llllllllllllllllllllllllllllllllllllllllllllllll 31 93 00061 1966 This is to certify that the thesis entitled The Relationship Between Life Stress and Symptomatology in Alcoholic and Nonalcoholic Families presented by Constance N. Neil has been accepted towards fulfillment of the requirements for Master of Arts degreein Psychology Date Nay 9. 1984 0-7639 MSU is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State Us.“ :23“? . 3'5 ll. MSU RETURNING MATERIALS: Place in book drop to LJBRAjJES remove this checkout from .—:_ your record. FINES will be charged if book is returned after the date stamped below. L “ C r H -" 03‘! Hg, 5:;I‘ r 1‘? . ' . THE RELATIONSHIP BETWEEN LIFE STRESS AND SYMPTOMATOLOGY IN ALCOHOLIC AND NONALCOHOLIC FAMILIES by Constance M. Neil A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1984 ABSTRACT THE RELATIONSHIP BETWEEN LIFE STRESS AND SYMPTOMATOLOGY IN ALCOHOLIC AND NONALCOHOLIC FAMILIES by Constance M. Neil The present study examines the relationship between life stress and symptomatology among adults and children in alcoholic and nonalcoholic families. Children in the alcoholic families are at heightened risk for later alcoholism because of their male gender and having an alcoholic father. A variety of self-report and interview measures of life stress, physical health and psychological symptoms were utilized (with special attention to anti-social and depressive characteristics in adults and a broader range of symptoms in children). Results show a positive relationship between parents' life stress and children's life stress and symptomatology. There is some evidence that this connection is stronger between mothers and children and, in alcoholic versus nonalcoholic familes, but replication is needed. When comparing groups, both parents in alcoholic families experience more stress and more symptomatology than their matched controls. No differences were found when comparing children's symptomatology between groups. To the struggle and the discovery of self. - ACKNOWLEDGMENTS I would like to express special thanks to Dr. Robert A. Zucker for his continuous support and understanding throughout the process of this work. I would also like to express thanks to the other members of my committee, Drs. Ellen A. Strommen and John Paul McKinney. A special thanks to my sister, Susan, whose love and patience made editorial revisions of this work possible. A special thanks also to Tom Packard whose statistical and computer talents made the data analyses much less stressful. TABLE OF CONTENTS Page LIST OF TABLES . ....................... vi LIST OF FIGURES ....................... vii CHAPTER 1. Review of the Literature . . . . . . ..... . . . . 1 Review of Stress Literature . ........... 3 Life Stress and its Measurement in Children . . . . 3 Life Stress and Symptomatology in Children . . . . . 4 Moderator Variables in the Child's Stress-Illness Relationship . . . . ..... . . 7 Life Stress and its Measurement in Adults . . . . . 12 Moderator Variables in the Adult's Stress-Illness Relationship . . . . . . . . . . 18 Life Stress and Symptomatology in Adults . . . . . . 19 Review of Alcoholism Literature . . . . . . . . . . 25 Behavior Problems in Children of Alcoholics . . . . 25 Characteristics and Behavior of the Alcoholic Father and the Nonalcoholic Wife . . . . 26 Interactions and Characteristics of Alcoholic~Families . . . . . . . . . . . . . . . . 28 Alcoholism and Stress Theory . . . . . . . . . . . . . 30 Formal Predictions . . . . . . . . . . . . . . . . . . 31 11. Method . . . . . . . . . . . . . . . . . . . . . . . . 35 Subjects . . . . . . . . . ....... . . . 35 Rationale . . . . . . . . . . . . . . . . . . 35 Selection . . . . . . . . . . . . . . . . . . . . 35 Procedure . . . . . . . . . . . . . . . . . . . . 35 Measures . . . . . . . . . . . . . . . . . . . . 40 Children's Measures . . . . . . . . . . 45 Measurement of Children' s Stressful Life Events . . . . . . . . . . . . . . . . . 45 Measurement of Children s Symptomatology . . . . 47 Parent Measures . . . . . . . . . . . . . . . . . 50 Measurement of Stressful Life Events . . . . . . 50 Measurement of Adult Symptomatology . . . . . . 51 III. Results . . . . . . ..... . . . . . . . . . . . . 58 Analysis . . . . . . . . . . . . . . . . . . . . . . 58 Life Stress Measures . . . . . . . . . . . . . . . . 59 Adult Symptom Measures . . . . . . . . . . . . . . . 60 Child Symptom Measures . . . . . . . . . 68 The Relationship Between Life Stressors and Symptoms . . . . . . . . . . . . . . . . . . . 68 Individual Perceptions and the Stress-Symptom Relationship . . . . . . . . . . . 73 iv IV. Discussion . Adults Adults . Life Stress Findings . . . . . . Children . . . . . . ............ Symptom Findings . . . . . . . . Children . . . . . . . . . . Stress-Illness Findings A Theoretical Framework . Methodological Issues . . Measurement Issues . . . . Future Directions . . . . APPENDICES I. Contact and Data Collection Schedule . II. Individual Frequencies of Life Events 111. Correlations of a Family Life Event Measure and Adult Symptomatology . . . . . . . . . IV. The Anti-Social Behavior Questionnaire . . . BIBLIOGRAPHY . . . 87 9O 92 93 95 Table 10 11 12 13 14 LIST OF TABLES Average Amounts of Stress Experienced in a One Year Period by Children From a Normal Population . . . . . . . . . . . . . . . . . . Significant Correlations Between Four Types of Subjective Evaluations of LIfe Events . . . Sociodemographic Characteristics of Alcoholic and Community Control Families . . . . . . . . . Alcoholic Diagnosis, Drinking Problem Scores, and Drinking Pattern Scores in Alcoholic Families and Community Control Familiesn . . Research Diagnostic Criteria (RDC) for Diagnosis of Alcoholism . . . . . . . . . . Summary of Life Stress Study Measures . Amount of Life Stress Experienced by Adults in Alcoholic and Community Control Families Symptomatology Reported by Adults in Alcoholic and Community Control Families Symptomatology Experienced by the Target Children of Alcoholic and Community Control Families . . . . . . . . . . . . The Relationship Between Family's Life Events and Children's Life Events . . . The Relationship Between Family's Life Events and Children's Symptomatology . . Amount of Life Stress - Corrected by Subjective Weighting - Among Alcoholic and Community Control Families . . . . . . Individual Frequencies of Life Events . Correlations of a Family LIfe Event Measure and Adult Symptomatology . . . . . . . . . . vi Page 17 38 41 43 44 61 63 69 74 75 77 9O 92 Figure LIST OF FIGURES Behavioral Patterns Resulting From the Interaction of Strong Versus Weak Self-Control and Active Versus Passive Responding O O O O O O O O I O O O O I O O O O 0 Life Events Mediation in the Family System . . Model of the Life Stress-Illness Relationship vii 23 Chapter 1 Review of the Literature One objective of research on life stress is to demonstrate an association, over time, between the onset of illness and a recent increase in the number of life events that require a certain amount of readjustment or adaptive responses from the individual (Rabkin and Struening, 1976). A life stressor temporarily upsets the balance of an individual's life and he/she needs to exhibit some kind of c0ping response in order to reestablish that equilibrium. Theoretically, an “unhealthy" or unsuccessful coping response will result in some kind of mental or physical illness. The effects of life stress on behavior and ongoing functioning can take many forms. A quick survey of the Journal of Human Stress shows a connection between stress and return to heroin use, hypertension, infectious disease, depression, cancer, heart disease, accidents, athletic injuries, diabetes, and ulcers, to name a few. However, the connection between life stress and physical or mental illness is not one of direct cause and effect. Various mediating factors thought to be involved include age, sex, amount of readjustment needed, whether or not the change is desirable or undesirable, whether the stress involves an entrance or exit in the individual's life, personality factors, feelings of control, whether or not the event is short lived or has persisting consequences, past experience, and social support (Rabkin et al., 1976; Rutter, 1981). (:The relationship of stress to functioning seems particularly pertinent and interesting for the alcoholic and his family. It may be that the alcoholic experiences a tremendous amount of stress, or is unable to cope with it successfully, so he turns to drinking as a way of "managing" the stress. This, in turn, would create a number of new stressors or symptoms: marital conflicts, health problems, loss of friends, loss of job, financial problems. The alcoholic's family would also be experiencing additional stressors: increased number of arguments, loss of income, illness of a close family member (Calahan and Cisin, 1976; Ablon, 1976). The present study1 is an examination of differences in the amount and type of stressors experienced by alcoholic2 and normal families, and the difference in symptoms manifested by each of these groups. Although subjects are being studied at only one point in time, 1The present study is one part of the Michigan State University Vulnerability Study directed by Robert A. Zucker, Ph.D. 2In this paper, the alcoholic family refers to a family containing an alcoholic father. The wife may or may not be alcoholic, although in the present work, by far the majority are nonalcoholic. Any children resulting from this marriage are referred to as chldren of alcoholics or high-risk children. Likewise, the phrase "alcoholic parents" refers to an alcoholic father, and, more likely than not, a nonalcoholic wife. The rationale for focusing on the male alcoholic is reviewed in a later section. 1"“ IS Re fc Ca Dc 91' Cr Ch Ho an the families being studied are sufficiently young that etiological leads may be suggested by the nature of the differences observed. The following literature review will cover the areas of stress, mediating variables, and maladjustment in children and adults, and the characteristics of alcoholic families and their children. Review of Stress Literature Life Stress and its Measurement in Children: Normal stress research on children is still relatively new. Specific events such as maternal deprivation, divorce, and birth of a sibling have been studied separately, but the first life stress inventories for children were developed by R. D. Coddington in 1972 (Bowlby, 1973; Hetherington, 1972; Rutter, 1981; Coddington, 1972a, 1972b). These scales were patterned after the Holmes and Rahe Social Readjustment Rating Scale (Holmes et al., 1967), and are constructed for four different age groups: preschool age children, elementary age, junior high age, senior high age. Average amount of stress was calculated for each of the above age groups (N = 3620) in a normal population. No significant differences were found between boys and girls but racial differences were present. Specifically, black children in the two younger groups faced more life change than white children while the situation was reversed for the two older groups. However, the outstanding diference was the interaction between age and amount of stress with major jumps in amount of stress OCC SOT? 31 W 19 ill occurring between preschool and elementary school and again between elementary school and junior high school (Table 1). There is clearly some developmental effect. Coddington's explanation is as follows, "As a child expands his social sphere he risks the occurrence of more life events, good and bad" (Coddington, 1972b, p. 212). This conclusion is reasonable; yet, the age differences found in this study are questionable. It is not clear whether or not the measures of stress are comparable across age levels and if the list of life events are comprehensive for each age group. These studies do give us a general idea of what a “normal“ amount of stress may be for a child. Life Stress and Symptomatology in Children: [:Abnormal amounts of stress or an inability to cope with stress may produce behavioral problems in a child. Forbes categorizes a list of problems for parents that she calls signs of stress. Briefly, these include types of infantile behavior, withdrawal and depression, abnormally poor school behavior, constant drug use, aggression, and physical complaints (Forbes, 1978). All of these symptoms, with the exception of drug use, appear in a major study on childhood life changes reported by Hurme in 1981 Significant differences were found in the frequency with which subjects exhibited the above symptoms when the subjects were divided into groups according to the amount of life change experienced. Using both weighted and unweighted measures of life change, a child is significantly more likely to exhibit serious illness, fatigue, nervousness, bedwetting, crying, and aggression, if Table l Average Amounts of Stress Experienced in a One Year Period by Children From a Normal Population Age group No. of life events No. of life change units1 Preschool 1.73 64.99 Elementary 2.63 102.80 Junior High 4.11 195.66 Senior High 4.71 226.80 1Life change units (LCUs) are equal to the amount of readjustment required for the total number of life events. Adapted from Coddington, 1972. he has a "high“ level of life change. The researcher then factor analyzed the behaviors that significantly correlated with various indices of life change and labeled the categories as follows: weak self-control, sleeping problems, psychosomatic symptoms, diseases factor. Although the researchers cited above are generally in agreement about the type of symptoms related to stress, there is less research on whether there are specific stressors associated with specific symptoms. One conclusion from studies on maternal deprivation is that normal infants who are separated from their mothers prematurely are unable to relate to anyone; they are inhibited, exhibit aggression and speech impairments, and are afraid to explore (Bowlby, 1973; Wolff, 1973). Similar behaviors are manifested by a child whose parent had died. They show increased shyness, timidity, and withdrawal. However, children from separated or divorced families are more likely to have aggressive, antisocial problems (Rutter, 1981). A different set of responses has been reported for children experiencing birth of a sibling. Over half of a sample of two to three year olds with new siblings cried more and showed new toileting problems, while one-fourth developed sleeping difficulties (Dunn et al., 1981). 1:5freview of the literature suggests that type of parenting and family interaction can also be stressful for the child. Bowlby comes to this conclusion as well; he believes it is not the succession of different caretakers but the lack of continuous mothering that affects the child (Bowlby, 1973). Parent response may also be the cause of the stress related to a divorce or the birth of a sibling. In the first case, mothers often become depressed, self-centered, erratic, less supportive, and more ineffectual as a disciplinarian. Mothers may also be depressed and more inattentive to an older child at the birth of a new baby. Finally, in the instance of a hospital admission, a persisting disturbance is more likely if the child comes from a disturbed family or if the previous parent-child relationship was poor (Rutter, 1981). A more methodical analysis of parental contribution to stress is reviewed in Wolff (1973). Wolff reviewed a study by Hewitt and Jenkins who found children's symptoms clustered to form three syndromes that significantly correlate with recognizable situational patterns. The results show that l) unsocialized aggressive behavior is found in solitary children with backgrounds of parental rejection. 2) Social- ized delinquent behavior (e.g. stealing, truancy) is found in children with backgrounds in parental neglect and delinquency in the family. And, 3) overinhibited and neurotic behavior in children is associated with a family containing an inconsistent father, a dominating mother and lack of sociability in either parent. Or, this behavior may also be associated with a chronic physical defect or illness in the child. Moderator Variables in the Child's Stress-Illness Relationship: In determining what a life stressor is for a child and what kind of problem, if any, will develOp because of that stressor, we need to consider the child's resources for coping with any life change. Any general definition of stress must include the individual being stressed. That is, a situation is or is not a stressor depending on the person's response to it, and that response is in part determined by the person's ability to cope with that situation (Coddington, 1972a; Holmes and Rahe, 1967; Rutter, 1981). The various factors that may be important in determining how an individual deals with a situation include age, sex, genetic factors, temperament, intelligence, social support, amount of stress, past experience, and other problem solving skills. The relationship between amount of stress and caping ability is not linear, as one might expect. Hurme's research points to a U-shaped model as applied to life events. Specifically, large amounts of stress do correlate with a larger number of behavioral problems but so do small amounts of stress. A medium amount of stress is optimal for maximum adjustment. Hurme concludes that life events not only constitute negative threats but also function as develOpmental tasks (Hurme, 1981). (This writer notes that definitions of small, medium, and high amounts of stress vary from study to study.) Sex of the child may have a direct or indirect influence on a child's ability to cope with change. Boys tend to be more vulnerable to stress than girls, though the reasons for this may vary. This may be biologically determined along with the increased male vulnerability to physical hazards. The salience of stressful events may sometimes be greater for boys. Or, it may be that parents are less supportive of boys in their attempts to cope with life changes. Parents may also respond more negatively to boys' distress reactions. These patterns of interaction might have to do with the kind of behavior parents expect from sons, or might be determined by temperamental differences associated with sex (Rutter, 1981). The argument continues concerning whether or not temperament is genetically determined and if it changes over time. In either case it is still an important element in adaption to stress. For example, a child's temperament can reliably be used to predict his behavior after birth of a sibling (Rutter, 1981). Age and developmental stage are other variables that may influence the impact of a life change. For example, the age of greatest risk for maladjustment resulting from a hospital admission is six months to four years. A child younger than six months has not yet formed any selective attachments, while a child greater than four years of age can cognitively understand that a separation is not the same as abandonment. Likewise, the younger child is also more likely to show disturbed behavior at the birth of a sibling, whereas in divorce situations, children of all ages exhibit problems (although type of problem may vary with age). Finally, at the death of a parent, younger children have immediate grief reactions that are milder and shorter compared to those experienced by adolescents and adults. This pattern is explained by the younger child's less developed cognitive level (Rutter, 1981). A moderator variable in the stress relationship that is more in the category of learned competency is self-control. That is, stress leads to emotional and behavioral reactions which, in humans, are under 10 cognitive control; therefore, reactions to stress are under self- control. However, children and adults vary with respect to the amount of self-control they have internalized. (The amount of internalized self-control may be measured by locus of control scales.) In using this concept, Hurme has rated children on two dimensions, strong versus weak self-control and active versus passive responding. These two dimensions combine to form different behavioral patterns that produce differences in coping with life situations. These different adaptions are diagrammed in Figure 1 (Hurme, 1981). A study using this model on groups at risk for aggression show that children with weak self-control have experienced more life changes than those with strong self-control (Hurme, 1981); These changes include divorced parents, mothers with shift work, fathers with shift work, and moving five or more times during the child's lifetime. The conclusion drawn from this study is that children with weak self-control have a more unstable environment and less parental guidance. In sum, weak self-control is associated with large amounts of change and leads to uncontrolled aggressive behavior. ([33; Hurme study implies that parent behavior and family type influence how a child manages change. Modeling is one mechanism in this relationship. In other words, a child learns to manage stress in much the same way his parent does. Also, a child will learn to deal with situations more effectively if there have been rules and regulations to follow since infancy. Without these rules, it is hard for a child to discern behavioral expectations, i.e., to learn ll Passive Responding (Small No. of Overt Responses) Controlled Deliberateness Anxiety Uncontrolled Inhibition Inhibition of Submissiveness Fearfulness of Impulses Impulses ’Strong Weak Self- Self- Control Control Controlled Defensive Uncontrolled Expression Constructive and Expression of Activity Offensive of Impulses Aggression Impulses Active Responding (Large No. of Overt Responses Reproduced from Hurme, 1981. Figure 1. Behavioral Patterns Resulting from the Interaction of Strong vs. Weak Self-control and Active vs. Passive Responding. 12 appropriate caping mechanisms. Lastly, a child needs parental trust and positive support::>Without positive support of his other strong points, it will be difficult for a child to develop self-esteem. And, without trust, a child may develop feelings of uncertainty. Finally, the capacity to cope with frustration and stress in the presence of insecurities and in the absence of self-esteem is hindered (Forbes, 1978; Hurme, 1981).?5 The above review of stress and coping in childhood is full of variables and it is not always clear which are dependent and which are independent variables. The most comprehensive model this writer found was developed by Hurme and is diagrammed in Figure 2. It is described as a family model because the main world for the younger child is his/her family. Life Stress and its Measurement in Adults: The research on life stress in adulthood is much more extensive than that for childhood; yet, the type of research varies. In the adult literature there is much controversy on how to measure life stress and fewer studies, proportionally, on the specific psychological effects of life stress. The most frequently cited measure of life events is the Social Readjustment Rating Scale by Holmes and Rahe (1967). Holmes et a1. had 194 subjects rate 43 life events on the amount of readjustment required for each event. Readjustment scores were obtained by taking the mean score for each event and dividing by ten. Because this scale l3 soumxw »_.5aa «ea c. eo.aa.uaz mucus” ac_4 .~ acsm.a .gmag .uscar seep vouauocaom ucocoa may go; moucoaaomcou mmugaomwg m:_aou peacocoa u—.=u 0:» cow moucoauomeou O—ucsuwm as» :u': o:_aou m.upvgu u_.gu a ”coca; »: co.uo=u.m us» go acoEmmomm< mucouuu opp. .asco: u__gu ecu acocoa Co copuaacmpo hep—ocemco m.e_.gu ago Co a: u_.=m P »_.2ou coo—oac on» ou_mu=o ucoaaam .opuom a..s~. ago he poeucou use uv_mu:c mcouunu aco>o ouva .m~uc=¢moc mcpmoo can «cocoa» m=.moum.uocn mo m:.uu~ .mccuuuu xuppocomcoa peacocua m . j xvi NA 14 is non-interval and has a skewed distribution of arithmetic means, the geometric mean was considered the best measure of central tendency. It discounts the extreme score but takes into account the distribution of scores (Masuda and Holmes, 1967). There is strong agreement on the rank order of events and the magnitudes of the means as evidenced by high Pearson product moment correlations among various groups in the sample. Correlations by sex, marital status, age, generation, education, social class, and religion were all above .90. The lowest correlation was .82 between whites and blacks. All correlations were statistically significant (Holmes and Rahe, 1967). A second study by Casey et al. examined the reliability of recall on life events. A sample of 54 subjects was given the same question- naire nine months apart. Scores that differed by more than 40 life change units between time one and time two were considered discrepant. Individual items were also examined for consistency. Correlations between time one (1964), and time two (1965), focusing on recall of three separate years were .669, .638, and .744 for the years 1957, 1960, and 1963 respectively. Clearly, the more recent years are recalled more reliably; yet all three scores are significantly related at the .005 level. Despite the statistical significance, there were discrepancies. Thirty-two of 44 subjects scored non-discrepantly for one or two of the three years, with no consistent change in direction (i.e. more or less life events), for the discrepant years. More specifically, the items with higher readjustment ratings were recalled more majz int the wit am re (.0 15 more consistently. Also, items that have qualifiers such as major/minor and more/less in the question, are prone to subjective interpretation and this may cause change in recall over time. In fact, there was a significant difference in recall consistency between items with qualifiers and those without. However, the saliency of the items and not the presence of qualifying words affected the consistency of recall the most. Items with double questions were also recalled less consistently. Finally, the amount of life stress at the time of recall did not influence the magnitude of scores recalled (Casey et al., 1967). In sum, there is a strong reliability in the recall of life events. This conclusion is given additional support by Brown and Harris' study of life events. Their results show that the reported rate of events did not significantly drop off in the year before the events interview. That is, there was not a significant increase in the number of events reported in the more recent months (Brown and Harris, 1978). Though recall over time is fairly consistent, one identified bias in the initial reporting of events is the personal characteristic of denial. Those higher on the denial measure do report fewer life events than those low on denial (Cobb, 1974). The main controversy in measuring life stress centers around the question, what is stressful about an event? Holmes et al. originally believed it was the amount of readjustment needed after an event occurred. Others believe it is the desirability or undesirability of the change, whether or not the event involves affil the < in a Burt 1918 impe it ‘ tha and unc an an de be t? CI 16 affiliative needs, if the event has short or long term effects, the degree of threat involved, or the number of events experienced in a certain period of time (Suls, 1981; Rahe, 1979; Cobb, 1974; Burchfield et al., 1982: Sarason et al., 1978: Brown and Harris, 1978). These researchers have objectively rated the desirability, impact, etc. of certain life events; yet, there are those that believe it is primarily the individual's subjective interpretation of an event that leads to its stressfullness or not. Fontana et al. have based their research on the above hypothesis and measured individual perception of l) the desirability or undesirability of an event, 2) the amount of readjustment required for an event, 3) the degree of anticipation prior to an event, and 4) the amount of control over an event. In sum, results show the subjectively desirable events were also those that required little adjustment, could be anticipated, and fell under the control of the individual. That is, the four types of evaluations are linked in some way. Also, various combinations of the four subjective evaluations significantly correlated with different areas of psychosocial adjustment. For example, events perceived as undesirable are strongly related to psychological disturbance. The specific significant relationships are presented in Table 2 (Fontana et al., 1979). These results might imply that certain types of subjective evaluations predict poor psychosocial adjustment in certain areas. All of the studies reviewed in this section show a significant relationship between their own measure of stress and various types of 17 Table 2 Significant Correlations Between Four Types of Subjective Evaluations of Life Events and Psychosocial Adjustment1 Subjective Evaluation Psychosocial Perceived Perceived Perceived Perceived Adjustment Desirability Adjustment Anticipation Control Extent of: Psychological Disturbance -.35** .48** -.26* Interpersonal Involvement -.28* Alcohol Abuse .33** Outside Social Participation .29* Employment . -.41** * p<.TO **p<.05 1Subjective evaluations involve the individual's perception of the desirability of the event, the amount of readjustment it will require, how much the event was anticipated, and the degree of control he/she felt over the event. Adapted from Fontana et al., 1979. sul £01 del 1116: thl M was ser PM see of int 9631 Ext 19E C01 the Inc lnc 18 illness. However, Rahe's comment in 1979 still applies now. “The superior utility of any life change scaling method over a simple counting of the number of events, per unit time, has yet to be demonstrated convincingly" (Rahe, 1979; p. 4). This statement is not meant to negate the value of these different classifications but implies that they may be better understood as moderator variables in the stress-illness relationship. Moderator Variables in the Adult's Stress-Illness Relationship: A major study on moderator variables in the stress-illness area was conducted in 1981 by Cooley et al. Briefly, their results show low sensation seekers have a stronger relationship between life events and physical disorders compared to high sensation seekers. Low sensation seekers are described as avoiding novel situations. On a health locus of control measure, there was no significant difference between internals and externals in the stress-illness relationship. On more general measures of locus of control, previous studies have shown externals have higher correlations between stress and illness (Suls, 1981). Cooley also found significantly higher stress-illness correlations for introverts vs. extroverts, and for those who perceive the world through their senses vs. by intuition. Another variable that moderates the effects of stress on an individual is the extent of one's social support. This support includes relationships in the family, at work, at church, and with friends. Research results support the hypothesis that decreases in 19 social support in family and work environments would be linked to increases in psychological maladjustment over a one year period (Holahan et al., 1981). Lastly, age is also a characteristic to be considered in assessing the importance of a life event. Forty-five to sixty year old subjects showed an inverse relationship between life stress scores and age. This may mean stressful life events decrease with age. Or else, the event scales do not include events relevant to older people or events that are age related (e.g. empty nest, menopause, retirement). Age related events may allow "anticipatory coping" which may decrease their potentially stressful impact. Age might also mediate stress because it involves the accumulation of life experience and a larger repertoire of caping skills. Clearly, more detailed research needs to be done to fully understand what aspects of age are mediating stressful life events (Lazarus and DeLongis, 1983). What these studies do demonstrate is that individuals will have different characteristics that will make them more or less vulnerable to the negative effects of stressful life events. Life Stress and Symptomatology in Adults: One of the most thorough studies of life stress and symptomatol- ogy was done by Brown and Harris in 1978. They set out to study the etiology of depression in women and analyzed the effects of different types of events in great detail. Their samples included depressed patients and general population controls. The only events that an re de no pe lh pr pr 61 91 We Of US 20 occurred at a higher rate in the depressed sample involved moderate or marked long term threat and/or a focus on the individual alone or in a relationship. (Long term threat implies that the consequence of the event follows a week or longer after its occurrence.) These "severe" events usually involve some kind of loss or disappointment and are most likely to cause depression. Brown et al. also examined the additivity of events, their long and short term effects, and their causal vs. triggering effects. The results are as follows: in comparing number of events experienced by depressed patients and normals, 60% of depressed patients and 75% of normals experienced one severe event in a 38 week period. Twenty-one percent and 8%, respectively, experienced three or more severe events. The difference between depressed patients and normals, when comparing proportions experiencing three or more events, was significant at the .05 level. Significantly more depressed patients also experienced three or more nonsevere events compared to controls. Therefore, an increased number of events in a certain time period is more likely to produce maladjustment. By examining the number of weeks before depression onset that an event occurred, one can determine the short or long term effect of that event. Results show that 30% of the severe events occur within nine weeks of onset while 4.6% of these events occur 37-45 weeks before onset. That is, most severe events quickly lead to depression. This usually happens within nine weeks and in almost all instances, within six m0 term E havihl this, which deprl more was anal the roll sch per Eve not of VS th Ea BE 21 six months. The less severe events do not appear to have the same long term effects. This phenomena is described in more detail below. Finally, Brown and Harris wondered if their severe events were having a triggering or formative effect on depression. To determine this, the researchers computed a Time brought forward (be) equation which estimates how long it would have been before the onset of depression if an event had not occurred. The longer this time, the more likely the event played a formative role because the depression was hastened to such a large degree. The results of this type of analysis show severe events play a formative role in depression with the average be equal to 2.13 years. Minor events play a triggering role with a be averaging 10 weeks. The role of events in the onset of schizOphrenia differs somewhat. That is, only events in the three week period right before onset appear capable of influencing schi20phrenia. Events play a triggering role in this disorder. We may conclude that not only do different types of events play different roles in the onset of disorder, similar events may lead to different disorders (depression vs. schi20phrenia) based on another moderator variable. Minor events clearly function differently than severe events but there may be a connection between these two types of events. Specifi- cally, a minor event nay cause a woman to reassess the impact of an earlier major event. This may lead to hopelessness and depression. For example, going to a friend's husband's funeral may trigger a repressed memory of her own husband who died five years earlier (Brown and Harris, 1978). stres persc anvil thei occu are for bet' exa 22 All of the above findings complicate the relationship between life stress and illness. First, an event may or may not be stressful to a person depending on his individual characteristics and those of his environment. Second, events themselves may vary in effect depending on their severity of threat, their term of effect, and whether or not they occur alone or in close proximity to other events. These statements are based on the Brown and Harris research (1978). A possible model for these relationships is outlined in Figure 3. The following studies help to clarify the complex relationship between stress and symptomatology. More characteristics of stress are examined as well as other types of maladjustment. Like Brown and Harris, E. S. Paykel also conducted a study on depression but focused completely on the specific types of life events that preceded the depression (Paykel, 1974). In an overall analysis, depressives reported three times as many events as controls. Suicide attemptors report more events than depressives who in turn report more events than schizophrenics. There were also significant differences among these groups on the category or type of event they experienced. That is, depressives had significantly more exits from their social field than controls. Depressives also experienced significantly more negative events, and health, and law-related events. In comparing both experimental groups, depressives reported experiencing more exits, negative events, and finance and health related events compared to the schizophrenic group. According to Paykel, 23 a.;m=o_ua_a¢ mma=___-mmocum oc.a men an .aua: .m oe=m_a moucaomoc .o—couue coguo .ucoaaam —o_uom paau.>puc. us“ go» moucoauomcou _~=c¢>.u=* on» we au__ucomcoa covuo=u_m age gap: ucwaou m.u—:u< T popcmo eo_aa=a.m as» no acoEmmomm< "mcouuou m>.u—:mou aco>o may; m__.xm mcwmmu xcoumwg an»; aeoEQopo>ou so macaw ao< "mu.um.couuocego w._a=u.>.u:~ new in 1 Pepi 011$ exa at tel ex 1n ”E U . W- i 24 These findings do give some indication of specific relationships. Only certain kinds of events precede depression. It is not just a question of magnitude of life change: the direction of the change and its desirability are also important. However, the link between event and disorder is far from exact. (Paykel, 1974; p. 138) Rahe did a study comparing life stress in schizophrenics and neurotics and found similar results. Both groups reported an increase in the number of life events prior to symptomatic onset, with neurotics reporting twice as many life changes during the three months prior to onset (Rahe, 1979). Sarason et al. expanded the types of maladjustment they were examining but limited the type of events they focused on. They looked at the relationship between adjustment and positive, negative, and total amount of change. Results showed one significant correlation with positive change: positive change is significantly related to extraverted expression. Significant correlations with negative change include greater trait and state anxiety, social nonconformity, neuroticism, current depression, and an increased external locus of control. The only significant correlations for total change scores were with the two anxiety measures (Sarason et al., 1978). The most general conclusion to draw from all of these studies is that different types of stressors influence different types of maladjustment. The specific relationships are not clear and need to be focused on more directly. One such relationship may be the interaction between an individual's perception of stress and pathology. The fairly consistent relationship between an external locus of control and incre event lndll extel late llOl‘lll res 9P0 add and COO 25 increased pathology may indicate that individuals who see stressful events as out of their control may be more vulnerable than those individuals who believe they have control over those same events. This external locus of control is also found in alcoholic families (reviewed later) making it another area to explore when comparing alcoholic and normal control families. Review of Alcoholism Literature Research in this area has produced varied and often contradictory results. Part of this is due to poor methodology: few matched control groups, strong reliance on retrospective or self-report data. In addition, samples include alcoholics at different stages in the disease and children of different ages. However, there are results that consistently reappear in the research and are presented below. Behavior Problems in Children of Alcoholics: [TE] presenting problems of these children cover a wide range and fall into the following general categories: physical complaints, identity problems, emotional problems, behavioral problems. The identity problems include lower self-esteem, a reluctance to grow up and assume age appropriate roles, more external locus of control, an inability to form relationships, and confusion over sexual identity (Blane and Hewitt, 1979; Jacob et al., 1972). Physical problems moi de 111' 26 include increased numbers of accidents and illnesses, headaches, and increased fatigue (El-Guebaly et al., 1977: Jacob et al.. 1978; Nylander et al., 1982). The behavioral problems presented cover the widest range of complaints. They include underachieving in school, enuresis, nightmares, nailbiting, speech disorders, impulsiveness, hyperactivity, and more aggressive acting out. This latter problem involves temper tantrums, fights with peers, and delinquency that involves the police and courts (Blane and Hewitt, 1977; Herjanic et al., 1977; Jacob et al., 1978; El-Guebaly et al., 1977). Lastly, the more emotional symptoms include neurosis, hysterical symptoms, depression, increased dependency, anxiety, emotional lability, and withdrawal (Blane and Hewitt, 1977; Jacob et al., 1978; Nylander et al., 1982). None of the above problems have been consistently linked to the severity of the alcoholic's drinking or to any set type of family environment; yet, general patterns of parent and family interactions and characteristics have been formed. One aim of this study is to examine the behavioral and emotional problems exhibited by these high-risk children and any link that may exist between their symptomatology and the stress present in their family. Characteristics and Behavior of the Alcoholic Father and the Nonalcoholic Wife: Early research indicated that wives of alcoholics had a certain personality type--domineering and "masculine“. The more recent 1‘91 f0 27 research takes a more social/environmental perspective and offers the following theories. Jacob et al. demonstrates moderate support for the "stressed wife" theory, e.g. a wife's tendency to ignore changes in her husband's drinking behavior is seen as adaptive--it is a realistic recognition of his undependability. Ablon's study of Al-Anon wives might support this theory. She proposes that wives' behavior will change over time, and may change with the husband's drinking stage. Five reasonably separate types of behavior observed include 1) family protectiveness, 2) withdrawal within the marriage, 3) attacking others, 4) acting out, 5) safeguarding family interests. Adding to the stress theory, more women married to alcoholics vs. nonalcoholics reported having an inadequate mother and an unhappy childhood. This also made them more likely to experience both personal sources of stress and symptoms of personality disorder. Wives of alcoholics without unhappy childhoods managed to handle their situation without the personality decompensation (Ablon, 1976). Perhaps these women never learned good coping skills as children so any added stress caused greater maladjustment. Ablon also presents characteristics of an alcoholic father. First, he has an inability to take appropriate responsibility within the family. Second, he lacks self-discipline and third, he is overly dependent. Fourth, the alcoholic father is preoccupied with himself, and fifth, suffers from a sense of inadequacy. Sixth, he has a negative attitude toward authority as well as taking an unrealistic and 28 immature approach to things. Finally, he has limited interests and relates to other pe0ple in a superficial manner (Ablon, 1976). These researchers have listed the above as characteristic of the alcoholic and his spouse. Even so, these behaviors/characteristics/ events seen in marriages complicated by alcoholism may be similar to those seen in other marriages. That is, the alcoholism is an added complication, or stress, but it does not necessarily influence parents' behavior (Jacob et al., 1978). Whatever the cause, individual behaviors do influence marital and family interactions. 'This raises questions about the interaction of alcoholism, stress, and behavior. Does alcoholism cause more or different kinds of stress in families or does it influence how people react to already present stressors and modify their behavior? The current study will begin to explore these questions. Interactions and Characteristics of Alcoholic Families: <<:The marital relationship between an alcoholic and his wife is full of conflict, with poor communication and resolution of that conflict. Alcoholic couples do report more frequent disagreements and quarrels as compared to normal population controls. These quarrels are character- ized by physical or verbal abuse, silence, walking out, or moodiness (Ablon, 1976). Any attempt to settle conflicts meets with little success because of the rigid, competitive style of communication these couples have. Each spouse is sending "one-up“ messages to the other and tries to gain control of the situation. The alcoholic commonly 29 uses a "responsibility-avoiding" style in an attempt to obtain control through passive, dependent-appearing ways. On the other hand, the nonalcoholic spouse uses an opposite, “responsibility accepting" style to gain control with action and independence. With this competitive but complimentary manner of communicating, the alcoholic couple have difficulty in achieving joint goals (Gorad, 1971). [Eihldren of these marriages also see their parents in conflict and rate their families significantly lower in harmony than children of nonalcoholics (Jacob et al., LEE8E3> In addition to high levels of conflict, these families often include fathers who change jobs frequently. They also have more separations and divorces, more mothers working outside the home, and parents who demonstrate inconsistency, unpredictability, and a failure to fulfill their parental responsibilities (Blane and Hewitt, 1977; Jacob et al., 1978; Zucker, 1976). .40 with the scale to which it belonged. Sixty-six of the remaining items measure hostility and 9 measure guilt. The social desirability of this instrument was determined by having another sample rate how desirable each item was. Then, the original sample's response were used to determine the probability of endorsement for each item, 5 = .27 for men and .30 for women. Both of these correlations are small but significant. Next, a factor analysis was run and produced two factors, an "emotional" or attitudinal component and a "motor" component that includes various aggressive behaviors. Factor one includes the subscales of assault, indirect aggression, irritability, and verbal aggression. Factor two includes the resentment and suspicion subscales (Buss et al., 1957; Buss et al., 1962). Relationships between a psychiatrist's rating and other hostility measures were also examined. The psychiatrist's ratings were based on 56 the patient's history and a lengthy interview. Significant correlations appeared for the irritability, negativism, resentment, and verbal hostility scales. Also, the Elizur measure of hostile content, the Iowa Picture Interpretation Test (IPIT), a scrambled sentences task, and an operant conditioning task were all significantly correlated with the BDHI. The above instruments are all measures of aggressiveness. By examining the patterns of results, the authors conclude that patients' irritability and negativism before they are hospitalized is related to measures of aggressiveness (Buss et al., 1962). (3) Measurement of Antisocial Behavior The Antisocial Behavior Scale (Zucker & Noll, 1983) represents a modification of an earlier instrument, the Adolescent Antisocial Behavior Questionnaire (Zucker & Barron, 1973), that incorporates 18 of the items from this earlier inventory, as well as a larger number of social and antisocial items that are more suitable for an adult population. These items were drawn from a variety of sources, including antisocial items from the Diagnostic Interview Schedule (Robins, 1981) and from areas of behavior included in antisocial personality disorder, as described in the Research Diagnostic Criteria (Feighner, Robins, Guze, Woodruff, Winokur and Munoz, 1972). The final instrument (Appendix 4) is composed of 46 items that have been categorized by content into 9 subscales including parental defiance, sexual behavior, delinquent behavior, leaving the field, serious physical aggression, excitement and sensation seeking, job related anti-social behavior, school related anti-social behavior, and 57 trouble with the law. Each individual item is rated for the frequency the respondent engaged in that activity. Again, both parents complete this questionnaire. Psychometric pr0perties of the instrument appear adequate. Test-retest reliability over a four week interval with a college student population is 0.81: the coefficient alpha is 0.84 (Zucker & Noll, unpublished manuscript). Due to the nature of these items, social desirability may be a factor in responses. That is, one might assume subjects would under report engaging in the listed activities because of their undesirable nature. To minimize this effect, the directions were phrased in an accepting manner with the expectation that many people engage in the following behaviors. (4) Measurement of Health History The final questionnaires used in this study are health history forms completed by both parents. This extensive self-administered history questionnaire was developed by the Rutgers Longitudinal Study (Carpenter and Lester, 1980) to assess health and illness status in fifteen areas: hospitalization history; current medication use; allergies; prior illnesses; skin and hair problems; eye, ear, nose and throat symptoms; heart and lung; G.I. tract; urinary tract; skeleton and joints; nervous system; alcohol and drug use; general health care patterns; diet and weight control; physical fitness activities. The wife's form, in addition, contains questions regarding the target child's birth and early developmental history. Chapter 3 Results Analysis: The overall design of this study is a cross-sectional design with matched experimental and control group conditions. The high-risk and community control families were compared on measures of stressful life events and symptomatology. Group differences on these variables were determined by a matched pair analysis of variance (BMDP-ZV) because of the careful matching of control and experimental families (Table 3). Matched pairs analyses of covariance controlling for chronological age of parents and/or children were also performed to determine if differences were significant with this factor controlled for. Results from the analysis of variance and covariance were quite similar. Certain variables subjected to the above analysis showed large variances between groups. Levine's test for homogeneity of variance was used on any variables where group variances differed by a factor of ten or more. Few variables violated the assumption of homogeneity of variance; however, analysis of variance with logarithmic transforma- tions were done when this assumption was violated. After controlling for the large number of computations, no significant results were obtained. Thus this analysis did not influence the final results. 58 59 Life Stress Measures: In the first hypothesis it is predicted that parents and children of high-risk families will experience more life stressors than normal controls. The data gathered by the Holmes 8 Rahe SRRS and a slightly modified version of Coddington's life events measure1 were used to test this idea. Various indices of life stress were derived from the above measures: number of life change units (LCU's) and frequency of life stressors in the last year, last six months, between six and twelve months ago, for husband and wife together; amount of LCU's in the areas of health, work, home and family, personal and social realm. No significant differences were found with an analysis of covariance between high-risk and control adults on these variables. Nonetheless, all of the differences were in the predicted direction 1Two life events were added to Coddington's scale, item numbers 10 and 19: "My child has been involved in a serious accident during the past year“ and “One of my child's pets died during the past year". These two items are from the Life Events Questionnaire developed by N. Garmezy for Project Competence at the UhiVersity of Minnesota. This measure was also adapted from Coddington and Holmes and Rahe. Life change unit values were determined for each event by selecting two events this author felt would be more and less stressful than each new event. The greater and lessor events were temporarily given LCU values of 10 and one respectively. Next, two independent licensed clinical psychologists assigned an LCU value between one and ten to the new events depending on how stressful they felt each event was in comparison to the already weighted events. The newly assigned values from the two raters were averaged for each event and the average was used in the computation of actual LCU values. Specifically, an equation was set up where the average value of the new event divided by the difference between temporary values of more 8 less stressful events equals the actual LCU value of the new event divided by the difference between actual LCU values of more 8 less stressful events. 60 with high-risk adults reporting more life stressors. Sign tests show that these trends are significant for amount of stress (NéB, i = 0, p$.004) and by the category of stress (N_= 5, §_= 0, p$.03). This same pattern of results is found when amount of stress is measured in six-month versus twelve-month intervals. Means and standard deviations for the above analysis may be found in Table 7. There was a significant difference in life stressors reported for the children, with control children experiencing a higher frequency of stressors in the last six months when compared to the high-risk children, [(1,8) = 9.26, pk.02. Based on the frequency of occurrence for individual events (Appendix 2), the three events which account for most of the difference are: #B-My child has had at least one outstanding personal achievement during the past year, #6-There has been a change (up or down) in the family's financial status in the past year, and #16-My child has a new brother or sister who was born in the past year. Adult Symptom Measures: The second hypothesis states that alcoholic parents show more symptoms of both depression and hostility. No statistically significant differences were found between groups on either the Beck Depression Inventory or the Hamilton Rating Scale for depression (Tables 8A thru 80). In addition, no significant differences appeared on the Buss-Durkee Hostility Inventory. 61 Table 7A Amount of Life Stress Experienced by Adults in Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Velue1 (n:9) (n=9) F-Parent F-Risk F-PxR Overall Life Stress Measures Total life change Mothers 1;. 333.33 248.67 units (LCU's) 5.0. (125.95) (110.24) experienced in __ 0.56 1.36 3.55+ the last year Fathers X 347.33 246.89 - family 5.0. (138.56) (107.43) Frequency of life Mothers ‘7' 10.78 9.33 events experienced 5.0. (4.44) (3.71) in the last year __ 0.64 0.42 0.95 - family Fathers X 10.78 9.22 5.0. (4.44) (3.53) LCU ' s experienced Mothers )7 193.22 142.22 by each adult in 5.0. (84.43) (88.42) the last year __ 0.35 1.03 0.07 Fathers X 237.78 175.11 5.0. (145.78) (73.58) Frequency of life Mothers 'x' 5.22 5.33 events experienced 5.0. (3.56) (2.60) by each adult in __ 0.71 0.35 0.08 the last year Fathers X 7.89 6.67 5.0. (4.54) (2.24) 62 Table 78 Amount of Life Stress Experienced by Adults in Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Value1 (n=9) (n=9) F-Parent F-Risk F-PxR Homogeneous Content Scales of Life Stress Health LCU's Mothers '7 36.33 16.44 experienced by 5.0. (36.35) (18.59) each adult in __ 2.82 3.21 0.99 the last year Fathers X 36.33 16.44 5.0. (36.35) (18.59) Work LCU's Mothers '7 77.11 49.67 experienced by 5.0. (62.52) (46.98) each adult in __ 2.69 0.60 0.78 the last year Fathers X 77.11 49.67 5.0. (62.52) (46.98) Financial LCU's Mothers '2 44.00 29.00 experienced by 5.0. (18.74) (25.17) each adult in __ 1.90 2.04 0.19 the last year Fathers X 44.00 29.00 5.0. (18.74) (25.17) Home a Family Mothers 'R 110.09 104.44 enced by __ 2.76 0.10 0.12 each adult in Fathers X 125.00 104.44 the last year 5.0. (60.42) (69.46) Personal 8 Mothers '7 17.56 22.22 Social LCU's 5.0. (17.93) (12.17) experienced by __ 9.19* 1.50 1.86 each adult in Fathers X 80.33 45.55 the last year 5.0. (61.10) (24.36) *p<.05 + p<.10 1F-value = main effect for parent and risk status and interaction from matched pairs analysis of covariance (BMDP-ZV) with adult age as covariate. 63 Table 8A Symptomatology Reported by Adults in Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Vslue1 (0:9) (n:9) F-Parent F-Risk F-PxR Depression Current Mothers .X 3.33 2.89 depression - 5.0. (3.46) (2.20) Beck Depression __ 0.03 0.73 0.64 Inventory Fathers X 2.00 2.67 5.0. (2.12) (3.97) Current Mothers IX 6.56 5.44 depression - 5.0. (4.39) (3.78) Hamilton Rating __ 0.95 1.62 0.01 Scale Fathers X 7.44 5.67 5.0. (4.61) (5.24) Worst ever Mothers 1? 18.67 19.44 depression - 5.0. (9.31) (9.79) Hamilton Rating __ 0.16 1.46 0.15 Scale Fathers X 14.89 12.67 5.0. (11.73) (10.61) 64 Table 88 Symptomatology Reported by Adults in Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Value1 (n=9) (n=9) F-Parent F-Risk F-PxR Antisocial behaviors2 Total Mothers X 8.67 4.89 Anti-social S.D. (4.30) (2.20) Behavior __ 7.66* 11.08* 0.20 Fathers X 16.22 10.67 5.0. (5.74) (5.38) Parental Mothers 3? 2. 33 1.89 Defiance 5.0. (1.41) (1.05) __ 0.41 0.57 0.01 Fathers X 2.89 2.44 5.0. (1.27) (1.13) Sexual Mothers X 0.11 0.11 Behavior 5.0. (0.33) (0.33) 1.04 0.80 1.36 Fathers '5? 0.67 0.33 5.0. (0.87) (0.71) Delinquent Mothers 11' 0.89 0.67 Behavior 5.0. (0.78) (0.50) 3.23 10.74* 0.77 Fathers '1? 1.89 1.11 5.0. (0.93) (1.05) Leaving Mothers IX 0.89 0.44 the field 5.0. (0.78) (0.53) __ 0.32 8.32* 0.49 Fathers X 1.33 0.56 5.0. (1.22) (0.73) Serious Mothers 'X 1.22 0.33 Physical 5.0. (0.83) (0.50) Aggression __ 3.51+ 7.38* 0.00 Fathers X 2.33 1.56 5.0. (1.41) (1.42) Excitement Mothers '1? 0.44 0.44 And sensation 5.0. (0.53) (0.53) Seeking __ 3.67+ 0.01 0.09 Fathers X 1.33 1.33 5.0. (0.87) (0.71) 65 Table 88 (cont'd.) Excitement Mothers X 0.44 0.44 And sensation 5.0. (0.53) (0.53) Seeking __ 3.67+ 0.01 0.09 Fathers X 1.33 1.33 5.0. (0.87) (0.71) Job related Mothers 7 0.89 0.11 Anti—social 5.0. (0.78) (0.50) Behavior __ 1.19 12.58** 0.28 Fathers X 1.22 0.67 5.0. (0.97) (.083) School related Mothers '1? 1.22 0.78 Anti-social 5.0. (0.83) (0.83) Behavior __ 4.941’ 12.34** 0.44 Fathers X 3.00 2.00 5.0. (1.73) (1.41) Trouble with Mothers “X 0.89 0.22 the law 5.0. (0.78) (0.44) 21.62** 6.369 0.62 Fathers 7 2.44 1.22 5.0 (1.59) (1.09) 66 Table 8C Symptomatology Reported by Adults in Alcoholic and ommunity Control Families Community Alcoholic Control Families Families F-Value1 (n=9) (n=9) F-Parent F-Risk F-PxR Hostility3 Assaultive Mothers 'X 3.78 3.22 Hostility 5.0. (2.64) (2.05) 0.44 1.06 0.01 Fathers X 4.33 3.78 5.0. (1.87) (2.49) Indirect Mothers 'X 4.89 5.00 Hostility 5.0. (1.69) (1.22) 2.14 0.01 0.06 Fathers 7? 4.33 4.11 5.0. (1.80) (1.76) Irrit ability Mothers X 5.00 5.67 5.0. (2.60) (2.12) 3.13 1.07 3.03 Fathers 1? 5.56 3.44 5.0. (2.60) (2.74) Negativism Mothers 'X 1.89 1.67 5.0 (0.60) (1.12) 0.23 1.40 0.09 Fathers 7 2.67 2.00 5.0. (1.66) (1.12) Resentment Mothers 7? 2.56 2.11 5.0. (2.30) (2.03) 0.99 0.28 0.21 Fathers 1? 2.00 1.89 5.0. (1.07) (2.32) Suspicion Mothers X 1.44 3. 33 5.0. (0.53) (2.00) 3.82+ 1.74 12.93** Fathers 1T 2.44 1.67 5.0. (1.94) (1.87) Verbal Mothers X 6. 11 4. 67 Hostility 5.0. (2.57) (1.58) 2.37 3.94+ 0.09 Fathers X 7.78 6.00 5.0. (2.33) (2.87) Guilt Mothers X 3.22 3.33 5.0. (1039) (1001) 0.26 0.17 0.02 Fathers 7? 3.67 3.11 5.0. (2.12) (1.96) 67 Table 80 Symptomatology Reported by Adults in Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Value1 (n:9) (n:9) F-Parent F-Risk F-PxR Physical health“ Number of Mothers 'x' 0.22 0.00 general health 5.0. (0.44) (0.00) symptoms __ 3.27 16.94" 0.04 Fathers X 0.78 0.78 5.0. (0.67) (0.83) Number of Mothers 7 0.11 0.11 urinary 5.0. (0.33) (0.33) tract problems _. 0.01 4.45+ 0.12 Fathers X 0.67 0.44 5.0. (1.19) (0.73) Nmber of Mothers Y 2.56 2.44 hours of sleep 5.0. (0.53) (0.53) per night __ 0.13 24.98** 0.01 Fathers X 3.22 2.88 5.0. (0.83) (0.60) * p<.05 ** p<.01 + p<.10 1This F-value is the main effect for parent and risk status and interaction computed by a matched pairs analysis of covariance (BMDP-ZV) with age as covariste. zAll measures of anti-social behavior were derived from the number of items checked in each category on the Anti-Social Behavior Scale (A58). 3A11 measures of hostility were derived from the Buss-Durkee Hositility Inventory (801). “All measure of physical health were derived from the Health History Forms. 68 Nonetheless, when a sign test was performed on all of the symptom measures together, a significant difference was found with adults in alcoholic families exhibiting more symptomatology (N = 22, X_= 1, 26.001). Statistically significant differences in means were found on the Anti-Social Behavior scale. High-risk adults reported significantly more overall antisocial behavior as well as more delinquent type behavior, leaving the field, serious physical aggression, job and school related antisocial behavior, and trouble with the law. Specific statistics may also be found in Tables 8A-BD. Child Symptom Measures: The third hypothesis predicts that children of alcoholics show more symptoms of impulsivity and aggression as well as fewer peer relations when compared to controls. No significant differences appeared between risk groups on either the Carey-McDivett child temperament questionnaire or on the Achenbach Child Behavior Checklist. Likewise, no significant trends appeared on any of these measures. That is, children of alcoholics apparently do not show higher levels of impulsivity and aggression or poorer peer relations; these results are summarized in Tables 9A and 9B. The Relationship Between Life Stressors and Symptoms: Hypothesis four predicts a positive relationship between parents' level of stress and children's level of stress and symptomatology. Past research has shown a link between stress and illness in an 69 Table 9A Symptomatology Experienced by the Target Children of Alcoholic and Community Control Families Community Alcoholic Control Families Families F-Value1 (n:9) (n:9) F-Parent F-Risk F-PxR The Carey-McDevitt Behavioral Style Questionnaire (BSQ) Activity level Mothers X 2.67 2.44 T-score 5.0. (0.71) (0.73) _ 0.37 0.00 1.16 Fathers X 2.56 2.78 5.0. (0.73) (0.83) Rhythmicity Mothers '7 2.22 2.67 T-score 5.0. (0.83) (0.87) __ 2.48 2.79 0.04 Fathers X 2.56 3.11 5.0. (0.88) (0.78) Approach/ Mothers .X 2.33 2.89 withdrawal 5.0. (1.00) (0.60) T-score __ 4.26+ 4.583 0.00 Fathers X 1.78 2.33 5.0. (0.44) (0.87) Adaptability Mothers 'i 2.67 2.44 T-score 5.0. (1.60) (1.01) __ 0.39 0.27 0.20 Fathers X 2.33 2.33 5.0. (0.71) (0.87) Intensity Mothers 'i 2.56 2.22 T-score 5.0. (1.13) (0.83) __ 3.57+ 1.19 0.00 Fathers X 2.00 1.67 5.0. (0.87) (0.71) Mood T-score Response Threshold T-score Persistence T-score Distractability T-score Mothers Fathers Mothers Fathers Mothers Fathers Mothers Fathers 5.0. 5.0. 5.0. SOD. SOD. 5.0. 5.0. 5.0. 70 2.67 (1.22) 1.89 (0.78) 2.11 (0.93) 1.78 (0.83) 2.78 (0.97) 2.67 (1.22) 2.78 (0.67) 2.67 (0.50) Table 9A (cont'd.) 2.33 (0.86) 2.33 (0.71) 2.56 (1.01) 2.44 (0.53) 2.44 (1.01) 2.67 (0.87) 2.56 (1.01) 2.33 (0.71) 2.02 8.24* 0.03 0.80 0.02 0.01 0.54 0.91 2.48 2.80 0.44 0.00 71 Table 9B Symptomatology Experienced by the Target Children of Alcoholic and Community Control Families Communitd Alcoholic Control Families Families F-Value1 (n:9) (n:9) F-Parent F-Risk F-PXR The Achenbach Child Behavior Check List (CBCL) Depression Mothers X 55.11 51.78 T-score 5.0. (14.32) (4.58) __ 0.96 0.42 0.49 Fathers X 51.11 50.67 5.0. (7.69) (9.10) Imaturity Mothers -X- 56.44 59.33 T-score 5.0. (9.07) (8.05) __ 2.43 0.14 0.27 Fathers X 53.89 53.67 5.0. (7.18) (9.19) Sex-related Mothers 'X 64.67 59.22 T-score 5.0. (12.01) (5.56) _ 3.23 0.68 1.71 Fathers X 58.11 60.44 5.0. (4.11) (8.44) Schizoid Mothers 'x' 59.33 59.44 Symptoms 5.0. (6.40) (5.25) T-score __ 2.94 0.16 0.12 Fathers X 56.11 57.67 5.0. (4.81) (5.81) Aggression Mothers 7 53.78 55.78 T-score 5.0. (16.53) (6.94) _ 0.28 0.02 0.94 Fathers X 54.22 51.33 S.D. (7.45) (10.56) Delinquency Mothers 3? 53.22 54.67 T-score 5.0. (6.72) (7.09) __ 1.32 0.23 0.00 Fathers X 54.89 56.00 5.0. (4.70) (6.95) 72 Table 98 (cont'd.) Other Mothers 'X 5.00 7.11 Problems 5.0. (3.67) (3.29) 0.09 4.88+ 0.14 Fathers it 5.22 6.44 5.0. (2.64) (3084) Nunber of Mothers X 25.11 28.67 Total 5.0. (16.50) (8.01) Problems 0.95 0.31 0.05 Fathers X 22.22 24.22 5.0. (8.22) (13.94) Internalizing Mothers 'X' 54.78 55.00 Problems 5.0. (12.47) (6.08) T-score 2.31 0.01 0.01 Fathers )( 50.89 50.44 5.0. (7.78) (9.51) Externalizing Mothers 'X 51.22 54.67 Items 5.0. (12.49) (5.98) T-score 0.13 0.09 0.93 Fathers )( 52.56 51.33 5.0. (5.80) (8.94) Withdrawal Mothers X 55.11 56.67 T-score 5.0. (8.07) (6.44) 0.27 0.00 0.51 Fathers 7? 55.67 54.11 5.0. (7.57) (8.88) Somatic Mothers X 59.67 64.11 Couplaints 5.0. (6.20) (9.20) T-BCOI‘B 2000 0096 1070 Fathers 11 58.11 57.67 5.0. (6.84) (3.50) Number of Mothers X 1.11 1.11 Friends 5.0. (0.60) (0.60) 0.22 0.13 0.13 Fathers X 0. 89 1.00 5.0. (0.78) (0.71) Amount of Mothers X 1.56 1.67 Time spent 5.0. (0.73) (0.71) With friends 0.64 0.80 0.31 Fathers )1 1.22 1.44 5.0. (0.97) (0.88) How well the Mothers 'X 1.67 1.44 Child gets 5.0. (0.50) (0.53) Along with 1.18 0.00 1.39 Friends Fathers )( 1.22 1.44 5.0. (0.44) (0.53) + p<.10 9 p<.05 1This F-value is the main effect for parent md risk status md interaction computed by a matched pairs analysis of covariance with child's chronological age a covariate. 73 individual person. Current results support this finding (Appendix 3) and show a relationship between the stress experienced by a parent and the stress and symptomatology experienced by his/her child. In this study, the amount of stress in LCU's experienced by both parents over a one year period1 was significantly correlated with the amount of stress in LCU's experienced by the child over the same year, R(l,18)=.35, £5.02. (See Table 10 for a breakdown of these results.) This same measure of family life stress was also significantly correlated, in the positive direction, with parent report measures of child activity level, distractability and response threshold level. These results are outlined in Table 11. The above relationships appear stronger between mother and child vs. father and child as well as appearing stronger for high-risk vs. control families. However, these differences are not yet conclusive and should be viewed with caution. Individual Perceptions and the Stress-Symptom Relationship: Hypothesis five predicts that individual perceptions moderate the stress-symptom relationship. When amount of stress was weighted by individual perceptions of the immediate reaction to the event, adequacy 1After finding no differences between mothers' and fathers' individual levels of stress, this researcher decided to use the best estimate of stress experienced by the family as the stress measure in these analyses. This variable includes stressful events reported by one or both parents with no event counted more than once. 74 Table 10 The Relationship Between Family's Life Events and Children's Life Events1 Best estimate of total LCU's experienced by the family in the last year Community Alcoholic and Control Alcoholic Families Control Families Families Combined (n:9) (n:9) (n=18) Number of LCU's Experienced by the Target child in .51+ .21 .32+ One year Number of events Experienced by the .54+ .18 .25 Target child in One year Number of events Experienced by the -.03 .49+ .07 Target child in The last six months + p<.10 1Pearson Product Moment correlations were computed between a family life events measure, total life change units from the adults in one year, and various various measures of child life stress. 75 Table 11 The Relationship Between Adults‘ Life Events/Stressors and Children's Symptomatology1 Community Alcoholic Control Families Families (n:9) (n:9) Family Stress Total number of behavior Mothers -0.01 0.03 problems (CBCL) Fathers -0.16 0.22 Number of other Mothers 0.19 0.06 problems (CBCL) Fathers -0.23 0.364' Activity level Mothers 0.03 0.61* raw score (850) Fathers 0.20 0.35+ Rhythmicity Mothers -0.01 -0.31 raw score (850) Fathers 0.471' 0.06 Approach/Withdrawal Mothers -0.36+ -0.52+ raw score (BSQ) Fathers 0.16 -0.22 Adaptability Mothers -0.23 -0.13 raw score (850) Fathers 0.37+ 0.13 Intensity Mothers -0.25 —0.31 raw score (850) Fathers -0.32* 0.20 Mood Mothers 0.08 -0.12 raw score (850) Fathers 0.02 0.22 Persistence Mothers -0.331' -0.23 raw score (850) Fathers -0.04 -0.38+ Distractability Mothers 0.58* 0.48+ raw score (850) Fathers -0.23 0.28 Response threshold Mothers 0.62* 0.41 raw score (850) Fathers -0.30 0.65“ + p<.10 * p<.05 1These results are from Pearson Product Moment correlations between various measures of child symptomatology and the best estimate of total LCU's experienced by the family in the past year. 76 of preparation, degree of anticipation, and amount of control they felt, analysis of covariance still failed to show significant differences in stress levels between high-risk and control groups (Table 12). To determine the mediating effect of individual perceptions on the stress-symptomatology relationship, partial correlations were computed between the weighted life stress scores and adult symptom scores with amount of life stress partialed out. Because of the number of correlations computed, the two significant correlations which appeared could be due to chance. However, the majority of partialed out correlations were larger than their zero-order counterparts, indicating that individual perceptions affect symptomatology independently of stress instead of in a linear fashion (Hunter, 1980). 77 Table 12 Amount of Life Stress - Corrected by Subjective Weighting - Among Alcoholic and Community Control Family Members Community Alcoholic Control Families Families F-Vslue1 (n:9) (n:9) F-Parent F-Risk F-PxR Subjective Perceptions Immediate Mothers '7 810.22 722.67 Reaction to 5.0. (412.20) (438.94) The life event __ 0.17 0.07 0.00 Fathers X 877.89 898.00 5.0. (625.64) (439.36) Amount of Mothers '§ 885.11 698.33 Preparation 5.0. (501.02) (405.42) For life event __ 0.19 0.44 0.01 Fathers X 936.00 809.44 5.0. (520.79) (394.84) Degree of Mothers 7 1004.00 709.78 Anticipation S.D. (627.67) (401.51) For life event __ 0.01 0.66 0.03 Fathers X 909.33 804.44 5.0. (520.57) (387.37) Amount of Mothers '7 749.33 587.22 Control felt 5.0. (336.04) (328.67) Over life event __ 0.89 0.43 0.00 Fathers X 793.00 779.44 5.0. (514.58) (505.47) lThis F-value is the main effect for parent and risk status and interaction computed by a matched pairs analysis of covariance (BMDP-ZV), with adult's age as covariate. Chapter 4 Discussion There is one immediate flaw in this research that was known at its inception. This is the small sample size which creates a high probability of type II error. Thus, except in cases of very strong effect, such a design is likely to yield a greater proportion of nonsignificant affects even when true differences may exist in the population being studied. There is a fair amount of evidence that suggests this is an issue for the current data set and it needs to be kept in mind in dealing with all that follows. It also needs to be clear at the outset that this problem was known at the beginning and that data are currently being collected to enlarge the sample size and rectify the problem. The careful matching of the high-risk and control groups on background characteristics was done with this in mind; controlling part of the error variance helps compensate for the small sample size. Nonetheless, in most respects, the current report needs to be regarded as a pilot study rather than as a definitive last word. Life Stress Findings: Adults: The hypothesis that adult alcoholic family members experience more life stressors than their controls received partial support from these data. This support comes from a significant sign 78 79 test for direction of effect, and so must be regarded as a tentative conclusion. More specifically, the data suggest that high-risk adults are experiencing greater amounts of stress in general; more Specifically, more stress appeared in the health, work, financial, family, and personal/social realms. It seems that all areas of life for the alcoholic family are more stressed. If this is in fact a replicable effect, a next step in the research would be to investigate the causal relationship between elements in the alcoholic's life and alcoholic consumption that connect the alcoholism to the stress. Children: The data on children's life stressors present a slightly different picture. The one significant difference in the measurement of the children's level of stress was in the opposite direction to the one predicted: parents reported that control children had experienced more stressors than high-risk children in the last six months. Presuming this is a real difference, a look at the three specific events which account for most of the variance in this factor may help to explain it. The three events (having an outstanding personal achievement, a change in the family's financial status, gaining a new brother or sister) may be seen as positive changes, indicative of a supportive family environment. The alcoholic family would probably not have as supportive an atmosphere which could facilitate this type of positive change or stress. In regard to level of stress, Hurme has noted that a certain amount of stress can act as a developmental 8O task. If this is the case, the higher level of stress reported for control children and the higher level of cognitive development reported on this same subsample by Noll (1983) are compatible results. Symptom Findings: Agglgg: The hypothesis that alcoholic family members exhibit more symptomatology also received partial support from these data. The four main areas of adult symptomatology measured include depression, anti-social behavior, hostility, and physical health. Significant differences between alcoholic and nonalcoholic adults were found, in the predicted direction, in the areas of antisocial behavior and health. Because of this, one could assume that these are extremely robust effects. On another level, a significant trend is also present when the various symptom measures are viewed together. In sum, it appears that higher levels of pathology in high-risk adults are real differences. Then, one again needs to question the role of alcoholism in this effect. Children: The hypothesis that high-risk children would exhibit more symptoms of impulsivity, aggressiveness, and poorer peer relations than their controls did not receive support from these data. This is unexpected considering past research has shown behavioral problems in children of alcoholics. There are a number of potential measurement and methodological problems that may be responsible for these results 81 and will be reviewed later. On the other hand, it is possible that this is a real effect and no behavioral or temperament differences exist between these two groups of children at this age. Stress-Illness Findings: The hypothesized link between stress and symptomatology also found support in these data. The significant positive relationship between stress and symptoms in adults was expected and helped to verify the findings of other researchers. However, the hypothesis that an individual's subjective perception of the stress experience would moderate this relationship did not receive support. In fact, this variable may not be a factor in the stress-illness interaction; conversely, subjective perception may be confounded by poor recall and/or cognitive dissonance -- the need to remember unpleasant past events as positive. This stress-illness connection is not as obvious when one looks across parent and child. Because this idea is a relatively new one, this study first looked at the relationship between parents' level of stress and children's level of stress. A significant and positive relationship was expected and found. In addition, this connection appears stronger in the alcoholic families as compared to their controls (Table 10). This may indicate that children's experiences in high-risk families are more influenced by the experiences of their parents. The relationship between parent stress and child symptomatology points to a similar conclusion. Again, the link is a 82 positive one and appears stronger in the high-risk families (Table 11). Yet, the stress-illness link is only significant for certain types of symptom behaviors: activity level, distractability, response threshold. These symptoms all have a common theme relating to behavior; they are connected to concepts like impulsivity and poor attention span. More simply, the results indicate that high levels of family stress are in some way related to having impulsive children. The data need to be considered as tentative until they are replicated. A Theoretical Framework: The picture of the alcoholic family emerging from these data is a complex and, as yet incomplete, one. The adults in these families are experiencing amounts of stress which are potentially higher than the average. Higher amounts of stress are coexisting with higher amounts of symptomatology, most obviously aggression and hostility, and depression in lesser amounts. It is also clear that stress and illness are influencing each other in these high-risk adults. Curiously enough, the above picture is not reproduced when the children in the alcoholic families are studied. At this point, the target children are not reported as experiencing more stress or pathology than their nonalcoholic counterparts and the link between stress and symptoms in children is not as strong as it is in the adults. Because adults, and parents in particular, are such a major part of children's lives in the preschool years, it makes sense to 83 investigate the interaction of stress and illness across parents and children. Again, this interaction is significant and complex. Higher amounts of parental stress are related to higher amounts of stress in children. In addition, higher amounts of parental stress are related to higher levels of activity/impulsivity in children. These results reinforce the idea that there is an influential bond between parents and children and might shed some light on the nature of this bond. More specifically, there could be an enmeshed character to it; whatever affects the parent affects the child too. And, if this bond is indeed stronger in alcoholic versus nonalcoholic families, the alcoholic family is the more enmeshed family. Following this line of reasoning, there may be some other characteristics of the alcoholic family that prevent them from being more autonomous. For example, poor coping skills and low levels of social support may cause an individual to turn to his/her child for support. In turn, in alcoholic families it may be more difficult for the child to resist this dependency of his parent. This might happen for a variety of reasons: the child himself has strong dependency needs, is depressed, has limited cognitive skills that impair his ability to develop the concept of independence. Data from another section of the larger MSU Family Study suggest that dependency and depression are stronger factors in high-risk children and support the hypothesis that high-risk children are behind control children in their level of cognitive development (Noll, 1983). Although there are 84 current data to support the above model, it is only a model until more research is done in this area. Methodological Issues: The problems of a small sample and type II error were mentioned at the outset of this discussion. Yet, with a small sample there is also the possibility of a biased sample through selective recruitment, selective participation, 8 selective population, or some other unknown idiosyncratic characteristics of the sample. The potential bias in recruitment was eliminated by systematically going door to door in search of control families and going straight down the list of potential high-risk families. Selective participation can also be ruled out because of the 100% participation rate of alcoholic families and the participation of ten out of eleven control families approached. The pepulation is restricted given that the sample was selected from court records in a midwestern state and consisted of subjects who were principally blue-collar semi-skilled workers. In these respects, the results may not be generalizable to other social statuses, or conceivably to other geographic areas. Another potential problem is a bias in the data which are based on adult self-report. While it is true that the present data set is based heavily on self-report instruments, it is also true that both alcoholic and control families engaged in this task. Thus, although the data may not be as reflective of the true behavioral state of affairs in these families, it is reasonable, nonetheless, to believe that the 85 differences we found were in fact true perceived differences between the two groups. Measurement Issues: In trying to understand the pattern of results in this data set, the actual measures used must be considered. Our literature review on stress raised the question "What about a specific event makes it stressful?". No firm conclusion was reached so this researcher derived numerous types of variables from the Social Readjustment Rating Scale (SRRS) and the Coddington Life Events Questionnaire. Stress was then measured, across different time periods, by life change units, frequency of occurence, type of event, and subjective perceptions. None of these methods of measurement made a difference in the end result, for this data set. This may be because the differences found in this sample are real but are hindered by sample size etc. or that the measure of major life events does not touch what is stressful for these families. If this latter point is correct, the issue is a conceptual one and not a measurement one. For this reason, a measure of more frequent/common day to day stressors may be more useful in a study such as this. This is planned in future data collection on the project. The target children's age adds another problem to the measurement process. Specifically, the Achenbach Child Behavior Checklist (CBCL) was not designed for use with children under the age of four years. This makes it an instrument of lesser reliability and unestablished 86 validity for use with the present sample. Also, while there may be no behavioral or temperament differences between high-risk and control children between the ages of 2 1/2 to 6 years of age, differences may appear at later ages: as the high-risk children grow older, they will be exposed to more aggression and confusion in their families. In addition, there may be sleeper effects--what children are exposed to ggg_nay not show up as behavioral consequences until later. This line of reasoning is compatible with other research findings in that the behavior problems reported in the earlier literature appeared in children who were ten years of age and older. Future Directions: The next phase of this project is designed to address a number of issues raised in this chapter. First, the sample size is currently being increased in order to reduce the possibility of type 11 error. As mentioned earlier, a measure of life stress that taps more frequent and common events will be utilized as a way of approaching the issue of what is stressful for these families. In addition, measures of adult caping skills and social support networks will be added. Finally, the California Child Q-sort will be completed by two independent raters in order to have some observational data on the children's behavior and temperament. With these changes, it will be possible to further clarify differences in stress and symptomatology between alcoholic and nonalcoholic families, and to expand the model of the stress-symptom relationship. APPENDICES 3A1 3A2 38 4A 48 5A Appendix 1 CONTACT AND DATA COLLECTION SCHEDULE B/N/T sends initial contact/introduction letter from RAZ, or else makes phone contact (or home visit) directly. Phone contact made by core staff (B/N/T) and appointment is set up for initial contact-explanation interview (respondents' home). Home visit by B/N/T with both parents (child briefly accessible) to explain study, obtain consent, and collect initial background information to allow for quick screening for family appropriateness. If family agrees to take part, the following measures are obtained at this time: (1) chi1d(ren)'s verbal consent to participate; (2) signed project consent forms; (3) signed video consent form, if applicable; (4) Health Questionnaires -Hu 8 Wi-includes SMAST related items and birth and early developmental history on child.; (5) Demographic Background Questionnaire (Hu 8 Wi). Time: Approx. 45' to 1:30: if time not available, E may only be able to obtain consent forms and part of Health Questionnaires. These are of higher priority than the demographic information. If all questionnaires are not completed at this time, schedule the ones still to be done for session 38. E completes “First Visit Rating Form“ following visit 3A. Follow-up visit for medical history and Health Questionnaire and Demographic information if needed. Participant observation and video taping if family agrees (and this is applicable); if family refuses or this procedure is not applicable, then go to step 5. Otherwise, step 5 begins when initial observations and video taping are completed. If participant observation, Caldwell HOME is done by JB after first visit. (Otherwise see step 8). H0 8 W1 - Questionnaire session (done by CW) Administer (1) Beck Questionnaire; (2) Leisure time Inventory A58 (3) Drinking and Drug History; (4) Achenbach Child Behavior Checklist; (5) Carey and McDevitt Behavioral Style Questionnaire; (6) Work Satisfaction Inventory (administer only to respondents who work full or part time outside the home). Time: Approx. 2:00 to 2:30. 87 58 6A 6A1 68 7A 78 10A 108 88 Appendix 1 (cont'd) Child and RBN: If possible, done simultaneously with (5A) Home visit to administer Yale Developmental Battery. Mo 8 Child and DT: Trip to MSU; Lab visit for structured play evaluation and behavioral ratings (done by 2 raters). Time: Approx. 1:00. Resistance to temptation task. California Q-sort on child done by DT immediately following observation session. Mo 8 Child - Home visit. Cognitive assessment on child for alcohol concepts, impulsivity, etc. administered by RBN. Mo completes Coddington while child competes cognitive assessment. Calif. Q-sort on child done by RBN immediately following session 7A. Mo and Child - Home visit to obtain Caldwell HOME. (Done by DT or J8: Only necessary if step 48 not applicable). Home participant observation visit - JB and family; only done if 8 not applicable. Questionnaire and Interview Session #2 - At home; Hu and W1 and CW and RAZ. Questionnaires: Benjamin Interpersonal Rating Scales (Parts A, B, C, done in order of: Self Description; Spouse Description; Spouse's Description of Self; Life Events Questionnaire, Part A; Moos Family Environment Scale; Life Events Questionnaire, Part B; Buss-Durkee Questionnaire. (N.B. All of these questionnaires to be administered to Wi with CW while RAZ does Diagnostic Interview Schedule with Hu. When 015 is completed, the sequence is reversed and the W1 is interviewed while Hu competes questionnaires. (Questions administered during the Diagnostic Interview also allow for the completion of (1) Hamilton Rating Scale for Depression; (2) Temperament Rating Form; (3) WAIS Information and Digit Symbol subtests: and (4) Genogram.) Time: Approx. 2:30 to 3:00; if family is too fatigued this session is spread over two adjacent days. RAZ does Calif. Q-sort on Hu and W1 immediately after the interview. 89 Appendix 1 (cont'd) 11 Final home visit done on families where participant observation took place; also complete genogram with family. 12A Payment and feedback answering and ending session; also have Hu and Ni fill out (1) recontact forms, and if necessary; (2) genogram; 3) provide information on potential neighborhood control families: 4) leave taking. (Core staff). Time: Approx. 1:00. 128 For families with participant observation, CW does Calif. Q-sort on Hu and Wi immediately following session. Fathers N=9 Control Families (N=9 Mothers Appendix 2 Table 13A Fathers Individual Frequencies of Life Events (N=9 as Reported by Adults on the Social Readjustment Rating Scale High-Risk Families Mothers (N=9 3. 2. 1. 3. 2. 1. Time*: 4 003112 00010 20200 20200 000]] 00311.]. 101110 04.1.0] 112303 06403 0001.0 2010] 234567890123456789” 011 01.103211000202001 20 1'0 000 1112000110001010002000 026 0140310011130414211010031| 12402101001214030030110]ml. 00 02102220000030011000024 0111.11101011012011011010032 .110211131220120100210001] 01111]010120100010110011002 0221101312121304112222022 0130111201111302011113020 10040104422011112021000]2 03001011011101020111200]3 90 to twelve months ago . . . hecked by spouse but not 1nd1v1dual era to six months ago 2 Six C *1 2 3 91 Appendix 2 (cont'd.) Table 138 Individual Frequencies of Life Events as Reported for Children On the Coddington Life Events Questionnaire High-Risk Families Control Families (N=9) (N=9) Timer: 1 a e 1 2 2 Item # l l 0 l l 2 0 0 0 O 3 0 l 2 5 2 4 l O l 0 5 l 0 l l 1 6 l O l 4 l 7 0 l l 0 O 8 O 1 0 l 9 O 0 O l 10 l l l 1 2 ll 0 0 1 l 0 12 0 l O O 13 l l 1 l 0 l4 0 0 l 0 15 1 O l l 0 l6 0 O 0 4 l7 0 l 2 2 l 18 0 0 l O 0 *1 = Zero to six months ago 2 = Six to twelve months ago 3 = Not specified Appendix 3 Table 14 ”Best Estimate ofETotaT [00's Experienced by the Family in the Last Twelve Months Wife Symptom: (N=18) Depression: Current - Beck 0.20 Current - Hamilton 0.29 Worst ever - Hamilton 0.67 Hostility: Assault 0.37 Indirect 0.16 Irritability 0.16 Negativism 0.08 Resentment 0.07 Sus icion 0.00 Ver a1 0.47 Guilt -0.22 Anti-social behavior: Total anti-social behavior 0.55 Delinquent role 0.05 Leaving the field 0.19 Serious physical a gression 0.70 Job re ated 0.57 School related 0.24 Trouble with the law 0.73 Health: No. of illnesses 0.16 No. of other general symptoms 0.14 No. of eye, ear, nose, throat problems 0.03 No. of urinary tract problems -0.l6 Hours of sleep per night -0.24 *‘A' + * Husband (N=18) . . C dowowNO—I + 4. 00090090 0 (bamboo-Human Wife + Husband (N=36) + p<.10 *<.05 **p<.Ol 92 Appendix 4 The Antisocial Behavior Checklist Many of us have had adventures during our lives...times that were exciting and carefree, even though they may have been a bit impulsive or happy-go-lucky. Please read each of the following items. Indicate (with a check) if you have ever done any of the following activities and how often. NEVER - you have never done this RARELY - done once or twice in your life SOMETIMES - done three (3) to nine (9) times in your life OFTEN - done more than ten (10) times in your life N R S O E A O F V R M T E E E E R L T N Y I M E S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Skipped school without a legitimate excuse for more than 5 days in one school ear. . Been suspended or expelled from school for fighting. Been suspended or expelled from school for reasons other than fighting. Lied to a teacher or principal. Cursed at a teacher or principal (to their face)? Hit a teacher or principal. Re eated a rade n school. Ta en part n a ang fight. "Beaten up" anot er person. Broken street lights, car windows, or car antennaes just for the fun of it. Gone for a ride in a car someone else stole. leased or killed an animal (like a dog or cat) just for the fun of it. Defied your parent's authority (to their face). Hit your parents. Cursed at your parents (to their face). Stayed out overnight without your parent's permission. un away from home for more than 24 hours. Lied to your parents. Snatched a woman's purse. Rolled drunks just for the fun of it. 93 94 Appendix 4 (cont'd) NEVER - you have never done this RARELY - done once or twice in your life 2171-1710 SOMETIMES - done three (3) to nine (9) times in your life 30m:a mmxH—Imzom OFTEN - done more than ten (10) times in your life 21. Shoplifted merchandise vaTUed over $25. 22. Shoplifted merchandise valued under $25. 23. Received a speeding ticket. 24. Been questioned by the police. 25. Taken part in a robbery. 26. Taken part in a robbery involving physical force or a weapon. 27. Been arrested for a felony. 28. Resisted arrest. 29. Been arrested for any other nontraffic police offenses (except fighting or a felony). 30. Been convicted of any nontraffic police offense. 31. Defaulted on a debt. 32. Passed bad checks for the fun of it. 33. Ever used an alias? 34. Gone AWOL from the military. 35. Received a bad conduct or indesirable discharge from the military. 36. Performed sexual acts for money. 37. Engaged in homosexual acts. 38. Had intercourse with more than one person in a single day. 39. "Fooled around" with other women/men after you were married. 40. Hit your husband/wife during an argument. 41. Lied to your spouse. 42. Spent six months without any job or permanent home. 43. Been fired for excessive absenteeism. 44. Been fired for poor job performance (except absenteeism). 45. Changed jobs more than 3 times in one year. 46. Lied to your boss. Thank your for your cooperation. 9Zucker, R. A. Michigan State University Family Study, Michigan State University, 1984. BIBLIOGRAPHY BIBLIOGRAPHY Ablon, J. (1976), Family Structure and Behavior in Alcoholism: A Review of the Literature. In B. Kissin 8 H. Begleiter (Eds.), The Biology of Alcoholism: Vol. 4. Social Aspects of Alcoholism. Neinork: Plenum Press, 205-242. Achenbach, T. and Edelbrock, C. (1981). Behavioral Problems and Competancies Reported by Parents of Normal and Disturbed Children aged four through sixteen. Monographs of the Society for Research in Child Development, 4§(SerialNo. 188). Beck, A., Rial, W. and Rickels, K. (1974). Short Form of the Beck Depression Inventory: Cross Validation. Psychological Reports,.§4, 1184-1186. Beck, A. T., Ward, C. H., Mendelson, M., Mock. J. and Erbaugh, J. (1961). An Inventory for Measuring Depression. Archives of General Psychiatry, 4, 5-63. Blane, H. T. and Hewitt, L. E. (1977). Alcohol and Youth: An Analysis of the Literature, 1960-1975. National Institute on Alcohol Abuse and Alcoholism; Alcohol, Drug Abuse and Mental Health Administration. Rockville, MD: Department of Health, Education and Welfare. Block, J. (1961). The Q:sort Method in Personality Assessment and Psychiatric Research. Springfield, IL: CharlesTC. Thomas PUbliSher. Block, J. H. and Block, J. (1980). The Role of Ego-control and Ego-resiliancy in the Organization of Behavior. In Collins, W. A. (Ed.), Development of Cognition, Affect, and Social Relations. The Minnesota Symposia on Child Psychology. New Jersey: Erlbaum Associates, E, 39-1 02. Block, J. and Block, J. H. (1981). The California Child Q-set: Orientation and Introduction. university of california,lBerkeley (in mimeo). Bowlby, J. (1973). Attachment and Loss. London: Hogarth. Britt, D. and Campbell, E. (1977). A Longitudinal Analysis of Alcohol Use, Environmental Conduciveness and Normative Structure. Journal of Studies on Alcohol, 38, 1640-1647. Brown, G. W. and Harris, T. (1978). Social Origins of Depression: A Study of Psychiatric Disorder in Women. Londonz’Tavistock Publications, 63-l40. 95 96 Burchfield, S., Hamilton, E. and Banks, W. (1982). Affiliative Needs, Interpersonal Stress and Symptomatology. Journal of Human Stress, 8, 5-9. Buss, A., Fisher, H. and Simmons, A. (1962). Aggression and Hostility in Psychiatric Patients. Journal of Consulting Psychology, 26, 84-89 Buss, A. and Durkee, A. (1957). An Inventory for Assessing Different Kinds of Hostility. Journal of ConsultingPsychology, 21, 343-349 a Calahan, D. and Cisin, I. (1976). Drinking Behavior and Drinking Problems in the United States. In Kissin, B. and Begleiter, H. (Eds.) The Biology of Alcoholism: Social Aspects of Alcoholism, Vol. 4. lNew York: Plenum’Press,’77-115. Cahalan, D., Cisin, I. and Crossley, H. (1969). American Drinking Practices: A national study of drinkipg behavior and attitudes. New Brunswick, N.JZ: Publications Division, Rutgers Center of Alcohol Studies. Carpenter, J. and Lester, D. (1980). The etiology of alcoholism: a longitudinal study. Mimeo, New Brunswick, N.J.: Rutgers Center for Alcohol Studies. Carey, W. and McDevitt, S. (1978). Stability and Change in Individual Temperament Diagnosis from Infancy to Early Childhood. Journal of the American Academy of Child Psychiatry, 29, 331-337. Carrol, 8., Fielding, J. and Blashki, T. (1973). Depression Rating Scales: A Critical Review. Archives of General Psychiatry, 28, 361-366. Casey, R., Masuda, M. and Holmes, T. (1967). Quantitative Study of Recall of Life Events. Journal of Psychosomatic Research, 11, 239-247 a Chiriboga, D. and Dean, H. (1978). Dimensions of Stress: Perspectives from a Longtudinal Study. Journal of Psychosomatic Research, 22, 47-55. Cobb, S. (1974). A Model for Life Events and Their Consequences. In Dohrenwend, B. S. and Dohrenwend, B. P. (Eds.) Stressful Life Events: Their Nature and Effects, New York: JDhn Wiley and Sons, 151-156. Coddington, R. D. (19728). The Significance of Life Events as Etiologic Factors in the Diseases of Children I - A Survey of Professional Workers. Journal of Psvchosomatic Research, 16, 7-18. 97 Coddington, R. D. (1972b). The Significance of Life Events as Etiologic Factors in the Diseases of Children 11 - A Study of a Normal Population, Journal of Psychosomatic Research, 16, 206-213. Cooley, E. and Keesey, J. (1981). Moderator Variables in Life Stress and Illness Relationship. Journal of Human Stress,.7, 35-40, Cotton, N. S. (1979). The Familial Incidence of Alcoholism: A Review. Journal of Studies on Alcohol, 39, 89-116. Dunn, J., Kendrick, C. and MacNames, R. (1981). The Reaction of First-born Children to the Birth of a Sibling: Mother's Reports. Journal of Child Psychology and Psychiatry, 22, 1-18. El-Guebaly, M. and Offord, D. (1977). The Offspring of Alcoholics: A Critical Review. The American Journal of Psychiatry, 134, 357-365. Feighner, J., Robins, E., Guze, S., Woodruff, R., Winokur, G. and Munoz, R. (1972). Diagnostic criterion for use in psychiatric research. Archives of General Psychiatry, 26, 57-63. Forbes, R. (1979). Life Stress. Garden City, New York: Doubleday, 104-113. Fontana, A., Hughes, L., Marcus, J. and Dowds, 8. (1979). Subjective Evaluation of Life Events, Journal of Consulting_and Clinical Garmezy, N. (1981), Children Under Stress: Perspectives on Antecedants and Correlates of Vulnerability and Resistance to Psychopathology. In Rabin, A., Aranoff, J., Barkley, A., and Zucker, R. (Eds.) Further Explorations in Personality. New York: Wiley, 196-269. Gersten, J., Langner, T., Eisenberg, J. and Orzek, L. (1974). Child Behavior and Life Events: Undesirable Change or Change Per $8? In Dohrenwend, B. S. and Dohrenwend, B. B. (Eds.) Stressful Life Events: Their Nature and Effects. New York: John Wiley and Sons. Guilford, J. (1956). Fundamental Statistics in Psychology and Education. New York: McGraw-Hill Book Company Inc. Hamilton, M. (1960). A Rating Scale for Depression. Journal of Neurology, Neurosurgery and Psychiatry, 23, 56-62. Hamilton, M. (1967). Development of a Rating Scale for Primary Depressive Illness. British Journal of Social and Clinical Psychology, s, 278-296. 98 Hamilton, M. (1969). Standardized Assessment and Recording of Depressive Symptoms. Psychiatry, Neurolqgia, and Neurchirurgia, 72, 201-205. Hatzenbuehler, L., Parpal. M. and Matthews, L. (1983). Classifying Students as Depressed or Nondepressed Using the Beck Depression Inventory: An Empirical Analysis. Journal of Consulting and Clinical Psychology, 51, 360-366. Hetherington, E. M. (1972). The Effects of Father Absence on Personality-development in Adolescent Daughters. Developmental Psychology,.Z, 313-326. Herjanic, 8., Herjanic, M., Penick, E., Tomelleri, C. and Armbruster, R. (1977). Children of Alcoholics. In Seixas, P. (Ed.) Currents in Alcoholism. Vol. II. New York: Grune and Stratton. Hesselbrock, M., Hesselbrock, V., Tennan, H., Meyer, R. and Workman, K. (1983). Methodological Considerations in the Assessment of Depression in Alcoholics. Journal of Consulting and Clinical Psychology,.§l, 399-405. Holahan, C. and Moos, R. (1981). Social Support and Psychological Distress: A Longitudinal Analysis. Journal of Abnormal Psychology, 20, 365-370. Holmes, T. and Rahe, R. (1967). The Social Readjustment Rating Scale. Journal of Psychosomatic Research, 11, 213-218. Hubert, N., Wachs, T., Peters-Martin, I. and Gandour M. J. (1982). The Study of Early Temperament: Measurement and Conceptual Issues. Child Development, 53, 571-600. Hunter, J. (1980). Partial Correlations. Unpublished manuscript, East Lansing, Michigan: Department of Psychology, Michigan State University. Hurme, H. (1981). Life Changes During Childhood. University of Jyvaskyla,TJyvaskyla. Jacob, T., Favorini, A., Meisel, S. and Anderson, C. (1978). The Alcoholic's Spouse, Children, and Family Interactions. Journal of Studies on Alcohol, 3, 12231-1251. Lazarus, R. and DeLonges, A. (1983). Psychological Stress and Coping in Aging. American Psychologist, 38, 245-254. Masuda, M. and Holmes, T. (1967). The Social Readjustment Rating Scale: A Cross-Cultural Study of Japanese and Americans. Journal of Psychosomatic Research,_ll, 219-225. 99 McCubbin, H. (1979). Integrating Coping Behavior in Family Stress Theory. Journal of Marriage and the Family, 41, 237-244. McDevitt, S. and Carey, W. (1978). Measurement of Temperament in 3- to 7-year-old children. Journal of Child Psychology and Psychiatry and Allied Disciplines, 12, 245-253. Mehrabian, A. and Ross, M. (1979). Illnesses, Accidents, and Alcohol Use as Functions of the Arousing Quality and Pleasantness of Life Changes. Psychological Reports, 55, 31-43. Murphy, I. 8. and Moriarty, A. E. (1976). Vulnerability, Coping and Growth from Infancy to Adolescence. New Haven: YalelUniVersity Press. Noll, R. and Zucker, R. (1983). Developmental findings from an alcoholic vulnerability study. Paper presented at the Annual Meeting of the American Psychological Association, Anaheim, California. Nylander, I. and Rydelius, P. A. (1982). A Comparison Between Children of Alcoholic Fathers from Excellent Versus Poor Social Conditions. Acta Paediatria Scandinavia,.11, 809-813. Paykel, E. S. (1974). Life Stress and Psychiatric Disorder: Applications of the Clinical Approach. In Dohrenwend, B. S. and Dohrenwend, B. P. (Eds.) Stressful Life Events: Their Nature and Effects. New York: John’Wiley andlSons. Rabkin, J. and Streuning, E. (1976). Life Events, Stress, and Illness. Science, 121, 1013-1020. Rahe, R. (1979). Life Change Events and Mental Illness: An Overview. Journal of Human Stress, 5, 2-10. Rutter, M. (1981). Stress, Coping and Development: Some Issues and Some Questions. Journal of Child Psychology and Psychiatry, 22, 323-356. Sands, J. and Parker, J. (1979-80). A Cross-sectional Study of the Perceived Stressfullness of Several Life Events. International Journal on Aging and Human Development, 10, 335-34l. Sarason, 1., Johnson, J. and Siegel, J. (1978). Assessing the Impact of Life Changes: Development of the Life Experiences Survey. Journal of Consultingand Clinical Psychology, 16, 932-946. 100 Seigel, S. (1956). Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill 800k CDmpany, Inc., 68-74. Selzer, M., Vinokur, A. and VanRooyen, L. (1975). A Self-administered Alcoholism Screening Test (SMAST). Journal of Studies on Alcohol, 36, 117-126. Suls, J. and Mullen, B. (1981). Life Events, Perceived Control and Illness: The Role of Uncertainty. Journal of Human Stress,_l, 30-34. Vaillant, G. and Milofsky, E. (1982). The Etiology of Alcoholism: A Prospective Viewpoint. American Psychologist,_37, 494-503. Whitehead, L. (1979). Sex Differences in Children's Responses to Family Stress: A Re-evaluation. Journal of Child Psychology and Psychiatry, 20, 247-254. Wolff, S. (1973). Children Under Stress. Great Britain: Penguin Books. Zucker, R. (1966). Community adolescent drinking study-research questionnaires. Mimeo, New Brunswick, N.J.: Rutgers Center for Alcohol Studies. Zucker, R. (1976). Parental Influences upon Drinking Patterns of their Children. In Greenblatt, M. and Schuckit, M. (Eds.) Alcoholism Problems in Women and Children. New York: Grune and Stratton. Zucker, R. (1980). The Michigan State University Vulnerability Study: Overview, Design, and Method. Department of Psychology, Michigan State University. Zucker, R. and Barron, P. (1973). Parental Behaviors Associated with Problem Drinkin and Antisocial Behavior Among Adolescent Males. In Chafetz, M. Ed.) Research on Alcoholism I. Clinical Problems and S ecial Po ulations. WashingtoniDLC.: 0.5. Government Printlng Office (DHEW Publications 74-675). Zucker, R., Baxter, J., Noll, R., Theado, D. and Weil, C. (1982). An alcoholic risk study: Design and early health reIated findings. Paper presented at the annual meeting of the American Psychological Association, Washington, D.C.. Zucker, R. and N011, R. (1980). The_Antisocial Behavior Checklist. Mimeo, East Lansing, Michigan: Department of Psycthogy, Michigan State University. 101 Zucker, R. and N011, R. (1982). Precursors and Developmental Influences on Drinking and Alcoholism: Etiology from 8 Longitudinal Perspective. In Deluca, J. (Ed.) Alcohol and Health Monographs - 1: Alcohol Consumption and Related Problems. National Institute on Alcohol Abuse and Alc0holism,lwashington D.C.: U.S. Government Printing Office. Zucher, R. and N011, R. (1984). Psychometric Properties of the Antisocial Behavior Scale. Unpublished manuscript, East Lansing, Michigan: Department of Psychology, Michigan State University. Zucker, R., Noll, R., Draznin, T., Baxter, J., Weil, C., Theado, D., Greenberg, G., Carlot, C. and Reider, E. (1984). The Ecology of alcoholic families: conceptual framework for the Michigan State University Longitudinal Study. Paper presented at the National Council on Alcoholism. National Alcoholism Forum. Detroit, Michigan.