PLACING FOSTER SIBLINGS TOGETHER OR APART? A PROSPECTIVE STUDY OF SIBLINGS IN FOSTER CARE.. Lourdes Oriana Linares1, Angelique R. Torres1, Minmin Li1, Evelyn E. Segura1, Yosef Salvay1, 1NYU Child Study Center, New York, NY United States
During periods of disruption and isolation from family and peers, siblings provide each other with support and facilitate individual development (Price & Brew, 1998). From a family systemic and cultural perspective, Minuchin & Fishman (1981), and Bank & Kahn (1982) emphasize that the development of strong sibling bonds is strengthened when parental influences are insufficient as may be the case with children in neglectful families. The placement of siblings in foster homes, together or apart, provides a prospective natural experiment to investigate the course of sibling positivity and sibling negativity on child outcomes (depressive symptoms, loneliness, behavior problems, and school competence).
Method: The preliminary sample consisted of 42 sibling dyads (n=84 siblings) recruited from 13 foster care agencies in a large eastern city and assessed 2.6 (SD=2.3) months after admission to care (Time1) and one year later (Time2). Retention from Time1 to Time2 was 94%. At Time1, older sibling was 9.7(2.8) and younger sibling was 7.5(2.7) years of age; siblings were of African American 48%, Latino 34%, and Mixed 18% ethnicity. Repeated measures included: the Sibling Relationship Questionnaire, Child Depression Inventory, Loneliness Scale, Eyberg Child Behavior Inventory and Harter Social Competence. Informants included the biological and foster mother, child, and classroom teacher.
Results: Siblings placed apart experienced a two-fold increase from Time 1 (14%) to Time 2 (30%). ANCOVAs showed that foster mothers (143 vs. 119) and classroom teachers (157.0 vs. 118.2) reported higher problems for siblings placed apart than siblings placed together Time2. After controlling for baseline sibling relationships, there were no differences for sibling positivity or negativity by placement condition at Time2. Hierarchical mixed regression analyses predicting child behavior problems at Time2 indicated that sibling conflict (âβ=10.31), parental favoritism in the foster parent (âβ= -1.4), and stability of placement (âβ=10.31) independently predict child behavior problems at Time2.
Conclusion: Controlling for Time1 child problems, the data suggest that siblings placed apart have higher behavior problems in the foster home and classroom at Time2. Placing siblings together or apart does not seem to change the sibling relationship at Time2, after controlling for Time1 relationship. Beyond known sibling conflict and parental favoritism predictors, stability in the family and sibling subsystems is a factor which uniquely impacts foster sibling adjustment to care.
A REVIEW OF THE HEALTH, SOCIAL, AND PSYCHOLOGICAL CONSEQUENCES OF DRUG USE AND ABUSE. Thomas Locke1, Michael Newcomb1, 1University of California, Los Angeles, Los Angeles, CA United States
It is believed (and sometimes established) that there are consequences of drug use in many areas of life. This presentation reviews and categorizes findings in three broad areas: health, psychological, and social consequences. Various studies are included in this review to include short-term and long-term consequences.
Various short-term health consequences of drug use have been established, and include over-doses, aggression, accidents. Long-term consequences have been more difficult to test adequately, the physiological mechanisms are more difficult to determine. The use of cigarettes is associated with cancer, emphysema, and heart disease. Cigarette smoking is responsible for over 400,000 deaths per year. Alcohol use is responsible for over 400,000 alcohol-related traffic crash deaths per year and over 25,000 people die from cirrhosis of the liver (NIAAA, 2004). It is estimated that illicit drug use resulted in approximately 17000 deaths in 2000 (Mokdad et al., 2004).
Psychological problems are often observed among drug abusers. However, the directional effects between drug use and psychological impairment are not always clear. (Locke & Newcomb, 2001). Psychological consequences can include many domains such as mood disorders, anxiety, other psychiatric disorders, and disruption in adaptive coping mechanisms, attachment disruption, anger management, and suicidal ideation. For example, Newcomb et al., (1993) found that teenage polydrug use had few effects on adult mental health. However, increased polydrug use exacerbated later psychoticism, suicidal ideation, and other markers of poor adult mental health.
Social problems have also been noted among those who use drugs. Domains that are often implicated include marital and relationship satisfaction, employment stability and satisfaction, educational attainment, legal issues / criminal behavior, parenting, and social support. Moderate-term social consequences are evident in a paper by Newcomb et al., (2002). They found that tobacco use in 8th grade predicted absenteeism and drop-out in 12th grade. This was a unique finding over-and-above the strong influence of academic achievement, SES variables, classroom behavior, and general deviance.
Understanding and documenting both the short-term and long-term consequences of drug use and abuse are difficult, as drug use is intertwined with other norm-violating behaviors and attitudes (McGee & Newcomb, 1992). Further, drug use is only one influence that shapes and alters personal development.
THE INFLUENCES OF DEMOGRAPHIC FACTORS AND PERCEIVED STRESS ON DEPRESSION AMONG BLACK AND WHITE ADULT MEN AND WOMEN. Robert Jagers1, Pamela Scott-Johnson1, Terra Bowen-Reid1, 1Morgan State University, Center for Health Disparities Solutions, Baltimore, MD United States
This individual poster session addresses the SPR conference cross cutting theme of epidemiology. It describes an investigation of the ways in which demographic factors and perceived stress influence depression among Black and White men and women from low- and middle-income backgrounds. The study seeks to clarify mixed findings on mental health disparities by disentangling race, socioeconomic status and related issues, and examining their contributions to risk for depression.
As part of larger study, 1,198 Black (715) and White (483) adult men and women from two adjacent census tracts completed individual interviews using well-research measures of depression, perceived stress and various demographic characteristics. Results indicated that the racial/ethnic groups were comparable on most demographic indicators. Hierarchical regression procedures were used to examine the influence of ascribed (race/ethnic group and gender) and attained (age, income, marital status, education and employment status) demographic factors and perceived stress on depression scores for the total sample, and for each ethnic group separately. For the total sample, results indicated that Whites, older respondents, and those not in the labor force reported significantly greater depression. However, these factors explained only a small portion of the variance in depression scores (roughly 7%). The addition of perceived stress into the regression equation accounted for another 22% of the variance. The overall pattern of findings was similar when the racial/ethnic groups were considered separately. However, differences emerged among the demographic factors predictive of depression. Among Whites, not being in the workforce and having a lower income were found to be significant individual predictors. Not being in the workforce also was a significant predictor of depression for Blacks. In addition, age and being widowed, separated or divorced were associated with greater depression.
Results are discussed in terms of attained demographic status, basic and applied research on stress and coping, and the diagnosis and treatment of mental health disorders among racially and economically diverse groups.
DOES THE RELATION BETWEEN PARENTAL AND ADOLESCENT CIGARETTE SMOKING VARY AMONG DEMOGRAPHIC GROUPS. Kellie Loomis1, Olivia Ashley1, Michael Penne1, Molly Aldridge1, Karl Bauman1, Scott Novak1, Joe Gfroerer2, 1RTI International, Research Triangle Park, NC United States; 2Office of Applied Studies, Substance Abuse and Mental Health Services Administration, Rockville, MD United States
Parental cigarette smoking has been identified as a risk factor for adolescent smoking, although there is considerable speculation regarding the mechanisms that influence the intergenerational transmission of smoking. This study examines the extent to which sociodemographic characteristics (adolescent age, gender, race/ethnicity; parent education; family structure) modify the association between parent and adolescent smoking. Data are from nationally representative samples of mothers (n=2,302) and fathers (n=1,563) interviewed along with their adolescents aged 12 to 17 in the 2002 National Survey on Drug Use and Health (NSDUH), a survey of the U.S. civilian, non-institutionalized population aged 12 or older. Newly developed sample weights were used to account for overlap due to two-parent households. Descriptive statistics were calculated using SUDAAN to account for the complex NSDUH sampling design and construct estimates that are generalizable to U.S. adolescents living with a mother or living with a father. Multivariate modeling was used to examine relations between parental and adolescent smoking and demographic characteristics, controlling for family income. Results showed that 17 percent of adolescents living with a mother, and 19 percent of adolescents living with a father, had smoked during the year before the survey. Adolescent age, adolescent gender, parent education, and family structure did not modify the relation between mother or father smoking and adolescent smoking. However, adolescent race/ethnicity modified the relation between fathers´ smoking and adolescent past year smoking. Fathers´ reports of their smoking status were associated with self-reported past year smoking among black adolescents (OR = 8.02, 95% CI = 1.89, 33.93) but not among adolescents from other racial/ethnic groups (white: OR = 1.58, 95% CI = 0.94, 2.64; other: OR = 1.96, 95% CI = 0.64, 5.94). In contrast, adolescent race/ethnicity did not significantly modify the relation between mother smoking and adolescent smoking. Mothers´ smoking was associated with adolescent past year smoking (OR = 2.17, 95% CI = 1.55, 3.04). Results from this study that parental influence on adolescent smoking varies across racial/ethnic groups and differs for fathers and mothers may inform interventions targeting adolescent smoking. Such interventions should take into account parent smoking, and prevention programs addressing smoking among fathers may be more effective for black adolescents than adolescents from other racial/ethnic groups. As this study employed cross-sectional data, additional research is needed to investigate differences in family processes among blacks in contrast to other racial/ethnic groups that may modify the relation between parent smoking and adolescent smoking.
BELIEFS AMONG YOUTHS ABOUT RISKS FROM MARIJUANA USE: MEDIATION OF RACIAL/ETHNIC DIFFERENCES IN PREVALENCE OF USE. Olivia Ashley1, Jennie Harris1, David Heller1, Jason Williams1, 1RTI International, Research Triangle Park, NC United States
Racial/ethnic differences in adolescent marijuana use are well established, although the mechanisms that account for these disparities are largely unexplained. This paper seeks to address this gap by investigating the relationships between race/ethnicity, risk beliefs about marijuana use, and current marijuana use among youths. Data are from the public use file of the 2001 National Household Survey on Drug Abuse (NHSDA), a survey of the U.S. civilian, non-institutionalized population aged 12 or older. Analyses were restricted to the 17,230 youths aged 12 to 17 reporting risk beliefs about marijuana use. The survey asked about demographic characteristics, beliefs about the risks associated with marijuana use, and participants´ marijuana use during the month prior to the survey. Inferential, descriptive statistics were calculated using SUDAAN to account for the complex sampling design. The NHSDA design utilizes clusters of households, and the data are generalizable to all persons of a specified age or age group – in this study, youths aged 12 to 17. Youths were oversampled in the NHSDA to improve the precision of estimates among them. All models controlled for age, gender, past month alcohol use, and past month tobacco use. Bivariate analyses and multivariate modeling and Sobel tests of mediation were used to examine relationships between study variables. Results showed that race/ethnicity was significantly associated with marijuana use during the month before the survey among youths; the prevalence was highest among American Indians or Alaska Natives (20 percent) and lowest among Asians (4 percent). Multivariate modeling showed that risk beliefs about marijuana use mediated the relationship between race/ethnicity and marijuana use among American Indians or Alaska Natives, blacks, and Hispanics in relation to whites, controlling for potential confounding variables. Results indicate that risk beliefs may explain variation in youth marijuana use across racial/ethnic groups. Interventions targeting youth marijuana use may need to consider cultural and contextual differences in perceived risk. As this study employed cross-sectional data, additional research is needed to provide empirical evidence about cultural and contextual influences that may contribute to formation of risk beliefs about drug use and to investigate a causal relationship between risk beliefs and drug use.
PREVENTION ON THE HIGHWAYS: USING SOBRIETY CHECKPOINTS TO REDUCE DRINKING AND DRIVING. John Lacey1, Tara Kelley-Baker1, Robert Voas1, 1Pacific Institute for Research and Evaluation, Calverton, MD United States
For more than two decades, U.S. police have been using sobriety checkpoints to enforce drunk-driving laws. In June 1990, the U.S. Supreme Court upheld the use of sobriety checkpoints to detect and deter impaired drivers. Sobriety checkpoints are enforcement activities where police stop drivers to evaluate them for alcohol or other drug impairment. Drivers are asked a few questions designed to allow the officer to observe their general demeanor. Those appearing unimpaired are waved on; those showing signs of impairment are held for further investigation.Research has indicated that well publicized and highly visible sobriety checkpoints can serve as a general deterrent to impaired driving. In fact, studies have confirmed that such activities can substantially reduce alcohol-related crashes by 10 to 15%.Because of the heightened visibility sobriety checkpoints give to DUI/DWI law enforcement, they are particularly valuable as a general deterrent. Publicity about the checkpoint program beforehand and publication of the arrests afterwards further increase the general deterrent effect. In sum, if the public is made aware that police are conducting checkpoints, they tend to be much more cautious about drinking and driving (i.e., they drink less and/or find alternative transportation).This study reports on a demonstration program in Tennessee developed to determine if highly publicized checkpoints conducted weekly throughout the state could have an effect on impaired driving statewide. Four sets of three fully staffed and equipped checkpoints were conducted throughout the state every weekend. On five weekends during the project year, checkpoints with reduced staff—not involving as many officers or as much equipment, but rather served to reinforce the “blitz” concept—were scheduled at each of the 95 counties in the state. The checkpoints were coordinated and conducted primarily by the Tennessee Highway Patrol with support from local law enforcement agencies. Results revealed a 20% reduction in alcohol-related fatal crashes that continued for 21 months after conclusion of the formal program.Promoting sobriety checkpoints is an effective prevention strategy. Checkpoints are conducted in 39 states and Washington, DC. Although the Supreme Court has ruled that checkpoints are constitutional, 11 states do not allow them. Removing these legal barriers should be pursued. Until recently, checkpoints have been conducted at the local level only. Statewide efforts would greatly enhance their effectiveness.
CLUSTERING OF DRUG INVOLVEMENT AMONG HISPANICS IN THE US.
. Fernando Wagner1, Huibo Shao1, Dorothy C. Browne1, 1Morgan State University, Baltimore, MD United States
Introduction. Less acculturated Hispanics seem to have lower rates of drug use (e.g., Burnam et el., 1987; Amaro et al., 1990; Vega et al., 1993; Wagner et al., 1994; Vega et al., 1998; De la Rosa et al., 2000; Epstein et al., 2001; Vega et al., 2002, among many others). Community-level influences have rarely been studied in a sample representative of Hispanics in the US. The present study offers new insights on acculturation and neighborhood among Hispanic youth by: (a) exploring if drug use differences can be traced back to reduced occurrence of exposure to drug use opportunities among those less acculturated; (b) estimating the degree of clustering of opportunities and use by neighborhood among Hispanics; and, (c) assessing the strength of the association, once neighborhood characteristics are held constant.
Methods. The data are from the 1991-94 National Household Survey on Drug Abuse (NHSDA), a series of annual independent surveys that were representative of the whole US population, with over sampling of youth, African Americans, and Hispanics. This study focuses upon Hispanic youth 12-25 years at assessment. The NSHDA has only one measure to approximate acculturation: language used at the time of assessment. Stratified Cox discrete-time survival analyses were used to estimate the risk of drug use, by ethnic subgroups and acculturation. In addition, Alternating Logistic Regression models are used to estimate the risk of drug use by ethnic subgroups and acculturation, simultaneously estimating and accounting for the degree of clustering of drug involvement by neighborhood.
Results. Hispanic-Americans preferring to complete the survey in English were more likely to be offered opportunities to use marijuana, than those preferring to take the survey in Spanish, controlling for sex, age, NHSDA year, neighborhood characteristics, and ethnic subgroups (aRR=2.0; 95% CI, 1.7, 2.3). Once the initial opportunity occurred, those Hispanics more acculturated were more likely to actually use marijuana, compared to those who were less acculturated (aRR= 2.7; 95% CI, 1.9, 3.7). Similar estimates were obtained when the level of clustering by neighborhood (PWOR= 1.9; 95%CI, 1.4, 2.5) was taken into account.
Comment. Hispanics who are more acculturated into the mainstream American society are more likely to have opportunities to use drugs, and are more likely to actually transit into drug use if presented with a chance to try, even when accounting for clustering of marijuana use by neighborhood with different analytical tools.
Acknowledgements. NIDA Grants DA 17796-01, DA12390, and K05DA015799.
EFFECT OF TEENAGE DRUG USE ON LATER SOCIAL ROLE FUNCTIONING. Kerry Green1, Margaret Ensminger1, 1Johns Hopkins University, Baltimore, MD United States
This study examines the effects of adolescent drug and alcohol use on young adult social role functioning. Gender differences in consequences are given particular attention. Using a life course conceptual framework, the study focuses on effects of beer or wine, hard alcohol, and marijuana on marriage and family formation and education and employment. Based on Precocious Development Theory, it is hypothesized that early drug and alcohol use leads to premature involvement in adult social roles. The study analyzes longitudinal data collected primarily through interviews with a cohort of first graders in the Woodlawn community of Chicago who were followed from age 6 to age 32. Participants are all African American (N=1,242, 51% female). Logistic regression and structural equation modeling results show that adolescent drug use has long-term effects for both males and females. After controlling early demographic and behavioral variables, adolescent marijuana was associated with teen parenthood, premarital pregnancy, divorce, and poor educational attainment. For males only, adolescent marijuana use was also related to poorer job status. Consequences of adolescent beer or wine use were similar to those seen with marijuana use but to a lesser extent. Additional gender differences were found for adolescent beer and wine use. Few effects were seen for adolescent hard alcohol use. Adulthood drug use and educational attainment only partially explained these findings. Health status did not mediate the effects. Findings suggest that effects of early drug use carryover into adulthood. Results provide insight for the structuring of prevention programs. Specifically, programs need to encompass a broad scope. The association of early drug use with teen parenting and school dropout suggests that programs need to be integrated and each risk can benefit from addressing the other risks simultaneously. For example, professionals treating adolescents with substance use problems need to be aware of an increase risk of teenage pregnancy, while doctors treating teenage mothers need to be sensitized to possible co-occurring drug use. Also, school counselors who attempt to prevent drop out need to take the role of drugs into consideration. Results also suggest that programs need to focus their efforts on the escalation of use, as this study found that heavier use increases the risk of a poor outcome. Finally, intervention programs need to take the specific drug into consideration as results show that outcomes varied by drug type.
PREDICTED AGGRESSION IN HYPOTHETICAL PROVOKING SITUATIONS: THE INTERACTION BETWEEN PERCEIVED ALCOHOL INTOXICATION, SOCIAL FACTORS, AND INDIVIDUAL DIFFERENCES. Paul Tremblay1, Kathryn Graham1, Ljiljana Mihic1, Jennifer Jelley1, 1Centre for Addiction and Mental Health, London, Ontario Canada
Although a considerable amount of empirical evidence exists for the association between alcohol and aggression, little attention has been paid to the motivation for reactive aggression and to the interaction between alcohol, the drinking environment, and individual characteristics such as gender and trait aggressiveness. Random samples of 1500 full-time undergraduate students at six Canadian universities (total N = 9000) were invited by email to participate in a web-based aggression study. As part of the study, participants (N = 2647) read three vignettes describing conflict situations: (1) a person cuts in front of another waiting in line to get into a bar/club, (2) a jealousy situation involving intimate partners, and (3) a racial slur in a university residence. Each vignette included manipulations of intoxication of the hypothetical person reacting to the provocation (i.e., sober vs. intoxicated) and social environmental variables such as time waiting in line (10 min vs. 60 min) in the first vignette, location of conflict (house party vs. formal event) in the second vignette, and presence of a residence assistant in the third vignette. Participants were randomly assigned to vignette conditions. The experimental design also involved a manipulation consisting of self vs. other evaluations. Half of the participants were asked to imagine themselves as the hypothetical individual reacting to the provocation. The other half were asked to predict how another individual of the same sex would respond. This study focuses on the self-evaluations. Dependent measures consisted of 11-point scale responses to the vignettes assessing affect, motivation to retaliate, and likelihood of aggressive retaliation. This study presents the structural equation models with sub-group analyses supporting the hypothesis that in hypothetical provoking social situations, the perceived likelihood of retaliatory verbal and physical aggression is influenced by perceived alcohol intoxication of the hypothetical person responding to provocation, gender and trait aggression of the respondent, and motivational variables such as anger and perceived importance to retaliate. In addition, a number of social-environmental factors were found to influence the perceived motivation to retaliate. For example, when a person is waiting in line to get into a bar and an instigator cuts in front of that person, factors such as amount of time waiting in line, sex of the instigator, and whether the instigator was intoxicated all contributed to the perceived motivation of the offended individual to retaliate. Implications for prevention initiatives related to university student activities are discussed.