The Co-occurrence of Intimate Partner Violence and Syphilis among Pregnant Women in Bolivia

Abstract

Background:

Intimate partner violence (IPV) and sexually transmitted infections (STI) are epidemics that disproportionately affect women. This study determined IPV prevalence and the association between IPV and positive syphilis tests among pregnant women attending antenatal clinics in three provinces in Bolivia.

Methods:

We administered structured questionnaires to women after syphilis testing. The questionnaire included sociodemographic variables and four questions form the modified version of the Abuse Assessment Screen (AAS) to assess physical and sexual violence.

Results:

Of 6002 women who completed the violence questionnaire and had a syphilis test, 20.4% (n = 1227) reported physical or sexual abuse or both committed by their partner in the past year. Prevalence of positive syphilis tests was twice as high among women who reported IPV (8%) than among women who did not (4%) (p < 0.01). Women's age (younger), education level (lower), occupation as homemaker, being in a consensual union, more previous pregnancies, lower economic status, and language spoken at home (Spanish and an indigenous language), as well as history of IPV in the past 12 months, were significantly associated with positive syphilis tests in bivariate analysis. History of IPV remained significantly associated with positive syphilis tests in multivariate analysis (OR 1.59, 95% CI 1.23-2.07). In addition, low education among women's partners and having at least one previous pregnancy were positively associated with IPV in multivariate analysis.

Conclusions:

There is a significant association between history of partner violence and a positive syphilis test among pregnant women, suggesting that syphilis can be an important negative health consequence of IPV. Bolivia's new maternal and infant health program in antenatal clinics, which includes universal syphilis screening, should also provide screening and follow-up care for IPV.

Introduction

Violence against women is a serious public health and human rights problem worldwide. Intimate partner violence (IPV) is the act or threat of physical, sexual, and psychological attacks against past or present intimate partners, and most IPV is committed by men against women.¹ IPV has been associated with a variety of health problems in women, specifically depression and stress, gynecological problems, sleep disorders, chronic pain, unintended pregnancy, and sexually transmitted infections (STIs).^2

–6 A groundbreaking 10-country study on women's health and IPV carried out by the World Health Organization (WHO) found that most gender-based violence was perpetrated by an intimate partner and that overall mental and physical health was worse among women who have experienced abuse compared with women who have not.⁷

It is widely documented that IPV is not only a human rights issue but also can yield negative health outcomes for the victim.^8
–10 In developing countries, reported rates of IPV range from 3.8% to 31.7%,^7,11 and research shows that women who experience IPV are at an increased risk of acquiring STIs.^5,12 Previous studies in Latin America have shown that IPV occurs at a greater rate during pregnancy. In 1998, the Demographic and Health Surveys conducted in Nicaragua reported that 11% of the women who participated in the survey had been physically abused by an intimate partner.¹³ Similarly, a study on maternal syphilis and IPV in Bolivia mentioned that 27% of male partners reported using physical force with their wife when she was pregnant.¹⁴ A study conducted in Nicaragua found an association between IPV against mothers before or during pregnancy and increased risk of under 5 child mortality.¹⁵ These studies show that IPV can have severe health implications for maternal and child health because of an increased risk of STI transmission and physical trauma during pregnancy.

IPV also increases women's vulnerability to STIs, as violence has been shown to limit women's power over their sexual health decisions. Maternal syphilis, which can be potentially fatal in infants and severely debilitating for women, continues to be a leading cause of adverse pregnancy outcomes in many developing countries despite recent advances in screening and treatment.¹⁶ Maternal syphilis is estimated to contribute to 29% of perinatal deaths, 11% of neonatal deaths, and 26% of stillbirths around the world.¹⁶ Research also shows an increased risk of acquiring HIV if syphilis is present.⁷ Antenatal care settings can be important sites of early detection and treatment of maternal syphilis.¹⁷ In areas where most women make at least one visit to antenatal clinics before giving birth, syphilis detection and screening programs can be cost-effective and feasible.¹⁸ Such screening programs also should provide women with appropriate skills to notify their partners of their positive syphilis test to reduce the risk that male partners may react violently to this information.¹⁴

In Bolivia, limited empirical evidence exists about the co-occurrence of IPV and syphilis among pregnant women. Bolivia has comparatively poor health indicators in the region, including having the second highest maternal and infant mortality rates in the hemisphere, the second lowest proportion of women receiving antenatal care, and nearly 60% of its population living in poverty.¹⁷ Accurate national estimates of IPV in Bolivia are unavailable; however, studies have estimated 28%–40% of women report IPV before and during pregnancy.^14,19 Studies in Bolivia also have documented high rates of maternal syphilis.^20,21 One study conducted among seven Bolivian hospitals detected maternal syphilis in 4.3% of women who had live births and in 26% of women whose babies were born dead, and another study reported that 26% of mothers with stillbirths tested positive for syphilis and 15% of children born to mothers who tested positive for syphilis had congenital infection.^20,21 Bolivia has had a universal package of basic health insurance for pregnant women (Seguro Universal Materno-Infantil or SUMI) since 1998 that includes syphilis testing during the antenatal visit. A syphilis rapid test was introduced in 2005 to facilitate standardized screening and treatment screening throughout the country.¹⁸ This evidence suggests that the antenatal care visit may be a feasible time to screen for both IPV and syphilis in Bolivia.

This study had two primary objectives: (1) to determine the prevalence of IPV and positive syphilis serology among pregnant women in antenatal clinics in Bolivia and the sociodemographic factors associated with each condition and (2) to determine the relationship between positive syphilis diagnoses and reports of IPV within the past year. We hypothesize that a positive syphilis serology in this sample of pregnant women is associated with a greater likelihood of IPV in the past year than in women who test negative for syphilis.

Materials and Methods

We conducted a nested cross-sectional study recruiting women who were part of a larger study that assessed the feasibility and acceptability of the Abbot Determine^® rapid syphilis test in Bolivia.¹⁸ The immunochromatographic strip test (ICS) is a fingerstick test that gives results in about 20 minutes. It has a demonstrated specificity of 93%–100% and a sensitivity of 92%–100%.²²

We recruited women from study sites used in the larger study, which included 4 urban hospitals and 37 rural health care centers in La Paz, Santa Cruz, Cochabamba, and Chuquisaca provinces.¹⁸ Inclusion criteria were women who were currently pregnant, who had not had a previous syphilis test, and who had just completed a syphilis rapid test. We recruited women after they had received rapid tests and results to participate in a quantitative interview about IPV conducted in Spanish or an indigenous languages (Aymara or Quechua). Any woman who had refused a syphilis rapid test in the larger study was excluded from the present study. In addition, women who refused to respond to the violence questionnaire were not included in the analysis. Because the test cannot specify when the woman was infected, we do not know if infection occurred during or prior to the current pregnancy. We asked all women interested in participating in the study if they had ever tested positive for syphilis in the past, and none of them reported a history of syphilis or a positive syphilis test. All participants gave written informed consent. If a woman needed immediate assistance for IPV at any time during the interview, she was referred to local domestic violence services.

The survey included questions about education, occupation, marital status, and household economic conditions, languages spoken at home, and previous pregnancies that were not in the syphilis rapid test questionnaire. We also asked women questions about their male partners, specifically age, occupation, education, and years spent together. To assess IPV history, we used questions from the modified version of the Abuse Assessment Screen (AAS), which measures physical, sexual, and psychological violence and has been standardized to similar populations in Latin America.^23,24 The original screening tool, which consisted of five questions to assess each form of violence, has been found to detect abuse as effectively (65%–70% of all women with a history of IPV) as longer instruments.²⁴

For our survey, we selected four questions from the instrument to assess physical and sexual violence: (1) In the last year, were you hit, slapped, kicked, or otherwise physically hurt frequently? (2) Who was the person who hit you? (3) Within the last year, did someone force you to have sexual intercourse against your will? (4) Who was this person? We defined a positive case of IPV in the past 12 months if physical or sexual abuse or both were performed by the woman's partner or ex-partner (boyfriend, husband, lover) and negative for IPV if physical or sexual abuse was perpetrated by some other person (relatives or strangers) or if they declared not to be victims of physical or sexual abuse in that period. Women also were asked about frequency and severity of abuse.

We calculated the prevalence of IPV, syphilis diagnosis, and sociodemographic variable. We performed bivariate analysis using chi-square tests to assess associations among sociodemographic variables, positive syphilis serology, and exposure to IPV. To explore the correlates of positive syphilis serology and the potential association with a history of IPV and other sociodemographic characteristics, we performed logistic regression. In this model, the syphilis test result was the dependent variable and sociodemographic characteristics significant in bivariate analysis and IPV history were included as independent variables. Finally, we tested for colinearity among variables of interest, although this is less likely with categorical variables (all variables in this study) than continuous ones. We developed correlation tables to test correlations between variables. We then constructed a dummy variable for each categorical one. Only one variable—years living with partner—was correlated with two other independent sociodemographic variables. We excluded this variable from the multivariate analysis. We performed all analysis in SPSS 13.0 (Chicago, IL). This project was approved by the ethics committee at the hospitals in Bolivia and by the Population Council's institutional review board.

Treatment for women testing positive by rapid test

All participants in the study who tested positive by ICS rapid test had a confirmatory blood test that was processed at the national reference laboratory in La Paz. Treatment for women (as well as partners and infants) was offered after the test and was in accordance with WHO guidelines.²⁵ Women who tested positive for syphilis based on the ICS test were offered standard syphilis treatment, which consisted of three doses of benzathine penicillin (one dose per week for 3 weeks) given intramuscularly. Those who consented received the first dose immediately and were asked to return to the clinic for the other two. Women with penicillin allergy received the alternative treatment (500 mg of erythromycin four times per day for 7 days). A description of presumptive treatment for male partners is described in a forthcoming report (C. Díaz-Olavarrieta et al., unpublished observations). Any woman who gave birth during the study period also received free treatment for her infant according to WHO treatment guidelines/treatment of neonatal syphilis.²⁵ In cases where the results were positive, we requested informed consent from the mother to treat the child. A single dose of procaine benzylpenicillin or sodic penicillin 50,000 U/kg body weight, intramuscularly, for 10 days was administered. In cases of allergy to penicillin, erythromycin was administered.

Results

Between July 2004 and February 2005, 7572 women completed the survey. Of these 7572 women, 19.5% did not receive the ICS test. Additionally, 1.2% refused to respond to the violence questionnaire, giving a total of 6002 woman included in this study. Overall, 20.4% of the women reported sexual or physical abuse committed by their partners, and 79.6% women did not. Of the women reporting IPV in the past year, 52% experienced only physical violence committed by their partners, 29% experienced only sexual violence committed by their partners, 18% experienced both types of violence committed by their partners, 0.3% experienced sexual violence by their partner and physical violence by some other person, and 0.2% experienced physical violence by their partners and sexual violence by some other person (Table 1). We present the demographic characteristics of the study participants by IPV status in Table 2. IPV in the past year was significantly associated with several sociodemographic variables (p < 0.01): younger age, home region, lower education (secondary or less), employment as homemaker, consensual union, lower income, more years living with current partner, at least one previous pregnancy, and earlier gestation week at time of testing. Partner characteristics significantly associated with committing IPV in the past year were younger age (p < 0.05), employed (p < 0.05), and lower education (secondary or less) (p < 0.01). In general, most of the women who reported a history of IPV also were <24 years old, had less education, worked in the home, lived in a consensual union, had low economic status, and had more children. In this study, abused and nonabused women were demographically more similar than different. Partners' demographic characteristics were also more similar than different.

Table 1.

Violence and Syphilis Profile of Women Included in Study

	n	%
Total women interviewed	7572
Women excluded because they did not have rapid syphilis test	1478	19.5
Women excluded because they did not respond to violence questionnaire	92	1.2
Women selected	6002	79.3
Women who report physical or sexual violence committed by partner^a	1227	20.4
Experienced only physical violence	640	52.2
Experienced only sexual violence	358	29.2
Experienced both types of violence	222	18.1
Experienced only sexual violence by partner and physical violence by some other person	4	0.3
Experienced only physical violence by partner and sexual violence by some other person	3	0.2
Women who report no violence in past 12 months	4659	77.6
Women who report sexual violence committed by someone, but not physical violence	46	0.8
Women who report physical violence committed by someone, but not sexual violence	60	1.0
Women who report physical and sexual violence committed by someone	10	0.2
Total included in final sample	1227	100.0

In the analysis, we did not differentiate by type of IPV, simply whether it was reported or not.

Table 2.

Demographic Characteristics of Participants by Physical or Sexual Abuse Committed by Partners in Last 12 Months

	Intimate partner violence
	No		Yes
Characteristic	n	%	n	%
Total	4775		1227
Woman's age^**, years
<15	41	0.9	8	0.7
15–19	1047	21.9	238	19.4
20–24	1559	32.6	350	28.5
25–29	1033	21.6	270	22.0
30–34	656	13.7	206	16.8
35–39	314	6.6	108	8.8
40–44	112	2.3	46	3.7
45–49	13	0.3	1	0.1
Place of residence
Urban	4775	100.0	1227	100
Rural	0	0.0	0	0.0
Region^**
La Paz	2261	47.4	627	51.1
Santa Cruz	1648	34.5	469	38.2
Cochabamba	866	18.1	131	10.7
Chuquisaca	0	0.0	0	0.0
Education level^**
No schooling	74	1.5	25	2.0
Secondary or less	1701	35.6	581	47.4
High school	2325	48.7	537	43.8
Technical school or college or more	675	14.1	84	6.8
Occupation^**
Homemaker	2768	58.0	777	63.3
Employed	1404	29.4	387	31.5
Unemployed	603	12.6	63	5.1
Marital status^**
Single	965	20.2	147	12.0
Married	1313	27.5	294	24.0
Consensual union	2461	51.5	773	63.0
Divorced/separated/widowed	36	0.8	13	1.1
Economic status^**
Low income	2660	55.7	770	62.8
Middle income	2115	44.3	457	37.2
Years with current partner^**
<1 year	869	18.2	179	14.6
1–4	1579	33.1	461	37.6
5–9	726	15.2	217	17.7
10–14	371	7.8	125	10.2
15+	222	4.6	85	6.9
Not currently in union	1001	21.0	160	13.0
NE	7	0.1	0	0.0
Languages spoken at home
Only Spanish	2516	52.7	613	50.0
Spanish and some indigenous language	2251	47.1	613	50.0
Only indigenous languages	8	0.2	1	0.1
Previous pregnancies^**
None	2006	42.0	337	27.5
1–2	1774	37.2	478	39.0
3+	995	20.8	412	33.6
Gestation period^*
First trimester	1206	25.3	290	23.6
Second trimester	1861	39.0	480	39.1
Third trimester	1417	29.7	356	29.0
Unsure	291	6.1	101	8.2
Partner's age^*
15–19	346	7.2	82	6.7
20–24	1378	28.9	334	27.2
25–29	1285	26.9	288	23.5
30–34	816	17.1	222	18.1
35–39	436	9.1	149	12.1
40–44	269	5.6	78	6.4
45–49	101	2.1	38	3.1
50–54	41	0.9	10	0.8
55+	21	0.4	7	0.6
NE	82	1.7	19	1.5
Partner's occupation^**
Employed	4195	87.9	1139	92.8
Unemployed	484	10.1	76	6.2
NE	96	2.0	12	1.0
Partner's education level^**
Secondary or less	779	16.3	293	23.9
High school	2385	49.9	618	50.4
Technical school or college or more	937	19.6	122	9.9
NE	674	14.1	194	15.8
Syphilis test result^**
Negative	4576	95.8	1128	91.9
Positive	199	4.2	99	8.1

Significant at p < 0.05; ^**significant at p < 0.01.

Syphilis diagnosis was twice as frequent among abused women as nonabused women (8.1% vs. 4.2%), and this difference was statistically significant (p < 0.01). We present variables significantly associated with positive syphilis serology from bivariate analysis in Table 3. Variables associated with a positive syphilis serology were very similar to those associated with a history of IPV (all p < 0.01 unless specified): younger age (p < 0.05), women's education level (secondary or less), home region (p < 0.05), occupation (homemaker), marital status (consensual union), economic status (low), number of previous pregnancies (one or more), more years with current partner (p < 0.05), languages spoken at home (Spanish and an indigenous language), and earlier gestation week (p < 0.05). Male partner characteristics significantly associated with women's positive syphilis serology were employment status (currently employed, p < 0.05) and education level (secondary or less) (p < 0.01).

Table 3.

Demographic Characteristics of Participants by Syphilis Test Result

	Syphilis test result
	Negative		Positive
	n = 5704	%	n = 298	%
Woman's age^*
<15	49	0.9	0	0.0
15–19	1241	21.8	44	14.8
20–24	1804	31.6	105	35.2
25–29	1239	21.7	64	21.5
30–34	818	14.3	44	14.8
35–39	393	6.9	29	9.7
40–44	147	2.6	11	3.7
45–49	13	0.2	1	0.3
Place of residence
Urban	5704	100.0	298	100
Rural	0	0.0	0	0.0
Region^*
La Paz	2721	47.7	167	56.0
Santa Cruz	2023	35.5	94	31.5
Cochabamba	960	16.8	37	12.4
Chuquisaca	0	0.0	0	0.0
Education level^**
No schooling	88	1.5	11	3.7
Secondary or less	2110	37.0	172	57.7
High school	2761	48.4	101	33.9
Technical school or college or more	745	13.1	14	4.7
Occupation^**
Homemaker	3368	59.0	177	59.4
Employed	1683	29.5	108	36.2
Unemployed	653	11.4	13	4.4
Marital status^**
Single	1068	18.7	44	14.8
Married	1549	27.2	58	19.5
Consensual union	3041	53.3	193	64.8
Divorced/separated/widowed	46	0.8	3	1.0
Economic status^**
Low income	3224	56.5	206	69.1
Middle income	2480	43.5	92	30.9
Years with current partner^*
<1 year	1014	17.8	34	11.4
1–4	1932	33.9	108	36.2
5–9	888	15.6	55	18.5
10–14	462	8.1	34	11.4
15+	287	5.0	20	6.7
Not currently in union	1114	19.5	47	15.8
NE	7	0.1	0	0.0
Languages spoken at home^**
Only Spanish	3031	53.1	98	32.9
Spanish and some indigenous language	2664	46.7	200	67.1
Only indigenous languages	9	0.2	0	0.0
Previous pregnancies^**
None	2271	39.8	72	24.2
1–2	2131	37.4	121	40.6
3+	1302	22.8	105	35.2
Gestation period^*
First trimester	1444	25.3	52	17.4
Second trimester	2213	38.8	128	43.0
Third trimester	1675	29.4	98	32.9
Unsure	372	6.5	20	6.7
Partner's age
15–19	413	7.2	15	5.0
20–24	1636	28.7	76	25.5
25–29	1502	26.3	71	23.8
30–34	977	17.1	61	20.5
35–39	555	9.7	30	10.1
40–44	323	5.7	24	8.1
45–49	133	2.3	6	2.0
50–54	47	0.8	4	1.3
55+	26	0.5	2	0.7
NE	92	1.6	9	3.0
Partner's occupation^*
Employed	5061	88.7	273	91.6
Unemployed	543	9.5	17	5.7
NE	100	1.8	8	2.7
Partner's education level^**
Secondary or less	984	17.3	88	29.5
High school	2881	50.5	122	40.9
Technical school or college or more	1034	18.1	25	8.4
NE	805	14.1	63	21.1
Intimate partner violence^**
No	4576	80.2	199	66.8
Yes	1128	19.8	99	33.2

Significant at p < 0.05; ^**significant at p < 0.01.

In the multivariate analysis, a positive syphilis serology was significantly associated with IPV exposure within the past year, even after controlling for sociodemographic variables (OR 1.61, 95% CI 1.24–2.08) (Table 4). In addition, sociodemographic variables that remained significantly associated with positive syphilis serology were language spoken at home (Spanish and an indigenous language) (OR 1.83, 95% CI 1.3–2.45), more previous pregnancies (OR 1.74, 95% CI 1.23–2.46), gestation week at time of testing (first trimester) (OR 1.41, 95% CI 1.01–1.97), and lower education level among partners (high school or more OR 0.68, 95% CI 0.05–0.93). Interestingly, living in Cochambaba (OR 0.67, 95% CI 0.46–0.98) was significantly associated with lower odds of positive syphilis serology among women. No other sociodemographic variables remained independently associated with syphilis serology in multivariate analysis.

Table 4.

Logistic Regression Results for Women Who Had Syphilis

		95% CI
		Lower	Upper
Woman's age
≤19
20–24	1.35	0.91	2.00
25–29	0.96	0.59	1.56
30+	0.82	0.48	1.40
Region
La Paz
Santa Cruz	0.76	0.56	1.03
Cochabamba	0.67^*	0.46	0.98
Education level
No schooling
Secondary or less	0.78	0.39	1.55
High school	0.54	0.26	1.12
Technical school or college or more	0.45	0.18	1.16
Occupation
Homemaker
Employed	1.21	0.94	1.57
Unemployed	0.88	0.47	1.67
Marital status
Divorced/separated/widowed
Married	0.61	0.18	2.10
Consensual union	1.01	0.30	3.38
Single	1.16	0.34	4.03
Economic status
Low income	1.00
Middle income	0.80	0.61	1.04
Languages spoken at home
Only Spanish	1.00
Spanish and/or some indigenous language	1.83^**	1.37	2.45
Previous pregnancies
None
1–2	1.74^**	1.23	2.46
3+	2.42^**	1.55	3.76
Gestation period
First trimester
Second trimester	1.41^*	1.01	1.97
Third trimester	1.23	0.86	1.76
Unsure	1.06	0.61	1.86
Partner's age
≤24
25–29	0.95	0.66	1.35
30–34	1.18	0.78	1.78
35+	1.01	0.64	1.59
NE	1.47	0.67	3.25
Partner's occupation
Employed
Unemployed	0.87	0.51	1.48
NE	1.48	0.65	3.39
Education level
Secondary or less
High school	0.68^*	0.50	0.93
Technical school or college or more	0.57^*	0.34	0.96
NE	0.89	0.62	1.28
Intimate partner violence
No	1.00
Yes	1.61^**	1.24	2.08
Constant	0.04
Number of observations	(6002)

Significant at p < 0.05; ^**significant at p < 0.01.

Discussion

This study supports growing evidence of the association between current or past history of IPV and STIs among women of reproductive age.^{4,7,21,26,27,34,36} This is one of the first large-scale studies on co-occurrence of IPV and STIs in Latin America and the first conducted in Bolivia. Sexual and physical violence by a male partner have been shown to increase a woman's risk for STIs because power imbalance in the relationship may diminish her ability to demand condoms or refuse sex.^28,29 Other studies have suggested that increasing severity of IPV correlates with increasing HIV risk behavior.²⁶ Furthermore, previous studies have shown that the cycle of violence creates feelings of shame, isolation, and disempowerment, which make it extremely difficult for women to leave the relationship regardless of the health risks of remaining in the relationship.^5,29,30 Our study shows that positive syphilis serology is significantly associated with IPV in the past year in this clinic-based sample of pregnant women. Additional sociodemographic factors significantly associated with positive syphilis serology were place of residence, number of previous pregnancies, gestation week, language spoken at home, and male partners' education levels. Interestingly, sociodemographic variables significantly associated with positive syphilis tests are very similar to those significantly associated with a history of IPV, which suggests that both problems emerge from the same socioeconomic and cultural contexts. Overall, 20.4% of the participants reported exposure to IPV in the past year, and 8% tested positive for syphilis, compared with 4% who did not report IPV in the past year. The IPV prevalence in this sample is similar to figures reported in other Latin American countries.⁷ A study conducted in 27 prenatal clinics in Morelos, Mexico, reported that nearly 25% of pregnant women had suffered IPV during pregnancy.³² A study conducted in Peru using an ecological framework found that individual predictors of IPV were low education, early marriage, and history of family violence.³² Other studies have found significant associations among low education, young age, poverty, high parity, and IPV.^4,13,22,31 It is plausible that women's low education may make them vulnerable to IPV because they have less decision-making power, information resources, and financial autonomy.^4,33 Low educational attainment of both partners, or of the male partner only, has been shown to be a predictor of IPV in a multicountry study as well as in population-based and clinic-based studies in Mexico and Kenya, respectively.^7,34,35

A positive syphilis test remained significantly associated with IPV after controlling for sociodemographic characteristics of the woman and her partner. The syphilis prevalence found in this study also is within the range reported in other Latin American countries, between 1.7% and 11.5%.³¹ The additional significant correlates of a positive syphilis test are noteworthy as well. Having more previous pregnancies may suggest these women have had more exposure to unprotected sex and, therefore, syphilis and other STIs compared with women with no previous pregnancies. In contrast, that earlier gestation week was significantly associated with a positive syphilis test may suggest that these are women who have had more contact with prenatal care and more opportunities for early testing. Spanish plus an indigenous language spoken in the home suggest that these women are from indigenous families/communities. As indigenous communities continue to be marginalized in Bolivia, speaking an indigenous language at home may be an indicator of poverty and lack of access to health services, which can increase vulnerability to health problems, including syphilis and other STIs. Finally, the significant association between positive syphilis serology and lower education among male partners suggests that these men may have fewer resources and information to protect themselves and their partners from STIs compared with better educated men. Additional sociodemographic variables significantly associated with a positive syphilis test (and IPV in bivariate analysis) are important to consider and warrant future research. However, the most salient finding from this study, which has immediate programmatic implications for pregnant women, is the significant association between positive syphilis serology and history of IPV. These results suggest that providers should be particularly aware of a potential history of IPV when a pregnant woman tests positive for syphilis and also support the argument for screening for both IPV and syphilis during the antenatal care visit in order to detect and address both issues. This is especially important given that for many Bolivian women, prenatal visits may be their only interactions with the healthcare system.

This study has limitations. Because this was a clinic-based sample, the findings cannot be generalized beyond women attending antenatal clinics in Bolivia. The prevalence of IPV, syphilis, and associated risk factors may differ in pregnant women who do not get antenatal care, an estimated 30% of the pregnant population in Bolivia, and among pregnant women who opt out of syphilis testing.¹⁷ Specifically, women included in this study came from self-reported urban areas of residence and may have a different pattern of IPV or syphilis prevalence from that of women in rural areas. The reported syphilis prevalence may be slightly higher in urban areas simply because more women are likely to get tested there. However, women in the study included participants from three different regions of Bolivia (cities with both urban and periurban areas), ethnic groups, and educational and economic levels, which may have helped reduce selection bias based on area of residence. Because this was a cross-sectional study, we cannot infer causality between IPV and a positive syphilis test result. It is clear, however, that these conditions coexist in pregnant women, and a positive syphilis diagnosis could be an indicator of partner abuse. It is possible that there was underreporting of IPV in this sample, which is more common than overreporting because of the stigma associated with victimization and fear of partner retribution.¹⁵ However, we asked women about IPV history after they received their syphilis test results, which may have introduced recall bias and potential overreporting of IPV. Finally, we focused only on physical and sexual violence. Including psychological or emotional violence in the assessment may have increased the prevalence of reported IPV.²⁴ Other studies have shown that emotional violence increases during pregnancy, usually occurring with other types of violence, but this relationship should be further explored.^13,29,31

Conclusions

There are several implications for improving screening for IPV and STIs among pregnant women in Bolivia. First, the Bolivian health sector should take concrete steps to increase access to screening for both IPV and syphilis in antenatal clinic settings. Both are major problems for women in Bolivia and have the potential to escalate further. Early prevention, detection, and treatment of women and their male partners are the primary ways to reduce these epidemics. There are simple and cost-effective screening tools already available that can be standardized in clinical settings. The Bolivian government already has included the rapid syphilis test in the universal maternal and infant insurance program. Expanding access to these screening options for pregnant women will require ongoing political will as well as high-quality training, monitoring, and evaluation.¹⁸ Before scaling up IPV screening, there need to be mechanisms in place to refer women to IPV resources in the event a pregnant woman screens positive for partner violence.

Second, as this and other studies have shown, antenatal clinics are important locations for addressing a number of sensitive health issues, such as IPV and STIs. The antenatal clinic often is a woman's first encounter with the medical system and provides a less stigmatizing environment than an STI clinic. Efforts are needed to extend access especially to pregnant women in rural and other remote areas where antenatal care is not routine. Finally, despite domestic violence laws in many Latin American countries, including Bolivia, legal systems are ineffective, and enforcement is inconsistent.¹⁴ The persistence of syphilis and IPV in Bolivia reinforces the need for a more comprehensive national policy that strengthens enforcement of existing laws and engages the health sector. Our study demonstrated that the association between positive syphilis serology and IPV among pregnant women persists beyond the effects of other sociodemographic characteristics. Syphilis is not simply an issue among lower marginalized women in Bolivia but is also related to the problem of partner violence, which transcends socioeconomic and ethnic groups. In sum, our findings provide evidence for the importance of screening and management of these important and co-occurring problems among pregnant women as an important step to averting negative health impacts on both woman and child.

Footnotes

Acknowledgments

We thank Jorge Valencia for his contribution to the analysis in this article as well as all the women who participated in this study.

Disclosure Statement

The authors have no conflicts of interest to report.

References

Basile

, Arias

, Desai

, Thompson

. The differential association of intimate partner physical, sexual, psychological, and stalking violence and posttraumatic stress symptoms in a nationally representative sample of women. J Trauma Stress, 2004; 17:413–421.

Silverman

, Raj

, Mucci

, Hathaway

. Dating violence against adolescent girls and associated substance use, unhealthy weight control, sexual risk behavior, pregnancy, and suicidality. JAMA, 2001; 286:572–579.

, El-Bassel

, Witte

, Gilbert

, Chang

. Intimate partner violence and HIV risk among urban minority women in primary health care settings. AIDS Behav, 2003; 7:291–301.

Jewkes

. Intimate partner violence: Causes and prevention. Lancet, 2002; 359:1423–1429.

Campbell

, Jones

, Dienemann

et al. Intimate partner violence and physical health consequences. Arch Intern Med, 2002; 162:1157–1163.

Plichta

. Intimate partner violence and physical health consequences: Policy and practice implications. J Interpers Violence, 2004; 19:1296–1323.

Garcia-Moreno

, Jansen

, Ellsberg

, Heise

, Watts

. WHO multicountry study on women's health and domestic violence against women. Geneva: World Health Organization, 2005.

Coker

, Davis

, Arias

et al. Physical and mental health effects of intimate partner violence for men and women. Am J Prev Med, 2002; 23:260–268.

Heise

, Garcia-Moreno

. Krug

, Dahlberg

, Mercy

et al. Violence by intimate partners. World report on violence and health. Geneva: World Health Organization, 2002; 87–121.

10.

Johnson

, Leone

. The differential effects of intimate terrorism and situational couple violence. J Fam Issues, 2005; 26:322–349.

11.

Campbell

, Webster

, Koziol-McLain

et al. Risk factors for femicide in abusive relationships: Results from a multi-site case control study. Am J Public Health, 2003; 93:1089–1097.

12.

Garcia-Moreno

, Watts

. Violence against women: Its importance for HIV/AIDS. AIDS, 2000; 14,Suppl 3:S253–S265.

13.

Valladares

, Peña

, Persson

, Högberg

. Violence against pregnant women: Prevalence and characteristics. A population-based study in Nicaragua. Br J Obstet Gynaecol, 2005; 112:1243–1248.

14.

Díaz-Olavarrieta

, García

, Feldman

et al. Maternal syphilis and intimate partner violence in Bolivia: A gender-based analysis of implications for partner notification and universal screening. Sex Transm Dis, 2007; 34,Suppl 7:S42–S46.

15.

Asling-Monemi

, Peña

, Ellsberg

, Persson

. Violence against women increased the risk of infant and child mortality: A case-referent study in Nicaragua. Bull WHO, 2003; 81:17–18.

16.

Terris-Prestholt

, Watson-Jones

, Mugeye

et al.

Is antenatal syphilis screening still cost effective in sub-Saharan Africa?

Sex Transm Infect, 2003; 79:375–381.

17.

World Health Organization. WHO reproductive health indicators database: Country-specific data, Bolivia. June 30, 2006. www.who.int/reproductive_indicators/countrydata.asp?page=36. 2007 November 12.

18.

Garcia

, Tinajeros

, Revollo

et al. Demonstrating public health at work: A demonstration project of congenital syphilis prevention efforts in Bolivia. Sex Transm Dis, 2007; 34,Suppl 7:S37–S41Erratum on Sex Transm Dis 2007;34:938.

19.

McCarraher

, Bailey

, Martin

. The relationship between birth predictedness and violence during pregnancy among women in La Paz and El Alto, Bolivia. Matern Child Health J, 2005; 9:101–112.

20.

Deperthes

, Meheus

, O'Reilly

, Broutet

. Maternal and congenital syphilis programmes: Case studies in Bolivia, Kenya and South Africa. Bull WHO, 2004; 82:410–416.

21.

Southwick

, Blanco

, Santander

et al. Congenital and maternal syphilis in Bolivia: Prevalence and risk factors. Bull WHO, 1996; 79:33–42.

22.

Lien

, Tien

, Chanpong

et al. Evaluation of rapid diagnostic tests for the detection of human immunodeficiency virus types 1 and 2, hepatitis B surface antigen, and syphilis in Ho Chi Minh City, Vietnam. Am J Trop Med Hyg, 2000; 62:301–309.

23.

Díaz-Olavarrieta

, Ellertson

, Paz

, Ponce de Leon

, Alarcon

. Prevalence and correlates of battering among outpatients at a Mexican tertiary care institution. Soc Sci Med, 2002; 55:1589–1602.

24.

McFarlane

, Christoffel

, Bateman

, Miller

, Bullock

. Assessing for abuse: Self-report versus nurse interview. Public Health Nurs, 1991; 8:245–250.

25.

WHO. Sexually transmitted and other reproductive tract infections—A guide to essential practice, 2005 Geneva: World Health Organization www.who.int/reproductive-health/publications/rtis_gep/. 2008 October 30.

26.

Pronyk

, Hargreaves

, Kim

et al. Effect of a structural intervention for the prevention of intimate-partner violence and HIV in rural South Africa: a cluster randomized trial. Lancet, 2006; 368:1973–1983.

27.

Dunkle

, Jewkes

, Nduna

et al. Perpetration of partner violence and HIV risk behaviour among young men in the rural Eastern Cape, South Africa. AIDS, 2006; 20:2107–2114.

28.

Wingood

, DiClemente

, Raj

. Identifying the prevalence and correlates of STDs among women residing in rural domestic violence shelters. Womens Health, 2000; 30:15–26.

29.

El-Bassel

, Gilbert

, Krishnan

et al. Partner violence and sexual HIV-risk behaviors among women in an inner-city emergency department. Violence Vict, 1998; 13:377–393.

30.

Ellsberg

, Peña

, Herrera

, Liljestrand

, Winkvist

. Candies in hell: Women's experiences of violence in Nicaragua. Soc Sci Med, 2000; 51:1595–1610.

31.

Fleury

, Sullivan

, Bybee

. When ending the relationship does not end the violence. Women's experiences of violence by former partners. Violence Women, 2000; 6:1363–1383.

32.

Castro

, Peek-Asa

, Garcia

, Ruiz

, Kraus

. Risks for abuse against pregnant Hispanic women: Morelos, Mexico and Los Angeles County, California. Am J Prev Med, 2003; 25:325–332.

33.

Flake

. Individual, family, and community risk markers for domestic violence in Peru. Violence Women, 2005; 11:353–373.

34.

Pulerwitz

, Amaro

, De Jong

, Gortmaker

, Rudd

. Relationship power, condom use and HIV risk among women in the USA. AIDS Care, 2002; 14:789–800.

35.

Rivera-Rivera

, Lazcano-Ponce

, Salmeron-Castro

, Salazar-Martinez

, Castro

, Hernandez-Avila

. Prevalence and determinants of male partner violence against Mexican women: A population-based study. Salud Pública Mex, 2004; 46:113–122.

36.

Fonck

, Els

, Kidula

, Ndinya-Achola

, Temmerman

. Increased risk of HIV in women experiencing physical partner violence in Nairobi, Kenya. AIDS Behav, 2005; 9:335–339.