Psychological Reports, 1993, 73, 19-26. @ Psychological Reports 1993

COGNITIVE FUNCTIONING, FAMILY HISTORY OF ALCOHOLISM, AND ANTISOCIAL BEHAVIOR I N FEMALE POLYDRUG ABUSERS ' MARINA RODR~GUEZ University of Santiago de Compostelo Summary.-Family history of alcoholism and antisocial behaviour had both independent and synergic negative relationships with abstract verbal reasoning for a group of 42 female polydrug abusers whose main drug was heroin. This finding suggests that the observed lower cognitive performance may reflect the family history of alcoholism or antisocial behaviour rather than just drug abuse and is consistent with theories predicting a relative deficiency in high language skill by persons with antisocial personality disorders.

Recent neuropsychological investigations have strongly emphasized the multifactorial nature of drug-related neur~ps~chological impairment (Tarter & Edwards, 1986; Grant, 1987). Family and psychopathological variables are considered to be relevant as well as sex, age, and drug-abuse record. It has been found, for example, that male alcoholics reporting a family hstory of alcoholism are more likely to have neuropsychological deficits than those reporting no such history (Goodwin, 1983; Schaeffer, Parsons, & Yohman, 1984; Tarter, Hedegus, Goldstein, Shelly, & Alterman, 1984). However, Jones-Saumty, Parsons, and Fabian (1980) reported that, although women alcoholics with family history of dcoholism tended to perform worse on some neuropsychological tests, these tendencies were not statistically significant, and other studies also have not yielded differences between subjects with and without family histories of alcoholism at risk for alcoholism (Alterman, Bridges, & Tarter, 1986; Workman-Daniels & Hesselbrock, 1987). This has led to speculation that factors other than family history of alcoholism might account for the reported differences in cognitive function. One possibility is that the relatively poor performance of subjects with positive family histories on neuropsychological tests may be attributable to comorbid psychopathology. In particular, the high prevalence of antisocial behaviour among drug users has been identified-as a possible confounding factor in previous studies (Langevin, Ben-Aron, Wortzman, & Dickey, 1987). Some studies have indeed indicated poorer neuropsychological performance in psychopaths, violent criminals, and other subjects exhibiting antisocial behaviour than by control groups, regardless of whether the antisocial subjects exhibited drug abuse behaviour (Gorenstein, 1982; Yeudall, Fromm'Address correspondence to Marina Rodriguez, Ph.D., Departament Psicologia Clinica y Psicobiologia, University of Santiago de Compostela, 15705 Santiago de Cornpostela, Spain.

Auch, & Davies, 1982) or were substance abusers (Malloy, Noel, Longabaugh, & Beattie, 1990) or specifically alcoholics (Hesselbrock, Weidenman, & Reed, 1985; Malloy, Noel, Rogers, Longabaugh, & Beattie, 1989). Malloy, et al. (1989, 1990) attribute the antisocial personality of their subjects to their having indulged in abuse earlier and more heavily than nonantisocial abusers (rather than to any pre-abuse deficits). To place the above findings and hypotheses in focus, it must also be borne in mind that several studies have shown the neuropsychological function of alcoholics, substance abusers, and criminal psychopaths to be no poorer than that of control groups (Hare, 1984; Fields & Fullerton, 1974; Bruhn & Maage, 1975; Hasin & Grant, 1987; Hoffman, Hall, & Bartsch, 1987; Hart, Forth, & Hare, 1990). With a view to throwing light on the possible roles of antisocial behaviour and family history of alcoholism to cognitive dys€unction of substance abusers, in this work we studied their separate and synerglc influence with a sample of female polydrug abusers whose main drug was heroin.

METHOD We studied right-handed Spanish white female polydrug abusers whose main drug was heroin and who had voluntarily requested dehabituation treatment in a Drug Addiction Outpatient Centre at Orense, Spain, where they had been diagnosed as opiate-dependent according to DSM-I11 criteria and were assigned to a drug-free treatment programme. Within this group, prospective subjects were excluded from the study if they had a history of (a) dependence on drugs other than heroin, (b) cranioencephalic traumatism with loss of consciousness, (c) diagnosed neurological alterations, (d) alterations affecting the CNS, or (e) psychatric disorders. The final sample consisted of 42 subjects (mean age 22.9 y t ) who had begun drug abuse at a mean age of 15 yr. (SD = 2.4) and had been dependent on heroin (daily self-administered intravenous doses of about .25 mg) for 1 to 9 yr. (M= 2.8, SD = 2.1). Before starting on heroin, alI had taken cannabis, amphetamines, cocaine, or barbiturics, either simultaneously or consecutively, without becoming dependent on any of these other drugs. In a 1-hour session held three weeks after the start of treatment (to avoid distortion of replies due to the effects of drugs), prospective subjects were individually given a structured interview to elicit demographic, medical, social, drug abuse, and data on family history of alcoholism, and their cognitive performance was evaluated. Subjects were considered positive on histories if their biological father was an alcoholic or had had social, family, work, or medical problems attributable to alcohol consumption and negative if no member of their immediate family was an alcoholic or had had alcohol-related problems. Following Cadoret, O'Gorman, Thoughton, and Theywood (1985) and Hart, et al. (19901, subjects were classed as positive, i.e., as

FEMALE POLYDRUG USERS

21

having exhbited antisocial behaviour had they been convicted of drug-related offences (mostly theft and sale of drugs), and as negative if they neither had convictions for any kind of offence nor exhibited any type of antisocial behaviour (all 42 subjects fell into one or other of these classes). Antisocial behaviour defined in terms of convictions was studied, rather than antisocial personality, because not all subjects' records were detailed enough for application of DSM-I11 criteria for the latter. Seven subjects (17%) were positive for history and antisocial behaviour, 11 (26%) positive on alcoholism but negative on antisocial behaviour, 11 (26%) negative on history but positive on behaviour, and 13 (31%) negative on both. Subjects' neuropsychological performance was evaluated individually with the Spanish version of the Wechsler Adult Intelligence Scale (WAIS) (Wechsler, 1984), which includes tests of expressive vocabulary (Vocabulary), general information (Information), formation of verbal concepts (Similarities), verbal abstraction (Comprehension), attention-concentration (Digit Span, Arithmetic), spatial and constructional skills (Block Design, Picture Completion, Object Assembly), visual sequences/nonverbal reasoning (Picture Arrangement), and concentration and perceptual and motor speed (Digit Symbol). The WAIS has frequently been used for neurotoxicological examination because some of its tasks (notably the Digit Symbol, Block Design, Similarities, and Digit Span tests) are particularly sensitive to the subclinical effects of exposure to toxicants (Hartman, 1988). The four subject groups described above under procedure make up a 2 x 2 experimental design for which analysis of variance was carried out with the WAIS subtest scores as dependent variables. All statistical calculations were performed using the Statistical Package for the Social Sciences (SPSS-PC + ).

RESULTS Table 1 describes the whole sample and the groups with positive and negative family history of alcoholism and of antisocial behaviour by age, duration of formal education, age at start of drug abuse, and time since onset of heroin addiction. This last was statistically significant and 2 years longer for subjects positive on antisocial behaviour than for subjects negative on antisocial behaviour (F,,,,= 9.74, p < .05). Analysis of variance gave no other significant "effects" of either classificatory criteria separately or synergistically on the other variables shown in this table. Table 2 lists the WAIS subscores of the whole sample and the four groups. All scores were within the ranges expected for nonaddicts. Only three mean scores were found by analysis of variance to be significantly affected by the independent variables: those measuring formation of verbal concepts (Similarities), verbal abstraction (Comprehension), and expressive vocabulary (Vocabulary). Family history of alcoholism alone affected only the

TABLE 1

DESCRIF~ION OF SAMPLEA N D SUBGROUPS: MEANSA N D STANDARD DEVIATIONS IN YEARS Measures

Family History of Alcoholism Yes No

Age M

SD Education M

SD Age first drug M

SD Addiction time M

Antisocial Behaviour

Total Group N, 42

n, 18

n, 24

Yes n , 18

n, 24

No

22.1 2.3

23.5 3.0

22.9 2.8

22.8 2.9

22.9 2.8

8.9 1.7

9.4 1.7

9.1 1.7

9.2 1.7

9.2 1.6

14.3 1.8

15.6 2.6

14.7 2.7

15.3 2.1

15.0 2.4

2.2

3.2

3.9*

1.9*

2.8

Comprehension score, positive subjects scoring significantly lower than negative subjects (F,,,, = 6.08, p < .05). Antisocial behaviour alone showed in positive subjects scoring significantly lower than negative subjects on both Corn~rehension(F,,,, = 5.50, p < .05) and Similarities (F,,,,= 4.38, p < .05). The interaction was statistically significant to the detriment of subjects positive on Comprehension (F,,,, = 9.10, p < .05), Similarities (F,,,,= 4.73, p < .05), and Vocabulary (F,,,, = 5.08, p < .05). TABLE 2

MEANSAND STANDARD DEVUTIONS ON COGNITIVE MEASURES WAIS Subtests

Information M

SD Comprehension M

SD Arithmetic M

SD

Family History of Alcoholism Yes No n. 18 n. 24

Antisocial Behaviour Yes No n . 18

n. 24

10.9 2.3

11.0 1.7

10.8 2.4

11.1 1.4

10.9 1.9

9.7 3.5

11.4 2.3

9.7 3.6

11.4 2.2

10.7 2.9

9.3 1.5

9.7 1.6

9.0 1.7

9.9 1.4

9.5

(continued on next ~ a e e ) OFamily history of alcoholism effect, p < .05. b~ntisocialbehaviour effect, p < .05. 'Interaction effect, p < .05.

P

Total Group N, 42

1.6

"9

b,c

23

FEMALE POLYDRUG USERS TABLE 2 (CONT'D) MEANSA N D STANDARD DEVIATIONS ON COGNITIVE MEASLJRES WAIS Subtests

Family History of Alcoholism Yes No n, 18 n, 24 -

~~

-

Antisocial Behaviour Yes No n, 18 n, 24

Total Grou~ N, 42

P

Similarities

M SD Digit Span

M SD Vocabulary

M SD Digit Symbol M

SD Picture Completion M

SD Block Design M

SD Picmre Arrangement

M

SD Object Assembly

M SD 'Family history of alcoholism effect, p < .05. b~ntisocialbehaviour effect, p < .05. CInteraction effect, p < .05.

As the time since onset of heroin addiction was significantly greater for subjects scoring positively than negatively on antisocial behaviour, we investigated the possible relationship between this antisocial behaviour and cognitive function. None of the Pearson product-moment correlations between time since onset of addiction and the WAIS subscores was significant ( p > ,051.

DISCUSSION The subgroups were highly similar in age, education, and age at which drug abuse began (and sex, of course). The only descriptive variable for which significant dependence on antisocial behaviour was detected, time since onset of addiction, was unrelated to any of the measures of cognitive function. The finding that the group with positive famdy history were worse on

verbal abstraction than the negative group, together with similar results for alcoholic women (Jones-Saumty, et al., 1980), suggests that family history of alcoholism is a risk factor for neuropsychological deterioration in substance abusers. Likewise, the fact that female polydrug abusers with antisocial behaviour were worse on verbal abstraction and concept formation than the group without antisocial behaviour is in keeping with reports of a relative deficiency in higher-level language skills of delinquents or subjects with a diagnosis of antisocial personality disorder (Yeudall, et al., 1982). I n our work, these effects of these two variables were synergic: subjects positive on both exhibited poorer abstract verbal reasoning than other subjects even after the direct effects of the two variables had been taken into account. Although none of our subjects exhibited cognitive deficiency as such, the effects of f a d y history of alcoholism or antisocial behaviour were nevertheless associated with a lower capacity for abstraction. This finding suggests that the verbal-abstraction deficits observed in heroin addicts by other authors (Ahmad, Ahmad, & Brinda, 1989) may have been related more to antisocial behaviour or family history of alcoholism than to drug abuse as such or were at least. not solely caused by drug abuse. Discrepancies among various studies as to whether drug abuse is associated with neuropsychological deficits may have arisen partly because most research on neuropsychological function of polydrug abusers has not taken these variables into account. Our results then emphasize the multifactorial nature of neuropsychological vulnerability to drugs. The observed synergism between history of alcoholism and antisocial behaviour in their effects on cognitive functioning in polydrug abusers is in keeping with the findings of McCarthy and Anglin (1990). Subjects with positive antisocial behaviour are likely to have had alcoholic, antisocial parents and are likely to have suffered from the effects of parental neglect as children (e.g., malnutrition and poor health care), leaving them vulnerable to causes of neuropsychological deficiency. Although the differences between the performances of subjects positive on both variables and those of the other groups were only statistically significant for Comprehension, Similarities, and Vocabulary, differences of like sign (but not significant) appeared for other WAIS subscores. The above findings suggest several avenues for research. Firstly, replication of this study using a larger sample and evaluating other cognitive areas would further clarify the relationship of family history of alcoholism and antisocial behaviour to cognitive functioning and their usefulness as indicators of risk of neuropsychological impairment in polydrug abusers. Secondly, we intend to investigate the relationship between apparent neuropsychological impairment and physiological measures of cerebral performance (specifically, event-related potentials). Finally, it also remains to be examined whether the

FEMALE POLYDRUG USERS

25

effects of the two variables observed in this study represent a premorbid decline in capacity of subjects who subsequently became polydrug abusers to the selective vulnerability of the affected cognitive processes to the effects of drug abuse or to a combination of both factors. REFERENCES ALTERMAN, A. I., BRIDGES,R., & TARTER,R. (1986) The influence of both drinking and familial risk status on cognitive functioning of social drinkers. Alcoholism: Clinical and Experimentul Research, 10, 448-451. AHMAD,S., AHMAD,H., & BRINDA,G. (1989) Cognitive impairment in chronic heroin addicts. Journal o/ Personalily and Clinical Studies, 5, 237-240. BRIJHN,P., & MAAGE,N. (1975) Intellectual and neuropsychological functioning in young men with heavy and long-term patterns of drug abuse. American Journal of Psychiatry, 132, 397-401.

CADORET,R. J., O'GORMAN,T. W., THOUGHTON, E., & THEYWOOD, E. (1985) Alcoholism and antisocial personality. Archives of General Psychiatry, 42, 161-167. FIELDS,F. R. J., & FULLERTON, J. R. (1974) Influence of heroin addiction on neuropsychological functioning. Journal of Consulting and Clinical Psychology, 43, 114. GOODWIN, D. W. (1983) Familial alcoholism: a separate entity? Alcohol Actions Misuse, 4, 129136.

GORENSTEIN, E. E. (1982) Frontal lobe function in psychopaths. Journal of Abnormal Psychology, 91, 368-379. correlates. Journal o/ Consulting GRANT,I. (1987) Alcohol and the brain: neuropsy~holo~ical and Clinical Psychology, 53, 310-324. HARE,R. D. (1984) Performance of psychopaths on cognitive tasks related to frontal lobe function. Journal of Abnormal Psychology, 93, 133-140. HART, S. D., FORTH,A. E., & HARE,R. D. (1990) Performance of criminal psychopaths on selected neuropsychological tests. Journal of Abnormal Psychology, 4, 374-379. HARTMAN, D. E. (1988) Neuropsychological toxicology. New York: Pergamon. functioning in alcoholics: psychiatric HASIN,D. S., & GRANT,B. F. (1987) Ne~rops~chological comorbidity, drinking history and demographic characteristics. Comprehensive Psychiatry, 28, 520-529. HESSELBROCK, M. N., WEUIENMAN, M. A,, & REED, H . B. C. (1985) Effect of age, sex, drinking hjstory and antisocial personality on neuropsychology of alcoholics. Journal of Studies on Alcohol, 46, 313-320. HOFFMAN, J. J., HALL,R. W., & BARTSCH,T. W. (1987) On the relative importance of "psychopathic" personality and alcoholism on neuropsychological measures of frontal lobe dysfunction. Journal of Abnormal Psychology, 96, 158-160. JoN~s-SArrrvlTY,D.J., PARSONS, 0 . A,, & FABIAN,M. S. (1980) Familial alcoholism, drinking behavior, and neurotoxicological performance in alcoholic women. Alcohol Technical Report, 9, 29-34. LANGEVIN, R., BEN-ARON,M., WORTZMAN, G., & DICKEY,R. (1987) Brain damage, diagnosis and substance abuse among violent offenders. (Special Issue: Homicidal behavior) Behavioral Sciences and the Law, 5, 77-94. MALLOY,l?, NOEL,N., LONGABAUGH, R., & BEAITIE,M. (1990) Determinants of neuropsychological impairment in antisocial substance abusers. Addictive Behaviors, 15, 431-438. MALLOY,l?, NOEL, N., ROGERS,S., LONGABAUGH, R., & BEAT~IE, M. (1989) Risk Factors for neuropsychological impairment in alcoholics: antisocial personality, age, years of drinking and gender. Jozrrnal of Studies on Alcohol, 50, 422-426. MCCARTHY, W. J., & ANGLIN, M. D. (1990) Narcotics addicts: effect of famly and parental risk Factors on timing of emancipation, drug use onset, pre-addiction incarceration and educational achievement. Journal of Drug Issues, 20, 93-12). SCHAEFFER,F. W., PARSONS,0. A,, & YOHMAN, J. R. (1984) Ne~rops~chological differences between familial and nonfarnilial alcoholics and nonakoholics. Alcoholism: Clinical and Experimental Research, 8, 347-35 1.

TARTER, R. E.,

& EDWARDS, K. L. (1986) Multifactorial etiology of neuropsychological impairment in alcoholics. Alcoholism: Clinical and Experimental Research, 10, 128-135. TARTER, R. E., HEDEGUS, A. M., GOLDSTEIN, G., SHELLY, C., & ALTERMAN, A. I. (1984) Adolescent sons of alcoholics: neuropsy~holo~ical and personality characteristics. Alcoholism: Clinical and Experirnentul Research, 8, 216-222. WECHSLER, D. (1984) Escala de Inteligencia de Wechsler para Adultos. Madrid: TEA. WORKMAN-DANIELS, K. L., & HESSELBROCK, V. (1987) Childhood problem behavior and neuropsychological functioning of persons at risk for alcoholism. Jotrrnal of Studies on Alcohol,

48, 187-193.

YEUDALL, L. T., FROMM-AUCH, D., & DAVIES,P. (1982) Neuropsychological impairment in persistent delinquency. Journal of Nervous and Mental Direase, 170, 257-265.

Accepted May 10, 1993