belief model. Health Education Monographs, §(A), 39A-A05. Cummings, M., Kirscht, J., Binder, L., 8 Godley, A. (1982). Determinants of drug treatment maintenance among hypertensive persons in inner city Detroit. Public Health Reports, 31(2), 99-106. 173 Davis, M. (1968). Physiologic, psychological and demographic factors in patient compliance with doctors orders. Medical Care, §(2), 115-122. Davis, M. (1982). Forward. In M. DiMatteo 8 D. DiNicola (Eds.), Achievingypatient compliance (pp. xi-xiii). New York: Pergamon Press. Dawber, T. (1980). The Framingham study: The epidemiology of atherosclerotic disease. Cambridge: Harvard University Press. Dehn, M. (1980). Rehabilitation of the cardiac patient: The effects of exercise. American Journal of Nursing, §Q(3), A35-AAO. DiMatteo, M., 8 DiNicola, D. (1982). Achievingipatlent compliance. New York: Pergamon Press. DiMatteo, M., 8 Friedman, H. (1982). Sociopsychology and medicine. Cambridge, MA: Oelgeschlager, Gunn and Hain. Devney, A. (1980). Bridging the gap between inhospital and outpatient care. American Journal of Nursing, §Q(3), AA6-AA9. Dracup, K., 8 Meleis, A. (1982). Compliance: An interactionist approach. NurelngiResearch, 31(1), 31-36. Dracup, K., Meleis, A., Baker, K., 8 Edlefsen, P. (198A). Family-focused cardiac rehabilitation. NursingyClinics of North America, 13(1), 113-12A. Dunbar, J. (1980). Assessment of medication compliance. In R. Haynes, M. Mattson, 8 T. Engebretson (Eds.), Patient compliance to prescribed antihypertensIveegegimens. Bethesda, MD: National Institutes of Health No. 81-2102. 17A Edel, M. (1985). Noncompliance: An appropriate nursing diagnosis? NursinglOutlook, 33(A), 183-185. Fletcher, S., Papplus, E., 8 Harper, 5. (1979). Measurement of medication compliance in a clinical setting: Comparison of three methods In patients prescribed digoxin. Archives of Internal medicine, 139(6), 635-638. Gentry, W., Baider, L., Oude-Weme, J., Musch, F., 8 Gary, H. (1983). Type A/B differences In coping with acute myocardial infarction: Further considerations. Heart 8 Lung, 13(3), 212-21A. Gentry, W., 8 Williams, R. (1979). Psychologlgpl aspects of myocardial infarction and coronary care. St. Louis: C.V. Mosby Company. Gillum, R., 8 Barsky, 0. (197A). Diagnosis and management of patient noncompliance. Journal of the American Medical Association, 32302), 1563-1567. Given, B., 8 Given, C. (1982). Patient contribution to care: Link to process and outcome. Unpublished summary report #5 R01 NU 0062. East Lansing, Michigan: Michigan State University. Glanz, K. (1980). Compliance with dietary regimens: Its magnitude, measurement and determinants. Preventive Medicine, 3(6), 787-80A. Glanz, K., Kirscht, J., 8 Rosenstock, I. (1981). Linking research and practice in patient education for hypertension. Medical C_ar_e_,13(2), 1111-152. Goldbarg, A. (1973). Rehabilitation of the coronary patient. Medical Clinics of North America, 51(1), 231-2A1. 175 Gordis, L. (1979). Conceptual and methodological problems in measuring patient compliance. In R. Haynes, D. Taylor and D. Sackett (Eds.), Compliance in health care (pp. 23-A5). Baltimore: The Johns Hopkins University Press. Goroll, A., May, L., 8 Mulley, A. (1981). Primary care medicine. Philadelphia: J.B. Lippincott Company. Greene, J., Weinberger, M., Jerin, M., 8 mamlin, J. (1982). Compliance with medication regimens among chronically ill, inner city patients. Journal of Community Health, 1(3), 183-193. Grundy, 5. (198A). Hyperlipoproteinemia: Metabolis basis and rationale for therapy. American Journal of Cardiolegy,.5fl(A), 20C-26C. Haefner, D., 8 Kirscht, J. (1970). Motivational and behavioral effects of modifying health beliefs. Public Health Reports, 35(6), A78-A8A. Haynes, R. (1979). Introduction. In R. Haynes, D. Taylor 8 D. Sackett (Eds.), Compliance in health care (pp. 1-7). Baltimore: John Hopkins University Press. Haynes, R. (1982). Improving patient compliance: An empirical view. In R. Stuart (Ed.), Adherence, compliance and generalization in behavioral medicine (PP. 56-78). New York: Brunner/Mazel Publishers. Haynes, R., Taylor, D., Sackett, 0., Gibson, E., Berhzolz, C., 8 Mukherjee, J. (1980). Can simple clinical measures detect patient noncompliance? Hypertension, 3(6), 757-76A. 176 Hellerstein, H., 8 Franklin, B. (1978). Exercise testing and prescription. In N. Wenger 8 H. Hellerstein (Eds.), Rehabilitation of the coronary patient (PP. 1A9-202). New York: John Wiley and Sons. Hirsch, E. (197A). Physical training and coronary heart disease. In R. Morse (Ed.), Exercise and the heart (PP. 106-187). Springfield, IL: Charles C. Thomas, Publisher. Hijeck, T. (198A). The health belief model and cardiac rehabilitation. Nursinnglinics of North America, 13(3), l1A9-1157. Hilbert, G. (1985). Spouse support and myocardial infarction patient compliance. Nursing Research, 33(A), 217-220. Hjalmarson, A., Herlitz, J., Malek, I., Ryden, L., Vedin, A., Waldenstrom, A., Wesel, H., Elmfeldt, D., Holmberg, S., Nyberg, G., Swedberg, K., Waagstein, F., Waldenstrom, J., 8 Wilhelmsen, L. (1981). Effect on mortality of metroprolol in acute myocardial infarction. The Lancet, 8250(2), 823-827. Hyman, M., Insull, W., Palmer, R., O'Brien, J., Gorden, L., 8 Levine, B. (1982). Assessing methods for measuring compliance with a fat-controlled diet. American Journal of Public Health, 13(2), 152-160. Janz, N., 8 Becker, M. (198A). The health belief model: A decade later. Health Education Quarterly. 1 (l), 1-A7. 177 The Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. (198A). The 198A report of the Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure. Archives of Internal Medicine, 1AA(5), 10A5-1057. Jones, D., Dunbar, C., 8 Jirovec, M. (1978). Medical-sumgical nursing: A concepgual approach. New York: McGraw-Hill Book Company. Kallio, V. (1978). Results of rehabilitation in coronary patients. Advances in cardiology: Cardiac rehabilitation, 33, 153-163. Karllnger, J., 8 Gregoratos, G. (Eds.). (1981). Coronary care. New York: Churchhill Livingstone. Kasl, 5. (197A). The health belief model and behavior related to chronic illness. In M. Becker (Ed.), The health belief model and pemsonal health behavior (pp. 106-127). Thorofare, NJ: Charles B. Slack, Inc. Kegeles, S., Kirscht, J., Haefner, D., 8 Rosenstock, I. (1965). Survey of beliefs about cancer detection and taking Papanicoloau tests. Public Health Reports, §Q(9), 815-82A. Kung, I. (1981). A theogyyfor nursing. New York: John Wiley and Sons, Inc. Kline, N., 8 Warren, B. (1983). The relationship between husband and wife perceptions of the prescribed health regimen and level of function in the marital couple post-myocardial infarction. Family Practice Research Journal, 3(A), 271-280. 178 Leventhal, H. (1973). Changing attitudes and habits to reduce risk factors in chronic disease. American Journal of Cardiology, 31, 571-581. Levy, R. (198A). Causes of the decrease in cardiovascular mortality. American Journal of Cardiology, 33(A), 35C-A1C. Levy, R. (198A). Discussion: The decrease in coronary heart disease mortality: Status and perspectives on the role of cholesterol. American Journal of Cardiology, 5&(A), 35C-A1C. Lewin, K. (19A8). Resolvingesocial conflict. New York: Harper 8 Row. Loustau, A. (1979). Using the health belief model to predict patient compliance. Health Values, 3(5), 2A1-2A5. Marston, M. (1970). Compliance with medical regimens: A review of the literature. Nursigg Research, 13(A), 312-323. Mayou, R., Foster, A., 8 Williamson, B. (1979).' Medical care after myocardial infarction. Journal of Psychosomatic Research, 33(1), 23-26. Mayou, R., Williamson, B., 8 Foster, A. (1976). Attitudes and advice after myocardial infarction. British Medical Journal, 1, 1577-1579. McGurn, W. (1981). People with cardiac problems: Nursing concepts. Philadelphia: J.B. Lippincott Company. Mikhail, B. (1981). The health belief model: A review and critical evaluation of the model, research and practice. Advances in NursimgeScience,Ig(10), 65-82. 179 Miller, P., Wikoff, R., McMahon, M., Garrett, M., 8 Johnson, N. (1982). Development of a health attitude scale. Nursing Research, 31(3), 132-136. Miller, P., Wikoff, R., McMahon, M., Garrett, M., 8 Ringel, K. (1985). Indicators of medical regimen adherence for myocardial infarction patients. NursingiResearch, 33(5), 268-272. Morisky, D., Bowler, M., 8 Finlay, J. (1982). An educational and behavioral approach toward Increasing patient activation in hyeprtensive management. Journal of Community Health, 1(3), 171-181. Mulder, J. (1981). Patient compliance to individualized home exercise programs. Journal of Family Practice, 11(6), 991-996. National Heart, Lung 8 Blood Institute. (1982). National Heart, Lung_ and Blood Institute fact book for fiscal year 1982 (NIH Publication #83-1525). Bethesda, MD: National Institute of Health. Niccoli, A., 8 Brammel, H. (1976). A program for rehabilitation in coronary heart disease. Nursing Clinics of North America, 11(2), 239-250. Oberman, A. (1985). Exercise and the primary prevention of cardiovascular disease. American Journal of Cardiology, 35(10), lOD-ZOD. Owens, J., McCann, C., 8 Hutelmyer, C. (1978). Cardiac rehabilitation: A patient education program. Nursing Research, 11(3), 1A8-150. I80 Payne, F., 8 Boineau, J. (1980). Cardiac rehabilitation. American Family Physician, 31(A), 152-156. Peck, C., 8 King, N. (1982). Increasing patient compliance with prescriptions. Journal of the American Medical Association, 133(21), 287A-2878. Petersdorf, R., Adams, R., Braunwald, E., Isselbacher, K., Martin, J., 8 Wilson, J. (Eds.). (1983). Harrison's principles of internal medicine. New York: McGraw-Hill, Inc. Polit, D., 8 Hungler, B. (1983). Nursinglresearch: Principles and methods. Philadelphia: Lippincott Company. Rifkind, B. (198A). Lipid research clinics coronary primary prevention trial: Results and implications. American Journal of Cardiolo , 33(A), 30C-3AC. Rosenstock, 1. (197A). Historical origins of the health belief model. In M. Becker (Ed.), The health belief model and personal health behavior, (pp. 1-8). New Jersey: Charles S. Slack, Inc. Rudd, P. (1979). In search of the gold standard for compliance measurement. -Archives of Internal Medicine, 139(6), 627-628. Ruffalo, R., Garbedian-Ruffalo, S., 8 Pawlson, G. (1985). Patient compliance. American Family Physician, 31(6), 93-100. Sackett, 0., 8 Snow, J. (1979). The magnitude and measurement of compliance. In R. Haynes, 0. Taylor, 8 0. Sackett (Eds.), Compliance In health care (pp. 11-22). Baltimore: Johns Hopkins University Press. 181 Schneck, 0. (1985). Calcium-entry blockers: A review of their basic and clinical pharmacology and therapeutic applications. Rational DrugeTherapy, 15(5), 1-5. Shephard, R., Corey, P., 8 Kavanagh, T. (1981). Exercise compliance and the prevention of a recurrance of myocardial infarction. Medicine and Science in Sports and Exercise, 13(1), 1-5. Stanitis, M., 8 Ryan, J. (1982). Noncompliance: An unacceptable diagnosis? American Journal of Nursin , 15(3), 187-197. Stunkard, A. (1981). Adherence to medical treatment: Overview and lessons from behavioral weight control. Journal of Psychosomatic Research, 15(3), 187-197. Taylor, W. (1979). A test of the health belief model. In B. Haynes, W. Taylor 8 0. Sackett (Eds.), Compliance in health care. Baltimore: Johns Hopkins University Press. Thomas, S., Sappington, E., Gross, H., Noctor, M., Friedmann, E., 8 Lynch, J. (1983). Denial in coronary care patients: An objective reassessment. Heart 8 Lung, 11(1), 7A-80. Tyroler, H. (198A). An overview of lipid research clinics (LRC) epidemiologic studies as background for the LRC coronary primary prevention trial. American Journal of Cardiology, 55(A), 1AC-19C. U.S. Bureau of the Census (198A). Money income of households, families and persons in the United States: 1982 (Current Population Reports, Series P-60, #1A2). Washington, DC: U.S. Government Printing Office. 182 U.S. Department of Health and Human Services (1983). Unived States health and prevention profile (PHS Publication #8A-1232). Washington, DC: U.S. Government Printing Office. Wenger, N. (1978). Early ambulatlon after myocardial infarction; Rationale, program components, results. In N. Wenger 8 H. Hellerstein (Eds.), Rehabilitation of the coronary patient (PP. 53-66). New York: John Wiley 8 Sons. Wenger, N. (1979). Research related to rehabilitation. Circulation, 92(7), 1636-1639. Wenger, N. 8 Hellerstein, H. (Eds.). (1978). Rehabilitation of the coronary patient. New York: John Wiley 8 Sons. Winslow, E., 8 Weber, 0. (1980). Rehabilitation of the coronary patient: Progressive exercise to combat the hazards of bed rest. American Journal of Nursing, 39(3), AAO-AAS. Yoos, L. (1981). Compliance: Philosophical and ethical considerations. Nurse Practitioner, 5(5), 27-3A. APPENDIX A Letter of Introduction to the Project MICHIGAN STATE UNIVERSITY COLLEGE 01' NURSING EAST LANSING ' MICHIGAN ' WM January 28. 1981 Mr. Mark Vsenendaal. 11.0. Cardiac Cllnlc, Sulte 220 405 W. Greenlawn Lansing, MI 48910 Dear Dr . Veenendaal : As a protesslcnal working with post. myocardial lntsrctlon and other cardiac impala-ed clients. you are well aware or the pervaslveness or these problems upon the lltestyle ct their victims and tallies. We are in the process of conductlng a research project. designed to collect descriptive data tron marital couples who have recently experienced myocardial lntarctlon. Specifically. we are studying men between the ages of. 39 and 65 who have experienced an N: wlthln the past year and their wlves. It is 9gp mandatory that both members or the marltal unit particlpsta in the study. Volunteers are being asked to complete a tlvc part guestlcnnalro which would take approximately 45 minutes o: thelr plus. the materlals covered lnclude: perception ct and adherence to the health regimen, level or tuncclonlng as a marltal unit, and general scale-demographic data. You could greatly ald our ottorts by asking for partlclpatlon from among your clients who meet the above crlterla. We wlll gladly set. up appoint- ments to administer the guestlomalre to the HI vlctlns and their wives. at. thelr convenience. Subjects wlll be tree to withdraw from the study at. any tine. Also, we wlll be more than happy to share our tlndlngs with you and the partlclpants. We will be contacting you wlthln the next few weeks in order to explain the study further, answer any questions you might. have, and to clarity the administrative protocol. Our research methodology ls designed to require pp the iron you personally. Thank you tor your consideration! Sincerely, Brlgld )1. Warren, 8.11., 11.8. and Nancy W. Kline. 3.14.. 81.11. Assistant. Professor 0! Nursing Assistant Professor ct Nursing Co-Pr lnclpal lnvestlgaeor Oo-?rlnci.pal Investigator 831111: Six/dds N-iUises W MWWWHM MICHIGAN STATE UNIVERSITY fi COLIEG‘ 0' NURSING WT LANSING ' MICHIGAN ' “I“ January 28. 1981 George Kleiber. 0.0. Cardiac Clinic. Suite 220 405 w. Creenlawn Lansing. MI 48910 Dear Dr. kleibers This is a follow-up to the letter of January 33. 1980 in which we described our research project about the post-myoardial infarction rehabilitation experience. Since that time we have 'disoussed the project with Dr. Walter baird who advised us that our project had been approved by the when subjects heview Committee and the board of Directors at Ingham IIedical Center. Dr. Baird has given us permission to contact his patients to' participate as subjects in this investigation. We are hoping that you will also agree to allow us access to your patient population. We'will be happy to share the results with you upon completion of the study. Again. this project will require _n2_ time from you personally. beyond the initial discussion and approval. we will be contacting you soon to discuss the study further. Thank you for your consideration! Sincerely. Brigid 5. Warren. 8.)!" 34.3. and Nancy W. kline. 3.8.. 86.8. Assistant Professor ' Assistant Professor aawmwx/sje “M I II as NW“? .4. “no 1'7qu W; Immune APPENDIX B Physician Letter to Potential Study Participants October 23. 1980 Dear a I am writing to you about a research study being conducted by Nancy kline and Brigid Warren. Assistant Professors in the Michigan State University College of Nursing. They are gathering information about how the experience of a heart attack has affected the lives of married couples like yourself. Through this study. they hope to gain information that will help nurses more effectively assist couples who have recently undergone a heart attack experience. Agreeing to participate in this study would be of no cost to you other than approximately forty-five minutes of your time. Both of you would individually complete a five-pert questionnaire which asks about yourself. your spouse. and the post-heart attack treatment plan. You would be free to withdraw from the study at any time and your identity would be kept strictly confidential. Please consider their request for your participation. Complete the enclosed postcard and return it to the College of Nursing at your earliest convenience. Thank you. . Sincerely. Dr. Hartin Jones. H.D. Made APPENDIX C Instruction Letter to Study Participants MICHIGAN STATE UNIVERSITY COLLEGE OI‘ NATUIl-AL SCIENCE ' m 0' NURSING EAST LANSING ' MICHIGAN ' WM Hello: Thank you very much for agreeing to participate in our research study. Your assistance will help guide nurses and other health professionals in the care of people who have had a heart attack. Following this letter. you should find five (5) questionnaires asking different questions about yourself. your spouse and the post-heart attack treatment plan. they will take about 65 minutes to complete. Please answer each question as honestly as possible. There are no right or wrong answers. PLEASE DO NOT CONSULT YOUR SPOUSB "BIL! YOU All FILLING OUT THE FORMS. Thank you again! If you have any questions or comments. feel free to contact any of the investigators. Sincerely. Nancy H. Kline, 3.8.. “.8. WM WW Br gid A. Warren. 3.8.. 3.8.8. NUKIBAH/jlm xii i APPENDIX D Instruments Pt. No. 9 a. 1-6 . Site 546 QUESTIONNAIRE I 1 D‘“ 7_9 Card No. 01 10-11 Rehab. Code 12 U‘ Since your heart attack you may have been told to follow a treatment plan by doctors, nurses. and other health professionals. Your plan was individually designed to meet your health needs. we are interested in finding out about your treatment plan. This questionnaire asks what you feel you are sugggged gouge as part of your treatment plan. It does NOT ask what you are actually doing. for each statement below, indicate whether you feel this is or is not part of your current treatment plan by checking (X) ”yes" or "no". Remember. there are no right or wrong anewers. Since my heart attack. the treatment plan prescribed for me includes: 1. modifying my smoking habits. 2. controlling my weight. Yes 3. following a special diet. Yes No ? 1 Yes l 6. reading food labels in order to decide if the food is allowed on .y dt.‘ s Y“ T No S. preparing foods according to my dietary restriction. Yes 6. taking a specific amount of medication. 7. taking medications on a daily basis. Yes Yes 8. taking medications at the same time every day. 9. using medications that I have been instructed to take when I feel I need them. Yes No ___§__ 10. waiting a few days to report side effects of my medications. No Yes "'2_' 11. following a program of regular exercise at least three times a week. Yes No "'I" --2- 12. continuing an activity when I know my body isn't tolerating the activity very well. Yes 1 13. restricting certain activities. No Yes 2 -T_- 16. planning time for recreational activities. l5. organising my daily activities to minimise stress. 0 1980 Nancy II. Kline, Brigid A. Warren No 1 No l 2 N0— 1 2 Yes No l 2 No __!_. Yes No ._.r_. '_-2—' Yes No ‘15" 13 16 15 16 17 18 19 20 22 23 24 25 26 27 current treatment plan by checking (X) ”yes" or "no". 16. 17. 18. 19. 21. 22. 23. permitting tense situations to build up rather than dealing with them as they come. Yes No ._I__. '-_2" practicing ways to reduce stress and tension. Yes setting short-term goals in order to reduce stress. 1 Yes No l postponing sexual intercourse when tired. upset and after heavy meals Yes 2 using medically recommended positions for sexual intercourse. Yes Yes 1 No T 1 2 ' modifying my daily work activities to reduce stress. Yes 1 2 )(V limiting sy alcohol intake to two drinks a day or less. lo Yes organising activities to keep stair climbing at a minimum. lo 1 for each statement below, indicate whether this i; or is not a part of your 2 No 28 29 30 31 32 33 34 35 Pt. No. ’ H. 1-6 ‘ Site 5-6 . Date QUESTIONNAIRE I 2 7-3.9 Card he. 1E1 lO-ll . Code 12 Since your heart attack you may have been told to follow a treatment plan by doctors. nurses and other health professionals. Your plan was individually designed to meet your health needs. For each statement below. please circle the appropriate response to indicate the extent to which you follow that pgrt of your treatment plan. You will be using the following scale: 1) all the time 2) more than 1/2 the time 3) about 112 the time ‘) less then 112 the time 5) never 6) not applicable to my treatment plan Remember, there are no right or wrong answers. 1. I am smoking. 13 1 2 3 A 5 0 all the time more than l/2 about 112 less than never not applicable the time the time ll2 the to my treatment time plan 2. I am reducing my smoking. 14 1 2 3 I S 0 all the time more than 1’2 about ll2 less than never not applicable time the time l/2 the to my treatment a time plan 3. I am taking measures to reduce or control my weight. 15 l 2 3 b ' 5 0 all the time more than 112 about 112 less than never not applicable the time the time 112 the to my treatment time plan. 6. i follow my diet when eating at home. 15 l 2 3 6' 5 0 all the time more than 112 about 112 less than never not applicable time the time '1/2 the to my treatment time plan 5. I eat foods without reading food labels to decide if they are allowed 17 on my diet. , - '1 2 3 6 5 0 all the time more than 1/2 about 1/2 less than never not applicable time the time 1]! the to my treatment time plan Q9 1980 Yancy H. lline l Brigid A. Darren X\II For each statement. please circle the appropriate response to indicate the extent 6. 10. ll. 13. 14. all the time more than 1/2 the time I ignore my diet when arting away from.hcme all the time more than 1/2 the time to which you follow your treatment plan. about l/2 the time about l/2 the time i takalthe right amount2 of medicine3(s). all the time more than 112 the time I take my medicine(s) on time. 1 2 all the time more than 112 the time 1 report side effects of my medicins(s) to the doctor. 1 2 3 i all the time more than 1/2 the time about ll2 the time 3 about 1/2 the time about 1/2 the time 1 exercise at least three times a w3eek. 1 2 all the time more than l/2 the time about l/2 the time I do my prescribed exercises regularly. 1 2 3 all the time more than 112 the time about l/2 the time I eat ffcds which are prfpared according to my dietary restriction. I discontinue an activity when symptoms indicate poor activity 1 2 3 A 5 all the time more than 112 the time' I rest whenever I feel tired. l 2 all the time more than 1/2 the time about 112 the time xvii 18 0 less than never not applicable l/2 the to my treatment time plan (i.e.‘. restaurasnt. friends).0 19 less than never not applicable 1/2 the to my treatment time plan I S 0 20 less than never not applicable 112 the to my treatment time plan 21 i 3 0 less than never not applicable 1/2 the to my treatment time plan 5 22 S 0 less than never not applicable 1/2 the to my treatment time plan 23 6 5 0 less than never not applicable 1/2 the to my treatment time plan 26 I S 0 less than never not applicable 1/2 the to my treatment time plan tolerance. 25 0 less than never not applicable 1/2 the to my treatment time plen 26 6 5 0 less than never not applicable 1/2 the to my treatment time plan For each statement, please circle the appropriate response Lo ludicatc the extent to which you follow your treatment plan. 15. I follow activity restrictions. 27 1 2 3 t. s 0 all the time more than 1/2 about 1/2 less than never not applicable the time the time l/2 the to my treatment time plan l6. 1 am physically active after meals. Ba 1 4 5 0 all the time more than 1/2 about 1/2 less than never not applicable the time the time 1/2 the to my treatment time plan 17. After I exercise. I do not check my pulse. :9 I 2 6 S 0 all tho time more than 1/2 about 1/2 less than never not applicable the time the time 1/2 the to my treatment time plan l8. 1 make time for recreational activities. 30 l 2 4 S 0 all the time more than 1/2 about 1/2 loss than never not applicable the time the time 1/2 the to my treatment clan: plan 19. I let Ccnuc situationza build up rulhcr than dun! with tin-m .13; Hwy ruuw along. :1 I J 6 5 0 all the time more than 1/2 about 1/2 loss than nuvvr not dpplluuhlu the time the time 1/2 the to my treatment L lint: 1: I..|1 20. I talk freely with my wife about our daily lchs in order to reduce stress. 5: l 2 3 4 5 U all the time more than l/2 about 1/2 less than nuvor not applicable the time the time 1/2 the to my treatment time plan 21. I use relaxation eXerciScs once or twice a day in order to reduce stress. 33 l 2 3 4 S 0 all the time more than l/2 about 1/2 less than never not applicable the time the time 1/2 the to my treatment time plan 22. I pet short-term goals about my daily life in order to reduce stress. 3; . l 2 3 b 5 0 .all the time more than 1/2 about l/2 less than never not applicable the-tine the time 1/2 the to my treatment time plan for each statement. plaese circle the appropriate response to indicate the extent to which you follow your treatment plan. 23. He use medically recommended positions for having sexual intercourse 35 l 2 3 b S 0 all the time more than 1’2 about 1]! less than never not applicable the time the time 1/2 the to my treatment time plan 24. He postpone sexual intercourse when I am tired. upset or after heavy meals. 36 3 4 5 0 all the time more than l/2 about l/Z less than never not applicable the time the time 1/2 the to my treatment time plan 25. Hy alcohol intake is two drinks a day or less. 37 l 2 3 a S 0 all the time more than l/2 about 1/2 less than never not applicable the time the time l/2 the to my treatment time plan 26. I modify my daily work activities to reduce atrees. 33 l 2 3 k S 0 all the time more than l/Z about 1]! less than never not applicable the time the time 1/2 the to my treatment time plan 27. I organise my activities to keep stair climbing at a minimum. 39 2 3 A b all the time more than l/2 about l/2 less than never “the time the time ll; the time 0 not applicable to my treatment plan Pt. No. 9 "1-2-3—4* Site ’SFG QUESTIONNAIRE 15 Date 47-8-9 Card no.11 lO-ll Rehab . Code ’12 SOCIO-DIHOGRAPBIC The following queationa deacribe general thinga about youraelf and your wife. Pleaee anewer.all the queationa to the beat of your ability. There are no right or wrong anawere. All intorlation will be confidential! 1. Ale: 13-14 2. Ethnic background: (Pleaee check (X) appropriate category) 15 white Black M 3 Oriental Hexican-Aeericnn .____5_____ Indian ’ Other ____1r____. 3. Marital etatue: (Pleaec check (X) appropriate category) 16 Harried Separated widowed """I"" """3"" ""'5“"‘ Single Divorced ""_'I"" 4. If earried. age of your wife: 17'18 5. How long have you been married to your current wife? 19'2° 6. Your educational level: (Pleaae check (X) higheet grade completed) 21 1 lower than aeven yeara of echool (gradee 1-6) 2 Junior high echool (gradee 7-9) partial high echool (gradee lO-ll) 3 high echool (co-pieced 12th grade) 4 S __ partial college education (3 years or less) 6 college education (A yeara) 7 beyond 4 yeare of college GD 1980 Nancy H. Kline Brigid A. Warren JXX The following questions describe general things about yOurself and your wife. Please answer all the questions to the best of your ability. There are no right or wrong answers. All infornation will be confidential! 7. wife's level of education: (Please check (X) highest grade coepleted) 22 fewer than seven years of school (grades l-6) i junior high school (grades 7-9) 2 partial high school (grades lO-ll) 1 high school (co-plated 12th grade) a partial college education (3 years or less) 5 college education (i years) 6 s beyond i years of college 7 8. Are you presently working outside the hoes? Yes No 23 I 2 a) If yes. are you working: full tile ~ part time 24 b) What is your current occupation (check(X) one)? 25 clerical 0 professional 1 executive in large-to-eediun-aized business 2 skilled worker 3 4 seliskilled or unskilled worker 5’ owner of business establishment retired 6 7 currently unemployed. but looking for work 8 other (please specify): c) How would you rate the stress associated with your job (check (X) one) high stress eoderata stress low stress no stress xxi -3... The following questions describe general things about yourself and your wife. Please answer all the questions to the best of your ability. There are no right or wrong answers. All information will be confidential! 9. is your wife currently working outside the home? Yes No 27 -- ——1 “T— a) If yes. is your wife working: full time part time 28 . l 2 b) Uhat is your wife's current occupation (check (X) one)? 29 clerical 0 professional l executive in large-to-medium-siaed business 2 skilled worker I semiskilled or unskilled worker 4 owner of business establishment S retired 6 currently unemployed. but looking for work 1 other (please specify): 8 c) how would you rate the stress associated with your wife's job (check (X) one)? 30 high stress 1 moderate stress 2 low stress 3 no stress 4 10. What is your combined annual income? ' 31 '_—5__ 0 - 9.999 ______50.000-59.999 1 l0.000-l9.999 5 60.000-69.999 20.000-29.999 6 70.000-79.999 : 30.000-39.999 7 greater than 80.000 60.000-69.999 8 ll. How many children do you have? 32‘33 -4- The following questions describe general things about yourself and your wife. Please answer all the questions to the best of your ability. There are no right or wrong answers. All information will be confidential! ’ 12, How many children do you have living at home? 34'35 13. Living arrangement: (please check(X) the description which best fits your 35 current living arrangement) I married. living with wife alone 2 married. living with wife and children 3 married. living with wife. children and other relatives ‘ unmarried. living alone (include widower. single. divorced) S unmarried. living with relative(s) or non-related person 6 other (please specify): 14. have you been hospitalized more than one time for a heart attack? 37 Yes No 38- 3‘) 15. How long ago did you have y0ur 1235 heart attack? 40-41 0 years I months 42 16. How severe was your last heart attack? Very severe; l have a lot of heart damage l Hoderately severe: I have some heart damage. but not a lot 2 Hild; l have only a little heart damage 3 I have no heart damage 4 -5- The following questions describe general things about yourself and your wife. Please answer all the questions to the best of your ability. There are no right or wrong answers. All information will be confidential! 17. when you were in the hospital for your last heart attack. did you participate in a heart teaching program? (1) Yes (2) No If yes. please check (X) what topics were covered and if the information was presented to you alone. to your wife alone. or to both of you together. myself ey wife ey wife and medication: alone alone and myself 1 myself 2 my wife 3 ey wife and diet: alone alone and myself 1 myself 2 my wife 3 my wife and smoking: alone alone and myself 1 myself 2 my wife 3 my wife and alcohol: alone alone and myself l myself 2 my wife 3 my wife and exercise: alone alone and myself 1 myself 2 my wife 3 my wife and stress: alone alone and myself 1 myself 2 my wife 3 my wife and work activity: alone alone and myself 1 2 3 18. Did your instruction include ways to deal with problems that could develop at home? Yes No 1 2 19. These statements deal with your current activity level. Please check the one statement that best describes your current activity level. Completely disabled. 1 to bed or chair. Cannot carry on any self-care; totally confined Capable of only limited self-care; confined to bed or chair more 2 than 501 of waking hours. walking about and capable of all self-care; but unable to carry out 3 any work activities; up and about more than 501 of waking hours. Restricted in physically strenuous activity. but walking and able to 4 carry out work of a light or quiet nature. i.e.. light housework. office'work. Fully active: able to carry oat all pro-heart attack activities 5 without restriction. xxi\/ 43 44 45 46 47 48 49 SO 51 52 The following questions describe general things about yourself and your wife. Please answer all the questions to the best of your ability. There are no right or wrong answers. 20. 21. All information will be confidential! Do you eat a special diet? Yes No ""I"" ""2""'- a) If yes. please check(X) all the responses that apply: low salt I low calorie T low cholesterol 3’ other (please specify): T In some co-anities there are special organized prograes for people who have had Are you presently participating in any organized heart program? heart attacks. ___r__ Yes No ____!___. a) If yes. does the program include (Please check (X) all that apply): physical exercise relaxation techniques methods of stress reduction diet instruction general information about a heart attack participation of wives other (please specify): Do you have any chronic health problems? Yes 1 No 2 a) If yes. please check (X) all that apply: arthritis cancer high blood pressure lung disease (asthsma. bronchitis. emphysema) sugar diabetes ether (please specify): X)fll 53 54 55 56 57 58 59 6O 61 62 63 64 65 66 67 68 69 The following questions describe general things about y0urse1f andjyour wife- Please answer all the questions to the best of your ability. There are no right or wrong answers. All information will be confidential! 23. Before your heart attack. did you drink alcoholic beverages? Yes No a) If yes. please specify: occasional 1 weekends only 2 several times a week 3 one or two drinks a day 4 two to five drinks a day 5 more than five drinks a day 24. Before your heart attack. did you smoke cigarettes? Yes No a) If yes. please specify: less than 1/2 pack a day 1 1/2 - 1 pack a day 2 1 - 1-1/2 packs a day 3 1-1/2 - 2 packs a day 4 more than 2 packs a day 5 CARD 12 (Keypunch: Dup. 1-9, 1 2 Dup. 12) 10 ll xxv i 70 71 '72 73 25. Below is a list of things which happen in many families. you experienced in your family during the past year? Please check (X) all that apply. menopause 1 pregnancy 1” an addition in the household 1 retirement ( your retirement 1 a moving 1 marital problems I divorce or separation froe your wife 1 major sickness or injury in your family If death of a close friend or family member 41 children left home 1 got laid off or fired from work 1 concern over aged parents or inlaws I change in work hours or responsibility 1 PLEASE TURN TO THE NEXT PAGE xxn/ii Which of these have 13 14 15 your wife's retirement)16 17 18 19 20 21 22 24 25 For questions 26 through 32. circle the response which best describes how you feel about that statement. 26. 27. 28. 29. 30. 31. 32. I am able to make contact with my doctor easily when I need medical attention. 1 2 3 4 5 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree I have cultural traditions which are a support to me. 1 2 3 4 5 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree My religion is a source of comfort and support during difficult times. 1 2 3 4 . 5 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree Socialiaing with friends is an important part of my life. 1 2 3 4 5 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree Since my heart attack, it has been necessary for me to limit socializing with friends. 1 2 3 4 S 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree Hy treatment plan will allow me to return to my pro-heart attack level of physical activity. 1 2 3 4 S 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree My current physical activity level is no different than before my heart attack. 1 2 3 4 S 6 strongly moderately agree disagree moderately strongly agree agree disagree disagree PLEASE TURN TO THE NEXT PAGE xxviii 26 27 28 29 30 31 32 APPENDIX E Consent Form and Investigators' Statement 33. -lo- 33 Is there anything about yourself or your wife. not covered in this questionnaire. that you would like to tell the investigators? Yes No ""'T"' "'Tf"" If yes. please describe below. YOU HAVE COMPLETED ALL 5 QUESTIONNAIRES. PLEASE CHECK AND MAKE SURE YOU HAVE ANSWERED ALL QUESTIONS. THANK YOU!! MICHIGAN STATE UNIVERSITY COLLEGE OF NURSING CON FORM Investigators: Brigid A. warren. B.N.. B.S.. N.S.N. Assistant Professor of Nursing. Michigan State University A10? Life Sciences Building Telephone: 353-8686 Nancy Xline. B.N.. B.8.. N.N. Assistant Professor of Nursing. Michigan State University A129 Life Sciences Building Telephone: 353-6499 Inves a era 8 ateman The experience of a heart attack may affect peoples' lives in many ways. The purpose of this study is to gather more information about the experience of a heart attack and how it affects the lives of married couples like yourself. we hope that the information gained'fromnthis study will provide a better understanding of how nurses can assist couples who have recently undergone a heart attack experience. Participation in this research study will require approximately forty-five minutes of your time. You will be asked to complete a questionnaire about the health information you have received. your present health practices and your view of your relationship as a couple since your heart attack. In addition. you will be asked information about yourself and your spouse. Your participation in this study will in no way interfere with the care you are now receiving. There will be minimal risk or expense to you. You are free to ask questions now and throughout the study. You may withdraw from the study at any time without jeopardiaing your future care. Your identity will be kept confidential and 92 information that could identify you will be used in any reports of the study. Responses you make on your question- naire will not be revealed to your physician or spouse with any identifying information. unless you so request. The results of the study will be made available to you upon request. \Zr26haasa, (‘4’. £:£!:;‘L;> ture of Investigator XXX‘ Subject's Statement I. the undersigned. agree to participate in this study about people who experience heart attacks. I understand that this study may not benefit me personally but could help future patients with heart attacks. I have been given the opportunity to ask questions and I understand that I may ask questions at any time during the study. I understand that this study will not affect the care I amxnow receiving. I also understand that my anonymity‘will be maintained and that my responses will be kept confidential. I understand my participation in this study is voluntary and that I may withdraw at any time. Signature of Subject Date Hitness Date xxoci APPENDIX F Project Approval Letters from UCRIHS MICHIGAN STATE UNIVERSITY vswrsmv cuwurrrva on amass" mvotvmc w'r tmuuc - woman - um “was tumors cusses. us aouwsnnien suitemc nruumn «flyl3.l98l Professor Nancy w. Kline College of Nursing Dear Professor Kline: Subject: Proposal Entitled. "The Relationship Between Husband and Wife Perceptions of the Prescribed health Regimen and Level of Function in the Marital Couple Post- Hyocardial infarction" The above referenced project was recently submitted for review to the UCRIHS. Us are pleased to advise that the rights and welfare of the human subjects appear to be adequately protected an? the Committee. therefore. approved this project at its meeting on J“'V . 198' . ' Projects involving the use of human subjects must be reviewed at least annually. if you plan to continue this project beyond one year. please make provisions for- ebtaining appropriate UCRIHS approval prior to the anniversary date noted above. Thank you for bringing this project to our attention. If we can be of any future help. please do not hesitate to let us know. Sincerely. H nry E. Bredeck Chairman. UCRIHS HEB/jms cc: Professor Brigid A. Harren xxs