Abstract
Background:
We sought to understand whether women’s empowerment and male partner engagement were associated with use of antenatal care (ANC).
Methods:
Women presenting for ANC in Nyanza province of Kenya during June 2015-May 2016, were approached for participation. A total of 137 pregnant women and 96 male partners completed baseline assessments. Women’s empowerment was measured using the modified Sexual Relationship Power Scale. ANC use measures included timing of the first ANC visit and number of visits. Male engagement was based on whether a husband reported accompanying his wife to one or more antenatal visits during the pregnancy. Multiple linear and logistic regression analyses were used to identify factors independently related to use and timing of ANC.
Results:
Women with higher mean empowerment scores were likely to have more than one ANC visit in the index pregnancy (Adjusted Odds Ratio (AOR)=2.8, 95% Confidence Interval (CI): 1.1–7.3), but empowerment was not associated with early ANC use. Women who were more empowered were less likely to have a husband who reported attending an ANC visit with his wife (AOR=0.1, 95% CI: 0.03–0.8).
Conclusions:
Women’s empowerment is important and may be related to ANC use and engagement of male partners in complex ways.
Keywords: Male engagement, women’s empowerment, prevention of mother-to-child transmission, HIV, antenatal care
Background
The influence of male partner engagement on maternal and child health, and women’s health in general, has been shown extensively in the literature. These relationships are particularly important in countries, such as Kenya, with cultural norms that may encourage male dominance or discourage women from taking an active part in decision-making processes (Kurniati et al. 2017, Yargawa and Leonardi-Bee 2015). Not only has positive male involvement been shown to enhance women’s use of antenatal care (ANC) visits, male involvement aids in the prevention of mother-to-child transmission (PMTCT) of HIV and also has been related to improved child health (Aluisio et al. 2011, Katz et al. 2009, Amano and Musa 2016, Msuya et al. 2008). Male involvement in ANC services and PMTCT, however, remains low in most sub-Saharan African settings, with studies reporting between 0.6%- 31% of males engaging in ANC and PMTCT services in various parts of Africa (Sherr and Croome 2012).
Challenges to women’s empowerment have long been pervasive in many low-resource settings, including Kenya. Despite significant efforts over the past few years, about 7% of women compared with 2.9% of men have no education (2014 Demographic and Health Survey (DHS) data). The same survey also showed that an estimated 1 in 5 women do not make decisions about their own health care, while approximately 42 % of women and 36 % of men justified wife beating. Other non-DHS studies have also found similar trends (Kenya National Bureau of Statistics et al. 2015). Despite over 75% of women being employed in the work force, only 1% of women in Kenya possess land titles in their names (Gaafar 2014, Kenya National Bureau of Statistics et al. 2015). Unfortunately, lack of women’s empowerment has been postulated to be the crux of challenges to improved outcomes for women’s health and general well-being (Ross et al. 2015, Mitroi et al. 2016). Empowered women are more likely to use ANC and postnatal care services (Sado, Spaho, and Hotchkiss 2014, Zaky, Armanious, and Hussein 2014, Hou and Ma 2013). However, it is a unclear whether women’s empowerment (often defined as the ability of women to make independent decisions about their own health) is in conflict with male engagement approaches, which encourage male partners to be involved in the decision-making processes and encourage joint decision-making concerning maternal and child health (MCH) (Jennings et al. 2014, Conroy et al. 2016). The risk exists that males will tend to dominate pregnancy-related decision-making once they are more involved (Turan et al. 2001b).
Although pregnant women’s contact with ANC services is relatively high in Kenya (96% of pregnant women have at least one ANC visit), ANC use continues to be sub-optimal, with many women presenting late for their first ANC visit and having fewer than four ANC visits throughout the course of pregnancy as suggested by the Kenyan national guidelines (Babalola 2014, International. 2015b). According to the 2014 Demographic and Health Survey (DHS) for Kenya, only an estimated 58% of women attended at least four ANC visits (Babalola 2014, International. 2015a).
While the several studies noted above have elucidated the importance of women’s empowerment and increased male involvement in ANC use in the reduction of maternal and infant mortality and morbidity, to the best of our knowledge, none of these studies have examined the relationship between these factors for both HIV-positive and HIV-negative pregnant women in the context of a high HIV prevalence setting. In addition, gaps remain in our knowledge of potential facilitators of pregnancy health care-seeking behaviors and male engagement in ANC care, which could inform efforts to promote PMTCT and improve overall ANC attendance. Although a study by Jennings et al in 2014 explored the relationship between women’s empowerment and male involvement using DHS data from eight African countries, excluding Kenya, it had mixed findings in terms of the direction of the association between male involvement and women’s empowerment (Jennings et al. 2014). Our analyses also bridge another gap in knowledge by using data collected directly from male partners of HIV-positive and HIV-negative pregnant women.
The aim of this paper was thus to explore the associations between women’s empowerment and indicators of ANC use, including timing of the first visit, number of visits, and male engagement in ANC, as well as to examine other important factors related to these positive health-seeking behaviors during pregnancy, and to make relevant policy recommendations.
Methods
Study setting, population, design and sampling procedure
Data for this paper were obtained from baseline assessments conducted for the Jamii Bora Study: a home-based couple’s intervention study conducted in Kenya (clinical trials number- NCT02403583). The Jamii Bora Study is a randomized controlled pilot intervention study of pregnant women and their partners with two arms: one arm was assigned to receive three couple home visits during pregnancy and postpartum to facilitate couple HIV testing and counseling (CHTC) as well as counseling on family health topics, and the other arm received standard care (invitation to return to the clinic with their male partner).
Women presenting for antenatal visits at five different rural health facilities in the former Nyanza Province of Kenya were approached for participation in the study during the period June 2015- May 2016. A total of 250 eligibility screening interviews were conducted. Pregnant women were eligible for inclusion in the study if they were at a gestational age less than or equal to 36 weeks at baseline, had a history of being offered HIV testing at an ANC visit, were 18 years of age or older, were currently living with a male partner with whom they had been in a stable relationship for at least six months, had not yet participated in CHTC with their male partner during the current pregnancy, and had not yet disclosed their current HIV status (positive or negative) to that partner. Women who were with a current male partner whom they already knew to be HIV-positive were ineligible for the study, as they did not have a need for CHTC and facilitated disclosure support.
Interested women who were found to be eligible signed an informed consent and then completed a baseline questionnaire. Women who had not experienced severe intimate partner violence (IPV) in the past six months (as assessed in the baseline questionnaire) were eligible for randomization after the baseline questionnaire. Women with a history of severe IPV (n=10) in the past six months were excluded from the randomized part of the study to avoid situations that might put women at further risk. In addition, 113 other women were excluded from the study for various reasons including; gestational age > 36 weeks, age <18 years, not being in a stable relationship for at least six months, male partner did not live in the same household for at least one night per week, had already had CHTC with male partner, had already disclosed HIV status to male partner, and knew for certain that male partner was HIV positive. The Study Coordinator (a trained counselor) provided counseling and linkage services to participants who reported recent severe IPV or depression, and they were referred to additional supportive resources at the local health facilities and in the nearest town. The present analyses used data from baseline questionnaires completed by 137 women (54.8% of those screened, approximately half HIV-positive and half HIV-negative) as well as 96 eligible male partners who could be contacted and agreed to participate in the study.
Data collection methods
Trained research assistants administered the baseline questionnaires, which were collected on tablet computers using the Open Datakit (ODK) platform. The research assistants, two males and two females, were all proficient speakers of the English language in addition to being native speakers of Dholuo and Kiswahili, the most frequently spoken languages in Nyanza province. The questionnaires were available on tablets in all three of these languages.
Dependent variables
Dependent variables considered in this study were indicators of use of health-seeking behavior during pregnancy, including early use of ANC (timing of first visit in the first trimester of pregnancy, a continuous variable as weeks of pregnancy, or dichotomized as less than 14 weeks--defined as first trimester--versus later), completion of only one versus more than one ANC visit during pregnancy, and male engagement in ANC. One versus more than one ANC visit was used as the outcome variable in these analyses, as well as controlling for current weeks of gestation in multivariate analyses, due to the fact that women completed baseline questionnaires at different stages in pregnancy, ranging from 4 to 36 weeks (mean 25 weeks). Male partner engagement in ANC was measured by the male partner’s report on his baseline questionnaire of whether he accompanied his pregnant partner on any antenatal visits during the current pregnancy. At least one accompanying ANC visit by a male partner was categorized as a “yes” and coded as “1”, while no accompanying ANC visit was categorized as a “no” and coded as “0”.
Independent variables
As a measure of women’s empowerment, the questionnaire for the Study used the modified Sexual Relationship Power Scale (SRPS) consisting of two subscales: the relationship control (RCS) and the decision-making dominance (DDS) subscales developed and validated by Pulerwitz, Gortmaker and DeJong in 2000 (Pulerwitz, Gortmaker, and DeJong 2000). The decision to examine relationship control and decision-making dominance stemmed from evidence from previous studies, including those using data from 2003 and 2008–2009 DHSKenya that showed that these two dimensions were important in assessing women empowerment and power dynamics in HIV/AIDS research (Voronca, Walker, and Egede 2018, Ghose et al. 2017, Upadhyay and Karasek 2012, Hindin, Kishor, and Ansara 2008). Furthermore, the SRPS scale has been extensively used to assess relationship and empowerment dynamics, including in low resource settings, such as Kenya (Pulerwitz, Mathur, and Woznica 2018, McMahon et al. 2015, Stephenson, Bartel, and Rubardt 2012). In this scale, a total of 15 items were included in the RCS which was reverse coded, with a maximum sub-scale score of 60 and a minimum of 15. Seven items were included in the DDS with a maximum sub-scale score of 21 and a minimum of 7. Mean scores were calculated for each subscale and then rescaled to obtain a final SRPS score, with a higher score signifying higher relationship power (Pulerwitz, Gortmaker, and DeJong 2000).
Other variables selected due to their importance in the literature and/or in theory included woman’s age, gravidity, education, polygynous relationship, occupation, women’s empowerment, the woman’s HIV status, household income, and male involvement.
Statistical analysis
Data analyses were conducted using SPSS version 23.0. Independent t-tests and bivariate logistic regression (unadjusted analyses) were initially used to assess the relationship between each dependent variable and individual independent variables. Multicollinearity and test assumptions were assessed. Statistically significant relationships, as well as additional variables that were deemed theoretically important mentioned above were included in multivariate linear and logistic regression analyses.
For multivariate analyses, we aimed for the best and most parsimonious models, given the findings from our review of literature and our preliminary correlational and bivariate analyses. We included education in our models because the literature has shown a strong association between level of education and pregnancy health care-seeking behaviors (Ahmed et al. 2010, Olayemi et al. 2009). Age was highly correlated with the number of pregnancies, and either age or gravidity was included in the models, based on theoretical considerations. Age (younger and older than 30 years) has been associated with increased risks for unhealthy pregnancy health care-seeking behaviors (Blanc, Winfrey, and Ross 2013), and lower parity has been associated with use of maternity services in Kenya (Turan et al. 2012). Weeks of pregnancy was included as a covariate to control for the differing timing during pregnancy when women entered the study for the male engagement in ANC and number of ANC visits outcomes. Polygynous relationship status (men having more than one wife) was also included because polygyny has been found to be associated with lower empowerment status and various situations that might contribute to adverse pregnancy health outcomes (Bove et al. 2015, Ditekemena et al. 2012). Household radio ownership and women’s occupation were used as proxies for socio-economic status (Turan et al. 2012, Ditekemena et al. 2012).
HIV status was also included in the models due to findings in the literature suggesting that women’s HIV status might affect pregnancy health care-seeking behaviors (Ditekemena et al. 2012, Turan et al. 2012). We decided to exclude marital status and ethnicity because almost all the women in our study were married and of the Luo ethnic group, respectively. Religion was also excluded as studies emanating from Sub-Saharan Africa have not consistently shown an association between religion and pregnancy health care-seeking behaviors (Ditekemena et al. 2012, Nwosu et al. 2012, Al-Mujtaba et al. 2016, Tarekegn, Lieberman, and Giedraitis 2014). We ran separate models examining the effects of the RCS, DDS, and the overall SRPS on our outcomes. Adjusted R-squared and the Hosmer-Lemeshow test were used to assess goodness of fit for the linear regression models and the logistic regression models, respectively.
Ethical Considerations
This study was conducted after obtaining ethical approval of the study protocol from the Kenya Medical Research Institute (KEMRI) Scientific and Ethical Review Unit (SERU) and the University of Alabama at Birmingham (UAB) Institutional Review Board (IRB). Signed informed consent was provided by all participants in the study. Data were de-linked from all personal identifiers, and data were stored in password protected/encrypted devices and servers, which were only accessible by IRB-approved personnel.
Results
Socio-demographic characteristics
A total of 223 participants, 137 women and 96 male partners, consented to take part in the study. The majority of participants were married and were of Luo ethnicity (Table 1). Most had primary school or less education, owned a radio and were in a monogamous marriage.
Table 1:
Socio-demographic and HIV-related characteristics by gender
| Variablesa | Female (n=137) | Male (n=96) |
|---|---|---|
| Age (years), mean (median, range) | 25.3 (24, 18–42) | 32.8 (31.5, 20–60) |
| N (%) | N (%) | |
| Marital Status | ||
| Currently married | 132 (96.4) | 96 (100) |
| Not married | 5 (3.6) | |
| Education | ||
| Primary school or less | 100 (73.0) | 51 (53.1) |
| More than primary school | 37 (27.0) | 45 (46.9) |
| Religion | ||
| Roman Catholic | 18 (13.1) | 14 (14.6) |
| Seventh Day Adventist | 39 (28.5) | 26 (27.1) |
| Others | 80 (58.4) | 56 (58.3) |
| Household goods ownership | ||
| Mobile Phone | 105 (76.6) | 88 (91.7) |
| Electricity | 28 (20.4) | 15 (15.6) |
| Radio | 114 (83.2) | 82 (85.4) |
| Television | 35 (25.5) | 18 (22.5) |
| Occupation | ||
| Housework | 35 (25.5) | - |
| Selling/Fish monger | 35 (25.5) | 9 (9.4) |
| Farming/manual labor | 21 (15.3) | 30 (31.2) |
| Others | 59 (43.1) | 57 (59.4) |
| Polygynous relationship | ||
| Yes | 32 (23.4) | 20 (20.8) |
| No | 105 (76.6) | 76 (79.2) |
| HIV status | ||
| HIV-positive | 72 (52.6) | 8 (8.3) |
| HIV-negative | 65 (47.4) | 78 (81.3) |
| Unknown | 10 (10.4) | |
| Maternal Characteristics | ||
| Weeks of pregnancy at first ANC, mean (median, range) | 22.1 (24, 4–36) | |
| Weeks at first ANC by trimesterb | 13 (9.6) | |
| 1st trimester | 87 (64) | |
| 2nd trimester | 36 (26.5) | |
| 3rd trimester | 25.0 (26, 4–36) | |
| Current weeks of pregnancy by trimester | ||
| Current weeks of pregnancy by trimester | ||
| 1st trimester | 6 (4.4) | |
| 2nd trimester | 66 (48.2) | |
| 3rd trimester | 65 (47.4) | |
| Number of pregnancies, mean (median, range) | 3.5 (3, 1–10) | |
| Live births, mean (median, range) | 2.3 (2, 0–9) | |
| Number of living children, mean (median, range) | 2.1 (2, 0–6) |
For categorical variables, n was reported followed by percentages in parentheses. For continuous variables, the mean followed by the median and the range, both in parentheses were reported.
N is less than 137 because of a missing value
Men tended to be older and more educated than women and were more likely to own a mobile phone (Table 1). While 52.6% of the women were HIV-positive (by study design), only 8.3% of the men self-reported being HIV-positive at baseline. Mean number of pregnancies and mean weeks of pregnancy at first antenatal care (ANC) were 3.5 pregnancies and 22.1 weeks, respectively. Approximately two-thirds, 64%, of the pregnant women had presented for their first ANC visit in the second trimester.
Empowerment measures and women’s characteristics
On a range of 1 (lowest) to 4 (highest), the mean RCS score for the overall sample was 2.7 ± 0.5; the mean DDS score was 1.7 ± 0.6, and the final weighted mean total SRPS score was 2.2 ± 0.5. Variables related to SRPS and subscale scores included women’s occupation, radio ownership and gravidity (Table 2). Women who engaged in farming and or manual labor (as compared to other occupations) tended to have lower scores on the RCS and overall SRPS scale. Surprisingly, women who reported that their household owned a radio had significantly lower mean DDS and SRPS scores, as compared to women who did not report household ownership of a radio. Lastly, women who were carrying their first pregnancies had a statistically significant higher RCS, DDS and SRPS scores than women who had been pregnant more than once. Age, education, polygynous relationship and HIV status were not significantly associated with the overall SPRS or subscale scores.
Table 2:
Women’s sociodemographic and HIV-related characteristics by their empowerment scores (n=137)
| Sociodemographic and HIV-related characteristicsa | Mean Relationship Control Subscale Score (SD) | Mean Decision-Making Dominance Subscale Score (SD) | Mean Total Sexual Relationship Power Scale Score (SD) |
|---|---|---|---|
| Age, years | |||
| Less than 30 | 2.7 (0.5) | 1.7 (0.6) | 2.2 (0.5) |
| Older than 30 | 2.6 (0.5 t=0.6, p= 0.6 |
1.7 (0.6) t=0.2, p= 0.8 |
2.1 (0.5) t=0., p= 0.7 |
| Education | |||
| Primary or less | 2.7 (0.4) | 1.7 (0.6) | 2.2 (0.5) |
| More than primary | 2.7 (0.5) t=−0.4, p= 0.7 |
1.7 (0.6) t=−0.1, p= 0.9 |
2.2 (0.5) t=−0.3, p= 0.8 |
| Radio ownership | |||
| Yes | 2.8 (0.4) | 1.6 (0.6)* | 2.1 (0.5)* |
| No | 2.6 (0.5) t=1.9, p= 0.06 |
2.0 (0.5) t=2.9, p= 0.004 |
2.4 (0.4) t=2.7, p= 0.01 |
| Occupation | |||
| Farming/Labor | 2.5 (0.4) | 1.5 (0.4) | 2.0 (0.4) |
| Others | 2.7 (0.5) t=−2.0, p= 0.05 |
1.7 (0.6) t=−1.8, p= 0.07 |
2.2 (0.5) t=−-2.1, p= 0.04 |
| Polygynous relationship | |||
| Yes | 2.7 (0.5) | 1.7 (0.6) | 2.2 (0.5) |
| No | 2.7 (0.5) t=−0.002, p= 1.0 |
1.7 (0.7) t=0.2, p= 0.8 |
2.2 (0.5) t=0.1, p= 0.9 |
| HIV status | |||
| HIV-positive | 2.6 (0.5) | 1.6 (0.6) | 2.1 (0.5) |
| HIV-negative | 2.7 (0.4) t=−0.9, p= 0. |
1.8 (0.6) t=−1., p= 0.2 |
2.2 (0.5) t=−1.3, p= 0.2 |
| Gravidity | |||
| 1st pregnancy | 2.9 (0.4)* | 2.0 (0.6) | 2.4 (0.5)* |
| 2nd or more | 2.6 (0.4) t=2.8, p= 0.01 |
1.7 (0.6) t=2.1, p= 0.04 |
2.1 (0.5) t=2.6, p= 0.01 |
Independent t-test significant at p <0.05.
Mean followed by standard deviation (SD) reported.
Antenatal Visits
Of the 136 women with non-missing data on ANC visits, 71.3% (n=97) had attended ANC only once, whereas only 28.7% (n=39) had attended more than one ANC visit. In bivariate analyses, radio ownership, HIV status, RCS and SRPS score were associated with having more than one ANC visit. Those with radios had a 0.3 decreased odds of having more than one ANC visit (p=0.01), while HIV-positive women had a 2.3-fold increased odds of multiple ANC visits (p=0.03). For every unit increase in RCS and SRPS, the odds of multiple ANC visits increased by 3.0 (p=0.01) and 2.3 (p=0.03), respectively.
The multivariate logistic regression analyses permitted examination of the association of SRPS with having more than one ANC visit, adjusted for other factors. In these adjusted analyses, low educational level (primary or less) was associated with a 0.3 decreased odds of attending more than one ANC visit (p=0.05) compared with women who had more than a primary education (Table 3.) HIV-positive women also had an increased odds (4.2, p=0.01) of having more than one visit. Every unit of increase in the SRPS score was associated with a 2.8 increase in the odds of more than one ANC visit (p= 0.04), adjusting for other factors. Repeating these analyses separately for the RCS and DDS revealed a similar significant association for the RCS (p=0.01) but not for the DDS (p=0.20). A sensitivity analysis was conducted excluding women in their first trimester, although the effect size was very similar (adjusted OR of 2.5 compared with an adjusted OR of 2.8) the significance level decreased to p=0.08, which may have been due to the reduced sample size.
Table 3:
Multivariate logistic regression of variables associated with having more than one ANC visit (N=136)>*
| Variable | Number of Women included (n) | Adjusted Odds Ratio for More than one ANC visit | 95% CI | p- Value |
|---|---|---|---|---|
| Empowerment | ||||
| Sexual Relationship Power Scale (Total) | 136 | 2.8 | 1.1–7.3 | 0.04 |
| Education | ||||
| Primary or less | 99 | 0.3 | 0.1–0.99 | 0.05 |
| More than a primary educationa | 37 | |||
| Age per year | 136 | 1.0 | 1.0–1.1 | 0.3 |
| Weeks of pregnancy (at time of interview) | 136 | 1.1 | 1.1–1.3 | 0.000 |
| Polygamous relationship | ||||
| Yes | 32 | 1.2 | 0.4–3.4 | 0.8 |
| Noa | 104 | |||
| Women’s occupation | ||||
| Farming/Manual labor | 21 | 1.5 | 0.5–6.3 | 0.4 |
| Othersa | 115 | |||
| HIV status | ||||
| Positive | 71 | 4.2 | 1.4–12 | 0.01 |
| Negativea | 65 | |||
| Radio ownership | ||||
| Yes | 113 | 0.4 | 0.1–1.3 | 0.1 |
| Noa | 23 | |||
N is less than 137 because of a missing value
Reference category.
Timing of the first antenatal care visit
A very low proportion of women in the study reported a first ANC visit in the first trimester of pregnancy, i.e., at less than 14 weeks (9.6%, n=13), with the mean timing of the first ANC visit being around 22 weeks. In bivariate analyses (data not shown), the only variables with significant associations with early ANC were those in polygamous relationships having a 3.2-fold increased odds of having their first ANC visit later (equal to or greater than 14 weeks) (p=0.05), and those experiencing their first pregnancy being more likely to access ANC early (mean weeks at first ANC visit was 19.3 weeks for primigravidas versus 22.6 weeks for multigravidas). After controlling for the other variables in the model, SPRS score was not significantly associated with timing of the first ANC visit (p=0.8) (Table 4). The only variable retaining significance in the final model was gravidity, with first pregnancies being significantly associated with earlier initiation of ANC (beta=0.2, p=0.04), although positive HIV status was also related (beta=0.18, p=0.05). Results were very similar for the RCS and DDS (data not shown).
Table 4:
Multivariate linear regression of variables associated with weeks of pregnancy at first ANC visit (N=136)*
| Variable | Number of Women included (n) | Beta- Coefficient | T-statistic | p- Value |
|---|---|---|---|---|
| Empowerment | −0.02 | −0.2 | 0.8 | |
| Sexual Relationship Power scale (Total) | 136 | |||
| Education | −0.0 | −0.1 | 0.7 | |
| Primary or less | 100 | |||
| More than a primary educationa | 36 | |||
| Gravidity | 0.2 | 2.0 | 0.04 | |
| First pregnancy | 19 | |||
| Second or greater pregnancya | 117 | |||
| Polygamous relationship | −0.09 | −1.0 | 0.3 | |
| Yes | 32 | |||
| Noa | 104 | |||
| Women’s occupation | −0.01 | −0.2 | 0.9 | |
| Farming/Manual labor | 21 | |||
| Othersa | 115 | |||
| HIV status | 0.18 | 2.0 | 0.05 | |
| Positive | 72 | |||
| Negativea | 6 | |||
| Radio ownership | −0.2 | −1.9 | 0.06 | |
| Yes | 113 | |||
| Noa | 23 |
N is less than 137 because of a missing value
Reference category.
Male Engagement in ANC
Overall, the male partners of 96 of the pregnant women were also enrolled in the study by the end of the recruitment period for pregnant women. Of these, only 12.5% (n=12) reported that they had ever accompanied their wives to at least one ANC visit in the index pregnancy. Although the factors were not significantly associated with male partner ANC attendance in bivariate analyses (data not shown), multivariate analyses revealed a significant association of women’s empowerment with male ANC attendance, after adjusting for other variables (Table 5). A statistically significant decreased odds of 0.14 was observed for male engagement in ANC associated with each unit increase in SRPS score (p=0.03). Repeating these analyses separately for the RCS and DDS revealed a similar significant association for the DDS sub-scale (p=0.03) but not for the RCS (p=0.07).
Table 5:
Multivariate logistic regression of variables associated with male partner attendance at ANC (N=96)
| Variable | Number of Women included (n) | Odds Ratio for Male Engagement | 95% CI | p- Value |
|---|---|---|---|---|
| Empowerment | ||||
| Sexual Relationship Power Scale (Total) | 96 | 0.1 | 0.03–0.8 | 0.03* |
| Education | ||||
| Primary or less | 67 | 0.6 | 0.1–2.7 | 0.5 |
| More than a primary educationa | 29 | |||
| Age per year | 96 | 0.9 | 0.8–1.1 | 0.2 |
| Weeks of pregnancy (at time of interview) | 96 | 1.0 | 0.9–1.1 | 0.7 |
| Polygamous relationship | ||||
| Yes | 20 | 2.6 | 0.5–14.7 | 0.3 |
| Noa | 76 | |||
| Women’s occupation | ||||
| Farming/Manual labor | 12 | 0.5 | 0.0–6.1 | 0.6 |
| Othersa | 84 | |||
| HIV status | ||||
| Yes | 47 | 0.5 | 0.1–2.2 | 0.3 |
| Noa | 49 | |||
| Radio ownership | ||||
| Yes | 80 | 0.2 | 0.04–1.4 | 0.1 |
| Noa | 16 |
Reference category.
Discussion
The findings presented here suggest that women’s empowerment was associated with pregnancy health care-seeking behaviors in the former Nyanza Province of Kenya, but in complex ways. More empowered women were more likely to have more than one ANC visit during the current pregnancy, but empowerment was not associated with early ANC use (with gravidity being the only significantly related factor). On the other hand, more empowered women were less likely to have a male partner who reported attending an ANC visit together with his wife.
In the current rural Kenyan sample, male partner accompaniment to ANC was very low at only 12.5%. This differs from the Jennings et al findings from the DHS from eight other African countries, which reported a range between 18.2% in Burundi to 86.8% in Rwanda (Msuya et al. 2008, Katz et al. 2009, Aluisio et al. 2016, Jennings et al. 2014, Aluisio et al. 2011). However, the current study included a select group of pregnant women who had not shared their HIV status (positive or negative) with their male partner. Thus, their reasons for not sharing their HIV test results might also be related to their lack of male partner accompaniment at ANC visits.
Our analyses revealed that women with higher DDS and SRPS scores were less likely to be accompanied to ANC by their male partner, even while controlling for other possible factors of male engagement in ANC. These findings are contrary to those observed in the Jennings et al. paper, although similar to those of studies conducted in Nepal, where increased women’s autonomy measured by decision-making power was associated with lower likelihood of male partner ANC accompaniment (Thapa and Niehof 2013, Jennings et al. 2014). It is possible that more empowered women did not feel the need to involve their male partner in ANC visits, being comfortable with making health-related decisions and health care visits on their own.
Our findings also indicate that women with higher SRPS and RCS scores were more likely to have more than one ANC visit, after controlling for other possible ANC visit factors. These findings are similar to those observed elsewhere (Ahmed et al. 2010, Singh and Singh 2014) and provides additional evidence for the contention that women’s empowerment can facilitate use of MCH services. Ahmed et al. found a 1.52 increased odds of attending four or more ANC visits in empowered women, using data from the DHS from 31 countries, including Kenya (Ahmed et al. 2010). Likewise, data from India have shown that low autonomy in women was associated with low levels of ANC use (Singh and Singh 2014).
In our sample, we did not find an association of empowerment measures with earlier use of ANC services during pregnancy. However, another recent Kenyan study did find a relationship between women’s autonomy and trimester of initiating ANC. Women who were involved in decision-making regarding the use of ANC services in their last pregnancy had a 2.5 increased odds of early initiation of ANC (Asweto et al. 2014).
In addition, our study elucidated other key factors that have been related to pregnancy health care-seeking behaviors, including education and HIV status (Asweto et al. 2014, Turan et al. 2012, Olayemi et al. 2009). More educated women and HIV-positive women were more likely to have more than one ANC visit compared with less educated women and HIV-negative women, respectively, after adjusting for other factors. The finding regarding HIV status is contrary to findings from Lesotho, where researchers did not observe a statistically significant difference in ANC attendance between HIV-positive and HIV-negative women (Gill et al. 2015). Contrary to many other studies (Khanal et al. 2015, Tekelab and Berhanu 2014, Turan et al. 2012, Olayemi et al. 2009), we did not find an association between other indicators of socio-economic status and any of the pregnancy health care-seeking behaviors examined.
Our study adds to the literature that reveals the complexity of relationships between women’s empowerment, male engagement, and health care use. As has been seen elsewhere, involving men in MCH in some cases may lead to more male domination and controlling behaviors in a sphere of life that has traditionally been under female control (Turan et al. 2001a). Thus, it appears that positive male engagement needs to be fostered, while at the same time encouraging women’s empowerment and full participation in decision-making.
The results of this study should be viewed within the context of several limitations. The overall study was a small intervention development and pilot study. Thus, the sample size did limit the statistical power of the current study to detect modest but potentially meaningful relationships. Also, as discussed above, the parent study selected a more vulnerable group of pregnant women who, despite being in a stable partnership, had not disclosed the result of their recent HIV test to their male partners, and had not tested for HIV together. On the other hand, male partners who were eligible (not currently a perpetrator of IPV) and actually participated in the baseline questionnaires for the study may have been those who tended to be more supportive, as compared to those who declined or could not be located. Nevertheless, this population represents an important sub-group in need of support and services and understanding of the role of women’s empowerment in their ANC use provides insights on methods to improve their service use and outcomes. In addition, 41 male partners of women in the sample were not able to be enrolled in the study (due to the woman’s baseline report of recent severe domestic violence (n=10) or being unreachable/unavailable (n=31) and did not complete baseline questionnaires providing data on their attendance at ANC visits. Furthermore, we used non-validated, self-reported HIV test results of male partners in the analysis for this paper, which might not be completely accurate. In addition, other unmeasured variables may account for women’s use of ANC services and male attendance at ANC. Finally, these analyses are based on cross-sectional data such that we examined the dependent and independent variables at the same point in time. This prevents assessment of temporality, and causal inferences cannot be made based on the findings in this paper.
Conclusions
The findings from this study provide support for the contention that women’s empowerment is an important construct related to use of antenatal care by vulnerable populations. However, the findings also suggest empowerment may be related to women’s health care use and engagement of male partners in complex ways, different from the general notion that had previously been published in literature. Our findings suggest that women’s empowerment may be associated with less male partner ANC attendance, which has been negatively associated with mother-to-child transmission of HIV, women’s ANC attendance and overall child health. Our results suggest that programs and services for pregnant women and infants in low-resource settings, especially western Kenya, should support efforts to empower women, but in addition, should also look for other pathways—preferably strategies that directly address couples and men as expectant fathers--to promote positive male engagement in maternal and child health, with important potential benefits for maternal, infant, and family health.
Acknowledgements
We thank the dedicated staff of the KEMRI Research Care and Treatment Program (RCTP) and Family AIDS Care and Education Services (FACES) who contributed their time and logistical support to this project. Special acknowlegement goes to the baseline data collection team, including Moses Okombo Ayany, Celestine Adhiambo Ngerhe, Julius Odiwour June, Jane Atieno June, and Irene Awuor Jimbo. We acknowledge the important role of the KEMRI-UCSF Collaborative group, the Director of KEMRI, the Director of KEMRI’s Centre for Microbiology Research, and the Kenya Ministry of Health Teams in the study locations for their important support in conducting this research. The research described in this manuscript was supported by the U.S. National Institute of Mental Health (NIMH), though grant R34MH102103. The content is solely the responsibility of the authors and does not necessarily represent the official views of the U.S. National Institutes of Health.
Funding Statement
The research described in this manuscript was funded by the U.S. National Institute of Mental Health (NIMH), though grant R34MH102103.
Footnotes
The authors declare no competing interests.
Contributor Information
Fatimah Oluwakemi Bello, Email: fobello@uab.edu.
Pamela Musoke, Email: pmusoke@uab.edu.
Zachary Kwena, Email: zkwena@kemri-ucsf.org.
George O. Owino, Email: gowino@kemri-ucsf.org.
Elizabeth A. Bukusi, Email: Ebukusi@kemri.org.
Lynae Darbes, Email: lynaed@umich.edu.
Janet M. Turan, Email: jmturan@uab.edu.
References
- Ahmed Saifuddin, Creanga Andreea A., Gillespie Duff G., and Tsui Amy O.. 2010. “Economic Status, Education and Empowerment: Implications for Maternal Health Service Utilization in Developing Countries.” PLoS ONE 5 (6):e11190. doi: 10.1371/journal.pone.0011190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Al-Mujtaba Maryam, Cornelius Llewellyn J., Galadanci Hadiza, Erekaha Salome, Okundaye Joshua N., Adeyemi Olusegun A., and Sam-Agudu Nadia A.. 2016. “Evaluating Religious Influences on the Utilization of Maternal Health Services among Muslim and Christian Women in North-Central Nigeria.” BioMed Research International 2016:3645415. doi: 10.1155/2016/3645415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aluisio AR, Bosire R, Betz B, Gatuguta A, Kiarie JN, Nduati R, John-Stewart G, and Farquhar C. 2016. “Male Partner Participation in Antenatal Clinic Services is Associated with Improved HIV-free survival Among Infants in Nairobi, Kenya: A Prospective Cohort Study.” J Acquir Immune Defic Syndr. doi: 10.1097/qai.0000000000001038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aluisio A, Richardson BA, Bosire R, John-Stewart G, Mbori-Ngacha D, and Farquhar C. 2011. “Male antenatal attendance and HIV testing are associated with decreased infant HIV infection and increased HIV-free survival.” J Acquir Immune Defic Syndr 56 (1):76–82. doi: 10.1097/QAI.0b013e3181fdb4c4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Amano A, and Musa A. 2016. “Male involvement in PMTCT and associated factors among men whom their wives had ANC visit 12 months prior to the study in Gondar town, North west Ethiopia, December, 2014.” Pan Afr Med J 24:239. doi: 10.11604/pamj.2016.24.239.8460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Asweto CO, Aluoch JR, Obonyo CO, and Ouma JO. 2014. “Maternal Autonomy, Distance to Health Care Facility and ANC Attendance: Findings from Madiany Division of Siaya County, Kenya.” American Journal of Public Health Research 2 (4):153–158. [Google Scholar]
- Babalola S 2014. “Women’s education level, antenatal visits and the quality of skilled antenatal care: a study of three African countries.” J Health Care Poor Underserved 25 (1):161–79. doi: 10.1353/hpu.2014.0049. [DOI] [PubMed] [Google Scholar]
- Blanc Ann K., Winfrey William, and Ross John. 2013. “New Findings for Maternal Mortality Age Patterns: Aggregated Results for 38 Countries.” PLoS ONE 8 (4):e59864. doi: 10.1371/journal.pone.0059864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bove JM, Golden MR, Dhanireddy S, Harrington RD, and Dombrowski JC. 2015. “Outcomes of a Clinic-Based Surveillance-Informed Intervention to Relink Patients to HIV Care.” J Acquir Immune Defic Syndr 70 (3):262–8. doi: 10.1097/qai.0000000000000707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Conroy AA, McGrath N, van Rooyen H, Hosegood V, Johnson MO, Fritz K, Marr A, Ngubane T, and Darbes LA. 2016. “Power and the association with relationship quality in South African couples: Implications for HIV/AIDS interventions.” Soc Sci Med 153:1–11. doi: 10.1016/j.socscimed.2016.01.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ditekemena John, Koole Olivier, Engmann Cyril, Matendo Richard, Tshefu Antoinette, Ryder Robert, and Colebunders Robert. 2012. “Determinants of male involvement in maternal and child health services in sub-Saharan Africa: a review.” Reproductive Health 9 (1):32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gaafar Reem. 2014. “Women’s land and property rights in Kenya.” Landesa: http://landwise.landesa.org/guides/8/generate_pdf. [Google Scholar]
- Ghose B, Feng D, Tang S, Yaya S, He Z, Udenigwe O, Ghosh S, and Feng Z. 2017. “Women’s decision-making autonomy and utilisation of maternal healthcare services: results from the Bangladesh Demographic and Health Survey.” BMJ Open 7 (9):e017142. doi: 10.1136/bmjopen-2017-017142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gill MM, Machekano R, Isavwa A, Ahimsibwe A, Oyebanji O, Akintade OL, and Tiam A. 2015. “The association between HIV status and antenatal care attendance among pregnant women in rural hospitals in Lesotho.” J Acquir Immune Defic Syndr 68 (3):e33–8. doi: 10.1097/qai.0000000000000481. [DOI] [PubMed] [Google Scholar]
- Hindin Michelle J., Kishor Sunita, and Ansara Donna L.. 2008. Intimate partner violence among couples in 10 DHS countries: Predictors and health outcomes In DHS Analytical Studies No. 18. Calverton, Maryland, USA: Macro International. [Google Scholar]
- Hou X, and Ma N. 2013. “The effect of women’s decision-making power on maternal health services uptake: evidence from Pakistan.” Health Policy Plan 28 (2):176–84. doi: 10.1093/heapol/czs042. [DOI] [PubMed] [Google Scholar]
- International., National Bureau of Statistics-Kenya and ICF. 2015a. “2014 KDHS Final Report.” [Google Scholar]
- International., National Bureau of Statistics-Kenya and ICF. 2015b. 2014 KDHS Key Findings. [Google Scholar]
- Jennings L, Na M, Cherewick M, Hindin M, Mullany B, and Ahmed S. 2014. “Women’s empowerment and male involvement in antenatal care: analyses of Demographic and Health Surveys (DHS) in selected African countries.” BMC Pregnancy Childbirth 14:297. doi: 10.1186/1471-2393-14-297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Katz DA, Kiarie JN, John-Stewart GC, Richardson BA, John FN, and Farquhar C. 2009. “Male perspectives on incorporating men into antenatal HIV counseling and testing.” PLoS One 4 (11):e7602. doi: 10.1371/journal.pone.0007602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kenya National Bureau of Statistics, Ministry of Health/Kenya, National AIDS Control Council/Kenya, Kenya Medical Research Institute, National Council for Population, and Development/Kenya. 2015. Kenya Demographic and Health Survey 2014. Rockville, MD, USA. [Google Scholar]
- Khanal V, Brites da Cruz JL, Mishra SR, Karkee R, and Lee AH. 2015. “Under-utilization of antenatal care services in Timor-Leste: results from Demographic and Health Survey 2009–2010.” BMC Pregnancy Childbirth 15:211. doi: 10.1186/s12884-015-0646-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kurniati A, Chen CM, Efendi F, Elizabeth Ku LJ, and Berliana SM. 2017. “Suami SIAGA: male engagement in maternal health in Indonesia.” Health Policy Plan 32 (8):1203–1211. doi: 10.1093/heapol/czx073. [DOI] [PubMed] [Google Scholar]
- McMahon James M., Volpe Ellen M., Klostermann Keith, Trabold Nicole, and Xue Ying. 2015. “A systematic review of the psychometric properties of the Sexual Relationship Power Scale in HIV/AIDS research.” Archives of sexual behavior 44 (2):267–294. doi: 10.1007/s10508-014-0355-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mitroi LR, Sahak M, Sherzai AZ, and Sherzai D. 2016. “The Women’s Health Care Empowerment Model as a Catalyst for Change in Developing Countries.” Health Care Women Int 37 (3):273–87. doi: 10.1080/07399332.2014.926903. [DOI] [PubMed] [Google Scholar]
- Msuya SE, Mbizvo EM, Hussain A, Uriyo J, Sam NE, and Stray-Pedersen B. 2008. “Low male partner participation in antenatal HIV counselling and testing in northern Tanzania: implications for preventive programs.” AIDS Care 20 (6):700–9. doi: 10.1080/09540120701687059. [DOI] [PubMed] [Google Scholar]
- Nwosu BO, Ugboaja JO, Obi-Nwosu AL, Nnebue CC, and Ifeadike CO. 2012. “Proximate determinants of antenatal care utilization among women in southeastern Nigeria.” Niger J Med 21 (2):196–204. [PubMed] [Google Scholar]
- Olayemi O, Bello FA, Aimakhu CO, Obajimi GO, and Adekunle AO. 2009. “Male participation in pregnancy and delivery in Nigeria: a survey of antenatal attendees.” J Biosoc Sci 41 (4):493–503. doi: 10.1017/s0021932009003356. [DOI] [PubMed] [Google Scholar]
- Pulerwitz J, Mathur S, and Woznica D. 2018. “How empowered are girls/young women in their sexual relationships? Relationship power, HIV risk, and partner violence in Kenya.” PLoS One 13 (7):e0199733. doi: 10.1371/journal.pone.0199733. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pulerwitz Julie, Gortmaker StevenL, and DeJong William. 2000. “Measuring Sexual Relationship Power in HIV/STD Research.” Sex Roles 42 (7–8):637–660. doi: 10.1023/A:1007051506972. [DOI] [Google Scholar]
- Ross Kara L, Zereyesus Yacob A, Shanoyan Aleksan, and Amanor-Boadu Vincent. 2015. “The Health Effects of Women Empowerment: Recent Evidence from Northern Ghana.” International Food and Agribusiness Management Review 18 (1):127. [Google Scholar]
- Sado L, Spaho A, and Hotchkiss DR. 2014. “The influence of women’s empowerment on maternal health care utilization: evidence from Albania.” Soc Sci Med 114:169–77. doi: 10.1016/j.socscimed.2014.05.047. [DOI] [PubMed] [Google Scholar]
- Sherr Lorraine, and Croome Natasha. 2012. “Involving fathers in prevention of mother to child transmission initiatives – what the evidence suggests.” Journal of the International AIDS Society 15 (Suppl 2):17378. doi: 10.7448/IAS.15.4.17378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Singh PK, and Singh L. 2014. “Examining inter-generational differentials in maternal health care service utilization: insights from the Indian Demographic and Health Survey.” J Biosoc Sci 46 (3):366–85. doi: 10.1017/s0021932013000370. [DOI] [PubMed] [Google Scholar]
- Stephenson R, Bartel D, and Rubardt M. 2012. “Constructs of power and equity and their association with contraceptive use among men and women in rural Ethiopia and Kenya.” Glob Public Health 7 (6):618–34. doi: 10.1080/17441692.2012.672581. [DOI] [PubMed] [Google Scholar]
- Tarekegn SM, Lieberman LS, and Giedraitis V. 2014. “Determinants of maternal health service utilization in Ethiopia: analysis of the 2011 Ethiopian Demographic and Health Survey.” BMC Pregnancy Childbirth 14:161. doi: 10.1186/1471-2393-14-161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tekelab Tesfalidet, and Berhanu Balcha. 2014. “Factors Associated with Late Initiation of Antenatal Care among Pregnant Women Attending Antenatal Clinic at Public Health Centers in Kembata Tembaro Zone, Southern Ethiopia.” Science, Technology and Arts Research Journal 3 (1):108–115. [Google Scholar]
- Thapa DK, and Niehof A. 2013. “Women’s autonomy and husbands’ involvement in maternal health care in Nepal.” Soc Sci Med 93:1–10. doi: 10.1016/j.socscimed.2013.06.003. [DOI] [PubMed] [Google Scholar]
- Turan JM, Hatcher AH, Medema-Wijnveen J, Onono M, Miller S, Bukusi EA, Turan B, and Cohen CR. 2012. “The role of HIV-related stigma in utilization of skilled childbirth services in rural Kenya: a prospective mixed-methods study.” PLoS Med 9 (8):e1001295. doi: 10.1371/journal.pmed.1001295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Turan JM, Nalbant H, Bulut A, and Sahip Y. 2001a. “Including expectant fathers in antenatal education programmes in Istanbul, Turkey.” Reprod Health Matters 9 (18):114–25. [DOI] [PubMed] [Google Scholar]
- Turan Janet Molzan, Nalbant Hacer, Bulut Ayşen, and Sahip Yusuf. 2001b. “Including expectant fathers in antenatal education programmes in Istanbul, Turkey.” Reproductive health matters 9 (18):114–125. [DOI] [PubMed] [Google Scholar]
- Upadhyay UD, and Karasek D. 2012. “Women’s empowerment and ideal family size: an examination of DHS empowerment measures in Sub-Saharan Africa.” Int Perspect Sex Reprod Health 38 (2):78–89. doi: 10.1363/3807812. [DOI] [PubMed] [Google Scholar]
- Voronca D, Walker RJ, and Egede LE. 2018. “Relationship between empowerment and wealth: trends and predictors in Kenya between 2003 and 2008–2009.” Int J Public Health 63 (5):641–649. doi: 10.1007/s00038-017-1059-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yargawa J, and Leonardi-Bee J. 2015. “Male involvement and maternal health outcomes: systematic review and meta-analysis.” J Epidemiol Community Health 69 (6):604–12. doi: 10.1136/jech-2014-204784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zaky HH, Armanious DM, and Hussein MA. 2014. “Testing for the endogenous nature between women’s empowerment and antenatal health care utilization: evidence from a cross-sectional study in Egypt.” Biomed Res Int 2014:403402. doi: 10.1155/2014/403402. [DOI] [PMC free article] [PubMed] [Google Scholar]