Intimate partner violence: A key correlate of women's physical and mental health in informal settlements in Nairobi, Kenya.

Globally, one billion people live in informal settlements, and that number is expected to triple by 2050. Studies suggests that health in informal settlements is a serious and growing concern, yet there is a paucity of research focused on health outcomes and the correlates of health in these settlements. Studies cite individual, environmental and social correlates to health in informal settlements, but they often lack empirical evidence. In particular, research suggests that high rates of violence against women (VAW) in informal settlements may be associated with detrimental effects on women's health, but few studies have investigated this link. The purpose of this study was to fill this gap by empirically exploring associations between women's experiences of intimate partner violence (IPV) and their physical and mental health. Data for this study were collected in August 2018 in Mathare Valley Informal Settlement in Nairobi, Kenya. A total of 550 randomly-selected women participated in surveys; however, analyses for this study were run on a subpopulation of the women (n = 361). Multivariate logistic regressions were used to investigate the link between psychological, sexual, and emotional IPV and women's mental and physical health. Results suggest that while some socioeconomic, demographic, and environmental variables were significantly associated with women's mental and physical health outcomes, all types of IPV emerged key correlates in this context. In particular, women's experiences of IPV were associated with lower odds of normal-high physical health component scores (based on SF-36); higher odds of gynecological and reproductive health issues, psychological distress (based on K-10), depression, suicidality, and substance use. Findings from this study suggest that policies and interventions focused on prevention and response to VAW in informal settlements may make critical contributions to improving health for women in these rapidly growing settlements.

Introduction

Around one-quarter of the world’s urban population live in informal settlements—and that number is rising [1]. An informal settlement is defined as a residential area that lacks: durable housing, sufficient living and public spaces, access to basic infrastructure and services like water and sanitation, and secure tenancy [1]. These settlements exist in urban contexts around the world in various forms, scales, and locations, and they go by a range of names (e.g., slums, squatter settlements, favelas, barrios bajos) [1]. The formation and expansion of informal settlements are driven by a range of interrelated factors including population growth; rural-urban migration; lack of affordable housing; poor policy, planning, and land management; economic vulnerability; under- and unemployment; discrimination and marginalization; displacement caused by conflict, natural disasters, and climate change; and, in the case of many sub-Saharan African countries, colonialism [1-3].

In sub-Saharan Africa (SSA) more than half of urban dwellers live in informal settlements, and this number will likely triple by 2050 [4]. In Kenya, specifically, approximately 56% of urban dwellers live in informal settlements [2], and the proportion is expected to reach two-thirds by 2030 [5]. Research conducted in Kenya suggests that residents of these settlements have the worst physical and mental health outcomes of any population in the country [6, 7]. Research also suggests women living in these settlements experience higher rates of violence [8], unemployment [9], and poverty [10] than women in other contexts—variables that evidence shows are linked to poor physical and mental health outcomes [11-13]. Some other individual, socioeconomic, political, social and environmental factors have also been identified as potential correlates of health in informal settlements [14-16]. To date, however, there have been few studies empirically investigating women’s physical and mental health and their correlates in informal settlements in Kenya. The purpose of this study was to empirically explore correlates of women’s physical health and mental health in a large informal settlement in Nairobi—focusing especially on intimate partner violence (IPV) as a potential correlate.

Women’s physical and mental health in informal settlements

Most of the health-related research carried out in informal settlements has focused on child health and mortality, reproductive health, and maternal and prenatal conditions [6, 15–17], with an emphasis on communicable diseases [16]. In one notable exception, Fink, Arku, and Montana [18] found that women living in informal settlements in Ghana report higher levels of health-related quality of life than people living other urban contexts. On the other hand, Gibbs, Govender, and Jewkes [19] found that 72% of women in informal settlements in South Africa reported moderate to high levels of depressive symptomology and 57.9% reported very high levels, compared to only 26.4% of women in a nationally-representative sample [20]. Some research suggests that, due to lack of access to adequate water, sanitation, and hygienic environments in informal settlements, female residents are also especially vulnerable to reproductive tract infections like vaginal and urinary tract infections (UTIs) [21-25]. One study, reporting on the prevalence of UTIs among women in informal settlements in Bangladesh, for example, found that 46% of women in the sample tested positive for a UTI [26].

Correlates of health in informal settlements

A growing body of literature cites low-quality housing and infrastructure, and a lack of access to water, sewage, garbage collection, health care, and other basic services as factors associated with poor health in these settlements [14, 15]. Individual demographic characteristics such as age, gender, and socioeconomic status have also been linked to poor physical health in informal settlements [14], and stress has been cited as a factor associated with poor mental health outcomes in these settlements [16]. Still other studies cite everyday hazards combined with rapid urbanization, insecurity, lack of social cohesion, climate change, and inadequate local government response to these hazards as factors associated with health risks in informal settlements [15].

Some research also suggests that lack of educational and employment opportunities and absence of public sector and law enforcement can foster violence and unrest in settlements and expose residents to greater health risks by limiting their ability to easily and safely navigate their environments and manage their health and daily tasks [6]. In particular, a 2008 study carried out in Nairobi, Kenya found that interpersonal violence was the second leading cause of mortality in informal settlements after infectious diseases (i.e., AIDS and tuberculosis) [27]. Violence against women (VAW)—IPV, in particular—can have deleterious effects on women’s health and well-being [28, 29]—effects that may be exacerbated in informal settlements [14].

Violence against women in informal settlements

VAW in lower- and middle-income countries (LMICs) is receiving growing scholarly attention, and several studies have documented prevalence rates of VAW in informal settlements [8, 17, 30–32], and several studies have linked VAW and crime to environmental and living conditions in these settlements [33-35]. Data collected in informal settlements in Nairobi in 2000 reported that 7% of female and male youth had been the victims of violence by a stranger and 16% had been the victim of violence by a partner or family member [36]—rates that have since risen. In a 2011 study carried out by Corburn and Hildebrand [17] in Mathare Valley—a large informal settlement in Nairobi, Kenya—68% of respondents reported violence as a common physical burden faced by women in this settlement. In 2012, Swart et. al. [8] recorded that 85% of women in Kibera, another large informal settlement in Nairobi, reported having experienced some form of violence in their lifetime compared to 39% of women in the general population in Kenya—conjecturing that these higher levels of violence in these settlements are potentially associated with lack of police and social services, limited access to healthcare, overcrowding, and the living conditions of residents. Finally, data collected from informal settlements in eThekwini Municipality, South Africa, in 2016–2017 showed that 65% of women reported physical and/or sexual violence in the past year [37].

When trying to account for the higher levels of VAW in informal settlements, scholars often employ socio-ecological frameworks to highlight the interplay of factors from multiple levels that shape its occurrence [38]. Some research links factors such as poverty in these settlements to high levels of VAW [33, 39, 40]. In addition, lack of durable housing, insecure tenancy, and lack of opportunity for education and employment may exacerbate incidence of ‘stress-induced violence’ in informal settlements [33]. Women with irregular, low-paid and casual jobs, like those likely to be found in these settlements, are also more likely to experience IPV [33]. Social instability, weak social ties, high levels of mobility, high unemployment for men, and armed conflict are other factors that have been associated with VAW in informal settlements.[33, 40]. In particular, findings suggests women in these settlements tend to have smaller friendship groups and are, consequently more isolated and less likely to respond to or exit from situations of IPV [33]. Some research suggests that psychological factors, such as “impaired masculinity,” often learned by observing violence as a normal behavior in childhood, may also be associated with IPV in informal settlements [33]. Still other socio-ecological theories suggest that IPV is a product of cross-cutting gender-power inequalities that are often exacerbated in economically and socially deprived settings such as these settlements [33, 40, 41]. Gender-based inequities at the structural level include fewer opportunities for women for education, employment, and political power [42]. This is accompanied by community and cultural norms that often reinforce the unequal status of women, with expectations that men play a dominant role within the household and community and VAW is acceptable [42, 43]. Individual and relationship level factors such as childhood experiences of violence, high levels of stress, beliefs and attitudes that condone violence, mental health issues and use of alcohol and drugs can exacerbate the risk of IPV occurring [38].

Violence against women and health

Numerous studies have explored associations between violence and poor health outcomes for women [28, 29, 44–46]. The majority of articles discussing the link between VAW and health are focused on IPV and mental health with fewer studies focused on physical health [29]. According to recent reviews, mental health outcomes associated with intimate partner violence include depression, posttraumatic stress disorder (PTSD), anxiety, suicide and self-harm, self-perceived mental health and psychological distress, sleep disturbance, and substance use disorders [28, 29]. Physical health outcomes associated with IPV include HIV/AIDS and other sexually transmitted infections; induced abortion; non-fatal injuries; fatal injuries (homicide); self-perceived limitations in physical; chronic physical health conditions such as chronic pain, fatigue, allergies, loss of hearing or eyesight, cardiovascular issues, diabetes, and gastrointestinal conditions; somatoform disorders and psychosomatic complaints; gynecological symptoms; memory loss, problems with concentration, and dizziness [28, 29].

Violence against women and health in informal settlements

While there is a growing body of literature focused on VAW in informal settlements and a well-established body of literature linking VAW with poor health, there remains a paucity of research focused on the associations between VAW and health in these settlements. This represents a critical gap in the research given that literature suggests women in these settlements experience much higher rates of violence than women in the general population [8]. The purpose of this study was to help fill this gap by exploring potential associations between women’s experiences of violence and their mental and physical health in informal settlements in Nairobi, Kenya.

Methods

The research was guided by the following research question: to what extent, if at all, is intimate partner violence (IPV), associated with women’s mental and physical health outcomes in a large informal settlement in Nairobi, Kenya when controlling for other potential correlates and covariates?

Data and sample

This study used data collected in Mathare Valley Informal Settlement (Mathare) in Nairobi, Kenya in 2018. Although the exact population and boundaries of the settlement are contested, some sources suggest Mathare is home to approximately 200,000 residents living on less than 3-square kilometers of land [47]. The settlement is divided into 11 villages. A sample of 550 women was selected using a stratified, random sampling technique. Geographic information systems (GIS) was used to randomly select 50 households from each of Mathare’s 11 villages and, subsequently, Kish methodology was used to randomly select one woman for each household [48]. To be included in this study, women had to be at least 18 years old and have resided in Mathare for a minimum of six months. Because a key variable of analysis was IPV, analyses for this study were run on a sub-population of the sample (n = 361). This subpopulation included all women who were married, living with a partner; in a relationship, but not living with a partner; recently separated, divorced, or widowed; or in a casual relationship (had a boyfriend) at some point in the 12 months leading up to the survey. All participants provided verbal informed consent. Verbal consent was deemed appropriate by the Internal Review Board at Rutgers University because a written consent document would have been the only identifying piece of information linking otherwise anonymous surveys to participants (US regulation 45 CFR 46.117). If participants consented, the survey interviewer signed a document affirming consent was given.

Surveys were conducted by female residents in Mathare who were trained on the principles of ethical research and rigorous data collection. Given the sensitive nature of the topic, women were also trained according to the World Health Organization’s (WHO) ethical and safety recommendations for research on VAW and sensitive topics [49, 50]. In accordance with these recommendations, investigators, field staff, and a local collaborative board agreed on safety protocols in the event that a participant reported violence and/or adverse mental health outcomes including depression and suicidality [51]. Participants who reported violence or adverse mental health were covertly provided referral to relevant services. In addition, a local counselor was contracted to be in the field throughout data collection. The study was approved by ethics committees at Rutgers, The State University of New Jersey, and the National Commission on Science, Technology, and Innovation in Nairobi, Kenya.

Measures

Mental health

Women’s perceived mental health was measured using a Swahili version of the 36-Item Short Form Health Survey (SF-36). The four domains that make up the mental health factor of the SF-36, i.e., vitality, social functioning, limited emotional functioning, and emotional well-being/mental health subscales were aggregated into a single mental health component score [52]. In a method used by other researchers [53], the mental health component scores were then dichotomized using the median as a cut-off. Participants with a mental health score below the median cut-off were defined as having a low mental health component score while patients with a summary score above or equal to the median cut-off were defined as having a normal-high mental health component score. Women’s psychological distress was measured using a Swahili version of the Kessler Scale of Psychological Distress (K10) [54], and those scores were also dichotomized using the median-cut-off method we used for the SF-36 mental health component scores. A Swahili version of the Patient Health Questionnaire-9 Depression Scale (PHQ-9) was used to measure major depressive disorder (MDD) [55]. Respondents were also asked questions about whether they had, in the last 2 weeks, experienced suicide ideation or had ever attempted suicide. Reponses to these variables were combined into a single ‘suicidality’ variable for which participants were given a score of ‘1 = has had suicidal thoughts in the past 2 weeks and/or has ever attempted suicide’ or ‘0 = had never attempted suicide nor had suicidal ideation in the past 2 weeks.’ Finally, participants were asked if they currently use alcohol and/or tobacco.

Physical health

Women’s perceived physical health was measured using the remaining four subscales of the Swahili version of the SF-36 including the physical functioning, role physical functioning, pain, and general health subscales [52]. Again, scores on these sub-scales were aggregated into a single physical health component score and then dichotomized using the median-cut-off method. Additionally, reproductive health and gynecological outcomes were measured using a series of binary yes/no questions about whether women had received medical diagnoses of common conditions such as urinary tract infections, vaginal infections, hemorrhoids, candidiasis/yeast infection, e-coli infection, vaginitis, and/or bacterial vaginosis in the previous 12 months. The frequencies for some of these conditions were not large enough to analyze separately; thus, a single “reproductive health issues” variable was created for which a woman was given a score of ‘1 = received medical diagnosis of at least one of these conditions in the last 12 months’ and ‘0 = did not receive medical diagnosis of any of these conditions in the last 12 months.’

Violence against women

This study used a modified version of the domestic violence module used in the Demographic and Health Surveys (DHS) [56]. The measure explores three types of violence (psychological, physical and sexual) perpetrated by an intimate partner. IPV scores were based on women's report of their experience of violence within the past 12 months. The emotional IPV sub-scale consisted of three items, the physical IPV scale had seven items, and the sexual IPV scale consisted of three items. Responses to each of the IPV items were dichotomized. A score of '1' on a sub-scale of violence (psychological, physical and sexual) reflected at least one affirmative response to questions on that sub-scale. A score of '0' reflected the absence of any affirmative response to questions on the sub-scale.

Sociodemographic and demographic characteristics

Finally, a number of sociodemographic variables including household income (measured as a respondent’s estimate of monthly income in KES), aware of household finances, education, marital status, number of children, age, employment and having a business were included in the models.

Environmental factors

Research has shown that lack of access to water and sanitation may be associated with poor health outcomes for women in informal settlements [17, 23, 25, 57, 58]. Access to water and sanitation are also used as alternative or additional sociodemographic or wealth measures in these settlements [59, 60]. Thus, women were asked about their primary source for drinking water and sanitation in 24-hr period, and these variables were included in the models.

Analysis strategy

Descriptive statistics provided information about the study sample and the prevalence rates of IPV in Mathare. In order to answer the key research questions in this study, multivariate logistic regressions were run in Stata statistical software (version 15) [61] to look at associations between women’s experiences of recent (in the last 12 months) IPV and their mental and physical health, controlling for other potential sociodemographic and environmental correlates. Two multivariate logistic regression models (Models 1 and 2) were run for each of the key physical health outcomes (1. dichotomized physical health component score from the SF-36 and 2. reproductive health issues) and each of the mental health outcomes (1. dichotomized mental health component score from the SF-36, 2. dichotomized psychosocial distress from the K10, 3. MDD, 4. suicidality, 5. alcohol and 6. tobacco use). In Model 1, only sociodemographic and environmental variables were included in the multivariate logistic regression. In Model 2, the IPV variables were added to the multivariate logistic regressions from Model 1. Wald tests were used to compare model fit between Models 1 (sociodemographic and environmental variables only) and 2 (IPV variables added to Model 1). Results from the Wald tests are presented in S1 Appendix.

In order to account for the clustered nature of the sample, the descriptive statistics and models were adjusted using the complex survey commands (svy) in Stata. Specifically, villages were used as a clustering variable in the svyset function in Stata. Additionally, because the analysis was conducted on a sub-population of the data (women in a long-term or casual relationship at some point in the past 12 months, n = 360), all analyses, including the descriptive statistics and logistic regressions were adjusted according to the svy and subpop() functions in Stata.

Results

Sample characteristics

A summary of the characteristics of the sample is provided in Table 1. According to these results, about 80% of the women in the sample were between the ages of 25 and 44 years and more than half of the women had at least completed primary education. Just over half the women were employed and almost three-quarters of the sample were married. Only 14% of the sample did not have children and close to half of the women had household incomes less than KES 10,000 per month (about $100). Average scores on the SF-36 measures were relatively high compared to women in informal settlements in Ghana [18], but were lower than scores for women in similar settings in Tanzania [52]. Almost 46% of women reported having been diagnosed with at least one reproductive-related health condition in the last 12 months. Almost 18% of women met the criteria for MDD, over 24% had experienced suicidal thoughts in the previous 2 weeks and/or had attempted suicide at least once in their lifetime. Just under two-thirds of women (66.2%) had experienced some form of IPV in the past year.

Table 1

Adjusted characteristics of sample (n = 361).

Socioeconomic demographic variables	Adjusted proportions
Age
18–24	13.3
25–34	44.6
35–44	35.5
45–54	6.6
Education
Less than complete primary school	42.9
Completed primary school, no secondary	21.6
At least some secondary school	35.5
Employed	49.9
Respondent has a business	34.4
Marital status
Married	69.0
Living with a man, not married	7.8
Regular partner, live apart	18.0
Casual boyfriend	5.3
Number of children
None	13.6
1–2 children	46.3
3–4 children	29.9
5 or more children	10.3
Monthly household income
Less than 10,000 KES/month	47.9
10,000–15,000 KES/month	31.0
More than 15,000 KES/month	21.1
Aware of household income	91.1
Has access to a toilet at all times	40.7
Primary drinking water source
Inside tap/well	7.2
Outside tap/well	36.0
Public tap/well	47.9
Vendor/tanker/burst pipe	8.9
Physical health outcomes
Normal-high physical health (from SF-36)	54.0
Reproductive Issues	45.7
Mental health
Normal-high mental health (from SF-36)	50.4
Normal-high psychosocial distress (K-10)	50.1
MDD	17.7
Suicidality	24.4
Alcohol use	21.1
Tobacco use	7.8
Intimate partner violence
Psychological violence	52.6
Sexual injury from IPV	42.1
Physical injury from IPV	64.5

Correlates of women’s health

Results from the multivariate logistic regression of women’s mental health outcomes on potential socioeconomic, environmental and IPV measures are summarized in Table 2. More detailed results including confidence intervals (95%) for both Models 1 and 2 are shown in Table A in S1 Appendix. Similarly, results from the multivariate logistic regressions of women’s physical health outcomes on potential correlates are summarized in Table 3 and detailed in Table B in S1 Appendix. Findings suggest that while some sociodemographic and environmental variables were significantly associated with mental and physical health outcomes, the IPV variables emerged as key correlates of women’s physical and mental health in Mathare.

Table 2

Summary of adjusted odds ratios from logistic regression of women’s mental health on their experiences of IPV and other potential correlates and covariates.

	Binary Mental Health Score	Major Depressive Disorder (MDD)	Suicidality	Alcohol Use	Tobacco Use	Binary Psychosocial Distress
Socio-economic variables
Monthly household income
10,000–15,000 KES/month	0.70	1.22	1.07	2.21*	8.80**	0.71
More than 15,000 KES/month	0.75	0.35†	1.02	3.86**	3.75†	0.55†
Aware of household finances	0.60	0.97	1.13	5.60†	5.04	1.74
Education
Completed primary school	1.08	1.19	1.84	0.83	1.57	1.16
At least some secondary	0.90	0.54	1.02	1.47	1.15	1.03
Marital status
Living with a man, not married	0.74	0.48	0.59	0.76	3.58	1.21
Regular partner, live apart	0.78	1.30	1.56	2.25*	1.96	1.03
Casual boyfriend	0.23*	2.14	1.84	3.21*	3.08	1.30
Number of children
1–2 children	0.82	0.63	0.51	2.26	1.07	1.14
3–4 children	0.37*	0.75	1.01	2.17	1.36	1.12
5 or more children	0.31*	0.81	0.81	1.88	3.96	1.23
Age	1.01	1.01	1.02	1.01	0.94	1.00
Respondent is employed	1.24	1.18	0.88	1.12	2.04†	1.02
Respondent has a business	0.68	2.84**	1.10	0.67	3.82*	1.44
Access to toilet
Has access to a toilet at all times	1.77*	1.21	0.97	0.52†	1.05	0.52*
Access to water
outside tap/well	3.17*	1.00	0.49	0.3†	0.59	0.56
public tap/well	3.46*	0.34	0.31†	0.69	0.50	0.41†
vendor/tanker/burst pipe	2.87	2.62	1.67	2.32	3.03	0.25*
Violence Variables
Intimate partner psychological violence	0.81	2.63*	2.36*	2.63**	3.77*	1.53
Intimate partner sexual violence	0.82	1.51	1.27	1.07	1.33	2.39**
Intimate partner physical violence	0.68	3.14**	3.74***	0.82	0.80	1.05

† p < .1

*p < .05

** p < .01

*** p < .001

Table 3

Summary of adjusted odds ratios from logistic regression of women’s physical health on their experiences of IPV and other potential correlates and covariates.

	Binary Physical Health Score	Reproductive Health
Socio-economic variables
Monthly household income
10,000–15,000 KES/month	0.75	0.82
More than 15,000 KES/month	1.32	0.74
Aware of household finances	2.44†	1.38
Education
Completed primary school	1.24	1.35
At least some secondary	0.92	0.67
Marital status
Living with a man, not married	0.63	0.78
Regular partner, live apart	0.78	1.28
Casual boyfriend	0.5	1.01
Number of children
1–2 children	0.79	1.08
3–4 children	0.5	0.62
5 or more children	0.65	0.65
Age	0.98	0.99
Respondent is employed	0.99	0.95
Respondent has a business	0.57*	1.18
Access to toilet
Has access to a toilet at all times	1.38	0.87
Access to water
outside tap/well	1.93	1.3
public tap/well	2.26	1.69
vendor/tanker/burst pipe	2.07	1.79
Violence Variables
Intimate partner psychological violence	1.7†	1.23
Intimate partner sexual violence	0.83	1.97*
Intimate partner physical violence	0.36***	0.95

† p < .1

*p < .05

** p < .01

*** p < .001

Associations between violence and women’s mental health

IPV was not associated with respondent’s mental health component score from the SF-36. Recent psychological IPV was associated with about two and one-half times the odds of meeting the criteria for MDD (OR = 2.6, p<0.01), suicidality (OR = 2.4, p<0.05), and alcohol use (OR = 2.6, p<0.05) and close to four times the odds of tobacco use (OR = 3.8, p<0.05). Sexual IPV was associated with almost two and one half times the odds of having normal-high psychosocial distress (OR = 2.4, p < .001). Finally, physical IPV was associated with about three times the odds of meeting the criteria for MDD and almost four times the odds of suicidality (OR = 3.7, p<0.001). Furthermore, results from the Wald tests comparing Models 1 and 2 (shown in Table B in S1 Appendix) suggest that including IPV variables in the dichotomized psychosocial distress (K10, MDD, suicidality, and alcohol use models significantly improves model fit, but does not improve model fit for the dichotomized mental health (SF-36) or tobacco use models.

Associations between violence and women’s physical health

Physical IPV was associated with 64% lower odds of having a normal-high physical health (SF-36). Sexual IPV was associated with just under two-times the odds of having been diagnosed with a recent reproductive health condition (OR = 1.97, p<0.05). Results from the Wald tests comparing Models 1 and 2 (shown in Table A in S1 Appendix) further suggest that including IPV variables in the dichotomized physical health (SF-36) and reproductive health conditions models significantly improves model fit.

Discussion

This study adds to the literature demonstrating poor mental and physical health outcomes for women in informal settlements in Nairobi. Specifically, the results suggest that while some sociodemographic and environmental variables were significantly associated with women’s mental and physical health outcomes, psychological, sexual, and physical IPV emerged as key correlates of women’s health in this environment. In particular, women’s experiences of IPV were associated with lower odds of having normal-high physical health; higher odds of gynecological and reproductive health issues; and higher odds of MDD, suicidality, normal-high psychosocial distress, and substance use. These findings potentially corroborate research that suggests women living in these settlements may have worse health outcomes compared to other populations in Kenya [6]. These results also suggest that the prevalence of IPV in these communities (66.2%) is higher than in the general population (39%) [8] and is likely linked to poor physical and mental health for these women—findings that have implications for health and community-based interventions and policy in these settlements.

Results from this study suggest that physical IPV is associated with low physical health scores and that sexual IPV is associated with gynecological and reproductive health issues for women in informal settlements in Nairobi. Both of these findings are consistent with research that suggests that physical IPV is associated with poor self-perceived physical health [see 28, 29] and that sexual IPV is associated with endogenous reproductive tract infections (RTIs) [62] and gynecological problems [63-65]. These findings, while perhaps not surprising, suggest there is a critical need for both IPV prevention and response interventions in these settlements.

Although there is a critical need for IPV prevention in these settlements, the overwhelming majority of evaluation research focused on the prevention of IPV has been carried out in the Global North [66, 67]. Physical and legal challenges, e.g. environmental hazards, land tenure disputes, and population densities, present different barriers to violence prevention in informal settlements [8, 68]. A small, but growing body of research focused on designing and testing violence prevention interventions in informal settlements in LMICs, mostly in South Africa [38, 40, 42, 69, 70], suggests that structural and behavioral interventions such as gender trainings and small-scale interventions, e.g. micro-loan programs, support groups, and job- and skills-trainings, might be effective strategies to help to empower women and serve as models for larger future projects in these settlements [8]. A review of IPV interventions in LMICs also found that 13 out of 16 studies that evaluated structural interventions found positive effects including decreased IPV [42]. A central facet of structural interventions is addressing the key areas of economic empowerment and gender norms, and a small yet increasing body of evidence points to the effectiveness of these approaches [40]. Although there is a need for more research focused on the specific causes of IPV in Mathare—factors that were beyond the scope of this study—small-scale, structural, and behavioral interventions, such as those tested in informal settlements in South Africa, may provide guidance about interventions that, if adapted to the local setting, might also be effective at helping to prevent violence, particularly IPV, in Mathare and similar settlements in Nairobi.

In addition to prevention, the high prevalence rates of violence in combination with the associated poor physical and mental health outcomes found in this study suggest there is also an urgent need for effective response strategies to help survivors of IPV. Specifically, there is a need to explore interventions to screen for, treat, and manage the physical and mental health outcomes associated with experiencing IPV and to provide safety planning and other advocacy strategies to prevent further occurrence of IPV for victims. According to some studies, healthcare utilization is higher among women who are currently experiencing or have recently experienced IPV [71, 72] while other research suggests that IPV is associated with foregone healthcare (being in need of healthcare services, but not seeking help) [73, 74]. This is likely the case for women experiencing IPV in informal settlements. While some of the women seek healthcare services, others may sustain injuries that affect their overall health status; yet, for a variety of reasons, including shame, embarrassment, fear of disclosure to others, or fear of partner retaliation, lack of access to adequate healthcare facilities, and lack of finances, do not seek healthcare and/or omit the potential cause of their physical and mental health symptoms if they do [73]. For this reason, there is likely a need for a multisectoral response to IPV in these settlements.

Recent research has highlighted the critical role of health systems in responding to and treating health outcomes associated with IPV [75]. The Kenyan Ministry of Health has already taken steps towards recognizing the crucial role of health systems in responding to IPV by developing national guidelines on the management of sexual violence in Kenya [76]; however, the guidelines do not yet address IPV nor recognize the unique healthcare requirements of specific populations, e.g., women in informal settlements. Given the findings in this study that suggest there are potentially serious physical and mental health consequences associated with IPV, there is a need to further involve the healthcare sector in developing appropriate interventions, practices, guidelines, and policies to screen for IPV and provide appropriate advocacy, treatment, and referrals for women seeking healthcare in response to IPV in informal settlements. Additionally, there is a need for further research to explore alternative, low-cost interventions to help expand healthcare services to women experiencing IPV in informal settlements who are unable or unwilling to seek healthcare in hospitals and clinics, this is particularly true for women who experience mental health challenges associated with IPV.

Results from this study suggest that all forms of IPV are associated with negative mental health outcomes and/or substance use. Psychological and physical IPV, in particular, were associated with several poor mental health outcomes including MDD, suicidality, and substance use. Findings also suggest that sexual IPV was associated with normal-high psychosocial distress. These findings are consistent with research from other contexts that suggests women’s experiences of violence, especially IPV, are associated with poor mental health and psychological distress [see 28, 29, 77]. Unfortunately, mental health services are often quite difficult to access in informal settlements. There are approximately 100 psychiatrists serving a population of 45 million in Kenya, and the majority of these, while located in urban areas, work in private practice [78]—charging fees well above those affordable to the majority of households in informal settlements. Consequently, despite high levels of IPV in informal settlements, services for survivors are minimal [79]. Recent research carried out in informal settlements in Nairobi, however, suggest there may be low-cost, effective behavioral interventions that can be carried out by lay community members to treat common mental disorders associated with IPV in these settlements [80, 81]. These findings suggest there may be feasible treatment options for women experiencing mental health disorders associated with IPV in informal settlements, but there is clearly a need for more research. There is, especially, a need to develop and test additional low-cost interventions which can be implemented by lay community members (e.g., community health volunteers, traditional healers, or non-medical pharmacy attendants) and/or to adapt and scale up existent interventions for variety of informal settlement contexts.

While the findings from this study provide an important contribution to the literature on IPV and women’s health in informal settlements in Kenya, the study has several limitations. The RTI and gynecological variables used in this study, for example, were measured by asking women whether or not they had received medical diagnoses for any of the conditions in the 12 months leading up to the study. There are several barriers that could prevent women from receiving a medical diagnosis for RTI or gynecological conditions in informal settlements. First, findings from a study carried out in informal settlements in Bangladesh [26] suggest that women may not be able to recognize the symptoms of common gynecological conditions and, therefore, not seek treatment. Second, women in informal settlements may not, due to a number of factors including lack of financial resources, seek medical advice even when they are aware of the symptoms. Finally, even if women recognize the symptoms and seek treatment, there are often limited healthcare services available in these settlements and those that are available may not have adequate diagnostic capacity [82]. Furthermore, in lieu of adequate healthcare services or the ability and willingness to seek treatment at a healthcare facility, many women in informal settlements seek treatment from local chemists or non-traditional healers who may not have the knowledge, capacity, or resources to make a medical diagnosis [82, 83]. Consequently, the rates of reported reproductive health conditions in this study are likely underestimating the actual prevalence rates of these conditions in informal settlements. These findings suggest there may be a need for interventions that help women to better recognize the symptoms of RTIs and gynecological conditions and for more accessible and adequate healthcare services as well as alternative prevention and treatment options for women living in these settlements. In addition, there may be a need for follow-up research that uses objective measures, e.g., urinalysis and urine and vaginal cultures, to capture more accurate prevalence rates of reproductive health conditions in these settlements. These recommendations also extend to other physical and mental health conditions beyond RTIs and gynecological conditions. Interventions to help women recognize or screen for symptoms of mental disorders, for example, may help women to seek treatment faster and improve their overall health and well-being. Some research has shown that mobile health (mhealth) interventions may help to expand healthcare services, e.g., screening and diagnostic tools, to areas with limited or poor access to health systems such as informal settlements [84], especially for women experiencing IPV [85].

Additional limitations of the study include, first, that the data are cross-sectional; thus, we cannot make any causal claims about the associations between women’s experiences of IPV and their health outcomes in informal settlements in Nairobi. Second, that this study focused on violence that occurred within the past 12 months; yet, we know that previous violence can be associated with poorer, long-term health outcomes as well. This limitation highlights the need for more longitudinal studies on VAW and health, especially in communities where poor health and rates of violence are high, e.g. informal settlements [28]. Third, that while this study used adapted versions of the violence measures from the domestic violence module of the DHS, a common measure for violence, it did not include measures for partner controlling behaviors, which might also affect women’s mental and physical health in this context—presenting a potentially important area for future research. Fourth, that while some of the health and violence measures used in this study have been validated in a Swahili context and in Kenya, specifically, others are new, modified, or have not. Fifth, for the purposes of these analyses the physical and mental health components of the SF-36 and the psychological distress scores (K10) were dichotomized. While some information may be lost in the dichotomization of continuous variables, the categorization of these variables allowed for a meaningful and accurate interpretation of the results. Finally, that while this study incorporated a variety of measures to explore different forms of IPV, these measures may not provide a comprehensive illustration of VAW in informal settlements in Kenya. For example, we did not include measures to explore financial abuse, which could play an important role in informal settlements where external factors such as poverty may exacerbate all forms, especially financial, abuse in intimate relationships [33, 39, 40].

Conclusion

The purpose of this study was to explore whether women’s experiences of recent IPV were associated with the physical and mental health of women in informal settlements in Nairobi, Kenya. The deprivations common to urban informal settlements including lack of adequate access to clean water and sanitation, housing, healthcare and emergency services, and transport, pose serious health risks to those who live there. A range of socio-ecological factors have also been associated with higher risk of women experiencing violence, particularly IPV, in these communities, which can have additional deleterious effects on women’s mental and physical health. Findings from this study suggest that women not only experience high rates of IPV, but that all forms of IPV are associated with poor physical and mental health outcomes for women in a large informal settlement in Nairobi. These findings suggest that policies and interventions focused on preventing VAW in informal settlements may make critical contributions to improving health for women in these rapidly growing environments. In addition, results from this study highlight a critical need for research focused on adapting existing and designing new interventions focused on low-cost, appropriate, and effective screening, advocacy, and safety planning for women experiencing IPV and treatment for related physical and mental health conditions in and outside of formal healthcare systems in informal settlements.

(DOCX)

Click here for additional data file.

(XLS)

Click here for additional data file.

31 Oct 2019

PONE-D-19-26708

Intimate partner and non-partner violence: Key correlates of women’s physical and mental health in informal settlements in Nairobi, Kenya

PLOS ONE

Dear Dr. Winter,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Importantly, please provide more theoretical backing including appropriate literature, as well as more detail on the statistical analysis that you conducted, in order to fulfill the respective PLOS One publication criterium. Further details are given in the individual reviews.

We would appreciate receiving your revised manuscript by Dec 15 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Hajo Zeeb

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

3. Thank you for stating in your Funding Statement:

"SCW - This study was partially supported by a global health seed grant from Rutgers Global Health Institute, Rutgers University. The funder did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.".

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

* Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript on violence against women in informal settlements of Nairobi brings attention to an important social problem that only promises to worsen in coming years. I applaud the authors for their work on this project and the effort undertaken to complete the data collection. However, this manuscript suffers from several major issues that preclude its publication as-is. Adding a theoretical basis, answering some serious methodological/analysis questions (and reanalyzing if needed), and a more robust Discussion would help warrant publication after additional review.

Introduction:

Major:

1. Considering PLOS one is targeted to a general audience, the authors might consider adding a brief primer on what informal settlements are and the major reasons for their existence Kenya. This would help ground non-experts in the need for this research.

2. The paper seem to lack any sort of theoretical grounding. What are the theoretical underpinnings for why violence is so high in informal settlements and what can be done about it? There are many potential theories others have used to situate similar studies- I suggest framing the paper using some sort of theoretical construct appropriate to the setting and weaving it throughout the Discussion as well.

Minor:

1. the second sentence uses semicolons to separate clauses instead of commas

2. fractions should be written as words in formal writing- i.e. “two-thirds”

3. Line 75: how can interpersonal violence be second behind both AIDS and TB? Doesn’t that make it third?

4. Line 82- VAW in the developing world has been receiving serious scholarly attention for decades- I don’t think you can say it’s “beginning”

Methods:

Major:

1. Why include only women currently in a relationship? Women who are separated, divorced, widowed, or in casual relationships within the last 12 months could also have experienced IPV. Restricting the sample to women currently in relationship is not uncommon, but this should be listed in the limitations if the team cannot include women who may have experienced IPV in the past year but are not currently in a relationship

2. The use of “medical diagnosis” for the physical health variable(s) is problematic when studying informal settlements since, by the team’s own admission, medical care is scant in these areas. While other methods of measuring these issues may have been unavailable, it stands to reason that a limitation of the study is an underrepresentation of these diagnoses since many women may be living with them but not have had the chance to have them diagnosed

3. The WHO definition of IPV also includes controlling behaviour as a major form of IPV. Is there a reason the researchers did not include controlling behaviour as a form of violence since it is included in the DHS? .

4. Why did the team choose to include both IPV and NPV? The drivers for these and theoretical underpinnings can be quite different- especially in developing country settings. It would be important to justify the inclusion of both kinds of violence.

5. The analysis strategy needs more detail to ensure the analysis methods are correct. Specifically:

a. How did the authors set up the svyset command- specifically, how did the authors take account of weighting, clustering, and stratification to ensure correct point estimates? If the survey set command does not correctly account for these things and multilevel modeling techniques are not used, the results could be largely inaccurate.

b. When models are run, what are the control variables included in each model? How many models were run total?

Minor:

1. Line 210: “thus” should be capitalized

2. Line 217 the use of “stratified” should be replaced with “clustered” or “hierarchical” to avoid confusion with pre-determined strata in the sampling technique

Results:

Major:

1. Why did the authors choose to show only the significant results in the Tables?

2. Do these results reflect adjusted b-coefficients and odds ratios? If so, this should be reflected in the title of the Table

Discussion

Major:

1. The Discussion is devoid of any real connection to why informal settlements are such hotspots for poor health outcomes and violence. Why the discussion of help-seeking is interesting, it is a small part of the larger picture for why this study was conducted, why health outcomes are so poor in informal settlements, and what can be done to reduce violence in these areas. As it stands, the discussion is largely a restating of the results and does little to add to where the literature and science should go next.

2. Line 387- policies and interventions like what?

3. Line 390- is the healthcare sector really the best place to situate future interventions considering the extensive strains and thin coverage already experienced in informal settlements? Can the authors provide some evidence for this suggestion?

Reviewer #2: This is an important contribution to the literature and with some targeted work can be revised for the journal.

1. The gaps in the literature are often stated broadly, such as “Few studies have empirically examined the correlates of health in informal settlements.” I think this needs to be toned back throughout the Introduction.

2. In line 85, please also cite:

Hatcher, A. M., H. Stockl, R. S. McBride, M. Khumalo and N. Christofides (2019). "Pathways From Food Insecurity to Intimate Partner Violence Perpetration Among Peri-Urban Men in South Africa." Am J Prev Med 56(5): 765-772.

Baiocchi, M., R. Friedberg, E. Rosenman, M. Amuyunzu-Nyamongo, G. Oguda, D. Otieno and C. Sarnquist (2019). "Prevalence and risk factors for sexual assault among class 6 female students in unplanned settlements of Nairobi, Kenya: Baseline analysis from the IMPower & Sources of Strength cluster randomized controlled trial." PLoS one 14(6): e0213359.

3. I’m not certain the section starting on line 99 is necessary. Instead of a whole paragraph, this could perhaps be summarized in 2-3 sentences?

4. In line 147, is it “at least 50 households” or can the authors provide the exact number?

5. With this sampling technique, can the estimates be considered population-based? If so, that would be an important distinction of this study vs the extant literature on VAW in informal settlements. Would highlight that throughout

6. I think the DHS items on VAW are accessible widely and do not need to be listed in full in the text. However, perhaps a table in a Supplementary Appendix would be valuable?

7. What is the timeframe for the VAW questions (ever or past 12-months)? How was household income assessed? And how about access to water and sanitation?

8. How were the linear and logistic models built? Did the linear regression outcomes meet assumptions of normality or did they require transformation?

9. In the ethics section please also mention how researchers were trained to assess mental health and violence exposure, and what steps were taken in the cases of current VAW or major depression / suicidality.

10. This is a personal preference (so feel free to ignore) but perhaps Table 1 socio-demographic variables could be dichotomized to save space?

11. Given that non-partner violence has weaker association with the outcomes of interest, I would consider only examining IPV in this paper, and combining the IPV variable to be “ever physical and/or sexual IPV”.

12. To simplify the analysis and make more of a statement about how violence frames outcomes for women, could you combine all the SF-36 items into a single continuous outcome (“physical health”) and say “Any sexual or reproductive health problems”? That way, Table 3 would only have two columns and be easier to interpret. Similarly, I might be tempted to drop Alcohol and Tobacco in Table 2, combine suicidal ideation and attempts into one column (“suicidality”) and report on the Short Form Health survey as one single continuous outcome.

13. Would start Discussion with your own findings, rather than citing literature. You have already made the case for the need for new research, so don’t need to do it again here. First paragraph should sum up your key findings.

14. In line 343 would cite:

Hatcher, A. M., A. Gibbs, R. Jewkes, R. S. McBride, D. Peacock and N. Christofides (2019). "Effect of childhood poverty and trauma on adult depressive symptoms among young men in peri-urban South African settlements." Journal of adolescent health 64: 79-85.

15. The Conclusion could be a bit shorter, and may not require citations (again, this is stylistic). Be sure there are no new ideas introduced here and it’s rather a summary of what’s already been stated.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Abigail M Hatcher

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

19 Dec 2019

Response to Reviewers’ Comments

We want to thank both the reviewers for the detailed comments and recommendations. We have made a number of changes to the paper. Please note: line numbers referenced here refer to the clean document (with all track-changes accepted).

Reviewer #1

Introduction

Comment 1:

Considering PLOS one is targeted to a general audience, the authors might consider adding a brief primer on what informal settlements are and the major reasons for their existence Kenya. This would help ground non-experts in the need for this research.

Response:

We have provided a very brief primer on informal settlements in the introduction section with specific reference to some resources for additional reading. See lines 26-36.

Comment 2:

The paper seems to lack any sort of theoretical grounding. What are the theoretical underpinnings for why violence is so high in informal settlements and what can be done about it? There are many potential theories others have used to situate similar studies- I suggest framing the paper using some sort of theoretical construct appropriate to the setting and weaving it throughout the Discussion as well.

Response:

The reviewer makes an important point. To date there is still relatively little research focused on violence against women in informal settlements, particularly the theoretical underpinnings of why violence is high in these settlements. That being said, we have provided a discussion of the theory presented in literature thus far to help explain the high prevalence of violence in informal settlements in the introduction (see lines 99-121) and throughout the discussion (see, for example, lines 326-344).

Comment 3:

The second sentence uses semicolons to separate clauses instead of commas

Response:

We have made the corrections (see lines 27-29).

Comment 4:

Fractions should be written as words in formal writing- i.e. “two-thirds”

Response:

We have made the corrections throughout.

Comment 5:

Line 75: how can interpersonal violence be second behind both AIDS and TB? Doesn’t that make it third?

Response:

The measures used in the burden of disease study utilized a single category for AIDS and Tuberculosis. As they explain it, “AIDS and tuberculosis were combined in the analysis because about 35% of deaths from AIDS and tuberculosis were due to a probable combination of HIV/AIDS and TB, and hence were coded as "AIDS with TB," while seven cases were coded as "unspecified TB/AIDS." We tried to rephrase the reference in such a way as to reduce confusion for readers (see line 77-79).

Comment 6:

Line 82- VAW in the developing world has been receiving serious scholarly attention for decades- I don’t think you can say it’s “beginning”

Response:

We have made a correction to help clarify the statement. Please see line 83 for revision.

Methods

Comment 7:

Why include only women currently in a relationship? Women who are separated, divorced, widowed, or in casual relationships within the last 12 months could also have experienced IPV. Restricting the sample to women currently in relationship is not uncommon, but this should be listed in the limitations if the team cannot include women who may have experienced IPV in the past year but are not currently in a relationship

Response:

In the study all women who reported having had any type of relationship, long-term or casual (e.g., married, living with a partner; in a long-term relationship; separated, divorced, or widowed within the last 12 months, or in casual relationship), at any point in the 12 months leading up to the survey were asked question related to violence. All of these women (n=361) were included in the sample. We have rewritten the language in the manuscript to more accurately reflect the measure (see lines 162-166)

Comment 8:

The use of “medical diagnosis” for the physical health variable(s) is problematic when studying informal settlements since, by the team’s own admission, medical care is scant in these areas. While other methods of measuring these issues may have been unavailable, it stands to reason that a limitation of the study is an underrepresentation of these diagnoses since many women may be living with them but not have had the chance to have them diagnosed

Response:

We have added several statements and citations to the discussion of limitations associated with using the self-report, medical diagnosis variables to capture information about women’s experiences of reproductive tract and gynecological conditions. Please refer to lines 394-423.

Comment 9:

The WHO definition of IPV also includes controlling behaviour as a major form of IPV. Is there a reason the researchers did not include controlling behaviour as a form of violence since it is included in the DHS?

Response:

Unfortunately, the partner controlling behavior questions from the DHS module were not included in the measures of IPV used in this study. The survey was meant to provide preliminary information on the state of women’s physical and mental health, including their experiences of violence in informal settlements. Not all of the questions could be retained in the final version of the survey. Women from the community were involved in the process of creating and cutting down the survey. We have included a statement in the limitations of this study to acknowledge that the controlling behavior by partner questions from the DHS domestic violence module were not included in this study. See lines 430-434.

Comment 10:

Why did the team choose to include both IPV and NPV? The drivers for these and theoretical underpinnings can be quite different- especially in developing country settings. It would be important to justify the inclusion of both kinds of violence.

Response:

We appreciate all reviewers’ suggestion to focus on IPV. We have, therefore, narrowed the scope of the study to include only the IPV correlates.

Comment 11:

The analysis strategy needs more detail to ensure the analysis methods are correct. Specifically:

a. How did the authors set up the svyset command- specifically, how did the authors take account of weighting, clustering, and stratification to ensure correct point estimates? If the survey set command does not correctly account for these things and multilevel modeling techniques are not used, the results could be largely inaccurate.

Response:

We appreciate the recommendation for more detail about the analytic strategy and the use of the svy functions in Stata. We have provided a much more detailed description of our models, analytic strategy, and use of the svy functions in our analysis strategy section (see lines 243-248). We are happy to provide additional information if there are still gaps in our description.

b. When models are run, what are the control variables included in each model? How many models were run total?

Response:

We have provided a more detail description of the exact models that were run in the analytic strategy. We have also included detailed results from all of the models in the Appendix so that readers have more information about the composition of the models, model fit, and findings (see Tables A and B in Appendix A).

Comment 12:

1. Line 210: “thus” should be capitalized

Response:

The correction has been made (see line 227).

Comment 13:

Line 217 the use of “stratified” should be replaced with “clustered” or “hierarchical” to avoid confusion with pre-determined strata in the sampling technique

Response:

The correction has been made (see line 243).

Results

Comment 14:

Why did the authors choose to show only the significant results in the Tables?

Response:

We chose to present on the significant results in the table in the body of the paper because they were much easier to read; however, we modified the tables to include a summary of both significant and non-significant results. We have also included tables with all key information, including confidence intervals in Appendix A. Hopefully this will ensure that all readers have access to the relevant information. Please see Appendix A.

Comment 15:

Do these results reflect adjusted b-coefficients and odds ratios? If so, this should be reflected in the title of the Table

Response:

All descriptive statistics are adjusted for the clustered nature of the data and the subpopulation analysis. All coefficients are adjusted for the clustered nature of the data and the subpopulation analysis as well as the covariates in the models. These are now reflected in the titles of Tables 1, 2, and 3 and in the column headings in Tables A and B in the Appendix.

Discussion

Comment 16:

The Discussion is devoid of any real connection to why informal settlements are such hotspots for poor health outcomes and violence. Why the discussion of help-seeking is interesting, it is a small part of the larger picture for why this study was conducted, why health outcomes are so poor in informal settlements, and what can be done to reduce violence in these areas. As it stands, the discussion is largely a restating of the results and does little to add to where the literature and science should go next.

Response:

We appreciate this comment. We have, at the request of both reviewers, rewritten a large portion of the discussion and conclusion sections of the paper. We have made considerable effort to talk more extensively about the socio-ecological factors that might contribute to poor health and higher rates of violence against women in informal settlements in the introduction/background/literature review section of the paper (see lines 66-121). In the discussion, we have focused on providing a more detailed set of recommendations about what sorts of policies and interventions might be feasible and/or effective in informal settlements for preventing and responding to IPV and its related physical and mental health consequences for women. Please see revised discussion (lines 307-442).

Comment 17:

Line 387- policies and interventions like what?

Response:

We have tried to provide much more detail about potential policy and interventions strategies that might start to help address the high VAW prevalence rates and related mental and physical health consequences for women in informal settlements throughout the discussion. Please see revised discussion (lines 307-442).

Comment 18:

Line 390- is the healthcare sector really the best place to situate future interventions considering the extensive strains and thin coverage already experienced in informal settlements? Can the authors provide some evidence for this suggestion?

Response:

We have provided a more detailed discussion about the rationale for including the health sector in preventing and responding to IPV in informal settlements (see, for example, lines 347-375 and 413-422). We have also included additional discussion about alternative strategies to help expand treatment and resources to women who cannot access health clinics and hospitals (see, for example, lines 372-395 and 422-425).

Reviewer #2:

Comment 1:

The gaps in the literature are often stated broadly, such as “Few studies have empirically examined the correlates of health in informal settlements.” I think this needs to be toned back throughout the Introduction.

Response:

We thank the reviewer for picking up on the redundancy. We removed a number of these sentences from the background section of the paper including in the introduction, the women’s physical and mental health in informal settlements, correlates of health in informal settlements, violence against women in informal settlements, and the violence against women and health in informal settlements sections.

Comment 2:

In line 85, please also cite:

Hatcher, A. M., H. Stockl, R. S. McBride, M. Khumalo and N. Christofides (2019). "Pathways From Food Insecurity to Intimate Partner Violence Perpetration Among Peri-Urban Men in South Africa." Am J Prev Med 56(5): 765-772.

Baiocchi, M., R. Friedberg, E. Rosenman, M. Amuyunzu-Nyamongo, G. Oguda, D. Otieno and C. Sarnquist (2019). "Prevalence and risk factors for sexual assault among class 6 female students in unplanned settlements of Nairobi, Kenya: Baseline analysis from the IMPower & Sources of Strength cluster randomized controlled trial." PLoS one 14(6): e0213359.

Response:

We have incorporated these additional resources into the paper as suggested. We appreciate that the reviewer brought these important articles to our attention.

Comment 3:

I’m not certain the section starting on line 99 is necessary. Instead of a whole paragraph, this could, perhaps, be summarized in 2-3 sentences?

Response:

We have cut the section down to 4 sentences (see lines 123-134). We are happy to make additional adjustments if that will increase the readability of the paper.

Comment 4:

In line 147, is it “at least 50 households” or can the authors provide the exact number?

Response:

We have made the correction (see lines 157-160).

Comment 5:

With this sampling technique, can the estimates be considered population-based? If so, that would be an important distinction of this study vs the extant literature on VAW in informal settlements. Would highlight that throughout

Response:

While the estimates may be considered population based, there may be additional sampling considerations that would need to be taken into account to truly assume these are population-based. Thus, for the purposes of this paper, we are not claiming that they are population-based.

Comment 6:

I think the DHS items on VAW are accessible widely and do not need to be listed in full in the text. However, perhaps a table in a Supplementary Appendix would be valuable?

Response:

We have cut down the description of VAW measures considerably and cited the DHS domestic violence module. Please see lines 210-218.

Comment 7:

What is the timeframe for the VAW questions (ever or past 12-months)? How was household income assessed? And how about access to water and sanitation?

Response:

The timeframe is IPV in the past 12-months. We have noted this in the measures section (lines 212-213). Measures for household income, access to sanitation, and primary drinking water source were described in the measures section (see lines 220-221 and 226-227).

Comment 8:

How were the linear and logistic models built? Did the linear regression outcomes meet assumptions of normality or did they require transformation?

Response:

We have provided a more detailed description of our analytic strategy to better explain the process of building the multivariate models used to analyze the data in this study (see lines 229-248). Original versions of some of the SF-36 outcomes did require transformations; however, based on the reviewer’s recommendations to combine/collapse some of the physical and mental health outcome variables, we combined four sub-scales of the SF-36 into an aggregate physical health component score variable and the remaining four sub-scales of the SF-36 into a separate aggregate mental health component score variable as is commonly done with the SF-36. Additionally, these component scores, as well as the remaining continuous variable (psychological distress measured with the K-10 scale), were dichotomized using a median-cut-off method described in detail in the methods section (see lines 179-190 and 198-202) and employed by other researchers using similar methodological approaches with these health outcome variables [1].

Comment 9:

In the ethics section please also mention how researchers were trained to assess mental health and violence exposure, and what steps were taken in the cases of current VAW or major depression / suicidality.

Response:

We have added more detail about the protocols for handling instances where women reported violence and/or adverse mental health outcomes (see lines 167-177).

Comment 10:

This is a personal preference (so feel free to ignore) but perhaps Table 1 socio-demographic variables could be dichotomized to save space?

Response:

We have dichotomized all of the continuous variables in Table 1, but we did leave the factorial structure for the categorical/nominal variables. The size of the table was greatly reduced, but still contains some level of detail. We are happy to make further adjustments if the table is still cumbersome to read.

Comment 11:

Given that non-partner violence has weaker association with the outcomes of interest, I would consider only examining IPV in this paper, and combining the IPV variable to be “ever physical and/or sexual IPV”.

Response:

We have removed NPV from the analysis. We are focusing solely on IPV. We did, however, decide to keep the three different levels of the IPV variable because they have differential associates with various mental and physical health outcomes.

Comment 12:

To simplify the analysis and make more of a statement about how violence frames outcomes for women, could you combine all the SF-36 items into a single continuous outcome (“physical health”) and say “Any sexual or reproductive health problems”? That way, Table 3 would only have two columns and be easier to interpret. Similarly, I might be tempted to drop Alcohol and Tobacco in Table 2, combine suicidal ideation and attempts into one column (“suicidality”) and report on the Short Form Health survey as one single continuous outcome.

Response:

We appreciate these suggestions about how to simplify the results for this study. We combined four sub-scales of the SF-36 into an aggregate physical health component score variable and the remaining four sub-scales of the SF-36 into a separate aggregate mental health component score variable as is commonly done with the SF-36. We also combined responses from all of the reproductive and gynecological conditions into a single dichotomous variable and presented only these results in the tables and text. We also created the “suicidality” variable as the reviewer suggested. We did, however, keep the alcohol and tobacco use variables in the results as substance use is a health outcome commonly associated with experiences of IPV [2, 3]. Pleased see revised measures sections (see lines 179-209)

Comment 13:

Would start Discussion with your own findings, rather than citing literature. You have already made the case for the need for new research, so don’t need to do it again here. First paragraph should sum up your key findings.

Response:

We have the adjustment to our discussion. In fact, we have, based on comments from both reviewers, rewritten a large portion of the discussion (see lines 307-442).

Comment 14:

In line 343 would cite:

Hatcher, A. M., A. Gibbs, R. Jewkes, R. S. McBride, D. Peacock and N. Christofides (2019). "Effect of childhood poverty and trauma on adult depressive symptoms among young men in peri-urban South African settlements." Journal of adolescent health 64: 79-85.

Response:

We appreciate the suggestion. We have made the addition.

Comment 15:

The Conclusion could be a bit shorter, and may not require citations (again, this is stylistic). Be sure there are no new ideas introduced here and it’s rather a summary of what’s already been stated.

Response:

We have shortened the conclusion considerably, removed references to literature, and provided more concrete recommendations for future policy and intervention. The discussion also contains considerably more recommendations for policy and intervention strategies to help prevent IPV in informal settlements to respond to physical and mental health outcomes associated with IPV in these communities. See lines 443-459 for revised conclusion.

References

1. Sineke T, Evans D, Schnippel K, van Aswegen H, Berhanu R, Musakwa N, et al. The impact of adverse events on health-related quality of life among patients receiving treatment for drug-resistant tuberculosis in Johannesburg, South Africa. Health and quality of life outcomes. 2019;17(1):94. doi: 10.1186/s12955-019-1155-4.

2. Abramsky T, Watts CH, Garcia-Moreno C, Devries K, Kiss L, Ellsberg M, et al. What factors are associated with recent intimate partner violence? findings from the WHO multi-country study on women's health and domestic violence. BMC public health. 2011;11(1):109.

3. Oram S, Khalifeh H, Howard LM. Violence against women and mental health. The Lancet Psychiatry. 2017;4(2):159-70.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

23 Jan 2020

PONE-D-19-26708R1

Intimate partner violence: A key correlate of women’s physical and mental health in informal settlements in Nairobi, Kenya

PLOS ONE

Dear Dr. Winter,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the remaining few points raised during the review process.

Please ensure that the issue of dichotomizing the SF-36 is adressed in your revised version.

We would appreciate receiving your revised manuscript by Mar 08 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Hajo Zeeb

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript focuses on the correlates of intimate partner violence (IPV) among women in informal settlements of Nairobi. I thank the authors for their conscientious responses to the reviewer comments and feel the paper is much improved. I do have a couple of comments below, but will leave it to the editor to ascertain whether these constitute another round of revisions of if the paper can be provisionally accepted.

The paper is much improved and sets the stage well for the analysis to follow. A few notes:

a. Line 83- “developing countries” (which I admit I used in my own review of this paper) may not be the most appropriate term. Check with PLOS for their preferred term to refer to what are commonly referred to “low- and middle-income countries” or LMIC.

b. Line 138- the same sentence regarding the Gibbs article from earlier in the introduction is mentioned again here- one reference is likely sufficient

c. Line 234: a comma should be placed after “study”

d. I am concerned that substantial information is being lost in the SF-36 by dichotomizing at the median. Is there a reason the authors decided to dichotomize instead of leaving it as a continuous variable and using linear regression for these models? A sentence expounding on this decision prior to its justification using prior studies would be helpful.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

5 Feb 2020

Response to Reviewers’ Comments

We want to thank the reviewer, again, for the insightful recommendations. We have made the appropriate adjustments to the paper and responded to the specific comments below. Please note: line numbers referenced here refer to the clean document (with all track-changes accepted).

Reviewer #1

Comment 1:

Line 83- “developing countries” (which I admit I used in my own review of this paper) may not be the most appropriate term. Check with PLOS for their preferred term to refer to what are commonly referred to “low- and middle-income countries” or LMIC.

Response:

Thank you for this comment. We have made the adjustment to use LMIC. (See ln 83)

Comment 2:

Line 138- the same sentence regarding the Gibbs article from earlier in the introduction is mentioned again here- one reference is likely sufficient

Response:

We have made the adjustment.

Comment 3:

Line 234: a comma should be placed after “study”

Response:

We have make the adjustment.

Comment 4:

I am concerned that substantial information is being lost in the SF-36 by dichotomizing at the median. Is there a reason the authors decided to dichotomize instead of leaving it as a continuous variable and using linear regression for these models? A sentence expounding on this decision prior to its justification using prior studies would be helpful.

Response:

We really appreciate this point and spent a great deal of time considering our approach to this issue both in the first revision and in this one. While some information may be lost in the dichotomization of the SF-36 at the median, the combined physical health and mental health components are not normally distributed and cannot be transformed in such a way to ensure the normality assumptions of a linear regression are being met; thus, the findings in the linear regression models may be biased or not interpretable. Following previous methods employing the SF-36, we felt the best approach to ensure the results were meaningful and accurate was to dichotomize the scores [1-3]. However, we have also acknowledged this limitation in the limitations section (please see lines 428-431).

References

1. Casso D, Buist DS, Taplin S. Quality of life of 5–10 year breast cancer survivors diagnosed between age 40 and 49. Health and quality of life outcomes. 2004;2(1):25.

2. Cioncoloni D, Innocenti I, Bartalini S, Santarnecchi E, Rossi S, Rossi A, et al. Individual factors enhance poor health-related quality of life outcome in multiple sclerosis patients. Significance of predictive determinants. Journal of the neurological sciences. 2014;345(1-2):213-9.

3. Sineke T, Evans D, Schnippel K, van Aswegen H, Berhanu R, Musakwa N, et al. The impact of adverse events on health-related quality of life among patients receiving treatment for drug-resistant tuberculosis in Johannesburg, South Africa. Health and quality of life outcomes. 2019;17(1):94. doi: 10.1186/s12955-019-1155-4.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

12 Mar 2020

Intimate partner violence: A key correlate of women’s physical and mental health in informal settlements in Nairobi, Kenya

PONE-D-19-26708R2

Dear Dr. Winter,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Hajo Zeeb

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

18 Mar 2020

PONE-D-19-26708R2

Intimate partner violence: A key correlate of women’s physical and mental health in informal settlements in Nairobi, Kenya

Dear Dr. Winter:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Hajo Zeeb

Academic Editor

PLOS ONE