PEDIATRICS Vol. 106 No. 4 October 2000, pp. 792-797

Patterns of Alcohol and Drug Use in Adolescents Can Be Predicted by Parental Substance Use Disorders

Joseph Biederman, MD^{*, §}, Stephen V. Faraone, PhD^{*, , §}, Michael C. Monuteaux, BA^*, , and Jennifer A. Feighner, BA^*

From the ^* Pediatric Psychopharmacology Unit, Massachusetts General Hospital;  Harvard Institute of Psychiatric Epidemiology and Genetics; ^§ Harvard School of Medicine; and  Harvard School of Public Health, Boston, Massachusetts.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  To examine the specificity of risk for alcohol or drug abuse or dependence (substance use disorders [SUDs]) in offspring exposed to particular subtypes of parental SUDs.

Methods.  The original sample was derived from 2 groups of index children: 140 attention-deficit/hyperactivity disorder (ADHD) probands and 120 non-ADHD comparison probands. These groups had 174 and 129 biological siblings and 279 and 240 parents, respectively.

Results.  Independent of familial risk, exposure to parental SUDs predicted SUDs in the offspring. Controlling for duration of exposure, we found that adolescence was a critical developmental period for exposure to parental SUDs. Because all our analyses controlled for social class, ADHD status, and parental lifetime history of SUDs, these results show that exposure to parental SUDs predicts offspring SUDs independently of these risk factors.

Conclusions.  These results support the critical importance of familial environmental risk factors for the development of SUDs in youth in general and particularly in those at high risk for these disorders. These results highlight adolescence as a critical period for the deleterious effects of exposure to parental SUDs, supporting the need to develop preventive and early intervention strategies targeted at adolescents at high risk for SUDs.  Key words:  substance use, adolescent, exposure, critical period.

Juvenile-onset alcohol or drug abuse and dependence (substance use disorders [SUDs]) continue to be a major clinical and public health concern. Current statistics document the enormity of the problem. Treatable addictive disorders exact tremendous human, social, and economic costs: individual suffering, family breakups, suicide, crime, violence, homelessness, physical illness, impaired work performance, and total disability. Mental and addictive disorders cost the US economy more than $170 billion annually, including productivity losses of $14 billion, health care costs of $82 billion, and other costs of $80 billion. These are strategic national concerns, affecting productivity, employment, education, mortality, and public safety. Given the risk of severe dysfunction, morbidity, and mortality associated with SUDs in youth,¹ prevention and early identification efforts are needed.

One group of youth at high risk for developing SUDs are the children of parents with SUDs.² Although a large body of literature documents this association,^2-4 the extant literature on the subject is limited in important ways. Most studies about children of substance-abusing parents have studied adult offspring. Much less is known about the impact of parental SUDs on preadolescent and adolescent children. In addition, most previous studies examined children of alcoholic parents, typically selected from alcohol treatment centers. Results from such studies may not generalize to samples of parents not selected for clinical referral or to children of drug-abusing parents.

In one of the few reports of its kind⁵ that specifically evaluated substance use outcomes in adolescent offspring at risk for SUDs, we demonstrated SUD transmission from parents to young children. Our results also suggested a possible common diathesis that is expressed as comorbid alcohol and drug use in the children of alcoholic parents. This work could not clarify the mode of transmission because family studies cannot disentangle genetic and environmental sources of transmission.⁶ Although all children of parents with SUDs are at risk for SUDs through genetic transmission,^7-11 these children also are exposed to parental SUD behaviors to varying degrees. For example, some parents may stop abusing substances before a child's birth, whereas others abuse substances to varying degrees throughout the child's lifetime.

To follow up on our previous results, we sought to clarify the mode of intergenerational SUD transmission by investigating the role of exposure to parental SUDs as a risk for SUDs in children. We examined 3 research questions:

1. Is exposure to parental SUDs associated with SUDs in offspring independently from the genetic risk imparted by parental SUDs and associated conditions?
2. Is exposure to parental SUD subtypes associated with the same SUD subtypes in offspring independently from the genetic risk imparted by parental SUDs and associated conditions? and
3. Does the timing of exposure to parental SUDs matter independently from the genetic risk imparted by parental SUDs and associated conditions?

	METHODS

Top Abstract Methods Results Discussion References

Families were selected from 2 groups of index children participating in a family study of ADHD: 140 ADHD probands and 120 non-ADHD comparison probands.¹² We recruited children with ADHD for this project because its initial goal was to study the long-term outcome and family history of ADHD and associated conditions such as substance abuse. These groups had 174 and 129 biological siblings and 279 and 240 parents, respectively. Probands were white, non-Hispanic boys between the ages of 6 and 17. Potential probands were excluded if they had been adopted or if their nuclear family was not available for study. Probands also were excluded if they had major sensorimotor disabilities (paralysis, deafness, blindness), psychosis, autism, or a full-scale IQ <80. Two independent sources provided the probands: psychiatric and pediatric referrals. Within each setting, non-ADHD controls were selected from active outpatients at pediatric medical centers. A 3-stage ascertainment procedure was used to select all probands regardless of their source of referral (ie, psychiatric or pediatric).

Probands, their siblings, and their parents were assessed at baseline. Probands and their siblings were assessed again 1 and 4 years later. As described by Biederman et al,¹³ the rates of successful follow-up at 1 and 4 years did not differ between proband groups. Moreover, at 4-year follow-up, 87% of the siblings of ADHD probands and 91% of the siblings of controls seen at baseline were recruited successfully.¹⁴ Rates of successful follow-up at 4 years and acquisition of new siblings did not differ significantly between the groups. There were no significant differences between probands and siblings successfully recruited and those lost to follow-up on any of the measures used in this investigation (detailed information available on request).

We assessed socioeconomic status (SES) with the Hollingshead-Redlich scale.¹ All diagnostic assessments were made using structured interviews based on the Diagnostic and Statistical Manual-III-Revised (DSM-III-R). Psychiatric assessments (including alcohol and drug abuse or dependence) of probands and siblings were made with the Schedule for Affective Disorders and Schizophrenia for School-Age Children, Epidemiologic Version (Kiddie-SADS-E).¹⁵ All follow-up assessments were blind to previous evaluations of the same subject. Diagnoses were based on independent interviews with the mothers and direct interviews of probands and siblings, except for children younger than 12 years, who were not interviewed directly. All assessments were made by raters who were blinded to proband diagnosis (ADHD or control) and ascertainment site. All participants older than 12 years gave written informed consent for participation. Parents gave written informed consent for participation of children under 12, and these children participated only if they assented to the study procedures. The Subcommittee on Human Studies of Massachusetts General Hospital approved this study.

All adults were interviewed with the Structured Clinical Interview for DSM-III-R (SCID)¹⁶ to assess psychopathology. To assess childhood diagnoses in adults, an addition to the SCID interview derived from the Kiddie-SADS-E to cover childhood diagnoses was administered. Previous work from our group has shown that these retrospective diagnoses can be made in a reliable and valid manner.¹⁷ As suggested by others,^18,19 we diagnosed major depression only if the depressive episode was associated with marked impairment. Because the anxiety disorders comprise many syndromes with a wide range of severity, we considered 2 or more anxiety disorders to indicate the presence of a clinically meaningful anxiety syndrome.²⁰

Interviews were conducted by raters with undergraduate degrees in psychology who had been trained to high levels of interrater reliability. We computed kappa coefficients of agreement by having 3 experienced, board-certified child and adult psychiatrists diagnose 173 participants from audiotaped interviews made by the assessment staff. The kappa coefficients of agreement were excellent, with a mean kappa of .85.

All diagnoses (including alcohol and drug abuse or dependence) were considered positive if, based on the interview results, DSM-III-R criteria were unequivocally met. All diagnostic uncertainties were resolved by a committee of 4 board-certified child and adult psychiatrists who were blinded to the participant's ascertainment group, ascertainment site, all data collected from other family members, and all nondiagnostic data (eg, cognitive functioning). Diagnoses presented for review were considered positive only if the consensus was that criteria were met to a degree that would be considered clinically meaningful. For children older than 12, data from direct and indirect interviews were combined by considering a diagnostic criterion positive if it was endorsed in either interview.

Exposure to parental SUDs was measured as the proportion of the child's life exposed to any SUD (ie, alcohol abuse or dependence or other drug abuse or dependence) in either parent. Exposure must have occurred during the child's lifetime (ie, after birth) to be incorporated into this variable. This proportion was calculated by dividing the child's age in years when first exposed to the parental disorder divided by the child's age in years at the time of the interview. Similarly, exposure to parental SUD subtypes was examined by looking at whether the child was exposed to an alcohol use disorder (abuse or dependence) or another drug use disorder (abuse or dependence) in either parent during his or her lifetime.

All offspring seen at either baseline or the 4-year follow-up (N = 556) were included in the data analyses. Key demographic and psychiatric characteristics of these children and their parents are listed in Table 1.

The associations between exposure to parental SUDs and the SUD outcomes of interest were examined using logistic regression. All analyses were controlled for ADHD status, SES, and familial risk (ie, parental lifetime SUD history). By statistically controlling for this familial risk, we could examine the transgenerational association of SUDs via exposure while accounting for these potentially confounding variables. In addition, because familial risk refers to lifetime diagnoses in the parents, the child may have never been exposed directly to the parent's psychopathology. Finally, we adjusted for duration of exposure to parental SUDs in analyses examining the critical periods in the life of the offspring.

We avoided the statistical problems associated with correlated family data (ie, offspring from the same family) by using Huber's²¹ formula as implemented in Stata²² to produce robust statistical tests for logistic regression models. Multiple comparisons were avoided by testing only the variables believed to be relevant a priori. All analyses were 2-tailed. Statistical significance was determined at the P = .05 level.

	RESULTS

Top Abstract Methods Results Discussion References

Association Between Exposure to Any Parental SUD and SUD in Offspring

Figure 1 shows the association between offspring SUDs and the proportion of their lives during which they were exposed to any parental SUD (alcohol or drug abuse or dependence). Offspring with a SUD spent a greater proportion of their lives exposed to parental SUDs than offspring who did not have a SUD (33% vs 18%, odds ratio [OR] = 2.4, P = .032). This association remained significant after adjustment for ADHD status, familial risk, and SES. Figure 1 also shows an elevated proportion of exposure to parental SUDs among offspring with an alcohol use disorder, with a trend toward statistical significance (OR = 2.0; P = .076). Finally, there was a positive but nonsignificant association between exposure to any parental SUD and childhood drug use disorders (OR = 1.8; P = .252).

View larger version (53K): [in this window] [in a new window]   Fig. 1.   Relationship between exposure to parental substance use disorders and substance use disorder subtype at follow-up.

Association Between Exposure to Parental SUD Subtypes and the Same SUD Subtypes in Offspring

Figure 2 shows more exposure to parental alcohol use disorder among offspring who had an alcohol use disorder (27% vs 15%), but the result was not statistically significant (OR = 1.9; P = .106). A similar pattern of results was found for the relationship between exposure to a parental drug use disorder and drug use in the offspring (OR = 1.1; P = .856).

View larger version (44K): [in this window] [in a new window]   Fig. 2.   Relationship between exposure to parental substance use disorder subtypes and same substance use disorder subtype in offspring at follow-up.

The Timing of Exposure to Parental SUDs

To evaluate whether there are critical periods during development in which offspring are at particular risk from parental SUDs, we examined whether exposure to parental SUDs occurred during the preschool, (0-5), latency (6-12), or adolescent (13+) years. In an analysis that controlled for children's ADHD status, familial SUD risk, SES, and duration of exposure, we found that exposure during adolescence was significantly associated with SUDs in offspring (OR = 3.0; P = .036). In contrast, there was no significant association between exposure to parental SUDs during preschool years or latency years and SUDs in offspring (OR = 0.7, P = .558, and OR = 1.7, P = .308, respectively). As can be seen in Fig 3, 53% of the offspring who were exposed to parental SUDs during adolescence had SUDs, compared with 15% of those who were not exposed as adolescents.

View larger version (50K): [in this window] [in a new window]   Fig. 3.   Critical periods of exposure to parental substance use disorder.

	DISCUSSION

Top Abstract Methods Results Discussion References

Our systematic evaluation of the impact of exposure to parental SUDs found that such exposure predicted SUDs in offspring. Because our analyses controlled for parental lifetime SUD history, ADHD status, and SES, our results show that exposure to parental SUDs predicts offspring SUDs independently of these risk factors. These results support the critical importance of familial environmental risk factors for the development of SUDs in youth in general and particularly in those at high risk for these disorders. Controlling for duration of exposure, we also found that adolescence was a critical developmental period for exposure to parental SUDs. These results especially highlight adolescence as a critical period for the deleterious effects of exposure to parental SUDs.

These results extend previous findings by our group and others documenting familial SUD transmission.^5,7,8,23 Although twin studies have shown that genes account for some familial SUD transmission, they also implicate environmental factors. Our work provides evidence that exposure to parental SUDs is one of these environmental risk factors.

Although positive associations were seen for exposure to parental alcohol and drug use disorders in predicting offspring SUD type, statistical evidence for specificity of exposure could not be established with these data. These results could be interpreted to mean that the effects of exposure to parental SUDs are nonspecific, conferring a general risk for SUDs to offspring. However, these results could have been limited by insufficient statistical power because we were examining exposure to subtypes of parental disorder. Therefore, the relationship between exposure to parental alcohol disorder and alcohol disorder in the offspring should be examined further because our results suggest the possibility of a meaningful effect (27% of life exposed in those with an alcohol use disorder vs 15% in those without). Regarding the drug use disorder findings, another possible explanation is that because drugs are illegal, parental drug use may be more likely to be covert than is parental alcohol use, limiting its detrimental effect on the child.

Although many details about the cause of SUDs remain unknown, our data and the literature are consistent with the hypothesis that SUDs may develop when a genetic predisposition is triggered by environmental events.^7-11 Exposure to parental SUDs is only one of many environmental risk factors that may be relevant to SUDs. It is likely that many environmental risk factors combine to elevate the risk of SUDs in youth, as has been found for the prediction of mental disorders. For example, Rutter et al^24,25 concluded that it was the aggregate of adversity factors, rather than the presence of any single one, that increased the risk of mental disorders in youth. Other studies found similar results.^26,27

Although our data cannot identify a specific mechanism underlying the effects of exposure to parental SUDs, they are consistent with social learning theories of substance use, especially with regard to parenting styles and parental modeling, which have been shown to be important risk factors in the substance use of young people.^28-30 Exposure to laissez-faire parenting, with poor and inconsistent family management practices, is one of the major risk factors for alcohol and drug problems in adolescence.²⁹

A crucial finding was that not all exposures to parental SUDs had the same impact on offspring. Although exposure in childhood conferred almost a twofold risk, this was not significant. In contrast, exposure during adolescence conferred a threefold risk that was highly significant clinically and statistically. It is well-known that adolescence is a critical developmental period for the emergence of SUDs.^31-34 One major reason is that adolescents are more likely than younger children to have access to alcohol and other drugs. Thus, we would expect factors that trigger the genetic susceptibility to SUDs to be more potent during adolescence than during childhood. Our data are consistent with this idea and stress the critical importance of adolescence as a major developmental period of highest risk for SUDs in youth.

The importance of when risk factors occur also has been seen in the link between bipolar disorder and SUDs. Although it has been known for some time that bipolar disorder is associated with substance use, we showed that adolescent-onset bipolar disorder but not child-onset bipolar disorder predicted onset of SUDs.^35,36

These findings about the timing of risk factors suggest that etiologic models for SUDs must consider the developmental sequence of gene expression from conception through birth and from childhood through adulthood. In this regard, our results are consistent with a general etiologic model proposed by Faraone et al.⁶ In Fig 4 we apply this model to SUDs but extend it by adding some of the many risk factors relevant to SUDs. The top of the figure defines the putative starting point of SUDs: the set of disease-predisposing genes that are determined at conception. This predisposition in conjunction with prenatal events and infant complications may lead to brain abnormalities that predispose to SUDs.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Hypothetical developmental sequence of the cause of substance use disorders.

The next step in the pathophysiologic chain occurs when the brain is exposed to environmental events in childhood, before adolescence. These include specific stressful life events such as child abuse, natural disasters, and exposure to crime as well as broader, chronic exposures to social adversity such as poverty and lack of parental supervision. These could lead to a deterioration of coping ability, additional brain abnormalities, and the onset of psychopathology. The figure also shows that during adolescence, the effects of access to substances, parental modeling, episodes of psychopathology, and peer influences can lead to SUD onset in adolescence.

Given the personal and public health consequences of SUDs and their costs to society, it is likely that our results will have implications for both government policy and pediatric primary care initiatives. Clearly, our finding that exposure to parental SUDs predicts SUDs in their offspring supports any policy that seeks to identify children at risk for preventive interventions. Although our study was naturalistic, not experimental, it suggests that reducing a child's exposure to substance use behaviors should reduce the child's risk for developing SUDs. Because both the government and medical sectors have limited resources, our findings further suggest that because adolescence is a critical developmental period for exposure to parental SUDs, funds should be focused on children in this age range.

Given its poor prognosis, any signs of substance use in children and adolescents should be treated aggressively because it is much easier to treat a SUD during its formative stages than after it has become a severe and chronic condition. Although it is premature to craft exact guidelines for practicing pediatricians, our results provide some suggestions for helping children at high risk of substance abuse. Taking a family history of substance abuse is fairly straightforward and would help pediatricians identify the children at highest risk. Although family studies suggest that parents may underreport their own substance use and abuse, a positive report can be considered a valid and useful source of data.⁶ Considering the covert nature of substance use in parents and teenagers, eliciting an accurate SUD history is critical to preventing adolescent-onset SUDs, which are known to have a very poor prognosis.³⁶

There are 2 main values of this information: education and referral. Substance-abusing parents need to know that they are placing their children at high risk for substance abuse and that they should seek treatment for the disorder not only for their sake but also for the well-being of their children. Even if referral to a mental health professional does not stop the parent's addiction, it could be useful in modifying the parenting styles and modeling behaviors that may lead to substance use in high-risk children.³⁷

Our findings should be interpreted with caution in light of methodologic limitations. Because we studied families selected for the presence or absence of ADHD in a proband child, these findings may not generalize to other samples. Moreover, because of the known familial link between ADHD and SUDs,^12,38,39 these families provide a unique perspective on parent-child SUD transmission. Most importantly, because the parents of children with ADHD and SUDs were not selected for having sought treatment for SUDs, they represent a broader range of parental SUD cases and a broader range of generalizability than has been possible in previous studies of substance-abusing families. In fact, because the rates of treatment seeking were low in these parents (18%), our data are a useful addition to the extant literature, which derives primarily from studies of referred parents.

Another limitation of this investigation is that the offspring were young at their last assessment (mean age: 16 years). Because they were still in the period of risk for developing alcohol and drug use disorders, we need additional follow-up data to determine whether further SUD onsets are consistent with the findings reported in this article. Because children under age 12 were assessed using only indirect reports from parents, we may have underestimated the degree of substance use in this sample.

Despite these limitations, our findings suggest that exposure to parental SUDs places offspring at high risk for a SUD, especially if the exposure occurs during adolescence. These findings suggest that preventive and early intervention strategies should be targeted at adolescents at high risk for SUDs. They also suggest that primary care physicians and mental health professionals should screen adolescent patients and their parents routinely for SUDs.

	ACKNOWLEDGMENTS

This work was supported in part by USPHS NIMH grants R01 MH41314-07 (J.B.), RO1 MH-50657-04 (J.B.), R01 MH57934 (S.V.F.), and R13 MH59126 (S.V.F.); NICHD grant R01HD37694 (S.V.F.); and NIDA grant K01 DA00294-01 (to Sharon Milberger, ScD).

We thank Sharon Milberger, ScD, for her assistance with this manuscript.

	FOOTNOTES

Received for publication Aug 19, 1999; accepted May 19, 2000.

Reprint requests to (J.B.) Pediatric Psychopharmacology, ACC-725, Massachusetts General Hospital, 15 Parkman St, Boston, MA 02114.

	ABBREVIATIONS

SUD, substance use disorder; ADHD, attention-deficit/hyperactivity disorder; SES, socioeconomic status; DSM-III-R, Diagnostic and Statistical Manual III-Revised; Kiddie-SADS-E, Schedule for Affective Disorders and Schizophrenia for School-Age Children, Epidemiologic Version ; SCID, Structured Clinical Interview for DSM-III-R; OR, odds ratio.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Hollingshead AB. Four Factor Index of Social Status. New Haven, CT: Yale Press; 1975
2. Wilens T, Biederman J Psychopathology in preadolescent children at high risk for substance abuse: a review of the literature. Harv Rev Psychiatry. 1993; 1:207-218 [Medline]
3. Earls F, Reich W, Jung KG, Cloninger R Psychopathology in children of alcoholic and antisocial parents. Alcohol Clin Exp Res. 1988; 12:481-487 [CrossRef][Medline]
4. Weintraub S. Children and adolescents at risk for substance abuse and psychopathology. Int J Addict. 1990-1991;25:481-494
5. Milberger S, Faraone SV, Biederman J, Chu MP, Feighner JA Substance use disorders in high-risk adolescent offspring. Am J Addict. 1999; 8:211-219 [CrossRef][Medline]
6. Faraone SV, Tsuang D, Tsuang MT. Genetics and Mental Disorders: A Guide for Students, Clinicians, and Researchers. New York, NY: Guilford; 1999
7. Tsuang MT, Lyons MJ, Eisen SA, Genetic influences on DSM-III-R drug abuse and dependence: a study of 3,372 twin pairs. Am J Med Genet. 1996; 67:473-477 [CrossRef][Medline]
8. Prescott C, Kendler K Genetic and environmental contributions to alcohol abuse and dependence in a population-based sample of male twins. Am J Psychiatry. 1999; 156:34-40 [Abstract/Free Full Text]
9. Kendler K, Prescott C, Neale M, Pedersen N Temperance board registration for alcohol abuse in a national sample of Swedish male twins, born 1902 to 1949. Arch Gen Psychiatry. 1997; 54:178-184 [Abstract]
10. Pedersen N. Twin similarity for usage of common drugs. In: Gredda L, Paris P, Nance W, eds. Twin Research 3: Epidemiological and Clinical Studies. New York, NY: Liss; 1981
11. Pickens RW, Svikis DS, McGue M, Lykken DT, Heston LL, Clayton PJ Heterogeneity in the inheritance of alcoholism: a study of male and female twins. Arch Gen Psychiatry. 1991; 48:19-28 [Abstract]
12. Biederman J, Faraone SV, Keenan K, Further evidence for family-genetic risk factors in attention deficit hyperactivity disorder. Patterns of comorbidity in probands and relatives in psychiatrically and pediatrically referred samples. Arch Gen Psychiatry. 1992; 49:728-738 [Abstract]
13. Biederman J, Faraone S, Milberger S, A prospective 4-year follow-up study of attention-deficit hyperactivity and related disorders. Arch Gen Psychiatry. 1996; 53:437-446 [Abstract]
14. Faraone SV, Biederman J, Mennin D, Gershon J, Tsuang M A prospective four year follow-up study of children at risk for attention deficit hyperactivity disorder: psychiatric, neuropsychological and psychosocial outcome. J Am Acad Child Adolesc Psychiatry. 1996; 35:1449-1459 [CrossRef][Medline]
15. Orvaschel H, Puig-Antich J. Schedule for Affective Disorders and Schizophrenia for School-Age Children: Epidemiologic 4th Version. Ft Lauderdale, FL: Nova University, Center for Psychological Study; 1987
16. Williams JBW, Gibbon M, First MB, Structured Clinical Interview for DSM-III-R (SCID). Arch Gen Psychiatry. 1992; 49:630 [Abstract]
17. Biederman J, Faraone SV, Spencer T, Patterns of psychiatric comorbidity, cognition, and psychosocial functioning in adults with attention deficit hyperactivity disorder. Am J Psychiatry. 1993; 150:1792-1798 [Abstract/Free Full Text]
18. Gershon ES, Hamovit J, Guroff JJ, A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands. Arch Gen Psychiatry. 1982; 39:1157-1167 [Abstract]
19. Weissman MM, Gershon ES, Kidd KK, Psychiatric disorders in the relatives of probands with affective disorders. The Yale University-National Institute of Mental Health Collaborative Study. Arch Gen Psychiatry. 1984; 41:13-21 [Abstract]
20. Edelbrock C, Costello M, Dulcan M, Conover NC, Kala R Parent-child agreement on child psychiatric symptoms assessed via structured interview. J Child Psychol Psychiatry. 1986; 27:181-190 [Medline]
21. Huber PJ The behavior of maximum likelihood estimates under non-standard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability. 1967; 1:221-233
22. Stata Corporation. Stata Reference Manual. Release 5. College Station, TX: Stata Corporation; 1997
23. Grove WM, Eckert ED, Heston L, Bouchard TJ, Segal N, Lykken DT Heritability of substance abuse and antisocial behavior: a study of monozygotic twins reared apart. Biol Psychiatry. 1990; 27:1293-1304 [CrossRef][Medline]
24. Rutter M, Cox A, Tupling C, Berger M, Yule W Attainment and adjustment in two geographical areas: Vol. 1. The prevalence of psychiatric disorders. Br J Psychiatry. 1975; 126:493-509 [Abstract/Free Full Text]
25. Rutter M, Quinton D. Psychiatric disorder: ecological factors and concepts of causation. In: McGurk H, ed. Ecological Factors in Human Development. Amsterdam, The Netherlands: North Holland; 1977:173-187
26. Biederman J, Milberger S, Faraone SV, Family-environment risk factors for attention deficit hyperactivity disorder: a test of Rutter's indicators of adversity. Arch Gen Psychiatry. 1995; 52:464-470 [Abstract]
27. Blanz B, Schmidt MH, Esser G Familial adversities and child psychiatric disorders. J Child Psychol Psychiatry. 1991; 32:939-950 [CrossRef][Medline]
28. Bailey GW Current perspectives on substance abuse in youth. J Am Acad Child Adolesc Psychiatry. 1989; 28:151-162 [CrossRef][Medline]
29. Hawkins JD, Catalano RF, Miller JY Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: implications for substance abuse prevention. Psychol Bull. 1992; 112:64-105 [CrossRef][Medline]
30. Petraitis J, Flay BR, Miller TQ Reviewing theories of adolescent substance use: organizing pieces in the puzzle. Psychol Bull. 1995; 177:67-86 [CrossRef]
31. Groves JB, Batey SR, Wright HH Psychoactive-drug use among adolescents with psychiatric disorders. Am J Hosp Pharm. 1986; 43:1714-1718 [Abstract]
32. Kandel DB, Logan JA Patterns of drug use from adolescence to young adulthood: I. Periods of risk for initiation, continued use and discontinuation. Am J Public Health. 1984; 74:660-666 [Abstract/Free Full Text]
33. Newcomb DR, Maddahian E, Bentler PM Risk factors for drug use among adolescents: concurrent and longitudinal analyses. Am J Public Health. 1986; 76:525-531 [Abstract/Free Full Text]
34. Yamaguchi K, Kandell DB Patterns of drug use from adolescence to young adulthood: II. Sequences of progression. Am J Public Health. 1984; 74:668-672 [Abstract/Free Full Text]
35. Biederman J, Faraone SV, Mick E, Attention deficit hyperactivity disorder and juvenile mania: an overlooked comorbidity? J Am Acad Child Adolesc Psychiatry. 1996; 35:997-1008 [CrossRef][Medline]
36. Wilens T, Biederman J, Abrantes AM, Spencer TJ Clinical characteristics of psychiatrically referred adolescent outpatients with substance use disorder. J Am Acad Child Adolesc Psychiatry. 1997; 36:941-947 [CrossRef][Medline]
37. Steinhausen HC, Metzke CW Frequency and correlates of substance use among preadolescents and adolescents in a Swiss epidemiological study. J Child Psychol Psychiatry. 1998; 39:387-397 [CrossRef][Medline]
38. Milberger S, Faraone SV, Biederman J, Chu MP, Wilens T Familial risk analysis of the association between attention-deficit/hyperactivity disorder and psychoactive substance use disorders. Arch Pediatr Adolesc Med. 1998; 152:945-951 [Abstract/Free Full Text]
39. Biederman J, Faraone SV, Keenan K, Knee D, Tsuang MT Family-genetic and psychosocial risk factors in DSM-III attention deficit disorder. J Am Acad Child Adolesc Psychiatry. 1990; 29:526-533 [Medline]