Using nationally representative data, our analyses illustrated that childhood maltreatment is not uncommon in the United States: 31.9% of respondents reported some instance of physical abuse, sexual abuse, or neglect. The most prevalent maltreatment and the least prevalent maltreatment were neglect and sexual abuse, respectively. The prevalence of physical abuse in the Add Health sample (14.3%) was similar to several previous estimates (Bensley et al., 2000
; Diaz et al., 2002
; Goodwin et al., 2003
; Goodwin & Stein, 2004
), in which the prevalence of physical abuse ranged from 10.6% to 15.8%. The Add Health estimate of sexual abuse (4.3%) was also comparable to the results of a study by Finkelhor and Dzibua-Leatherman (1994)
which showed that 3.3% of a nationally representative sample (two thousand children aged 10 – 16 years) reported contact sexual abuse by a family member. There have been relatively few studies that have examined the prevalence of child neglect using a nationally representative sample. The Add Health estimate of childhood neglect (20.7%) was much higher than that of other studies (e.g., 2.89% in Goodwin & Stein, 2004
). The discrepancy may have resulted from the fact that other studies such as Goodwin & Stein's study (2004)
measured a relatively severe form of neglect (i.e., “seriously neglected”) while the Add Health neglect measure was more inclusive (e.g., supervisory neglect, physical neglect). In fact, among those experiencing neglect (n=938), 77% reported supervisory neglect only, 12% reported physical neglect only, and 11% reported both. Hence, our findings regarding neglect may actually reflect the fact that neglect was measured predominantly by supervisory neglect. Moreover, it is important to note that the measure of supervisory neglect is similar to the construct of parental monitoring used in other studies (DeVore & Ginsberg, 2005
; Fisher et al., 2003
Our results demonstrated that physical abuse was related to illicit drug use and drug-related problems in young adulthood. Compared to those not abused, participants with a physical abuse history in childhood were 37% more likely to use illicit drugs in the past 30 days. This finding is consistent with the results in a recent retrospective study by Thompson and colleagues using data from National Violence against Women Survey. They found that physical abuse in childhood was significantly related to past month illicit drug use in adulthood and reported no significant childhood physical abuse by gender interaction (Thompson et al., 2004
). Although the results from a longitudinal community study by Silverman and colleagues suggested physical abuse was related to drug abuse/dependence at age 21 for men, but not for women (Silverman et al., 1996
), it should be noted that in our study there was no significant physical abuse by gender interaction therefore results by gender were not reported.
Although previous studies have found that childhood sexual abuse is associated with lifetime illicit drug use and drug abuse/dependence (Wilsnack et al., 1997
; Burnam et al., 1988
), we did not observe a relationship between childhood sexual abuse and illicit drug use and drug-related problems in young adulthood. This finding was consistent with some previous prospective and community-based studies. For example, the prospective findings of a study by Widom and colleagues (1999)
suggested that sexual abuse was not related to current drug abuse/dependence. The results from a longitudinal community study by Silverman and colleagues (1996)
also indicated that, among women, sexual abuse was not associated with drug abuse-dependence at age 21.
There have been relatively few studies that have examined the relationship of childhood neglect and illicit drug use in young adulthood. In our study, child neglect was indirectly related to adulthood illicit drug use for females, but not for males. This result is in agreement with the findings from two studies by Widom and colleagues (Widom & White, 1997
; Widom et al., 2006
), which showed that for females, but not for males, child abuse/neglect was related to substance abuse/dependence and illicit drug use. Similarly, the Adverse Childhood Experiences study also showed that neglect (emotional and physical neglect) was independently related to lifetime use of illicit drugs (Dube et al., 2003
). However, a study by Kang and colleagues reported that neglect was not related to drug-related HIV risk behaviors, including heroin use and injection drug use (Kang et al., 2002
). Comparisons of findings among these studies can be challenging due to different definitions used for neglect and illicit drug use outcomes.
The link between childhood maltreatment and subsequent illicit drug use in young adulthood and why the relationships differ by gender are still poorly understood. Risky behaviors, including substance use may serve as a coping mechanism or a self-medication strategy by which victims deal with the feelings of pain, anxiety, anger, depression and low self-esteem associated with childhood maltreatment and these risky behaviors can be used chronically if found beneficial (Felitti et al., 1998
; Lebling 1986
; Triffleman et al., 1995
; Miller et al., 1987
). There is evidence that females are more vulnerable to childhood maltreatment, and therefore exhibit a higher frequency of maladaptive coping strategies. Our finding about the effect of neglect is consistent in this regard where females who experience childhood neglect are more likely to engage in illicit drug use during the past year prior the Wave 3 survey. However, our study presents an even more complicated scenario whereby males and females respond similarly to physical abuse. Previous research also show that female victims are more inclined to experience internalizing symptoms as a response to childhood maltreatment (e.g., depression, alcohol and drug abuse) whereas males victims are more inclined to experience externalizing symptoms (e.g., aggressive behaviors, violence, antisocial personality disorder) (Horwitz et al., 1996
; Luntz & Widom, 1994
; Chandy et al., 1999). Our finding regarding the effect of neglect is also in agreement with the hypothesis that females who experience childhood neglect are more likely to engage in illicit drug use during the past year prior the Wave 3 survey. A previous study hypothesized that gender differences in the prevalence of adverse childhood experiences, other than maltreatment, may cause the effects of maltreatment to differ by gender (MacMillan et al., 2001
). In our study, however, important family and peer contextual variables were statistically controlled for both males and females in the analyses, thus our findings cannot be interpreted by differences in other adverse childhood experiences. Research also suggests that the difference in the severity of maltreatment between males and females may contribute to the differences of childhood maltreatment effects (Cutler & Nolen-Hoeksema, 1991
). In our study, there were no significant differences by gender in the percentage of participants reporting physical abuse and neglect items that happened either 5 to 10 times or more than 10 times between male and female victims. Thus, it is unlikely that the observed differences in the effects of neglect on illicit drug use were due to difference in the severity of maltreatment between males and females. However, without a more detailed quantitatively-measured maltreatment severity history, this hypothesis could not be tested. In summary, these inconsistent findings underscore the need for further research to determine the ways in which males and females differ in the consequences of childhood maltreatment.
Our study attempted to address the mechanisms that might mediate the relationship between childhood maltreatment and subsequent illicit drug use in young adulthood. In our study, illicit drug use in adolescence partially mediated the effects of childhood physical abuse on past year illicit drug use and drug-related problems, and fully mediated the effect of childhood neglect on past year illicit drug use for females. Although our study did not examine specifically which contextual or individual factors mediated the relationship between childhood maltreatment and illicit drug use in young adulthood, our results reinforced the importance of early illicit drug use intervention in adolescence, which can interrupt a critical point in the developmental path of illicit drug use in young adulthood.
Our study is among only a few studies at the national level that have examined the relationship of childhood maltreatment to subsequent illicit drug use and drug-related problems in adulthood. The present study has several strengths and addresses some of the gaps in previous studies. First, we employed a large nationally representative sample of young adults. Second, the inclusion of a large sample of men and women allowed us to investigate whether the association between maltreatment in childhood and illicit drug use and drug related problems in adulthood varied by gender. Third, in addition to childhood maltreatment experiences, other (prospectively) measured family and peer contexts were statistically adjusted in our study to control for these potential sources of confounding.
This study has several limitations. First, although the Add Health survey prospectively collected data on family and peer contexts during childhood, the measures of childhood maltreatment relied on retrospective recall in the subsequent Wave 3 survey. However, several concerns accompany the prospective collection of childhood maltreatment. First, it may seem offensive to collect information of this sort from parents or care-givers and second, the discovery of child abuse during data collection typically signals a concerted effort to protect the child, which, in turn, will contaminate the effects of the maltreatment being studied (Fergusson et al., 1996
). The Add Health study prospectively collected details on family and peer contexts during the Wave 1 survey, and retrospectively collected childhood maltreatment history when the participants reached young adulthood. This method is consistent with the one used in Fergusson's study (Fergusson et al., 1996
). Nevertheless, responses may have been subjected to recall bias. Recall bias may have been alleviated somewhat by use of an audio-CASI technique for collecting answers to sensitive questions concerning maltreatment experiences in the Add Health survey (Turner et al., 1998
). Furthermore, studies have shown that childhood maltreatment experiences are generally underreported (Williams, 1995
). This random misclassification would have probably biased our results toward the null. Thus our estimates of the association between childhood maltreatment experiences and adult illicit drug use were probably underestimated. Second, to our knowledge, there is no consensus on measures of childhood maltreatment, and the measures used in the Add Health study may not be comparable to the ones used in other studies. The childhood maltreatment measures used in the Add Health study also have some limitations. For example, the two items assessing neglect were based on respondent's personal opinion or perception and not on observation. The item on physical abuse bundled “slap, hit, and kick” into one question. This can be problematic as it encompasses a wide range of severity of behaviors from physical discipline to life-threatening assaults and, therefore, may mask significant differences between these subgroups. Also, childhood physical abuse and sexual abuse in Add Health study were measured by a single items and thus the reliability and validity of these constructs may be significantly limited. Another limitation of this study is that it only examined the long-term effects of childhood maltreatment that were committed by parents or primary caretakers. Maltreatment by nonfamily members was not collected as part of the Add Health study, and consequently was not evaluated in this study. Finally, the Add Health Wave 3 survey did not measure drug abuse/dependence based on the 4th
edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). Thus, we did not investigate the relationship between childhood maltreatment experiences and drug abuse/dependence in young adulthood.
We found direct effects of physical abuse on subsequent illicit drug use and drug-related problems in young adults even when a host of contextual factors and adolescent drug use were controlled for. Our study also found that neglect was indirectly associated with a higher likelihood of illicit drug use for females. Our findings indicate the importance of considering childhood maltreatment when developing drug abuse preventive interventions for adolescents.
- Childhood physical abuse was associated with a higher likelihood of past 30-day illicit drug use, past year illicit drug use and drug-related problems in young adulthood.
- Childhood neglect was associated with a higher likelihood of past year illicit drug use in young adulthood for females.
- Illicit drug use in adolescence mediated the effects of childhood physical abuse and neglect on illicit drug use and drug-related problems in young adulthood.