.r it“. y .. . u... | : Adrsuycfi “Lain“: .- I Psnn..(..;fl.v~ol.)u .3... A a fi 91 \. uric . i.“ u . v , . , VI. 1’ ‘1', “2.. E .12.: , , .3}. 1.3.5.: . 3!? . 123.... vs. ‘ “1'0 3’. z t u re: .. Qua-.0... .5! 35.33%": 2...: lit 11:... . v1 .x... Jr... ’3‘}... t7!!! : . . ‘Innli. It: . 7!"! 21:1! flsl.‘ 1:1)...(v .. 13:52.! 1V to?!) 1.17 p! a I. ; 151!’ ‘ {.832 V. n\!-.. . . I . ‘ O. t. :1". THESIS 3 1293 01417 1940 This is to certify that the dissertation entitled DIETARY INSTRUCTIONS IN THE PRESCRIPTION DRUG REGIMENS OF A GROUP OF COMMUNITY DWELLING OLDER WOMEN: INDICATIONS, FREQUENCY, DELIVERY CHARACTERISTICS AND ADHERENCE presented by Donna Frances McLean has been accepted towards fulfillment of the requirements for Ph.D. degree in Human Nutrition a 6nd Major ofessor Date JuIy 28, 1995 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Mlchlgan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. I DATE DUE DATE DUE DATE DUE I MSU Is An Affirmative Action/Equal Opportunity Institution Wilma-o. DIETARY INSTRUCTIONS IN THE PRESCRIPTION DRUG REGIMENS OF A GROUP OF COMMUNITY DWELLING OLDER WOMEN: INDICATIONS, FREQUENCY, DELIVERY CHARACTERISTICS AND ADHERENCE. BY Donna Frances McLean A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1995 ABSTRACT DIETARY INSTRUCTIONS IN THE PRESCRIPTION DRUG REGIMENS OF A GROUP OF COMMUNITY DWELLING OLDER WOMEN: INDICATIONS, FREQUENCY, DELIVERY CHARACTERISTICS AND ADHERENCE. BY Donna Frances McLean The effects and efficacy of prescription drugs can be changed by drug-food or drug-nutrient interactions, particularly when physiological changes related to aging are concurrent. While these interactions can put older adults at risk for increased morbidity and mortality, they also may produce small decrements in physical and social functioning which must be avoided for optimal quality of life. In this study, community dwelling older women in a subsidized apartment complex were surveyed to determine whether they were given dietary instructions indicated for the prescription drugs they took. When dietary instructions had been given, data were collected to characterize the delivery of the instructions, assess adherence and identify variables which predicted adherence. Data were collected in two phases: the first, a self- administered survey that addressed demographic information and prescription drug use screening; the second, a computer- assisted personal interview in the apartment which addressed instruction frequency and delivery characteristics, adherence and adherence prediction variables. Women who returned surveys in the first phase and met specified inclusion criteria formed the sample for the second phase. Delivery of indicated dietary instructions was assessed by interviewer examination of drug containers and medication fact sheets and by interview respondent report. Adherence was assessed with a one-day food and medication record. Only 3% of respondents received all indicated dietary instructions for their prescription drugs, leaving 97% of respondents at risk for drug-food and/or drug nutrient interactions. The mean percent of indicated dietary instructions given per person was 24.6%. Fifty-six percent of drugs which had dietary instructions indicated had no dietary instructions at all given. When dietary instructions were given, respondents adhered to 81.9%. Respondents viewed the doctor as the ultimate source of drug information and the doctor as a source of oral instruction was significantly associated with adherence. Diabetic meal plans were significantly associated with nonadherence. The only significant predictor of adherence was negative: a dietary instruction intended to prevent a reduced drug efficacy interaction. Copyright by Donna Frances McLean 1995 With heartfelt thanks to my parents, Eileen and Edward McLean, and my grandparents, Helen and George Gardiner. They believed that I could never acquire enough knowledge and that I could do whatever I wanted to do. ACKNOWLEDGMENTS I would like to offer sincere appreciation to my major advisor, Dr. Jenny Bond, and the members of my committee, Dr. Wanda Chenoweth, Dr. Larry Krupka, Dr. Lois Lund and Dr. Howard Teitelbaum, who provided guidance, support and encouragement throughout the development and implementation of my research plan. Special thanks are due to Dr. Bond who provided support and counsel even before I began my doctoral program, to Dr. Teitelbaum who went far above and beyond the call of duty to enhance my statistical expertise, and to Dr. Lund who has been a devoted mentor from the beginning. Essential support was also provided by many other people. Dr. Margaret Holmes—Rovner allowed her own research expertise and resources to rub off on me during my tenure as her research assistant. The Center for Health Promotion and Chronic Disease Prevention at the Michigan Department of Public Health (MDPH), and especially my supervisor Dr. Judith Anderson, provided access to the public health library and flexibility in my student assistant position to permit completion of my doctoral program. Norm Keon, also vi at MDPH, solved my “mystery” Epi Info data analysis problem and, in communicating its resolution to The Centers for Disease Control, probably helped other researchers. Mary Perrin at the MSU Computer Center found me a portable computer and maneuvered schedules to allow me to keep it long enough to finish data collection. At the senior apartment complex, Marva Britten and Virginia McClure cheerfully endured endless visits, phone calls and requests for help; residents cheerfully submitted to survey deliveries and interview appointments. A “little help from my friends” assured the successful completion of my dissertation too. My dear, longtime friends, Lee and Nicole Zalewski made a midnight run to deliver their brand new computer to my house for long term loan when my own collapsed from overwork at a critical time. Fellow students Karen Schmitz and Karen Petersmarck provided company and encouragement when the going got rough. WordPerfect expert Lisa Wells patiently taught me how to use the graphics function. Undergraduate researcher Ann Shanks provided second round data entry and expert application of PharmAssist software. Debbie Lecato typed forms, fielded phone calls, bird-dogged details and contributed to my program in uncounted ways. Vii Financial support for my doctoral program and research was provided by Department of Food Science and Human Nutrition graduate assistantships and fellowships; College of Human Ecology graduate assistantships, scholarships and dissertation fellowships; a Department of Medicine research assistantship; and American Dietetic Association Foundation scholarships. Finally, special acknowledgment must go to my mother, Eileen, who welcomed me back home and spent many hours filling multiple roles as cook, bottle washer, bookkeeper, proofreader and cheerleader as I hit the books and the computer. viii TABLE OF CONTENTS LIST OF TABLES ........................................... xii LIST OF FIGURES .......................................... xiv INTRODUCTION ............................................... 1 Statement of the Problem .............................. 1 Research Questions, Objectives and Hypotheses ......... 4 REVIEW OF LITERATURE ....................................... 7 Prescription Drug Use ................................. 7 Physiological Changes in Aging ........................ 9 Adverse Effects of Prescription Drug Use .............. 13 Role of Dietary Modifications in Optimizing Drug Effectiveness and Minimizing Adverse Drug Events ..... 15 Prescription Drug Regimens ........................... l8 METHODS ................................................... 27 Approval to Conduct Study ............................ 27 Study Population ..................................... 27 Measures ............................................. 28 Data Collection ...................................... 36 Statistical Analysis ................................. 46 RESULTS AND DISCUSSION .................................... 49 Sample Characteristics ............................... 49 ix Prescription Drug Use ................................ 51 Dietary Instructions ................................. 55 Medical Outcomes Study SF-36 Survey Scores ........... 58 Characteristics of Dietary Instruction Delivery ...... 62 Attitudes, Beliefs and Knowledge ..................... 66 Adherence ............................................ 70 Summary of Results ................................... 76 CONCLUSIONS AND IMPLICATIONS .............................. 79 Conclusions .......................................... 79 Implications ......................................... 8O BENEFITS, STRENGTHS AND LIMITATIONS OF THE STUDY ........... 89 Benefits ............................................. 89 Strengths ............................................ 89 Demographic Characteristics ..................... 89 Measures and Data Collection Procedures ......... 91 Hypothesis Testing .............................. 91 Limitations .......................................... 92 Sample Size ..................................... 92 Nonresponse Bias ................................ 92 Selection Bias .................................. 93 APPENDICES ................................................ 95 A. Approval to Conduct Study ........................ 95 B. Phase 1 Self—administered Survey ................. 96 C. Phase 2 Computer-assisted Personal Interview Questionnaire and Interviewer Instructions ...... 102 D. Data Entry Code Sheets for Dietary Instructions and Drug Categories ............................. 107 E. Phase 2 Medical Outcomes Study SF—36 Survey ..... 110 F. 1—day Food and Medication Record Instructions and Form ........................... 115 G. Consent Forms ................................... 118 H. Drug Container Labels Used for the Phase 2 Reading Test .................................... 120 I. Anecdotal Information/Interviewer Observations and Over—the—counter Drug Use for Selected Respondents by ID ............................... 121 LIST OF REFERENCES ......................................... 130 xi 10. 11. 12. 13. 14. 15. 16. LIST OF TABLES Sample Characteristics Comparisons of Phase 1 and Phase 2 ........................................... 49 Prescription (Rx) Drug Use Characteristics ............ 53 Prescription Drug Use by Therapeutic Category ......... 53 Relationship of Phase 2 Demographic Variables to Drug Therapeutic Categories Used ...................... 54 Length of Time of Prescription Drug Use ................ 55 No Indicated Dietary Instructions Given by Drug Therapeutic Category .................................. 56 Dietary Instructions Indicated and Given by Drug- food or Drug-nutrient Interaction (DFNI) Category ..... 58 Medical Outcomes Study SF-36 Health Survey Physical Component Summary (PCS) Scores and Norm Scores for U.S. Population of Equivalent Age and Gender .......... 60 Medical Outcomes Study SF-36 Health Survey Mental Component Summary (MCS) Scores and Norm Scores for U.S. Population of Equivalent Age and Gender .......... 61 Relationship of Selected Variables to Medical Outcomes Study SF-36 Health Survey Summary Scores ..... 61 Dietary Instruction Delivery Forms by Potential Drug-food Drug—Nutrient Interaction (DFNI) Category...62 Sources of Dietary Instruction Other Than Labels ...... 64 Dietary Instruction Formats Used by Health Professionals ............................... 64 Respondents' Beliefs About Dietary Instructions Given for Each Drug ................................. 67 Side Effects Reported by Respondents That They Attributed to One of Their Drugs ...................... 7O Adherence to Diabetic Meal Plan Instructions Compared to Adherence to All Other Dietary Instructions .......................................... 71 xii l7. l8. 19. 20. 21. 22. Adherence to Dietary Instructions Given for Hyperglycemia Agents Compared to Adherence to Dietary Instructions Given for All Other Drugs ....... Relationship of Adherence to Selected Variables ...... Adherence by Source of Oral Dietary Instruction ...... Parameter Estimates for Logistic Regression Model.... Comparison of Selected Demographic Characteristics of Phase 1 Respondents and Michigan Needs Assessment of the 60 and Over Population Respondents ............ Comparison of Study Respondents to All Women in the Apartment Complex by Three Key Characteristics... xiii .72 .74 .75 .75 .90 .93 LIST OF FIGURES 1. Data Collection Phases ................................ 37 2. Cross-classification Format for Statistical Tests in the Results and Discussion Section ................. 50 xiv INTRODUCTION STATEMENT OF THE PROBLEM The United States has been characterized as a nation devoted to youth and youthful attitudes. Although youthful attitudes may prevail, U.S. age demographics indicate that 12% of the population is aged 65 years and over. Today, women over age 65 outnumber men in their age group 3 to 2. The percentage of older adults is steadily increasing and is expected to continue to rise over the next decades as post-World War II and Vietnam War era "baby boomers" reach senior citizen status. By 2025, older Americans (those over age 65) should represent almost 20% of the population (House of Representatives 1987). Personal independence and optimal quality of life are key objectives for aging Americans and their health care system (Institute of Medicine 1991; Carruth and Boss 1990; Department of Health and Human Services 1990). Until recently, scant attention has been given to the nutrition status and nutrition-related needs of older adults 2 in the United States. A substantial proportion of older Americans have dietary intakes or diseases which place them at high risk of malnutrition. Among the risk factors that predetermine poor nutritional status is prescription drug use (Dwyer 1991). Prescription drugs can influence nutritional status directly through drug-nutrient or drug-food interactions; indirectly through side effects such as appetite change, fatigue or confusion. Conversely, nutrients and foods can alter drug effects and efficacy. Identifying individuals at risk for such interactions provides opportunities for implementation of preventative measures before clinical manifestations of problems are evident and reduces dysfunction, disability, decreased quality of life and, in some cases, morbidity and mortality. The approach for risk screening advocated by the Nutrition Screening Initiative is multi-factorial, emphasizing medical, social, economic and lifestyle factors that are equally important in determining risk (Dwyer 1991). An understanding of the factors which predict adherence to recommended prescription drug dietary modifications is difficult to achieve. Many factors may be involved in producing the desired adherence outcome. It is clear that individuals who never receive instructions for dietary 3 modification cannot adhere. For those who do receive instructions, the role of environmental factors potentially impinging on individual adherence behavior is also unclear, but clues to factors involved in the adherence of older adults as a group have been presented and are reviewed in a later section. If the probability of adherence can be determined by a manageable number of variables, such variables could help identify major targets of intervention and strategies for outcome enhancement (Glanz and Eriksen 1993). For example, if self-efficacy is a variable in a predictive model, group discussions or written brochures sharing the successes of others in adherence and offering their tips to make adherence easier could provide an improvement over didactic instruction. Other variables, like multiple medication use, could target individuals who will probably need special instructional attention. The use of a prediction model can improve allocation of intervention resources since all patients may not need the same level of intervention to achieve an outcome (Fedder 1982). Its development could also help fill a gap in research identified by the Nutrition Screening Initiative: core indicators for precise functional assessment of eating—related behaviors of older people (Dwyer 1991). 4 RESEARCH QUESTIONS, OBJECTIVES AND HYPOTHESES This study addressed the following research questions and sought to accept the accompanying hypotheses: Research Queetien I: When drugs are prescribed for community dwelling older women, are patients instructed to follow indicated dietary modifications to optimize drug effectiveness or minimize adverse drug effects, including known potential drug-food or drug-nutrient interactions? Objective I: To determine whether respondents received recommended dietary instructions indicated for the prescription drugs they take and to describe the sources and methods of instruction. Hypothesis I: In the majority of cases (51% or more), respondents did not receive instruction for all indicated dietary modifications. Reeeereh_Queetien_IIz When older women receive dietary instructions, do they adhere? Objective II: To determine whether respondents adhere to dietary instructions. 5 Hypothesis II: Respondents adhere to a majority (51%) of dietary instructions. Re r h e ' II: What factors predict adherence of older women to dietary instructions recommended for the prescription medications they take? Objective III: To identify variables related to dietary instruction adherence and formulate a prediction model for adherence. Hypothesis III: A regression equation can be constructed which will predict dietary instruction adherence. Review of literature suggests the following domains which may predict adherence: 1) knowledge of the modification's purpose 2) knowledge of consequences of nonadherence 3) understanding of instructions 4) instruction methods 5) instruction sources 6) Medical Outcomes Study SF-36 health status survey scores: Physical Component Summary (PCS) and Mental Component Summary (MCS) self-efficacy number of medications duration of medication use assistance with medication regimen attitude toward medication demographic variables significantly related to adherence REVIEW OF LITERATURE Preseriptien Drug Use As increasing numbers of people live to old age, the proportion of the population subject to the discomforts of chronic and acute diseases common to aging also increases. To reduce discomfort and the progress of disease, the use of prescription and non-prescription drugs is widespread. It has been estimated that individuals over 65 years of age consume 30% of the total medications prescribed in the U.S. and the average older person has three diseases treated by an average of eight prescription and/or over-the-counter medications (Anderson 1990). Epidemiological studies have shown that among older adults prescription drugs were used by 60 to 68% of men and 68 to 78% of women; five or more prescription medications were used by 4.6 to 11.8% of men and 6.5 to 13.9% of women (Chrischilles et al. 1992; Magaziner et al. 1989); a significantly greater number of drug classes were used by women than men (Hale et al. 1979). This heavy use of drugs is accompanied by a high rate of 8 drug-related hospitalizations among older adults. Drug— related problems may contribute to up to one third of hospital and one half of nursing home admissions of older adults (Cooper 1990; Colt and Shapiro 1989). In a Canadian study, older women had significantly more drug-related adverse patient events than men (Grymonpre et al. 1988). Patients aged 65 years and older account for 10 to 28% of hospital admissions for drug-induced symptoms; a situation which does not seem to be improving (Hallas et a1. 1990; Col, Fanale, and Kronholm 1990; Grymonpre et a1. 1988; Lamy 1988; Ives, Bentz, and Gwyther 1987; Nygaard et al. 1986). Stewart et al. (1991) conducted a longitudinal study of participants in a Florida geriatric health screening program and determined that the mean number of prescribed and nonprescribed drugs used per participant increased by more than 40% over a ten-year period. It is clear that drug-related problems increase with age, probably due to heavier use of multiple drugs, greater frequency of illness and altered physiology (Michigan Governor's Task Force 1991; Denham 1990; Hale et al. 1979; Williamson and Chopin 1980). 9 Phyeielegieal Changes in.Aging Physiological changes related to aging, although highly individual (Lamy 1982a), can have a direct effect on the efficacy of a drug or its side effects. Aging produces changes in pharmacokinetics that may cause a specific dose of a drug to lead to a higher steady-state concentration in an older adult than a younger person (Blumberg 1985; Schmucker 1984; Greenblatt, Sellers and Shader 1982). It also produces changes in pharmacodynamics that may lead to modifications in drug response. Pharmacokinetics The major effect of aging on pharmacokinetics occurs in excretion or drug clearance. There is an overall decline of about 35% in glomerular filtration rate between the ages of 20 and 90 years, although there is considerable individual variability (Rowe, Andres, and Tobin 1976). There are similar declines in effective renal plasma flow and tubular excretory capacity (Davies and Schock 1950). Since most drugs and their metabolites are eliminated via urinary excretion, reduced renal function produces slowed drug clearance and potential drug accumulation. Many of the conditions exhibited by older adults, including heart disease, are treated by the use of drugs with narrow therapeutic indexes which may be toxic at high serum 10 concentrations (Tregaskis and Stevenson 1990; Lamy 1982a). Clearance is complicated further when specific drugs are cleared through the liver, particularly via Phase 1 reactions. Substantial research over the past two decades has demonstrated that most medications metabolized through the hepatic mixed oxidase system tend to be metabolized more slowly with age due to reduced induction of enzymes or decreased affinity of enzymes for substrate (Wynne et al. 1988; Kamataki, Maeda, and Shimada 1985; Vestal 1978; Salem et a1. 1978; Greenblatt, Allen, and Harmatz 1975; Whittaker and Evans 1970). In addition, a decline in liver size and blood flow may also contribute to reduced biotransformation capacity (Wynne et al. 1988; Kitani 1986; Calloway, Foley, and Lagerbloom 1965). These changes in hepatic function can cause a decrease in the metabolic clearance of drugs and a resulting high serum level. Changes in body composition with aging, beyond the liver and kidney, include declines in total body size, total body water and lean body mass, accompanied by an increase in body fat stores (Tregaskis and Stevenson 1990). Such changes may also lead to changes in the volumes of distribution of highly lipid or water soluble drugs. Volume of distribution is an essential element in the plasma half life of a drug 11 (Tregaskis and Stevenson 1990) and the concentration of drug per unit volume (Lamy 1982a). Age-related declines in serum albumin levels have been associated with reduced plasma protein binding of such drugs as antipyrine, meperidine, diazepam, phenytoin, propranolol, warfarin and salicylates (Tregaskis and Stevenson 1990; Cohen 1986). Increased free drug fractions are seen since total plasma clearance is negatively correlated with serum albumin for many highly bound, lipophilic drugs. Conversely, some drugs which bind preferentially to alpha—1- glycoproteins, which increase with aging and under stress conditions, show increased plasma protein binding (Tregaskis and Stevenson 1990; Mitenko 1986; Lamy 1982a). Other drugs show no change in binding at all in the elderly (Wallace and Verbeeck 1987). A few studies have examined gastrointestinal changes related to aging and the absorption and bioavailability of drugs, but have not shown a direct relationship (Mitenko 1986; Rikans 1986; Rubin, Scott, and Reid 1981; Cusack et al. 1979; Fulton, James, and Rawlins 1979). Studies of the effects of age on drug absorption are very limited, probably due to methodological problems. Available information has primarily come from simple comparison of oral and 12 intravenous plasma drug concentration against time curves (Tregaskis and Stevenson 1990). Pharmacodynamics There is evidence that, beyond alterations in pharmacokinetics, age—associated changes in homeostatic mechanisms and sensitivities of specific receptors and target organs can modify drug responsiveness in older adults (Blumberg and Suter 1991). Age appears to produce a gradual reduction in the ability to adapt to changes in the environment, including drug use. Important homeostatic mechanisms which demonstrate reduced vitality include postural control, orthostatic circulatory responses, thermoregulation, visceral muscle function and higher cognitive function (Blumberg and Suter 1991; Swift 1990). Pharmacologic challenge to these mechanisms in an older person can produce altered and possibly undesired drug effects. At the site of drug action, age-related changes in drug receptors or in the intrinsic properties of the target tissue can result in altered drug effects, particularly when responses of younger and older adults are compared (Rikans 1986). Although in vivo research with humans is rarely possible, recent work has been done in vitro and through 13 animal studies to describe age-associated receptor declines in density (numbers), binding sites, binding affinity and transduction of signals (Blumberg and Suter 1991; Swift 1990; Rikans 1986). Adverse Effeets of Prescription Drug Use The adverse effects of prescription drug use are difficult to identify, categorize or quantify. Various terms and definitions have been used. For reporting, the Food and Drug Administration (1985) defines an adverse drug experience as: any adverse event associated with the use of a drug in humans, whether or not considered drug related, including the following: an adverse event occurring in the course of the use of a drug product in professional practice; an adverse event occurring from drug overdose, whether accidental or intentional; an adverse event occurring from drug abuse; an adverse event occurring from drug withdrawal; and any significant failure of expected pharmacological action. This definition is broad and open to interpretation. Kramer et al. (1979) suggested that clinical identification of an adverse drug reaction (ADR) is a "nonreproducible act of unspecified subjective judgment.” It is difficult then to sort out drug-age, drug-drug, drug— nutrition, drug-food, drug-allergy, drug-regimen and drug- disease effects (Institute of Medicine 1991; German and Burton 1989). Few drugs produce distinctive adverse effect 14 signs. If a drug is known to have an association with a particular reaction, it is hard to prove that it is the cause in a particular patient short of unethically challenging the patient with the drug again. Even laboratory tests can seldom differentiate between an adverse drug reaction and underlying disease (Denham 1990). In addition, older adults and health care professionals may be slow to identify transient symptoms and signs related to medications (German and Burton 1989). British investigators have estimated that three fourths of adverse drug reactions (ADRs), which are carefully and formally tracked through the British National Health System (NHS), are simply an exaggeration of the normal drug response. Even though ADRs do not always result in a hospital admission, it has been suggested that their cost in emotional terms to the patient and the financial cost to the NHS is considerable (Denham 1990). Elderly individuals often manifest their medical problems in vague and nonspecific symptoms that are hard to interpret (Denham 1990; Ouslander 1981; Hodkinson 1973) and are all too often simply attributed to "getting older" (Blumberg and Suter 1991; American Medical Association Council on Scientific Affairs 1990; German and Klein, 1984). 15 Complaints like malaise, memory loss, confusion, anorexia, bone pain, changed sleep patterns, gastrointestinal upset and incontinence are frequently considered inevitable signs of old age (Lamy 1982a). Klein et a1. (1984) found that 30% of elderly outpatients taking medications reported adverse reactions to at least one drug and established that the most frequently reported were urinary symptoms, head or chest discomfort, mood/sleep disturbance, balance problems and gastrointestinal upset. Although severe drug reactions are very likely to receive attention, small non-life threatening decrements in functioning may go unnoticed or be considered unimportant. Small decrements may however cause ripple effects, diminishing physical or social functioning (German and Klein 1984) and lowering the quality of life. The Nutrition Screening Initiative has reported that particular attention needs to be paid to the effect of health interventions on function and performance rather than solely on morbidity and mortality (Dwyer 1991). I] E 1 E I' I H T'E' !' n . : !' . . E 'v n M n'm' in A v D n Diet composition and meal pattern can influence drug efficacy and effects through chemical interactions and a 16 wide range of physiologically mediated interactions at many stages of drug action. The type, combination, amount and timing of foods consumed can alter the dissolution, absorption, distribution, metabolism and excretion of many drugs (MacLeod and Soldin 1986; Hathcock 1985; Roe 1985; Smith and Bidlack 1984; Lamy 1982b). Since altered pharmacokinetics and pharmacodynamics may already put older adults at risk for reduced drug efficacy or adverse effects, risk enhancement or risk reduction inherent in dietary practices is an important consideration in their drug regimens. This is particularly true since most information about drugs is obtained through drug testing done on younger adults (McAllister 1986; Roe 1985). Although much research is still needed on the specific effects of prescription drugs on older adults, a considerable amount is already known about the potential interplay of medications and dietary components and reference materials are readily available (Physicians' Desk Reference 1995; United States Pharmacopeia Dispensing Information 1995; Pronsky 1993; Roe 1992). Since response to prescription drugs and age-associated physiological Changes are both highly individual, the absolute need for precise dietary modifications to avoid serious alterations 17 in drug effects is difficult to predict but the need for risk reduction in the case of older adults is definite. One of the health behaviors cited by the Nutrition Screening Initiative as important to malnutrition prevention is appropriate use of medications, especially those with nutritional effects, and adherence to medication schedules (Dwyer 1991). In institutional settings, two studies have examined indications for dietary instructions in the prescription drug regimens of adults and subsequent adherence as assessed by examination of drug administration data on medical charts. Lewis, Frongillo and Roe (1995) studied men and women in three long—term care facilities. Most of the subjects were older; mean ages of subjects in the three facilities were 86.5 years, 91.3 years and 84.5 years. Charts were audited and data collected on fifty-three patients from each of the facilities over a period of six months. A computerized algorithm was developed to assess the risk for eight classes of potential drug—nutrient interactions (DNIs). Patients were at risk for a mean of 1.43, 2.69 and 1.43 potential DNIs in each of the facilities, respectively. The most commonly observed DNIs were gastrointestinal interactions affecting drug bioavailability and interactions affecting electrolyte I! 18 status. Although the researchers had planned to measure the rate of actual DNIs, they could not do so because of the infrequent availability of laboratory data and sparse documentation of patients' conditions in the records. In the other study, Strong et a1. (1991) investigated adherence to indicated dietary instructions which specified when each of five cardiovascular drugs should be taken in relation to mealtimes. Subjects were adults of various ages, primarily older, at two short—term and two long-term care facilities. Data obtained from medical records revealed that 85 to 93% of 183 patients received one or more drug doses incorrectly in relation to mealtimes during the study period. The researcher concluded that arbitrary medication schedules were used and that there is a need for schedules to be designed with care to achieve the greatest drug bioavailability. E . I' L E . Prescription drug regimen adherence is an important issue since as many as half of prescribed drugs fail to have their desired effects because they have been used improperly, most often because of poor communication to the patient or caregiver. This is especially common when the patient is elderly (Michigan Governor's Task Force 1991; American 19 Medical Association Council on Scientific Affairs 1990). One of the more serious adherence problems encountered is the practice of many older adults of simply discontinuing a medication when unpleasant side effects occur (Clark 1991; Schoenberger et al. 1990; Spagnoli et a1. 1989; Darnell et al. 1986; Ostrom et al. 1985). As part of a prescription drug regimen, dietary modifications are probably subject to many of the same adherence perils as are other regimen components. Although older adults' adherence to recommended dietary modifications has not been studied specifically, overall prescription drug regimen adherence has been documented in limited population studies. Also, research unrelated to drug regimens has been done to identify factors associated with adherence of older adults to recommended health promoting dietary behavior changes and may be relevant to drug regimens as well. Prescription Drug Regimen Adherence Prescription drug regimen adherence (or compliance) studies have taken many different approaches but have universally suffered from any agreed upon standard for defining adherence (German and Burton 1989). Within individual studies comparing community dwelling older and younger adults who administer their own medications, drug regimen 20 adherence based on a single defined standard is not greatly different for the two groups, ranging from 52% to 69% (German et al. 1982; German and Klein 1984). Darnell et al. (1986) found that there was no correlation of adherence with age per se in their study of medication use by ambulatory elderly, although there was correlation with the number of medications taken. This may have accounted for the fact that men adhered more frequently than women, since the mean number of drugs in this study per man was 3.3 compared to 5.0 per woman. For older adults, factors that contribute to nonadherence include polypharmacy (German and Klein 1984; Inui et al. 1980; Fletcher et al. 1979), complex drug taking schedules (Hulka et al. 1976), side effects (Clark 1991; Inui et al. 1980; Darnell et al. 1986), patient functional limitations (Darnell et al. 1986) or taking specific drug classifications like central nervous system drugs, antibiotics, antihistamines and gastrointestinals (Hulka et al. 1976). Unfortunately, the degree of correlation of such factors to nonadherence is often clouded by the use of statistical methods that do not control for correlations among the factors themselves (Spagnoli et al. 1989). Factors reportedly involved in adherence to prescription 21 drug regimens are the knowledge of medication purpose (Helling et al. 1987; Klein, German, and McPhee 1982), knowledge of consequences of nonadherence (German and Burton 1989), patient's understanding of the regimen (Col, Fanale, and Kronholm 1990; Spagnoli et al 1989), patient's attitude toward the medication (Darnell et al. 1986; Inui et al. 1980), patient's level of education (German et al. 1982), source of prescription (ie. general practitioner vs. specialist) (Spagnoli et al. 1989) and regimen counseling and/or memory aids (Michigan Governor's Task Force 1991; Gilchrist, Lee, and Tam 1987; Macdonald, Macdonald, and Phoenix 1977). Since patient knowledge of drug information is a factor in adherence, it follows that sources of information may also be important. In a random sample of 1100 adults with new prescriptions, Morris et al. (1984) discovered that 70% had received some information about the drug; 95% of those receiving information said that it came from a physician. According to the Iowa 65+ Rural Health Study (Semla et al. 1991), however, 3 percent of all drug purposes reported by participants in that study were inappropriate. When participants cited appropriate purposes, there was no association to whether the medication had been dispensed by a pharmacy or a physician. Interestingly, the highest 22 percentage of appropriate purpose responses were for drugs dispensed by mail-order pharmacies. An earlier report from the ongoing Iowa study (Helling et al. 1987) cited mail order as a matter of concern since communication between the pharmacist and patient is limited. Studies have also examined directions given on prescription drug containers. In the Iowa report, the most frequently noted directions were for scheduled daily dosing (>75%), followed in frequency by "as directed/as needed" (8%) and nonspecific directions such as "take with meals"(5.2%). Lundin (1978) found that 25% of 50 people over 60 years of age in a community—based study took medications differently than suggested on container labels, usually because they said their physicians had given them oral instructions to do so. In a study of elderly (mean age of 71.6 years) residents of an urban subsidized apartment building, Darnell et al. (1986) found that standard pharmacy container labels could be read by 79.7% of those surveyed, but inability to read labels did not measurably affect adherence. There have been recent attempts to expand the sources of information offered to prescription drug users. In England, 3410 adults and 254 pharmacies participated in a national mail survey of the effect of prescription information 23 leaflets (Gibbs, Waters, and George 1990). The 1809 people who received leaflets knew more about their medicines, especially the side effects, and were significantly more satisfied than the 1601 patients who had not been given additional written information. Ninety—seven percent of participants thought it was a good idea to be given an information leaflet with their prescriptions whether they had been given one or not in the study. Other studies have reported actual improvements in prescription drug regimen adherence when patient educational aids such as specific counseling (Hawe and Higgins 1990; Gilchrist, Lee, and Tam 1987; Hammarlund, Ostrom, and Kethley 1985; Macdonald, Macdonald, and Phoenix 1977), drug guides (Ross 1991) and drug calendars (Wandless and Davie 1977) were used. German and Burton (1989) listed the following factors which were considered in successful prescription drug information and education strategies: information/education topic relevance to a patient's need; appropriate degree of individualization of the teaching process; use of feedback and reinforcement; and facilitation of action by the patient. 24 Changing Health Promoting Dietary Behavior of Older Adults Since changing behavior in any context is a multi—factorial issue, the concept of changing health promoting behavior has been addressed by many researchers with the subsequent publication of many theories about how such change, including dietary change, might be most effectively accomplished (Glanz and Eriksen 1993). Using meta-analysis of nutrition education research, Johnson and Johnson (1985) demonstrated that there is strong evidence that nutrition education promotes positive changes in dietary behavior. More recent studies have also reported successful nutrition education interventions to achieve health—related dietary modifications (Gans et al. 1990; Hackman and Wagner 1990; Hermann et al. 1990; Williams, Kim, and McMullen 1987), although no studies have been directed specifically to dietary modifications recommended for prescription drug use. Whether reported dietary behavior changes are the sole result of increased nutrition knowledge has been questioned, since meta—analysis has also shown that the theoretical relationship has a small effect size and does not appear to be strong (Axelson, Federline, and Brinberg 1985). Confidence in the relationship of results of tests of nutrition knowledge and dietary behavior change requires confidence in the validity of the measures. Knowledge '1 25 measures must reflect the information that is required by the individual to exhibit the dietary behaviors of interest. This confidence may exist in intervention research projects, although meta-analysis shows that it most often does not (Axelson and Brinberg 1992). It would be difficult to achieve in retrospective research of prescription drug dietary modifications which require specific nutrition knowledge and involve instruction for widely different regimens from varied sources, in differing formats, through different methods. A single knowledge measure could not be used to predict adherence to a wide variety of recommended dietary modifications. In studies examining predictors of other health promoting behaviors, including weight loss, self—efficacy is frequently cited as an important predictor (Bernier and Avard 1986; Strecher et al. 1986; O'Leary 1985; Jeffrey et al. 1984; Weinberg et al. 1984). The value of perceived self-efficacy as a factor in predicting medical recommendations adherence behavior has also been demonstrated (Meichenbaum and Turk 1987; Bandura 1986; Strecher et al. 1986). Self-efficacy theory proposes that a person's perception of capability to perform a task and the likely outcome of such action can affect behavior (Stanley and Maddux 1986; Bandura 1977). 26 The factors influencing self—efficacy have never been clearly determined, but Matheson et al. (1991) developed a theoretical model to predict the self—efficacy of older adults toward nutrition behaviors. Self-efficacy toward nutrition behavior was defined as people's judgments of their capabilities to organize and execute behaviors to enhance the quality of their diets. '( METHODS Approvel to Conduct Study Approval to conduct this study was granted in June 1994 by the Michigan State University Committee on Research Involving Human Subjects (UCRIHS) prior to administration of surveys or collection of data (Appendix A). S u P 1 ion Female residents of an East Lansing, Michigan, subsidized senior apartment complex (total = 154) comprised the study population. Residents' 1994 incomes were limited to two levels under HUD Section 236: $11,500 or less for subsidized apartments; $11,500 to $29,650 for sliding scale rental apartments. Income was calculated by the apartment manager according to HUD Section 236 guidelines, which provide formulas for deducting Medicare, health insurance and prescription drug expenses from gross income. Based on the calculated income figure, residents were classified at one of the two income levels. The population was controlled for: gender, geographic location, cooking facilities, public transportation, availability of in-house meal service and 27 28 availability of health promotion programs. Measures Data collection was done in two phases, as illustrated later in Figure 1. Separate measures were developed and used for each phase. Phase 1 Self-administered Questionnaire Prescription drug use screening and baseline demographics information were obtained through a self-administered large- type (6 characters per inch) questionnaire (Appendix B). The questionnaire comprised prescription drug use screening questions and demographics questions taken from the Iowa and Washington Counties baseline questionnaire of the National Institute on Aging Established Populations for Epidemiologic Studies of the Elderly (EPESE) (Cornoni-Huntley et al. 1986). The questionnaire also directed respondents taking prescription drugs to list those by name along with the name of the prescribing doctor on a simple two-column form at the end of the questionnaire. An additional demographics question was developed to determine whether the respondent's apartment was subsidized or sliding scale as a proxy for income. Income proxy was used since two levels of income were adequate for this study 29 and efforts to determine specific income information from older adults are often not successful (Wallace 1991; Gibson and Aitkenhead 1983). Previous studies have shown no relationship between medication use and income (Chrischilles et al. 1992; Nolan and O’Malley 1988), so it is unlikely that there is a relationship with dietary modification. In the EPESE study care was taken to assure reporting of income, offering dollar income range categories to address this sensitive issue, but "refused/don't know/missing" responses were still high (Cornoni-Huntley et al. 1986). Questions about income may also tend to prejudice respondents against further participation in a study for fear that too many personal details will be exposed (Wallace 1991; Gibson and Aitkenhead 1983). Landry et al. (1988) demonstrated that the self-administered mail questionnaire, which is analogous to the Phase 1 self— administered questionnaire, can be a valid source of information about prescription drugs. They compared data obtained from community dwelling older adults, aged 65 and over, by three methods: in—home examination of drug containers and a personal interview; telephone interview; and mail questionnaire. When overall percent agreement rates were calculated for each person comparing in-home data to the mail questionnaire data, the median and modal percent 3O agreement rate was 100%. Respondents with less than 100% agreement on prescription drug use by mail and in—home assessment were older and lower on two functional status measures, Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL). The questionnaire was submitted to a panel of experts for review. They each provided one or more of the following areas of expertise: survey research, data entry and analysis, nutrition and food consumption of older adults, prescription drugs and drug regimens, demographics, and epidemiology. A focus group evaluation of the questionnaire along with its introduction/instructions letter, which incorporated informed consent for Phase 1, was conducted with seven women volunteers from a nearby town's senior citizen center. The women evaluated the questionnaire for appropriate language, instruction clarity, question sensitivities, response alternatives, sources of confusion, difficulty/time burden, and format/type size. They commented on the questionnaire as a whole and on individual questions. The revised questionnaire was submitted to the same women who participated in the focus group to complete it as a 31 pilot. There were no additional revisions necessary before the questionnaire was distributed in Phase 1. There were several comments noting the ease of reading the questionnaire from respondents who actually completed it during Phase 1. Phase 2 Computer-assisted Personal Interview The basic computer-assisted personal interview questionnaire (Appendix C) comprised the following: 0 drug and drug container information sections filled in by the interviewer which were adapted from the EPESE studies' forms discussed earlier 0 questions addressed to respondents about hypothesized variables which were adapted from similar questions used in the EPESE studies 0 specific questions addressed to respondents about dietary instructions which were developed for the interview 0 self—efficacy toward dietary instruction questions which were addressed to respondents and adapted from the alternate measures of self-efficacy described by Lust, Celuch, and Showers (1993), based on the concepts of Bandura (1986) and Ozer and Bandura (1990) ('1 (I) 32 The personal interview computer questionnaire was submitted for review to the same group of experts and the same focus group who reviewed the Phase 1 self-administered questionnaire. It was revised initially based on their comments. It was then converted into a format suitable for direct data into Epi Info software and adjusted so that computer monitor screen views allowed smooth question transitions, legible interviewer instructions, and clear sight of legal code values used for responses. The focus group members and two members of the apartment complex resident advisory council, who did not meet the inclusion criteria for Phase 2, were respondents in a pilot of the entire computer—assisted personal interview session. The session included administration of the health status survey, review of the 1-day diet and medication record, and review of the consent forms, all discussed in following subsections. The pilot respondents provided feedback to the interviewer about individual questions and the personal interview questionnaire as a whole, considering appropriateness of language, clarity, question sensitivities, response alternatives, and sources of confusion. The interview pilot was also used to assess the feasibility ,1 33 of data collection procedures and convenience and flexibility of the procedures for coding and data entry (Aday 1989). Phase 2 respondents said that the computer entry during the actual interview did not bother them, that they were used to seeing computers used. Phase 2 Medical Outcomes Study 36-Item Short Form Health Survey Health status was assessed during the Phase 2 personal interview session with the self—administered Medical Outcomes Study 36—Item Short Form (MOS SF-36) Health Survey (Ware et al. 1993) (Appendix E) which has been extensively researched and validated (McHorney et al. 1992), including its use with older adults (Weinberger et al. 1991). Norms for MOS SF-36 summary scores have been established for various age groups based on representative national samples (Ware, Kosinski, and Keller 1994). Pretesting of the MOS SF-36 survey was part of the pilot of the entire computer- assisted personal interview session. Phase 2 1-day Food and Medication Record A 1-day, non-quantitative, large—type (6 characters per inch) food and medication record with instructions for completion (Appendix F) was developed to assess dietary adherence. Since most dietary modifications were to be 34 followed on a daily basis and involved absolute inclusion or exclusion of specific foods or modified eating patterns (Pronsky 1993), a 1-day food record could assess adherence without concern for inadequate reflection of "usual" intake (Hankin 1989). The instructions did, however, specify that the record be kept for a weekday to try to avoid unusual complications which special occasions might impose on food and beverage consumption and drug regimens. The short record form with instructions addressed information of interest in assessing adherence to dietary instructions. It was pretested during the pilot computer- assisted personal interview session. Pilot respondents were provided with copies of the instructions and form 2 to 4 days in advance. They were asked to complete the record according to the instructions and bring it to the personal interview. None of the respondents reported having trouble following the instructions and completed the record properly, although minor revisions were made to the record instructions as a result of their comments offered for improvements. Phase 2 Consent Forms Two consent forms (Appendix G) were developed for Phase 2 and pretested by the focus group and in the pilot personal () 93 P} FIE 35 interview session. One was for informed consent; the other for the respondent's permission to contact her physician(s) if that was necessary. No revisions were needed. Phase 2 Drug Container Labels Reading Test To assess the ability of respondents to read the information on drug container labels, a reading test was developed using copies of two actual drug container labels (Appendix H)--one with black letters on a white background and another with black letters on a red background. The reading test was added to the end of the personal interview questionaire and pretested along it. Phase 2 Assessment of Respondent Cognitive Ability Respondent cognitive reliability was assessed after the personal interview session by consistency of answers to duplicate questions, count of "don't know/remember" answers per respondent and by interviewer assessment of confusion or memory problems and level of confidence in information received (Wallace 1991; Gibson and Aitkenhead 1983). All respondents were considered cognitively reliable. This assessment was consistent with the independent living capability required for apartment dwellers in the complex. 36 Da ll ion Data were collected in two phases, illustrated in Figure 1. Phase 1 In cooperation with the resident advisory council, the questionnaires were delivered to women residents along with instructions to return completed questionnaires to the Resident Manager's office by a specified day. Although an incentive had been considered for questionnaire return, the council advised against it in the belief that incentives had been ineffective in previous surveys and that residents participated in such activities for the reward of helping other people. Communications to residents about the study were designed with this participation motive in mind. The questionnaire included a cover letter explaining the purpose of the study and advising that the results would provide confidential background information for a later study. No reference was made to diet or drug regimens. Announcements concerning the survey were made by the resident council president at in-house meals during the week the questionnaire was distributed and the week the completed 37 PHASE 1 61 Subjects Self—administered Questionnaire Demographics Prescription Drug Screening Inclusion criteria not met or interview not completed End I > 32 Subjects PHASE 2 29 Subjects Computer—assisted Personal Interview Quantitative Data on Hypothesized Variables Qualitative Data Diet and Medication Record Figure 1: Data Collection Phases 38 Sixty-one women returned completed surveys by the original return due date, a response rate of 39.6%. Reminder notices were delivered to each apartment two weeks after the original return due date, but no additional surveys were returned. Residents who returned questionnaires with "Don't Know" or blank answers were contacted by phone or in person to attempt to determine the correct response. A microcomputer database form was prepared so that the data from the self—administered questionnaire could be entered into Epi Info (1994), Version 6: a word processing, database and statistics software program for epidemiology on microcomputers. A rigorous checking system was programmed for the database entry to provide error checking, mathematical operations, rejection of illegal response values, automatic skip/jump patterns, and autosearching for duplication of data which should have been unique. All data were entered a second time by another trained person and the two data files were processed through the software's data entry validation feature. Inconsistent entries were referenced to the written response and corrected. Prescription drug information from the self—administered survey was also entered into PharmAssist software (1993) to 39 estimate the number of potential drug-drug interactions for each Phase 1 participant. Phase 2 Women who listed prescription drugs with indicated dietary instructions on their Phase 1 questionnaire and who met one or more of the following inclusion criteria comprised the population for this phase. The inclusion criteria were: 0 Took one or more of the most frequently listed drugs 0 Took one or more drugs in drug categories listed seven or more times in the Phase 1 responses (representing more than 10% of respondents) O Took drugs with serious consequences for nonadherence to dietary regimen Women meeting these criteria were contacted by telephone and asked to participate in an interviewer-administered computer-assisted personal interview. Specific appointments were made for interviews to be conducted in the individual apartments or in an assigned room in the apartment complex. All but one participant chose to be interviewed in their apartments. Many indicated that they found it difficult to get around the building and an interview at home would be more convenient. 40 When appointments were made, participants were asked to complete the 1—day food and medication record prior to the interview. Confirming letters were delivered to participants' apartments along with food and medication record forms and instructions. Instructions noted that the records should include all prescription and nonprescription drugs, vitamin and/or mineral supplements, water and alcoholic beverages ingested on the day of record. Participants were asked to have their food and medication records available for the interview to minimize record— keeping bias that might result from the dietary modification questions in the interview. Since the average educational level of the complex's residents was quite high, almost 12 years, this group was likely to be able to keep food records without direct assistance (Elahi et al. 1983; Garry et al. 1982). On the telephone and in the confirming letter, participants were also asked to make all prescription drug containers and any written information they had about their drugs available at the time of the interview. Food records are one of the screening measures recommended for use with older adults by the Nutrition Screening Initiative (Dwyer 1991). The 24—hour food record is preferable to the 24—hour recall technique since information obtained from the latter is always subject to the ability to 41 accurately remember foods eaten in the short-term past. Despite a lack of convincing evidence, reviewers have noted that memory deficits may particularly influence 24-hour recall performance by older adults (Krall, Dwyer, and Coleman 1988; Dwyer, Krall, and Coleman 1987; Campbell and Dodds 1967). In a study of adults aged 60 years and over (Gersovitz, Madden, and Smiciklas-Wright 1978), the internal validity of a 24-hour recall was compared to that of a 7-day food record. Although the study was not directed to individual nutrient intake, results suggested that both methods provide about equally accurate estimates of the mean intake; the recall was subject to flat slope syndrome. The authors also found that 85% of the sample returned usable records for at least two days, but the percentage of usable records declined from the third to seventh days. They suggested that the 1—day food record may provide more accurate information than the 24-hour recall. Fanelli and Stevenhagen (1986) specifically compared data collected using interviewer administered 24—hour recalls and 1-day food records from 2,667 older adults who participated in the 1977—78 Nationwide Food Consumption Survey. Data were analyzed for energy and 14 nutrients. Both assessment 42 methods gave similar estimates of intakes for the groups of older adults, even though data were collected on different days, indicating a lack of variability in the diet from day to day. The interview questionnaire was converted to an Epi Info (1994) database relational file entry form. Programming was completed for the form which allowed conditional switching between three related form levels: the base information form, the drug information form and the dietary instruction form. The hierarchal form levels provided flexible data entry for respondents' records which varied considerably in number of drugs and number of dietary instructions indicated for each drug. A check system was programmed which provided for automatic error checking, mathematical and logical operations, rejection of illegal response values, automatic and conditional skip/jump patterns, autosearching for duplication of entries which should be unique, automatic coding, automatic entry under specified conditions, and pop- up windows to provide data entry coding information for the interviewer. Based on information reported by respondents in the initial Phase 1 questionnaire, the list of drugs for each respondent participating in a Phase 2 interview was entered by drug 43 name into Computerized Food-Medication Interactions (FMI) software (1993) to obtain indicated dietary instructions for each drug. Indicated dietary instructions were then entered and coded on each respondent's computer record prior to the interview, along with drug name, drug category code and name of prescribing physician. Dietary modifications listed as "may be needed" (ie. based on lab tests) were not included. FMI was used in preference to other resources since it provides information obtained from 72 authoritative references, including recent professional journals, in one handbook which is available to health professionals. It is edited by well—known experts in the field: J.P. Crowe, Pharm.D., R.Ph., Director of Pharmacy, Carmilla Hall, Immaculata PA; S. Epstein, M.D., F.R.C.P., F.A.C.P., Head, Endocrinology and Metabolism, Albert Einstein Medical Center and Professor of Medicine, Temple University; and C.H. Smith, Ph.D., R.D., Professor, The Marilyn Magaram Center for Food Science, Nutrition and Dietetics, California State University. When the interview was conducted, the interviewer followed an introductory, prepared script which was integrated into the interview questionnaire and appeared on the computer monitor screen. Prior to asking any questions, the 44 ithernriewer reviewed the general purpose of the study and obtained the informed consent of each participant. Permission to contact the participant's physician, if that shcnnhi be deemed advisable after analysis of the survey results, was requested at the end of interview. The iruxarviewer advised the participant that her responses to added other information from the drug container label whhjiwas required for the questionnaire. When the respmmbnt was finished with the health survey, the 45 interviewer conducted the personal interview, entering responses to questions directly into the computer program. At the end of the interview session, the interviewer reviewed the 1—day food and medication record and asked the respondent to clarify information as necessary. After the interview session was completed, the food and medication record was compared to each of the dietary instructions given and adherence to each instruction was determined. Adherence was scored yes (1) or no (0) and entered into the Epi Info database record. The MOS SF—36 survey responses were also entered into Epi Info by two separate individuals and the two record files were processed and reconciled with the original responses through the program's data entry validation feature. Information for both over—the-counter drugs (Appendix I) and prescription drugs was entered into PharmAssist software (1993) and the total number of potential drug—drug interactions estimated for each Phase 2 respondent. Of the 34 women residing in the complex who met the Phase 2 inclusion criteria, interviews were completed and diet records obtained for 29 of them. The length of the 46 interview sessions ranged from one to two hours. Of the five women who were not interviewed, one was no longer taking prescription drugs so did not qualify for inclusion, two had relocated to Florida, one was in the hospital, and one refused because of illness. The Phase 2 response rate was 87.9%. Stetieticel Analysis: Data were analyzed using Epi Info software (1994), Version 6.0, as noted previously, and SPSS/PC+ (1990) software, Version 4.0. Results were considered significant at the d=.05 level. Subjects in Phases 1 and 2 were compared to determine if there were any significant differences in their demographic characteristics. Comparisons were done using anova, Satterthwaite's t-tests for samples with unequal variances, and chi square tests for proportions (Dean 1994). Frequencies were determined for all variables; associations were examined using chi square and Fisher exact tests. Further analysis of selected variables was completed to assess the magnitude of their contribution to probability of adherence as predicted by a best fit logistic regression model. 47 Ixogisfl:ic regression can be used effectively to identify a nmxiel.tdhich describes the relationship between predictor or independent variables and a dichotomous outcome or dependent variable; flutexmmfle, adherence or nonadherence (Hosmer and Lemeshow 1989) . Logistic regression is the preferable rmiltivariate technique to predict a binary dependent 'variable from a set of independent variables. If the a! dependent variable can have only two values, the assumptions 2 necessary for hypothesis testing in multiple regression analysis are violated. Multiple regression analysis also does not allow predicted outcome values to be interpreted as probabilities since they are not constrained to fall in the interval between 0 and 1. While linear discriminant analysis allows direct prediction of outcome in terms of group membership, it requires the assumption of multivariate normality of the independent variables and equal variance-covariance matrices for its prediction rule to be optimal. Even when the assumptions ramured for discriminant analysis are satisfied, logistic regressbdn still performs well (Norusis 1990). In Hus study, independent variables which were sLyflficantly associated with adherence were entered shmfltaneously to obtain a fully saturated logistic 48 regression model. A subsequent forward stepwise entry Inetlnod. Unith a test for backward elimination) identified ‘varijfloles for inclusion or exclusion from the model based on the significance level of the Wald statistic. This prtxxaiure produced a statistically best fitted model with time fewest variables. The final model was compared to the sattunated model through goodness—of—fit statistics and by comparing'predictions to observed outcomes in a classification table (Hosmer and Lemeshow 1989). The same procedure was followed for other selected variables in domains suggested by the literature review. The logit (log of the odds of the outcome occurring) of the multiple logistic regression model is given by the equation: g(x)=BO+lel+Bzx2+...Bpxp. The logistic coefficient (B) can be interpreted as the change in the log odds associated with a one unit change in the predictor variable (x). The estimated probability of the outcome is 1 / 1 + e-gUU where e is the base of the natural logarithm 2.718). (approximately The probability of the outcome not occurring is 1 mhmnsthe probability of the outcome occurring. A pmobafllity of 0.5 or greater estimates that the outcome willcxmur (Hosmer and Lemeshow 1989). RESULTS AND DISCUSSION Statistical test results in this section appear either in tables or summary tables. The format for tables not in summary form is shown in Figure 2 on page 50. Sample Characteristics There were no significant differences in the demographic characteristics of the subjects in Phases 1 and 2 which are summarized in Table 1. Table 1: Sample Characteristics Comparisons of Phase 1 and Phase 2 Characteristic Phase 1 Phase 2 Test p—Value (n=61) (n=29) Mean Age 76.97 76.03 t—test .642 % 77 Yrs. or More 52.50 48.30 chi square .711 % Race = White 83.60 79.30 chi square .618 Mean Yrs. Education 11.75 11.31 t-test .427 % Not Married 90.20 89.70 chi square .810 % Income s$11,500* 55.70 62.10 chi square .570 *Gross income adjusted for Medicare, health insurance and medication expenses. 49 50 I B1 B2 1 _______ +___.__.___________+______ A1 | Cell Cell | R1 A1/B1 A1/B2 l Cell Cell | A2 A2/B1 A2/B2 R2 _______ +________________+_-____ l C1 C2 | Total [Chi square test statistic or Fisher's Exact test] [p—Value = ] Legend; A = Independent variable A1 and A2 = Independent Variable Categories U3 ll Dependent variable (potentially influenced by A) B1 and B2 = Dependent Variable Categories Cells = Counts of cases with both characteristics C1 and C2 = Column subtotals R1 and R2 = Row subtotals Statistic = Test statistic used for association p-Value = The probability of a result at least as large as that observed if the null hypothesis is true Figure 2: Cross-classification Format for Statistical Tests in the Results and Discussion Section 51 The mean age of subjects was 76.97 years; age range was 54 to 93 years. In Phase 1, 52.5% of respondents were 77 years of age or over; in Phase 2, 48.3%. All subjects tended to be currently unmarried (widowed, divorced or never married) and have an average of somewhat more than 11 years of education. Most were white: 83.6% of Phase 1 respondents, 79.3% of Phase 2. More than half of respondents had incomes equal to or less than $11,500. Maximum household income allowed for apartment residents was $29,650. All but one respondent had known access to health expense coverage: 57 were covered by Medicare, 7 by Medicaid, 43 had private health insurance coverage (some respondents had more than one form of coverage). WW Phase 1 Among Phase 1 respondents (n=61), 53 individuals were taking at least one prescription drug at the time they filled in the questionnaire. The mean number of drugs prescribed for respondents was 3.3. This rate is consistent with a range of mean drug use from 1.5 to 4.5 reported in other studies of community dwelling older women (Chrischilles et a1. 1992; Helling et al. 1987; Darnell et al. 1986). The mean number of potential drug-drug interactions estimated by PharmAssist software was 1.7 (Table 2). 52 An examination of the drug therapeutic categories represented in Phase 1 responses showed that the categories taken most frequently were Hypertension Agents (28.0%), Other Cardiovascular Agents (14.2%), and Arthritis Agents (11.0%) (Table 3). Individual drugs taken by 7 or more respondents, representing more than 10% of respondents, were: digoxin (ll), furosemide (9), ranitidine (7) and insulin (7). There were no significant differences in using or not using prescription drugs by age grouping (less than 77 years; 77 years or older) or by race (white or other); comparison chi square test p—values, .095 and .084 respectively. Phase 2 Phase 2 respondents, by selection, took at least one prescription drug on a daily or as needed basis. The mean number of prescription drugs was 4.9; the range, 1 to 13. More than half of respondents (18) took 5 drugs or less on a daily or as needed basis; 11 respondents took more than 5 drugs. Less than one quarter (22.6%) of the prescriptions were filled with generic drugs. PharmAssist software (1993) estimated that the mean number of potential drug-drug interactions (DDIs), including both prescription and over- the-counter drugs, was 2.7 (Table 2). 53 Table 2: Prescription (Rx) Drug Use Characteristics Characteristic Phase 1 Phase 2 (n=61) (n=29) Subjects Using One or More Rx Drugs 53 29 Mean Number Rx Drugs 3.3 4.9 Potential Drug—Drug Interactions (DDIs)a Mean 1.7b 2.70 Range 0-13 0-13 Std.Dev. 2.919 3.646 DDIs = o 27 9 DDIs = 1-13 26 20 aEstimated with PharmAssist software bRx drugs only cRx and over-the-counter drugs Table 3: Prescription Drug Use by Therapeutic Category m1 Phase; Therapeutic Category Freq % Freq % Hypertension Agents 53 28.0% 38 26.8% Other Cardiovascular Agents 27 14.2% 16 11.3% Arthritis Agents 21 11.0% 21 14.8% GI Agents 17 9.0% 15 10.6% Hyperglycemia Agents 12 6.3% 11 7.7% Other 60 31.5% 41 28.9% Total 190 100.0% 142 100.0% There was no significant relationship between the number of prescription drugs used and the number of potential DDIs (comparison Fisher exact test p—value 0.140). As in 54 Phase 1, the therapeutic categories used most frequently were Hypertension Agents (26.8%), Arthritis Agents (14.8%) and GI Agents (10.6%) (Table 3). There was no significant difference in therapeutic categories used by age group, race, income level or education level (Table 4). Table 4: Relationship of Phase 2 Demographic Variables to Drug Therapeutic Categories Used Variable Chi Square df p-Value Age: < or 277 10.14 5 .071 Race: white or other 5.00 5 .416 Income: 5 or > $11,500 0.65 5 .985 Education: { or a high school 2.41 5 _.790p_ Most drugs were taken once a day (61.3%) or twice a day (20.4%), although some (14.8%) were prescribed on an as needed basis. Respondents reported taking prescription drugs for durations ranging from less than 1 month to 480 months; the mean duration was 65.4 months (Table 5). The length of time a prescription drug had been taken was significantly related to income level, but not to age group or race. Respondents at the lower income level, $11,500 or less, were more likely to take a drug for more than 36 months. This may have been due to the availability of Medicaid and coverage for the costs of prescription drugs. 55 Table 5: Length of Time of Prescription Drug Use Months Number of Drugs % Cum.% <12 36 25.4% 25.4% 13-36 43 30.3% 55.6% >36 63 44.4% 100.0% Total 142 100.0% Mean Duration 65.39 Std. Dev. 88.430 Di r In r c ions All of the results in this section on dietary instructions were based on interviewer examination of drug container labels and medication fact sheets and on respondent reports. All indications for dietary instructions were according to Computerized Food—Medication Interactions (1993). Of the 142 drugs taken by Phase 2 respondents, 6 did not have any modification to the diet indicated as part of the drug regimen; 136 drugs did have indicated dietary instructions. More than half (56.3%) of the drugs with indicated dietary instructions had no instructions at all given for them. This was independent of age, race, income level and number of drugs taken; but significantly related to certain drug therapeutic categories (p=.000)(Table 6). 56 Table 6: No Indicated Dietary Instructions Given by Drug Therapeutic Category Di In 1 1v n Therapeutic Category I No Yes (Some/All)! Total AEEQEEEEJAQEE; """"""""" TNT—‘15 __________ T31" GI Agents l 11 4 | 15 Hyperglycemia Agents | 4 7 I 11 Hypertension Agents | 27 9 | 36 Other I 18 19 | 37 Other Cardiovascular Agents | 14 2 | 16 ‘ """""""""""""" ESE—T77"; """""" T1325" Chi square = 30.99 Degrees of freedom = 5 p—Value = .000 Comparing categories, Arthritis Agents had proportionately fewer drugs with no indicated dietary instructions given. This may have been attributable to the fact that most Arthritis Agents need to be taken with meals or with food or milk. Many respondents said that their doctors told them to take all of their drugs with food or meals. Individual codes were assigned to each dietary instruction indicated for a drug during data entry (Appendix D). In data analysis, the instruction codes were grouped into one of the six categories listed below. The categories 57 described the type of potential drug-food or drug-nutrient interaction (DFNI) the instruction was designed to prevent and were similar to those used in a previous research study (Lewis, Frongillo, and Roe 1995). 0 Alcohol Incompatibility O GI-Interference (drug malabsorption when administered with food) 0 GI—Meal Omission (reduced drug bioavailability or G1 irritation when administered on an empty stomach) O Inappropriate Supplement or Intake (toxicity and/or reduced drug efficacy with concurrent administration) 0 Drug-Induced Nutrient Deficiency 0 Reduced Drug Efficacy (drug effect or duration) The total number of specific dietary instructions indicated for the 142 prescription drugs taken by Phase 2 respondents was 394. According to respondent reports and interviewer examination of drug container labels and medication fact sheets, the total number of instructions actually given was 80 (20.3%). The DFNI category representing the highest number of indicated dietary instructions was Reduced Drug Efficacy (Table 7). The highest percentage of dietary instructions given was in the GI-Meal Omission category. 58 Table 7: Dietary Instructions Indicated and Given by Drug- food or Drug-nutrient Interaction (DFNI) Category DFNI Category W Indicated Given % Given Alcohol Incompatibility 90 25 27.8% GI-Interference 10 1 10.0% GI-Meal Omission 88 37 42.1% Inappropriate Supplement/Intake 64 5 7.8% Nutrient Deficiency 26 0 0.0% Reduced Drug Efficacy 116 12 10.4% Total 394 80 20.3% For individual respondents, the mean percent of indicated dietary instructions given was 24.6% with a range from 0% to 100%. Only one individual received 100% of indicated instructions; she was taking one drug which had one instruction indicated. Among multiple drug users (n=27), the two respondents with the highest percentages of dietary instructions given took 4 drugs (69.2% given) and 3 drugs (30.0% given) respectively. Medical Qutgemee Study 5E-36 Survey Seerea SF-36 scores were calculated as Physical Component Summary (PCS) scores and Mental Component Summary (MCS) scores. The PCS comprised subscales which addressed respondent assessment of personal physical functioning, physical role, 'l 59 bodily pain and general health; the MCS subscales addressed vitality, social functioning, emotional role and mental health. The summary scores have been recommended as more reliable measures of individual health status than the subscale scoring system originally applied to SF-36 results for groups (Ware, Kosinski, and Keller 1994). Phase 2 respondents presented PCS scores which were generally lower than the U.S. general population norms (or averages) for equivalent age and gender (Table 8). This may have been due to the high proportion of respondents who were over 77 years of age (48.3%), and in particular to the 10 respondents who were over 85 years of age. Individuals over 85 years of age are usually under represented in surveys and Ware, Kosinski, and Keller (1994) noted that the oldest old age groups comprising the norms were collapsed into the 65 years and over age group to provide a large enough sample to assure precision. These authors also noted that in the norm population there was an almost linear decline in PCS scores with advancing age beyond 65 years. Given these considerations, the physical health status of Phase 2 respondents probably did not differ greatly from others of the same age. There was no significant relationship between PCS score and demographic variables, number of drugs, any dietary instructions being given or percent of dietary 60 instructions given (Table 10). Table 8: Medical Outcomes Study SF-36 Health Survey Physical Component Summary (PCS) Scores and Norm Scores for U.S. Population of Equivalent Age and Gender PCS Phase 2 Norm for Women Norm for Women :65 Yrs. of Age 55-64 Yrs. of Age n 29 413 164 Mean Score 33.64 41.02 45.03 Std.Dev. 11.56 11.52 11.57 Range _ l6i56 _ 8-59 _ ___ I.._._-_.137§2___ MCS scores for Phase 2 respondents were also generally below the norms (Table 9). Ware, Kosinski, and Keller (1994) stated that, unlike PCS scores, MCS scores did not decline for the norm population older old respondents. For Phase 2 respondents, there was a very significant relationship between taking more than 5 prescription drugs and lower MCS score (Table 10). Respondents taking more than 5 drugs had a mean MCS score of 39.70, while those taking 5 or fewer drugs had a mean MCS score of 52.29. These mean scores were below the norm or average score for the U.S. general population of equivalent age and gender. It is not clear why taking more than 5 drugs might have this effect, but a clue may have been provided by one of the respondents who volunteered the comment that she was upset about having to take so many medicines and sometimes just ‘l 61 didn't take them (Appendix I). Since older adults are the biggest users of prescription drugs, having to take drugs may be an unpleasant reminder of one's advancing age and inevitable mortality. There was no significant relationship found between MCS scores and demographic variables, any dietary instructions being given or percent of dietary instructions given (Table 10). Table 9: Medical Outcomes Study SF-36 Health Survey Mental Component Summary (MCS) Scores and Norm Scores for U.S. Population of Equivalent Age and Gender MCS Pase Norm for Wome Norm rnorwome"_ 265 Yrs. of Age 55-64 Yrs. of Age n 29 413 164 Mean Score 47.52 51.44 50.56 Std.Dev. 13.09 10.54 10.16 Table 10: Relationship of Selected Variables to Medical Outcomes Study SF—36 Health Survey Summary Scores Variable Test PCS MCS p-Value p-Value Age: < or 277 anova .969 .488 Race: white or other anova .990 .403 Income: 5 or > $11,500 anova .342 .117 Education: < or a high school anova .947 .620 Number of Drugs: 15 or >5 t—test .213 .009 Diet Instructions: Yes or No anova .570 .286 % Instructions Given t-test .428 .192 62 Characteristics of Dietary Instruction Delivery Dietary instructions were given on drug container labels, in other written forms—-most often medication fact sheets--and orally. Instructions were occasionally given more than one way (Table 11). Table 11: Dietary Instruction Delivery Forms by Potential Drug—food Drug-Nutrient Interaction (DFNI) Category Label=1 Other Written=2 Oral=3 DFNI Category 1 2 3 1&2 1&3 2&3 All Total Alcohol Incompatibility l7 8 6 2 3 - - 36 GI-Interference - 1 — - - - — 1 GI—Meal Omission 13 10 18 2 2 1 1 47 Inappropriate Supplement/Intake — 5 — — — - — 5 Nutrient Deficiency — — - — — - - _ Reduced Efficacy 6 7 6 1 - — — 20 Total 36 31 30 5 5 1 l 109 To assess the ability of respondents to read container labels, each was asked to read the information on two real labels obtained from a pharmacy. Label 1 contained alcohol avoidance information in black text on a red background. Of 29 respondents, all but 4 (13.8%) could read the label although some readers had to find a brighter source of light. The 4 respondents who could not read the label knew that it said something about not drinking alcohol. 63 All could read label 2 which said to take the drug with food or milk and had black text on a white background. Although the different colors of the labels offer some visual association with the instructions themselves, it would probably be helpful for older adults to have the instructions reinforced verbally at the time the prescription is written or filled. More research might determine if the black on red label presents a significant reading difficulty for a larger population of older adults. For the purposes of this study, however, it is unlikely that not being able to read a label seriously impaired the ability of respondents to know what the instructions were. The most frequent source of written dietary instructions (Table 12) other than container labels was the pharmacist (64.5%) who used medication facts sheets as the format (Table 13). Oral instructions were most often given by physicians (56.7%) using individual counseling, although dietitians provided group instruction for 7 dietary instructions related to the use of Hyperglycemia Agents. Occasionally, written instructions were obtained by the respondent herself from independent sources. These sources of dietary instruction included a book about prescription drugs (1 instruction), the Physician's Desk Reference (1 instruction), sample package inserts (2 instructions), and a 64 label about drinking orange juice or eating a banana that was on the container of another, previously prescribed, diuretic drug which the respondent still had in her medicine cabinet (1 instruction). Table 12: Sources of Dietary Instruction Other Than Labels Source Written (not label) % Oral % Dietitian 6 19.4% 8 26.6% Doctor 1 3.2% 17 56.7% Pharmacist 20 64.5% 5 16.7% Other 4 12.9% - - Total 31 100.0% 30 100.0% Table 13: Dietary Instruction Formats Used by Health Professionals Format Frequency Label 36 Individual Counseling 23 Medication Fact Sheet 20 Brochure 7 Group Instruction 7 Sample Package Insert 2 Book 2 In addition to dietary instructions indicated for the 142 drugs taken by Phase 2 respondents, other dietary instructions were given for 39 of the drugs (27.5%). In these instances, other instructions were provided by the .l 65 doctor (26 drugs), the pharmacist (10 drugs), a container label for an earlier purchase (1 drug), a brochure provided by a nephew (1 drug), and a senior citizen organization's meeting presentation (1 drug). The methods used to provide extra instructions included individual counseling (28), container label (6), fact sheets (4) and group instruction (1). - ‘11 1:3 9. The extra instructions did not conflict with those indicated for 30 of the drugs, but for 9 drugs conflicting instructions were given. All conflicting instructions would pose a risk for DFNI and were associated with individual counseling by respondents' physicians (p=.018). In five instances, instructions had been given for individuals to take all their medications with food, probably in an effort to avoid gastrointestinal discomfort. Twice instructions were given for individuals to take all medications together in the morning before breakfast, probably to simplify the dosing schedule. One individual was told to take a specific drug with food which was recommended for administration on an empty stomach, again probably in an effort to avoid gastrointestinal distress. The extra instructions which did not conflict with those indicated were: take all medications with food (11), eat a 66 banana and/or drink a glass of orange juice every day (6), take all medications together in the morning (4), take with food or milk for GI distress (3), take with a full glass of water (2), do not drink alcoholic beverages (1), take with breakfast (1), general information on how to take medications (1), increase cereal/fiber intake to help control diabetes (1). Many of these instructions were listed as “may be needed," rather than indicated, by Computerized Food-Medication Interactions software (1993) and would be helpful for the respondent in managing her drug regimen. The instruction to take all medications with food was given as a blanket statement and, in the case of these instructions, would not create a potential DFNI. Attitudes, Beliefs and KquLedge When respondents were asked if they knew why a drug had been prescribed, they indicated 83.1% of the time that they did (118 of 142 drugs). They further indicated that 78.9% of the prescribed drugs (112) helped them. As expected, there was a strong association between knowing why a drug was prescribed and the belief that it helped (p=.000). When dietary instructions were given for a drug (59 drugs), most of the time respondents believed that the drug could be taken as directed and that they understood the dietary 67 instructions for their drugs and followed them. Responses were divided about knowing why dietary instructions were given or what problems could develop if the instructions were not followed, and about having trouble following the instructions (Table 14). Table 14: Respondents' Beliefs About Dietary Instructions Given for Each Drug Survey Question Respeuses Yes % No % Do you know why certain eating or drinking habits are included in your instructions? 32 54.3% 27 45.7% Do you know what problems might develop if you don't follow these instructions? 30 50.8% 29 49.2% Do you understand the instructions about eating and drinking habits? 59 100.0% - - Do you have any trouble following the instructions? 10 16.9% 49 80.1% As a general rule, do you follow the instructions? 54 91.5% 5 8.5% There was no significant relationship between knowing what problems could develop and income level, but there was a significant relationship to education level (p=.027). Those respondents who had completed 12 or more years of school were less likely to say that they knew what problems could develop than those who had less education. This was unexpected since education level is often considered a proxy 68 for income. The relationship may just be an artifact or health care providers may perceive that less educated individuals require more information or that information needs to be presented in a different way to achieve a positive outcome. Self—efficacy scores were very similar for all of the respondents who received dietary instructions with 93.2% feeling completely confident in their ability to adjust their eating or drinking habits in order to take their prescription drug(s) exactly as instructed. Respondents who needed more help thought that additional written information and someone explaining what they needed to do would make them fully confident. Seven respondents, taking a total of 10 drugs, gave the following reasons for trouble with dietary instructions: can't remember can't afford foods specified (diabetic meal plan) can't eat all the food listed (diabetic meal plan) exchange system is confusing (diabetic meal plan) activities interfere (2) instructed to take medicine four times a day with meals but only eat 3 meals. 69 Only 2 respondents indicated that they receive any help in taking their prescription drugs: one regularly from a neighbor, one occasionally from a home health aide. Respondents believed that they had experienced side effects with 30 of the 142 drugs taken (21.1%), including 8 complaints of GI distress (Table 15). Three complaints of GI distress came from respondents who should have received dietary instructions to help prevent the complaint, but did not. .I 70 Table 15: Side Effects Reported by Respondents That They Attributed to One of Their Drugs Side Effect Reported Frequency GI distress 8 Shakes/sweating 5 Fatigue 2 Drowsiness/very dry mouth 2 Puffiness 2 Reduced urine output 2 Colonic hemorrhage 1 Loss of taste sensation 1 Diminished Vision 1 Tender breasts 1 Hair thinned/facial hair formed 1 Ringing in ears/heavy head 1 Irritable/can’t think 1 Frequent urination (often can't find bathroom) 1 Dizziness 1 Adherence Adherence was assessed for each dietary instruction which was given, using the completed 1—day diet and medication records. Although 80 instructions were given, 8 instructions were for medications prescribed on an as needed basis and not taken on the record day. Adherence was scored yes or no. 71 Respondents adhered to 59 (81.9%) of the 72 dietary instructions given. Almost all of the instances of nonadherence involved respondents who believed that they understood their dietary instructions and followed them. These individuals may just have needed further explanation in order to adhere. A glaring exception in nonadherence was the dietary instruction to follow a diabetic meal plan. There was a significant difference (p=.000) between adherence to diabetic meal plans and adherence to all other dietary instructions (Table 16). Table 16: Adherence to Diabetic Meal Plan Instructions Compared to Adherence to All Other Dietary Instructions Dietary Instruction | No r Yes I Total BESSIES'EEQ """" 15116 All Other Instructions | 8 58 | 66 """"m""7551""T"’IS""’§;"T"3£" Fisher exact 2-tailed p-value = .000 The difference between adherence to instructions given for drugs in the Hyperglycemia Agents category, which included both oral hypoglycemics and insulin, and those in other categories was also significant (p=.0496) (Table 17), but less so because respondents did adhere to instructions for ' '1 BF" ‘l 72 those drugs other than the diabetic meal plan (ie. in relation to meals, avoid/caution with alcohol). Table 17: Adherence to Dietary Instructions Given for Hyperglycemia Agents Compared to Adherence to Dietary Instructions Given for All Other Drugs Adherence Drug Therapeutic Category | No Yes I Total ___________________________ +..______.__-__+.._____ Hyperglycemia Agents I 5 8 I 13 All Other | 8 51 | 59 ___________________________ +____________+______ Total I 13 59 I 72 Fisher exact 2—tailed p-value = .0496 Six respondents, 4 using insulin and 2 taking oral hypoglycemics, were given diabetic meal plan instructions, but only one adhered. Those who did not adhere said in the interview that they did not follow the instruction. Two also said that they were not completely confident that they could follow the instruction even if they wanted to. One of these indicated that she thought she could become confident if someone explained the meal plan to her again, talked about applying it when eating out, and gave her additional information to refer to at home. The other said that it wouldn't matter how much more help she had, that her 2000 kilocalorie diet was just more than she could eat and she did not intend to give up everything she liked (Appendix I). 73 Most of those who did not adhere also said that their doctors did not ask them if they followed their meal plans. If their blood sugar levels were high, the doctor just said to stop eating sweets (Appendix I). In one case, however, the doctor had scheduled the respondent to return to diabetic meal plan group instruction with a dietitian. The one respondent, on an oral hypoglycemic, who followed her meal plan said that she had lots of support. Her doctor always asked about how she was doing with her diet, had her meet with the office nurse to discuss it each time she came in, and went so far as to have her to dinner at his house to show her how to apply the plan to a real meal (Appendix I). None of the demographic variables were significantly related to adherence (Table 18). Of the variables hypothesized, the only one which proved to have a significant relationship to adherence was the doctor as a source of oral dietary instruction (Table 19). Other variables were also tested for a relationship to adherence revealing only one of significance, the reduced drug efficacy DFNI category (Table 18), which was negatively associated. The methods of instruction used for this category were not significantly related to adherence. 74 Table 18: Relationship of Adherence to Selected Variables Variable Test p-Value Assistance with drug regimen chi square .406 Belief that drug helps chi square .654 DFNI category chi square .002 Drug therapeutic category chi square .932 Hyperglycemia Agents category Fisher exact .0496 Knowledge of instruction's purpose chi square .108 Knowledge of nonadherence consequences chi square .229 Length of time drug taken chi square .158 Number of drugs chi square .932 Doctor as source of oral instruction Fisher exact .044 PCS score anova .831 MCS score anova .859 Self-efficacy Fisher exact .106 Side effects reported Fisher exact .336 Trouble following instruction chi square .254 75 Table 19: Adherence by Source of Oral Dietary Instruction Source of Oral Instruction | Ye: r n No I Total 52.35; """"""""""""""" I”; """""" 771%" Dietitian or Pharmacist | 6 6 | 12 ‘ _________________ EEQETWEB""'""§"T"£§" Fisher exact 2-tailed p—value = .044 Reduced efficacy DFNI category, Hyperglycemia Agents therapeutic category, and doctor as source of oral instruction were entered into forward stepwise logistic regression as dichotomous variables to determine their significance as predictor variables in a model for adherence. Reduced efficacy DFNI was the only variable included in the final model (Table 20) where it decreased the probability of adherence: Probability of adherence = 1 / 1 + e'gUU g(x) = 2.3224 — 2.3224(reduced drug efficacy DFNI) Table 20: Parameter Estimates for Logistic Regression Model Variable 8 S.E. Wald df Sig R Exp(8) Reduced Drug Efficacy DFNI —2.3224 .6853 11.4849 1 .0007 —.3735 .0980 Constant 2.3224 .4686 24.5596 1 .0000 The final model correctly classified 81.94% of cases. 76 Groups of other variables in domains suggested by literature review were also entered into logistic regression models but none were significant predictors of adherence. Summaty cf Results All of the results related to dietary instructions were based on interviewer examination of drug container labels and medication fact sheets and on respondent reports. Indications for dietary instructions were according to Computerized Food-Medication Interactions (1993). Hypotheses I, II and III were accepted: Hypothesis I: In the majority of cases (51% or more), respondents did not receive instruction for all indicated dietary modifications. This hypothesis was accepted based on the following results: 0 IMmmtyreight respondents (96.6%) did not receive all indicated dietary instructions. Only one respondent (3.4%) received all indicated dietary instructions; she was taking one drug with one instruction indicated. 0 The mean percent of indicated dietary instructions given per respondent was 24.6% (range 0.0% to 100.0%). ' I 77 O Cfi'the 394 dietary instructions indicated for 136 drugs, only 80 (20.3%) were actually given. '0 Seventy-seven (56.3%) of the 136 drugs which had dietary instruction indicated had no dietary instructions at all given. Hypothesis II: Respondents adhere to a majority (51%) of dietary instructions. This hypothesis was accepted since respondents adhered to 81.9% of instructions given. Hypothesis III:.A regression equation can be constructed which will predict dietary instruction adherence. This hypothesis was accepted since a logistic regression equation was constructed which correctly predicted adherence for 81.94% of the dietary instructions received by respondents in the study. The only predictor variable included in the logistic regression model was Reduced Drug Efficacy DFNI category, a variable not hypothesized. Its inclusion may have been enhanced by the presence of the diabetic meal plan instruction as a category component, since that instruction demonstrated a very high rate of nonadherence. 'I 78 The inclusion of only one variable in the final logistic regression model does not necessarily indicate that the hypothesized variables do not contribute to adherence or that they are unimportant in practice. In analysis, the doctor as a source of oral instruction was significantly related to adherence. This was consistent with the many respondents who mentioned during interview probing that their doctors told them anything important that they need to know about their prescription drugs. The significant relationship of the Hyperglycemia Agents category to adherence was similarly consistent with the very low rate of adherence to diabetic meal plan instructions. The failure of these variables to enter into the logistic regression model and of other variables to reach the d=.05 significance level of relationship may have been attributable to small sample size. CONCLUSIONS AND IMPLICATIONS Conclusions The following conclusions were reached in this study: Subjects did not receive most of the dietary instructions indicated for their prescription drugs, putting them at risk for drug-food and/or drug nutrient interactions. This was true despite a conservative approach to tabulating indicated instructions. If preventative measures, like increasing intake of foods high in potassium with use of potassium-depleting diuretics before laboratory tests detect hypokalemia, were added to the tabulation, the percentage of instructions given would have been even lower. Although the risk posed by many of the potential drug—food and/or drug-nutrient interactions was not necessarily life—threatening, even small decrements in functioning must be avoided for older adults to attain the highest levels of physical and social functioning. 79 80 0 When dietary instructions were given as part of the prescription drug regimen, they tended to be followed, except for diabetic meal plans. 0 Even when subjects did not adhere to instructions, they most frequently believed that they understood and followed them. 0 The instructions least likely to be followed were those intended to prevent reduced drug efficacy interactions, especially diabetic meal plans. 0 Oral instructions from doctors were significantly related to adherence and the doctor was viewed as the ultimate source of drug information. 0 Dietary instructions which conflicted with those recommended were most likely to be given by doctors. I 1' !' Dietary Instruction Delivery and DFNI.Awareness Since individuals cannot adhere to instructions they never receive, the key issue in reducing risk for drug-food and drug—nutrient interactions is delivery of appropriate dietary instructions. Strategies to assure delivery should focus on increasing awareness of DFNI risk and prevention measures in relevant arenas. Awareness of DFNIs, the means to prevent them and the ripple 81 effect of small decrements in functioning, should be defined components of continuing and higher education drug related courses for dietitians, physicians, pharmacists and nurses. This study indicated that the physician is viewed by the patient as the ultimate source of persuasive drug information and can play an effective role in enhancing patient awareness of the importance of dietary instructions. Whether physicians themselves have an adequate awareness is unknown. The evidence presented here suggests that they may not. For the physician who is educated to look for it, there is a wealth of information about DFNIs available in the form of handbooks and computer software like that used in this study. The Working Group on Health Education and High Blood Pressure (1987) has acknowledged the important role of the physician in improving adherence to high blood pressure treatment regimens, including those for drugs and dietary modifications. Even when the physician relies on other health professionals to communicate DFNI information to the patient, however, it is clear that the physician’s persuasive caution to the patient about the importance of following instructions from other sources may improve adherence. Although the dietitian is responsible for assuring 82 communications to the patient about DFNIs upon hospital discharge (Joint Commission on Accreditation of Hospitals 1986), pharmacists and nurses frequently assume the role of DFNI educator. Community dwelling individuals who receive outpatient drug prescriptions are unlikely to have the advantage of a dietitian's counsel since health insurance reimbursement is currently not allowed for such service. The nutrition community may best influence DFNI communications in these circumstances by enhancing awareness of DFNIs in other health professionals with whom they deal and by promoting public education messages through professional organization venues. The Physician Education Project of the American Dietetic Association and the public service announcements produced by The Society for Nutrition Education represent the kind of avenue which might be used. Like the issue of drug-drug interactions, DFNI awareness and education also need to be widely addressed at the level of the consumer. Older women, and all other consumers, should know enough about the possibility of DFNIs to question the need to modify eating and drinking habits in prescription drug regimens. Many people are now informed consumers who question the combinations of drugs they take. Publications like those from the American Association of Retired Persons, 83 “The Smart Consumer's Guide to Prescription Drugs” (American Association of Retired Persons 1989a) and “Healthy Questions” (American Association of Retired Persons 1989b), are a step toward DFNI awareness. These publications tell readers to ask pharmacists and doctors about taking drugs with certain food and beverages, but they do not tell them why they should ask or mention that quality of life issues may be involved. Measures being taken by the Food and Drug Administration to mandate readable, understandable information about over—the-counter drugs is another step toward DFNI awareness. The most effective ways to achieve increased awareness and knowledge of DFNIs among older women need to be identified. As part of this study, the resident advisory council and participants of Phase 2 were questioned informally about their interest in attending a presentation about the relationship of foods to prescription drugs. They were polite, but unanimous in their disinterest. The disinterest may have reflected the views of several Phase 2 respondents who, in response to the interviewer script section suggesting that they might find it helpful to ask the pharmacist or physician about the need for dietary modifications in their prescription drug regimens, said that their doctors tell them everything they need to know about 84 their medicines. The resident manager said that there are often health related presentations in the complex and that they are generally not well attended. Diabetic Meal Plans Results of this study related to diabetic meal plans were very alarming. A low percentage (just more than 50%) of diabetic meal plan instructions were given and there was very poor adherence. This is in direct opposition to the position of the American Diabetes Association (ADA) that medical nutrition therapy is integral to total diabetes care and management (American Diabetes Association 1994). Reasons cited for nonadherence (see p.72) suggested that respondents did not understand how to apply their meal plans, that meal plans had not been individualized and that skills had not been developed in adapting meal plans to special situations. The Food Guide Pyramid has been recommended for use in diabetic meal planning (American Diabetes Association 1994) and could prove to be a simpler, yet effective, approach for the older adult. The one respondent who actually adhered to her diabetic meal plan mentioned that she had been given the Food Guide Pyramid to use and that she liked it (Appendix I). Four respondents who reported that they received no dietary 85 instructions were taking oral hypoglycemics; two were using insulin. Although the need for coordinating insulin use with food and beverage intake is generally acknowledged, the inclusion of dietary modifications in oral hypoglycemic .regimens may have been viewed as optional. The American IDiabetes Association position on medical nutrition therapy (ZUnerican Diabetes Association, 1994) in non—insulin- -U)[\)l—‘ 109 DRUG CATEGORIES CODE SHEET Drug Categories diuretic-K depleting calcium channel blocker ACE inhibitor combination diuretic beta blocker adrenergic stimulator ACE inhibitor with diuretic alpha adrenergic blocker rauwolfia combination antianginal cardiotonic antihyperlipidemic antiarrhythmic antiplatelet antiulcer antiinflammatory/antiarthritis antidepressant antidiabetic/insulin bronchodilator potassium preparation thyroid estrogen antineoplastic analgesic oral hypoglycemic muscle relaxant corticosteroid anticonvulsant antianxiety antihistamine prescription vitamin/mineral supplement antispasmodic antibiotic antianemic stool softener/laxative sedative antinauseant antigout antisecretory psoralen opthalmic diuretic—K sparing expectorant APPENDIX E PHASE 2 MEDICAL OUTCOMES STUDY SF-36 SURVEY 110 Phase 2 Medical Outcomes Study SF—36 Survey THE MOS 136-ITEM SHORT-FORM HEALTH SURVEY (SF-36) INSTRUCTIONS: This survey asks for your views about your health. This information will help keep track of how you feel and how well you are able to do your usual activities. Answer every question by marking the answer as indicated. If you are unsure about how to answer a question, please give the best answer you can. 1. In general, would you say your health is: (circle one) Excellent ......................... 1 Very good ........................ 2 Good ............................ 3 Fair ............................. 4 Poor ............................ 5 2. We. how would you rate your health in general hog? (circle one) Much better now than one year ago ..... 1 Somewhat better now than one year ago . . 2 About the same as one year ago ........ 3 Somewhat worse now than one year ago . . 4 Much worse now than one year ago ..... 5 W'iMWSTMUnc. ”Macaw“. 111 3. The following items are about activities you might do during a typical day. Does your health noleimit you in these activities? If so, how much? (circle one number on each line) Yes, Yes, No, Not ACTIVITIES Limited Limited Limited ' A Lot A Little At All a. Vigorous activities, such as running, lifting heavy objects, participating in strenuous 1 2 3 sports b. Moderate activities, such as moving a table, pushing a vacuum cleaner, bowling, 1 2 3 or playing golf c. Lifting or carrying groceries 1 2 3 d. Climbing several flights of stairs 1 2 3 e. Climbing one flight of stairs 1 2 3 f. Bending, kneeling, or stooping 1 2 3 9. Walking more than a mile 1 2 3 h. Walking several blocks 1 2 3 i. Walking one block 1 2 3 j. Bathing or dressing yourself 1 2 3 4. During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities es a result of your physical health? (circle one number on each line) YES NO a. Cut down the amount of time you spent on 1 2 work or other activities ' b. Accomplished less than you would like 1 2 c. Were limited in the kind of work or other 1 2 activities d. Had difficulty performing the work or other 1 2 activities (for example, it took extra effort) Copynqnt ° 1992 M05 Trust. Inc. All ngnts reserved. 112 5. During the past 4 weeks. have you had any of the following problems with your work or other regular daily activities as a result of a_nv emotional problems (such as feeling depressed or anxious)? (circle one number on each line) YES NO a. Cut down the amount of time you spent on work or other 1 2 activities b. Accomplished less than you would like 1 2 0. Didn't do work or other activities as carefully as usual 1 2 6. During the past 4 weeks. to what extent has your physical health or emotional problems interfered with your normal social activities with family, friends, neighbors, or groups? (circle one) Not at all ............................... 1 Slightly ................................. 2 Moderately .............................. 3 Quite a bit .............................. 4 Extremely ............................... 5 7. How much bodily pain have you had during the past 4 weeks? (circle one) None ........................ 1 Very mild ..................... 2 Mild ......................... 3 Moderate ..................... 4 Severe ....................... 5 Very severe ................... 6 Copyright ° 1992 M05 Trust. Inc. AII ngnts reserved. 113 8. During the east 4 weeks, how much did oain interfere with your normal work Gncluding both work outside the home and housework)? (circle one) Not at all ..................... 1 A little bit ..................... 2 Moderately .................... 3 Quite a bit .................... 4 Extremely ..................... 5 9. These questions are about how you feel and how things have been with you durinq the past 4 weeks. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time during the past 4 weeks I (circle one number on each line) A Good A All Most Bit of Some Little None of the of the the of the of the of the Time Time Time Time Time Time a. Did you feel full of pep? 1 2 3 4 5 6 b. Have you been a very 1 2 3 4 5 6 nervous person? 0. Have you felt so down in the dumps that 1 2 3 4 5 6 nothing could cheer you up? d. Have you felt calm and 1 2 3 4 5 6 peaceful? e. Did you have a lot of 1 2 3 4 5 6 energy? ' f. Have you felt 1 2 3 4 5 6 downhearted and blue? 9. Did you feel worn out? 1 2 3 4 5 h. Have you been a happy 1 2 3 4 5 person? ' i. Did. you feel tired? 1 2 3 4 5 6 Cooyrignt ° 1992 M05 Trust. Inc. All rights reserved. 114 10. During the past 4 weeks, how much of the time has your physical health or emotional problems interfered with your social activities (like visiting with friends, relatives, etc.)? (circle one) All of the time ................. 1 Most of the time ............... 2 Some of the time ............... 3 A little of the time .............. 4 None of the time ............... 5 11. How TRUE or FALSE is each of the following statements for you? (circle one number on each line) Definitely Mostly Don’t Mostly Definitely True True Know False False a. I seem to get sick a little easier than other 1 2 3 4 5 people b. I am as healthy as 1 2 3 4 5 anybody I know c. I expect my health to 1 2 8 4 5 get worse d. My health is excellent 1 2 3 4 5 Copyright ’ 1992 M05 Trust. Inc. All rights reserved. APPENDIX F l-DAY FOOD AND MEDICATION RECORD INSTRUCTIONS AND FORM 115 One-day Food and Medication Record Instructions and Form ONE-DAY FOOD AND MEDICINE RECORD Directions: Choose one weekday between now and your interview. On that day, use the following pages to record all foods, beverages, water, prescription medicines and the time of day when they are consumed. Don't forget to include snacks, nibbles, condiments (ie. catsup, salt), alcoholic beverages, liquid diet supplements (ie. Instant Breakfast or Ensure) and foods you eat away from home. If you take vitamin/mineral preparations (ie. One-A-Day, Centrum, calcium, Vitamin E) or non- prescription medicines (ie. aspirin, Tylenol, cough syrup), record them in the medicine column. Please bring this record with you to your scheduled interview. 116 Time Food/Beverage/Water Name of Medicine Time 116 Food/Beverage/water Name of Medicine Time 117 Food/Beverage/Water Name of Medicine APPENDIX G CONSENT FORMS 118 Consent Forms Research Study Participation This study is being condiJcted by a health research team from Michigan State University. The information obtained will be used to identify health education needs among older women and to develop materials or programs to meet those needs. Participation is completely voluntary. Any information you give will be held in strict confidence. No one outside the ' research team will ever be able to connect your answers with you or your address. You do not have to answer any questions that make you uncomfortable in any way. You can end your participation at any time you wish. The interview you are scheduled for today will take adet one, hour. If you understand the explanation on this page and agree to participate in the study, please sign your name below. .EAME: DATE: 119 PERMISSION TO CONTACT PHYSICIAN During your participation in this research study, we may have obtained information from you which your doctor should know. We will not give that information to him/her unless you agree that it is alright for us to do so. Please sign your name below if you will allow us to contact the doctorls) who prescribed your medication. Date APPENDIX H DRUG CONTAINER LABELS USED FOR THE PHASE 2 READING TEST 120 Drug Container Labels Used for the Phase 2 Reading Test Label 1: Black letters on a red background X/po NOT DRINK ; ALCOHOLIC BEVERAGES WHEN mane mus MEDICATION Label 2: Black letters on a white background APPENDIX I ANECDOTAL INFORMATION/INTERVIEWER OBSERVATIONS AND OVER-THE- COUNTER DRUG USE (OTCS) FOR SELECTED RESPONDENTS BY ID 121 ANECDOTAL INFORMATION/INTERVIEWER OBSERVATIONS AND OVER-THE- COUNTER DRUG USE (OTCS) FOR SELECTED RESPONDENTS BY ID 12 rm i n v i ns QTCS 1 She went to the library and read up on HTN and on her medication in the PDR. She is an RN. Doctor said she was more concerned about her watching her fat intake than sodium. She is quite slender and looks much younger than her age. OTCs: calcium & vitamin D (500mg & ZOOIU X 2) aspirin lOgr Centrum 5 She is a heavy smoker—-can't quit. She didn't sign physician consent form. She says the doctor just lectures her about smoking and she can't quit. They had a falling out and she hasn't been back since. She does everything he wants her to do but just can't quit smoking. The doctor said she should eat salt, that her body is low in salt. She runs a test tape on urine am & pm. When she gets the shakes she eats a small candy bar. She had the shakes just before interview, delaying it 20 minutes. She was on insulin years ago and "weaned" herself off. Doctor didn't give her a meal plan this time because he knew she still had one from before (2000 kcal-she can't eat that much now). She eats dinner in the complex dining room every day. OTCs: garlic-parsley pill aspirin 6 She was in the hospital with ulcers last summer—-has had ulcer problems for years. She also has had gall bladder attacks, but doctor reluctant to do surgery-~said she is high risk because of gastric problems and high blood pressure. Doctor told her to take all meds with food unless otherwise instructed. OTCs: docusate sodium (2) X 2 garlic pill calcium—magnesium-phosphorus-D (650mg-400mg-500mg—ZOOIU X 2) Phillips milk of magnesia Tessalon Perles X 4 10 13 16 122 She received a meal plan when she started on hypoglycemics about 5 years ago, but never really followed it. When she does follow the plan she finds it discouraging that blood sugar readings don't go down (she uses glucometer). Also she is distressed that even when readings are low at night (94) they are high again by morning (134). Doctor doesn't ask about meal plan adherence and she doesn't tell him that she does not follow. She did not know where the meal plan was for me to see. She says the doctor is pleased with her blood sugar. Monitor readings usually run 120-130 at home—-highest about 160. OTCs: One—a-day vitamin X 2 Fibre—con (2) X 2 She had some medication sheets for some of the meds, she thinks, but she says she always puts those away somewhere where she can't find them. OTCs: multi—vitamin (100% RDA) She has emphysema--oxygen tank with her at all times. Doctor not happy with blood sugar. She monitors am & pm—-runs low 2005 am; about 250 pm. Doctor is sending her to dietitian on Fri. He does not routinely ask about her diet, but she has told him she does not stay on meal plan. She was on an oral hypoglycemic for several years first; insulin later. She says she just gets lonely and food is solace. She said if someone "really leaned on" her she would probably stick to the diet. Meal plan not available to see-—didn't know where it was. She takes 2 Slo-bid if she has a cold; extra atarax (total 4) when especially nervous. She drinks herb & regular tea, and takes Lanoxin. DTCS: aspirin She seems to have no problem scheduling medicine correctly. OTCs: 1/2 aspirin calcium & vitamin D (600mg & ZOOIU) X 2 19 20 21 123 She worked as a nurse's aide for many years so is familiar with medications. She has some financial problems getting fresh fruit and some medications. Doctor ordered potassium supplement but she has not had prescription filled because she would rather not spend the money. She has had some signs of potassium deficiency though so she guesses she will. Thinks she may be having an allergic reaction to oranges and will have to stop eating them for potassium. OTCs: none She has her medication fact sheets and she says she refers to them regularly. She took triamterene one time with just water and it bothered her GIT--read fact sheet and makes sure she takes with food now when fact sheet says to. She takes ferrous sulfate 325mg three times per day--after meal with full glass water--per doctor who says she is anemic. She is a very small, thin person. OTCs: ferrous sulfate 325mg X 3 She is upset about having so many medicines so sometimes she just doesn't take them. Potassium supplement bothers her stomach so she only takes it once in a while. There are far too many pills left in her containers for the dates on the labels. She takes Zantac only when her ulcer bothers her and then takes one instead of two as directed. Her daughter-in-law is a doctor and she prescribed the Ambien. It is very hard for her to get around because of arthritis. Used to have help coming in for housework before but not cooking. Apartment was an unbelievable mess, with litter everywhere, every kitchen cupboard/oven door standing Open, an Old beef roast sitting out with fat congealed. OTCs: Advil (2) X 3 acetominophen 28 29 31 124 She was hospitalized for "bladder problems" which she says is why she now takes the K—Dur and Demadex. Many years ago she took another kind of thryoid that she says was much stronger. She has talked to the doctor about increasing her dose, but he doesn't think she needs it. She feels very tired all the time and isn't able to do hardly anything around the house. (She is over 90.) She has an aide who comes in every day and helps with her bath, dishes, and whatever she needs. She's not sure what days she should take the K-Dur. She thinks maybe her Meijer medicines came with fact sheets. She's not sure. If so, they were thrown away--but she "knows what they said". DTCs: docusate sodium Advil (2) Given 1200 kilocalorie diabetic diet some 30 years ago, but says she couldn't eat that much food. She never has talked to a dietitian or nurse about her diet through the years. Doctors have talked about her diet, but mostly just tell her she has to cut out the sweets when her suger goes up. Sugar only goes up occasionally. Her diet is now limited by availability of congregate meals in the apartment complex-4 or 5 per week. The rest of the time she does pickup food like canned soup. Asked her about meals on wheels --says she gets enough of that kind of food (made a face) in the complex dining room. She says her daughter wants her to move to Burcham Hills where 3 meals/day are provided. She is over 90. OTCs: Meijer vitamin/mineral supplement for those over 50 years old Very alert and seems to have no problems with her drug regimen. Her granddaughter was there Visiting and said she seems to be just fine on her own. Using Ensure once a day. She is over 90. DTCs: none 33 34 35 125 Takes her Darvocet-N with food because it upsets her stomach otherwise--but was not instructed to do so. Uses Fiber-con to correct constipation. She was in an auto accident 5 years ago and damaged her hip. It still bothers her a lot and she takes Darvocet for pain. Had hysterectomy many years ago. OTCs: Fibre—con (2) X 2 -- takes with her Lanoxin iron supplement once a week (not available to see) aspirin She believes that the many supplements she takes keep her healthy. She certainly is very spry. She is a senior companion and keeps very busy. She thinks firmly that her garlic capsules have helped her blood pressure. She has many books on medications and health and says she has read them, but doesn't seem to know anything about potential drug-food or drug-supplement interactions. OTCs: high potency multi-vitamin/mineral B-complex pill antioxidant pill garlic capsule calcium-magnesium (500mg-250mg X 4) vitamin E 400mg vitamin C 500mg zinc 50mg Metamucil She is very systematic with putting pills for each day in little cups to be sure she takes them correctly. She is using mail order pharmacy. She gets medication fact sheets and reads them, but then throws them away. None were available to see. She says she'll keep them from now on, it's just that she seems to collect so much paper and there is no room for it all in her small apartment. DTCs: aspirin (l) 37 39 41 126 She was given a diabetic doesn't follow now. She thinks she eats too much diet many years ago, but keeps away from sugar, fat but bread—-but isn't going to change that. Her doctor thinks she does fine. She does not monitor her sugar, just has lab tests when the doctor orders. She cannot get out of the house much since her husband is housebound, apparently with dementia. OTCS: Centrum Silver She is very alert and bright. She eats lots of bran flakes and oatmeal, and takes Lanoxin. Centrum Silver Tums EX with Calcium Vitamin C 250mg baby aspirin (1) docusate sodium (1) Excedrin (1) OTCs: (Ca=750mg) X 3 She says she has a lot of trouble with memory from hardening of brain blood vessels. She could not handle medicines if her friend did not lay them out for her with signs every day. Her diet record is very sparse and she says that she has no appetite at all. I talked to her when the interview appointment was made but she did not remember the date. She wasn't in the building at appointment time, had been to the doctor. She came home later in the day and was willing to do the interview then. She had not been told to avoid alcohol and said it would be difficult for her to do so since she and her gentleman friend like to go out for dinner and drinks. OTCs: none 43 45 52 127 Almost 90 years old, she is very bright and alert. She moved to assisted living after she fell and needed therapy on her leg. She is eager to get back to her own apartment. Her niece says they want her to stay where she is--that she made a mistake taking her medicine while she was there and now they are worried. She was taking her own drugs, now the nurse dispenses. Her medication containers were not available to see, but medication dosing schedule was available from her chart. There were no diet instructions at all. The only drug she's still taking from her list on the first survey is prednisone--she remembered that she is supposed to take that with food or milk. She said her doctor told her that and she makes sure that if they hand her the pills before there is food on table she waits to take the pills until she has food. OTCs: baby aspirin (1) Tylenol (2) When she was found to be diabetic, she was given a diet plan but ever since her doctors have said she's doing ok as she is. She used to monitor her sugar with urine tape, but has not done so in many years. She "can tell" when her sugar gets high. She says it usually is in good control. She tends to get leg cramps. She says she lost 3 pounds over the holidays and that worries her. She is very tiny-~says she now weighs 105 lbs, but used to weigh 120. She lost weight after her heart bypass 1.5 years ago. Uses Ensure once a day. OTCs: Meijer Advanced Formula multi-vitamin/mineral aspirin (1) Doctor told her to always take all drugs with food and she always does. Aspirin and vitamin/mineral supplement cause GI distress if she doesn't take with food or milk. She was very slow about filling in the MOS survey, but didn't seem to have any trouble answering questions. She says she has a lot of trouble with eyes, and holds papers right up to her nose in order to see. She takes her Lanoxin with bran flakes, drinks herbal tea. OTCs: Bufferin (1) Myadec high potency vitamin/mineral supplement 53 54 55 56 128 Her husband is very ill-—home health aide coming in. He has had multi-strokes and cannot be left alone. OTCs: enteric-coated aspirin (1) She is not taking Norvasc anymore only rarely takes Tylenol #4 (did not have it available for label). She is taking Excedrin PM for sleeplessness and "funny dreams"—- 2 at bedtime. If she does not take it, she has horrible dreams, ie. about dead husband with glowing green eyes. She only has experienced the dreams since first taking Excedrin PM-—she didn't have dreams before. Her diet record is very sparse but she is quite heavy—set. She did have Tylenol #4 medication sheet available to see. OTCs: Excedrin PM (2) Doctor told her to just take all drugs with food as a rule. She also takes Advil as needed for leg cramps--sometimes it helps more than Flexeril. She is not taking peri—colace or colace any more. Neighbor suggested that she eat oatmeal instead and she says that seems to work. Although not instructed to do so, she does take Oxsoralen with food, says it upsets her stomach if she doesn't. She forgot to include insulin on original survey list. She was given a diabetic diet years ago but does not follow it now~-in moving she lost it. She monitors blood sugar herself every am. She has been upset that it varies widely. Her doctor is pleased if readings are under 200. She was in the hospital with ketoacidosis when put on insulin. OTCs: Advil She takes a 812 injection (1000mg) every 2 weeks. She was given a diabetic meal plan initially and she knows what it was, but didn't have it around any more. She monitors her blood sugar every day--usual range 200-270, once in a while up in the 3005. Doctor thinks her meter may not be working right because she has big swings. Last night it read 116. She is not taking Quinamm or Tagamet any more. OTCs: calcium-Vitamin D (600mg—2001U) X 2 57 6O 61 129 She believes firmly that her chromium supplement improves blood sugar control. She has a meal plan and full instructions but does not follow it. She says she never has money for groceries--gives money to her daughter who is always in trouble. She used to follow the meal plan carefully and understood it completely. She thinks she should inject insulin twice a day instead of just in the morning because she gets irritated, tired and can't think in afternoon. Doctor said it was ok to do, but she found she couldn't remember to do pm injection, so went back to all in am. She uses a glucometer which she says ranges 130-154 with chromium; 160-214 without. OTCs: Vitamin C 1000mg One-a-Day vitamin/mineral supplement 400mcg chromium picolinate enteric-coated aspirin (2) She continues to go to KMart for prescriptions, even though it’s far away, because the pharmacist takes time to talk to her about her drugs. She had knee surgery in 1978 and had to have potassium for 10 days before they could operate. She had been falling a lot before that. Now she always eats a banana or drinks orange juice as her old medicine bottle.said many years ago. She used to be diabetic but lost 60 lbs and walked a lot and now is ok without insulin or pills. She is not taking Quinamm or Zantac now since Flexeril and Prilosec work better and doctor said to not take both. Her doctor retired. She was taking Titralac antacid 20 per day until she started the Prilosec. OTCs: Vitamin C 500mg X 2 Family history of diabetes--brother especially had it very bad. Knowing that encourages her to follow her diet and be careful. She has difficulty with English but goes to a doctor and nurse who speak Spanish and who help her with her diet. Her doctor asks her on a regular basis about how she is following her diet and even has her over to dinner. When the MSU health fair came to the apartment complex, a woman there talked to her about her diet plan and gave her a copy of the new food guide pyramid which she incorporates into her meal plan. She is not taking Premarin any more and is to have hysterectomy later this month. OTCs: aspirin (l) LIST OF REFERENCES LIST OF REFERENCES Aday LA. Designing and conducting health surveys: a comprehensive guide. San Francisco: Jossey—Bass Inc.; 1989. American Association of Retired Persons. The smart consumer's guide to prescription drugs. Washington, D.D.: American Association of Retired Persons; 1989A. American Association of Retired Persons. Healthy questions: how to talk to and select physicians, pharmacists, dentists and vision care specialists. Washington, D.C.: American Association of Retired Persons; 19898. American Diabetes Association. Position Statement: nutrition recommendations and principles for people with diabetes. Diabetes Care. 1994; 17(5): 519-522. American Diabetes Association. Consensus statement: self- monitoring of blood glucose. Diabetes Care. 1995; 18: 47—52. American Medical Association Council on Scientific Affairs. American Medical Association white paper on elderly health. Arch Intern Med. 1990; 150: 2459-2472. Anderson EG. Keeping an eye on older patients and their drugs. Geriatrics. 1990; 45: 81-82. Axelson ML; Brinberg D. The measurement and conceptualization of nutrition knowledge. J Nutr Educ. 1992; 24(5): 239-246. Axelson ML; Federline; Brinberg D. A meta-analysis of food- and nutrition-related research. J Nutr Educ. 1985; 17: 51-59. Baer JS; Holt CS; Lichenstein E. Self-efficacy and smoking reexamined: construct validity and clinical utility. J Consult Clin Psychol. 1986; 54: 846-852. 130 131 Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev. 1977; 84: 191—215. Bandura A. Social foundations of thought and action. Englewood Cliffs, NJ: Prentice-Hall; 1986. Bernier N; Avard J. Self—efficacy, outcome and attrition in a weight reduction program. Cognitive Ther Res. 1986; 10: 319-338. Blumberg JB. A discussion of drug metabolism and actions in the aged. Drug-Nutr Interact. 1985; 4: 99-106. Blumberg JB; Suter P. Pharmacology, nutrition, and the elderly: interactions and implications. in: Chernoff R, Editor. Geriatric Nutrition. Gaithersburg, MD: Aspen Publishers, Inc.; 1991. Calloway NO; Foley CF; Lagerbloom P. Uncertainties in geriatric data: organ size. J Am Geriatr Soc. 1965; 13: 20-28. Campbell VA; Dodds ML. Collecting dietary information from groups of older people. J Am Diet Assoc. 1967; 51: 29—32. Carruth AK; Boss BJ. More than they bargained for: adverse drug effects. J Gerontol Nurs. 1990 Jul; 16(7): 27—31. Chrischilles EA; Foley DJ; Wallace RB; Lemke JH; Semla TP; Hanlon JT; Glynn RJ; Ostfeld AM; Guralnik JM. Use of medications by persons 65 and over: data from the established populations for epidemiologic studies of the elderly. J Gerontol. 1992; 47(5): M137—M144. Clark LT. Improving compliance and increasing control of hypertension: needs of special hypertensive populations. Am Heart J. 1991; 121(2): 664-669. Cohen J:. Pharmacokinetic changes in aging. Am J Med. 1986; 30: 31—38. Col N; Fanale JE; Kronholm P. The role of medication noncompliance and adverse drug reaction in hospitalization of the elderly. Arch Intern Med. 1990; 150(4): 841—845. Colt HG; Shapiro AP. Drug—induced illness as a cause for admission to a community hospital. J Am Geriatr Soc. 1989; 37: 322—326. 132 Computerized Food-Medication Interactions [software]. Pronsky ZM; Pronsky W; Steinman J; Steinman R. Coral Springs, FL: Poly-bites, Inc./Food-Medications Interactions, Inc.; 1993. Contento IR; Murphy BM. Psycho-social factors differentiating people who reported making desirable changes in their diets from those who did not. J Nutr Educ. 1990; 22(1): 6—14. Cooper JW. Drug-related problems in the elderly at all levels of care. J Geriatr Drug Ther. 1990; 4(3): 79-83. Cornoni-Huntley J; Brock DB; Ostfeld AM; Taylor JO; Wallace RB, editors. Established populations for epidemiologic studies of the elderly: resource data book. Washington D.C.: National Institute on Aging, U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health; 1986(NIH Publication No.86-2443). Cusack B; Kelly J; O'Malley K; Noel J; Lavan J; Horgan J. Digoxin in the elderly: pharmacokinetic consequences of Old age. Clin Pharmacol Ther. 1979; 25: 772—778. Darnell JC; Murray MD; Martz BL; Weinberger M. Medication use by ambulatory elderly: an in-home survey. J Am Geriatr Soc. 1986; 34(1): 1-4. Davies DF; Schock NW. Age changes in glomerular filtration rate, effective renal plasma flow, and tubular excretory capacity in adult males. J Clin Invest. 1950; 29: 496—507. Dean AG. Epi Info Version 6 Manual. Atlanta: Centers for Disease Control and Prevention; 1994. Denham MJ. Adverse drug reactions. Br Med Bull. 1990; 46(1): 53-62. Department of Health and Human Services (Public Health Service). Healthy people 2000: national health promotion and disease prevention objectives. Washington, DC: Superintendent of Documents, U.S. Government Printing Office; 1990. (DHHS Publication No. PHS 91—50213). Dwyer JT. Screening older Americans' nutritional health: current practices and future possibilities. Washington, D.C.: Nutrition Screening Initiative; 1991. 133 Dwyer JT; Krall EA; Coleman KA. The problem of memory in nutritional epidemiology research. J Am Diet Assoc. 1987; 87: 1509-1512. Elahi VK; Elahi D; Andres R; Tobin JD; Butler MG; Norris AH. A longitudinal study of nutritional intake in men. J Gerontol. 1983; 38: 162-180. Epi Info [software]. Dean AG; Dean JA; Coulombier D; Brendel KA; Smith DC; Burton AH; Dicker RC; Sullivan K; Fagan RF; Arner T6. 6. Atlanta: Centers for Disease Control and Prevention; 1994. Fanelli MT; Stevenhagen KJ. Consistency Of energy and nutrient intakes of older adults: 24-hour recall vs. l-day food record. J Am Diet Assoc. 1986; 86(5): 655-667. Fedder DO. Managing medication and compliance: physician-pharmacist-patient interactions. J Am Geriatr Soc. 1982; 30(11 suppl): $113-$117. Fletcher SW; Fletcher RH; Thomas DC; Hamann C. Patients' understanding of prescribed drugs. J Comm Health. 1979; 4: 183-189. Food and Drug Administration. Draft guideline for postmarketing reporting of adverse drug reactions. Rockville, MD: Division of Drug and Biological Products, Experience Center for Drugs and Biologics; 1985; Docket No. BSD-0249. Fowler FJ. Survey research methods. Beverly Hills: Sage Publications; 1988. Franz MJ; Horton ES; Bantle JP; Beebe CA; Brunzell JD; Coulston AM; Henry RR; Hoogwerf BJ; Stacpoole PW. Nutrition principles for the management of diabetes and related complications. Diabetes Care. 1994; 17: 490— 518. Fulton B; James O; Rawlins MD. The influence of age on the pharmacokinetics of paracetamol. Br J Clin Pharmacol. 1979; 4: 418-421. Funnell M; Anderson L; Arnold M; Fogler J; Merritt J. Participation in a diabetes education and care program: experience from the Diabetes Care for Older Adults Project (DCOAP). 54th Annual Meeting of the American Diabetes Association; 1994 Jun 11; New Orleans. 134 Gallup G; Cotugno HE. Preferences and practices of Americans and their physicians in antihypertensive therapy. Am J Med. 1986; 81(suppl C): 20-24. Gans KM; Lasater TM; Linnan L; LaPane K; Carleton RA. A cholesterol screening and education program: differences between older and younger adults. J Nutr Educ. 1990; 22(6): 275-282. Garry PJ; Goodwin JS; Hunt WC; Hooper EM; Leonard AG. Nutritional status in a healthy elderly population: dietary and supplemental intakes. Am J Clin Nutr. 1982; 36: 319-331. German PS; Burton LC. Medication and the elderly: issues of prescription and use. J Aging Health. 1989; 1(1): 4—34. German PS; Klein LE. Side effects of drugs in the elderly. Comp Ther. 1986; 12(4): 3—5. German PS; Klein LE. Drug side effects and doctor/patient relationship among elderly patients. J Soc Adm Pharm. 1984; 2: 67-73. German PS; Klein LE; McPhee SJ; Smith CR. Knowledge and compliance with drug regimens among the elderly. J Am Geriatr Soc. 1982; 30: 568-571. Gersovitz M; Madden JP; Smiciklas—Wright H. Validity of the 24-hr. dietary recall and seven-day record for group comparisons. J Am Diet Assoc. 1978; 73: 48—55. Gibbs S; Waters WE; George CF. Prescription information leaflets: a national survey. J R Soc Med. 1990; 83(5): 292-297. Gibson DM; Aitkenhead W. The elderly respondent: experiences from a large-scale survey of the aged. Res Aging. 1983; 5(2): 283-296. Gilchrist WJ; Lee YC; Tam HC. Prospective study of drug reporting by general practitioners for an elderly population referred to a geriatric service. Br Med J. 1987; 294: 289-290. Glanz K; Eriksen MP. Individual and community models for dietary behavior change. J Nutr Educ. 1993; 25(2): 80-86. 135 Greenblatt DJ. Disposition of cardiovascular drugs in the elderly. Med Clin North Am. 1989; 73(2): 487-494. Greenblatt DJ; Allen MD; Harmatz JS. The effects of age and liver disease on the disposition and elimination of diazepam in adult man. J Clin Invest. 1975; 55: 347—359. Greenblatt DJ; Sellers EM; Shader RI. Drug therapy: drug disposition in old age. N Engl J Med. 1982; 306: 1081-1088. Grymonpre RE; Mitenko PA; Sitar DS;.AOki FY; Montgomery PR. Drug-associated hospital admissions in older medical patients. J Am Geriatr Soc. 1988; 36(12): 1092-1098. Hackman RM; Wagner EL. The senior gardening and nutrition project: development and transport of a dietary behavior change and health promotion program. J Nutr Educ. 1990; 22(6): 262—270. Hale WE; Perkins LL; May FE; Marks RG; Stewart RB. Symptom prevalence in the elderly: an evaluation of age, sex, disease and medication use. J Am Geriatr Soc. 1979; 34: 333-340. Hallas J; Harvald B; Gram LF; Grodum E; Brosen K; Haghfelt T; Damsbo N. Drug related hospital admissions: the role of definitions and intensity of data collection, and the possibility of prevention. J Intern Med. 1990; 228(2): 83-90. Hammarlund ER; Ostrom JR; Kethley.AJ. The effects of drug counseling and other educational strategies on drug utilization of the elderly. Med Care. 1985; 23: 165-170. Hankin JH. Development of a diet history questionnaire for studies of Older persons. Am J Clin Nutr. 1989; 50: 1121-1127. Hathcock JN. Metabolic mechanisms of drug-nutrient interactions. Federation Proc. 1985; 44(1): 124-129. Hawe P; Higgins G. Can medication education improve the drug compliance of the elderly? evaluation of an in-hospital program. Patient Educ Couns. 1990; 16(2): 151-160. 136 Helling DK; Lemke JH; Semla TP; Wallace RB; Lipson DP; Cornoni-Huntley J. Medication use characteristics in the elderly: the Iowa 65+ rural health study. J Am Geriatr Soc. 1987; 35: 4-12. Hermann JR; Kopel BH; McCrory ML; Kulling FA. Effect of a cooperative extension nutrition and exercise program for older adults on nutrition knowledge, dietary intake, anthropometric measurements, and serum lipids. J Nutr Educ. 1990; 22(6): 271-274. Hodkinson HM. Non-specific presentation of illness. Br Med J. 1973; 4: 94-96. Hosmer DW; Lemeshow S. Applied logistic regression. New York: John Wiley & Sons; 1989. House of Representatives (100th Congress). Hearing before Select Committee on Aging. Planning for an aging America. ; 1987. Hulka BS; Cassel J; Kupper L; Burdette J. Communication, compliance and concordance between physicians and patients with prescribed medications. Am J Pub Health. 1976; 66: 847—853. Institute of Medicine; Division of Health Promotion and Disease Prevention; Committee on a National Research Agenda on Aging. Lonergan, Edmund T., Editor. Extending Life, Enhancing Life: A National Research Agenda on Aging. Washington, D.C.: National Academy Press; 1991. Inui TS; Carter WE; Pecoraro RE; Pearlman RA; Dohna JJ. Variations in patient compliance with common long—term drugs. Med Care. 1980; technical report 2. Ives TJ; Bentz EJ; Gwyther RE. Drug-related admissions to a family medicine inpatient service. Arch Intern Med. 1987; 147(6): 1117-1120. Jeffrey RW; Bjornson-Benson WM; Rosenthal BS; Lindqutis RA; Kurth CL; Johnson SL. Correlates of weight loss and its maintenance over two years of follow-up among middle-aged men. Prev Med. 1984; 13: 155-168. Johnson DW; Johnson RT. Nutrition education: a model for effectiveness, a synthesis of research. J Nutr Educ. 1985; 17(suppl 1): Sl-S44. 137 Joint Commission on Accreditation of Hospitals. Drug-food interaction counseling. JCAH Perspect. 1986; 6: 11. Kamataki T; Maeda K; Shimada M. Age-related alteration in the activities of drug-metabolizing enzymes and contents of sex-specific forms of cytochrome P—450 in liver microsomes from male and female rats. J Pharmacol Exp Ther. 1985; 233: 222-228. Kitani K. Hepatic drug metabolism in the elderly. Hepatology. 1986; 6: 316—319. Klein LE; German PS; Levine DM; Feroli R; Ardery J. Arch Intern Med. 1984; 144: 1185-1188. Klein LE; German PS; McPhee SJ. Aging and its relationship to health knowledge and medication compliance. Gerontologist. 1982; 22: 384-390. Krall EA; Dwyer JT; Coleman KA. Factors influencing accuracy of dietary recall. Nutr Res. 1988; 8: 824-841. Kramer MS; Leventhal JM; Hutchinson TA; Feinstein AR. An alogorithm for the operational assessment of adverse drug reactions. J Am Geriatr Soc. 1979; 242: 623-632. Kuusisto J; Mykkanen L; Pyorala K; Laakso M. NIDDM and its metabolic control predict coronary heart disease in elderly subjects. Diabetes. 1994; 43: 960-967. Lamy PP. Comparative pharmacokinetic changes and drug therapy in an older population. J Am Geriatr Soc. 1982A; 30(11): 811-319. Lamy PP. Effects of diet and nutrition on drug therapy. J Am Geriatr Soc. 1982B; 30(11): 899-8112. Lamy PP. Potential adverse effects of antihypertensive drugs in the elderly. J Hypertens. 1988; 6(suppl 1): 881-885. Landry JA; Smyer MA; Tubman JG; Lago DJ; Roberts J; Simonson W. Validation of two methods of data collection of self-reported medicine use among the elderly. Gerontologist. 1988; 28(5): 672-676. Lewis CW; Frongillo EA; Roe DA. Drug-nutrient interactions in three long-term-care facilities. J Am Diet Assoc. 1995; 95: 309-315. 138 Lundin DV. Medication taking behavior of the elderly: a pilot study. Drug Intell Clin Pharm. 1978; 12: 518-521. Lust JA; Celuch KG; Showers LS. A note on issues concerning the measurement of self-efficacy. J Appl Soc Psychol. 1993; 23(12): 1426-1434. Macdonald ET; Macdonald JB; Phoenix M. Improving drug compliance after hospital discharge. Br Med J. 1977; 2: 618-621. MacLeod SM; Soldin SJ. Determinants of drug disposition in man. Clin Biochem. 1986; 19(2): 67-71. Magaziner J; Cadigan D; Fedder DO; Hebel JR. Medication use and functional decline among community dwelling Older women. J Aging Health. 1989; 4: 470-484. Magnus MH; Roe DA. Computer-assisted instruction on drug-nutrient interactions for long-term caregivers. J Nutr Educ. 1991; 23: 10-17. Matheson DM; Woolcott DM; Matthews AM; Roth V. Evaluation of a theoretical model predicting self-efficacy toward nutrition behaviors in the elderly. J Nutr Educ. 1991; 23(1): 3-9. McAllister RG Jr. Age-related changes in drug handling in man. Am J Cardiol. 1986; 57(5): 59C-62C. McHorney CA; Ware JE; Rogers W; Raczek AB; Lu JFR. The validity and relative precision of MOS short- and long-form health status scales and Dartmouth COOP charts: results from the Medical Outcomes Study. Med Care. 1992; 30(5): M8253-M8265. Meichenbaum D; Turk DC. Facilitating treatment adherence. New York: Plenum Press; 1987. Michigan Governor's Task Force. Prescription drugs and older consumers: report of the governor's task force on prescription drugs and older consumers issued August 10, 1990. Michigan Medicine. 1991; 90(2): 27-31. Michigan Office of Services to the Aging. Michigan needs assessment of the 60 and over population: executive summary and independent variable analysis. Lansing: Michigan Office of Services to the Aging; 1987. 139 Mitenko PA. Drug monitoring in the elderly. Clin Biochem. 1986; 19(2): 145-149. Morris LA; Grossman R; Barkdoll GL; Gordon E; Soviero C. Survey of patient sources of prescription drug information. Am J Pub Health. 1984; 74: 1161-1162. Nolan L; O'Malley K. Prescribing for the elderly: part II-prescribing patterns: differences due to age. J Am Geriatr Soc. 1988; 36: 245-254. Norusis MJ. SPSS/PC+ Advanced Statistics 4.0 Manual. Chicago: SPSS Inc.; 1990. Nygaard TW; Sellers TD; Cook TS; DiMarco JP. Adverse reactions to antiarrhythmic drugs during therapy for ventricular arrhythmias. JAMA. 1986; 256(1): 55-57. O'Leary A. Self-efficacy and health. Behav Res Ther. 1985; 23: 437-451. Ostrom JE; Hammarlund ER; Christensen DB; Plein JB; Kethley AJ. Medication usage in an elderly population. Med Care. 1985; 23: 157-170. Ouslander JA. Drug therapy in the elderly. Ann Int Med. 1981; 95: 711—721. Ozer EM; Bandura A. Mechanisms governing empowerment effects: A.self-efficacy analysis. J Personality Soc Psychol. 1990; 58: 472-486. PharmAssist [software]. N8 Productions, Inc. : Software Marketing Corp.; 1993. Physicians' Desk Reference. Oradell, NJ: Medical Economics Co.; 1995. Pronsky ZM. Food-medication interactions. 8th ed. Pottstown, PA: Food—Medication Interactions; 1993. Rikans LE. Drugs and nutrition in old age. Life Sci. 1986; 39: 1027—1036. Roe DA. Handbook on drug and nutrient interactions. Chicago: American Dietetic Association; 1992. Roe DA. Therapeutic effects of drug-nutrient interactions in the elderly. J Am Diet Assoc. 1985; 85(2): 174-181. 140 Ross FM. Patient compliance-—whose responsibility? Soc Sci Med. 1991; 32(1): 89-94. Rowe JW; Andres R; Tobin JD. Age-adjusted standards for creatinine clearance. Ann Intern Med. 1976; 84: 567-569. Rubin PC; Scott PJW; Reid JL. Prazosin disposition in young and elderly subjects. Br J Clin Pharmacol. 1981; 12: 401-406. Salem SAM; Rajjaybun P; Shepherd AMM; Stevenson IH. Reduced induction of drug metabolism in the elderly. Age Ageing. 1978; 7: 68-73. Schmucker DL. Drug disposition in the elderly: a review of the critical factors. J Am Geriatr Soc. 1984; 32: 144-149. Schoenberger JA; Croog SH; Sudilovsky A; Levine S; Baume RM. Self-reported side effects from antihypertensive drugs: a clinical trial. Am J Hypertens. 1990; 3(2): 123-132. Semla TP; Lemke JH; Helling DK; Wallace RB; Chrischilles EA. Perceived purpose of prescription drugs: the Iowa 65+ rural health study. Drug Intell Clin Pharm. 1991; 25(4): 410-413. Smith CH; Bidlack WR. Dietary concerns associated with the use of medications. J Am Diet Assoc. 1984; 84(8): 901-914. Spagnoli A; Ostino G; Borga AD; D'Ambrosio R; Maggiorotti P; Todisco E; Prattichizzo W; Pia L; Comelli M. Drug compliance and unreported drugs in the elderly. J Am Geriatr Soc. 1989; 37(7): 619-624. SPSS/PC+ [software]. SPSS, Inc. 4.0. Chicago: SPSS, Inc.; 1990. Stanley MA; Maddux JE. Cognitive processes in health enhancement: investigation of a combined protection motivation and self-efficacy model. Basic Appl Soc Psychol. 1986; 7: 101-113. Stern MP; Patterson JK; Haffner SM; Hazuda HP; Mitchell BD. Lack of awareness of and treatment of hyperlidemia in type II diabetes in a community survey. JAMA. 1989; 262: 360-364. 141 Stewart RB; Moore MT; May FE; Marks RG; Hale WE. A longitudinal evaluation of drug use in ambulatory elderly population. J Clin Epidemiol. 1991; 44(2): 1353-1359. Strecher VJ; DeVellis BM; Hecker MH; Rosenstock IM. The role of self-efficacy in achieving health behavior change. Health Educ Quarterly. 1986; 13: 73-91. Strong A; Wolff H; Kinder S; Lubischer A. Drug administration in relation to meals in the institutional setting. Heart Lung. 1991; 20(1): 39-44. Swift CG. Pharmacodynamics: changes in homeostatic mechanisms, receptor and target organ sensitivity in the elderly. Br Med Bull. 1990; 46(1): 36-52. Tregaskis BF; Stevenson IH. Pharmacokinetics in old age. Br Med Bull. 1990; 46(1): 9-21. United States Pharmacopeia Dispensing Information. Rockville, MD: U.S. Pharmacopeial Convention; 1995. Vestal RE. Drug use in the elderly: a review of problems and special considerations. Drugs. 1978; 16: 358-382. Wallace RB. Epidemiology of aging seminar. American College of Preventative Medicine and University of Michigan School of Public Health. 1991; 28th Annual Graduate Summer Session in Epidemiology. Wallace SM; Verbeeck RK. Plasma protein binding of drugs in the elderly. Clin Pharmacokinet. 1987; 12: 41-72. Wandless I; Davie JW. Can drug compliance in the elderly be improved? Br Med J. 1977; 1: 359-361. Ware JE; Kosinski M; Keller SD. SF-36 physical and mental health summary scales: a user's manual. Boston: The Health Institute, New England Medical Center; 1994. Ware JE; Sherbourne CD. The MOS 36-item short-form health survey (SF-36): 1. conceptual framework and item selection. Med Care. 1992; 30(6): 473-483. Ware JE; Snow KK; Kosinski M; Gandek B. SF-36 health survey manual and interpretation guide. Boston: The Health Institute, New England Medical Center; 1993. 142 Weinberg RS; Hughes HH; Critelli JW; England R; Jackson A. Effects of pre-existing and manipulated levels of self-efficacy on weight loss in a self-control program. J Personality. 1984; 18: 351-358. Weinberger M; Samsa GP; Hanlon JT; Schmader K; Doyle ME; Cowper PA; Uttech KM; Cohen HJ; Feussner JR. An evaluation of a brief health status measure in elderly veterans. J Am Geriatr Soc. 1991; 39(7): 691-694. Whittaker JA; Evans DDP. Genetic control of phenylbutazone metabolism in man. Br Med J. 1970; 4: 323-333. Williams L; Kim SK; McMullen EA. Nutrition education to facilitate dietary modification in hypertensive elderly. J Nutr Elderly. 1987; 6(2): 13-30. Williamson J; Chopin JM. Adverse reactions to prescribed drugs in the elderly: a multicentre investigation. Age Ageing. 1980; 9: 73—80. Working Group on Health Education and High Blood Pressure Control. The physician's guide: improving adherence among hypertensive patients. Washington, D.C.: U.S. Department of Health and Human Services (PHS); 1987. Wynne HA; Cope E; Mutch E; Rawlins MD; Woodhouse KW; James OFW. The effect of age upon liver volume and apparent liver blood flow in healthy man. Hepatology. 1989; 9: 297-301. Wynne HH; Mutch E; James OFW; Wright P; Rawlins MD; Moorhouse KW. The effect of age upon the affinity of microsomal mono-oxygenase enzymes for substrate in human liver. Age Ageing. 1988; 17: 401-410. MICHIGAN STATE UNIV. LIBRARIES Illllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll 31293014171940