Director's Report to the National Advisory Council on Drug Abuse
Epidemiology, Etiology and Prevention Research
Profile of New York City Homicides
Results of toxicological testing of homicide victims reveals recent cocaine use by 19.0 percent of victims killed in the street (n=1,466 after exclusion of those whose survival time would have permitted elimination of assayed substances) and 15.8 percent of those killed in their own home (n=438, same exclusion). Rates of ethanol involvement were 21.4 and 16.0 percent, respectively, for victims in street and home homicides. Compared to those killed in their homes, homicide victims killed in the street were more likely to be male, to be 15-24 years of age, and to be African American. They also were more likely to be killed by a firearm. This study was based on medical examiner files for 4468 homicides that occurred in New York City in 1990 and 1991; results of primary interest pertain to homicides that occurred in the street or other outdoor places (50.1 percent of the 4468 cases) and in the victim's home (19.3 percent). Tardiff, K., Marzuk, P., Leon, A. Hirsch, C., Marina, S. Portera, L, and Hartwell, N. A Profile of Homicides on the Streets and in the Homes of New York City. Public Health Reports, 110(1): 13-17, 1995.
Feasibility of Half-Sibling Designs for Detecting a Genetic Component to a Disease
In genetic epidemiology, adoption and twin study designs are the most commonly used designs to identify genetic and common environmental components underlying familial aggregation but such families are not typical of families in drug abuse research. The feasibility of half-sibling designs for detecting a genetic (single-locus or multifactorial) source of familial aggregation of a disease were studied and compared to both adoption and twin study designs. Results for two types of design where the sample units were relative pairs reared apart (I) or reared together (II) indicated that the half-sibling designs were feasible but required more observations to achieve the same power. Designs involving half-siblings may be more feasible in situations in which greater availability exists of these relatives. This approach broadens opportunities for genetic research in nontraditional families. (Tierney, C, Merikangas, KR, and Risch, N. Feasibility of Half-Sibling Designs for Detecting a Genetic Component to a Disease. Genetic Epidemiology 11:523-538, 1994).
Co-morbidity and Co-transmission of Alcoholism, Anxiety, and Depression
Familial transmission analyses confirm findings regarding the specificity of transmission of alcohol and anxiety disorders. Relatives of probands with substance abuse/dependence showed increased rates of drug abuse/dependence and alcohol dependence (not alcohol abuse) compared to relatives of probands with alcohol abuse/dependence, anxiety disorders or normal controls. Specificity of transmission of drug abuse and alcohol abuse and specificity within specific drug classes also occurred. Results suggest that relatives of probands with substance abuse have more than a 2-fold increased risk of developing a substance abuse problem themselves. The rate of disruptive disorders was significantly elevated in offspring (age 7-18) of fathers with diagnoses of both substance abuse and antisocial personality disorder (44.4%) compared to offspring of fathers with anxiety or affective diagnoses (19.2%), both higher than rates of offspring of substance abusers without antisocial personality disorder (9.1%) and normals (7.1%). Analysis of subset of children (age 11-18) indicates that more than 27% of children of fathers with both antisocial personality disorder and substance abuse met diagnostic criteria for substance abuse or dependence whereas the risk for substance abuse was highest for offspring of mothers with a substance abuse diagnosis (30.8%), a 2-fold increase over offspring of mothers with anxiety or affective disorders and a 10fold increase over offspring of normals. (Merikangas, KR, Risch, NJ, and Weissman, MM. Comorbidity and Co-transmission of Alcoholism, Anxiety and Depression. Psychological Medicine 24:69-80, 1994).
Differential Gender-Related Treatment Needs of the Homeless
The historical status of the homeless as being primarily older white men with alcohol problems is changing as the proportion of women in this population increases. In a 19-site Los Angeles County study comparing homeless men (n=386) and women (n=145), researchers found that homeless men, compared to homeless women, reported more substance use, longer periods of homelessness, poorer housing quality when not literally homeless, and greater criminal involvement. Homeless women were more likely than homeless men to have children living with them. Homeless men, compared to homeless women, showed stronger relationships between mental illness and prior institutionalization, between mental illness and drug use, between drug use and victimization. Homeless women, on the other hand, showed stronger relationships between drug use and alcohol use and between criminal involvement and drug use than did homeless men. Among all the homeless, men were more likely to use drugs and alcohol than were women, but among the homeless who are substance abusers, women were more likely than men to use multiple substances. These results underscore the differences in substance abuse treatment needs between homeless men and women. Stein, J. and Gelberg, L. Homeless Men and Women: Differential Associations Among Substance Abuse, Psychosocial Factors, and Severity of Homelessness. Experimental and Clinical Psychopharmacology, 3(1): 75-86, 1995.
Substance Abuse Involvement Among Juvenile Murderers
In order to determine the extent to which juvenile offenders, convicted of murder or manslaughter, were involved with substance abuse at the time of offense, Fendrich et al., interviewed incarcerated juveniles to determine patterns of substance involvement among them as compared with patterns found in older offenders. Irrespective of age, close to one third of all homicide perpetrators reported that they were affected by alcohol prior to the offense. Alcohol was the substance showing the highest rate of "regular" lifetime use and the highest rate of ingestion in the week preceding the homicide. In many respects, the reported substance use patterns of the 16-17 year old age group were closer to the patterns demonstrated by the oldest (36+) age group than they were to the adjacent 18-20 year old group. Most of the juveniles who were using illicit drugs other than alcohol prior to the homicide attributed the homicide to the effects of those substances. Narrative accounts suggest that substances escalated impulsive, spontaneous violent outbursts. Fendrich, M., Mackesy-Amiti, M.E., Goldstein, P., Spunt, B., Brownstein, H. Substance Involvement Among Juvenile Murderers: Comparison with Older Offenders Based on Interviews with Prison Inmates. International Journal of the Addictions, in press, 1995.
Inconsistencies in Self-Reported Drug Use
Self-report information, in general, but particularly when provided by substance abusers, has traditionally been plagued with inconsistencies. Fendrich and his colleagues evaluated discrepancies in responses to questions about lifetime cocaine and marijuana use asked of nearly 10,000 respondents in the National Longitudinal Survey of Youth, a study that began in 1979. When examining responses from 1984 and 1988, they found that one in seven responses regarding cocaine use were inconsistent in some way, such as when it was first used or how often; one person in six did not tell the truth about marijuana use; African-Americans underreported drug use at higher rates than whites which Fendrich believes may have to do with how much the questions have done to enhance the level of trust. The types of inconsistencies varied according to the substance; cocaine reports yielded more inconsistencies with regard to timing of first use while for marijuana, most of the inconsistences were with respect to disclosure. Fendrich, M., Mackesy-Amiti, M.E. Inconsistencies in Lifetime Cocaine and Marijuana Use Reports: Impact on Prevalence and Incidence. Addiction, 90, 111-118, 1995.
Validity of Self-Reported Drug Use Among Injection Drug Users and Crack Cocaine Users
Norman Weatherby, Ph.D. and associates at the University of Miami School of Medicine examined the validity of self-reported drug use as a measure of behavior change for the evaluation of drug use prevention and HIV risk reduction programs. They recruited not-in-treatment IDUs and crack cocaine users through street outreach in four U.S. cities and compared self-reports of drug use in the past 48 hours from the Risk Behavioral Assessment questionnaire (developed by NIDA in 1991) with data from urine tests. Very high agreement was found between self-report and urine test results (86 percent agreement for recent cocaine use; 85 percent agreement for recent heroin use). The results suggest that self-reported drug use in not-in-treatment populations is accurate enough for measuring changes in risk behavior practices. Urinalysis may not be necessary if respondents are asked about their drug use in a nonthreatening manner, and if they are assured of the confidentiality of their results. Weatherby, N., Needle, R., Cesari, H., Booth, R., McCoy, C., Watters, J., Williams, M., and Chitwood, D. Validity of Self-Reported Drug Use Among Injection Drug Users and Crack Cocaine Users Recruited Through Street Outreach. Evaluation and Program Planning. 17(4):347-355, 1994.
Symptom and Substance Use Reporting Consistency
To investigate accuracy of recall, and factors that potentially affect recall, Drs. Fendrich and Warner from the Institute for Juvenile Research, University of Illinois at Chicago, examined 2-year recall reports of lifetime symptomatology and substance use questions on the K-SADS-E. Comparisons were made between those who forgot and those who remembered reports of screening symptoms made at an initial interview. In general, individuals with externalizing disorders (conduct disorder and substance use) recalled more of their presenting symptomatology than did those individuals suffering from internalizing disorders (depression and anxiety disorders. Conduct disorder and substance use symptoms were characterized by the highest overall median kappa and recall percentage statistics; median kappa values for symptoms constituting both disorders were well above .40. One explanation supported by the findings suggests that consistent retrospective reporting may depend on the extent to which symptoms and behaviors (e.g., drug use) create a lasting impression on the young respondent reporting them.
Fendrich, M., Warner, V. Symptom and Substance Use Reporting Consistency Over Two Years for Offspring at High and Low Risk for Depression. Journal of Abnormal Child Psychology, 22, 425-439, 1994.
Criminal Activities and Drug Using Behaviors
Dr. Richard Dembo and colleagues of the University of South Florida examined criminal activities and drug using behaviors among a sample of juvenile detainees. They found that familial drug use, mental health problems, involvement in crime, and sexual victimization or physical abuse were stronger predictors of initiation of alcohol and marijuana use compared to family socioeconomic status, composition, and size. Continued use of alcohol and marijuana among these youth was significantly related to the severity of early childhood psychological problems, physical and sexual abuse, and low selfesteem. Dembo, R., Williams, L., and Schmeidler, J. Key Findings of the Tampa Longitudinal Study of Juvenile Detainees: Contributions to a Theory of Drug Use and Delinquency Among High Risk Youth. In: Roberts, A. (ed.) Juvenile Justice: Policies, Programs, and Services. 2nd edition. Chicago, IL: Nelson-Hall (in press).
Problem Behavior Syndrome
The existence of a Problem Behavior Syndrome or clustering of deviant behaviors has been demonstrated in numerous research studies of a variety of populations. A very clear depiction of the hypothesized pattern was found in a study of 695 African American and 637 Puerto Rican adolescents. For both ethnic groups, the sequence of sexual behavior and its relationships to other problem behaviors were unidimensional, cumulative, and significantly related to the level of drug use and to the frequency of delinquent behaviors thereby suggesting an underlying problem behavior syndrome.
(Brook, JS, Balka, EB, Abernathy, T, Hamburg, BA. Sequence of Sexual Behavior and Its Relationship to Other Problem Behaviors in African American and Puerto Rican Adolescents. Journal of Genetic Psychology 155(1):107-114, 1994).
Friends' smoking exerts a stronger influence on adolescents' smoking initiation than does parents' smoking. Friends' smoking has both direct and indirect influences on adolescent initiation of smoking but only indirect effects on escalation. Parental smoking has only indirect effects on initiation and escalation, channelled through perceived approval. Parental approval of smoking mediated smoking initiation among females but not among males. Pathways were found to be somewhat different by race/ethnicity, with friends' smoking having both direct and indirect effects on initiation among white and Hispanic adolescents but only direct effects among African American, Asian and other adolescents. These findings are derived from a longitudinal study of 6,695 seventh-grade students recontacted as eighth graders (73.1 percent retention rate). Flay, B.; Hu, B.; Siddiqui, O; Day, L.; Petratis, J.; Richardson, J.; and Sussman, S. Differential Influence of Parental Smoking and Friends' Smoking on Adolescent Initiation and Escalation of Smoking. Journal of Health and Social Behavior, 35(3): 248-265, 1994.
Influences on Adolescent Smoking Initiation and Escalation.
Gender Differences in the Pattern and Progression of Substance Use in Conduct Disordered Adolescents
Conduct disordered adolescents tend to initiate drug use earlier than other adolescents, and, among those with conduct disorder, females are more likely to have experimented with nonprescription diet pills and caffeine while males are more likely to have experimented with snuff or chewing tobacco. Women are at greater risk for qualifying for a diagnosis of nicotine dependence than were men. Although women start drinking alcohol at a later age than men, the latency from initial alcohol use to diagnosis of alcohol abuse/dependence is shorter for women than men, and the age at which males and females qualify for a diagnosis of alcohol abuse/dependence does not differ. Compared to the men, women also show a shorter latency from cannabis use to diagnosis of cannabis abuse/dependence. These findings are based on a sample of 40 female and 42 male adolescents qualifying for a conjoint diagnosis of conduct disorder and substance abuse disorder. The findings suggest that although the prevalence of substance abuse is higher in men, the liability to develop a supra-threshold disorder is greater in women when they do initiate drug use. Mezzich, A., Moss, H., Tarter, R., Wolfenstein, M., Hsieh, Y-C., and Mauss, R. Gender Differences in the Pattern and Progression of Substance Use in Conduct-Distorted Adolescents. The American Journal of Addictions, 3: 289-295, 1994.
Social and Developmental Correlates of Tobacco and Alcohol Use Among Elementary School Students
To test the predictive strength of multiple risk profiles which could expand understanding of "high risk" youth and which could indicate program objectives for earlier intervention for substance use prevention, Jackson and Henriksen from the University of North Carolina report on baseline data for Kids FIRST -- a four year panel study of relationships between child initiation of alcohol and tobacco use. A significant finding related to the differences in rates of initiation of tobacco and alcohol use across gender and ethnic subgroups, and variation in substance use reported across students' levels of development. Also of interest were the findings of the associations between children's initiation of use and their parents' rule setting about tobacco and alcohol use, their parents' modeling of use and other family socialization variables. Jackson, C., Henriksen, L. Tobacco and Alcohol Use Among Elementary Grade Students: Social and Developmental Correlates in a Rural North Carolina Sample. Paper presented at APHA 122nd Annual Meeting, 1994.
Genetic Pathways to Drug Abuse. The hypothesis that genetic factors are etiologically important in drug abuse/dependency and that psychiatric problems in adoptive parents are associated with drug abuse/dependency was tested by Dr. Remi Cadoret and his colleagues. A sample of male adoptees, separated at birth from their biologic parents, were followed up as adults to determine their substance use/abuse and their psychiatric diagnosis. Half of the adoptees came from biologic parents known to be alcohol abuse/dependent and/or to have antisocial personalities (determined by hospital and/or prison records). The adoptive home environment assessment included psychiatric evaluations of adoptive parents. This study confirmed the involvement of two independent genetic factors in drug abuse/dependence. It also confirmed the previous finding that environmental factors defined by psychiatric conditions in adoptive families independently predicted antisocial personality disorder in adoptees. Adoptees born of alcohol-abusing mothers showed evidence of fetal alcohol syndrome, but this did not diminish the evidence for a direct genetic effect between an alcohol-abusing biologic parent and drug abuse/dependency in offspring. Cadoret RJ; Yates WR; Troughton E; Woodworth G; Stewart MA. Adoption Study Demonstrating Two Genetic Pathways to Drug Abuse. Archives of General Psychiatry, January, 52(1): 42-52, 1995.
Early Onset of Alcohol and Tobacco Use
Since early onset of alcohol and tobacco use is associated with continued use of both substances, risk of use of other substances and unresponsiveness to substance use prevention programs, Jackson and colleagues collected drug use prevalence and timing data directly from elementary grade children. These data included estimates of alcohol use onset and provided estimates for children living in rural areas. Of the 27% of 5th grade initiators of alcohol use, 59% first had a drink with alcohol while in 1st, 2nd or third grade; of the 21% of 5th grade initiators of tobacco use, 48% first smoked tobacco in 3rd grade or earlier. This study supports the recommendation that primary prevention of substance use begin in the primary grades. Jackson C., Dickinson, D. Early indeed: Onset of Alcohol and Tobacco Use in a Sample of Rural Elementary School Children. Journal of Drug Education, in press, 1995.
Impact of Parental Substance Use on Adolescents
Parental substance abuse increases the likelihood of adolescents' substance abuse both directly and through its association with mediating factors including heightened tolerance for deviance, lower behavioral control, increased perceived coping function of substance use, more negative life events, and greater affiliation with peer users. As an example of these relationships, children who observe parents smoking or drinking may be more likely to engage in these behaviors with age peers, and parental multisubstance use impacts deviance-related attitudes, behavioral control, expectancies about substance use, and stressful life events. These findings are based on a study of 1,775 urban adolescents (mean age 13.3 years) designed to test a mediational model including constructs from stress-coping theory (Cooper, Russell, & George, 1988), the problem behavior (deviancy) model (Jessor & Jessor, 1977), and social learning theory (Bandura, 1977). The findings underscore the need to consider parental substance use history in planning treatment for adolescents. Several treatment strategies are implied by the study, including replacing perceptions of substance use with more acceptable coping methods that also reduce the probability of affiliation with deviant peers. Wills, T., Schreibman, D., Benson, G., and Vaccaro, D. Impact of Parental Substance Use on Adolescents: A Test of a Mediational Model. Journal of Pediatric Psychology, 19(5):537-556, 1994.
Stability in Adolescent Friendship Group Context Across the School Year
The initiation and use of alcohol and cigarettes among adolescents is, to some extent, dependent on membership in friendship groups. Urberg and colleagues investigated the extent to which adolescents are in similar friendship group contexts in the Fall and Spring of a school year, and whether or not they are a member of the same friendship group. Gender and grade differences in the tendency to remain in the similar context were examined. Correlations across a range of variables between Fall and Spring were calculated separately for individuals who changed groups and for those who remained in the stable groups. For all but academic values, there was significant similarity in the group context from Fall to Spring, for both those who changed groups and those who remained in stable groups. The researchers conclude that adolescents who change peer groups do not dramatically change their peer context and that peer context is more stable for those who remain in the same group compared to those who change groups. Urberg, K., Rao, P., Mack, F., Stability in Adolescent Friendship Group Context Across the School Year. Paper presented at the Meeting of the Society for Research in Child Development, Indianapolis, March 1995.
Friendship Influence, Selection, and Deselection in Adolescent Substance Use
To untangle the often intertwined effects of friend selection versus influence in adolescent substance use, Urberg and colleagues followed Billy and Udry's model to ascertain which might be important processes in adolescent substance use. Students completed surveys in the Fall and Spring terms in which they named their best friend (BF) as well as other friends. Those students who listed a BF were matched with that BF so that the BF's data would be used. Variables of lifetime drug use and current drug use were dichotomized and both were found to increase with grade and prevalence of use was similar across waves. To determine whether influence was operating, nonusers with and without user friends were compared. Controlling for grade and gender, BF use did not make a significant contribution in predicting Spring use for the adolescent. To determine whether selection was operating, those who chose a new BF were examined. Those who had tried drugs before were most likely to chose a BF who had also tried, followed by the student who made his/her transition to use. The researchers conclude that there is more evidence for BF selection rather than BF influence on the basis of substance use. Degirmencioglu, S., Urberg, K., Friend Influence, Selection and Deselection in Adolescent Substance Use Over the School Year: A Two-Wave Analysis. Paper presented at the Meeting of the Society for Research in Child Development, Indianapolis, March 1995.
Structure of Adolescent Peer Networks
In order to study the effects of peers on any aspect of adolescent development, but particularly on the initiation of drug use, the most comprehensive data to date on the structure of school-based peer networks has been collected by Urberg and colleagues from Wayne State University. Their database consists of 90% or more of the population of three school systems with each subject naming up to 10 friends. Because subjects are matched to friends, mutuality could be assessed, and data from multiple sources could be used to assess friendship groups. The most robust finding was that females are more integrated into school social networks than males and that numeric minorities usually are less connected to school peer networks than the majority group. Of interest, few group effects replicated across the three schools. This strongly suggests that in order for programs to be maximally effective, attention must be given to school and community level differences that may substantially impact the organization of peer networks. Urberg, K., Degirmencioglu, S., Tolson J., Halliday-Scher, K., The Structure of Adolescent Peer Networks. Developmental Psychology, in press, 1995.
Reciprocal Relations Between Perceived Parenting and Adolescents' Substance Use
Deficits in parental support and control prospectively predict adolescent substance use, and adolescent substance use, in turn, is prospectively related to lower levels of parental support and control, These findings are based on data from a community sample of 441 adolescents (average age of 12.7 years at initial contact) and their parents; half of the adolescents were at risk for problem behavior because of parental alcoholism. The study highlights the need for early prevention efforts and suggests that manipulating parenting practices during childhood, before problem behaviors become ingrained, would be the most effective approach. Stice, E. and Barrera, M. A Longitudinal Examination of the Reciprocal Relations Between Perceived Parenting and Adolescents' Substance Use and Externalizing Behaviors. Developmental Psychology, 31(2): 1-13, 1995.
Attitudes and Health Behavior in Diverse Populations
Five different health behaviors (cigarette use, alcohol use, binge eating, illicit drug use, and drunk driving) were studied prospectively in 5 different groups of subjects. Associations between attitudes toward these behaviors and the behaviors themselves were investigated. Findings revealed that attitudes predicted behavior in 2 instances: alcohol use and marijuana use. Attitudes did not predict drunk driving, binge eating, or smoking behaviors. Past behavior predicted attitude in the domains of binge eating and smoking, but not in the domains of alcohol use, drunk driving, or marijuana use. Stacy, A., Bentler, P. and Flay, B. Attitudes and Health Behavior in Diverse Populations: Drunk Driving, Alcohol Use, Binge Eating, Marijuana Use, and Cigarette Use. Health Psychology, 13(1): 73-85, 1994.
[Office of the Director][Report Index][Next Report Section]