FERTILITY AND STERILITY威 VOL. 80, NO. 5, NOVEMBER 2003 Copyright ©2003 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A.
RECURRENT PREGNANCY LOSS
Violence does not influence early pregnancy loss Deborah B. Nelson, Ph.D.,a Jeane Ann Grisso, M.D., M.Sc.,a,b Marshall M. Joffe, M.D., Ph.D.,a Colleen Brensinger, M.S.,a Roberta B. Ness, M.D., M.P.H.,c Katherine McMahon, B.S.,d Leslie Shaw, Ph.D.,e and Elizabeth Datner, M.D.f University of Pennsylvania Health System, University of Pennsylvania, Philadelphia, Pennsylvania Received November 11, 2002; revised and accepted March 21, 2003. Supported by a grant from the National Institute of Child Health and Human Development (R01 HD36918-04). Presented at the Annual Meeting of the American Public Health Association, October 2002, Philadelphia, Pennsylvania. Reprint requests: Deborah B. Nelson, Ph.D., Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, 423 Guardian Drive, 921 Blockley Hall, Philadelphia, Pennsylvania 19104-6021 (FAX: 215-573-2265; E-mail:
[email protected]). a Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics and Epidemiology. b Present address: The Robert Wood Johnson Foundation, Princeton, New Jersey. c University of Pittsburgh, School of Public Health, Philadelphia, Pennsylvania. d School of Nursing, University of Pennsylvania. e Department of Pathology and Laboratory Medicine, University of Pennsylvania Health System. f Department of Emergency Medicine, University of Pennsylvania Health System. 0015-0282/03/$30.00 doi:10.1016/S0015-0282(03) 01180-4
Objective: To examine the relationship between physical violence, controlling behavior, and spontaneous abortion (SAB). Design: Nested case-control study. Setting: Emergency department of a university hospital. Patient(s): One thousand one hundred ninety-nine pregnant women. Intervention(s): Main Outcome Measure(s): Physical violence and controlling behavior. Result(s): Cases experienced a SAB (n ⫽ 392) and controls maintained their pregnancy through 22 weeks (n ⫽ 807). Fifteen percent of women reported violence during the pregnancy, and 49% had reported one or more past episodes of violence. We found no relationship between any measure of physical violence (past, current, or by perpetrator) and the risk of SAB. Conclusion(s): Although physical violence was very prevalent in the study population, exposure to violence did not influence the risk of SAB. (Fertil Steril威 2003;80:1205–11. ©2003 by American Society for Reproductive Medicine.) Key Words: Physical violence, intimate partner violence, abuse, spontaneous abortion, miscarriage
Over 40% of women of childbearing age have experienced at least one episode of violence in their lifetime; the rates of reported violence range from 8% to 40% of women (1–5). In the vast majority of the cases, the violence was perpetrated by an intimate partner, defined as a husband, boyfriend, ex-husband or ex-boyfriend (6, 7). Rates of reported violence during pregnancy have varied extensively, depending on the characteristics of the study population and the definition of violence as physical, emotional, or both (6 –11). O’Campo et al. (8) recruited primarily African American pregnant women at the time of their first prenatal care visit to an inner-city obstetrics clinic; over 65% of the women reported either physical or emotional abuse during their pregnancy: 45% reported experiencing verbal abuse, 11% reported moderate physical violence, and 9% reported severe physical violence. The major perpetrator of the abuse was the male partner, but siblings and parents were
also responsible for a significant proportion. In contrast, Smikle et al. (9) examined the prevalence of physical and sexual violence in an insured, middle-class obstetrics population and found that, although 18% reported ever experiencing violence, only 1% reported violence during pregnancy. Population-based self-report studies have reported a 6% prevalence of physical violence during pregnancy, and a higher rate of physical violence among younger women with unintended pregnancies (12, 13). Pregnancy during adolescence may impart a particularly high risk for violence (14 –16). Covington et al. (14) reported that 16% of adolescents experienced physical violence during pregnancy, with 9% of the violence severe (i.e., hitting, kicking, or stabbing). In comparison, 12% of adult women reported violence during pregnancy, 5% classified as severe. The effect of violence during pregnancy on subsequent pregnancy outcomes has also been 1205
examined, but these studies have primarily focused on the outcomes of low birth weight (LBW) or preterm labor (PTL). An increase in the risk of LBW and preterm labor among women who had experiencing violence during pregnancy has been shown in some but not all studies (17–25). A recent meta-analysis found that women who reported physical, sexual, or emotional abuse during pregnancy were 40% more likely to deliver an LBW baby compared with nonabused women (OR ⫽ 1.4, 95% CI: 1.1–1.8) (21). The relationship between poor pregnancy outcome and violence during pregnancy has been shown for both adult and adolescent pregnant women; in the vast majority of cases, the violence is perpetrated by an intimate partner (26, 27). Campbell et al. (20) used the Index of Spousal Abuse (ISA) to examine the relationship between LBW and controlling behavior by an intimate partner, and found an increased risk of LBW in full-term infants born to pregnant women who were involved in a controlling relationship. Studies have found that women experiencing violence are significantly more likely to report a low level of social support, a high amount of perceived stress, a high number of negative life events, and a higher proportion of partners with a current drinking problem (5, 28). Additionally, women who report experiencing violence are more likely to use cigarettes, alcohol, or both before and during pregnancy, and have an increased rate of sexually transmitted diseases and menstrual problems (3, 4, 29). Late entry into prenatal care, increased substance use, and poor obstetric health, in addition to violence, have been correlated with a significant increase in LBW, preterm delivery, PTL, chorioamnionitis, and delivery by cesarean section (17–19, 21–23). Limited research has been conducted to date on the role of physical violence in spontaneous abortion (SAB). One small study that recruited women from prenatal care clinics found that 14% of women self-reported physical violence during pregnancy but the violence was not related to SAB (30). Another study found that adolescents who had experienced violence were more likely to miscarry than adolescents who had not experienced violence, although this study relied solely on retrospective chart reviews (31). Case reports have documented that direct abdominal trauma can cause abruptio placentae, which, depending on the gestational age of the fetus, can lead to SAB or the early onset of labor (32). In theory, physical or sexual victimization may increase the risk of SAB through [1] elevated psychological stress; [2] isolation and inadequate access to prenatal care and other health care services; [3] behavioral risks such as cigarette smoking, alcohol use, and illegal drug use (in reaction to victimization); and [4] exacerbation of chronic illnesses such as hypertension, diabetes, or asthma. In this study, we describe the prevalence of physical violence experienced by inner-city, generally young, pregnant women and examine the influence of violence on the risk of SAB. 1206 Nelson et al.
Physical violence and spontaneous abortion
MATERIALS AND METHODS Study Design Adolescent girls and women aged 14 to 40 years who presented to the emergency department (ED) of the hospital of the University of Pennsylvania between January 1999 and August 2001, and who resided in selected zip codes, were screened for eligibility. Exclusion criteria included greater than 22 weeks of gestation, history of hysterectomy, or a normal menstrual cycle in the past 28 days. Urine pregnancy tests were conducted on all other women, regardless of the reason for the ED visit. Of the 1,532 eligible women with positive pregnancy tests, 283 women were deemed ineligible because of ectopic, molar, or twin pregnancy (n ⫽ 42); non–English-speaking status (n ⫽ 9); acute mental illness (n ⫽ 3); or birth of a child, therapeutic abortion, or complete SAB within 4 to 14 days before the ED visit (n ⫽ 229). The 4 to 14 day cut-off was used to ensure that exposure information had been obtained before a completed SAB. Among the 1,249 pregnant adolescent and adult women who were eligible for the study, 96% agreed to participate (n ⫽ 1,199). All participants provided written informed consent, and the protocol and consent forms were approved by the institutional review board of the University of Pennsylvania. At study enrollment, each participant completed an extensive, in-person interview administered by a nurse interviewer. The interview contained information regarding medical and reproductive history, sociodemographic factors, current level of social support, living arrangements, and complications of the pregnancy. Gestational age was calculated on the basis of self-reported date of the last menstrual period. Participants were also asked about their past and current history of violence, and information regarding the perpetrator was collected. Hair samples were collected to measure cocaine use, and urine samples were collected to assess recent cocaine and tobacco use. All interviews were conducted in a confidential, private location in the ED. Follow-up telephone interviews were conducted at 16 and 22 weeks of gestation to determine pregnancy status, and medical record reviews were conducted to confirm pregnancy outcome. A case of SAB was defined as any woman experiencing a noninduced pregnancy loss before 22 weeks of gestation. Pregnancy outcome information was obtained through follow-up interviews or medical record review for 93% of eligible participants. The 807 women who remained pregnant through 22 weeks of gestation were classified as controls (67%), and the 392 women who experienced a SAB were classified as cases (33%). Of the cases, 212 experienced a SAB at enrollment and 180 experienced a SAB during the follow-up period. Also, 61% of the SABs that occurred during the follow-up period were confirmed through medical record review. Vol. 80, No. 5, November 2003
Study Measures
Neighborhood Violence
In the baseline questionnaire, violence was broadly identified using the question: “During any argument or fight, did you get pushed, slapped, or hurt in any way?” The next set of questions identified the number of times this had happened, the relationship of the perpetrator to the woman (to classify the perpetrator as an intimate partner, family member, acquaintance, or stranger), where on the body the woman had been hit or hurt, and whether she was stabbed, cut, or shot at with a gun. In addition, we included a final question, “Did anyone else harm you in any other way?” to identify other acts of violence that may not have been included in responses to the previous questions. These questions were used to capture a wide range of degrees of physical violence, both during pregnancy and over the woman’s lifetime, and to allow us to classify the perpetrator(s). To our knowledge, these questions are the most comprehensive assessment of physical violence among pregnant women.
Perceptions of neighborhood violence were assessed through questions concerning the frequency of assaults in the neighborhood, how often the woman had heard gunfire or saw someone carrying a weapon, and how often she had been personally afraid in the previous year.
Episodes of rape were identified through two questions focusing on childhood or adult rape (i.e., “Before you were 16 years old, did anyone ever force you to have sex when you didn’t want to?”). The next questions identified the number of times this had happened and the relationship of the perpetrator to the woman. Intimate Partner Violence/Controlling Behavior Intimate partner violence was defined as being stabbed, shot at with a gun, pushed, slapped, or hurt in any way by the baby’s father, a boyfriend, an ex-boyfriend, a spouse, or an ex-spouse. The violence was measured and categorized according to violence that occurred during the pregnancy and lifetime violence. Controlling behavior by an intimate partner was measured using the Index of Spousal Abuse (ISA). The ISA measures the severity of physical and nonphysical abuse inflicted on a woman by her most recent intimate partner (33). In our project, we included the eight items that focused on nonphysical or controlling behavior, such as “My partner doesn’t want me to go out with my friends.” Participants were asked to indicate the frequency of these behaviors using a five-item Likert scale ranging from 1 (never) to 5 (very frequently). Items were summed with a possible range of scores from 8 to 40, with a higher score representing increasing severity of nonphysical abuse. Because previous items in the questionnaire identified violent acts, we used this modified version of the ISA to measure current controlling behavior experienced by women. Nonintimate Partner Violence (Violence by Others) Violence by others was classified as physical violence perpetrated by family members, acquaintances, or strangers during the pregnancy and over the woman’s lifetime. Again, women were asked if they had been pushed, slapped, or hurt in any way, how many times the violence had occurred, and whether they had been stabbed or shot at with a gun. FERTILITY & STERILITY威
Drug Exposure Urine samples collected at enrollment were tested for recent exposure to tobacco. All tests were performed in batches, with positive and negative controls, by a technician blinded to the woman’s case/control status. Urinary cotinine, the major metabolite of nicotine, was assessed by a microplate enzyme-linked immunosorbent assay (STC Diagnostics, Bethlehem, PA). The results of assay were considered positive if they exceeded the threshold concentration of 500 ng/mL of cotinine. The results for urinary analyses were available for 77% of participants. Cocaine was measured in hair specimens by Psychemedics (Culver City, CA) with a radioimmunoassay (34 –36). A sample of approximately 3.9 cm in length captured cocaine exposure in the prior 3 months, the period of early gestations for our study population (37). Results were considered positive above a threshold of 1.2 ng/mg of cocaine. Cocaine results using hair samples were available for 74% of participants. Social Support Social support was assessed through the self-reported number of telephone contacts per day with a friend or family member in the previous month.
Analytic Method Pregnant controls were compared with the total group of cases who had experienced a SAB, and separately with those cases who experienced a SAB during the follow-up period. Bivariate analysis examined the main study exposures (pregnancy and lifetime intimate partner violence, and pregnancy and lifetime violence perpetrated by others), the major risk factors for SAB (maternal age, gestational age, and prior SAB), social support, and drug exposure. Statistical significance was assessed using t-tests, Wilcoxon rank sum (ANOVA), and chi-square tests. To determine whether an independent relationship between violence and SAB was found after adjusting for confounding factors, multivariate logistic regression models were constructed examining the association between each of the measures of violence and SAB. Violence was classified to capture pregnancy-related and lifetime violence and to assess the impact of intimate partner violence versus violence perpetrated by family members, friends/acquaintances, and strangers. The models adjusted for other risk factors for SAB and potential confounders including maternal age (categorical), previous SAB (categorical), cocaine use detected in hair and cigarette use detected in urine (dichotomous), 1207
education (categorical), and gestational age (continuous). We also included a term to assess the level of social support (i.e., number of telephone contacts (⬍1 and 1⫹ per day), and a measure of neighborhood violence (i.e., number of times a woman felt afraid) (⬍1 or 1⫹ times per month) in the regression models. We used backward elimination with a P⬎.05 to remove variables not significantly associated with SAB.
RESULTS Women enrolled in the study presented to the ED with varying complaints. Three-quarters presented with pregnancy-related complications: 42% for vaginal bleeding, 21% for abdominal pain, and 11% for other pregnancy-related symptoms. The remaining 26% presented with other non–pregnancy-related complaints related to infection, accidents, and injury. Women participating in the study were predominantly under 30 years of age (75%) and African-American (92%), nearly 30% had not completed high school. Almost 75% of participants reported being single, 47% were receiving government assistance, and 66% reported one previous livebirth. Women who had a SAB had a slightly earlier gestational age at enrollment compared with pregnant controls (9 weeks vs. 11 weeks, P⬍.05) and were more likely to present to the ED with pregnancy-related symptoms. In addition, women experiencing a SAB were slightly older (25 vs. 23 years), more educated (71% with a high school education or higher vs. 64%), and had a higher number of previous pregnancy losses (37% vs. 34%). Overall, 15% of women reported experiencing physical violence since their last menstrual period, with 66% of the episodes were perpetrated by a current or past intimate partner. Nearly one-half of the total population (49%) reported one or more past episodes of intimate partner violence over the lifetime and approximately 58% of women reported at least one episode of previous violence by a family member, acquaintance, or stranger. Women experiencing intimate partner violence were almost two times as likely also to experience violence by a family member, acquaintance, or stranger over their lifetime (OR ⫽ 1.9, 95% CI: 1.5–2.4) and significantly more likely to experience controlling behavior by an intimate partner (P⬍.0001). In addition, 15% of women reported one or more episodes of rape and 14% reported sexual abuse as a child. Table 1 outlines the prevalence of violence over the woman’s lifetime and during the pregnancy by the type of perpetrator comparing the case and control groups. As shown, there were no statistically significant differences between cases and controls for any measure of violence. Fifty percent of the cases and 47% of the controls reported experiencing violence by an intimate partner at least once over their lifetime, and approximately 10% of both the cases and controls reported violence by an intimate partner occurring during the pregnancy. In addition, 5% of the cases and 1208 Nelson et al.
Physical violence and spontaneous abortion
TABLE 1 Violence assessments comparing cases of spontaneous abortion and pregnant controls.
Violence by typea Intimate partner violence, lifetimec Yes No Intimate partner violence since LMP Yes No Index of spousal abuse (8–40) (mean ⫾ SD) Violence by others, lifetime Yes No Violence by others since LMP Yes No Neighborhood violenced ⬍ once a month ⱖ once a month Childhood violence Yes No Rape (ever) Yes No
Total cases (n ⫽ 392)
Cases at follow-upb (n ⫽ 180)
Pregnant controls (n ⫽ 807)
50% 50%
57% 43%
47% 53%
9% 91% 14.0 ⫾ 4.7
10% 90% 14.4 ⫾ 4.9
10% 90% 14.7 ⫾ 5.2
58% 42%
58% 42%
58% 42%
5% 95%
4% 96%
6% 94%
91% 9%
89% 11%
92% 8%
30% 70%
32% 68%
29% 71%
14% 86%
16% 84%
14% 86%
Note: LMP ⫽ last menstrual period. a Violence does not include rape. Rape is a separate item. b Cases at follow-up included women experiencing a spontaneous abortion after enrollment. c P⬍.05 chi-square test (cases at follow-up vs. controls). d Neighborhood violence was defined as the number of times respondent felt afraid in the past month. Nelson. Physical violence and spontaneous abortion. Fertil Steril 2003.
6% of the controls reported at least one episode of violence by others during the pregnancy. The prevalence of lifetime physical violence by others (family members, acquaintances, and strangers) was also similar between the cases and controls (58% vs. 57%). The prevalence of neighborhood violence and rape was similar between the two groups, and 30% of both groups reported experiencing one or more episodes of childhood violence (see Table 1). The level of recent controlling behavior by an intimate partner, as measured by the ISA, was similar among the controls compared with the cases (14.7 vs. 14.0). After adjustment for relevant confounding factors, violence was still not related to SAB (OR ⫽ 0.86, 95% CI: 0.49 –1.54); however, early gestational age (OR ⫽ 0.92 per week, 95% CI: 0.88 – 0.96) and history of two or more SABs (OR ⫽ 2.37, 95% CI: 1.21– 4.64) were related to an increased risk of SAB. Maternal age, social support, education, Vol. 80, No. 5, November 2003
TABLE 2
TABLE 3
Multivariate analysis of intimate partner violence and spontaneous abortion among inner-city pregnant women.
Variablea Intimate partner violence since LMP Maternal ageb Gestational age Social support Prior spontaneous abortionc Education leveld Cigarette use Cocaine use Neighborhood violence
Total cases versus controls odds ratio (95% CI)
Cases at followup versus controls odds ratio (95% CI)
0.86 (0.49–1.54)
0.76 (0.35–1.67)
1.04 (0.58–1.89) 0.92 (0.88–0.96) 1.20 (0.80–1.81) 2.37 (1.21–4.64)
0.92 (0.42–2.00) 0.90 (0.85–0.96) 0.93 (0.56–1.55) 2.64 (1.21–5.80)
1.60 (0.96–2.65) 1.26 (0.83–1.90) 0.86 (0.56–1.32) 1.00 (0.53–1.88)
1.99 (1.04–3.84) 1.26 (0.74–2.15) 0.90 (0.51–1.60) 1.26 (0.59–2.68)
a
Variables included in the model were maternal age (categorical), gestational age (continuous), cigarette and cocaine use (dichotomous), prior spontaneous abortion (categorical), social support (dichotomous), education (categorical), neighborhood violence (dichotomous). b Maternal age: The comparison group was age 30 years and older versus under 20 years. c Prior spontaneous abortion: The comparison group was women with a prior pregnancy but no spontaneous abortions versus women reporting two or more spontaneous abortions. d Education: The comparison group was post high school training or higher versus less than high school education. Nelson. Physical violence and spontaneous abortion. Fertil Steril 2003.
Multivariate analysis of violence by others and spontaneous abortion among inner-city pregnant women.
Variablea Violence by others since LMPb Maternal agec Gestational age Social support Prior spontaneous abortiond Education levele Cigarette use Cocaine use Neighborhood violence
Total cases versus controls odds ratio (95% CI)
Cases at followup versus controls odds ratio (95% CI)
1.67 (0.82–3.39)
1.86 (0.74–4.71)
1.10 (0.61–2.00) 0.91 (0.87–0.96) 1.17 (0.78–1.77) 2.36 (1.20–4.62)
0.98 (0.45–2.13) 0.90 (0.84–0.95) 0.91 (0.54–1.52) 2.65 (1.21–5.82)
1.63 (0.98–2.71) 1.21 (0.80–1.84) 0.85 (0.55–1.31) 0.98 (0.52–1.85)
2.09 (1.08–4.02) 1.23 (0.72–2.10) 0.89 (0.51–1.58) 1.22 (0.58–2.59)
a
Variables included in the model were maternal age (categorical), gestational age (continuous), cigarette and cocaine use (dichotomous), prior spontaneous abortion (categorical), social support (dichotomous), education (categorical), neighborhood violence (dichotomous). b Violence was defined as violence perpetrated by others since respondent’s last menstrual period, includes family members, acquaintances, and strangers. c Maternal age: The comparison group was age 30 years and older versus under 20 years. d Prior spontaneous abortion: The comparison group was women with a prior pregnancy but no spontaneous abortions versus women reporting two or more spontaneous abortions. e Education: The comparison group was post– high school training or higher versus less than high school education. Nelson. Physical violence and spontaneous abortion. Fertil Steril 2003.
cigarette or cocaine use, and neighborhood violence were not related to SAB. The results were similar comparing the subset of cases who experienced a SAB at follow-up with the controls (Table 2). The regression model examining lifetime intimate partner violence also failed to show a significant relationship between violence and SAB (results not shown). The examination of violence perpetrated by a family member, acquaintance, or stranger provided similar results. As shown in Table 3, we did not detect a relationship between SAB and any measure of violence by non-intimates. Similar results were obtained comparing the cases at follow-up to the control group (see Table 3) and in the models examining violence separately for [1] family members, [2] acquaintances, or [3] strangers (results not shown). Finally, we assessed the influence of controlling behavior by a recent intimate partner on SAB, using scores from the ISA. A high level of controlling behavior by a recent intimate partner was associated with a decreased risk of SAB (OR ⫽ 0.65, 95% CI ⫽ 0.46 – 0.94 per 10 point increase in ISA score) (Table 4). Similar results were found comparing the cases at follow-up with the control group. FERTILITY & STERILITY威
DISCUSSION Regardless of the definition of violence, the perpetrator of the violence, or the time-frame (i.e., lifetime or during pregnancy), we did not find an independent relationship between violence and the risk of SAB in this prospective study of inner-city women. At the same time, we found a very high rate of violence, both among the case and control groups. More than one-half of women reported experiencing at least one episode of violence during their lifetime; 49% by an intimate partner and 58% by a family member, acquaintance, or stranger. In addition, 15% of women reported violence during the pregnancy with the majority of episodes of pregnancy-related violence perpetrated by the woman’s intimate partner (66%). Consistent with other studies, we did find a statistically significant relationship for earlier gestational age and a history of two or more SABs, and a future SAB (38, 39). We were also interested in examining whether previous SAB was related to violence exposure in our population. We did find a relationship between a history of SAB and intimate 1209
TABLE 4 Multivariate analysis of controlling behavior and spontaneous abortion among inner-city pregnant women.
Variablea Controlling behaviorb Maternal agec Gestational age Social support Prior spontaneous abortiond Education levele Cigarette use Cocaine use Neighborhood violence
Total cases versus controls odds ratio (95% CI)
Cases at followup versus controls odds ratio (95% CI)
0.65 (0.46–0.94) 1.13 (0.62–2.05) 0.92 (0.88–0.96) 1.08 (0.72–1.63) 2.32 (1.19–4.53) 1.54 (0.93–2.56) 1.27 (0.84–1.92) 0.90 (0.58–1.39) 1.09 (0.58–2.06)
0.69 (0.44–1.09) 1.05 (0.49–2.27) 0.90 (0.85–0.96) 0.90 (0.54–1.51) 2.59 (1.18–5.67) 1.82 (0.95–3.51) 1.36 (0.80–2.31) 0.90 (0.51–1.60) 1.36 (0.63–2.90)
standardized interview, obtaining confidential information in an ED setting, and interacting with women who had or are still experiencing violence in their lives. Violence was assessed by asking each woman “How many times during any argument or fight did you get pushed, slapped, or hurt in any way?” and then collecting information regarding the perpetrator(s) relationship to the woman and violence over the lifetime and since the last menstrual period. Numerous advantages exist in recruiting pregnant women from an ED setting, such as enrolling a group of women very early in gestation (our mean gestational age was 9 weeks), identifying early SABs, and including a group of women at high risk for both the study outcome and the study exposure (4).
a
Our control group consisted of pregnant women recruited from the ED who met the study eligibility requirements and maintained their pregnancy past 22 weeks’ gestation. Although more cases compared with controls presented with vaginal bleeding, a post hoc analysis comparing cases and controls on use of ED services found that over one-half of both groups had multiple urgent and non-urgent visits to the ED during the index pregnancy. Thus, both the case and control group used the ED for similar reasons.
Nelson. Physical violence and spontaneous abortion. Fertil Steril 2003.
This study population consisted of predominately low income, African-American pregnant women recruited from an urban ED and the study results may not be generalizable to other groups of pregnant women. In addition to the high prevalence of reported violence, this population also experienced a high incidence of SAB compared with a prenatal care population. Previous studies examining the relationship between violence and pregnancy outcomes collected information on past experiences of verbal abuse (such as swearing, insulting, or degrading behavior), but we did not obtain this information.
Variables included in the model were maternal age (categorical), gestational age (continuous), cigarette and cocaine use (dichotomous), prior spontaneous abortion (categorical), social support (dichotomous), education (categorical) b Controlling behavior was defined using the ISA (per 10 unit increase). c Maternal age: The comparison group was age 30 years and older versus under 20 years. d Prior spontaneous abortion: The comparison group was women with a prior pregnancy but no spontaneous abortions versus women reporting two or more spontaneous abortions. e Education: The comparison group was post– high school training or higher versus less than high school education.
partner violence. Women reporting a history of intimate partner violence were significantly more likely to report having had a SAB (33% vs. 24%, P⬍.001). However, experiencing rape, child abuse, or previous non-intimate partner violence was not related to a history of SAB. In addition, the multivariate models excluding history of SAB provided similar findings. Thus, current or previous spontaneous risk was not related to violence in our population, adjusting for other covariates. An unexpected finding was a reduced risk of SAB with increasing ISA scores. However, the magnitude of the difference in the mean ISA scores comparing the cases and controls was marginal (14.0 ⫾ 4.7 vs. 14.7 ⫾ 5.2, respectively). Campbell et al. (40) have examined the relationship between ISA scores and LBW and found an increased risk of LBW among women scoring high on the ISA, defined as a score of 10 or higher (OR ⫽ 3.29, 95% CI: 1.18 –9.18). Using this cut-off, our results show that both the case and control group experienced moderately high levels of controlling behavior by an intimate partner regardless of the slight protective relationship with SAB found in our study. In this study, we measured violence through the administration of an in-person interview conducted by nurse interviewers who had received intensive training in conducting a 1210 Nelson et al.
Physical violence and spontaneous abortion
In addition, our assessment of rape may not capture some episodes of forced sex acts, particularly by an intimate partner (i.e., date rape). This may result in misclassification and bias toward a null finding. The difference in measurement of violence may be one reason for the conflicting results relating violence to adverse pregnancy outcomes. For example, Jagoe et al. (25) compared women with a history of abuse with women with no history of abuse, and Cokkinides et al. (17) assessed violence through postpartum mail surveys. Other studies have measured violence multiple times throughout pregnancy, or found only weak or moderate relationships between violence and later pregnancy outcomes (8, 26). Perhaps violence, either physical or emotional, is more likely to affect women later in pregnancy by disrupting the vascular needs of the placenta. In conclusion, we found a high prevalence of lifetime violence and violence during the pregnancy in our group, with the vast majority of episodes of violence perpetrated by intimate partners. However, we did not detect a statistically significant relationship among physical violence, rape, and the risk of SAB. Although no association was found between Vol. 80, No. 5, November 2003
SAB risk and violence, violence has been related to numerous other adverse pregnancy outcomes (17–25). This is the largest prospective study, to our knowledge, examining the relationships among violence, controlling behavior, and SAB. In our study, physical violence was not related to SAB, but violence during pregnancy and over a woman’s lifetime continues to be very prevalent and represents a consistent, alarming public health issue.
Acknowledgments: The authors thank Lynn Beach, BSN, Patricia Douglas, BSN, Jennifer Hart, BA, Susan Meeker, MSW, and Chekemma Fulmore, MSW, for their extraordinary efforts as primary interviewers who recruited participants, collected data, and followed women in this study. The authors also thank Susan Primavera for technical assistance, and Leona Fields for laboratory assistance.
References 1. O’Campo P, Gielen AC, Faden RR, Xue X, Kass N, Wang MC. Violence by male partners against women during the childbearing year: a contextual analysis. Am J Public Health 1995;85:1092–7. 2. Bailey JE, Kellermann AL, Somes GW, Banton JG, Rivara FP, Rushforth NP. Risk factors for violent death of women in the home. Arch Intern Med 1997;157:777–82. 3. Plichta SB, Abraham C. Violence and gynecologic health in women ⬍50 years old. Am J Obstet Gynecol 1996;174:903–7. 4. Roberts GL, Lawrence JM, O’Toole BI, Raphael B. Domestic violence in the emergency department: two case-control studies of victims. Gen Hosp Psychiatry 1997;19:5–11. 5. Bullock L, McFarlane J, Bateman LH, Miller V. The prevalence and characteristics of battered women in a primary care setting. Nurse Pract 1989;14:47,50,53– 6. 6. Webster J, Sweett S, Stolz TA. Domestic violence in pregnancy. A prevalence study. Med J Aust 1994;161:466 –70. 7. Helton AS, McFarlane J, Anderson ET. Battered and pregnant: a prevalence study. Am J Public Health 1987;77:1337–9. 8. O’Campo P, Gielen AC, Faden RR, Kass N. Verbal abuse and physical violence among a cohort of low-income pregnant women. Womens Health Issues 1994;4:29 –37. 9. Smikle CB, Sorem KA, Satin AJ, Hankins GD. Physical and sexual abuse in a middle-class obstetric population. South Med J 1996;89: 983–8. 10. Gazmararian JA, Lazorick S, Spitz AM, Ballard TJ, Saltzman LE, Marks JS. Prevalence of violence against pregnant women. JAMA 1996;275:1915–20. 11. McFarlane J, Parker B, Soeken K. Abuse during pregnancy: frequency, severity, perpetrator, and risk factors of homicide. Public Health Nurs 1995;12:284 –9. 12. Martin SL, Mackie L, Kupper LL, Buescher PA, Moracco KE. Physical abuse of women before, during and after pregnancy. JAMA 2001;285: 1581–4. 13. Goodwin MM, Gazmararian JA, Johnson CH, Gilbert BC, Saltzman LE. Pregnancy intended ness and physical abuse around the time of pregnancy: findings from the pregnancy risk assessment monitoring system, 1996 –1997. Matern Child Health J 2000;4:85–92. 14. Covington DL, Justason BJ, Wright LN. Severity, manifestations, and consequences of violence among pregnant adolescents. J Adolesc Health 2001;28:55–61. 15. Parker B, McFarlane J, Soeken K, Torres S, Campbell D. Physical and emotional abuse in pregnancy: a comparison of adult and teenage women. Nurs Res 1993;42:173–8.
FERTILITY & STERILITY威
16. McFarlane J, Parker B, Soeken K, Bullock L. Assessing for abuse during pregnancy. JAMA 1992;267:3176 –8. 17. Cokkinides VE, Coker AL, Sanderson M, Addy C, Bethea L. Physical violence during pregnancy: maternal complications and birth outcomes. Obstet Gynecol 1999;93:661–6. 18. Shumway J, O’Campo P, Gielen A, Witter FR, Khouzami AN, Blakemore KJ. Preterm labor, placental abruption, and premature rupture of membranes in relation to maternal violence or verbal abuse. J Matern Fetal Med 1999;8:76 –80. 19. Fernandez FM, Krueger PM. Domestic violence: effect on pregnancy outcome. J Am Osteopath Assoc 1999;99:254 –6. 20. Campbell J, Torres S, Ryan J, King C, Campbell DW, Stallings RY, Fuchs SC. Physical and nonphysical partner abuse and other risk factors for low birth weight among full term and preterm babies. Am J Epidemiol 1999;150:714 –26. 21. Murphy CC, Schei B, Myhr TL, Du Mont J. Abuse: a risk factor for low birth weight? A systematic review and meta-analysis. CMAJ 2001;164: 1567–72. 22. Berenson AB, Wiemann CM, Wilkinson GS, Jones WA, Anderson GD. Perinatal morbidity associated with violence experienced by pregnant women. Am J Obstet Gynecol 1994;170:1760 –9. 23. McFarlane J, Parker B, Soeken K. Abuse during pregnancy: associations with maternal health and infant birth weight. Nurs Res 1996;45: 37–42. 24. Grimstad H, Schei B, Backe B, Jacobsen G. Interpersonal conflict and physical abuse in relation to pregnancy and infant birth weight. J Womens Health Gend Based Med 1999;8:847–53. 25. Jagoe J, Magann EF, Chauhan SP, Morrison JC. The effects of physical abuse on pregnancy outcomes in a low-risk obstetric population. Am J Obstet Gynecol 2000;182:1067–9. 26. Parker B, McFarlane J, Soeken K. Abuse during pregnancy: effects on maternal complications and birth weight in adult and teenage women. Obstet Gynecol 1994;84:323–8. 27. Quinlivan JA, Evans SF. A prospective cohort study of the impact of domestic violence on young teenage pregnancy outcomes. J Pediatr Adolesc Gynecol 2001;14:17–23. 28. Muhajarine N, D’Arcy C. Physical abuse during pregnancy: prevalence and risk factors. CMAJ 1999;160:1007–11. 29. Martin SL, English KT, Clark KA, Cilenti D, Kupper LL. Violence and substance use among North Carolina pregnant women. Am J Public Health 1996;86:991–8. 30. Hedin LW, Janson PO. Domestic violence during pregnancy. The prevalence of physical injuries, substance use, abortions and miscarriages. Acta Obstet Gynecol Scand 2000;79:625–30. 31. Jacoby M, Gorenflo D, Black E, Wunderlich C, Eyler AE. Rapid repeat pregnancy and experiences of interpersonal violence among low-income adolescents. Am J Prev Med 1999;16:318 –21. 32. Ribe JK, Teggatz JR, Harvey CM. Blows to the maternal abdomen causing fetal demise: report of three cases and a review of the literature. J Forensic Sci 1993;38:1092–6. 33. Hudson WW, McIntosh SR. The assessment of spouse abuse: two quantifiable dimensions. J Marriage Fam 1981;43:873–88. 34. Gillogley KM, Evans AT, Hansen RL, Samuels SJ, Batra KK. The perinatal impact of cocaine, amphetamine, and opiate use detected by universal intrapartum screening. Am J Obstet Gynecol 1990;163:1535– 42. 35. McCalla S, Minkoff HL, Feldman J, Delke I, Salwin M, Valencia G, Glass L. The biologic and social consequences of perinatal cocaine use in an inner-city population: results of an anonymous cross-sectional study. Am J Obstet Gynecol 1991;164:625–30. 36. Hadeed AJ, Siegel SR. Maternal cocaine use during pregnancy: effect on the newborn infant. Pediatrics 1989;84:205–10. 37. Feldman JG, Minkoff HC, McCalla S, Salwen M. A cohort study of the impact of perinatal drug use or prematurity in an inner-city population. Am J Public Health 1992;82:726 –8. 38. Kline J, Stein Z, Susser M. Conception to birth: epidemiology of prenatal development. New York: Oxford University Press, 1989. 39. Bracken MB. Perinatal epidemiology. New York: Oxford University Press, 1984. 40. Campbell DW, Campbell J, King C, Parker B, Ryan J. The reliability and factor structure of the index of spouse abuse with African-American women. Violence Vict 1994;9:259 –74.
1211
