Growing awareness of domestic violence during pregnancy has led organizations such as The American College of Obstetricians and Gynecologists (ACOG), the American Medical Association, and various public agencies to emphasize that verbal or physical abuse of women is a significant public health problem.1–3 For example, Surgeon General C. Everett Koop declared domestic violence to be a public health problem in 1985.4 Beginning in 1992, the Joint Commission on Accreditation of Healthcare Organizations emphasized identification, evaluation, and care of adult victims of domestic violence. Also in 1992, the Centers for Disease Control established a national center for prevention of domestic violence against women. About the same time, the American Medical Association published diagnostic and treatment guidelines and launched a national campaign against domestic violence.5
Domestic violence against women covers a broad spectrum of behaviors, including actual or threatened physical, sexual, or psychological abuse between family members or intimate partners. Although few studies have examined the potential association between violence during pregnancy and subsequent adverse outcomes, there is the belief that violence during pregnancy increases obstetric complications.6 Our purpose was to study domestic violence during pregnancy by using a standardized survey methodology to assess whether there are any associated adverse pregnancy effects.
MATERIALS AND METHODS
Between December 6, 2000, and March 31, 2002, women presenting to the Parkland Hospital Labor and Delivery Unit were approached to be interviewed concerning domestic violence during pregnancy. Parkland Hospital is a tax-supported institution serving the medically indigent women of Dallas County, Texas. The study protocol was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center. Written consent was waived on the grounds that the research employed a structured interview during which verbal consent was obtained.
The survey questionnaire used in this study was previously validated for the identification of female victims of domestic violence.7–8 The questionnaire, known by the acronym HITS, is composed of 4 questions that prompt a woman to indicate whether her partner or family member physically Hurt her, Insulted her or talked down to her, Threatened her with harm, or Screamed or cursed at her during this pregnancy. The HITS survey has been shown to have good internal reliability and concurrent validity when compared with other questionnaires.7 All women answering “yes” to any of the 4 questions were offered the phone number of a violence intervention center. This center is staffed 24 hours a day and is equipped to provide a full range of medical services and referrals to legal and community organizations. The phone number was written on a small sheet of paper that could easily fit into a woman’s shoe, with no identifying name or organization written on it to protect the woman from retaliation in case the information should be found by her abuser.
The interviews were conducted face-to-face by 1 of 5 female, bilingual, research personnel not involved in the patient’s care. These personnel were generally on duty 24 hours a day, 7 days a week. All interviews were conducted orally in an identical fashion, and patients were guaranteed confidentiality. Survey responses were simultaneously recorded by the interviewer using a preprinted form. Each survey was initiated with the interviewer stating “I would like to ask you a few questions about physical or emotional trauma because we know that these are common and affect women’s health.” The interview was not conducted if the woman was unable to be interviewed privately or if the woman declined to be interviewed. Women diagnosed with a stillborn were not interviewed in an effort to avoid magnifying the patients’ emotional duress. Only women thought to have singleton gestations of 24 weeks or greater were interviewed.
The HITS survey responses were entered into a computer and linked to delivery data using a preexisting electronic obstetrics database. These data sets contain obstetric and neonatal outcomes for all women delivering infants at Parkland Hospital. Nurses attending each delivery completed an obstetric data sheet, and research nurses prospectively reviewed all mother and infant charts to ensure completeness and accuracy of the data before electronic storage.
An interim analysis was done after 2,700 women had been interviewed to assess the incidence of physical abuse in our study population and project sample size, using low birth weight (≤ 2,500 g) as the outcome of interest. The incidence of physical abuse during pregnancy was 0.4%, and the incidence of low birth weight was 5%. To achieve a power of 80% for a test of 2-sided significance of 0.05 to test a 3-fold increase in low birth weight in women reporting physical abuse, a sample size of 16,000 women was projected.
Statistical analyses included χ2, Student t test, and Wilcoxon rank-sum test. P values less than .05 were considered significant. Analysis was performed with SAS 9.1 (SAS Institute Inc, Cary, NC).
A total of 21,483 women with liveborn singletons were delivered during the study period, and 16,041 (75%) of these were approached to be interviewed (Fig. 1). Several factors accounted for the women who were not approached. These factors included imminent delivery, maternal-fetal conditions requiring immediate obstetric evaluation, inability to ensure privacy, or unavailability of research personnel. Shown in Table 1 is the distribution of positive responses (“yes” answer) to each of the 4 survey questions. Some women gave positive responses to more than one question.
As shown in Figure 1, a total of 949 (6%) women responded affirmatively to one or more of the survey questions, and another 94 (.6%) declined to be interviewed. We combined women who answered “yes” to any of questions 1, 2, and 3 (Table 1) and designated this group “verbal abuse” (Fig. 1). Pregnancy demographics and outcomes were then compared between 4 study groups: no abuse, verbal abuse, physical abuse, and women who declined to answer the survey. Shown in Table 2 are maternal demographics according to these study groups. African-American women and younger women reported a significantly higher rate of physical abuse. Those women who declined to answer the survey were significantly less likely to have prenatal care. Approximately 30% of the women surveyed identified third-party health insurance (typically government-sponsored) at delivery, and the remainder were uninsured.
Shown in Table 3 are selected pregnancy complications in relation to domestic violence. Women who declined to answer the survey experienced significantly increased rates of placental abruption and preterm delivery. Women who declined to participate in the survey also had significantly increased rates of neonatal intensive care admission and low birth weight infants when compared with women in the no-abuse group (Table 4). The incidence of low birth weight infants was also significantly increased in women who reported verbal abuse compared with those reporting no abuse (7.6% versus 5.1%, respectively, P = .002). Reported physical abuse was associated with an increased neonatal death rate (1.5% versus 0.2%, physical-abuse group versus no-abuse group, respectively, P = .004).
There were 2 major findings in this observational study of domestic violence in more than 16,000 pregnant women encountered when presenting to the hospital for delivery. First, when compared with women denying domestic violence, women reporting verbal abuse had an increased rate of low birth weight infants, and neonatal deaths were significantly increased in women experiencing physical abuse. Second, women who declined to participate in our survey were found to have significantly increased rates of a variety of pregnancy complications that adversely affected their infants’ outcomes. These included predominately preterm birth and attendant neonatal complications of prematurity. Low birth weight, our primary study outcome, was significantly increased almost 3-fold (12.8% versus 5.1%, declined-survey group versus no-abuse group, respectively, P < .001) in women declining to be interviewed, compared with women denying domestic violence. Considering that fetal growth restriction, defined as either less than the 10th percentile or less than the 3rd percentile, was not related to the domestic violence survey results, we believe that the adverse effects of domestic violence, viz-a-viz low birth weight, occur via preterm birth as opposed to compromised fetal growth.
In a review of the literature between 1963 and 1995, Gazmararian and colleagues9 found 13 studies that met specific criteria with regard to the incidence of violence during pregnancy. The number of women surveyed in each study ranged from 290 to 12,612, and the incidence of violence during pregnancy ranged from 1% to 20%, with the majority reporting rates of 4–8%. Our finding of a 6% incidence (1% physical abuse and 5% verbal abuse) is consistent with these other reports. We found that the majority of pregnant women we approached for domestic violence screening were accepting of direct questioning because only 0.6% declined to be interviewed. This openness to questioning has also been reported by others.10
The outcome of interest in the majority of domestic violence studies during pregnancy has been low birth weight. For example, a meta-analysis of 8 studies published between 1990 and 1999, with sample sizes ranging from 178 to 1,897 women from North America and Europe, showed a weak but significant association between abuse during pregnancy and low birth weight (odds ratio 1.4, 95% confidence interval 1.1–1.8).6 Seven of the 8 studies in this meta-analysis showed no significant association between domestic violence and low birth weight. Altarac and Strobino,11 in a subsequent study involving over 800 women, also found no association between physical abuse during pregnancy and low birth weight. The Pregnancy Risk Assessment Monitoring System of the Centers for Disease Control and Prevention found that, in the 17 states where it investigated the prevalence of physical abuse during pregnancy, such domestic violence was unrelated to low birth weight.12 Therefore, our findings of no increase in low birth weight in those women reporting physical abuse is consistent with the preponderance of other published experiences.
We found that women who declined to be surveyed regarding domestic violence were most at risk for adverse pregnancy outcome. We can only speculate that those women who declined to be interviewed did so because they were fearful of retaliation. These women less frequently obtained prenatal care, a fact that has been reported to be an indirect surrogate marker for victimization and isolation during pregnancy.13 We conclude from this study that women most at risk, in terms of adverse pregnancy effects, from domestic violence are those who do not (cannot) respond to questions about their circumstances. We have not developed a strategy about how to deal with such women other than to appraise them of support avenues available via telephone if they choose this. Clearly, future efforts to study and prevent domestic violence during pregnancy should consider that the women who remain silent when questioned about the subject may, in fact, be speaking the loudest.
1. McAfee RE. Physicians and domestic violence can we make a difference? JAMA 1995;273:1790–1.
2. American College of Obstetricians and Gynecologists. Guidelines for women’s health care. 2nd ed. Washington, DC: ACOG; 2002.
3. Petersen R, Gazmararian JA, Spitz AM, Rowley DL, Goodwin MM, Saltzman LE, et al. Violence and adverse pregnancy outcomes: a review of the literature and directions for future research. Am J Prev Med 1997;13:366–73.
4. Titus K. When physicians ask, women tell about domestic abuse and violence. JAMA 1996;275:1863–5.
5. Salber PR, Taliaferro E. An introduction to domestic violence. In: Brown Z, editor. The physician’s guide to domestic violence. Volcano (CA): Volcano Press; 1995. p. 6.
6. Murphy CC, Schei B, Myhr TL, Mont JD. Abuse: a risk factor for low birth weight? A systematic review and meta-analysis. CMAJ 2001;164:1567–72.
7. Sherin KM, Sinacore JM, Li XQ, Zitter RE, Shakil A. HITS: A short domestic violence screening tool for use in a family practice setting. Fam Med 1998;30:508–12.
8. Bardwell J, Sherin K, Sinacore JM, Zitter R, Shakil A. Screening for domestic violence in family medicine: development of the verbal HITS scale. J Advocate Health Care 1999;1:5–7.
9. Gazmararian JA, Lazorick S, Spitz AM, Ballard TJ, Saltzman LE, Marks JS. Prevalence of violence against pregnant women. JAMA 1996;275:1915–20.
10. Johnson JK, Haider F, Ellis K, Hay DM, Lindow SW. The prevalence of domestic violence in pregnant women. BJOG 2003;110:272–5.
11. Altarac M, Strobino D. Abuse during pregnancy and stress because of abuse during pregnancy and birthweight. JAMWA 2002;57:208–14.
12. Beck LF, Morrow B, Lipscomb LE, Johnson CH, Gaffield ME, Rogers M, et al. Prevalence of selected maternal behaviors and experiences, Pregnancy Risk Assessment Monitoring System (PRAMS), 1999. MMWR Surveill Summ 2002;51:1–27.
13. Newberger EH, Barkan SE, Lieberman ES, McCormick MC, Yllo K, Gary LT, et al. Abuse of pregnant women and adverse birth outcome: current knowledge and implications for practice. JAMA 1992;267:2370–2.