MICHIGAN STATE UNlVE SITY L a ll! I ll Hill/WWI!Willi/l l 2““ Tu Hi 3 1293 00882 40518! Ill! I“. This is to certify that the thesis entitled ASSOCIATION OF FAMILY HISTORY OF CARDIOVASCULAR DISEASE AS RECORDED ON A FAMILY GENOGRAM AND THE LIPID AND LIPOPROTEIN PROFILE presented by ANNE PATRICIA CURRIE has been accepted towards fulfillment of the requirements for MASTER OF SCIENCE degree in NURSING Major professor JULY 29, 1992 Date 0.7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mlchlgan State Unlverslty PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or betore ode due. DATE DUE DATE DUE DATE DUE II II ' | IL. ll IV— l—j l MSU le An Affirmative Action/Equal Opportunity Institution cMma-nt l _ ,._ ___~___“__ _ __ w— ASSOCIATION OF FAMILY HISTORY OF CARDIOVASCULAR DISEASE AS RECORDED ON A FAMILY GENOGRAM AND THE LIPID AND LIPOPROTEIN PROFILE by Anne Patricia Currie A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN NURSING 1992 ABSTRACT ASSOCIATION OF FAMILY HISTORY OF CARDIOVASCULAR DISEASE AS RECORDED ON A FAMILY GENOGRAM AND THE LIPID AND LIPOPROTEIN PROFILE BY Anne Patricia Currie The usefulness to the primary health care provider of a positive family history of cardiovascular disease (CVD) recorded on a genogram as a predictor of a client's lipid and lipoprotein profile was explored. The sample was comprised of 360 records of medical students participating in a preventive cardiology study. Total family history was not a significant predictor; however, paternal history was significantly correlated with total cholesterol and low-density lipoprotein levels. Female sex was found to the best predictor of high-density lipoprotein levels, whereas male sex and age were found to best predict total cholesterol and low-density lipoprotein levels. Therefore, a primary health care provider can not use family history of CVD alone to determine the need for the lipid and lipoprotein profile. DEDICATED TO MIKEL B. CORDS who taught me through his life the meaning of forgiveness, humor and love. I am forever grateful. Thank You. 111 ACKNOWLEDGEMENTS There are many without whom this thesis would not be an accomplished work. First and foremost, I thank my most loyal and faithful friend Christ, whose love gives me the courage to "Dream The Impossible Dream" and reach for it. Rachel Schiffman, PhD, R.N., my advisor and chair, has been an invaluable resource. Her patience with my ignorance, procrastinations, and halting steps while learning the thesis process, has left me in awe of the educator's role. Thank You. Gabriele Kende, M.S., R.D., who has acquired a special place in my heart, for her unfailing encouragement. She, consistently, found ways to motivate me when I was_stalled out. She listened to me pour out my heart, which then left me room to work. Thank You. Jacqueline Wright, MSN, R.N., a woman with a heart of gold, who has helped me to understand and incorporate this new role. Thank You. Dr. Albert Sparrow, who was the principal investigator of the study where my data came from, iv provided me with the subject matter, humor and guidance which helped me to successfully complete this project. Thank You. Julie Tazzia, a dear friend, who helped me with her support, strength, friendship and assistance on the genogram. Thank You. Then, there are my family and friends in New Jersey and at Saint John's Student Parish who never failed to support and encourage me. They never stopped believing in me and supporting my hopes and dreams. Thank You. Cheryl Farguhar, R.N., my job-share partner, who allowed her schedule to change as I needed the time. Thank You. To all at Greater Lansing Visiting Nurse Services/Visiting Nurse Hospice, who allowed me to be so very flexible. Thank You. To all the others, you know who you are, this would not have been possible without your help. Thank You. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER I - The Problem Introduction Statement of the Problem Research Questions Overview of Chapters CHAPTER II - Conceptual Framework Introduction Definition of Concepts Cardiovascular Disease Lipid and Lipoprotein Profile Family History of CVD on a Genogram Health Belief Model Application of Health Belief Model to the Study Demographic Variables Structural Variables Sociopsychological Variables Summary CHAPTER III - Review of Literature Introduction The Relationship Between Family History of Cardiovascular Disease and the Lipid and Lipoprotein Profile of a Client Age and Sex Summary CHAPTER IV - Methods Overview Research Methods Operational Definitions of the Variables Family History/Genogram Lipid and Lipoprotein Profile vi viii ix GUI-FF H 4 Age/Sex Research Questions Sample Instrumentation Data Collection Data Analysis Methodological Assumptions and Limitations Human Subjects Summary CHAPTER V - Data Analysis Overview Descriptive Data Statistical Analyses of the Research Questions Summary CHAPTER VI - Interpretations, Implications and Recommendations Overview Interpretations Discussion Implications for Advanced Nursing Practice Recommendations for Future Research Summary APPENDICES Appendix A Appendix B Appendix C Appendix D LIST OF REFERENCES References vii 47 47 47 53 57 80 86 87 Table Table Table Table Table LIST OF TABLES Frequencies and Percentages of Incidences of Cardiovascular Diseases as Recorded on the Genogram Means and Standard Deviations of Incidences of Cardiovascular Diseases as Recorded on the Genogram Frequencies and Percentages of Lipids by NIH Groupings Correlations Between Incidences of Cardiovascular Disease and The Lipid and Lipoprotein Profile Multiple Regression Analyses of Age, Sex, Family History of CVD to Predict The Lipid and Lipoprotein Profile viii 49 50 51 54 56 Figure 1 LIST OF FIGURES Health Belief Model ix 18 CHAPTER I The Problem Introduction Cardiovascular disease (CVD) is a predominant cause of mortality and morbidity in modern societies (LaRosa, Becker, & Fitzgerald, 1990; Samuelsson, Wilhelmsen, Andersson, Pennert, & Berglund, 1987). CVD ranks first in terms of Social Security disability, and is second only to all forms of arthritis for limitation of activity and to all forms of cancer combined for total hospital bed days (Lipid Research Clinics Program, 1984). Several factors have been found to be associated with an individuals increased risk of developing CVD, including cigarette smoking, high blood pressure, an elevated blood lipid and lipoprotein profile, sex, age, family history of CVD, obesity, diabetes mellitus, and physical inactivity (Kwiterovich, 1989; Nichols, 1988). The axiom "an ounce of prevention is worth a pound of cure" takes on a more significant meaning when applied to the prevention of CVD in at risk persons 2 through the interventions of a primary health care provider. With the dramatic shift in the major goals of the national health policy to health promotion and disease prevention (Pender, 1987), health care providers should assess a client in relation to the number of risk factors present then, in collaboration with the client, establish interventions which will promote health and prevent disease. In this thesis, the relationship between the lipid and lipoprotein profile, age and sex of an individual, and the family history of CVD as shown on a genogram of incoming first year medical students was explored. Primary health care providers must weigh a client's potential and actual risk for developing cardiovascular disease with the known risk factors which are present in the client or client's family history before considering appropriate steps for intervention. One rationale for the original assessment of the lipid and lipoprotein profile, sex, age and a family history of CVD on a genogram of incoming first year medical students was to trigger an interest in and an awareness of preventive cardiology by students in the health care profession (Eaton et al., 1990). A Family Clinical Nurse Specialist (FCNS) in the role of a primary health care provider and educator 3 must have some knowledge in the field of preventive cardiology and the risk factors involved in the development of premature CVD to initiate health promotion and disease prevention within a practice. This knowledge will facilitate the appropriate linkage between the identification of the risk factors and interventions (Sackett, Haynes, & Tugwell, 1985), thereby hopefully decreasing the probability of an at risk client developing premature CVD. This longitudinal aspect of disease prevention through health promotion is a subject which in and of itself is a separate research topic and not the subject of this paper. A second rationale is to investigate the strength of a family genogram both in identifying the family history of CVD and as a predictor of the lipid and lipoprotein profile in incoming first year medical students. ‘Primary health care providers might be able to base client interventions for health promotion/disease prevention on a positive family history of CVD as evidenced by a genogram, and the lipid and lipoprotein profile which may or may not be abnormal in the client as he/she grows and develops. Statement of the Problem During the collaborative process between the primary health care provider and the client in the prevention of cardiovascular disease and promotion of cardiovascular health, the practitioner must have some knowledge base from which to practice. Part of the knowledge base utilized by the practitioner is the strength of the relationship which exists between the incidence of CVD on a genogram and the probability of a client having an abnormal lipid and lipoprotein profile. The purpose of this thesis was to explore the usefulness to the primary health care provider of a positive family history as a risk factor for a client's potential to prematurely develop a disease. It has been well documented that: 1) cholesterol and triglyceride levels increase with age and total cholesterol increases, on the average, more than 2mg/d1/year during early adulthood (Corrao, Becker, Ockene, 8 Hamilton, 1990; Levy, Wilson, Anderson, & Castelli, 1990; Sempos et al., 1989); 2) men have higher total cholesterol levels that women until age 50 (Corrao et al., 1990; Expert Panel, 1988); and 3) women carry a higher proportion of cholesterol in the form of 4 5 HDL (Corrao et al., 1990; Levy et al., 1989; Sempos et al., 1989). Although, the purpose of this thesis was not specifically to examine the effects of age and sex, they were considered confounding variables. Research Questions The two research questions which formed the problem statement were: 1) Is there a relationship between the reported incidence of cardiovascular disease on the family genogram and the lipid and lipoprotein profile in incoming first year medical students?; and 2) Is the incidence of cardiovascular disease on the genogram a valid predictor of the lipid and lipoprotein profile of these same medical students after accounting for age and sex? It is supposed that a relationship does in fact exist between a positive family history of CVD on the genogram and the lipid and lipoprotein profile indicating the risk of premature CVD in the first year medical students. Secondary data will be analyzed using correlation and regression analyses to describe the existing relationships. Overview of Chapters This thesis is presented in six chapters. Chapter I contains the introduction, background information, statement of the problem, purpose of the thesis, research questions, and an overview of the thesis. The conceptual framework will be presented in Chapter II. In this chapter, the concepts and variables will be defined as they relate to the conceptual model of nursing in advanced practice. Chapter III will be a review of the literature. The research design and methods will be included in Chapter IV, along with the assumptions and limitations of this thesis. Chapter V will contain the data analysis and results. In the concluding chapter, Chapter VI, a summary of the research findings, conclusions, recommendations and nursing implications will be presented. CHAPTER II Conceptual Framework Introduction This chapter contains a description and discussion of the variable definitions as they relate to the conceptual framework for this thesis. The variables defined are: cardiovascular disease, lipid and lipoprotein profile and, family history of CVD on the genogram. These variables and the conceptual framework which is the Health Belief Model (HBM) are discussed in the context of health promotion and disease prevention. A brief discussion of the HBM is presented with an indepth exploration of the concept variables as they relate to the Modifying Factor portion of the HBM. Definition of Concepts The following are the concepts being defined: cardiovascular disease, lipid and lipoprotein profile, and a family history of CVD on the family genogram. The variables are discussed as they relate to screening and disease prevention. Cardiovascular Disease The term cardiovascular disease encompasses many cardiac abnormalities related to heart, blood vessels and circulation. Many of these abnormalities or diseases such as congenital malformations and valve diseases can be placed within the category of CVD. However, they are not related to the lipid and lipoprotein profile, whose abnormalities are the precursors of atherosclerosis. Two terms, ischemic heart disease (IHD) and coronary heart disease (CHD) are frequently utilized to express CVD resulting from narrowing of the coronary arteries and therefore a decreased blood supply. IHD, CHD, and CVD are common terms found in literature discussions referring to heart disease resulting from atherosclerosis. In this thesis, the term CVD will be used when references are made to heart and peripheral vascular diseases resulting from atherosclerosis. Atherosclerosis is an arterial lesion characterized by intimal thickening due to localized accumulations by lipids, known as atheromas (Berkow, 1982). These atheromas consist of a mass of fatty material associated with fibrous connective tissue, very often with secondary deposits of calcium salts and blood products (Price 8 Wilson, 1986). The blood vessels 9 affected are the aorta, its large branches, and medium- sized arteries such as those supplying portions of the extremities, the brain, the heart, and the major internal viscera (Price 8 Wilson, 1986). Despite progressive luminal narrowing and the concurrent loss of vascular responsiveness, clinical manifestations of disease do not appear until the lumen is over 75% obstructed, which can take 20 to 40 years (Price 8 Wilson, 1986). For the purposes of this thesis, the pathophysiological consequence of atherosclerosis is cardiovascular disease. The cardiovascular diseases identified within this thesis in the recorded family history on the genogram of CVD are hypertension, hypercholesterolemia, heart disease, angina, and myocardial infarction. Lipid and Lipoprotein Profile Lipids are carbon and hydrogen containing compounds insoluble in water but soluble in organic solvents (Widmann, 1983). These plasma lipids (cholesterols, triglycerides, phospholipids, and free fatty acids) are derived from exogenous dietary sources and endogenous lipid synthesis (Price 5 Wilson, 1986). Cholesterol and triglycerides are the two 10 lipids of major clinical significance relative to atherogenesis (Price & Wilson, 1986). Since lipids are insoluble in plasma, lipids are bound to proteins as a mechanism for serum transport (Price 5 Wilson, 1986). Lipoproteins are a complex compound of lipids and proteins (Whitney & Cataldo, 1987). This complex bonding produces four major classes of lipoproteins: chylomicrons, very low-density lipoproteins (VLDL), low-density lipoproteins (LDL), and high-density lipoproteins (HDL) (Price & Wilson, 1986). Of the four lipoprotein classes, the chylomicrons and VLDL are richest in triglyceride, whereas LDL and HDL contain the largest portion of cholesterol (Price 5 Wilson, 1986). The presence within the body of lipids and lipoproteins in and of itself is not abnormal. The abnormality is when the levels within the bloodstream are such that they would facilitate the formation of atheromas the precursors for the development of cardiovascular disease. For the purposes of this thesis the lipid and lipoprotein profile of total cholesterol (TC), HDL and LDL levels were classified as risk factors according to the National Institute of Health's (1987) definition of abnormal within the National Cholesterol Education Program. 11 Family History of CVD on a Genogram The pathophysiological consequences of atherosclerosis clearly demonstrate the need for the primary health care provider to gather an adequate and accurate family history of CVD going back as many generations as possible when determining the need for an individual client to have a lipid and lipoprotein profile ordered. Health care providers dealing with high-risk families consistently find that most high- risk persons in these disease prone families are not receiving the benefit of established risk reduction interventions early enough to effectively prevent or delay serious consequences from the disease reinforcing the importance of promotion and prevention by the provider (Williams et al., 1988). Since heredity is one of the most important factors in the etiology of atherosclerosis (Farcot, Hashimoto, Meerbaum, & Corday, 1977; Griffin, Christoffel, Binns, McGuire, & The Pediatric Practice Research Group, 1989), family, for purposes of this thesis, was considered to be those persons on the genogram who were identified as genetically related to the subjects in the cases. That is not to say that environmental factors have no bearing on the development of CVD, and that family members who enter 12 into a high-risk family through means other than birth will not develop CVD. However, for the sake of brevity and clarity, environmental factors are not variables being explored in this thesis. An important first step in investigating a client's familial health history is to look for the classic mendelian patterns of inheritance allowing the provider to recognize the predominant genetic causation for the presence of a disease in question, in this case CVD, or its potential for occurence (Gordon, 1972). A family history is often gathered using a genogram which is a multi-generational pictorial representation depiciting sex, birth and death dates, diseases, causes of mortality, and relationships between individuals (McGoldrick & Gerson, 1980; Jolly, Froom, & Rosen, 1980). For the purposes of this thesis, family history on a genogram were the recorded multi-generational incidences of cardiovascular disease. If there are genetic risk factors present for CVD then the possibility exists that the presenting client may have a lipid and lipoprotein profile at such levels as to predispose the client to the development of atheromas, the precursors of CVD. For the primary health care provider to assist in the prevention of CVD and the 13 promotion of cardiovascular health in the presenting client, a complete family history discerning the possible presence of cardiovascular diseases within the family health/disease history must be acquired. Health Belief Model The Health Belief Model (HBM) was developed based on a theory explaining preventive health behaviors (Rosenstock, 1974). Behavioral scientists and health care providers saw an increasing need to understand why and under what conditions people take action to prevent, detect, and treat diseases (Mikhail, 1981). It was postulated that such a theory would have to deal with the behavior of individuals who were not currently suffering from a disabling disease (Rosenstock, 1974). The orientation of the theory encompasses the avoidance of disease and the potential role of barriers to accepting health services (Rosenstock, 1974). The HBM is a psychosocial formulation developed to explain health-related behavior at the level of individual decision-making (Mikhail, 1981). The HBM is viewed as a potentially useful tool which the primary health care provider can use to predict those individuals who would or would not use preventive measures and to suggest 14 interventions which might increase compliance in resistant individuals to engage in preventive or health—protecting behaviors (Pender, 1987). The researchers who postulated the Health Belief Model (HBM), were influenced considerably by the social psychological theory of Kurt Lewin (Rosenstock, 1974; Mikhail, 1981; Pender, 1987). The implicit conception following Lewin was of an individual existing in a life space composed of regions some of which were positively valued (positive valence), others of which were negatively valued (negative valence), and still others of which were relatively neutral (Rosenstock, 1974). Diseases were considered to be regions of negative valence which could be expected to exert a force moving a person away from the region, unless doing so would require him/her to enter an even greater region of negative valence (Rosenstock, 1974). An individual's daily activities were therefore viewed as a process of being pushed by positive forces (Rosenstock, 1974). It can thus be said that preventive behaviors are strategies for avoiding the negatively valued regions of illness and disease (Pender, 1987). For a person to engage in the appropriate behavior needed to avoid illness and disease they would need to believe the following: (1) they were personally susceptible to the 15 illness or disease, (2) that the occurrence of the disease would affect some component of their lives with moderate severity, and (3) that taking action would in fact be beneficial by reducing their susceptibility to the condition or if the disease occurred, by reducing its severity, and that it would not entail overcoming important psychological barriers such as cost, convenience, pain, embarrassment (Rosenstock, 1974). With respect to taking a test for early detection of a disease, the same factors were deemed necessary, but in addition there was also the requirement that the individual believe that they could have the disease even in the absence of symptoms (Hochbaum, 1958; Rosenstock, 1960; Rosenstock, 1974). The HBM is phenomenological in orientation and is anchored in the belief that people can only act on what they believe to exist even though this viewpoint may not match professional viewpoints (Mikhail, 1981). Given the phenomenological nature of the HBM and the fact that the components of the model are dealing with individual perceptions, it becomes apparent that without modifications, it is an inappropriate paradigm for health-promoting behavior. Health promotion focuses on movement toward a positively valenced state of enhanced health and 16 well-being (Pender, 1987). Whereas the negatively valenced states of illness and disease, while relevant to motivation for health-protecting (disease preventing) behavior, appear to have minimal motivational significance for health promoting behavior (Pender, 1987). However, because of the perceptions of the individual in relation to susceptibility and seriousness of a disease, benefits and barriers to action, the HBM can be very appropriate for disease prevention (Pender, 1987). This model may also be able to explain the primary health care provider's preventive actions, that is, using preventive services (e.g., diagnostic tools, genogram and laboratory testing, blood lipid screening) in a provider-consumer context (Pender, 1987). It is not the intention or purpose of this thesis to modify the HBM in totum thereby insuring a "perfect fit" of all the variables and concepts and their application in the context of health promotion/disease prevention. That modification is out of the scope of this work. Application of Health Belief Model to the Study The following section contains a discussion of the definitions of the concepts within the first box identified as Modifying Factors in Figure 1 which is a depiction of the HBM as it relates to this study. As mentioned earlier the use of the entire Health Belief Model without modifications would not be applicable for health promotion/disease prevention as a concept together. The variables and concepts have a unidirectional flow. Given this unidirectional flow, the difficulty of utilizing this model to depict both a positively and negatively valenced motivation for behavior poses a contradiction which can not be successfully defended by this model. This is why the HBM has been more successfully utilized depicting disease preventing (health protecting) behaviors by an individual. However, the variables of this thesis (cardiovascular disease, lipid and lipoprotein profiles and the family history of CVD on a genogram), along with the confounding variables of age and sex can be 17 .xepafl ..~u no unseen Iona caboose. 83...... 5.8.. 25.3.8... 3 38.3... 8 :33: I... 5.3.... I: . 2...... 8.x...- ecfieues. 3 3:33.02 35.: 3 .332: $58 3 32.:— 3...% 3 53.93.. EB. 23.8; 33:53— 823 32.. 3.2—; 2:363. 8.3:. one: 5.3.3. 8 839 L 63.2 5.3: , — :x: 0385 - 8 e335 .3 3.1939 on 94.5.85 voucoEEooez .3 access-om .3383.— 1 5...: a .3523 r .82.... 3...»...— 90 8.85 , a - a 3 asset—33m .8333.— Il .9. i 9.2. easel . cote- e>_.:e>2a .8»: use 88 .839 38. .3:- o. ghee voZooaem .322: 831...... 38332338» .Ieuuoceo e so go no Page; and-em 3...... A . 3:8.— ..«euouaoaS o..- 33.. eeeeeqa neusoee>ouuueu :38. 25:23.... 8315.; 33.25% 3 53:3 .8333.— se- e04 53...! E 7.9.84. .16 Door—4&2: . - mZOEflQ—mm Aida—>57: g gun: a ink 82 9—05.05.— @2230: 19 found within the Modifying Factor variables of the demographic variables, structural variables, and sociopsychologic variables. The Family Clinical Nurse Specialist (FCNS) during an assessment for the possibility of the client's developing the precursor of CVD will assess the interrelationship between the Modifying Factor variables. The client characteristics which used to assess the client's risk status are the demographic variables or age and sex and the structural variables of CVD, lipid and lipoprotein profile, and family history of CVD on the genogram. Demographic Variables Age and sex are some of the variables listed within this category. These variables define objective characteristics of an individual. The confounding and non-modifiable variables of this thesis, age and sex of the sample subjects have been mentioned as having a potential to affect the predisposition of an individual for developing CVD are included within the demographic variables of the HBM. 20 Structural Variables Two such variables presumed by the model to influence preventive behavior include knowledge about the target disease and prior contact with it (Pender, 1987). It is within the structural variables that CVD, the lipid and lipoprotein profile and family history of CVD on a genogram can be found. Although it can be argued that the lipid profile would be more a part of the perceived threat of disease as found within the Health Belief Model. It can also be argued that knowledge about one's lipid profile would allow a person to intellectually assess what is known regarding the probability of developing CVD based on the causation of this disease and that knowledge with or without a sense of perceived threat would motivate a persons preventive behavior. People learn that cardiovascular disease and a lipid and lipoprotein profile may result in CVD from books, primary health care providers, advertising, magazines, college courses, knowing one's own lipid and lipoprotein profile levels as a predisposition for developing CVD, and prior contact. Prior contact, in this case, is knowledge of a family history of CVD. 21 Sociopsychological Variables Social pressure or social influence appear to play a role in stimulating appropriate health actions even when low levels of individual motivation exist (Pender, 1987). Although sociopsychological variables, are not specifically studied within this thesis, it may be important information about an individual's motivation for action and can be ascertained by the primary health care provider of the client within the scope of health promotion/disease prevention. Summary In summary, the HBM is one in which a Family Clinical Nurse Specialist, as a primary health care provider can integrate a disease preventive practice with health promotion. Concepts and variables relating to a client's state of health are assessed by the FCNS within the Modifying Factors of the Health Belief Model. CHAPTER III Review of Literature Introduction This chapter contains discussions in the literature of the relationship that exists between the family history of cardiovascular disease on a genogram with the lipid and lipoprotein profile and the presence of CVD on the family genogram as a predictor of the lipid and lipoprotein profile. The concepts will be presented as they are found in the context of relationships or predictors within the family history of CVD. The Relationship Between Family History of Cardiovascular Disease and the Lipid Lipoprotein Profile of a Client Identification of client risk factors and prevention within the realm of CVD more parallels the purposes of this thesis. One area where client risk 22 23 factors can be assessed is within the family health/disease history as noted on a family genogram. There are numerous investigated and recorded risk factors, which when present in a client, a client's history or a client's family history, are significant indicators to the primary health care provider that CVD may be a potential problem. One risk factor found in the literature which is also a concept for this thesis is the presence of a positive genetic family history of CVD as evidenced by a genogram. There are three studies which contained discussions relating to the aspects of this thesis. Between 1981-1982, The Bogalusa Heart Study (Shear, Webber, Freedman, Srinivasan, & Berenson, 1985) investigated the relationship between parental history of vascular disease and the presence of CVD risk factors in children. The ages of 3,312 children, who were assessed, ranged between 5 and 17. The following outcome variables were investigated: systolic and diastolic blood pressure, total serum cholesterol, triglycerides, VLDL, LDL, and HDL (all of these factors were adjusted for age and sex). Univariate and multivariate analyses were performed. The Bogalusa study investigated only the relationship between parents and their children not grandparents. Serum 24 lipid/lipoprotein profile, as evidenced by the rise in the mean LDL level, increased in children based on parental history of myocardial infarction, stroke or diabetes mellitus. A study was conducted to evaluate the efficacy of family history factors as screening criteria for childhood hypercholesterolemia (Griffin et al., 1989). One thousand and five healthy prepubertal children between ages 2-13 years were assessed. Their parents were asked to complete a parental and grandparental history of cardiovascular risk factors (diabetes, hypercholesterolemia, hypertension, gout, obesity) and atherosclerotic complications (myocardial infarction, angina, sudden death, stroke and peripheral vascular disease) before the age of 55 years. Unknown information was considered a negative history. Maternal and paternal histories of hypercholesterolemia were significantly associated with elevated LDL in children (odds ratio = 7.3 for maternal and 2.9 for paternal), but had extremely low sensitivities and modest positive predictability as a screening tool for elevated LDL levels in children (Griffin et al., 1989). Grandparental histories of sudden death, peripheral vascular disease, and gout were associated with elevated LDL levels, but 25 sensitivities and positive predictive values for all of these factors were not significant (Griffin et al., 1989). Family history factors most commonly utilized as criteria for cholesterol screening in children failed to identify half of all children whose LDL levels were elevated and did not selectively identify the most severly affected children. Although both the Bogalusa Heart Study (Shear et al., 1985) and the study conducted by Griffin et a1. (1989) revealed some connection between a family history of CVD as defined within this thesis and an abnormal serum LDL level in children, the family history variables were not gathered using a genogram and were different for both studies. Age and sex were controlled for in only the Bogalusa study (Shear et al., 1985). The Bogalusa study (Shear et al., 1985) looked at blood pressure and heart attacks of the parents, while the study by Griffin et a1. (1989), investigated a family history of CVD before age 55, heart disease, hypercholesterolemia, and angina. The results of neither study had enough statistical significance to warrant using the family health history alone for determination of a client's risk of having an abnormal lipid and lipoprotein profile levels. 26 A genetic-epidemiologic study was undertaken using a white Colorado population of 207 patients, ages 35- 54, who had experienced a myocardial infarction before the age of 55 years (Nora, Lortscher, Spangler, Nora, & Kimberling, 1980). The highest significant risk ratio (10.4) was found to be associated with a family history showing a first degree relative who had an onset of CVD before the age of 55 and the next highest risk ratio (7.1) was for CVD was found in a first degree relative before the age of 65 (Nora et al., 1980). Family history in first and second degree relatives of CVD, heart attacks, hypertension, and heart disease were assessed as were the subjects' fasting cholesterol levels (controlling for age, sex and race). A statistically significant univariate relationship between a family history of CVD and cholesterol levels (P,.0001, r not given) but family history did not have any statistical significance for predicting elevated cholesterol levels (Nora et al., 1980). Cholesterol levels were not found to be indicative of risk. However, the levels used to indicate abnormal values were higher than those currently being recommended by the National Institute of Health (1987). This study adds support to the relationship between family history of CVD and risk of developing CVD. It 27 does not, however, support the premise of this thesis that a family history of CVD will put a client at risk for an abnormal lipid and lipoprotein profile. If the cholesterol levels in the study by Nora et al. (1980) were lowered to match those recommended by the NIH, it is unknown if the results would then support the premise of this thesis that a family history of CVD will put a client at risk for an abnormal lipid and lipoprotein profile. Neither was a genogram utilized to gather the information about the family health history. In the Tromso Heart Study, a population survey for risk factors associated with CVD among 6595 men aged 20-49 years, a family history of myocardial infarction was obtained by interview (Forde & Thelle, 1977). A high degree of concurrence (78%) was found between the reported and confirmed diagnosis of myocardial infarction (Forde & Thelle, 1977). Subjects who had evidence of a previous MI had a significantly higher frequency of first degree relatives with an MI compared to healthy men of the same age (Forde & Thelle, 1977). The mean serum cholesterol concentration was found to be slightly elevated when the MI was suffered before age 50 by a first degree relative of either sex (Forde & Thelle, 1977). There was no criteria recorded by 28 which elevation of serum cholesterol was determined. Since this study was conducted prior to the National Institute of Health (NIH) (1987) recommendations of serum cholesterol levels, the assumption was made that the acceptable serum cholesterol level is greater than that given by the NIH (1987). The Tromso Heart Study (Forde & Thelle, 1977) discerned that slight elevations in serum cholesterol contributed only a very small extent to the increased risk of subjects with a positive family history (Forde & Thelle, 1977), and that the risk increased slightly as the age of the subject increased. Subjects with a negative family history of an MI had significantly lower serum cholesterol compared to the total group of participants with a positive family history of MI (Forde 5 Thelle, 1977). The inverse of this negative family history and lower serum cholesterol relationship may in fact indicate the strength between a positive family history and higher serum cholesterol levels which the Tromso Heart Study failed to support. The Tromso Heart Study (Forde 8 Thelle, 1977), investigated only the family history of first degree relatives known to have had an MI, angina and heart disease. The variables used in this present study, hypertension and hypercholesterolemia were not 29 included, neither were second degree relatives (grandparents). A genogram was not utilized to gather the family health history. Discerning the strength in the ability of the incidence of a positive family history of CVD on the genogram to be a predictor of the lipid and lipoprotein profile is another purpose of the present study. Using the knowledge of this relationship, the primary health care provider can decide if serum testing is necessary. Based on this will be the decision for further interventions. Williams et a1 (1988) reported on the use of a Health Family Tree as a means of identifying high risk persons within families who are prone to CVD, strokes and common familial cancers for the purposes of preventing or delaying the serious consequences of these diseases. The purpose of the study by Williams et al. (1988) was to use the family tree to assess persons considered to be at risk for certain diseases based on family history. The "Health Family Tree" is a medical family history questionnaire that collects detailed disease and risk factor information of siblings, parents, aunts, uncles and grandparents (as does a family health/disease genogram). High school students enrolled in required health education classes 30 in selected school districts in Texas and Utah were given the questionnaires to take home for 1 week, after which 24,332 "trees" were completed by parents and students (Williams et al., 1988). Risk factors of CVD (hypertension, hypercholesterolemia, and myocardial infarction) were assessed for the students families and their parents' families. High-risk families were selected as those with a family history score (FHS) 3 1.0 who had at least 2 relatives with early CVD totaling 1,796 (7.5%) of the student families who completed the Health Family Tree (Williams et al., 1988). For CVD, high-risk families also had to have at least one early disease case (diagnosis before age 55) (Williams et al., 1988). These families were referred to a physician for further evaluation and screening. This study did not follow through to discern the predictive value of this type of screening. No attempt was made to discern if the student with a positive family history of CVD did in fact have the lipid and lipoprotein profile which put them at risk for developing CVD. Of the literature available, none contained any indepth discussions about the subjects involved in the original study from which the sample for this present study was drawn. Most of the studies in the literature 31 were conducted with participants who were children, adolescents or older adults. The differences in the ages of the samples means that this present study is not and can not be a replicated study. Children may not yet be exhibiting any changes in the lipid and lipoprotein profile, regardless of their parental history. The samples of this present study may be found to be old enough as young adults to exhibit changes in the lipid and lipoprotein profile based on parental history. The existing literature does establish the relationship between a positive familial history of CVD and an elevation of lipid and/or lipoprotein profile. For screening purposes, some of the literature validated the predictive value in terms of a positive family history of CVD and an abnormal lipid and lipoprotein profile. Age and Sex Age and sex are the confounding variables for this thesis. They are considered to be variables that apart from a positive family history of cardiovascular disease can and do affect the plasma lipid and lipoprotein profile in a client. 32 Levy et a1. (1990) discussed insights from the Framingham Heart Study on age and sex. In men between the ages of 30 and 62, the incidence of CVD increases in a linear fashion with increasing age (Levy et al., 1990). Whereas women were at a relatively low risk until menopause, after which the risk of CVD accelerated, but always remains lower than the risk in men (Levy et al., 1990). At any given level of risk, women lagged behind men by 15 to 20 years (Levy et al., 1990). In an offspring study of the Framingham Heart Study (Wilson, Christiansen, Anderson, Kannel, 1989), a median lipid value showed that cholesterol levels rise with age for both men and women. Total cholesterol levels are lower in women through the fifth decade of life, are similar in both sexes in the sixth decade of life and are higher in women in the seventh decade of life (Wilson et al., 1989). HDL levels decline slightly with age in men but are similar across the age ranges studied in women; whereas LDL levels rise with age in both sexes. The two offshoot studies from the Framingham Heart Study (Levy et al., 1990, Wilson et al., 1989) actually approach the confounding variations sex from different perspectives. Levy et a1. (1990) discussed the effect 33 age and sex have upon the risk of a client developing CVD without discussing the effect age and sex may have on the lipid and lipoprotein profile. On the other hand, Wilson et al. (1990) discussed age and sex as they effect a client's lipid and lipoprotein profile; which is the way age and sex were handled in the present study. Summary Since CVD is the number one cause of morbidity and mortality in the United States, it has been and will remain a subject under intense investigation. With understanding and knowledge of the disease process comes the hope that one day, CVD will be preventable. Until then, high risk groups will be defined and attempts will be made to educate these groups to promote health and prevent this disease. There was no study which covered the topic of this thesis perfectly so that this thesis would be a replicated study. However, most of the studies reflected a general trend of risk factors for the development of CVD especially a positive family history of CVD and the lipid and lipoprotein profile which coincide with the concepts of this thesis. 34 In the next chapter, Chapter IV, an operational definition will be presented. The statistical methods utilized to analyze the data will be discussed. CHAPTER IV Methods Overview Chapter IV is a description of the research methods and procedures utilized to assess the data acquired for this thesis. Also found within this chapter, are the operational definitions of the variables, the research questions, definition of the sample, along with the instrumentation utilized to collect the data and the methods and statistical tests to analyze the data. The purpose then of this chapter is to operationally define the variables, describe the research methods, and procedures for collecting and analyzing the pertinent data for this thesis. Research Methods The design of this thesis was a non-experimental, descriptive, correlational study. The variables, family history of CVD on a genogram, lipid and 35 36 lipoprotein profile, age and sex, were analyzed as secondary data gleened from original data procured during the first three years (1980-1982) of the Preventive Cardiology Academic Award (PHS grant HL00613-01, Dr. Albert Sparrow, principal investigator). The thesis method was descriptive and correlational because the aim was to describe the existing relationships among the variables, while exerting no control over the independent variable, rather than infer cause-and-effect relationships (Polit & Hungler, 1987). In 1978 the National Heart, Lung, and Blood Institute established the Preventive Cardiology Academic Award (PCAA) to emphasize the teaching of preventive cardiology in U.S. medical schools (Eaton et al., 1990). The data and concepts for this thesis were a subset of this original PCAA program. The original study was conducted yearly with first year medical students during pre-matriculation week starting in September of 1980 through September of 1989 except for 1986 when no data was gathered. The student participants in the PCAA program filled out questionnaires, including the genograms, had fasting blood lipids analyzed for a lipid and lipoprotein profile, took a physical stress test, had vital signs 37 assessed as well as triceps skin-fold caliper test, height and weight, and in addition a personality test was administered. Only the genogram and blood work results are used in this study. Operational Definitions of the Variables Of the variables in this study, family history of CVD on the genogram was the independent variable, while the dependent variable was lipid and lipoprotein profile of the first year medical student. Age and sex were confounding variables for the lipid and lipoprotein profile in the sample subjects. Family History/Genogram A family health history was obtained via a genogram (see Appendix A) which was numerically coded with 1 representing each incidence of a family history of CVD. An incidence of cardiovascular disease was presumed if any of the following diseases were present: angina, heart disease, hypercholesterolemia, hypertension, and myocardial infarction. No previously utilized and tested method for quantitatively coding a genogram was located. Therefore each piece of data was 38 assigned a number, starting with line A through line D on the genogram. The persons coded who are included within the data analysis for this study are the sample subject, birth parents, maternal and paternal birth grandparents. Although siblings, great-grandparents, aunts, and uncles, were available in the model, sufficient data was not collected initially to substantiate the use of these persons within the data to be analyzed for this thesis. A score was assigned to each subject and family members recorded on the genogram based on their relationship to the subject, their sex and any notation of CVD. Age was discerned for the identified patient, who was the sample subject, using the date of birth. Data for the age of parents and grandparents were not reported accurately enough to be utilized for data analysis without further clarification, which was impossible because of the secondary nature of the data. For an indepth view of the genogram coding scheme, see Appendix B. A summative score, for each subject's family genogram was obtained from recorded incidences of CVD and closeness of the relationship of the recorded family member with CVD to the sample subject. 39 Lipid and Lipoprotein Profile The final variable to be operationalized was the lipid and lipoprotein profile. The lipid and lipoprotein profile contains multiple levels which are the Total Cholesterol (TC), High-density lipoproteins (HDL), and Low-density lipoproteins (LDL). The blood samples for the lipid and lipoprotein profile were collected during the primary data collection from the sample subjects, after a 12 hour fast from the night before, during the same time frame as the genogram were completed. The results of the lipid and lipoprotein profile testing were found documented as actual blood levels located within the data stored on the computer disc labelled Med. Stud. Data I for the PCAA. Age/Sex Age and sex are those of the subject's sampled in the primary data gathering. Age is the self-reported age the sample subject gave to the recording partner at the time of completing the health history on the genogram. Sex is the self-reported identification the sample subject gave to the recording partner at the time of completing the health history on the genogram. This data was obtained from line A on the genogram 40 utilizing the specified symbol identifying the sample subject (Appendix A). Research Questions The research questions around which the concepts were formulated and the data collected for this thesis were as follows: 1) Is there a relationship between the reported incidence of CVD on the family genogram and the lipid and lipoprotein profile of the incoming first year medical students; and 2) Is the incidence of cardiovascular disease on the genogram a valid predictor of the lipid and lipoprotein profile of these same medical students after accounting for age and sex? Sample The sample was selected from 481 records of archival data of the PCAA Program from the years of 1980 through 1982. The cases selected for this thesis were 360 of the most complete records of genograms and the corresponding lipid and lipoprotein profiles of the incoming first year medical students in the Colleges of Human and Osteopathic Medicine at Michigan State 41 University during the fall semesters from the first three years of the study in 1980, 1981, and 1982. Instrumentation The blood samples were analyzed by The Lipid Reference Laboratory in Iowa. The internal and external quality control to ensure standardization of lipid and lipoprotein profile results was that used by the Lipid Research Clinics Program (of which The Lipid Reference Laboratory in Iowa was a part) in the Manual of Laboratory Operations. An identical system for quality control and surveillance was used in each laboratory in order to stabilize performance over the entire project, and to assure that each laboratory would obtain results comparable to those found in other laboratories (National Institute of Health, 1974). The genogram is an untested instrument in terms of statistical analyses. A search of the literature revealed no statistical data using the genogram, coding it, testing it's reliability and validity either externally or internally. Data Collection The data utilized for this thesis were archival data from 1980, 1981 and 1982. Specifically, the lipid and lipoprotein profile data were found recorded on computer disk. The data were originally obtained from fasting serum testing of each incoming freshman medical student who consented to participate in this activity of pre-matriculation orientation week. The records containing the genogram surveys were located in file cabinets and boxes. The results of the lipid and lipoprotein profile were matched with the coded genograms by a coded number assigned to each when the consent form was signed. The genograms were coded specifically for the purpose of the secondary data being utilized within this study using the schema identified under operational definitions and entered on a computer word processing program. Data Analysis Frequency, percentages, means and standard deviation were calculated for each variable. The lipid and lipoprotein profile was described by classifications according to National Institute of 42 43 Health (NIH) (1987). The first research question was analyzed using the Pearson product moment correlational coefficient between the family history of CVD score on the genogram and the three lipid and lipoprotein levels (TC, HDL, and LDL). This computation of the correlational coefficient provides a single measure of the degree of relationship between two variables, but this measure, by itself, may not provide a complete picture of the relationship between the variables (Wood, 1981). The second research question was addressed using regression analyses. A multiple linear regression can assist with the development of a regression equation by providing the best prediction possible, given the correlations among several variables (Polit & Hungler, 1987). Age and sex were entered first in order to control for their effect, followed by the family history of CVD. The criterion variables were the three lipid and lipoprotein profile results. Methodological Assumptions and Limitations Given the fact that the data utilized were archival, there were more assumptions and limitations than if the researcher had control over the initial 44 collection of the data from the subjects. The first assumption was that the data in the 360 archival records utilized within this study were recorded accurately following the instructions (Appendix A) for data collection, thereby enabling accuracy in the data to be statistically analyzed (eg. fasting, was done by each participant, from the night before lab testing for lipid and lipoprotein profiles, following directions for recording information in the genogram). The second assumption was that the recorded data of lipid and lipoprotein were accurate and equivalent and that the laboratory followed the standard operating procedures for testing, recording and reporting of the lipid and lipoprotein profile. The third assumption was that the method of coding the genogram was reliable and valid for discerning statistical information. The major limitation for this thesis was that the initial sample was limited to incoming medical students. Therefore, there cannot be generalizability of any findings to other populations. Another limitation was that other factors, such as race, diet, lifestyle, etc. may alter the outcome of the relationship between a positive history of CVD on the family genogram and the recorded lipid and lipoprotein profiles of the same subjects, were not studied. A 45 third limitation weakening the data base was the need to delete 121 cases due to the lack of information recorded on the genogram. This deletion of cases from the data base may have resulted in a change in the correlational relationship of the lipid and lipoprotein profile and family history of cardiovascular disease. Human Subjects The original study was approved by University Committee on Research Involving Human Subjects. The subjects voluntarily signed an informed consent (Appendix C) for participation in the original study. To insure confidentiality, each student was assigned a number which was used to identify the data. Approval to conduct the secondary data analysis was obtained from the University Committee on Research Involving Human Subjects (Appendix D). No attempt was made in the present study to link the coded data with individuals. Summary A discussion of the research methods and operational definitions of the concepts was presented 46 within this chapter. Also, found was an explanation of how the data was collected, the instruments utilized, the assumptions and limitations of this design. The data is analyzed in Chapter V utilizing the tests mentioned in this chapter. CHAPTER V Data Analysis Overview In this chapter the sample subjects are described and the relationship of the lipid and lipoprotein profile levels with/to the family history of cardiovascular disease recorded on a genogram, and the predictability of the relationship with the subjects lipid and lipoprotein profile levels is presented. The reliability of the genogram as an assessment tool is discussed.' Descriptive Data There were 481 records from the incoming first year medical students over the time ranging from 1980 through 1982 from which 360 records contained enough information on the genogram and the lipid and lipoprotein profile to statistically analyze. The 121 cases removed from the data base prior to analysis were the result of incomplete documentation. These cases 47 48 contained not enough or no information to enable them to be analyzed. Of the 360 cases represented, 209 were males (58%) and 149 were females (42%), with 2 cases having missing data on the variable. The ages ranged from 19 to 42 with 7 cases missing data for the variable. The mean age was 25.48 years with a standard deviation of 4.08. Recorded incidence of angina, hypertension, heart disease, hypercholesterolemia, and myocardial infarction for parents and grandparents is presented in Table 1. Hypertension and myocardial infarction were the two most frequently recorded cardiovascular diseases overall. Hypertension was found most frequently recorded in fathers, mothers, paternal and maternal grandmothers. Myocardial infarction was most frequently recorded in paternal and maternal grandfathers. Hypertension was mainly found in females, whereas, myocardial infarction was found most 4 often in males. The means and standard deviations for the recorded incidences of cardiovascular disease on the family genogram, include fathers, mothers, and grandparents (individually and collectively by relationships) are presented in Table 2. The means and standard deviations for the recorded number on incidences of CVD 49 Table l Frequencies and Percentages of Incidences of Cardiovascular Diseases as Recorded on the Genogram Frequency Percent Paternal Angina 14 3.9 Heart Disease 18 5.0 Hypertension 80 22.2 Hypercholesterolemia 28 7.8 Myocardial Infarction 30 8.3 Paternal Grandfather Angina 22 6.1 Heart Disease 38 10.6 Hypertension 39 10.8 Hypercholesterolemia 5 1.4 Myocardial Infarction 89 24.7 Paternal Grandmother Angina 16 4.4 Heart Disease 26 7.2 Hypertension 78 21.7 Hypercholesterolemia 6 1.7 Myocardial Infarction 29 8.1 Maternal Angina 8 2.2 Heart Disease 12 3.3 Hypertension 74 20.6 Hypercholesterolemia 7 1.7 Myocardial Infarction 6 1.7 Maternal Grandfather Angina 13 3.6 Heart Disease 42 11.7 Hypertension 28 7.8 Hypercholesterolemia 7 1.9 Myocardial Infarction 70 19.4 Maternal Grandmother Angina 7 1.9 Heart Disease 26 7.2 Hypertension 54 15.0 Hypercholesterolemia 9 2.5 Myocardial Infarction 35 9.7 50 Table 2 Means and Standard Deviations of Incidences of Cardiovascular Diseases as Recorded on the Genogram Relationship Mean S.D. Paternal . .47 .81 Maternal .30 .61 Paternal Grandfather .54 .73 Paternal Grandmother .53 .69 Maternal Grandfather .44 .74 Maternal Grandmother .36 .65 Both Parents .77 1.00 All Grandparents 1.72 1.72 Total Paternal Family 1.44 1.39 Total Maternal Family 1.11 1.25 Total Family 1.52 .69 51 are consistently higher in all males and the paternal family and all grandparents, as seen in Table 2. The mean total cholesterol level for the sample was 176.72 (S.D. 32.35). The mean low-density lipoprotein level for the sample was 107.58 (S.D. 28.47). While the mean high-density lipoprotein level was 55.00 (S.D. 14.30). As can be seen in Table 3, the lipid and lipoprotein profile were categorized by the standards of the National Institute of Health (NIH) in a panel report by National Cholesterol Education Program 1987. The sample means and the frequencies in Table 3 indicate that the majority of the sample subjects' lipid and lipoprotein profile levels were within the desirable limits set forth by the NIH. 52 Table 3 Frequencies and Percentages of Lipids by NIH Groupings Variable NIH Grouping Freq. % TC <200 mg/dl 273 75.8 200-239 mg/dl 64 17.8 240 or > mg/dl 14 3.9 missing 9 2.5 LDL <129 mg/dl 276 76.7 130-159 mg/dl 57 15.8 160 or > mg/dl 12 3.3 missing 15 4.2 HDL <35 mg/dl 19 5.3 35 or > mg/dl 327 90.8 missing 14 3.9 Note. TC=total cholesterol <200 mg/dL Desirable Blood Cholesterol 200-239 mg/dL Borderline-High Blood Cholesterol >240 mg/dL High Blood Cholesterol LDL=low-density lipoprotein <130 mg/dL Desirable LDL-Cholesterol 130-159 mg/dL Borderline-High-risk LDL- Cholesterol >160 mg/dL High-Risk LDL-Cholesterol HDL=high-density lipoprotein <35 mg/dl Risk Factor for HDL-Cholesterol NIH Classification of TC, LDL, and HDL Cholesterol Levels as recommended by the NIH (1987). Statistical Analyses of the Research Questions The data from the first research question concerning a relationship between the reported incidence of cardiovascular disease on the family genogram and the lipid and lipoprotein profile in incoming first year medical students was statistically analyzed by Pearson correlations between incidences of CVD on a genogram and the lipid and lipoprotein profile (Table 4). The magnitude of the correlation indicates the degree of the relationship which exists between two variables (Wood, 1981). As can be seen in Table 4, positive family history of CVD in the paternal blood line (although low) significantly correlated with TC and LDL levels. There were no significant correlations between family history and HDL. The second research question concerning the incidence of cardiovascular disease on the genogram as a valid predictor of the lipid and lipoprotein profile of these same incoming first year medical students after accounting for age and sex was tested by using a multiple regression analysis. Multiple regression analysis which is a method used to statistically understand the effects of two or more independent variables on a dependent measure (Polit & Hungler, 53 54 Table 4 Correlations Between Incidences of Cardiovascular Disease and The Lipid and Lipoprotein Profile Relationship/Hx TC HDL LDL (n=351) (n=346) (n=345) Paternal Hx .16* .09 .14* Maternal Hx .05 .01 -.06 Paternal Grandfather's Hx .05 .03 .02 Paternal Grandmother's Hx .05 -.02 .09 Maternal Grandfather's Hx .00 .00 -.02 Maternal Grandmother's Hx .11 .09 .06 Both Parents Hx .10 .07 .08 All Grandparents Hx .09 .04 .06 Paternal Grandparent's Hx .07 .01 .07 Maternal Grandparent's Hx .07 .05 .02 Paternal/Grandfather Hx .15* .08 .12 Paternal/Grandmother Hx .15* .05 .16* Maternal/Grandfather Hx -.03 .00 -.05 Maternal/Grandmother Hx .05 .07 .01 Paternal/Grandparents Hx .15* .06 .14 Maternal/Grandparents Hx .04 .05 -.01 Total Family's Hx .12 .07 .09 * P 5.01, 2 tailed 55 1987), was performed between the independent variable of family history of CVD on the genogram and the dependent variables, the levels of the lipid and lipoprotein profile, after accounting for the age and sex of the sample. This can be seen in Table 5 which depicts a multiple regression performed on the dependent variables: TC, LDL, and HDL; and the independent variable of the total family incidences of CVD recorded on the genogram known as family history, after accounting the confounding for variables of the sample subject's age and sex. After entering sex and age into the multiple regression equations, total family history was entered. Generally, the total family history did not significantly affect the variation for this sample with the exception of the increase in variance of 1% of the total cholesterol levels in older subjects. For the prediction of total cholesterol only age has a strong significant prediction accounting for 10% of the variance. The strength for the predictability in the variance of LDL was found in sex, specifically male, and age accounting for 10% of the variance in LDL levels. For HDL only the female sex was a significant predictor explaining 14% of the variance in the HDL score. In no equation does family history add 56 Table 5 Multiple Regression Analyses of Age, Sex, Family History of CVD to Predict The Lipid and Lipoprotein Profile N R2 b SE b TC 346 sex .00 4.39 3.35 age .10** 2.50* .41 famhis .11* 1.53 .79 constant 106.50 10.85 LDL 340 sex .03** 11.21** 2.99 age .10** 1.90** .33 famhis .11 1.10 .70 constant 49.71 9.68 HDL 341 sex .14** -10.73** 1.47 age .14 .15 .18 famhis .14 .25 .35 constant - 56.75 4.76 *p g .05 and u p 5 .01. 57 anything but a very small amount in a predictive fashion to the explanation of the lipid and lipoprotein profile levels. Summary The statistical tests performed on the variables of family history of CVD on a genogram (angina, heart disease, hypertension, hypercholesterolemia, and myocardial infarction), the levels of the lipid and lipoprotein profile (TC, LDL, and HDL) and the confounding variables of sex and age did not show that family history can predict the levels of the lipid and liprotein profile. Some correlations between variables were discovered but none of these correlations indicate predictability. Chapter VI contains interventions based on the statistical outcomes of this thesis and factors which may or may not have affected these statistical outcomes. CHAPTER VI Interpretations, Implications and Recommendations Overview This Chapter contains a summary of Chapter I through V with the interpretations of the statistical procedures and findings. The interpretations are viewed from a theoretical and statistical perspective. The implications for advanced nursing practice based on the findings of this thesis are presented. Finally, recommendations for future research are presented, along with a summary statement. Interpretations The purpose of this thesis was to explore the usefulness to the primary health care provider of a positive family history of CVD as a predictor of a client's lipid and lipoprotein profile. The two research questions formulated to explore the purpose of this thesis were: 1) Is there a relationship between the reported incidence of cardiovascular disease on a 58 59 family genogram and the lipid and lipoprotein profile in incoming first year medical students?; and 2) Is the incidence of cardiovascular disease on the genogram a valid predictor of the lipid and lipoprotein profile of these same medical students after accounting for age and sex? The variables of these research questions (family history of CVD on a family genogram, lipid and lipoprotein profile, age and sex) are found within the Modifying Factor variables of the Health Belief Model. The HBM was the conceptual framework used to investigate the variables within the context of health promotion/disease prevention. The secondary data of this non-experimental descriptive correlational study, as tested within this thesis, were not found to be strongly investigated as evidenced by a lack of studies within the literature, in relation to the predictive value of the genogram as an assessment tool. Most of the literature reviewed reflected a general trend of risk factors for the development of CVD especially a positive family history of CVD and undesirable levels on the lipid and lipoprotein profile. High risk groups for the precursors of CVD have been defined in the attempt to 60 identify and educate the subject, promote health and prevent disease. Angina, heart disease, hypercholesterolemia, hypertension, and myocardial infarction were the cardiovascular diseases recorded on the genogram for parents and grandparents. The fasting total cholesterol, low-density lipoprotein, and high-density lipoprotein levels of 360 first year medical students were tested for the relationship of these levels with a recorded family history of CVD on the genogram after accounting for age and sex of the sample. Discussion The frequencies and percentages of recorded incidences of CVD on the genogram were not as high as the potential. Given the possibility of 5 diseases assessed for 6 relatives of the 360 cases, a potential of 10,800 incidences of CVD could have been recorded. A total of 899 incidences of CVD among relatives were recorded, which is less than one-tenth the potential of recorded incidences. The total paternal family history of recorded incidences of CVD on the genogram occured in a frequency 120 times more than in the total maternal family history, which corroborates the 61 literature for male and female risk for CVD (Corrao et al., 1990; Levy et al., 1990; Sempos et al., 1989) especially given the young age of sample subject. The total individual frequency for hypercholesterolemia was 45 with 28 coming from father alone. The frequencies and percents (Table 1) indicated that females had more incidences of hypertension than males, except for fathers and that myocardial infarction tends to be found more in grandfathers than any other relative. The mean and standard deviation (Table 2) indicated that males and grandparents had a higher recorded incidence of CVD on the genogram than any other relationships. Given the potential pathophysiological interconnectedness of the CVD variables (angina, heart disease, myocardial infarction, hypertension and hypercholesterolemia), the total frequency in this study for hypertension and myocardial infarction numbered a little under three-fourths of the recorded incidences of CVD, a question is posited about what happened to the expected numbers of frequencies of the remaining variables. The mean (1.52) and S.D. (.69) for total family history incidences of CVD recorded on the genogram (Table 2) illustrate that overall there were 1 to 2 incidences of CVD recorded on the genogram for family 62 members. There is only a very small difference in the mean incidence of CVD between male relatives and female relatives. There are approximately 2 more mean recorded incidence of CVD for grandparents than parents. This difference in expected and actual frequencies and the low mean incidences have the potential of resulting in a failure to support the concepts of the structural variable of the Health Belief Model. There are multiple reasons which may in fact be the cause of this difference between the expected and actual frequencies and the low mean numbers of recorded incidence of hypertension, heart disease, and hypercholesterolemia: l) The diseases may actually not have been present in this sample, meaning this group may have been particularly healthy; 2) The data may have been poorly collected and/or recorded, the sample may have found the instructions on genogram to be confusing thereby resulting in a lack of accuracy in their recording of the information, or the subjects in fact did not know their family history and therefore were unable to record it (this reason would not support the use of family history of CVD on the genogram as a component within the structural variables of the HBM); 3) The age of the sample was too young for an evidenced 63 existence Of CVD in family members and some of the individual parents and grandparents may have been too young to manifest any physical symptoms or may not yet have been diagnosed; 4) The sample subjects were in the prematriculation orientation week of their first year in medical school, for many probably a stressful time when thinking and remembering a family history of disease and health, if thoroughly known, may not have been in the top priority of important items to remember and think about; 5) The variables on the genogram which were not included (gout, diabetes mellitus, cigarette use, cerebrovascular accident, cancer, and obesity of which diabetes mellitus and gout with their pathophysiological etiology and interaction with cholesterol may be secondary contributing factors to the development of atherosclerosis) may have significantly increased the mean and standard deviation of the recorded incidences of CVD on the genogram; 6) The degree of emphasis placed on the genogram in the two different colleges of medicine (ostoeopathic and allopathic) and the amount of orientation and information presented to each college prior to the presention of the instrument (genogram) to be filled- out may have accounted for the overall small number of recorded incidences on the genogram; and finally, 7) 64 The subjects may not have known their family health history and the interconnectedness of the diseases so had no reason to record what they knew nothing about. It is unknown which, if any, of these reasons were factors in this thesis or contributed to the low frequency for family history. The frequencies and standard deviations for the total cholesterol, low-density lipoprotein, and high-density lipoprotein levels (Table 3) indicated a large number of desirable or only slightly abnormal levels for total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. Perhaps the approximately 15% of those with borderline high-risk or high-risk levels of total cholesterol and low-density lipoprotein, may have a desirable level of HDL (90%) which would decrease their potential risk of the development of the precursor of CVD, however this was not tested. The large number of desirable total cholesterol, low-density lipoprotein, and high-density lipoprotein levels is undoubtedly affected by the young age of the sample subjects included for this thesis. Given that total cholesterol levels increase more than 2mg/d1/year during early adulthood (Corrao et al., 1990; Levy et al., 1990; Sempos et al., 1989), perhaps it can be speculated that if the cut-off points for 65 total cholesterol and low-density lipoprotein levels were lowered for those of younger ages, the frequencies of the levels on the lipid and lipoprotein profile which were within acceptable and only slightly abnormal ranges in this sample may have fallen into the at risk categories for development of the precursor of CVD. The correlations in Table 4, depict a significant positive correlation between a paternal family history of cardiovascular disease and total cholesterol and low-density lipoprotein levels of the sample. This correlation does not, however, carry through to the multiple regression to demonstrate that a family history may have a predictive value in determining a client's total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. The first hypothesis of this thesis that a relationship exists between a family history of CVD and lipid and lipoprotein profile levels was shown to exist to some extent for this sample. Some of the individual family history relationships, although shown on Table 4 to have statistical significance in a univariate, zero order, Pearson correlation with the lipid and lipoprotein profile, when brought together as a total family history fail to indicate any predictability for the 66 levels in the lipid and lipoprotein profile in the multiple regression equations of Table 5. This is not totally unexpected statistically, however, for if a significant zero order correlation is seen to exist in the Pearson correlation, the variables frequently do not have any significant predictability in the bivariate multiple regression equation (Polit et al., 1987). In this thesis there was no statistical significance for the relationship which existed with total family's hx of CVD and the subject's total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. Any statistical significance of the single order correlations may have been obliterated when only the total family's history which is the combination of all recorded family incidences of CVD was the predictor variable. The strength of predictability of family history on the lipid and lipoprotein profile of the subjects in the multiple regression (accounting for their sex and age) is computed from the correlational relationship of total family's history and total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. Since no statistically significant relationship exists between these two variables (Table 4) it is net surprising that no statistically 67 significant predictability was established either (Table 5). The results of the multiple regression performed (Table 6) indicated that family history of CVD does not significantly account for much of the variances in the levels of the lipid and lipoprotein profile (TC, LDL, and HDL) in this sample. The small total number of incidences of CVD recorded on the family genogram and the narrow standard deviations certainly explain why on the multiple regressions family history does not predict any significant amount of the variances of the total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. Therefore, incidence of family history of CVD on a genogram based on the statistical results of this thesis do not have a predictive ability of the total cholesterol, low-density lipoprotein, and high-density lipoprotein levels of offspring. The results of this thesis concur with the conclusions of the study by Griffin et al. (1989), which found that the undesirable total cholesterol, low-density lipoprotein, and high- density lipoprotein levels of the samples of children were not significantly explained by incidences of a positive family history of CVD. The Bogalusa Heart Study (Shear et al., 1985), The Tromso Heart Study (Forde & Thelle, 1977) and the study by Nora et al. 68 (1980), on.a univariate analysis revealed a relationship between a history of individual cardiovascular diseases (HTN, MI, and/or HRT DIS) and elevated total cholesterol and low-density lipoprotein levels in the sample. The results of these studies concur with the findings in this thesis for a statistically significant correlational relationship with the sample's lipid and lipoprotein profile (specifically total cholesterol and low-density lipoprotein levels) and individual diseases recorded on the genogram as a family history of CVD (e.g. MI, HRT DIS, HTN). The confounding variables of age and sex, however, were found to significantly account for the variance of the lipid and lipoprotein profile within this sample. The confounding variables of age and sex demonstrate more of an effect on the levels of the lipid and lipoprotein profile than a family history of CVD. For high-density lipoprotein only the female sex accounts for any variation (14%) in the HDL level. Studies recorded within the literature (Correo et al., 1990; Levy et al., 1990; Sempos et al., 1989) report that premenopausal women have a higher HDL level and that the role of family history does not have as strong an affect on the desirable levels a premenopausal 69 client may have. Age, if a woman is menopausal, can effect the HDL level but these sample subjects do not fall into that age group (Correo et al., 1990; Levy et al., 1990; Sempos et al., 1989). There are 5 cases unaccounted for in the total cholesterol, low-density lipoprotein, and high-density lipoprotein levels of the multiple regression which are accounted for in the frequency, percentages and correlation tests. The reason these 5 cases are unaccounted for is unknown. Whether these 5 unaccounted for cases would have significantly changed the multiple regression equation results is doubtful but the potential exists. Implications for Advanced Nursing Practice Assessment tools, in many forms, are consistently utilized by a primary health care provider to obtain objective data which enables the provider to make a diagnosis and in collaboration with the client devise and recommend interventions which can promote health, prevent and/or treat disease. One of the tools used to assess a client's family health/disease history is a genogram. 70 A Family Clinical Nurse Specialist (FCNS) assessing the risk of the precursor of CVD being present in a client utilizing the statistical findings of this thesis would need to consider the implications of the small number of frequencies of abnormal levels in the lipid and lipoprotein profile and recorded incidences of CVD on the genogram. If the acceptable range for the lipid and lipoprotein profile were, also, weighted according to age of the client, and the interconnectedness of the incidences of CVD were known and recorded then risk factors may increase. Clients of a younger age population could be targeted for health promotional education, hopefully preventing the need for disease prevention modality of interventions. Based on the multiple regression (Table 5) results of this thesis, a primary health care provider could not use positive incidences of CVD on a family genogram as rationale to order a lipid and lipoprotein profile on a client, but could use the client's age and sex, knowing that positive family history of CVD on a genogram does not account for variance in LDL and HDL levels but age and sex do significantly account for the variations in LDL and HDL levels. The lack of predictability of the family history of this sample in the multiple regression indicate the importance of the 71 FCNS acquiring a complete family history by sending the client home with the genogram to question relatives and ordering lipid and lipoprotein profile for risk factor screening regardless of (a positive) family history for CVD. Although the genogram may not be able to be used as a tool to determine the need for lab tests. It is an excellent pictorial representation of a client's family health/disease history and an extremely efficient mode of communication with other health care providers who may be utilizing the client's record about the client's family's health/disease history. The genogram utilizing the client's interviewing of family member's health/disease history allows the client and family to learn about information in their own history which otherwise may have remained unearthed. The review and discussion of the genogram with the client by the FCNS provides an educationally visually stimulating and pictorial representation of a client's risk factors for specific diseases based on family history as a risk factor alone. Given the strength of the predictability of female sex on high-density lipoprotein levels (Table 5), a FCNS assessing a premenopausal female with or without a positive family history of CVD on the genogram would 72 expect to find that female client's risk based on lipid and lipoprotein profile to be minimally based on an undesirable HDL level. Based on recent literature (Correo et al., 1990; Levy et al., 1990; Sempos et al., 1989), a FCNS assessing a pre-, peri-, and post- menopausal female needs to pay particular attention to the risk factors which may exist even if the HDL levels are not abnormal and intervene appropriately through health promotion and/or disease prevention education and more extensive assessment of other risk factors eg. lifestyle, family history, diet, or exercise. Based on the findings of the correlations (Table 4) of this thesis, a FCNS might expect to find that a client with a positive family history of CVD in the paternal family history on the genogram would have an undesirable total cholesterol and low-density lipoprotein level. Combining the known information of the correlations on Table 4 and the multiple regression on Table 5, an older male with a positive paternal family history of CVD would be expected to be at risk for an undesirable total cholesterol and low-density lipoprotein levels. A practicing Family Clinical Nurse Specialist, using the Health Belief Model, can educate all client's starting with health promotion and/or disease 73 prevention (wherever the presenting client is at) for any significant family history health/disease trends. A family history can be obtained from a genogram which is a non-threatening, non-invasive and relatively inexpensive source of information gathering from a historical perspective about the client and their family. Health promotion could best be utilized if a disease is not present in the family history and if the client who has a positive history for a disease has no other clinical or non-clinical manifestations of risk factors for the presence of the disease. If a client has clinical or non-clinical manifestations of disease (laboratory tests, lifestyle of sedentary activity level, high fat diet, etc.), then the FCNS must consider the education to be disease prevention for clinical manifestations or worsening of a preexisting condition.. Interventions for plan of care (education on lifestyle, diet, exercise, the use of drugs, further testing, and other modifiable factors) are acknowledged as important aspects in caring for a client but are not under consideration within this thesis. Recommendations for Future Research Cardiovascular disease is well documented as being the leading cause of morbidity and mortality in modern society (LaRosa et al., 1990 & Samuelsson, et al., 1987). Several factors have been found to be associated with an individual's increased risk of developing CVD and/or its precursors, including cigarette smoking, hypertension, an undesirable lipid and lipoprotein profile, sex, age, family history of CVD, obesity, diabetes mellitus and a sedentary lifestyle (Kwiterovich, 1989; Nichols, 1988). The findings of this thesis failed to support the predictability of positive family history of CVD on a genogram for the lipid and lipoprotein profile in the sample. A limitation of this thesis design lay in the secondary data recorded on the genograms. There was no way to go back and assess the reliability of what was recorded on the genogram or the investigator's accuracy in interpreting the meaning of what was recorded. A coding scheme which included more than presence or absence of a disease may elicit more strength in the correlational relationship. If the data of a genogram were coded for presence, absence, unknown and missing, this change in the coding scheme may change the 74 75 relationships which may or may not be found to exist with family history of CVD and a client's lipid and lipoprotein profile. Future research using a genogram would require that careful consideration be paid to the data collection for accuracy and reliability. The genogram in the PCAA were not collected for research purposes. If a genogram were sent home with a subject and returned within a predetermined length of time, reviewed with the client the reliability, validity and utility of the genogram as a statistical assessment tool may be more accurate. The accuracy of the interpreter can only be assessed by reviewing the recorded incidence of diseases with the client. Would a change in the method of data collection and a graduated coding scheme of the recorded data of the genogram provide strength in the data to enable the genogram to be a statistically significant tool? The genogram needs to be sent home with the client for at least one or two weeks to enable the client to investigate the family history with relatives. Next, the primary health care provider needs to go over the genogram assessing and clarifying the recorded information to insure accuracy of recording and interpretation of information recorded on the genogram. 76 If a genogram proves to be an accurate and reliable assessment tool, would the results have a more predictive ability where a client's physical test results are concerned? Using multiple regression analysis, accounting for specified confounding variables of the sample, recorded incidences of a disease in a genogram may be determined to have predictability in a sample's physical test results. The incidences and the relationships can be tested in the covariate analyses separately and together along with established risk indexes, thereby assessing the strength and predictability of family history and present sample's test results. Risk factors of sample subjects (race, diet, activity level, cigarette smoking and family history of cigarette smoking, cerebrovascular accident, gout, diabetes mellitus, and obesity) that were not assessed within this thesis may strongly affect the variance of total cholesterol, low-density lipoprotein, and high-density lipoprotein levels. Sex and age (the confounding variables in this thesis) were shown to have a significant effect on the variance of the TC, LDL and HDL levels. If more risk factors were assessed for the sample subject (eg. race, diet, weight, lifestyle, exercise, blood pressure, and percent of 77 body fat), the statistical strength of the significance of age and sex on the variance of TC, LDL, and HDL (Table 5) may be altered. For future research a reanalysis of the data using all the risk factors of the Preventive Cardiology Academic Award might reveal that a positive family history of CVD on the genogram (also including gout, DM, CVA, obesity, and cigarette smoking) would predict an undesirable lipid and lipoprotein profile (after also accounting for race, weight, height, diet, vital signs, and activity level) in the incoming first year medical students for the academic years 1980 through 1982. Another study using an older age group might reveal predictability of the lipid and lipoprotein profile by~a positive family history of CVD. Lower cut-off scores for risk assessment in the lipid and lipoprotein profile, different geographical locations, different age groups, along with more complete demographic information might affect the predictability of a positive history of CVD on a genogram for the lipid and lipoprotein profile. A longitudinal study using the variables of this thesis, a family history of cardiovascular disease on the genogram, and the lipid and lipoprotein profile, with the confounding variables of age and sex may reveal a positive and predictive 78 relationship between total family history of cardiovascular disease and the lipid and lipoprotein profile. Using the Health Belief Model as a conceptual framework for investigating the relationships which exist with a family history of CVD on the genogram and the lipid and lipoprotein profile of a client, the application of the variables of the HBM (eg. perceived seriousness of the disease, susceptibility of the disease, the perceived threat of the disease and likelihood of taking action) could be accomplished using a scale of appropriate questions to test each variable. The testing of the variables of the HBM might best be accomplished after a complete genogram has been gathered by the client and reviewed with the client by the primary health care provider. Within the Modifying Factor Variables of the Health Belief Model, cardiovascular disease was included as a separate concept variable within the structural variables; for future research the concept of CVD would not have to be separated out but instead could be included within the variable of family history of CVD on the genogram. Summary In summary, a positive family history of CVD on a genogram did not support the predicted undesirability in a client's total cholesterol, low-density lipoprotein, or high-density lipoprotein profile levels for this sample. A relationship between paternal family history of CVD and undesirable TC and LDL levels was shown to exist for this sample. The genogram is an instrument which requires further testing to prove any internal validity and reliability. It is an instrument that if utilized and assessed accurately will probably provide a wealth of information about a client's family health/disease history. Use of the genogram as a tool for statistical testing continues to require further investigation before any definitive conclusions can be made. 79 APPENDICES APPENDIX A 80 Michigan State University Preventive Cardiology Program lion t use the genogram: famii hx of can and risk factors (1) Record your partner's name in ink of one color along with the symbol "B'to indicate this to be the'identified patient": if partner is unmated, enter his/ her name on line A; a mated partner’s name should go on line 8. (2) Using the genogram symbols, systematically go back in the family lineage, establishing years of birth, marriages and siblings. Ilse a different color pen. ‘3) Siblings should be arranged in order of birth (from left-> right should indicated decreasing age of siblings); abortions should be placed in the appropriate chronological order also. flames of family members are not necessary but good to include if you have time to get them. include 003. (4) flow refer to the gen ogram code and note all illnesses and diseases of family members. Remember to use or® to show death. Where possible, in lcate cause of death by circling the diagnosis. e.g. Ii? “"" a heart attach a . a... o 32 a. 22:... K 9 a “gut. ¢ .0- .— nanacnasnncu 3:32. .3 so :3. . . :3. .3... .3 .3 a. .3200 e mné 3:33.! .3 .333 «new... rot-nu 3. . .nn_nnn an. .33-- I. r 0.0: .3 so ‘0“ 03.”... mu»- .3 .1 a. a 3.: a u... u ”a a a“ 2:79 K®%_® {anal-unevuuw a"... , 3...... so- «8. son. 5 0 an- 0 "an... o 5-8 o ease as a e. .e m. e a e x e e. a..- SEN; "can: .530... 3.30 "eh-z 339nm so; can x... 2.5!... "Encuoaou 035nm 5209:. 323.030 25:26:. 3.32.25 22m 59.3.5. Michigan State University Preventive Cardiology Program PCP GENOGRAM SYMBOLS C" male 0 telnaie 'M carriage Qfig divorce 0 so: not specified 0 adopted l spontaneous abortion 8 death f identified patient . A) twins (berm?) MiG angina CIG cixarette aser on diabetes mellitus 61‘ goat tilt‘l' DIS heart desease fCIIOL hypercholesterolemia f8? hypertension Ml myocardial infarction OBES obesity CVA stroke CA cancer DOB date-oi- birth it. _ _ _ l 2...: I I o 3.: a I I I a I a I: "can: 338.... am:— ucc 0333 too: 553.30 Io foam:- 27.5... Emaoi 322030 25:26:. 3383:: 25w 59:82 APPENDIX B Student ID! Line A male - female 000 5 84’ CODEBOOK FAMILY GENOGRAM you and siblings - 01 parents I 02 grandparents, paternal - 30 maternal - 31 great-grandparents, paternal - 40 maternal - 41 sex not specified - 7 . - missing data ,3 identified patient - 10 Q adopted - 11 (f spontaneous abortion . 12 8 E death - 13 ab twins - 14 For lines 8 thru D only: m marriage I 8 divorce - 9 c5" '75 85 ANG angina I 15 C16 cigarette user I 16 on diabetes mellitus I 17 GT gout I 18 HRT DIS heart disease I 19 f (2301. hypercholesterolemia I 20 f B/P hpertension I 21 MI myocardial infarction I 22 oars obesity - 23 CVA stroke I 24 Cause of death I 0 DOB date of birth (year only) Age (it given) ALL MISSING DATA I . APPENDIX C 86 Consent Form for Preventive Cardiology Program Experience PU'POse: The study in which you are being asked to participate is a mass screening oi “'Si'Year medical students. The purpose is to interest you in preventive medicine and the concept of rislr factor assessment and modiiication. This cardiology program will serve as a prototype you can use when attempting to Intergrate preventive measures in your clinical practice. Consent: 1. i agree to having height. weight. blood pressure. triceps, skiniold and elec- trocardiogram (EKG) measurements both at rest and with exercise taken. 2. i agree to having blood drawn for the purpose of determining glucose. cholesterol. HDL and triglyceride levels. 3. l understandthat certain health conditions preclude participating in tread. mill testing and i agree to provide the requested information about my health so that a determination can be made by a licensed physician as to my eligibility to participate. 4. l understand that in the unlikely event oi physical injury resulting irom research procedures. Michigan State University. its agents. and employees will assume that responsibility as required by law. Emergency medical treat- ment for injuries or illness is available where the injury or illness is incurred in the course oi an experiment. l have been advised that i should look toward my own health insurance program for payment of said medical ex- 0.0308. 5. i understand that no beneficial effects are guaranteed by participation in the program. , b. i understand that individual or group test results may be published. but ac- tual names shall not be used, so as to insure participant confidentiality. i have read the above consent form for the Preventive Cardiology Pro. gram Experience. fully understand the purpose and requirements of my participation. and freely consent to participate. Signature: Date: Witness: APPENDIX D 87 MICHIGAN STATE UNIVERSITY OIFICF UI ”(I PRESIDENT K)! RESEARCH EAST LANSING 0 MICHIGAN 0 “new“. \50 DEAN O! "I! GLADL AT! SCHOOL April 3, 1992 Ann Currie 1606 Snyder Rd. East Lansing, HI 48823 RE: ASSOCIATION OF FAMILY HISTORY OF CARDIOVASCULAR DISEASE AS RECORDED ON A FAMILY GEROGRAH AND LIPID AND LIPOPROTEIN PROFILES, IRB '92-141 Dear Ms. Currie: The above project is exempt from full UCRIHS review. The proposed research protocol has been reviewed by a member of the UCRIHS committee. The rights and welfare of human subjects appear to be protected and you have approval to conduct the research. You are reminded that UCRIHS approval is valid for one calendar year. If you plan to continue this project beyond one year, please make provisions for obtaining appropriate UCRIHS approval one month prior to March 23, 1993. Any changes in procedures involving human subjects must be reviewed by UCRIHS prior to initiation of the change. UCRIHS must also be notifed promptly of any problems (unexpected side effects, complaints, etc.) involving human subjects during the course of the work. Thank you for bringing this project to my attention. if I can be of any future help, please do not hesitate to let me know. 5 avid E. Hright, Ph. University Committee Human Subjects (UCRIHS DEH/pjm cc: Dr. Rachel Schiffman Sincerely, air search Involving "\L a an Vine-um r hum. Fund ()ppnflmlun Inuit-rune LI ST OF REFERENCES REFERENCES Berkow, R. (Ed.). (1982). The Merck manuel. New Jersey:’ Merck & Co. Becker, M. H., Drachman, R. 8., s Kirscht, J. P. (1974). A new approach to explaining sick-role behavior in low-income populations. American Journal of Public Health, 91, 205-216. Corrao, J. M., Becker, R. C., Ockene, I. S. 8 Hamilton, G. A. (1990). Coronary heart disease risk factors in women. Cardiology, 11 (Suppl. 2). 8-24. Eaton, C. 8., Schaad, D. C., Rybicki, 8., Pearson, T. A., Van Citters, R. L., Stone, E. J., Castle, C. H., Cohen, J. D., Davidson, D. M., Greenland, P., Krakoff, L. R. Riemenschneider, T. A. 5 Derby, C. (1990). Risk factors for cardiovascular disease in U.S. medical students: The preventive cardiology academic award collaborative data project. In E. J. Stone, R. L. Van Citters, & T. A. Pearson (Eds.), Perspectives in physician education (pp. 14-22). New York: Oxford University Press. Expert Panel. (1988). Report of the national cholesterol education program expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. Archives in Internal Medicine, 148, 36-69. Farcot, J-C., Hashimoto, R., Meerbaum, S. a Corday, E. (1977). Do risk factor interventions prevent or reverse_asteriosclerosis? In. E. Corday (Ed.), Controversersies in cardiology (pp. 1-10). Philadelphia: PA Davis. Forde, O. H. & Thelle, D. S. (1977). The Tromso Heart Study: Risk factors for coronary heart disease related to the occurrence of myocardial infarction in first degree relatives. American Journal of Epidemiology, ;, 192-199. 89 Gordon, H. (1972). The family history and the pedigree chart. The Journal of Applied Medicine, g, 123-125. Griffin, T. C., Christoffel, K. K., Binns, H. J., McGuire, P. A. a The Pediatric Practice Research Group. (1989). Family history evaluation as a predictive screen for childhood hypercholesterolemia. Pediatrics, 2, 365-374. Hochbaum, G. M.. (1958). Public participation in medical screening programs: A sociopsychological study. Public Health Service Publications, 572. Washington: United States Government Printing Office.' Jolly, W., Froom, J. E Rosen, M. G. (1980). The genogram. The Journal of Family Practice, 3, 251- 255. Kwiterovich, P. O. (1989). Beyond cholesterol: The Johns Hopkins complete guide for avoiding heart disease. Baltimore, MD: The Johns Hopkins University Press. LaRosa, J. H., Decker, D. M. & Fitzgerald, S. (1990). Elevated blood cholesterol a risk factor for coronary heart disease. American Association of Occupational Health Nurses, 5, 211-215. Levy, D., Wilson, P. W. F., Anderson, R. M. & Castelli, W. P. (1990). Stratifying the patient at risk from coronary disease: New insights from the Framingham Heart Study. American Heart Journal, 3(Suppl.2), 712-717. Lipid Research Clinics Program. (1984). The lipid research clinics coronary primary prevention trial results 1: Reduction in incidence of coronary heart disease. Journal of the American Medical Association, 251, 351-362. McGoldrick, M. s Gerson, R. (1985). Genograms in family assessment. New York: W. W. Norton. Mikhail, B. (1981). The health belief model: A review and critical evaluation of the model, research, and practice. Advances in Nursing Science, 3, 65-82. 90 National Institute of Health. (1974). Manual of laboratory operations lipid research clinics program, lipid and lipoprotein analysis. (DHEW Publication No. NIH 75-628). Bethesda, MD: National Heart and Lung Institute. National Institute of Health. (1987). National cholesterol education program: Highlights of the report of the expert panel on detection, evaluationy and treatment of high blood cholesterol in adults. (NIH Publication No. 88-2926). Nichols, A. A. (1988). Management of elevated serum cholesterol in a university-based family practice. The Journal of Family Practice, 3, 281-286. Nora, J. J., Lortscher, R. H., Spangler, R. D., Nora, A. H. a Kinberling, W. J. (1980). Genetic- epidemiologic study of early-onset ischemic heart disease. Circulation, 3, 503-508. Pender, N. J. (1987). Health promotion in nursing practice. Norwalk, Ct: Appleton & Lange. Polit, D. F. 8 Hungler, B. P. (1987). Nursing research principles and methods. St. Louis: J. B. Lippincott. Price, 8. A. 8 Wilson, L. (1986). Pathophysiology: clinical concepts of disease processes. St. Louis: McGraw-Hill Book. Rosenstock, I. M. (1960). What research in motivation suggests for public health. American Journal of Public Health, 59, 295-302. Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 328-335. Sackett, D. L., Haynes, R. B. 8 Tugwell, P. (1985). Clinical epidemiology7a basic science for clinical medicine. Boston: Little, Brown. BF 91 Samuelsson, 0., Wilhelmsen, L., Andersson, O. R., Pennert, K. a Berglund, G. (1987). Cardiovascular morbidity in relation to change in blood pressure and serum cholesterol levels in treated hypertension results from the primary prevention trial in Goteborg, Sweden. JAMA, 13, 1768-1777. Sempos, C., Fulwood, R., Haines, C., Carroll, M., Anda, R., Williamson, D. F., Remington, P. & Cleeman, J. (1989). The prevalence of high blood cholesterol levels among adults in the United States. JAMA, 262, 45-53. Shear, C. L., Webber, L. S., Freedman, D. S., Srinivasan, S. R. & Berenson, G. S. (1985). The relationship between parental history of vascular disease and cardiovascular disease risk factors in children: The Bogalusa Heart Study. American Journal of Epidemiology, 5, 762-771. Whitney, E. N. a Cataldo, C. B. (1987). Understanding normal and clinical nutrition. New York: West Publishing. Widmann, F. K. (1983). Clinical interpretation of laboratory tests. Philadelphia: F.A.Davis. Williams, R. R., Hunt, 8. C., Barlow, G. R., Chamberlain, R. M., Weinberg, A. D., Cooper, H. P., Carbonari, J. P. a Gotto Jr., A. M. (1988). Health family trees: A tool for finding and helping young family members of coronary and cancer prone pedigrees in Texas and Utah. American Journal of Public Health, 19, 1283-1287. Wilson, P. W. F., Christiansen, J. C., Anderson, K. M. 5 Kennel, W. B. (1989). Impact of national guidelines for cholesterol risk factor screening the Framingham offspring study. JAMA, ggg, 41-44. Wood, G. (1981). Fundamentals of psychological research. Boston: Little, Brown.