Share this article on:

Physical Partner Abuse During Pregnancy: A Risk Factor for Low Birth Weight in Nicaragua

Valladares, Eliette MD, MPH; Ellsberg, Mary PhD; Peña, Rodolfo MD, PhD; Högberg, Ulf MD, PhD; Persson, Lars Åke MD, PhD

Original Research

OBJECTIVE To assess whether being physically abused during pregnancy increases the risk of a low birth weight (LBW) infant.

METHODS We conducted a hospital-based case-control study in León, Nicaragua. Cases consisted of 101 newborns with a birth weight under 2500 g, and for each case two controls with a birth weight over 2500 g were selected randomly from infants born the same day. Anthropometry of newborns was done immediately after birth, and background information and data on experiences of violence and potential confounders were obtained through private interviews with mothers. Crude and adjusted odds ratios (ORs) and population-attributable proportion were calculated for exposure to partner abuse in relation to LBW. Multivariate logistic regression analysis was used to control for potential confounding.

RESULTS Seventy-five percent of LBW newborns (cases) were small for gestational age and 40% were preterm. Twenty-two percent of the mothers of LBW infants had experienced physical abuse during pregnancy by their intimate partners compared with 5% of controls. Low birth weight was associated with physical partner abuse even after adjustment for age, parity, smoking, and socioeconomic status (OR 3.9; 95% confidence interval 1.7, 9.3). Given a causal interpretation of the association, about 16% of the LBW in the infant population could be attributed to physical abuse by a partner in pregnancy.

CONCLUSION Physical abuse by a partner during pregnancy is an independent risk factor for LBW.

Physical abuse by a partner during pregnancy is an independent risk factor for a low birth weight infant.

From Epidemiology, Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden; Department of Obstetrics and Gynecology, León University Hospital, León, Nicaragua; Department of Preventive Medicine, National Autonomous University, León, Nicaragua; Obstetrics and Gynecology, Department of Clinical Sciences, Umeå University, Umeå, Sweden; Public Health Sciences Division, ICDDR, B: Centre for Health and Population Research, Dhaka, Bangladesh; and Program for Appropriate Technology in Health (PATH), Washington, DC.

Address reprint requests to: Eliette Valladares, MD, Epidemiology, Department of Public Health and Clinical Medicine, Umeå University, SE-901 85 Umeå, Sweden; E-mail:

Funding for this research was provided by SAREC (Swedish Agency for Research Cooperation with Developing Countries).

Received October 25, 2001. Received in revised form March 11, 2002. Accepted April 11, 2002.

Violence against women by male partners is recognized as one of the most common forms of gender-based violence and is a significant public health concern.1 International studies suggested that 1–25% of pregnant women are exposed to physical violence by intimate partners during pregnancy.2–5

Physical abuse during pregnancy is a potential health hazard both for the woman and the fetus.6 Women who experience violence during pregnancy are significantly more likely to have conditions such as sexually transmitted infections7,8; bleeding9–11; depression and anxiety8; inadequate prenatal care11; smoking, alcohol, or drug consumption8,10; unintended pregnancy; and poor weight gain.6,11,12 Most of these conditions are also associated with intrauterine growth restriction and low birth weight (LBW). Research regarding a possible independent t effect of partner abuse on birth weight has been inconclusive. Although some studies have found that abuse during pregnancy increases the risk of LBW,9,11,13–15 other studies did not find a significant association.16,17 This discrepancy might be due to differences in study populations (eg, relative importance of growth restriction among newborns with a low weight and the presence of other contributing factors to LBW), sample size, study design, measurements of violence, analytic approaches, and handling of potentially confounding variables.18

The mechanisms linking violence with LBW could be direct, through abdominal trauma linked to placental damage, premature rupture of membranes,19 or release of prostaglandin leading to preterm labor and LBW.20 Stress could also constitute an intermediate pathway from violence to LBW, acting through the neuroendocrine axis, causing the release of catecholamines, beta-endorphin, and cortisol, which can lead to vasoconstriction, fetal hypoxia, fetal growth restriction,21,22 as well as provoke the release of prostaglandin, thereby contributing to preterm labor.21,23 These potential pathways indicate the possibility of multiple mechanisms and multiple outcomes with varying gestational ages and degrees of growth restriction.

We have shown that 52% of ever-married Nicaraguan women have experienced physical violence by the current or former intimate partner, and that one in three of those who had experienced violence had been physically abused during pregnancy.5 Thus, exposure to violence was common, smoking, alcohol use, and chronic energy deficiency (body mass index [BMI] < 18.5 kg/m2) were relatively uncommon among pregnant women in that area. The aim of the present study was to assess whether the exposure to physical violence in pregnancy from an intimate partner increases the risk of giving birth to a LBW infant, considering potential confounders.

Back to Top | Article Outline


The study was conducted at the University Hospital of León, Nicaragua, from July to October 1996. Two-thirds of pregnant women in the area deliver at this hospital, which serves an estimated total population of 200,000. An incident case-control study was performed. Cases were defined as newborns with a birth weight less than 2500 g. Two controls for each case were selected randomly among women who gave birth on the same day as the case. Sample size (unmatched case-control study)24 was based on the following assumptions: exposure to violence in pregnancy was estimated at 15% among controls,9 and the estimated odds ratio (OR) was set at 2.5, with 95% confidence interval (CI) and 80% power. A sample of 90 cases and 180 controls was required, and to further allow for omissions a total of 104 cases and 208 referents were invited. Multiple births and stillbirths were excluded. Three cases with their respective controls were excluded because of missing data. All remaining women participated in the interviews.

Interviews with mothers were performed in privacy 8–24 hours after delivery by one of the authors (E.V.) and a trained female field worker. The same interviewer interviewed a triplet of case and controls. Ten percent of the study sample was reinterviewed by the other interviewer, who was blinded to case or control classification. The agreement between interviews was good (κ 0.86). The Abuse Assessment Screen25 was used to measure emotional and physical abuse by a partner during pregnancy. Emotional abuse was defined as repeated yelling, humiliating, and threatening acts by the male partner during the index pregnancy. Physical abuse was considered as one or more acts of physical aggression perpetrated by the male partner during pregnancy. The parts of the body injured by physical trauma were recorded. Sexual abuse was defined as aggression including sexual assault and was also consistently combined with emotional and physical aggression.

Low birth weight was defined as a birth weight below 2500 g.26 Preterm birth was considered as gestational age less than 37 completed weeks.26 Several mothers could not recall the time of last menstrual period (LMP). Therefore, gestational age at birth was assessed by the Capurro method by a pediatrician and one of the authors (EV).23 Gestational age by the Capurro method was compared with available LMP data, with good κ agreement (0.80). Intrauterine growth restriction was evaluated against reference curves that combined information on gestational age and birth weight (in grams). Small for gestational age (SGA) was defined as birth weight below the tenth percentile, appropriate for gestational age (AGA) as between the tenth and the 90th percentiles, and large for gestational age (LGA) as over the 90th percentile. We classified SGA newborns, based on ponderal index [(100 * birth weight)/(crown-heel length3)], into acute or chronic growth restriction. Acute growth restriction (asymmetric) was considered as ponderal index below the tenth percentile of the reference curve and chronic growth restriction (symmetric) as ponderal index at or over the tenth percentile.26

Socioeconomic status was determined by using a modified version of the Unsatisfied Basic Need Assessment.28 This includes four indicators (school enrollment among minors, housing conditions, availability of sanitary services, and household income) that were arbitrarily dichotomized into high (0–1) and low (2–4) socioeconomic status. Emotional distress during pregnancy was assessed by using a validated scale that was developed by a team of Nicaraguan and Swedish mental health experts to detect a variety of mental health problems, including depression, insecurity, anxiety, fear, and defenselessness.26,27 From four possible answers, a dichotomous variable was created to measure frequency, with never or sometimes = 0 and most of the time or all the time = 1. Of a total possible score of 5, a woman was considered to be positive for stress with a score of 3 or more. Mothers' weight and height were measured immediately before delivery. Body mass index was calculated as weight/height2. Primary education for three or more years was considered as having formal education. The number of visits to prenatal care clinics (offering routine health care to all pregnant women) was dichotomized into 0–3 and 4 and above. The questionnaire also included questions about age, marital status, urban or rural residency, and medical and obstetric history.

Crude and adjusted ORs with 95% CIs were calculated using the Statistical Package for the Social Sciences 8.0 (SPSS Inc., Chicago, IL) in bivariate and multivariate logistic regression analysis. Population attributable proportion was calculated as [(OR − 1)/OR](fe), where fe is the proportion of cases exposed to violence. Population attributable proportion combines the size of the risk estimate with the information on how common the exposure is in the target population and is, thus, a measure of the public health impact of the exposure.

Ethical approval was obtained from the Medical Faculty of León University. All respondents gave informed consent.

Back to Top | Article Outline


During the study period 7.5% (134 of 1794) of all live births were LBW, of which 75% fulfilled the inclusion criteria. Sixty percent of the LBW infants studied were SGA and term, 15% were SGA and preterm, and 25% were preterm but with adequate weight for gestational age. Among SGA newborns, 58% had symmetric or chronic growth restriction, and 42% had asymmetric or acute growth restriction.

Twenty-two percent of the mothers of LBW infants had been physically abused by their partners during pregnancy, compared with 5% of the control mothers. All women who reported physical violence by a partner also reported that they had suffered humiliations, threats, and verbal aggression. Moreover, 12% of mothers of cases and 3% of mothers of controls had been abused sexually by their partners in addition to being abused physically and emotionally. Fifteen percent of case mothers and 4% of controls had experienced more than three acts of physical abuse during the index pregnancy. Forty percent of abused women reported that physical abuse began during pregnancy, and 13% reported that violence had worsened during pregnancy. Most women who reported physical abuse had been injured in the extremities and not in the abdomen.

Mothers who experienced any physical violence during pregnancy gave birth more frequently to LBW infants (Table 1). Potential confounding factors were evaluated, and if they were associated with physical violence as well as the outcome (LBW), they were included in the multivariate analysis. The association between physical violence and LBW remained significant even after adjustment for parity, socioeconomic status, mother's age, and smoking habits (OR 3.98; 95% CI 1.7, 9.31). This association was also analyzed for subgroups of LBW according to type of growth restriction and gestational age at birth (Table 2). Physical abuse during pregnancy consistently was associated with a significant increase in LBW in all subgroups.

Table 1

Table 1

Table 2

Table 2

Perceived stress was associated with LBW. Abused women more frequently reported perceived stress during pregnancy compared with nonabused women (30% and 15%, respectively). Mothers of cases more frequently reported that the pregnancy was unwanted (Table 1). The main reasons given for the pregnancy to be unwanted were “conflict in the relationship with the partner” and “economic difficulties.” Mothers of LBW infants had attended prenatal care less frequently than control mothers, often stating that their partners did not allow it. The mother's height or BMI measured immediately before delivery was not associated with LBW.

Mothers who had preeclampsia during pregnancy or bleeding during the third trimester gave birth to significantly more LBW infants. No significant association was found between physical abuse and preeclampsia or bleeding.

If evaluating the public health impact of partner abuse in relation to LBW, the risk assessment can be combined with the information on the prevalence of the violence. Given a causal interpretation of the association between violence and LBW, 16% of the LBW in the study population could be attributed to partners' physical abuse in pregnancy.

Back to Top | Article Outline


Physical violence in pregnancy by intimate partners was independently associated with a birth weight below 2500 g, suggesting a causal link between abuse and intrauterine growth restriction. The study did not have sufficient power to confirm or refute an association to preterm delivery.

The target population of this study was newborn infants of women in urban and rural areas of León, Nicaragua. A major part of the deliveries in the area took place at the hospital, where cases and controls were recruited. There is still the possibility of different means of delivery of potential LBW cases and non-LBW controls, respectively. For example, individuals with increased risk of LBW could be referred more frequently to the hospital for delivery. A woman exposed to violence in pregnancy is more frequently restricted from access to health care or institutional assistance. Thus, such selection bias, if present, would result in an underestimate of the exposure to violence of the potential LBW-delivering mothers in the community and a subsequent underestimate of the OR.

The assessment of women's experiences of violence is sensitive and might be underreported. The instrument used, the Abuse Assessment Screen, has been applied in multiple settings and is considered appropriate for measuring violence during pregnancy.9,29 Masked, repeated interviews showed good agreement. Thus, there were no indications of differential misclassification of exposure.

Others have attempted to determine the effect of violence during pregnancy on birth weight, but with inconclusive results.9,11,16,17,20 Most of those studies were done in high-income countries, with relatively high prevalence of smoking and sometimes also other substance abuse, which are well-known risk factors for intrauterine growth restriction.16,17 In most cases, their analyses did not control for those factors or for other potential confounders. Other studies found a crude association between violence during pregnancy and LBW but did not show any independent effect when adjusting for smoking, alcohol, or other substance use.9,11,13 In the current study we made efforts to control for confounding in the analysis. Smoking and use of alcohol or other substances were uncommon but were still accounted for in the analysis. The socioeconomic status, measured as unsatisfied basic needs, represented other confounding factors that are socially stratified. The possibility of major residual confounding is less likely, given the strength of the effect after these adjustments.

Physical abuse during pregnancy was strongly associated with perceived stress. This finding suggests that a plausible causal pathway for the association between physical abuse during pregnancy and growth restriction of the fetus could be through the induced stress and activation of the neuroendocrine axis.21 This activation determines changes in different maternal-placental-fetal hormones that could lead to vasoconstriction and fetal growth restriction as well as uterine contractions and preterm labor.21,23 By these mechanisms, physical abuse could result in preterm birth as well as growth restriction. The timing of the events and the stress induced could lead to symmetric or asymmetric growth restriction. Women in this study rarely reported direct abdominal trauma that could cause preterm labor.

In the present study, three-quarters of the LBW infants were small for gestational age. In Sweden, only 31% of all LBW live births are small for gestational age (personal communication, Centre for Epidemiology, National Board of Health and Welfare, Sweden). Such differences in intrauterine growth and gestational age at birth between countries and regions of the world also reflect differences in contributing causes. This must be considered when interpreting earlier inconclusive results.9,11,13,16,17,20 The etiology of intrauterine growth restriction is not the same as that of preterm delivery, and growth restriction initiated early in pregnancy presumably has a different cause than growth restriction with a late onset. The presence of other contributing causal factors might be necessary for exposure to physical violence in pregnancy to result in intrauterine growth restriction. This might explain why no association could be found in a study of domestic violence and low birth weight in Norway.16

Thus, the results of the present study indicate that there is an independent effect of physical abuse during pregnancy on the birth weight of the offspring, mainly through intrauterine growth restriction. The biologic mechanisms are not known, but the association to perceived stress and earlier research on stress and fetal development indicate that a chronic stress-induced biological mechanism is plausible.

Back to Top | Article Outline


1. Heise L, Ellsberg M, Gottemoeller M. Ending violence against women. Population Rep 1999;27:43.
2. Leung WC, Leung TW, Lam YY, Ho PC. The prevalence of domestic violence against pregnant women in a Chinese community. Int J Gynecol Obstet 1999;66:23–30.
3. Ellsberg M, Peña R, Herrera A, Liljestrand J, Winkvist A. Candies in hell: Women's experiences of violence in Nicaragua. Soc Sci Med 2000;51:1595–610.
4. Gazmararian JA, Peterson R, Spitz AM, Goodwin MM, Saltzman LE, Marks JS. Violence and reproductive health: Current knowledge and future research directions. Matern Child Health J 2000;4:79–84.
5. Amaro H, Fried LE, Cabral H, Zuckerman B. Violence during pragnancy and substance abuse. Am J Public Health 1990;80:575–9.
6. 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.
7. Shei B, Samuelsen S, Bakketeig L. Does spousal physical abuse affect the outcome of pregnancy. Scand J Soc Med 1991;19:26–31.
8. Curry MA, Doyle BA, Gilhooley J. Abuse among pregnant adolescents: Differences by developmental age. MCN Am J Matern Child Nurs 1998;23:144–50.
9. Goodwin MM, Gazmararian JA, Johnson CH, Gilbert BC, Saltzman LE. Pregnancy intendedness and physical abuse around the time of pregnancy: Findings from the pregnancy risk assessment monitoring system, 1996–1997. Matern Child Health J 2000;4:85–101.
10. Bullock LF, McFarlane J. The birth-weight/battering connection. Am J Nurs 1989;89:1153–5.
11. Valdez SR, Sanin LH. La violencia doméstica durante el embarazo y su relación con el peso al nacer. Salud Pública Mex 1996;38:352–62.
12. Murphy C, Shei B, Myhr T, Du Mont J. Abuse: A risk factor for low birth weight? A systematic review and meta-analysis. Can Med Assn J 2001;164:1567–72.
13. Grimstad H, Shei B, Backe B, Jacobsen G. Physical abuse and low birth weight: A case-control study. Br J Obstet Gynaecol 1997;104:1281–7.
14. Dye TD, Tollivert NJ, Lee RV, Kenney CJ. Violence, pregnancy and birth outcome in Appalachia. Pediatric Perinat Epidemiol 1995;9:35–47.
15. Peterson R, Gazmararian J, Spitz A, 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.
16. 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.
17. Berenson AB, Wiemann CM, Wilkinson GS, Jones WA, Anderson GD. Perinatal morbidity associated with violence experienced by pregnant women. Am J Obstet Gynecol 1994;170:1760–6.
18. Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA. Prenatal psychosocial factors and the neuroendocrine axis in human pregnancy. Psychosom Med 1996;58:432–46.
19. Dejin-Karlsson E, Hanson BS, Ostergren PO, Sjoberg NO, Marsal K. Does passive smoking in early pregnancy increase the risk of small-for-gestational-age infants? Am J Public Health 1998;88:1523–7.
20. Austin MP, Leader L. Maternal stress and obstetric and infant outcomes: Epidemiological findings and neuroendocrines mechanism. J Obstet Gynecol 2000;40:331–7.
21. McFarlane J, Parker B, Soeken K, Bullock L. Assessing for abuse during pregnancy. Severity and frequency of injuries and associated entry into prenatal care. JAMA 1992; 267:3176–8.
22. World Health Organization. Physical status: The use and interpretation of anthropometry. Infants and children. World Health Organization 1995;854:161–262.
23. Capurro H, Konichezky S, Fonseca D, Caldeyro-Barcia R. A simplified method for diagnosis of gestational age in the newborn infant. J Pediatr 1978;93:120–2.
24. Lubchenco L, et al. Intrauterine growth as estimated from liveborn birth weight data at 26 to 42 weeks of gestation. Pediatrics 1963;32:793–800.
25. Renzi M, Agurto S. Mercado laboral y condiciones de vida de la población urbana de Managua León y Granada. FIDEG: Fundación Internacional para el Desafío Económico Global. Managua, Nicaragua: Editorial Nueva Nicaragua, 1993.
26. Mezzich JE, Caldera JT, Berganza CE. Psychiatric diagnosis in primary care and Personal Health Scale. In: Beigal A, López I, Costa e Silva JA, eds. Past, present and future of psychiatry: IX World Congress of Psychiatry. Singapore. Singapore: World Scientific Publication, 1994.
27. Ellsberg M, Caldera T, Herrera A, Winkvist A, Kullgren G. Domestic violence and emotional distress among Nicaraguan women: Results from a population based study. Am Psychol 1999;54:30–6.
28. Campbell J. Addressing battering during pregnancy: Reducing low birth weight and ongoing abuse. Semin Perinatol 1995;19:301–6.
29. Centre of Epidemiology. Medical birth registry, 1991–1996. National Board of Health and Welfare, Sweden, 2001.
© 2002 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.