INTRODUCTION
Many women, including those who are pregnant, have been victims of intimate partner violence (1-6). Each year, approximately 1.5 million US women are physically or sexually assaulted by their intimate partners (3,6), with such victimization being most likely when women are of reproductive age (1,6). Prenatal care based studies report that from 4% to 26% of patients are violence victims before pregnancy, whereas 1% to 17% are violence victims during pregnancy (7-27). A review of studies that examined the prevalence of violence during pregnancy reported that most of the prevalence estimates ranged between 4% and 8% (28). Community-based studies also have found high rates of violent victimization of women before and during their pregnancies. For example, a statewide survey of a representative sample of postpartum North Carolina women found that 7% had been physically abused during the year before pregnancy, whereas 6% had been physically abused during pregnancy (29,30).
Violence at any time in a woman's life, regardless of whether or not she is pregnant, can result in a myriad of health problems, including violence-related injuries, physical health problems, mental health problems, and substance use problems (21,31-34). Although it is clear that intimate partner violence is the direct cause of some of these types of health problems (e.g., a hit may cause an injury such as a black eye), the relationship between intimate partner violence and other types of health problems is less straightforward. For example, there is not one well-accepted, empirically documented explanation concerning why female domestic violence victims may abuse substances more than other women; however, several ideas on the topic have been put forth in both the clinical and scientific literature (35). One of the most popular of these explanations is the self-medication hypothesis, which contends that violence victims turn to substances as a means of coping with the emotional and physical pain caused by violence. In this perspective, violence is viewed as the cause of the women's substance abuse. Another explanation that has been used to understand potential links between intimate partner violence victimization and women's substance abuse contends that substance abusing women are at elevated risk for domestic violence since their substance use behaviors may be viewed by their partners (and other members of society) as being inappropriate, in particular, "obnoxious" or "unfeminine" (36,37), views that may lead the partner to physically "punish" the woman for her "unladylike" behavior. Thus this perspective views the women's substance abuse as triggering violent episodes.
Although substance abuse at any time during a woman's life is problematic because of the devastating negative health effects it may have, substance abuse during pregnancy is of special concern for a number of reasons. For example, substance abuse during pregnancy has been linked to poor pregnancy and birth outcomes (38-40). Moreover, after the birth of the infant, substance abuse by the mother may negatively impact on her ability to care for her infant, a situation that may ultimately result in impairment of the infant's development (41).
Even though both substance abuse and violence that occur during pregnancy may adversely affect the well-being of the mother and later the infant, only a limited amount of research has simultaneously examined violence and substance use in the lives of pregnant women. Several prenatal care based studies have found that women who have been victimized either before or during pregnancy are more likely than other women to drink alcohol or use illicit drugs before and/or during pregnancy (7-10,16,21,26,32). However, most of this past research has focused on only one form of violence, most typically physical assault. Thus less is known about how other important forms of intimate partner violence, such as psychological aggression and sexual coercion, may be associated with women's use of substances before and during pregnancy. Furthermore, additional research is needed to examine whether, among the women who use substances before or during pregnancy, various types of intimate partner violence are associated with the women having symptoms of substance abuse disorders (e.g., having delirium tremens after heavy drinking, having "blackouts" or "flashbacks" as a result of drug use), symptoms that may impair the women" s functioning and limit their ability to care for themselves and their newborn infants.
This article extends our knowledge in these areas by examining associations between women's experiences of various forms of intimate partner violence (including psychological aggression, physical assault, and sexual coercion) and substance use (including alcohol and illicit drug use), both before and during pregnancy. Studying a convenience sample of 85 pregnant women from North Carolina, this study asks whether, during the year before pregnancy and during pregnancy:
* Were the women who experienced various types of intimate partner violence more likely to use alcohol or illicit drugs compared with women who did not experience such violence?
* Among the women who drank alcohol, did those who experienced various types of partner violence drink more frequently and have a greater number of alcohol disorder symptoms compared with women who did not experience such violence?
* Among the women who used illicit drugs, did those who experienced various types of partner violence use drugs more frequently and have a greater number of drug disorder symptoms compared with women who did not experience such violence?
METHODS
Sample
Eighty-five research participants recruited from North Carolina prenatal care clinics that predominately serve low-income women were studied. After obtaining the women's clinical histories (including clinically screening for physical violence) and providing clinical care, the prenatal clinicians described the research study to eligible study participants and invited them to take part in the project. Patients were eligible to enter the study if they were at least 18 years of age, spoke English, and began prenatal care some time before their sixth month of pregnancy. In addition, because we hoped to have somewhat similar numbers of women who had experienced partner violence during pregnancy and those who had not experienced such violence during pregnancy, the clinicians were asked to recruit a nonvictimized woman for study after they had recruited each violence victim for study.
Measures
A structured research interview was administered to the study participants by well-trained female research staff. This assessment occurred when the women were approximately 6 to 7 months pregnant. Study interviews were conducted in private rooms in the health care clinics to help assure confidentiality. Great care was taken to establish rapport with the study participants prior to interview administration. The study was described to the women, and it was stressed that truthful answers were needed to potentially sensitive questions in order to gain accurate insights concerning women's health. An informed consent form was administered to the participants to assure them that their responses would be treated confidentially and that their participation (or nonparticipation) in the study would not affect their or their family's health care. All respondents were provided with a brochure describing several types of health-related services that they could access free of charge (including domestic violence services and substance abuse services) and they were given a modest monetary "thank you gift" for their involvement in the study.
The study interview included a wide range of topics. Information was collected concerning the women's experiences of intimate partner violence, their use of substances, and their sociodemographic characteristics.
