Trauma and Intraveneous Drug Use among Pregnant Alcohol/Other Drug Abusing Women: Factors in Predicting Child Abuse Potential

ABSTRACT

Because there is limited empirical literature addressing child abuse potential (CAP) in pregnant substance users, our study investigated associations between trauma, route of drug administration (IV use), and CAP in pregnant substance abusers. We explored this association using CAP as both a continuous and dichotomous measure, and with both narrow-band (i.e., PTSD) and broad-band (i.e., anxious arousal, depression, and anger/irritability) trauma symptoms. Participants were 44 ethnically diverse substance using pregnant women (61.4% Hispanic, 15.3% Native American, and 15.3% Caucasian women) who had completed a NIDA clinical trial. We found that women in our sample reported high CAP, with 43% reaching the conservative CAP cutoff, and the majority of women (56%) meeting the more liberal CAP cutoff. Trauma symptoms, both narrow- and broad-band, as well as IV status, accounted for significant and unique CAP variance. Specifically, women with greater trauma symptomatology and women who were IV users reported significantly greater CAP. To a lesser extent, trauma exposure also accounted for significant CAP variance. Exploratory analyses revealed that “problems with others” drove the CAP difference between the IV status groups.

Our results suggest that both trauma and IV use are significant predictors of child abuse potential. IV use was more highly associated with broad-band compared with narrow-band trauma symptoms, suggesting broader trauma spectrum problems in the form of affect dysregulation in this subsample. Overall, our findings were largely concordant when CAP was used as a continuous versus categorical variable. Regarding the level of trauma symptom specificity required to represent trauma in predicting CAP, our results suggest that narrow-band and broad-band trauma symptoms may be relatively equivalent, with clinical considerations determining which measure is employed in assessing trauma in this population. In addition, IV status is likely a proxy variable for “high-risk” status, and high-risk women, regardless of substance use, have been shown to evidence high levels of child abuse potential.

The results from this community sample of low SES, ethnically diverse pregnant women suggest that because of the high levels of child abuse potential among this group, preventative interventions targeted toward child abuse potential are warranted for these women. Although there has been a burgeoning of substance abuse intervention services for this population, they have not addressed child abuse potential specifically (e.g., in the form of targeting parental expectations or discipline strategies), nor have they systematically evaluated outcomes. Secondly, the high level of trauma exposure and trauma symptoms in this group suggests that trauma-specific interventions are warranted. Although our study was unable to address questions of causality, and the relationship between trauma symptoms and IV use is likely bidirectional, identifying individual characteristics that predict IV use as a means of coping with trauma would be an important area for future research. Furthermore, to the extent that IV-using women report greater difficulties with others, this may exacerbate child abuse potential, as both substance use and low social support have been identified as child abuse potential risk factors in this population.

KEYWORDS:

• Pregnant substance users
• trauma
• IV use
• child abuse potential

Notes

This study was supported by the National Institute of Drug Abuse (NIDA) Clinical Trials Network (CTN), Southwest Node grant #U10DA1533; and was conducted as an ancillary study to NIDA Clinical Trials Network Study 0013, Motivational Enhancement Therapy to Improve Treatment Utilization and Outcome in Pregnant Substance Users, Theresa Winhusen, Lead Investigator.

¹Chi square test was nonsignificant. N for this mean difference test was 41.

²ASI composite scores represent the mean severity rating on a 1--8 scale.

Effect sizes were calculated for small sample bias (Hedges, 1988).

Effect sizes are considered to be small (.2), medium (.5), or large (.8) ( Cohen, 1988 )

∗p < 0.05

∗∗p < 0.01

∗∗∗p < 0.001

∗p < 0.05

∗∗p < 0.01

∗∗∗p < 0.001