Differences and similarities in development of drinking behavior between alcoholic offspring of alcoholics and alcoholic offspring of non-alcoholics

Addictive Behaviors, Vol. 16, pp. 341-347, Printed in the USA. All rights reserved.

1991 Copyright

0306-4603/91 $3.00 + .oo e 1991 Pergamon Press plc

BRIEF REPORT DIFFERENCES AND SIMILARITIES IN DEVELOPMENT OF DRINKING BEHAVIOR BETWEEN ALCOHOLIC OFFSPRING OF ALCOHOLICS AND ALCOHOLIC OFFSPRING OF NON-ALCOHOLICS GEARY S. ALFORD University

of Mississippi

Medical Center

ERNEST N. JOURILES University

of Houston

SARA C. JACKSON University

of Southern Mississippi,

Gulf Park

Abstract - Self-reported initial, early, and long-term drinking behaviors, experiences, and consequences were obtained from mate alcoholics completing inpatient treatment. Subjects were recruited and selected on the basis that they met VSM-III criteria for diagnosis of alcohol dependency and that their biological fathers were alcoholic (FHP; Family History Positive) or that they had no biological family history of alcoholism (MN; Family History Negative). Results indicated that FHP subjects rated their initial taste of beer higher than FHN subjects, that FHP subjects began tasting and subsequently regularly drinking alcohol at an earlier age than FHN subjects and that there was significantly shorter elapsed time between initiating regular drinking and developing alcoholic-symptomatic problems in living among PHP alcoholics than FHN alcoholics. Although there were a few other significant differences, the drinking-behavioral histories of the two groups were remarkably similar and parallel. Taken together, results suggest that familial risk factors primarily influence the rate at which alcoholic drinking and alcoholism develop, rather than the form or pattern of alcoholic drinking.

INTRODUCTION

It has long been recognized and more recently documented that children of alcoholics are, themselves, more likely to become alcoholic than are offspring of non-alcoholics (Cotton, 1979). Social scientist have generally attributed this to parental modeling and social-environmental influences (e.g., Bandura, 1969), and have been supported in their position by some controlled laboratory and natural setting research. Candill & Marlatt (1975), for example, found that peer influence and exposure to a heavy drinking model significantly increased amount of alcohol consumption among research subjects. In their extensive review of the research literature on social learning and development of alcoholics, O’Leary, O’Leary, & Donovan (1979) concluded that parents of alcoholics had typically failed to expose their offspring to moderate, social drinking models. Although concepts of a specific and unique “alcoholic personality’ ’ have been abandoned, personality traits or factors continue to play a prominent role in etiological theories of alcoholism and drug abuse, more recently gaining some empirical support (e.g., Shedler & Block, 1990; Tarter & Edwards, 1986). In addition to psychosocial influences, a growing body of evidence indicates that genetic factors predispose at least some individuals to alcohol abuse and alcoholism. Goodwin and his colleagues (Goodwin, Schulsinger, Hermanson, Guze, & Winokur, 1973; Requests for reprints should be sent to Geary Alford, Department of Psychiatry and Human University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216. 341

Behavior,

342

GEARY S. ALFORD

et al

Goodwin et al., 1974) found that male adoptees who had an alcoholic biological father were significantly more likely to have become alcoholic than were a matched group of adoptees without such biological, parental alcoholic history. Some investigators have suggested that alcoholism has at least two subtypes related, in part, to a presumed genetic predisposition in one type but not the other (Cloninger, 1987). Although precisely how subtypes of alcoholics may differ biopsychologically remains speculative and controversial (e.g., Buydens-Branchey and Branchey, 1989; Buydens-Branchey, Branchey, & Noumair, 1989; Vaillant, 1989), there is general agreement now that some individuals who become alcoholics were genetically predisposed to becoming alcoholic while other alcoholics were not. However, it is also generally accepted that, unlike Huntington’s disease, alcoholism is not a strictly genetic disorder; and other familial factors as well as other individual behavioral characteristics or patterns can contribute to the development of alcoholism. Questions naturally arise as how genetic and other familial risk factors are actually expressed in the development of alcoholism: In what ways are actual drinking practices and drinking-related behaviors of alcoholic offspring of alcoholics different from those of alcoholics without such family histories ? Speculative answers to this question have included possible differences in taste preference for alcohol, behavioral reactions to alcohol, and behavioral style and drinking practices favoring alcoholism (Goodwin, 1979; Tarter & Edwards, 1986). The present study was designed to begin identifying both differences and similarities in early drinking behavior and in the development of abusive drinking patterns between alcoholics who have a positive biological family history of alcoholism and alcoholics who have no such biological family history. METHOD

Subjects Subjects were selected from 100 males who were undergoing inpatient treatment for alcoholism at a large, private hospital and who completed a demographic form that assessed their chemical abuse history, that of their biological parents, and, where applicable, their step-parents’ drinking habits. Subjects were selected for inclusion in the study on the following bases: 1. They met the DSM-ZZZ (American Psychiatric Association, 1980) criteria for alcohol dependency, diagnosed by detailed history of drinking pattern and biopsychosocial consequences, physical examination, and clinical laboratory studies. 2. Their parents’ drinking patterns met criteria for inclusion in either Family History Positive (FHP) group or the Family History Negative (FHN) group. Genetically relevant and adoption studies have found that a biological predisposition toward developing alcoholism tends to be transmitted by alcoholic parents to same sex offspring (Goodwin, 1979). Wishing to maximize the chances of genetic contribution in the FHP group, while wishing to minimize the chances of significant in utero alcohol influences, only males whose biological fathers were alcoholic but whose mothers were not alcoholic were used as FHP subjects. Then, in order to reduce possibly confounding sex differences, only male patients were used as FHN subjects. Criteria for inclusion in the FHP group required that a subject’s biological father was or an “alcoholic who became abstinent,” and whose either a “problem drinker/alcoholic” mother was not and had never been a “heavy or alcoholic drinker.” Criteria for the FHN group required that biological parents and grandparents had no problem drinking or alco-

Development

of

drinking behavior in alcoholic offspring

343

holic history and were either “non-drinkers (never drank)” or “occasional or light” social drinkers. Excluded from either category and the study were individual patients who rated a parent or grandparent as a “heavy” but not alcoholic drinker, as we considered it likely that at least some of these “heavy drinkers” were, in fact, alcoholics while others probably were not. Placing such patients in either FHP or FHN, then, would risk having placed them in an inappropriate category. On the basis of these criteria, 19 patients met criteria for the FHP group and 23 patients met criteria for the FHN group. This subject selection procedure, of course, did not necessarily produce two groups that were perfectly orthogonal with respect to potential genetic, parental modeling, or other familial risk factors. However, it is a reasonable presumption that selection criteria would result in differential loading of these factors into the FHP versus the FHN group. Measure

A 115item questionnaire was administered to each subject shortly before they completed inpatient treatment. Items covered the following eight topic areas: 1. Age of initial alcohol consumption and initial experiences of tasting alcohol. 2. Delinquent or maladaptive behavior prior to first year of regular drinking (alcohol consumption at least once a month). 3. Frequency of drinking during the first year of regular alcohol consumption. 4. Feelings following alcohol consumption during the first year of regular alcohol consumption. 5. Delinquent or maladaptive acts associated with drinking during the first year of regular alcohol consumption. 6. Negative life events associated with drinking during the first year of regular alcohol consumption. 7. Age subjects initially experienced problems with alcohol. 8. Drinking History and development of alcoholic drinking symptoms. Questions required either a yes or no response (e.g., During the firsr year of drinking, were family arguments/conflicts related in any way to your drinking?); a multiple choice response (e.g., When you first started drinking, the taste of beer was: [a] very bad, [b] a little bad, [c] a little good, or [d] very good); or a fill in response (e.g., What is the youngest age you can personally, clearly remember ever having a drink or even a sip of any alcoholic beverage? ). RESULTS

Separate multivariate analyses of variance (MANOVAs) were conducted on questionnaire items from the following content areas: age subjects initially consumed alcohol, taste experiences, frequency of drinking during the first year of regular drinking, feelings following alcohol consumption during the first year of regular drinking, and age subjects initially experienced problems with alcohol. Significant multivariate findings were followed with univariate comparisons (ANOVAs). To ensure that the problem of escalating Type I error rate did not occur in the multiple comparisons, the family-wise error rate was set at p = .05 for each set of comparisons and was adjusted with the Bonferroni procedure (Games, 1971). A set of chi square analyses were computed on items from the other three content areas (delinquent behavior prior to regular alcohol consumption, delinquent acts associated with drinking, and negative life events associated with drinking). For each set of analyses, the family-wise error rate was set at p = .05 and was adjusted

344

GEARY S. ALFORD

Table 1. Means and standard deviations

et al.

of self-reports

of FHP and MN

alcoholics FHN (n = 23)

(n “99, M

SD

M

SD

31.1 13.9

10.9 2.5

36.8 12.5

11.7 1.8

9.2 11 .o 14.7

3.0 4.3 2.2

12.7 14.5 16.5

3.6 3.0 2.2

Taste E.~perienc~es (Preferences)’ Beer Wine Whiskey

3.3 3.3 2.3

1.7 2.2 2.5

2.0 3.2 1.2

0.9 2.0 2.3

Frequency of Drinki@ During First Year Consumption Consumption to intoxication

2.6 2.2

1.0 1.1

2.9 2.5

0.9 1.2

Feelings After Consuming Alcohol’ During First Year Activated or energized Aggressive Emotional Socially outgoing Engage in Activities that normally would not been done

3.3 2.6 2.6 3.3 2.9

0.9 1.0 1.0 0.9 0.7

2.8 2.5 1.8 3.1 2.5

1.0 1 .o I.0 1.1 0.9

20.6 28.0

6.2 8.5

28.8 33.5

10.4 9.6

Demographics Age Education (years) Age ofInitial Alcohol Consumption First sip Sip “once in a while” Drink “fairly regularly”

(years)

Age Experienced Alcohol-related Problems (years) Lost control of drinking First help or treatment for drinking Percentage

of “JW” Responses FHN

FHP n

Delinquent Acts Prior to Regular Alcohol Consumption Arrested Served time in a youthdetention center or jail Suspended from school Failed a grade

6 2 6

%

32 5 11 32

Delinquent Acts Associated with Alcohol Consumption Physical fights Destructive behavior Legal involvement Missed/cut school or work

42 37 21 32

Negative Life Events Associated with Alcohol Consumption Family arguments Drop in grades or work performance Physical injury Loss of friend

47 32 32 32

n

6 I

9

%

26 4 4 39 26 26 22 35

4 6 4 6

17 26 17 26

“I = very bad, 2 = a little bad, 3 = a little good, 4 = very good. bl = about once a day, 2 = about twice a week, 3 = about once a week, 4 = about twice a month, 5 = less than once a month. ‘1 = never or almost never, 2 = occasionally, 3 = frequently, 4 = always or almost always.

with the Bonferroni procedure. Means and standard deviations for subjects’ demographic characteristics and responses to items in the eight content areas are presented in Table 1.

Development

of drinking behavior in alcoholic offspring

345

Age of initial alcohol consumption.

The MANOVA on items assessing age of onset of behavior yielded a significance between group difference F(3,38) = 3.87, p = .05. Using the Bonferroni procedure, the critical significance level for the individual F tests was computed as .05/3 = .017. Univariate analyses indicated that subjects in the PI-IP group reported younger ages for: (a) experiencing their first sip of alcohol F(1,40) = 8.73, p < .017; (b) sipping an alcoholic beverage once in a while, F(1,40) = 9.71, p < .017; and (c) initiating pattern of regular, intermittent alcohol consumption, F( 1,40) = 7.32, p < .017. drinking

Taste experiences. The results of the MANOVA for items assessing taste preference for alcoholic beverages yielded a significance between group effect, F(3,38) = 3.46, p < .05. The critical significance level for the individual F tests was computed as .05/3 = .017. Univariate analyses indicated that subjects in the FHP group preferred the taste of beer, F(1,40) = 9.78, p < .017, relative to subjects in the PI-IN group. There were no significant between group differences on subjects’ taste preferences for whiskey and wine. Delinquent behavior prior to regular alcohol consumption. The critical significance level for the chi square analyses examining the association between family history and deviant behavior prior to regular drinking was computed at .05/4 = .0125. FHP and FHN groups failed to differ significantly on any of the items in this content area. Frequency of drinking. The results of the MANOVA for the two measures of drinking frequency (frequency of alcohol consumption, frequency of drinking to intoxication during the first year) failed to yield a significant between group effect, F(2,38) = 0.70, p < .l. Thus, univariate analyses were not pursued with the individual items. Feelings following alcohol consumption. The MANOVA on items assessing subject’s reports of alcohol’s effects on feelings yielded a significant between group difference, F(5,35) = 2.45, p < .05. Using the Bonferroni procedure, the critical significance level for the individual F tests was computed as .05/S = .Ol. Univariate analyses indicate that when drinking relasubjects in the FHP group reported that they were more “emotional” tive to subjects in the PI-IN group F(1,39) = 6.17, p < .Ol. Other univariate analyses failed to yield significant differences. Delinquent acts and negative life events. The critical significance level was computed at .05/5 = .Ol for chi square analyses examining the association between family history and the relation between delinquent acts and drinking behavior as well as analyses examining the association between family history and the relation to negative life events and drinking behavior. Chi square analyses failed to yield significant associations for items in either content area. Age subject initially experienced alcohol-related problems. The MANOVA on items assessing the ages when problems with drinking occurred yielded a significant between group difference, F(2,35) = 3.27, p < .05. The critical significance level for univariate analyses was computed as .05/2 = .025. One additional comparison was made to determine whether FHP alcoholics drank for shorter periods of time relative to FHN alcoholics, prior to losing control of their drinking. In order to test this hypothesis, the age subjects reported that they began to drink regularly was subtracted from the age subjects reported they lost control of their drinking. FHN alcoholics reported drinking for a longer period of time (M = 13.4) prior to losing control of their drinking relative to FHP alcoholics (M = 8.0), t(38) = 1.86, p = .05.

346

GEARY S. ALFORD

et al.

DISCUSSION

Perhaps the most initially impressive findings from the present study is how very similar are the reported histories of alcoholics who have a positive family history for alcoholism (FHP) and alcoholics who do not have a family history of alcoholism (FHN). In almost every area addressed in the questionnaire few or no significant between-group differences were obtained. This was as true for behaviors and events prior to onset of regular consumption of alcohol as well as to most later alcohol related effects and consequences. Nevertheless, the present data also point to some potentially very important differences between the two groups. As can be seen in Table 1, FHP subjects rated their initial tastes of beer significantly higher than did FHN subjects. While the FHP higher numerical ratings for whiskey and wine did not achieve statistical significance, the positive finding for beer does suggest the possibility that some inherent olfactory-gustatory differences may exist between these two groups. Although environmental factors, such as differential parental comments about beverage tastes can be entertained, olfactory-gustatory ratings would appear more likely associated with inherent factors. It is likewise unclear to what extent genetic versus social-environmental differences might account for the findings that in contrast to FHN alcoholics, alcoholic offspring of alcoholics (FHP) report significantly younger ages for (a) first sip of an alcoholic beverage, (b) age at which they began taking a sip “once in a while,” and (c) age at which they started drinking alcoholic beverages fairly regularly (that is, “on average at least once a month”). More consistent with an environmental influence explanation, however. was the finding that while no significant difference was found between percentages of FHP and FHN subjects who occasionally asked their parents for a taste of alcoholic beverages, parents of FHP subjects were reportedly more likely to have allowed them to “sip alcohol on special occasions. ” It is interesting that both groups rated the effects of alcohol on their behavior quite similarly, except that significantly more FHP subjects reported becoming “more emotional” when drinking than did FHN subjects. However, they did not differ in their selfreport of alcohol’s effects on their becoming more stimulated, aggressive, socially outgoing, or in having done things when drinking they would not normally have done. In comparing drinking patterns during the first year of regularly recurrent, periodic drinking, no significant systematic differences between FHN and FHP groups were found in (a) frequency of alcohol ingestion, (b) frequency of intoxication, (c) social settings in which drinking occurred (e.g., alone vs. in the company of friends), (d) in reported desired sensory-emotional effects of drinking, or (e) in the type or frequency of adverse behavioral effects from drinking such as missed school or work, family arguments, physical fights, arguments, loss of friends, physical injury, motor vehicle accidents, or legal problems. In addition, the kinds of problems that eventually developed as a result of alcoholic drinking and that preceded and often precipitated their admission to treatment were quite similar in both groups. Taken together results reveal that male alcoholic offspring of male alcoholics rate their initial tastes of alcohol higher than did alcoholics without a biological family history of alcoholism, male alcoholics with a family history of alcoholism begin to sip and subsequently initiate regular, periodic consumption of alcohol at an earlier age than FHN alcoholics, and indicate a somewhat higher degree of emotionality resulting from drinking during the first year of regular, periodic drinking. However, in most respects, the pattern and consequences of drinking during the first year of regular, periodic drinking did not differ between the two groups. Likewise, both groups evidenced similar drinking patterns and adverse consequences by the time they were first admitted for treatment.

Development

of drinking behavior

in alcoholic

offspring

341

However, the FHP group reported more rapid development of abusive drinking pattern and alcohol-related problems which, as indicated above, tended to occur ufrer the first year (and often after several years) of regular, periodic drinking. The findings indicate that among male alcoholic offspring of male alcoholics, familial and possibly genetic factors were associated with younger age of first sipping, higher alcohol taste ratings of early sipping, earlier establishment of relatively frequent drinking, and shorter elapsed time between onset of regular, periodic drinking and the subsequent onset of abusive drinking, loss of control, and related adversive consequences from drinking. In contrast to the notion, especially popular with many disease model theorists, that genetically or familially predisposed alcoholics engage from the very first in drinking styles and practices that are distinguished in gross and obvious ways from drinkers not so predisposed, the present results revealed no such blatant, robust differences. Instead, the present results suggest that whatever early drinking behavioral differences may exist as manifestations of familial social learning or genetic expression (Blum, et al., 1990), they are relatively subtle and evidence their influences primarily on the age and rate at which abusive and subsequently alcoholic drinking patterns and symptoms develop. Finally, when viewed from still another perspective, the present results suggest that whether influenced by familial, biopsychological factors, or by other, environmental, or personal, idiosyncratic contributors, or by still other unknown variables, there appears to be a final common behavioral pathway for alcoholism: drinking too much, too often, for too long.

REFERENCES American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd ed. Washington, D.C.: American Psychiatric Association. Bandura, A. (1969). Principles of behavior modifkation. New York: Holt, Rhinehart, & Winston. Blum, K., Noble, E.P., Sheridan, P.J., Montgomery, A., Ritchie, T., Jagadeeswaran, P., Nogami, H., Briggs, A.H., & Cohn, J.B. (1990). Allelic association of human dopamine D, receptor gene in alcoholism. Journal of the American Medical Association, 263, 2055-2059. Buydens-Branchey, L., & Branchey, M.H. (1989). In reply. Archives of General Psychiatry, 46, 1151-l 152. Buydens-Branchey, L., Branchey, M.H., & Noumair, D. (1989). Age of alcoholism onset, I: Relationship to psychopathology. Archives of General Psychiatry, 46. 225-230. Candill, B.D., & Marlatt, G.A. (1975). Modeling influences in social drinking: An experimental analogue. Journal of Consulting and Clinical Psychology, 43, 405415, Cloninger, C.R. (1987). Neurogenic adaptive mechanisms in alcoholism. Science, 236, 410-416. Cotton. N. (1979). The familial incidence of alcoholism: A review. Journal of Studies on Alcohol, 40, 89-l 16. Games, PA. (1971). Multiple comparison of means. American Educational desearch Journal, 8, 531-565. Goodwin, D.W. (1979). Alcoholism and heredity: A review and hypothesis. Archives of General Psychiatry, 36, 57-61. Goodwin, D.W., Schulsinger, F., Hermansen, L., Guze, S.B., & Winokur, G. (1973). Alcohol problems in adoptees raised apart from alcoholic biological parents. Archives of General Psychiatry, 28, 238-243. Goodwin, D.W., Schulsinger, F., Moller, N., Hermansen, L., Winokur, G., & Guze, S.B. (1974). Drinking patterns in adopted and nonadopted sons of alcoholics. Archives of General Psychiatry, 31, 164-169. O’Leary, D.E., O’Leary, M.R., & Donovan, D.M. (1979). Social skill acquisition and psychosocial development of alcoholics: A review. Addictive Behaviors, 1, 11 l-120. Shedler, J., & Block, J. (1990). Adolescent drug use and psychological health: A longitudinal inquiry. American Psychologist, 45, 612-630. Tarter, R., & Edwards, K. (1986). Antecedents to alcoholism: Implications for prevention and treatment. tlehavior Therapy, 17, 346-361. Vaillant, G.E. (1989) To the editor. Archives of General Psychiatry, 46, 1151.