El Kady, Dina MD; Gilbert, William M. MD; Xing, Guibo PhD; Smith, Lloyd H. MD, PhD
Violence during pregnancy is a tragic, yet not uncommon event. Domestic violence is responsible for more deaths in pregnant women than any single medical complication associated with pregnancy.1 In the course of a year in the United States, more than 1.8 million women will be victims of assault,2 thereby affecting up to 335,000 pregnant women and their infants.3,4 With the prevalence of violence in pregnancy reported to be between 0.9% and 31%,5–9 it may exceed the prevalence of diabetes or preeclampsia and is therefore a major public health issue.1,3
There have been differences reported among studies regarding the frequency, prevalence, and outcomes of violence in pregnancy.1–11 This may be secondary to lack of standard definitions, variations in reporting, differences in the timing of the assault in relation to the delivery, and different populations studied.3,11 Nevertheless, there is reported to be an increased prevalence of violence during pregnancy1,3,12 and a debated association between assault and low birth weight (LBW) infants and adverse outcomes.12–22 Regardless of whether these poor outcomes are secondary to the physical assault or the associated psychological stress, violence, in and of itself, is a risk factor for poor outcomes.14 For health care professionals, and especially prenatal care providers, there is a potentially important window of opportunity for early detection, minimization, and potential elimination of assaults and associated adverse outcomes during pregnancy.1,3,23
Our study seeks to estimate the maternal, fetal, neonatal, and infant outcomes of women hospitalized for assault during pregnancy in a large obstetric population. The types, mechanism, and severity of the injuries sustained from assaults are assessed. Outcomes are analyzed as a function of the timing of obstetric delivery: 1) women admitted after sustaining an assault who are delivered during the assault hospitalization (ASSLT-DEL) and 2) women admitted after sustaining an assault prenatally who are discharged and delivered at a subsequent hospitalization (SUBSQ-DEL).
MATERIALS AND METHODS
A computerized database of hospital discharge records, linked to birth and death certificates in the state of California from 1990 to 1999, was used. The Vital Statistics–Patient Discharge Database is compiled from hospitals reporting to the California Office of Statewide Health Planning and Development. Linkage of maternal and neonatal/infant hospital discharge records with birth and death certificates has been shown to be successful in 98% of cases.24 This study was approved by Office of Statewide Health Planning and Development and by the Human Subjects Committee at The University of California, Davis.
All assaults during pregnancy resulting in hospitalization were identified. Women were included in the study if assault was identified as the mechanism of an injury using external causation codes (E-codes, E960–E969). Assaults that did not result in hospitalization and pregnancies resulting in delivery before 20 completed weeks of gestation are not captured in the Vital Statistics–Patient Discharge Database. Pregnancy outcomes for women delivering after 20 weeks were available for the entire length of gestation. Perpetrator codes, identifying whether the assault was committed by a partner or spouse (E967.0–E967.9), were identified for all assaults after 1997 (after which the code was initiated). Codes using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) were used to identify maternal and fetal outcome measures. These included preterm labor (644.0, 644.2), premature rupture of the membranes (658.2), placental abruption (641.2), cesarean hysterectomy (68.3, 68.4, 68.6, 68.8), uterine rupture before and during labor (665.0, 665.1), blood transfusion (99.0), antepartum hemorrhage (641.1–641.3, 641.8, 641.9), infectious complications (670, 646.6), cesarean delivery, and maternal death. Fetal/neonatal/infant outcomes included premature delivery (< 37 weeks of gestation), LBW (< 2,500 g), fetal distress (656.3, 768.2–768.4), birth asphyxia (768.5–768.7, 768.9), respiratory distress syndrome (769), and fetal, neonatal, and infant death. Because women sustaining assault are more likely to report a history of substance abuse,25 and because substance abuse may have a direct impact on fetal weight,26 codes 304–305 were used to identify substance abuse in these women.
Different injury types were identified, including fractures/dislocations/sprains/strains (800–849), intracranial injuries (850–854), internal injuries to the thorax/abdomen/pelvis (860–869), open wounds (870–897), injury to blood vessels (900–904), superficial injuries/contusions/crushing injuries (910–929), burns (940–949), and nerve and spinal cord injuries (950–957).
Injury severity scores have previously been used to assess outcomes in pregnant trauma patients.27 It is an anatomic scoring system where 6 body regions (head and neck, face, chest, abdomen/pelvis, extremity, external) are assigned scores (1–6) depending on the nature of the injury. The highest 3 scores are squared and summed to produce the injury severity score. Scores of 9 or higher have been previously used in the obstetric literature to designate severe injury.28 Injury severity scores were calculated using ICDMAP-90 software (Tri Analytics Inc, Forest Hill, MD).
Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated, and logistic regression was used to perform outcome analysis. Because confounding factors are known to affect outcomes, the odds ratios are controlled for maternal age, race, parity, education, prenatal care,29,30 and insurance type. Additional multivariate logistic regression analysis was performed to assess the effect of assault and intimate partner violence on LBW, controlling for gestational age and substance abuse. Maternal and neonatal outcomes were compared to pregnant women who did not sustain assault.
A total of 4,833,286 obstetrical deliveries were identified in the Vital Statistics–Patient Discharge Database from 1990–1999. A total of 2,070 women were hospitalized after sustaining an assault during pregnancy (occurrence rate 0.04%). The most common mechanism by which assault was sustained was an unarmed fight or brawl, as shown in Table 1. Thirteen percent of women admitted for assault delivered during that hospitalization.
Superficial injuries, accounted for the largest type of injury in the ASSLT-DEL group (28%), followed by internal injuries to the thorax, abdomen, and pelvis, but only 51% of the injury types in this group were reported in the database. Conversely, 98% of the types of injury in the SUBSQ-DEL group were reported, with open wounds the most common (28%), followed by superficial injuries, contusions and crushing injuries (27%), and fractures dislocations, sprains, and strains (20%).
Demographic data of pregnant women sustaining assault are shown in Table 2. For both groups, there were higher percentages of women aged 20 years and under, a higher percentage of African-American women, and a higher rate of multiparous women (parity > 3). Assaulted women were more likely to have less than a high school education, be a participant in California's state Medi-Cal program, and were more likely to deliver prematurely. In fact, only 59% of the ASSLT-DEL group delivered at term. The lowest rates of prenatal care are also seen in the ASSLT-DEL women, with only 58% initiating prenatal care in the first trimester compared with 82% in the control group.
Increases in adverse maternal and fetal outcomes are demonstrated for both ASSLT-DEL and SUBSQ-DEL women compared with pregnant women who did not sustain assault (Table 3). The ASSLT-DEL women had the worst outcomes, including an almost 20-fold increased risk of maternal death. Cases of maternal death were associated with multiple injury types: fractures, internal injury, and open wounds. All maternal deaths were associated with concomitant fetal death. Adverse outcomes in the ASSLT-DEL group included increased risks of abruption, stillbirth, blood transfusions, hysterectomy, maternal death, neonatal death, infant death, and almost all other adverse outcomes measured (Table 3). Of interest is a very high rate of uterine rupture before the onset of labor in these women. These women also had increased rates of readmission within 90 days postpartum (OR 3.19, 95% CI 1.75–5.80, P < .001).
At the time that the SUBSQ-DEL women were hospitalized for the assault, there were immediate, as well as long-term, adverse effects. Immediate adverse outcomes included a 7.6-fold increased risk of rupturing membranes (95% CI 3.9–14.7, P < .000) and a 28-fold increased risk of receiving a blood transfusion (95% CI 20.5–40, P < .000) compared with unassaulted women. In addition, SUBSQ-DEL women had an increased number of antepartum admissions (range 1–12, mean 1.6) compared with controls (P < .001) and a 3-fold increase in rates of readmission within 90 days of delivery (OR 3.33, 95% CI 2.64–4.21, P < .001). The average time from admission to subsequent delivery was 134 days. Long-term adverse outcomes persisted at the time of readmission for subsequent delivery, with increased risks of prematurity, abruption, and additional blood transfusions (Table 3). A nonsignificant increase in maternal death rate was seen in these women as well. There were no increased risks of fetal, neonatal, or infant death in the SUBSQ-DEL women.
The risks of LBW were increased for both ASSLT-DEL and SUBSQ-DEL women, with the highest risk seen in ASSLT-DEL women (Table 4). When controlling for gestational age and substance abuse, the LBW risk persisted. Perpetrator codes, available after 1997, revealed the fact that 50% of the assaults in the ASSLT-DEL group and 20% in the SUBSQ-DEL group were perpetrated by a spouse or partner (intimate partner violence). Regression analysis revealed that intimate partner violence resulted in increased risk of LBW for the ASSLT-DEL group (Table 4).
The mean injury severity score for the ASSLT-DEL group was 26, and for SUBSQ-DEL group it was 20. Further severity analysis was not performed because both of these scores are above 9 and therefore considered to be in the severe injury range. This demonstrates the severe nature of injuries that were inflicted secondary to an assault in pregnancy leading to hospitalization.
This study analyzes the clinical outcomes of women hospitalized for assault-associated injuries during pregnancy in a very large obstetric population. The ability to stratify by timing of the delivery in relation to the assault led to the observation that there are significant immediate and long-term adverse maternal, fetal, and neonatal outcomes for these women. Therefore, undelivered assaulted women should be monitored closely during the remainder of the pregnancy.
The risks of uterine rupture and maternal mortality were high in women delivering at the assault hospitalization. The rate of uterine rupture before the onset of labor was found to be much higher than previously reported and suggests that the clinician should be aware of the possibility of maternal assault when the diagnosis of uterine rupture is made before labor. The increased risk of death in the pregnant women could be explained by the risk of homicide in this age group.1 However, when controlling for age, assault still appears to have an effect, suggesting that the increased maternal mortality may be secondary to the assault sustained.
The etiologies of the long-term consequences of assaults seen in this study (increased risks of preterm labor, premature rupture of membranes, preterm delivery, LBW, abruption, hemorrhage, transfusion, and infection) can only be speculative. The persisting adverse outcomes could be explained by repeated assaults because some of these women may, in fact, return to abusive environments. Another theory is that the original assault has led to a subclinical partial placental abruption, which in turn progresses into a chronic abruption and placental insufficiency,31 leading to the adverse outcomes observed.
Efforts to decrease assaults in pregnant women should be focused around 2 main areas. First, because intimate partner violence accounts for approximately 50% of the assaults in women delivering at the time of the assault hospitalization and because this group has the worst outcomes, diligent efforts of prenatal care providers in increasing the detection and minimization of intimate partner violence is important. Second, because the remaining 50% of assaults are other from other sources, teaching violence prevention to the pregnant patient (ie, avoidance of fights and physical conflicts where possible) may also be beneficial. It is intuitive to advocate for diagnosis, counseling, and intervention of cases of assault early in the pregnancy because the group sustaining the severest assaults and worst outcomes were more likely to deliver prematurely.
There are certain demographic variables that are associated with increased risk of assault, ie, the younger, multiparous women with poor socioeconomic status presenting late to care. It would be important to target these women for violence screening and counseling. Although these women presented later for care, 85% of women with the worst outcomes had contact with the health care system at some point during the pregnancy and thus could have been identified and targeted with preventive measures.
Estimation of the true prevalence of assault during pregnancy is difficult because wide ranges are reported in the literature (0.9–31%).5–9 Our population-based study's occurrence rate of 4.3 per 10,000 deliveries is lower than expected. This can be explained, however, by the inclusion of only those women hospitalized for their assault injury and by the exclusion of women who lost their pregnancies at less than 20 weeks of gestation. Many women sustaining assault may not report their injuries, may not even seek medical care, and if they do, may not be admitted to the hospital for these injuries. However, it has been shown that pregnant women who sustained assaults were more likely to be admitted for less serious conditions than their nonpregnant counterparts.10
Our population of assaulted women, despite the large size, by no means captures the scope of violence during pregnancy, and we believe that the true prevalence of assault in pregnancy cannot be determined from this study. Our study may only reflect a subset of severe assaults, as noted by the high injury severity scores, yet few published studies characterize outcomes based on the severity of the assault. In one study by Covington et al,19 severe violence during pregnancy was elicited in 37 women by using a questionnaire and was associated with high rates of adverse outcomes, including preterm labor, preterm delivery, and LBW infants.
Our study has the limitations of reporting and coding errors inherent in a large retrospective database. This study's large numbers and the consistency of the demographics10,15,16 in comparison with the results of other studies mitigate concerns regarding the potential effect of coding errors.
There is controversy in the current literature in regard to assaults as a contributor to LBW infants. Our data suggest that assault is a significant immediate and long-term risk factor for reduction of birth weight, among other poor outcomes. Bullock and McFarlane13 were among the first to detect a connection between violence and LBW, and since then multiple studies using various methodologies have both supported and contradicted this association.12–22 The association of LBW with assault has been attributed to late or no prenatal care, poor socioeconomic status, or substance abuse. However, when controlling for all of these factors in our study, the poor outcomes persisted (Table 4). Although this study may have presented outcomes for severe assaults, once a pregnant patient is at risk for an assault, no one can predict the severity of the assault that she may encounter.
Maternal complications of pregnancy (ie, diabetes and preeclampsia) are not associated with maternal death rates as high as those seen with assault (1%), and yet these conditions receive much more interest and support for prevention. If we are to decrease mortality in pregnant women, we will need to direct resources in the area of assault prevention. Moreover, the long-term consequences warrant additional monitoring of women who have been assaulted and continue their pregnancy. Therefore, health providers should have the same enthusiasm for reducing the risks associated with assaults as they do for any other medical condition complicating pregnancy.
1. Liebschutz J, Frayne S, Saxe G, Charney P. Violence against women: a physician's guide to identification and management. Philadelphia (PA): American College of Physicians/American Society of Internal Medicine; 2003.
2. 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.
3. Peterson R, Saltzman L, Goodwin M, Spitz A. Key scientific issues for research on violence occurring around the time of pregnancy. National Center for Chronic Disease Prevention and Heath Promotion, and the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/reproductivehealth/wh_violenc.htm
. Retrieved November 19, 2004.
4. Lipsky S, Holt V, Easterling T, Critchlow C. Impact of police-reported intimate partner violence during pregnancy on birth outcomes. Obstet Gynecol 2003;102:557–64.
5. Fernandez FM, Krueger PM. Domestic violence: effect on pregnancy outcome. J Am Osteopath Assoc 1999;99:254–6.
6. Rachana C, Suriaya K, Hisham AS, Abdulaziz AM, Hai A. Prevalence and complications of physical violence during pregnancy. Eur J Obstet Gynecol Reprod Biol 2002;103:26–9.
7. Hedin L, Grimstad H, Moller A, Schei B, Janson P. Prevalence of physical and sexual abuse before and during pregnancy among Swedish couples. Acta Obstet Gynecol Scand 1999;78:310–5.
8. Martin S, Mackie L, Kupper L, Buescher P, Moracco K. Physical abuse of women before, during and after pregnancy. JAMA 2001;285:1581–4.
9. Evins G, Cheschier N. Prevalence of domestic violence among women seeking abortion services. Womens Health Issues 1996;6:204–10.
10. Weiss H, Lawrence B, Miller T. Pregnancy-associated assault hospitalization. Obstet Gynecol 2002;100:773–80.
11. Jasinski JL. Pregnancy and domestic violence: a review of the literature. Trauma Violence Abuse 2004;5:47–64.
12. Newberger E, Barkan S, Lieberman E, McCormick M, Yllo K, Gary L, Schechter S. Abuse of pregnant women and adverse birth outcome. JAMA 1992;267:2370–2.
13. Bullock L, McFarlane J. The birth weight/battering connection. Am J Nurs 1989;89:1153–5.
14. Murphy C, Schei B, Myhr T, Du Mont J. Abuse: a risk factor for low birth weight? A systematic review and meta-analysis. CMAJ 2001;164:1567–72.
15. 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.
16. Curry M, Perrin N, Wall E. Effects of abuse on maternal complications and birth weight in adult and adolescent women. Obstet Gynecol 1998;92:530–4.
17. Valladares E, Ellsberg M, Pena R, Hogberg U, Persson L. Physical partner abuse during pregnancy: a risk factor for low birth weight in Nicaragua. Obstet Gynecol 2002;100:700–5.
18. Janssen P, Holt V, Sugg N, Emanuel I, Critchlow C, Henderson A. Intimate partner violence and adverse pregnancy outcomes: a population-based study. Am J Obstet Gynecol 2003;188:1341–7.
19. Covington D, Hage M, Hall T, Mathis M. Preterm delivery and the severity of violence during pregnancy. J Reprod Med 2001;46:1031–9.
20. Cokkinides V, Coker A, Sanderson M, Addy C, Bethea L. Physical violence during pregnancy: maternal complications and birth outcomes. Obstet Gynecol 1999;93:661–6.
21. 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.
22. Jagoe J, Magann E, Chauhan S, Morrison J. The effects of physical abuse on pregnancy outcomes in a low risk obstetric population. Am J Obstet Gynecol 2000;182:1067–9.
23. Hillard P. Physical abuse in pregnancy. Obstet Gynecol 1985;66:185–90.
24. Herrchen B, Gould J, Nesbitt T. Vital statistics linked birth/infant death and hospital discharge record linkage for epidemiological research. Comput Biomed Res 1997;30:290–305.
25. Kearney M, Munro B, Kelly U, Hawkins J. Health behaviors as mediators for the effect of partner abuse on infant birth weight. Nurs Res 2004;53:36–45.
26. Berenson A, Stiglich N, Wilkinson G, Anderson G. Drug abuse and other risk factors for physical abuse in pregnancy among white non-Hispanic, black, and Hispanic women. Am J Obstet Gynecol 1991;164:1491–9.
27. Schiff MA, Holt VL. The injury severity score in pregnant trauma patients: predicting placental abruption and fetal death. J Trauma 2002;53:946–9.
28. El Kady D, Gilbert WM, Anderson J, Danielsen B, Towner D, Smith LH. Trauma during pregnancy: an analysis of maternal and fetal outcomes in a large population. Am J Obstet Gynecol 2004;190:1661–8.
29. Dietz P, Gazamarian J, Goodwin M, Bruce C, Johnson C, Rochat R. Delayed entry into prenatal care: effect of physical violence. Obstet Gynecol 1997;90:221–4.
30. Webster J, Chandler J, Battistutta D. Pregnancy outcomes and health care use: effects of abuse. Am J Obstet Gynecol 1996;174:760–7.
31. Williams JK, McClain L, Rosemurgy AS, Colorado NM. Evaluation of blunt abdominal trauma in the third trimester of pregnancy: maternal and fetal considerations. Obstet Gynecol 1990;75:33–7.