Intimate partner violence during pregnancy and associated mental health symptoms among pregnant women in Tanzania: a cross-sectional study
Correspondence: Ms B Mahenge, Department of Psychiatry and Mental Health, Muhimbili University College of Health Sciences, PO Box 65005, Dar es Salaam, Tanzania. Email firstname.lastname@example.org
Violence against pregnant women is a prevalent issue with severe health implications, especially during pregnancy. This study seeks to determine the prevalence of intimate partner violence against women during pregnancy and its associated mental health symptoms.
Cross-sectional survey conducted from December 2011 to April 2012.
Muhimbili National Hospital antenatal clinic in Dar es Salaam, Tanzania.
1180 pregnant antenatal care patients.
Trained interviewers conducted face-to-face standardised interviews with the women in a private room prior to their antenatal care appointment. (PTSD), anxiety and depressive symptoms were assessed through the Conflict Tactics Scale, the John Hopkins Symptom Checklist (25) and the Posttraumatic Diagnostic Scale.
Main outcome measures
The Conflict Tactics Scale, the John Hopkins Symptom Checklist (25) and the Posttraumatic Diagnostic Scale.
Of the 1180 women who were interviewed, 27% reported experiencing both physical and sexual intimate partner violence in the index pregnancy, with 18% reporting physical violence and 20% reporting sexual violence. After adjusting for the sociodemographic characteristics of women, women who experienced physical and/or sexual intimate partner violence during pregnancy were significantly more likely to have moderate PTSD (AOR 2.94, 95% CI 1.71–5.06), anxiety (AOR 3.98, 95% CI 2.85–5.57) and depressive (AOR 3.31, 95% CI 2.39–4.593) symptoms than women who did not report physical and/or sexual intimate partner violence during pregnancy.
About three out of ten women experienced physical or sexual intimate partner violence during pregnancy by an intimate partner, which was significantly associated with poor mental health symptoms. These rates are alarming, and justify training and education of antenatal care providers to raise awareness.
Provide feedback or get help You are viewing our new enhanced HTML article.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Violence against women, especially physical and sexual violence by an intimate partner, has been acknowledged as a serious global public health issue. Worldwide, the prevalence of lifetime experiences of physical and sexual intimate partner violence is estimated to range from 15 to 71%, with 1–28% of women reporting intimate partner violence during pregnancy.[1, 2] In Tanzania, the lifetime prevalence of intimate partner violence was reported to be 44% in 2010, with 9% reporting violence during pregnancy. An in-depth study of population-based surveys in Tanzania found that 7% of ever-partnered, ever-pregnant women in Dar es Salaam and 12% in Mbeya reported ever experiencing physical violence during pregnancy.
Intimate partner violence during pregnancy is not only concerning because of its high prevalence, it has also been associated with serious health effects and adverse fetal outcomes, including low birthweight, anaemia, preterm labour, low maternal weight, kidney infection, miscarriage, neonatal death and poor mental health.[5, 6] Although intimate partner violence at any period in life has been shown to be associated with women's adverse mental health,[7, 8] few studies have investigated its effect on women's mental health in Africa. In addition, despite studies showing a high prevalence of mental health disorders during pregnancy, for example, a study in Tanzania found that 40% of second to third trimester pregnant women had significant depressive morbidity, no study in sub-Saharan Africa has investigated the effect of intimate partner violence during pregnancy on women's mental health.
The aim of this study therefore is to determine the prevalence of intimate partner violence and its associated mental health symptoms among pregnant women attending an antenatal clinic in the largest national hospital in Tanzania.
The cross-sectional survey was conducted among women attending the antenatal care clinic of Muhimbili National Hospital, a national referral and university teaching hospital in Dar es Salaam, from December 2011 to April 2012. As a minimum of 80 pregnant women are seen in the clinic per day, only every fifth woman could be invited to participate in the study if she was physically fit and willing to participate. The research assistant, who was placed with the nurses who conducted the routine antenatal care check-up, recruited the women by asking them if they want to participate in a study on ‘Women's Health’. Non-participation was not monitored, but it was recalled that only approximately 15 women refused participation. The reasons women gave for non-participation included tiredness and not feeling like participating in this study; none of the women who declined cited illness as a reason for non-participation. Approximately 20 women had missing information in their questionnaire, as they were called for their doctor's appointment during the interview. Women who agreed to be interviewed on their own were accompanied to a separate room, where another research assistant read the consent form to the women. The consent form addressed the purpose, confidentiality and voluntary nature of the study. The research assistants also reiterated that participants were free to not answer any question or to end the interview if they felt uncomfortable, and encouraged them to ask questions. The questionnaire was administered in Kiswahili after the women gave verbal consent. At the end of the interview women received a complementary brochure on mental health services and the contact details of the study coordinator, whom they could contact in case of distress. Women who were perceived to be in immediate danger were referred to an existing intervention programme in the hospital. All research assistants were recently graduated medical doctors, and could therefore also answer medical questions that the women might ask, even if they were not related to the study. The research assistants who had prior research experiences received prior training in the adaptation of the instruments, with a special focus on asking women about their experiences of violence, and procedures for data collection. Their performance was monitored during the piloting of the questionnaire at Mwanayamala Hospital with 30 women.
The study protocol followed the WHO recommendations on safety when researching violence against women, and the study design, data collection and the analysis of findings complied with the Consolidated criteria for reporting qualitative research. It received ethical approval from Muhimbili University of Health and Allied Sciences Research and Publication Committee, and it was also supported by the Executive Director of Muhimbili National Hospital.
The structured questionnaire included women's sociodemographic information, such as age, marital status, level of education and form of employment. The Conflict Tactics Scale (CTS) was used to measure physical and sexual violence. It includes 12 questions on physical violence and six questions on sexual violence. The tool has been validated and used in Tanzania. To measure physical violence, women were asked if, in their index pregnancy, their partner threatened to hit or throw something at them, threw something at them that could hurt, twisted their arm or pulled their hair, shoved or pushed them, grabbed them, beat them up, threatened them with a knife or a gun, used a knife or a gun on them, punched or hit them with something that could hurt, choked them, burned or scalded them on purpose or kicked them. Women who reported that their partner threatened to hit or throw something at them, threw something at them that could hurt, twisted their arm or pulled their hair, shoved or pushed them or grabbed them, but who did not report any of the other acts of physical violence, were considered to have experienced moderate physical violence. Those who experienced any of the other acts of physical violence by their partner during pregnancy were considered to have experienced severe physical violence during pregnancy. To measure sexual violence during pregnancy, women were asked if their partner made them have sex without a condom, used force to make them have sex, used force to make them have oral sex, insisted on sex when they did not want to (but did not use physical force), used threats to make them have oral sex or used threats to make them have sex.
Post-traumatic stress disorder (PTSD) symptoms were captured by the Posttraumatic Diagnostic Scale (PDS), which first screened for experiences of traumatic events, which included severe forms of physical and sexual violence. When screened positive for a traumatic event, women were asked when the traumatic event happened, followed by questions on PTSD symptoms. The PDS captures symptoms such as hyper arousal, re-experiencing, arousal and hyper vigilance. Patients rated the frequency of symptoms in a week, which were coded by 0 (not at all), 1 (once a week or less), 2 (between two and four times a week) and 3 (five or more times a week). The numbers of symptoms endorsed were added up. According to the manual, symptoms were coded as none, mild (1–10), moderate (11–20), moderate to severe (21–30) and severe (36–51). None of the women reached the highest severity score. As only five women were rated to have moderate to severe symptoms, they were included in the moderate category. Frequencies were therefore run among women with no, mild and moderate symptoms only. The PDS has never been validated in Tanzania, but it has been used Uganda. The principal investigator and a medical doctor used forward-and-back translation to translate the PDS into Kiswahili.
To capture anxiety and depressive symptoms, the John Hopkins Symptoms checklist with 25 questions (HSCL 25) was used. The tool consists of ten questions for the anxiety symptoms and 15 questions for the depressive symptoms. A sum of means for each subscale was added up and the total sum of means was divided into three parts for each subscale. In the anxiety subscale the lowest score was ten and the highest was 31, with a mean score of 13.2. Anxiety symptoms were coded into three groups: scores of 11 and below were coded as low anxiety symptoms; scores of 12–14 were coded as medium symptoms of anxiety; and scores of 15–31 were coded as high symptoms of anxiety. For the depression subscale, the minimum score was 15 and the highest score was 50, with a mean score of 19.6. The depression subscale was divided into three groups: scores of 15–17 were coded as low symptoms of depression; scores of 18–21 were coded as medium symptoms of depression; and scores of 21–50 were coded as high symptoms of depression. The HSCL 25 has been validated among pregnant women who are HIV-positive in Tanzania.
All questionnaires were checked for completeness and consistencies daily, and were stored in a secure environment. The data entry and analysis was performed by the principal investigator using spss 17. The 20 questionnaires with missing information, as a result of women being called for their antenatal care appointment, were excluded from the analysis.
A woman was considered to have experienced physical and/or sexual intimate partner violence during pregnancy if she reported at least one act of violence by her partner during her index pregnancy. Cross tabulations and chi-square statistics were carried out to explore the association between intimate partner violence during pregnancy, women's sociodemographic characteristics, and PTSD, anxiety and depression. Women's socio-economic characteristics were also analysed using a multivariate logistic regression, controlling for the effect of sociodemographic characteristics on each other. To analyse the association between experiencing intimate partner physical and/or sexual violence during pregnancy with PTSD, anxiety and depressive symptoms, multivariate logistic regressions were conducted, controlling for a woman's age, marital status, level of education and employment status. Logistic regressions were also conducted to examine the associations between experiences of moderate intimate partner physical violence during pregnancy, severe physical intimate partner violence and sexual intimate partner violence only during pregnancy, as well as the cumulative effect of experiencing both physical and sexual intimate partner violence during pregnancy on symptoms of PTSD, anxiety and depression, controlling for the woman's age, marital status, level of education and employment status; P < 0.05 denotes significance, an odds ratio <1 represents a protective factor, whereas an odds ratio >1 was considered a risk factor.
A total of 1180 pregnant women attending antenatal care participated in this study. Participants were between 17 and 43 years old, with a mean age of 29 years. The majority of participants were married (83%), had secondary education (51%) and were employed or self-employed (37%).
The prevalence of intimate partner violence of physical and/or sexual violence in the index pregnancy was 27% (n = 315), with 18% (n = 209) reporting physical violence and 20% (n = 234) reporting sexual violence by their intimate partner. Moderate intimate partner physical violence during pregnancy only was reported by 12% of women (n = 138), and severe physical intimate partner violence during pregnancy was reported by 11% (n = 133); 11% (n = 128) reported both physical and sexual intimate partner violence during pregnancy. Young age, being married, low levels of education and being self-employed all significantly increased a woman's likelihood to experience physical and/or sexual intimate partner violence during pregnancy (Table 1).
Table 1. The sociodemographic characteristics and experiences of physical and/or sexual intimate partner violence of the participants in their index pregnancies (n = 1180)
| Age |
|35–43 years||170 (14.41%)||143 (16.53%)||27 (8.57%)||Ref.|| |
|26–34 years||715 (60.59%)||514 (59.42%)||201 (63.81%)||2.12||1.35–3.30|
|17–25 years||295 (25.00%)||208 (24.05%)||87 (27.62%)||1.98||1.20–3.27|
| Marital status |
|Married||971 (82.36%)||726 (84.77%)||245 (77.78%)||Ref.|| |
|Single||97 (8.23%)||63 (7.18%)||35 (11.11%)||1.32||0.85–2.05|
|Cohabiting||111 (9.41%)||76 (8.80%)||35 (11.11%)||1.57||0.99–2.49|
| Education level |
|College||183 (15.51%)||153 (17.69%)||30 (9.52%)||Ref.|| |
|Primary||391 (33.13%)||282 (32.6%)||109 (34.64%)||1.72||1.05–2.80|
|Secondary||606 (55.87%)||430 (49.61%)||176 (55.87%)||1.92||1.23–2.99|
| Employment status |
|Employed||440 (37.32%)||345 (39.93%)||95 (30.16%)||Ref.|| |
|Unemployed||299 (25.36%)||220 (23.35%)||80 (25.40%)||1.21||0.82–1.81|
|Self-employed||440 (37.32%)||300 (34.72%)||140 (44.44%)||1.48||1.07–2.06|
| Total ||1180 (100%)||865 (73.31%)||315 (6.70%)|| || |
As displayed in Table 2, 150 women (13%) had mild or moderate PTSD symptoms. Women who experienced physical and/or sexual intimate partner violence in the index pregnancy were significantly more likely to have mild (9%) and moderate (9%) symptoms of PTSD, compared with women who did not (7% and 3%, respectively). No woman had severe symptoms of PTSD. Even after adjusting for women's sociodemographic characteristics, women who experienced physical and/or sexual intimate partner violence during pregnancy were significantly more likely to report moderate symptoms of PTSD (AOR 2.94, 95% CI 1.71–5.06) than women who did not report physical and/or sexual intimate partner violence during pregnancy. The association with mild PTSD symptoms was not statistically significant (AOR 1.35, 95% CI 0.84–2.16).
Table 2. Physical and/or sexual intimate partner violence in the index pregnancy and its association with PTSD, anxiety and depressive symptoms (n = 1180)
| PTSD symptoms |
|None||1030 (87.3%)||772 (89.2%)||258 (81.9%)||6.29||0.016|
|Mild||92 (7.8%)||64 (7.4%)||28 (8.9%)|
|Moderate||58 (4.9%)||29 (3.4%)||29 (9.2%)|
| Anxiety symptoms |
|Low||418 (35.4%)||359 (41.5%)||59 (18.7%)||91.99||<0.001|
|Moderate||280 (23.7%)||223 (25.8%)||57 (18.1%)|
|High||482 (40.8%)||283 (32.7%)||199 (63.2%)|
| Symptoms of depression |
|Low||415 (35.2%)||359 (41.5%)||56 (17.8%)||60.22||<0.001|
|Moderate||105 (8.9%)||77 (8.9%)||28 (8.9%)|
|High||660 (55.9%)||429 (49.6%)||231 (73.3%)|
Results displayed in Table 2 also show that moderate levels of anxiety symptoms were present in 24% of women, with 41% having high levels of symptoms. Of the women who experienced physical and/or sexual intimate partner violence, 63% reported high levels of anxiety symptoms and 73% reported high levels of depressive symptoms. Both the levels of anxiety and the depressive symptoms were significantly lower among women who did not experience physical and/or sexual intimate partner violence during their index pregnancy. Among women with no experience of physical and/or sexual intimate partner violence during pregnancy, 33% reported high anxiety symptoms and 50% reported depression symptoms.
After controlling for sociodemographic factors, women who had experienced physical and/or sexual intimate partner violence in the index pregnancy were four times as likely to have a higher score of anxiety symptoms (AOR 3.98, 95% CI 2.85–5.57), and were three times as likely to have a higher score of symptoms of depression (AOR 3.31, 95% CI 2.39–4.59). Women who experienced physical and/or sexual intimate partner violence during pregnancy were not significantly more likely to have moderate levels of anxiety, compared with low levels, than women who did not experience physical and/or sexual intimate partner violence during pregnancy (AOR 1.48, 95% CI 0.98–2.21), but they were significantly more likely to report moderate symptoms of depression (AOR 2.20, 95% CI 1.30–3.69).
The analysis of the different levels of severity of physical and sexual intimate partner violence during pregnancy found that moderate physical intimate partner violence, sexual violence, and experiences of both physical and sexual intimate partner violence were significantly associated with moderate symptoms of PTSD, high symptoms of depression and high symptoms of anxiety, even after controlling for the socio-economic characteristics of the women. Moderate physical intimate partner violence during pregnancy also significantly increased a woman's likelihood to report medium symptoms of depression. Table 3 further displays that severe levels of physical intimate partner violence during pregnancy were significantly associated with a high level of depression and anxiety symptoms, adjusting for the socio-economic characteristics of the women.
Table 3. Associations between different severities of intimate partner violence during pregnancy with mental health outcomes
| PTSD symptoms |
|None||164||15.9||1|| || ||18||1.7||1|| || ||170||16.5||1|| || ||18||1.7||1|| || |
| Depression symptoms |
|Low||37||8.9||1|| || ||3||0.7||1|| || ||31||7.5||1|| || ||3||0.7||1|| || |
| Anxiety symptoms |
|Low||39||9.3||1|| || ||4||1.0||1|| || ||32||7.7||1|| || ||4||1.0||1|| || |
To our best knowledge, this is the first study in Tanzania that investigated the prevalence of intimate partner violence and its associations with mental health in antenatal care clinics in Tanzania.
The prevalence of physical and/or sexual intimate partner violence in the index pregnancy was found to be 27%, with a prevalence of 18% for physical violence and 20% for sexual violence. This rate is higher than in a previous population-based study, which found a prevalence of 7% for ever experiencing intimate partner violence during any pregnancy in Dar es Salaam. Reasons for the disparity are likely to be explained by the different conditions under which the studies took place. Whereas the current study sampled women attending antenatal care and asked specific and detailed questions about intimate partner violence during the women's current pregnancy, the previous study was a population-based survey asking all ever-pregnant women aged 15–49 years about their experiences of being physically abused during any of their pregnancies, regardless of when the woman was pregnant, using a single-item question restricted to physical intimate partner violence. However, the prevalence rates are comparable with those found in a systematic review on the prevalence of intimate partner violence during pregnancy in antenatal care studies in sub-Saharan Africa. This systematic review reported prevalence rates between 2.3 and 57.1% for physical and sexual violence, between 2.7 and 26.5% for sexual violence, and between 22.5 and 40% for physical violence during pregnancy.
The study further highlights that women who experienced physical and/or sexual violence in their index pregnancy are significantly more likely to report moderate symptoms of PTSD and higher levels of anxiety and depression symptoms in their index pregnancy than women who did not experience any intimate partner violence during pregnancy. The detailed analysis of different levels of violence during pregnancy showed that mental health effects of intimate partner violence during pregnancy are not limited to severe forms of physical intimate partner violence, or to the combined experiences of physical and/or sexual intimate partner violence, as even acts of moderate physical intimate partner violence during pregnancy or acts of sexual intimate partner violence during pregnancy by themselves can significantly increase women's symptoms of anxiety, depression and PTSD. These findings are supported by a number of antenatal care studies around the world.[17-21] This is especially concerning given that intimate partner violence during pregnancy is also known to have mental health effects beyond the duration of pregnancy. For example, longitudinal studies have shown that experiencing intimate partner violence during pregnancy increases a woman's risk of postpartum depression, and also increases the risk of the children's future behavioural and emotional problems.[23, 24]
Several limitations of this study must be taken into account. First, this is a cross-sectional study that limits interpretations regarding causality and temporality. It therefore remains unknown if the violence led to poor mental health outcomes or vice versa. Second, although great care has been taken to establish good rapport and provide participants with as much information as possible, experiences of intimate partner violence and mental health symptoms may have been underreported because of their sensitive nature, embarrassment, or fear of revealing violence or poor mental health. Biases might also have emerged because of the nature of the clinic women were recruited from for this study. As part of a national hospital in the capital city, the women who seek antenatal care here are more likely to live in urban areas and in some cases are referred to the clinic because of pregnancy complications. Furthermore, this study is only based on screening tools for mental health issues, but not on diagnostic tools and comprehensive procedures. This may explain the high percentage of women reporting depressive symptoms in this study. Although most of the tools used have been validated in Tanzania, the PDS tool has only been used in Uganda previously. One problematic aspect of the PDS tool was the low prevalence of traumatic events in this population, one of them being severe physical and sexual violence, which resulted in few women being screened for PTSD symptoms. Additional limitations include the lack of information on the women's trimester and on their experiences of other forms of abuse by their intimate partner, such as emotional and economic abuse.
About three out of ten women have experienced physical and/or sexual intimate partner violence in their index pregnancy, which is significantly associated with increased levels of PTSD, anxiety and depressive symptoms. These findings call for actions by healthcare professionals to integrate mental health services and programmes to address intimate partner violence in antenatal care, so as to improve women's and their unborn children's mental health and wellbeing. Although screening all women for mental health issues and intimate partner violence during pregnancy and offering them effective assistance would be the long-term goal, this is currently not achievable because of resource constraints, limited awareness on the issue of intimate partner violence during pregnancy and the lack of evidence on suitable antenatal care interventions in sub-Saharan African settings. The immediate aims, therefore, are to include intimate partner violence in the training curriculum of medical students and the continuing training of healthcare professions on intimate partner violence, with a specific focus on healthcare professional working in antenatal care.
Disclosure of interests
The authors declare that they have no conflicts of interest.
Contribution to authorship
All authors contributed to the article in a meaningful way: MB conducted the study; MB, LS and MJ contributed to the design of the study and its analysis; MB, LS and HS conducted the statistical analysis; MB and HS wrote the first draft of the article; MB, LS, MJ and HS contributed to the final version of the article. All authors have reviewed and approved the final article.
Details of ethics approval
The study received ethical approval from Muhimbili University of Health and Allied Sciences Research and Publication Committee, and was supported by the Executive Director of Muhimbili National Hospital: ref. no. MU/PGS/SAEC/Vol. VI/81, 3 November 2011.
BM was funded by the Tanzania Ministry of Health and Social welfare. HS is currently funded by a British Academy Postdoctoral Fellowship.
We want to acknowledge the important contributions and support from the members of the Department of Psychiatry and Mental Health, the Department of Obstetrics and Gynecology at Muhimbili National Hospital, the research assistants and all of the women who participated in this study.
Commentary on ‘Intimate partner violence during pregnancy and associated mental health symptoms among pregnant women in Tanzania: a cross-sectional study'
Over the past three decades, there has been an exponential increase in the scientific literature on violence against women (Jordan CE Violence Against Women 2009;15:393–419), with some research focused on violence during pregnancy. These studies have found that violence during pregnancy is an important concern because of the extent of the problem and the associated burden of suffering (Martin SL et al., VAWnet 2012, available at www.vawnet.org). Prevalence estimates of violence during pregnancy vary considerably, from approximately 1 to 40%, reflecting both true differences in the commonality of violence within various populations and methodological dissimilarities among studies. Such violence has been associated with women's physical and emotional health problems, as well as negative pregnancy and birth outcomes (Campbell JC Lancet 2002;359:1331–1336; Nwabuzor I et al. Journal of Interpersonal Violence, Published online before print January 11, 2013, doi: 10.1177/0886260512471080; Shah PS et al. J Womens Health 2010;19:2017–2031).
Most research on violence against women has been set within industrialised countries, but recent years have seen a growing number of studies from developing countries (Watts C et al. Lancet 2002;359:1232–1237). This global work is critical, as there are context-specific variations in some violence risk factors, the types and severity of violence perpetrated, and the prevalence of this violence.
The study of Mahenge and colleagues adds to this research by estimating the prevalence of violence and associated mental health symptoms among women attending an antenatal clinic in Tanzania, a country in which there are few such studies (for one example, see WHO 2005, WHO multi-country study on women's health and domestic violence against women: summary report of initial results on prevalence, health outcomes and women's responses). This study has several strengths, including: a fairly large sample size; standardised assessment instruments; descriptive and multivariable analyses; provision of information concerning mental health interventions to participants; and referrals to interventions for women in immediate danger from violence. The study also has limitations, most of which are described by the investigators: it is a clinic-based sample; there was no random selection of participants; there was no close monitoring of non-participation; and there were extremely small cell sizes for some analyses, compromising the statistical power. However, in my opinion, the research limitations are not so great as to put into question the overall study findings that pregnant patients in this Tanzanian clinic experience high levels of physical or sexual violence, and that this violence is associated with elevated levels of depression, anxiety and post-traumatic stress disorder symptoms. These findings are not surprising in that they are consistent with previous research from other settings; nonetheless, this study adds to the literature by shedding further light on the experiences of women in this under-studied geographic area.
Randomised controlled trials are beginning to show that various interventions are effective in reducing the levels of violence experienced by pregnant women, and in improving their mental health (Kiely M et al., Obstet Gynecol 2010;115:273–283; O'Reilly R et al. Trauma Violence & Abuse 2010;11:190–201; Zlotnick C et al. Arch Women's Ment Hlth 2011;14:55–65). It is hoped that Mahenge and colleagues will build on their work by collaborating with healthcare providers and others to develop, implement and evaluate interventions aimed at improving the lives of these women and children in Tanzania.
University of North Carolina at Chapel Hill, NC, USA