The influence of gender and household headship on voluntary health insurance: the case of North-West Cameroon

The influence of gender and household headship on voluntary health insurance: the case of... Abstract Within the existing health financing literature, males are typically categorized as the household‘s decision-makers. While this view accurately reflects many local sociocultural realities, approximately a quarter of sub-Saharan African households are now headed by females. In light of various efforts to expand health insurance coverage in the region, it is necessary to examine whether the factors influencing voluntary health insurance enrolment are analogous across male- and female-headed households. This study sought to identify the gendered determinants of voluntary enrolment into a church-run micro health insurance scheme. A cross-sectional survey of 550 households was carried out in Bui and Donga-Mantung Divisions of North-West Cameroon in May 2016. A structured questionnaire was administered on health insurance membership, household attributes, headship characteristics and health-seeking behaviour. We assessed the influence of gender on the associations between health insurance enrolment and the explanatory variables using logistic regression. This study found that voluntary health insurance demand was influenced by involvement in social networks regardless of gender. However, in line with entrenched household roles, men’s understanding of potential household health risks ultimately facilitated their enrolment decisions, while economically empowered women prioritised their direct knowledge of household health risks. Men’s demand for health insurance was correlated primarily with their education level (OR = 2.238 [CI 1.228–2.552]), as well as with their socioeconomic status (OR = 2.207 [CI 1.173–4.153]), age (OR = 2.238 [CI 1.151–4.352]) and trust of the insurance provider (OR = 4.770 [CI 2.407–9.453]). Conversely, women’s enrolment decision was primarily associated with their income levels (OR = 5.842 [CI 1.589–21.484]), as well as by the presence of children (OR = 3.734 [CI 1.228–11.348]). The influence of wealth on health insurance enrolment highlights the need for policymakers to subsidize health insurance schemes for vulnerable population groups. Further, it is imperative to develop sensitization campaigns that are simple and digestible to facilitate understanding of health insurance across all target groups. Voluntary health insurance, decision making, household headship, gender Key Messages Household heads are more likely to enrol into voluntary health insurance if involved in social networks, regardless of gender. Women prioritize their direct knowledge of potential household health risks when evaluating the decision to enrol into voluntary health insurance. However, income ultimately determines women’s ability to participate in health insurance schemes. When purchasing health insurance, men prioritize on their understanding of household health risks, which is linked to their education and age. Introduction Within the existing health financing research, males are typically categorized as the gatekeepers in the decision to enrol into voluntary health insurance. While this view accurately reflects local sociocultural realities in the developing world, many sub-Saharan African countries have been experiencing an increase in female-led households: In Cameroon, for example, women now head 25% of households (Institut National de la Statistique 2012). This necessitates an examination of whether the factors influencing voluntary health insurance enrolment are analogous across male- and female-headed households. This study seeks to identify if and how the gender of the household head influences enrolment into a voluntary micro health insurance scheme in North-West Cameroon. It further aims to analyse the influence, if any, of entrenched gender household roles on the decision to enrol. In many low- and middle-income countries, there is increasing recognition of the debilitating financial impact of illness on households (World Health Assembly 2005; United Nations General Assembly 2012; World Bank et al. 2016). Against this backdrop, governments and health financing stakeholders have endorsed prepaid healthcare as the preferred approach for reducing out-of-pocket expenditure and facilitating access to health services (World Health Organization 2010). This has largely taken the form of risk-pooling mechanisms in developing countries, with many governments striving to develop functional and inclusive health insurance schemes (Soors et al. 2010; Lagomarsino et al. 2012; Giedion et al. 2013). However, due to budgetary constraints and complexities in identifying the informal sector, many African governments have been unable to achieve large-scale health insurance coverage (Arhin-Tenkorang 2001; Acharya et al. 2013; Chuma et al. 2013). This has led to the proliferation of private micro health insurance schemes targeting those who would otherwise suffer from a disproportionate risk of impoverishment due to illness (Leive and Xu 2008; Wagstaff 2008). Crucially, these schemes are voluntary, and depend upon an active decision by households to purchase the health benefit package. In order to better understand the factors which influence one’s likelihood to seek health insurance coverage, a range of studies have investigated the determinants of voluntary enrolment across various African settings. These studies have largely focused on identifying the individual and household drivers of enrolment, with household size and composition; socioeconomic status (SES); and education level emerging as major determinants of voluntary health insurance enrolment (Criel et al. 1998; Musango et al. 2004; De Allegri et al. 2006a; Basaza et al. 2008; Jehu-Appiah et al. 2011, 2012; Onwujekwe et al. 2011; Kimani et al. 2012, 2014; Parmar et al. 2012; Mulupi et al. 2013; Dixon et al. 2014; Macha et al. 2014; Mebratie et al. 2015; ). Most of these studies have approached the analysis of health financing decision-making from a patriarchal point of view, reflecting the established sociocultural conventions in the study areas. This approach, however, may no longer be fully representative of current household structures, given that 26% of sub-Saharan African households are now headed by females (Beegle et al. 2016). The increase in female-led households in various sub-Saharan African countries may be attributed to several documented patterns in migration and epidemiology: widespread male economic migration to urban areas has resulted in de facto female household headship in many rural areas (International Organization for Migration 2013; Milazzo and Van de Walle 2015). Further, females are likely to live longer than males despite higher reported incidences of chronic diseases and exposure to death through childbirth (World Health Organization 2014). These realities underscore the need to explore the driving factors of voluntary health insurance enrolment not just from the perspective of normative social convention, but also with a pragmatic view of existing household headship structures. In this regard, there remains a dearth of literature on the role of women as primary agents in the decision to enrol into voluntary health insurance. Most relevant studies have investigated the ability of health insurance schemes to reach women as a vulnerable demographic, but not as the decision-makers driving the demand for health insurance (Kimani et al. 2014; Parmar et al. 2014). To the authors’ knowledge, only a single study has been carried out investigating the gendered determinants of voluntary health insurance enrolment in the African context. Dixon et al. carried out a comparative analysis of the factors associated with voluntary health insurance in Ghana, and found crucial educational, socioeconomic and marital differences between male and female individuals (Dixon et al. 2014). While this work presents an important first step in the analysis of the gender dynamics in health insurance demand, Dixon et al. did not consider the role of gendered household headship in the decision to enrol. This means that there is still a significant knowledge gap on the influence of gender on voluntary health insurance demand when decision-making autonomy is acquired. Precursory insight into the relationship between gender and healthcare decision-making may be gained from health-seeking behaviour literature, where several reviews have investigated the intersection between household roles and the decision to seek external health care. A systematic review by Colvin et al. on health-seeking behaviour found that timely treatment of sick household members is inextricably linked to the level of influence held by the mother on the final decision to seek external care (Colvin et al. 2013). This study also noted that male partners are typically only involved in the care-seeking pathway as the decision-makers of the economic and medical case for seeking outside care. Congruently, a global review of morbidity and mortality literature found that autonomy amongst mothers was associated with better overall health status amongst children in the household (Richards et al. 2013). These studies allude to potential gendered differences in the consideration of health-related decisions: they suggest that the divisional roles intrinsic within the household lead men and women to evaluate health risks differently. Based on this, we hypothesize that women are likely to prioritize their direct knowledge of the household’s potential healthcare needs when making healthcare decisions. Conversely, we hypothesize that men are likely to prioritize their understanding of potential health risks associated with their households. Materials and methods Research setting The Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) scheme is a micro health insurance scheme set up by the Roman Catholic Church in North-West and South-West Cameroon. It consists of four independently-operated schemes in Bamenda, Buea, Kumbo and Mamfe Parishes covering a total of 35 224 individuals as of November 2016 (BEPHA 2016). For the purposes of this study, we will be focusing on the BEPHA Kumbo Scheme. The BEPHA Kumbo Scheme is active in the Bui and Donga-Mantung administrative divisions of North-West Cameroon. The North-West region is plagued by poor health outcomes, including the highest HIV prevalence rate in Cameroon (8.7%). It also suffers from a high malaria prevalence of 20% (Institut National de la Statistique du Cameroun 2014). Poor health outcomes in this region are exacerbated by steep out-of-pocket costs: according to the country’s Demographic and Health Survey (DHS), North-West Cameroon has the second-highest level of health spending in Cameroon in the event of illness, in spite of accounting for 13% of the country’s poor (Institut National de la Statistique 2008, 2012). Membership of the BEPHA Kumbo Scheme is voluntary with a minimum enrolment unit of four individuals. The annual premium per individual is set at FCFA 4000 (US$6.80 in May 2016) and is paid in a maximum of three instalments. BEPHA annual coverage starts in either June or November. However, coverage within educational institutions begins in October in line with the annual academic calendar. The BEPHA benefit package covers three-quarters of the cost of inpatient services, delivery services, outpatient services and surgery, respectively. Enrolees are permitted to access BEPHA inpatient and outpatient care benefits twice annually, while surgery and maternity services can only be reimbursed once annually. BEPHA applies maximum cost ceilings for each of the offered services in order to ensure financial sustainability: FCFA 15 000 (US$25.50) for outpatient services and maternity services, respectively; FCFA 25 000 (US$42.50) for inpatient services; and FCFA 70 000 (US$119.00) for surgery. Study design A cross-sectional household survey was carried out in the Bui and Donga-Mantung divisions of North-West Cameroon between April and May 2016. All administrative sub-divisions of Bui and Donga-Mantung were eligible for sampling. For operational reasons, the study was conducted in the sub-divisions where >30 households were enrolled into the BEPHA Kumbo Scheme. The study had two distinct aims: (1) to estimate the proportion of the population with BEPHA health insurance and (2) to assess the effect of explanatory variables on this proportion. For the first aim, we based the sample size on the ability to estimate the proportion insured with a certain precision—in this case, with an expected BEPHA population coverage of 4%. This coverage rate was obtained from the BEPHA membership audit carried out in FY 2015–2016 (BEPHA 2016). Using the Hayes and Bennett equation and taking into account clustering in nine sub-districts, the required sample size for the first study aim was calculated to be 416 households with the precision of a 95% confidence interval with width of 2–6% (Hayes and Bennett 1999). The number of households interviewed in each sub-division was proportional to its demographic size (République du Cameroun and Agence Régionale du Nord-Ouest 2015), and we assumed that the measure of variability between clusters k, the standard deviation divided by the mean, was equal to 0.1 (Hayes and Bennett 1999). As a sampling frame was not readily available for the study area, we randomly selected starting points within outposts in each sub-division. We subsequently carried out systematic sampling, with every nth household interviewed. Approximately 98% of the targeted households could be interviewed. In order to achieve the second study aim of estimating the factors associated with health insurance enrolment, we sampled additional insured households from the BEPHA membership list. BEPHA’s internal client management system stratifies insured households according to their sub-division of residence and randomly assigns membership numbers. In order to optimize our sample of insured households to address the second study aim, we selected every nth number from the BEPHA membership list. In total, 174 insured households were randomly selected stratifying by sub-division, from the BEPHA Kumbo membership list. These households were chosen from the same sub-divisions within which the random population sampling was conducted. Approximately 80% of these households could be identified and interviewed. Survey tools and data analysis The survey was administered as a structured questionnaire using Open Data Kit software on handheld tablets to answer questions on household composition; household assets; household expenditure and consumption; and health-seeking behaviour. Research assistants who were fluent in Pidgin English and the local dialect, Lam Nso’, and who had knowledge of the local geographical and sociocultural context were hired. Training was provided to familiarize research assistants with the questionnaire and data collection using handheld tablets. Insured households were defined as those where at least one member was enrolled into the BEPHA Kumbo Scheme in the year preceding the study. The household head was defined as the household member whose income contributed to at least half of the household’s costs in the preceding year. The household head’s gender was defined as the self-reported sex of the household head. De facto female household headship was assumed in households where no male adults lived for the preceding year. The data were analyzed using STATA version 14.1 for Windows (STATA Corporation, College Station, Texas). The main outcome variable was health insurance enrolment into the BEPHA Kumbo Scheme. The explanatory variables were divided into five components: household composition and attributes, household head factors, mean perceived household health status and proxies for exposure to financial risk pooling. In order to assess asset-based wealth, we constructed an asset index from the Cameroonian Demographic and Health Survey (DHS) index which measures the relative wealth ranking of households (Cameroun Institut National de la Statistique 2011). Household per capita consumption was calculated as the annual food, non-food, and consumer durables consumption per household member. Mean perceived health status was calculated as the average self-reported health status value for all household members. The final value of the mean perceived health status was assigned a value of between 1 and 4, with 1 being ‘very good’ health status; 2 being ‘good’ health status; 3 being ‘poor’ health status; and 4 being ‘very poor’ health status. Chronic disease status in the context of this study was attributed on the basis of the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) (World Health Organization 1992). We used logistic regression to estimate the association between the explanatory variables and health insurance enrolment, with random effects for sub-division and outpost to account for the clustering in the sample. Our study utilized two regression models. The first model measured the association of each variable and health insurance enrolment within each gender group. The second model included interaction terms for gender and each explanatory variable to determine the influence of gender on the association of each variable and health insurance enrolment. Ethical approval Ethical approval for this study was obtained through the Institutional Review Board (IRB) of the Catholic University of Cameroon in Bamenda, North-West Cameroon (Ethics Reference No. 001/HEPM/CATUC-IRB/16; Date of approval: 17th May 2016). Results Study population characteristics Table 1 reports the differences in each test variable according to insurance status and gender of the household head. A total of 550 households were enumerated in the survey, with male-headed households accounting for 70% of the study population. 80% of male-headed and 89% of female-headed households lived in rural areas, reflecting the largely rural nature of the population in the Bui and Donga-Mantung Divisions. Approximately 28% of male- and 36% of women-headed households were enrolled into BEPHA micro health insurance. Table 1. Descriptive statistics by gender and insurance status   Male   Female   Insured (n = 105)  Uninsured (n = 275)  Combined (n = 380)  Insured (n = 61)  Uninsured (n = 109)  Combined (n = 170)  Household characteristics      Urban location  27%  18%  20%  11%  11%  11%      Mean household sizea  5.81 (2.33)  5.66 (2.79)  5.70 (2.67)  4.95 (2.17)  4.43 (2.48)  4.61 (2.38)      Household sex ratio (males to females)a  1.194 (0.969)  1.273 (0.979)  1.250 (0.974)  0.762 (0.788)  0.631 (0.669)  0.674 (0.714)      Households with children ≤5 years  57%  60%  59%  41%  51%  48%      Households with children ≤15 years  85%  85%  85%  90%  75%  81%      Households with elderly ≥60 years  32%  26%  28%  34%  28%  30%      Average SES scoreb  0.492 (0.965)  0.317 (0.834)  0.365 (0.875)  0.318 (0.674)  0.189 (0.677)  0.236 (0.677)      % of poor households (Q1–Q2)  36%  40%  39%  39%  45%  43%      % of rich households (Q5)  30%  20%  23%  20%  9%  13%  Household head characteristics      Average ageb  49 (13.23)  47 (15.11)  48 (14.60)  49 (12.46)  47 (16.00)  48 (14.95)      Literate  94%  76%  81%  77%  63%  68%      Primary-level education  50%  41%  44%  54%  40%  45%      Secondary-level education  21%  24%  23%  18%  19%  19%      Higher-level education  23%  9%  13%  5%  3%  4%      Married household heads  94%  90%  91%  28%  27%  27%      Low income (≤47 000 FCFA)  50%  48%  49%  57%  60%  59%      Proportion of Catholic faith  87%  57%  66%  87%  61%  71%  Household health status      Mean health statusa  1.96 (0.672)  1.96 (0.67)  1.96 (0.67)  2.01 (0.76)  1.86 (0.74)  1.92 (0.76)      Presence of chronic disease  8%  6%  6%  5%  5%  5%  Use of curative services      Outpatient service use in past 30 days  42%  35%  37%  33%  25%  28%      Inpatient service use in past year  42%  35%  37%  31%  28%  29%      Annual per capita health expenditure (FCFA)b  5000 (10 500)  4500 (11 750)  5000 (11 634)  5000 (9500)  5000 (12 500)  5000 (12 500)    Male   Female   Insured (n = 105)  Uninsured (n = 275)  Combined (n = 380)  Insured (n = 61)  Uninsured (n = 109)  Combined (n = 170)  Household characteristics      Urban location  27%  18%  20%  11%  11%  11%      Mean household sizea  5.81 (2.33)  5.66 (2.79)  5.70 (2.67)  4.95 (2.17)  4.43 (2.48)  4.61 (2.38)      Household sex ratio (males to females)a  1.194 (0.969)  1.273 (0.979)  1.250 (0.974)  0.762 (0.788)  0.631 (0.669)  0.674 (0.714)      Households with children ≤5 years  57%  60%  59%  41%  51%  48%      Households with children ≤15 years  85%  85%  85%  90%  75%  81%      Households with elderly ≥60 years  32%  26%  28%  34%  28%  30%      Average SES scoreb  0.492 (0.965)  0.317 (0.834)  0.365 (0.875)  0.318 (0.674)  0.189 (0.677)  0.236 (0.677)      % of poor households (Q1–Q2)  36%  40%  39%  39%  45%  43%      % of rich households (Q5)  30%  20%  23%  20%  9%  13%  Household head characteristics      Average ageb  49 (13.23)  47 (15.11)  48 (14.60)  49 (12.46)  47 (16.00)  48 (14.95)      Literate  94%  76%  81%  77%  63%  68%      Primary-level education  50%  41%  44%  54%  40%  45%      Secondary-level education  21%  24%  23%  18%  19%  19%      Higher-level education  23%  9%  13%  5%  3%  4%      Married household heads  94%  90%  91%  28%  27%  27%      Low income (≤47 000 FCFA)  50%  48%  49%  57%  60%  59%      Proportion of Catholic faith  87%  57%  66%  87%  61%  71%  Household health status      Mean health statusa  1.96 (0.672)  1.96 (0.67)  1.96 (0.67)  2.01 (0.76)  1.86 (0.74)  1.92 (0.76)      Presence of chronic disease  8%  6%  6%  5%  5%  5%  Use of curative services      Outpatient service use in past 30 days  42%  35%  37%  33%  25%  28%      Inpatient service use in past year  42%  35%  37%  31%  28%  29%      Annual per capita health expenditure (FCFA)b  5000 (10 500)  4500 (11 750)  5000 (11 634)  5000 (9500)  5000 (12 500)  5000 (12 500)  a Results reported as mean (standard deviation). b Results reported as median (interquartile range). Mean exchange rate for May 2016: US$1 = FCFA 588.24. Table 2. Logistic estimates for probability of purchasing BEPHA health insurance at household level Variable description  Male   Female   Interaction   Odds ratio  CI  Odds ratio  CI  P-value  Household characteristics    Location (reference group: Rural)      Urban  1.378  0.646–2.939  1.063  0.211–5.354  0.78    Number of children <5 years (Reference group: None)      One to two children  0.826  0.503–1.357  0.977  0.451–2.116  0.67      Three or more children  0.933  0.373–2.333  0.151  0.168–1.357  0.13      Presence of children <15 years  1.143  0.567–2.306  3.734  1.228–11.348  0.051      Presence of elderly household members  1.301  0.756–2.269  1.072  0.484–2.376  0.83    SES (reference group: Poor)      Middle wealth  0.722  0.431-1.210  0.760  0.355–1.630  0.78       Wealthiest quintile  2.207  1.173–4.153  2.698  0.825–8.826  0.94  Household head characteristics    Age (reference group: under 35)      35–50 years  1.223  0.743–2.013  1.543  0.728–3.271  0.46      Older than 50 years  2.238  1.151–4.352  1.916  0.697–5.264  0.82    Education (reference group: primary education or less)      Secondary education or more  1.770  1.228–2.552  1.554  0.903–2.675  0.008    Monthly income (reference group: no income)     <23 500 FCFA  0.576  0.322–1.029  0.559  0.257–1.214  0.93      23 500–47 000 FCFA  1.658  0.902–3.047  1.883  0.707–5.020  0.86     >47 000 FCFA  1.401  0.810–2.425  5.842  1.589 –21.484  0.037    Married  1.898  0.709–5.081  1.259  0.542–2.927  0.56    Catholic religion  4.770  2.407–9.453  2.926  1.076–7.953  0.33  Household health status    Mean perceived household health status      Good  1.043  0.621–1.751  0.860  0.399–1.854  0.49      Poor  0.959  0.571–1.601  1.163  0.539–2.507  0.49    Presence of chronic illness in household  1.519  0.574–4.018  0.795  0.128–4.924  0.99  Exposure to concept of financial risk protection    Member of informal savings group (njangi)  2.301  1.302–4.064  3.146  1.429–6.921  0.42    Trust BEPHA health insurance provider  5.170  2.365–11.301  2.830  0.884–9.061  0.23  Variable description  Male   Female   Interaction   Odds ratio  CI  Odds ratio  CI  P-value  Household characteristics    Location (reference group: Rural)      Urban  1.378  0.646–2.939  1.063  0.211–5.354  0.78    Number of children <5 years (Reference group: None)      One to two children  0.826  0.503–1.357  0.977  0.451–2.116  0.67      Three or more children  0.933  0.373–2.333  0.151  0.168–1.357  0.13      Presence of children <15 years  1.143  0.567–2.306  3.734  1.228–11.348  0.051      Presence of elderly household members  1.301  0.756–2.269  1.072  0.484–2.376  0.83    SES (reference group: Poor)      Middle wealth  0.722  0.431-1.210  0.760  0.355–1.630  0.78       Wealthiest quintile  2.207  1.173–4.153  2.698  0.825–8.826  0.94  Household head characteristics    Age (reference group: under 35)      35–50 years  1.223  0.743–2.013  1.543  0.728–3.271  0.46      Older than 50 years  2.238  1.151–4.352  1.916  0.697–5.264  0.82    Education (reference group: primary education or less)      Secondary education or more  1.770  1.228–2.552  1.554  0.903–2.675  0.008    Monthly income (reference group: no income)     <23 500 FCFA  0.576  0.322–1.029  0.559  0.257–1.214  0.93      23 500–47 000 FCFA  1.658  0.902–3.047  1.883  0.707–5.020  0.86     >47 000 FCFA  1.401  0.810–2.425  5.842  1.589 –21.484  0.037    Married  1.898  0.709–5.081  1.259  0.542–2.927  0.56    Catholic religion  4.770  2.407–9.453  2.926  1.076–7.953  0.33  Household health status    Mean perceived household health status      Good  1.043  0.621–1.751  0.860  0.399–1.854  0.49      Poor  0.959  0.571–1.601  1.163  0.539–2.507  0.49    Presence of chronic illness in household  1.519  0.574–4.018  0.795  0.128–4.924  0.99  Exposure to concept of financial risk protection    Member of informal savings group (njangi)  2.301  1.302–4.064  3.146  1.429–6.921  0.42    Trust BEPHA health insurance provider  5.170  2.365–11.301  2.830  0.884–9.061  0.23  The descriptive statistics for the sample population were similar across gender lines in terms of location, and the household head’s age, self-employment status and religion. Overall, male and female household heads had an average age of 48.0 years, with a self-employment rate of 65%. The sample was predominantly Catholic (67%), highlighting BEPHA’s affiliation with the Church. The gender groups also showed similarities in their reported health status, with 6% of all households reporting the presence of chronic disease. We found important distinctions between male- and female-headed households in several household composition and socioeconomic characteristics. Male-headed households had more uniformity in the number of males and females within the household compared with their female counterparts. They were also more likely to be in the wealthiest socioeconomic quintile and have higher incomes than female-headed households. Female-headed households, on the other hand, were typically smaller in size, had more female than male household members and were more likely to be located in a rural setting. Households also differed in their marital status, with 91% of male household heads living within a marriage setting compared with 40% of female household heads who were widowed. Factors associated with voluntary health insurance enrolment amongst male- and female-headed households Household-level regression results are presented in Table 2. This study identified several commonalities amongst male- and female-headed households in the characteristics associated with health insurance demand. Regardless of gender, we found that Catholic Church membership was correlated to one’s likelihood of enrolling into BEPHA health insurance (OR = 4.770 [CI 2.407–9.453] for males and OR = 2.926 [CI 1.076–7.953] for females). Practices in social solidarity were also found to be associated with BEPHA health insurance enrolment across genders: both male and female household heads showed higher odds of purchasing health insurance if they belonged to informal savings groups known as njangis (OR = 2.301 [CI 1.302–4.064] for males and OR = 3.146 [CI 1.429–6.921] for females). In spite of these broad convergences, we found that the influencing factors of BEPHA insurance enrolment differed depending on the gender of the household head. Amongst male household heads, demand for health insurance was correlated to their SES, age, education level and their trust of the insurance provider. This study found that men belonging to the highest asset-based wealth quintile had 2.207 higher odds of enrolling into health insurance compared to those with lower asset-based wealth (CI 1.173–4.153). Older age was also identified as a significant contributor to health insurance enrolment, with male household heads aged above 50 years showing increased health insurance demand compared with their younger counterparts (OR = 2.238 [CI 1.151–4.352]). Further, men educated to secondary school-level and above showed 2.238 higher odds of enrolment compared with those with lower education levels (CI 1.228–2.552). Finally, we found that male household heads were more likely to enrol if they expressed confidence in the Church as a potential health insurance provider (OR = 4.770 [CI 2.407–9.453]). Amongst female-headed households, the decision to enrol into BEPHA health insurance was associated with household composition and income. This study found that demand for BEPHA health insurance amongst female-headed households was correlated to the presence of children below the age of 15 (OR = 3.734 [CI 1.228–11.348]). Women who earned higher incomes were also more likely to purchase BEPHA health insurance: the study found that women who earned earning above 47 000 FCFA had 5.842 higher odds of enrolling into health insurance than those earning less (CI 1.589–21.484). When the interactions between gender, health insurance membership and the explanatory variables were taken into account, we found that Catholic Church and njangi membership was correlated to health insurance demand in both male- and female-headed households (P-value = 0.332 for Catholic Church membership, and P-value = 0.421 for njangi membership, respectively). Income and education, however, were found to have varied associations with the decision to enrol into health insurance depending on the gender of the household head. Amongst female-headed households, we found that health insurance enrolment was associated with high income levels (P-value = 0.037)—a finding that was non-significant amongst male-headed households. Conversely, male-headed households with higher levels of education were found to have higher demand for BEPHA health insurance (P-value = 0.008). This finding was non-significant amongst female-headed households. We found no further evidence of a relationship between the gender of the household head, BEPHA membership and other explanatory variables. Discussion In this study, we set out to expand the narrative on decision-making in voluntary health insurance by looking at how the gender of a household head influences voluntary health insurance enrolment. It is important to acknowledge from the outset that commonalities exist between male- and female-headed households, with the overarching influence of social networks on health insurance demand. Based on our findings, it is clear that involvement in networks that encourage solidarity and reciprocity, such as njangis and the Catholic Church, increase one’s likelihood of enrolling into voluntary health insurance regardless of gender. This reflects existing studies which have highlighted the importance of social solidarity as a key determinant of voluntary health insurance schemes demand (Basaza et al. 2007; Mladovsky et al. 2015). This notwithstanding, it is apparent that the decision to enrol into voluntary health insurance is multidimensional and complex, driven in part by aversion to the risk of illness (Schneider 2004). We propose that the evaluation of health risks manifests differently between male and female household heads due to their unique household roles in the study setting. We therefore postulate that women prioritize their direct knowledge of the household’s healthcare needs in the decision to enrol into health insurance. Conversely, we suggest that men prioritize their understanding of potential health risks associated with their households. How does this reconcile with existing knowledge of health insurance and the everyday realities of household power structures? Within existing literature, it is acknowledged that the burden of caregiving is borne primarily by women—a dynamic that is particularly pronounced in the occurrence of illness in the domestic framework (Colvin et al. 2013; Richards et al. 2013). This underlying knowledge of the physical, psychological and economic cost of illness elevates the female voice as imperative in understanding and assessing household health risks. In the traditional hierarchy dominant within North-West Cameroon, however, female influence is often relegated in decision-making due to their subordinate status (Goheen 1995). This means that, in the absence of direct knowledge of household health needs, male household heads are compelled to act on the basis of their knowledge and understanding of potential household health risks. Our findings suggest that this is primarily shaped by their level of education, which aligns to other health financing studies across Africa (De Allegri et al. 2006a; Jehu-Appiah et al. 2012; Kimani et al., 2012, 2014; Dixon et al. 2014). We posit that an advanced level of education enables male household heads to better evaluate the potential social and economic implications of illness, as well as to assess the financial protection afforded by health insurance. This study also found that men’s ability to evaluate potential health risks was correlated to a lesser degree to their old age. We suggest that older age provides male household heads with a better understanding of the potential health, economic and social costs of illness, thereby increasing their understanding of the utility of financial safety nets in adverse health events. In addition to the above characteristics, we found that male household heads place value on a specific return on investment, factoring in trust of a specific health insurance provider in their decision to enrol. This reflects other qualitative and quantitative health financing studies (De Allegri et al. 2006 b; Basaza et al. 2007; Jehu-Appiah et al. 2011, 2012), and highlights the reliance of many voluntary health insurance schemes on the reputation of affiliated entities. When women hold a high level of economic power, on the other hand, we posit that enrolment decisions are likely to prioritize their direct knowledge of potential household health risks. According to our findings, health insurance enrolment was correlated to the presence of children in female-headed households. This implies that women with dependent children are attuned to the healthcare needs of their households, ostensibly due to their traditional role as caregivers. This reflects the findings of various health-seeking behaviour studies, which have found that women’s role in caregiving provides them with awareness of the potential health risks associated with childhood illness (Colvin et al. 2013; Richards et al. 2013). While their willingness to prioritize the health needs of children is notable, poverty ultimately reduces women’s access to decision-making resources. Given that most asset-based wealth in North-West Cameroon has historically been held by males (Goheen 1995), income acts as a limiting factor for women’s participation in health insurance over and above the need to minimize potential household health risks. This aligns to the findings of various social science studies (Kishor and Nietzel 1997; Dixon et al. 2014; Milazzo and Van de Walle 2015), and suggests that access to financial resources serves as an important gateway for facilitating women’s decision-making autonomy. Given the wide variability in household structures between and within regions, care must be taken to avoid blunt generalizations on the nature of households. We concede that these findings are specific to the unique North-West Cameroonian context within which the research was carried out. We also appreciate that even within our designated gender groupings, there will inevitably be heterogeneity in the power relations that are associated with the decision to enrol into voluntary health insurance. That said, to the authors’ knowledge, this is one of the first studies investigating the role of household headship and gender in the context of voluntary health insurance enrolment. This study has several design limitations that may affect the external validity of our findings: the number of BEPHA-insured female-headed households was substantially less than those in male-headed households. However, we contest that this is symptomatic of the low number of female-headed households in the study area in general, as well as the limited uptake of insurance amongst this study population. It would therefore be necessary to carry out further research on a larger number of female-led households to further test our findings. In addition, translation and interpretation of interview questions into Lam Nso’ language may have biased the responses to the research questions. These issues notwithstanding, our findings provide precursory insight into the evolving role of gender in health insurance decision-making. Conclusion This study has embraced the realities of gender-specificity in household roles, and identified clear disparities in the way male- and female-headed households evaluate health risks in the decision to enrol into health insurance. Our findings suggest that voluntary health insurance demand amongst male household heads is associated with their ability to understand household health risks in the absence of a direct role in caregiving. Conversely, health insurance enrolment amongst women is correlated with the need to minimize potential household health risks based on their direct knowledge of household healthcare needs. The impact of education on male health insurance demand underscores the importance of simple and digestible sensitization programmes as a means of facilitating household decision-makers’ understanding of the concept of health insurance. Further, the influence of wealth on health insurance enrolment particularly amongst female-headed households provides important insight from a policy perspective, given the limited availability of subsidies to cover vulnerable populations in many sub-Saharan African health insurance schemes. Our findings highlight the need for partnerships between health insurance schemes and governments in order to develop a financial safety net to limit the impact of illness on the poor. Indeed, the ultimate goal of voluntary health insurance schemes such as BEPHA is to support governments in their quest towards achieving universal health coverage. It is through collaboration and responsiveness to the social, cultural and economic realities of the target population that health insurance coverage will reach sustainable levels. Acknowledgements The authors thank Felicia Ekpenyong Isang and the Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) Kumbo team for their support during the data collection process. The authors also thank Dr. Amanda Ross (Swiss TPH) for statistical support provided. Funding Data collection was financed by the Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) Provincial Office. T.O. is a recipient of the Swiss Government Excellence Scholarship (ESKAS), provided by the Swiss Federal Commission for Scholarships for Foreign Students (FCS). Conflict of interest statement. None declared. References Acharya A, Vellakkal S, Taylor F et al.   2013. The impact of health insurance schemes for the informal sector in low- and middle-income countries: a systematic review. World Bank Research Observer  28: 236– 66. Google Scholar CrossRef Search ADS   Arhin-Tenkorang D. 2001. Health insurance for the informal sector in Africa: design features, risk protection, and resource mobilization, Health, Nutrition, and Population Discussion Paper. The International Bank for Reconstruction and Development. The World Bank, Washington, DC. Basaza R, Criel B, Van der Stuyft P. 2007. Low enrolment in Ugandan Community Health Insurance Schemes: underlying causes and policy implications. BMC Health Services Research  7: 105. Google Scholar CrossRef Search ADS PubMed  Basaza R, Criel B, Van der Stuyft P. 2008. Community health insurance in Uganda: why does enrolment remain low? A view from beneath. Health Policy  87: 172– 84. Google Scholar CrossRef Search ADS PubMed  Beegle K, Christiaensen L, Dabalen A, Gaddis I. 2016. Poverty in a rising Africa. The World Bank. doi: 10.1596/978-1-4648-0723-7 BEPHA. 2016. BEPHA Membership Audit FY 2015–2016. Bamenda Ecclesiastical Provincial Health Assistance, Bamenda. Cameroun Institut National de la Statistique. 2011. Cameroon Wealth Index. Chuma J, Mulupi S, McIntyre D. 2013. Providing financial protection and funding health service benefits for the informal sector: evidence from sub-Saharan Africa, RESYST Working Paper 2. KEMRI, Nairobi. Colvin CJ, Smith HJ, Swartz A et al.   2013. Understanding careseeking for child illness in sub-Saharan Africa: a systematic review and conceptual framework based on qualitative research of household recognition and response to child diarrhoea, pneumonia and malaria. Social Science and Medicine  86: 66– 78. Google Scholar CrossRef Search ADS PubMed  Criel B, Van Dormael M, Lefevre P, Menase U, Van Lerberghe W. 1998. Voluntary health insurance in Bwamanda, Democratic Republic of Congo. An exploration of its meanings to the community. Tropical Medicine & International Health  3: 640– 53. Google Scholar CrossRef Search ADS   De Allegri M, Kouyate B, Becher H et al.   2006a. Understanding enrolment on community health insurance on sub-Saharan Africa: a population-based case-control study on rural Burkina Faso. Bulletin of the World Health Organization  84: 852– 8. Google Scholar CrossRef Search ADS   De Allegri M, Sanon M, Sauerborn R. 2006. “To enrol or not to enrol?”: a qualitative investigation of demand for health insurance in rural West Africa. Social Science and Medicine  62: 1520– 7. Google Scholar CrossRef Search ADS PubMed  Dixon J, Luginaah I, Mkandawire P. 2014. The National Health Insurance Scheme in Ghana‘s Upper West Region: a gendered perspective of insurance acquisition in a resource-poor setting. Social Science and Medicine  122: 103– 12. Google Scholar CrossRef Search ADS PubMed  Giedion A, Alfonso EA, Diaz Y. 2013. The Impact of Universal Coverage Schemes in the Developing World: a review of the existing evidence (No. UNICO Studies Series No. 25). Universal Health Coverage Studies Series (UNICO). The World Bank, Washington, D.C. Goheen M. 1995. Gender and accumulation in Nso’. Paideuma  41: 73– 81. Hayes RJ, Bennett S. 1999. Simple sample size calculation for cluster-randomized trials. International Journal of Epidemiology  28: 319– 26. Google Scholar CrossRef Search ADS PubMed  Institut National de la Statistique, 2012. Enquète Demographique et de Santé et à Indicateurs Multiples. République du Cameroun, Yaoundé. Institut National de la Statistique, 2008. Conditions de Vie des Populations et Profil de Pauvreté au Cameroun en 2007: Rapport Principal de l’ECAM3. République du Cameroun, Yaoundé. Institut National de la Statistique du Cameroun, 2014. La quatrième Enquête Camerounaise Auprès des Ménages (ECAM 4). République du Cameroun, Yaoundé. International Organization for Migration (Ed.), 2013. Migrant well-being and development, World Migration Report. IOM, Geneva. Jehu-Appiah C, Aryeetey G, Agyepong I, Spaan E, Baltussen R. 2012. Household perceptions and their implications for enrolment in the National Health Insurance Scheme in Ghana. Health Policy Planning  27: 222– 33. Google Scholar CrossRef Search ADS PubMed  Jehu-Appiah C, Aryeetey G, Spaan E et al.   2011. Equity aspects of the National Health Insurance Scheme in Ghana: who is enrolling, who is not and why? Social Science & Medicine  72: 157– 65. Google Scholar CrossRef Search ADS   Kimani JK, Ettarh R, Kyobutungi C, Mberu B, Muindi K. 2012. Determinants for participation in a public health insurance program among residents of urban slums in Nairobi, Kenya: results from a cross-sectional survey. BMC Health Services Research  12: 66. Google Scholar CrossRef Search ADS PubMed  Kimani JK, Ettarh R, Warren C, Bellows B. 2014. Determinants of health insurance ownership among women in Kenya: evidence from the 2008–09 Kenya Demographic and Health Survey. International Journal for Equity in Health  13: 27. Google Scholar CrossRef Search ADS PubMed  Kishor S, Nietzel K. 1997. The Status of Women: Indicators for Twenty-Five Countries - Gender and Headship, DHS Comparative Studies. Macro International Inc., Calverton, Maryland. Lagomarsino G, Garabrant A, Adyas A, Muga R, Otoo N. 2012. Moving towards universal health coverage: health insurance reforms in nine developing countries in Africa and Asia. The Lancet  380: 933– 43. Google Scholar CrossRef Search ADS   Leive A, Xu K. 2008. Coping with out-of-pocket health payments: empirical evidence from 15 African countries. Bulletin of the World Health Organization  86: 849– 56. Google Scholar CrossRef Search ADS PubMed  Macha J, Kuwawenaruwa A, Makawia S, Mtei G, Borghi J. 2014. Determinants of community health fund membership in Tanzania: a mixed methods analysis. BMC Health Services Research  14: 538. Google Scholar CrossRef Search ADS PubMed  Mebratie AD, Sparrow R, Yilma Z, Alemu G, Bedi AS. 2015. Enrollment in Ethiopia‘s community-based health insurance scheme. World Development  74: 58– 76. Google Scholar CrossRef Search ADS   Milazzo A, Van de Walle DP. 2015. Women left behind? Poverty and headship in Africa. Poverty Headship Afr. June 23 2015 World Bank Policy Research Working Paper. Mladovsky P, Ndiaye P, Ndiaye A, Criel B. 2015. The impact of stakeholder values and power relations on community-based health insurance coverage: qualitative evidence from three Senegalese case studies. Health Policy Planning  30: 768– 81. Google Scholar CrossRef Search ADS PubMed  Mulupi S, Kirigia D, Chuma J. 2013. Community perceptions of health insurance and their preferred design features: implications for the design of universal health coverage reforms in Kenya. BMC Health Services Research  13: 474. Google Scholar CrossRef Search ADS PubMed  Musango L, Dujardin B, Dramaix M, Criel B. 2004. Profile of members and non members of mutual health insurance system in Rwanda: the case of the health district of Kabutare. Tropical Medicine & International Health TMIH  9: 1222– 7. Google Scholar CrossRef Search ADS   Onwujekwe O, Onoka C, Uguru N et al.   2011. Socio-economic and geographic differences in acceptability of community-based health insurance. Public Health  125: 806– 8. Google Scholar CrossRef Search ADS PubMed  Parmar D, De Allegri M, Savadogo G, Sauerborn R. 2014. Do community-based health insurance schemes fulfil the promise of equity? A study from Burkina Faso. Health Policy Planning  29: 76– 84. Google Scholar CrossRef Search ADS PubMed  Parmar D, Reinhold S, Souares A, Savadogo G, Sauerborn R. 2012. Does community-based health insurance protect household assets? Evidence from rural Africa. Health Services Research  47: 819– 39. Google Scholar CrossRef Search ADS PubMed  République du Cameroun, Agence Régionale du Nord-Ouest, 2015. Annuaire Statistique de la Région du Nord-Ouest 2014. Richards E, Theobald S, George A et al.   2013. Going beyond the surface: gendered intra-household bargaining as a social determinant of child health and nutrition in low and middle income countries. Social Science Medicine  95: 24– 33. Google Scholar CrossRef Search ADS PubMed  Schneider P. 2004. Why should the poor insure? Theories of decision-making in the context of health insurance. Health Policy Planning  19: 349– 55. Google Scholar CrossRef Search ADS PubMed  Soors W, Devadasan N, Durairaj V, Criel B. 2010. Community health insurance and universal coverage: multiple paths, many rivers to cross. World Health Report. United Nations General Assembly. 2012. Global Health and Foreign Policy. Agenda Item 123. Wagstaff A. 2008. Measuring financial protection in health, Policy Research Working Papers. The World Bank. doi:10.1596/1813-9450-4554 World Bank, JICA, The Global Fund, African Development Bank, WHO. 2016. Universal health coverage in Africa: a framework for action, in: Joint 2016 TICAD Conference Working Paper. World Health Assembly. 2005. Resolution WHA58.33: sustainable health financing, universal coverage, and social health insurance. Presented at the Fifty-Seventh World Health Assembly. World Health Organization, 1992. International Statistical Classification of Diseases and Related Health Problems, 10th ed. World Health Organization, Geneva. World Health Organization (Ed.). 2014. The health of the people: what works: the African Regional Health Report 2014. World Health Organization, Regional Office for Africa, Brazzaville, Republic of Congo. World Health Organization. 2010. The world health report: health systems financing: the path to universal coverage. World Health Organization, Geneva. © The Author 2017. Published by Oxford University Press in association with The London School of Hygiene and Tropical Medicine. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Health Policy and Planning Oxford University Press

The influence of gender and household headship on voluntary health insurance: the case of North-West Cameroon

Loading next page...
 
/lp/ou_press/the-influence-of-gender-and-household-headship-on-voluntary-health-JXP0Ou04cJ
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press in association with The London School of Hygiene and Tropical Medicine. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0268-1080
eISSN
1460-2237
D.O.I.
10.1093/heapol/czx152
Publisher site
See Article on Publisher Site

Abstract

Abstract Within the existing health financing literature, males are typically categorized as the household‘s decision-makers. While this view accurately reflects many local sociocultural realities, approximately a quarter of sub-Saharan African households are now headed by females. In light of various efforts to expand health insurance coverage in the region, it is necessary to examine whether the factors influencing voluntary health insurance enrolment are analogous across male- and female-headed households. This study sought to identify the gendered determinants of voluntary enrolment into a church-run micro health insurance scheme. A cross-sectional survey of 550 households was carried out in Bui and Donga-Mantung Divisions of North-West Cameroon in May 2016. A structured questionnaire was administered on health insurance membership, household attributes, headship characteristics and health-seeking behaviour. We assessed the influence of gender on the associations between health insurance enrolment and the explanatory variables using logistic regression. This study found that voluntary health insurance demand was influenced by involvement in social networks regardless of gender. However, in line with entrenched household roles, men’s understanding of potential household health risks ultimately facilitated their enrolment decisions, while economically empowered women prioritised their direct knowledge of household health risks. Men’s demand for health insurance was correlated primarily with their education level (OR = 2.238 [CI 1.228–2.552]), as well as with their socioeconomic status (OR = 2.207 [CI 1.173–4.153]), age (OR = 2.238 [CI 1.151–4.352]) and trust of the insurance provider (OR = 4.770 [CI 2.407–9.453]). Conversely, women’s enrolment decision was primarily associated with their income levels (OR = 5.842 [CI 1.589–21.484]), as well as by the presence of children (OR = 3.734 [CI 1.228–11.348]). The influence of wealth on health insurance enrolment highlights the need for policymakers to subsidize health insurance schemes for vulnerable population groups. Further, it is imperative to develop sensitization campaigns that are simple and digestible to facilitate understanding of health insurance across all target groups. Voluntary health insurance, decision making, household headship, gender Key Messages Household heads are more likely to enrol into voluntary health insurance if involved in social networks, regardless of gender. Women prioritize their direct knowledge of potential household health risks when evaluating the decision to enrol into voluntary health insurance. However, income ultimately determines women’s ability to participate in health insurance schemes. When purchasing health insurance, men prioritize on their understanding of household health risks, which is linked to their education and age. Introduction Within the existing health financing research, males are typically categorized as the gatekeepers in the decision to enrol into voluntary health insurance. While this view accurately reflects local sociocultural realities in the developing world, many sub-Saharan African countries have been experiencing an increase in female-led households: In Cameroon, for example, women now head 25% of households (Institut National de la Statistique 2012). This necessitates an examination of whether the factors influencing voluntary health insurance enrolment are analogous across male- and female-headed households. This study seeks to identify if and how the gender of the household head influences enrolment into a voluntary micro health insurance scheme in North-West Cameroon. It further aims to analyse the influence, if any, of entrenched gender household roles on the decision to enrol. In many low- and middle-income countries, there is increasing recognition of the debilitating financial impact of illness on households (World Health Assembly 2005; United Nations General Assembly 2012; World Bank et al. 2016). Against this backdrop, governments and health financing stakeholders have endorsed prepaid healthcare as the preferred approach for reducing out-of-pocket expenditure and facilitating access to health services (World Health Organization 2010). This has largely taken the form of risk-pooling mechanisms in developing countries, with many governments striving to develop functional and inclusive health insurance schemes (Soors et al. 2010; Lagomarsino et al. 2012; Giedion et al. 2013). However, due to budgetary constraints and complexities in identifying the informal sector, many African governments have been unable to achieve large-scale health insurance coverage (Arhin-Tenkorang 2001; Acharya et al. 2013; Chuma et al. 2013). This has led to the proliferation of private micro health insurance schemes targeting those who would otherwise suffer from a disproportionate risk of impoverishment due to illness (Leive and Xu 2008; Wagstaff 2008). Crucially, these schemes are voluntary, and depend upon an active decision by households to purchase the health benefit package. In order to better understand the factors which influence one’s likelihood to seek health insurance coverage, a range of studies have investigated the determinants of voluntary enrolment across various African settings. These studies have largely focused on identifying the individual and household drivers of enrolment, with household size and composition; socioeconomic status (SES); and education level emerging as major determinants of voluntary health insurance enrolment (Criel et al. 1998; Musango et al. 2004; De Allegri et al. 2006a; Basaza et al. 2008; Jehu-Appiah et al. 2011, 2012; Onwujekwe et al. 2011; Kimani et al. 2012, 2014; Parmar et al. 2012; Mulupi et al. 2013; Dixon et al. 2014; Macha et al. 2014; Mebratie et al. 2015; ). Most of these studies have approached the analysis of health financing decision-making from a patriarchal point of view, reflecting the established sociocultural conventions in the study areas. This approach, however, may no longer be fully representative of current household structures, given that 26% of sub-Saharan African households are now headed by females (Beegle et al. 2016). The increase in female-led households in various sub-Saharan African countries may be attributed to several documented patterns in migration and epidemiology: widespread male economic migration to urban areas has resulted in de facto female household headship in many rural areas (International Organization for Migration 2013; Milazzo and Van de Walle 2015). Further, females are likely to live longer than males despite higher reported incidences of chronic diseases and exposure to death through childbirth (World Health Organization 2014). These realities underscore the need to explore the driving factors of voluntary health insurance enrolment not just from the perspective of normative social convention, but also with a pragmatic view of existing household headship structures. In this regard, there remains a dearth of literature on the role of women as primary agents in the decision to enrol into voluntary health insurance. Most relevant studies have investigated the ability of health insurance schemes to reach women as a vulnerable demographic, but not as the decision-makers driving the demand for health insurance (Kimani et al. 2014; Parmar et al. 2014). To the authors’ knowledge, only a single study has been carried out investigating the gendered determinants of voluntary health insurance enrolment in the African context. Dixon et al. carried out a comparative analysis of the factors associated with voluntary health insurance in Ghana, and found crucial educational, socioeconomic and marital differences between male and female individuals (Dixon et al. 2014). While this work presents an important first step in the analysis of the gender dynamics in health insurance demand, Dixon et al. did not consider the role of gendered household headship in the decision to enrol. This means that there is still a significant knowledge gap on the influence of gender on voluntary health insurance demand when decision-making autonomy is acquired. Precursory insight into the relationship between gender and healthcare decision-making may be gained from health-seeking behaviour literature, where several reviews have investigated the intersection between household roles and the decision to seek external health care. A systematic review by Colvin et al. on health-seeking behaviour found that timely treatment of sick household members is inextricably linked to the level of influence held by the mother on the final decision to seek external care (Colvin et al. 2013). This study also noted that male partners are typically only involved in the care-seeking pathway as the decision-makers of the economic and medical case for seeking outside care. Congruently, a global review of morbidity and mortality literature found that autonomy amongst mothers was associated with better overall health status amongst children in the household (Richards et al. 2013). These studies allude to potential gendered differences in the consideration of health-related decisions: they suggest that the divisional roles intrinsic within the household lead men and women to evaluate health risks differently. Based on this, we hypothesize that women are likely to prioritize their direct knowledge of the household’s potential healthcare needs when making healthcare decisions. Conversely, we hypothesize that men are likely to prioritize their understanding of potential health risks associated with their households. Materials and methods Research setting The Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) scheme is a micro health insurance scheme set up by the Roman Catholic Church in North-West and South-West Cameroon. It consists of four independently-operated schemes in Bamenda, Buea, Kumbo and Mamfe Parishes covering a total of 35 224 individuals as of November 2016 (BEPHA 2016). For the purposes of this study, we will be focusing on the BEPHA Kumbo Scheme. The BEPHA Kumbo Scheme is active in the Bui and Donga-Mantung administrative divisions of North-West Cameroon. The North-West region is plagued by poor health outcomes, including the highest HIV prevalence rate in Cameroon (8.7%). It also suffers from a high malaria prevalence of 20% (Institut National de la Statistique du Cameroun 2014). Poor health outcomes in this region are exacerbated by steep out-of-pocket costs: according to the country’s Demographic and Health Survey (DHS), North-West Cameroon has the second-highest level of health spending in Cameroon in the event of illness, in spite of accounting for 13% of the country’s poor (Institut National de la Statistique 2008, 2012). Membership of the BEPHA Kumbo Scheme is voluntary with a minimum enrolment unit of four individuals. The annual premium per individual is set at FCFA 4000 (US$6.80 in May 2016) and is paid in a maximum of three instalments. BEPHA annual coverage starts in either June or November. However, coverage within educational institutions begins in October in line with the annual academic calendar. The BEPHA benefit package covers three-quarters of the cost of inpatient services, delivery services, outpatient services and surgery, respectively. Enrolees are permitted to access BEPHA inpatient and outpatient care benefits twice annually, while surgery and maternity services can only be reimbursed once annually. BEPHA applies maximum cost ceilings for each of the offered services in order to ensure financial sustainability: FCFA 15 000 (US$25.50) for outpatient services and maternity services, respectively; FCFA 25 000 (US$42.50) for inpatient services; and FCFA 70 000 (US$119.00) for surgery. Study design A cross-sectional household survey was carried out in the Bui and Donga-Mantung divisions of North-West Cameroon between April and May 2016. All administrative sub-divisions of Bui and Donga-Mantung were eligible for sampling. For operational reasons, the study was conducted in the sub-divisions where >30 households were enrolled into the BEPHA Kumbo Scheme. The study had two distinct aims: (1) to estimate the proportion of the population with BEPHA health insurance and (2) to assess the effect of explanatory variables on this proportion. For the first aim, we based the sample size on the ability to estimate the proportion insured with a certain precision—in this case, with an expected BEPHA population coverage of 4%. This coverage rate was obtained from the BEPHA membership audit carried out in FY 2015–2016 (BEPHA 2016). Using the Hayes and Bennett equation and taking into account clustering in nine sub-districts, the required sample size for the first study aim was calculated to be 416 households with the precision of a 95% confidence interval with width of 2–6% (Hayes and Bennett 1999). The number of households interviewed in each sub-division was proportional to its demographic size (République du Cameroun and Agence Régionale du Nord-Ouest 2015), and we assumed that the measure of variability between clusters k, the standard deviation divided by the mean, was equal to 0.1 (Hayes and Bennett 1999). As a sampling frame was not readily available for the study area, we randomly selected starting points within outposts in each sub-division. We subsequently carried out systematic sampling, with every nth household interviewed. Approximately 98% of the targeted households could be interviewed. In order to achieve the second study aim of estimating the factors associated with health insurance enrolment, we sampled additional insured households from the BEPHA membership list. BEPHA’s internal client management system stratifies insured households according to their sub-division of residence and randomly assigns membership numbers. In order to optimize our sample of insured households to address the second study aim, we selected every nth number from the BEPHA membership list. In total, 174 insured households were randomly selected stratifying by sub-division, from the BEPHA Kumbo membership list. These households were chosen from the same sub-divisions within which the random population sampling was conducted. Approximately 80% of these households could be identified and interviewed. Survey tools and data analysis The survey was administered as a structured questionnaire using Open Data Kit software on handheld tablets to answer questions on household composition; household assets; household expenditure and consumption; and health-seeking behaviour. Research assistants who were fluent in Pidgin English and the local dialect, Lam Nso’, and who had knowledge of the local geographical and sociocultural context were hired. Training was provided to familiarize research assistants with the questionnaire and data collection using handheld tablets. Insured households were defined as those where at least one member was enrolled into the BEPHA Kumbo Scheme in the year preceding the study. The household head was defined as the household member whose income contributed to at least half of the household’s costs in the preceding year. The household head’s gender was defined as the self-reported sex of the household head. De facto female household headship was assumed in households where no male adults lived for the preceding year. The data were analyzed using STATA version 14.1 for Windows (STATA Corporation, College Station, Texas). The main outcome variable was health insurance enrolment into the BEPHA Kumbo Scheme. The explanatory variables were divided into five components: household composition and attributes, household head factors, mean perceived household health status and proxies for exposure to financial risk pooling. In order to assess asset-based wealth, we constructed an asset index from the Cameroonian Demographic and Health Survey (DHS) index which measures the relative wealth ranking of households (Cameroun Institut National de la Statistique 2011). Household per capita consumption was calculated as the annual food, non-food, and consumer durables consumption per household member. Mean perceived health status was calculated as the average self-reported health status value for all household members. The final value of the mean perceived health status was assigned a value of between 1 and 4, with 1 being ‘very good’ health status; 2 being ‘good’ health status; 3 being ‘poor’ health status; and 4 being ‘very poor’ health status. Chronic disease status in the context of this study was attributed on the basis of the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) (World Health Organization 1992). We used logistic regression to estimate the association between the explanatory variables and health insurance enrolment, with random effects for sub-division and outpost to account for the clustering in the sample. Our study utilized two regression models. The first model measured the association of each variable and health insurance enrolment within each gender group. The second model included interaction terms for gender and each explanatory variable to determine the influence of gender on the association of each variable and health insurance enrolment. Ethical approval Ethical approval for this study was obtained through the Institutional Review Board (IRB) of the Catholic University of Cameroon in Bamenda, North-West Cameroon (Ethics Reference No. 001/HEPM/CATUC-IRB/16; Date of approval: 17th May 2016). Results Study population characteristics Table 1 reports the differences in each test variable according to insurance status and gender of the household head. A total of 550 households were enumerated in the survey, with male-headed households accounting for 70% of the study population. 80% of male-headed and 89% of female-headed households lived in rural areas, reflecting the largely rural nature of the population in the Bui and Donga-Mantung Divisions. Approximately 28% of male- and 36% of women-headed households were enrolled into BEPHA micro health insurance. Table 1. Descriptive statistics by gender and insurance status   Male   Female   Insured (n = 105)  Uninsured (n = 275)  Combined (n = 380)  Insured (n = 61)  Uninsured (n = 109)  Combined (n = 170)  Household characteristics      Urban location  27%  18%  20%  11%  11%  11%      Mean household sizea  5.81 (2.33)  5.66 (2.79)  5.70 (2.67)  4.95 (2.17)  4.43 (2.48)  4.61 (2.38)      Household sex ratio (males to females)a  1.194 (0.969)  1.273 (0.979)  1.250 (0.974)  0.762 (0.788)  0.631 (0.669)  0.674 (0.714)      Households with children ≤5 years  57%  60%  59%  41%  51%  48%      Households with children ≤15 years  85%  85%  85%  90%  75%  81%      Households with elderly ≥60 years  32%  26%  28%  34%  28%  30%      Average SES scoreb  0.492 (0.965)  0.317 (0.834)  0.365 (0.875)  0.318 (0.674)  0.189 (0.677)  0.236 (0.677)      % of poor households (Q1–Q2)  36%  40%  39%  39%  45%  43%      % of rich households (Q5)  30%  20%  23%  20%  9%  13%  Household head characteristics      Average ageb  49 (13.23)  47 (15.11)  48 (14.60)  49 (12.46)  47 (16.00)  48 (14.95)      Literate  94%  76%  81%  77%  63%  68%      Primary-level education  50%  41%  44%  54%  40%  45%      Secondary-level education  21%  24%  23%  18%  19%  19%      Higher-level education  23%  9%  13%  5%  3%  4%      Married household heads  94%  90%  91%  28%  27%  27%      Low income (≤47 000 FCFA)  50%  48%  49%  57%  60%  59%      Proportion of Catholic faith  87%  57%  66%  87%  61%  71%  Household health status      Mean health statusa  1.96 (0.672)  1.96 (0.67)  1.96 (0.67)  2.01 (0.76)  1.86 (0.74)  1.92 (0.76)      Presence of chronic disease  8%  6%  6%  5%  5%  5%  Use of curative services      Outpatient service use in past 30 days  42%  35%  37%  33%  25%  28%      Inpatient service use in past year  42%  35%  37%  31%  28%  29%      Annual per capita health expenditure (FCFA)b  5000 (10 500)  4500 (11 750)  5000 (11 634)  5000 (9500)  5000 (12 500)  5000 (12 500)    Male   Female   Insured (n = 105)  Uninsured (n = 275)  Combined (n = 380)  Insured (n = 61)  Uninsured (n = 109)  Combined (n = 170)  Household characteristics      Urban location  27%  18%  20%  11%  11%  11%      Mean household sizea  5.81 (2.33)  5.66 (2.79)  5.70 (2.67)  4.95 (2.17)  4.43 (2.48)  4.61 (2.38)      Household sex ratio (males to females)a  1.194 (0.969)  1.273 (0.979)  1.250 (0.974)  0.762 (0.788)  0.631 (0.669)  0.674 (0.714)      Households with children ≤5 years  57%  60%  59%  41%  51%  48%      Households with children ≤15 years  85%  85%  85%  90%  75%  81%      Households with elderly ≥60 years  32%  26%  28%  34%  28%  30%      Average SES scoreb  0.492 (0.965)  0.317 (0.834)  0.365 (0.875)  0.318 (0.674)  0.189 (0.677)  0.236 (0.677)      % of poor households (Q1–Q2)  36%  40%  39%  39%  45%  43%      % of rich households (Q5)  30%  20%  23%  20%  9%  13%  Household head characteristics      Average ageb  49 (13.23)  47 (15.11)  48 (14.60)  49 (12.46)  47 (16.00)  48 (14.95)      Literate  94%  76%  81%  77%  63%  68%      Primary-level education  50%  41%  44%  54%  40%  45%      Secondary-level education  21%  24%  23%  18%  19%  19%      Higher-level education  23%  9%  13%  5%  3%  4%      Married household heads  94%  90%  91%  28%  27%  27%      Low income (≤47 000 FCFA)  50%  48%  49%  57%  60%  59%      Proportion of Catholic faith  87%  57%  66%  87%  61%  71%  Household health status      Mean health statusa  1.96 (0.672)  1.96 (0.67)  1.96 (0.67)  2.01 (0.76)  1.86 (0.74)  1.92 (0.76)      Presence of chronic disease  8%  6%  6%  5%  5%  5%  Use of curative services      Outpatient service use in past 30 days  42%  35%  37%  33%  25%  28%      Inpatient service use in past year  42%  35%  37%  31%  28%  29%      Annual per capita health expenditure (FCFA)b  5000 (10 500)  4500 (11 750)  5000 (11 634)  5000 (9500)  5000 (12 500)  5000 (12 500)  a Results reported as mean (standard deviation). b Results reported as median (interquartile range). Mean exchange rate for May 2016: US$1 = FCFA 588.24. Table 2. Logistic estimates for probability of purchasing BEPHA health insurance at household level Variable description  Male   Female   Interaction   Odds ratio  CI  Odds ratio  CI  P-value  Household characteristics    Location (reference group: Rural)      Urban  1.378  0.646–2.939  1.063  0.211–5.354  0.78    Number of children <5 years (Reference group: None)      One to two children  0.826  0.503–1.357  0.977  0.451–2.116  0.67      Three or more children  0.933  0.373–2.333  0.151  0.168–1.357  0.13      Presence of children <15 years  1.143  0.567–2.306  3.734  1.228–11.348  0.051      Presence of elderly household members  1.301  0.756–2.269  1.072  0.484–2.376  0.83    SES (reference group: Poor)      Middle wealth  0.722  0.431-1.210  0.760  0.355–1.630  0.78       Wealthiest quintile  2.207  1.173–4.153  2.698  0.825–8.826  0.94  Household head characteristics    Age (reference group: under 35)      35–50 years  1.223  0.743–2.013  1.543  0.728–3.271  0.46      Older than 50 years  2.238  1.151–4.352  1.916  0.697–5.264  0.82    Education (reference group: primary education or less)      Secondary education or more  1.770  1.228–2.552  1.554  0.903–2.675  0.008    Monthly income (reference group: no income)     <23 500 FCFA  0.576  0.322–1.029  0.559  0.257–1.214  0.93      23 500–47 000 FCFA  1.658  0.902–3.047  1.883  0.707–5.020  0.86     >47 000 FCFA  1.401  0.810–2.425  5.842  1.589 –21.484  0.037    Married  1.898  0.709–5.081  1.259  0.542–2.927  0.56    Catholic religion  4.770  2.407–9.453  2.926  1.076–7.953  0.33  Household health status    Mean perceived household health status      Good  1.043  0.621–1.751  0.860  0.399–1.854  0.49      Poor  0.959  0.571–1.601  1.163  0.539–2.507  0.49    Presence of chronic illness in household  1.519  0.574–4.018  0.795  0.128–4.924  0.99  Exposure to concept of financial risk protection    Member of informal savings group (njangi)  2.301  1.302–4.064  3.146  1.429–6.921  0.42    Trust BEPHA health insurance provider  5.170  2.365–11.301  2.830  0.884–9.061  0.23  Variable description  Male   Female   Interaction   Odds ratio  CI  Odds ratio  CI  P-value  Household characteristics    Location (reference group: Rural)      Urban  1.378  0.646–2.939  1.063  0.211–5.354  0.78    Number of children <5 years (Reference group: None)      One to two children  0.826  0.503–1.357  0.977  0.451–2.116  0.67      Three or more children  0.933  0.373–2.333  0.151  0.168–1.357  0.13      Presence of children <15 years  1.143  0.567–2.306  3.734  1.228–11.348  0.051      Presence of elderly household members  1.301  0.756–2.269  1.072  0.484–2.376  0.83    SES (reference group: Poor)      Middle wealth  0.722  0.431-1.210  0.760  0.355–1.630  0.78       Wealthiest quintile  2.207  1.173–4.153  2.698  0.825–8.826  0.94  Household head characteristics    Age (reference group: under 35)      35–50 years  1.223  0.743–2.013  1.543  0.728–3.271  0.46      Older than 50 years  2.238  1.151–4.352  1.916  0.697–5.264  0.82    Education (reference group: primary education or less)      Secondary education or more  1.770  1.228–2.552  1.554  0.903–2.675  0.008    Monthly income (reference group: no income)     <23 500 FCFA  0.576  0.322–1.029  0.559  0.257–1.214  0.93      23 500–47 000 FCFA  1.658  0.902–3.047  1.883  0.707–5.020  0.86     >47 000 FCFA  1.401  0.810–2.425  5.842  1.589 –21.484  0.037    Married  1.898  0.709–5.081  1.259  0.542–2.927  0.56    Catholic religion  4.770  2.407–9.453  2.926  1.076–7.953  0.33  Household health status    Mean perceived household health status      Good  1.043  0.621–1.751  0.860  0.399–1.854  0.49      Poor  0.959  0.571–1.601  1.163  0.539–2.507  0.49    Presence of chronic illness in household  1.519  0.574–4.018  0.795  0.128–4.924  0.99  Exposure to concept of financial risk protection    Member of informal savings group (njangi)  2.301  1.302–4.064  3.146  1.429–6.921  0.42    Trust BEPHA health insurance provider  5.170  2.365–11.301  2.830  0.884–9.061  0.23  The descriptive statistics for the sample population were similar across gender lines in terms of location, and the household head’s age, self-employment status and religion. Overall, male and female household heads had an average age of 48.0 years, with a self-employment rate of 65%. The sample was predominantly Catholic (67%), highlighting BEPHA’s affiliation with the Church. The gender groups also showed similarities in their reported health status, with 6% of all households reporting the presence of chronic disease. We found important distinctions between male- and female-headed households in several household composition and socioeconomic characteristics. Male-headed households had more uniformity in the number of males and females within the household compared with their female counterparts. They were also more likely to be in the wealthiest socioeconomic quintile and have higher incomes than female-headed households. Female-headed households, on the other hand, were typically smaller in size, had more female than male household members and were more likely to be located in a rural setting. Households also differed in their marital status, with 91% of male household heads living within a marriage setting compared with 40% of female household heads who were widowed. Factors associated with voluntary health insurance enrolment amongst male- and female-headed households Household-level regression results are presented in Table 2. This study identified several commonalities amongst male- and female-headed households in the characteristics associated with health insurance demand. Regardless of gender, we found that Catholic Church membership was correlated to one’s likelihood of enrolling into BEPHA health insurance (OR = 4.770 [CI 2.407–9.453] for males and OR = 2.926 [CI 1.076–7.953] for females). Practices in social solidarity were also found to be associated with BEPHA health insurance enrolment across genders: both male and female household heads showed higher odds of purchasing health insurance if they belonged to informal savings groups known as njangis (OR = 2.301 [CI 1.302–4.064] for males and OR = 3.146 [CI 1.429–6.921] for females). In spite of these broad convergences, we found that the influencing factors of BEPHA insurance enrolment differed depending on the gender of the household head. Amongst male household heads, demand for health insurance was correlated to their SES, age, education level and their trust of the insurance provider. This study found that men belonging to the highest asset-based wealth quintile had 2.207 higher odds of enrolling into health insurance compared to those with lower asset-based wealth (CI 1.173–4.153). Older age was also identified as a significant contributor to health insurance enrolment, with male household heads aged above 50 years showing increased health insurance demand compared with their younger counterparts (OR = 2.238 [CI 1.151–4.352]). Further, men educated to secondary school-level and above showed 2.238 higher odds of enrolment compared with those with lower education levels (CI 1.228–2.552). Finally, we found that male household heads were more likely to enrol if they expressed confidence in the Church as a potential health insurance provider (OR = 4.770 [CI 2.407–9.453]). Amongst female-headed households, the decision to enrol into BEPHA health insurance was associated with household composition and income. This study found that demand for BEPHA health insurance amongst female-headed households was correlated to the presence of children below the age of 15 (OR = 3.734 [CI 1.228–11.348]). Women who earned higher incomes were also more likely to purchase BEPHA health insurance: the study found that women who earned earning above 47 000 FCFA had 5.842 higher odds of enrolling into health insurance than those earning less (CI 1.589–21.484). When the interactions between gender, health insurance membership and the explanatory variables were taken into account, we found that Catholic Church and njangi membership was correlated to health insurance demand in both male- and female-headed households (P-value = 0.332 for Catholic Church membership, and P-value = 0.421 for njangi membership, respectively). Income and education, however, were found to have varied associations with the decision to enrol into health insurance depending on the gender of the household head. Amongst female-headed households, we found that health insurance enrolment was associated with high income levels (P-value = 0.037)—a finding that was non-significant amongst male-headed households. Conversely, male-headed households with higher levels of education were found to have higher demand for BEPHA health insurance (P-value = 0.008). This finding was non-significant amongst female-headed households. We found no further evidence of a relationship between the gender of the household head, BEPHA membership and other explanatory variables. Discussion In this study, we set out to expand the narrative on decision-making in voluntary health insurance by looking at how the gender of a household head influences voluntary health insurance enrolment. It is important to acknowledge from the outset that commonalities exist between male- and female-headed households, with the overarching influence of social networks on health insurance demand. Based on our findings, it is clear that involvement in networks that encourage solidarity and reciprocity, such as njangis and the Catholic Church, increase one’s likelihood of enrolling into voluntary health insurance regardless of gender. This reflects existing studies which have highlighted the importance of social solidarity as a key determinant of voluntary health insurance schemes demand (Basaza et al. 2007; Mladovsky et al. 2015). This notwithstanding, it is apparent that the decision to enrol into voluntary health insurance is multidimensional and complex, driven in part by aversion to the risk of illness (Schneider 2004). We propose that the evaluation of health risks manifests differently between male and female household heads due to their unique household roles in the study setting. We therefore postulate that women prioritize their direct knowledge of the household’s healthcare needs in the decision to enrol into health insurance. Conversely, we suggest that men prioritize their understanding of potential health risks associated with their households. How does this reconcile with existing knowledge of health insurance and the everyday realities of household power structures? Within existing literature, it is acknowledged that the burden of caregiving is borne primarily by women—a dynamic that is particularly pronounced in the occurrence of illness in the domestic framework (Colvin et al. 2013; Richards et al. 2013). This underlying knowledge of the physical, psychological and economic cost of illness elevates the female voice as imperative in understanding and assessing household health risks. In the traditional hierarchy dominant within North-West Cameroon, however, female influence is often relegated in decision-making due to their subordinate status (Goheen 1995). This means that, in the absence of direct knowledge of household health needs, male household heads are compelled to act on the basis of their knowledge and understanding of potential household health risks. Our findings suggest that this is primarily shaped by their level of education, which aligns to other health financing studies across Africa (De Allegri et al. 2006a; Jehu-Appiah et al. 2012; Kimani et al., 2012, 2014; Dixon et al. 2014). We posit that an advanced level of education enables male household heads to better evaluate the potential social and economic implications of illness, as well as to assess the financial protection afforded by health insurance. This study also found that men’s ability to evaluate potential health risks was correlated to a lesser degree to their old age. We suggest that older age provides male household heads with a better understanding of the potential health, economic and social costs of illness, thereby increasing their understanding of the utility of financial safety nets in adverse health events. In addition to the above characteristics, we found that male household heads place value on a specific return on investment, factoring in trust of a specific health insurance provider in their decision to enrol. This reflects other qualitative and quantitative health financing studies (De Allegri et al. 2006 b; Basaza et al. 2007; Jehu-Appiah et al. 2011, 2012), and highlights the reliance of many voluntary health insurance schemes on the reputation of affiliated entities. When women hold a high level of economic power, on the other hand, we posit that enrolment decisions are likely to prioritize their direct knowledge of potential household health risks. According to our findings, health insurance enrolment was correlated to the presence of children in female-headed households. This implies that women with dependent children are attuned to the healthcare needs of their households, ostensibly due to their traditional role as caregivers. This reflects the findings of various health-seeking behaviour studies, which have found that women’s role in caregiving provides them with awareness of the potential health risks associated with childhood illness (Colvin et al. 2013; Richards et al. 2013). While their willingness to prioritize the health needs of children is notable, poverty ultimately reduces women’s access to decision-making resources. Given that most asset-based wealth in North-West Cameroon has historically been held by males (Goheen 1995), income acts as a limiting factor for women’s participation in health insurance over and above the need to minimize potential household health risks. This aligns to the findings of various social science studies (Kishor and Nietzel 1997; Dixon et al. 2014; Milazzo and Van de Walle 2015), and suggests that access to financial resources serves as an important gateway for facilitating women’s decision-making autonomy. Given the wide variability in household structures between and within regions, care must be taken to avoid blunt generalizations on the nature of households. We concede that these findings are specific to the unique North-West Cameroonian context within which the research was carried out. We also appreciate that even within our designated gender groupings, there will inevitably be heterogeneity in the power relations that are associated with the decision to enrol into voluntary health insurance. That said, to the authors’ knowledge, this is one of the first studies investigating the role of household headship and gender in the context of voluntary health insurance enrolment. This study has several design limitations that may affect the external validity of our findings: the number of BEPHA-insured female-headed households was substantially less than those in male-headed households. However, we contest that this is symptomatic of the low number of female-headed households in the study area in general, as well as the limited uptake of insurance amongst this study population. It would therefore be necessary to carry out further research on a larger number of female-led households to further test our findings. In addition, translation and interpretation of interview questions into Lam Nso’ language may have biased the responses to the research questions. These issues notwithstanding, our findings provide precursory insight into the evolving role of gender in health insurance decision-making. Conclusion This study has embraced the realities of gender-specificity in household roles, and identified clear disparities in the way male- and female-headed households evaluate health risks in the decision to enrol into health insurance. Our findings suggest that voluntary health insurance demand amongst male household heads is associated with their ability to understand household health risks in the absence of a direct role in caregiving. Conversely, health insurance enrolment amongst women is correlated with the need to minimize potential household health risks based on their direct knowledge of household healthcare needs. The impact of education on male health insurance demand underscores the importance of simple and digestible sensitization programmes as a means of facilitating household decision-makers’ understanding of the concept of health insurance. Further, the influence of wealth on health insurance enrolment particularly amongst female-headed households provides important insight from a policy perspective, given the limited availability of subsidies to cover vulnerable populations in many sub-Saharan African health insurance schemes. Our findings highlight the need for partnerships between health insurance schemes and governments in order to develop a financial safety net to limit the impact of illness on the poor. Indeed, the ultimate goal of voluntary health insurance schemes such as BEPHA is to support governments in their quest towards achieving universal health coverage. It is through collaboration and responsiveness to the social, cultural and economic realities of the target population that health insurance coverage will reach sustainable levels. Acknowledgements The authors thank Felicia Ekpenyong Isang and the Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) Kumbo team for their support during the data collection process. The authors also thank Dr. Amanda Ross (Swiss TPH) for statistical support provided. Funding Data collection was financed by the Bamenda Ecclesiastical Provincial Health Assistance (BEPHA) Provincial Office. T.O. is a recipient of the Swiss Government Excellence Scholarship (ESKAS), provided by the Swiss Federal Commission for Scholarships for Foreign Students (FCS). Conflict of interest statement. None declared. References Acharya A, Vellakkal S, Taylor F et al.   2013. The impact of health insurance schemes for the informal sector in low- and middle-income countries: a systematic review. World Bank Research Observer  28: 236– 66. Google Scholar CrossRef Search ADS   Arhin-Tenkorang D. 2001. Health insurance for the informal sector in Africa: design features, risk protection, and resource mobilization, Health, Nutrition, and Population Discussion Paper. The International Bank for Reconstruction and Development. The World Bank, Washington, DC. Basaza R, Criel B, Van der Stuyft P. 2007. Low enrolment in Ugandan Community Health Insurance Schemes: underlying causes and policy implications. BMC Health Services Research  7: 105. Google Scholar CrossRef Search ADS PubMed  Basaza R, Criel B, Van der Stuyft P. 2008. Community health insurance in Uganda: why does enrolment remain low? A view from beneath. Health Policy  87: 172– 84. Google Scholar CrossRef Search ADS PubMed  Beegle K, Christiaensen L, Dabalen A, Gaddis I. 2016. Poverty in a rising Africa. The World Bank. doi: 10.1596/978-1-4648-0723-7 BEPHA. 2016. BEPHA Membership Audit FY 2015–2016. Bamenda Ecclesiastical Provincial Health Assistance, Bamenda. Cameroun Institut National de la Statistique. 2011. Cameroon Wealth Index. Chuma J, Mulupi S, McIntyre D. 2013. Providing financial protection and funding health service benefits for the informal sector: evidence from sub-Saharan Africa, RESYST Working Paper 2. KEMRI, Nairobi. Colvin CJ, Smith HJ, Swartz A et al.   2013. Understanding careseeking for child illness in sub-Saharan Africa: a systematic review and conceptual framework based on qualitative research of household recognition and response to child diarrhoea, pneumonia and malaria. Social Science and Medicine  86: 66– 78. Google Scholar CrossRef Search ADS PubMed  Criel B, Van Dormael M, Lefevre P, Menase U, Van Lerberghe W. 1998. Voluntary health insurance in Bwamanda, Democratic Republic of Congo. An exploration of its meanings to the community. Tropical Medicine & International Health  3: 640– 53. Google Scholar CrossRef Search ADS   De Allegri M, Kouyate B, Becher H et al.   2006a. Understanding enrolment on community health insurance on sub-Saharan Africa: a population-based case-control study on rural Burkina Faso. Bulletin of the World Health Organization  84: 852– 8. Google Scholar CrossRef Search ADS   De Allegri M, Sanon M, Sauerborn R. 2006. “To enrol or not to enrol?”: a qualitative investigation of demand for health insurance in rural West Africa. Social Science and Medicine  62: 1520– 7. Google Scholar CrossRef Search ADS PubMed  Dixon J, Luginaah I, Mkandawire P. 2014. The National Health Insurance Scheme in Ghana‘s Upper West Region: a gendered perspective of insurance acquisition in a resource-poor setting. Social Science and Medicine  122: 103– 12. Google Scholar CrossRef Search ADS PubMed  Giedion A, Alfonso EA, Diaz Y. 2013. The Impact of Universal Coverage Schemes in the Developing World: a review of the existing evidence (No. UNICO Studies Series No. 25). Universal Health Coverage Studies Series (UNICO). The World Bank, Washington, D.C. Goheen M. 1995. Gender and accumulation in Nso’. Paideuma  41: 73– 81. Hayes RJ, Bennett S. 1999. Simple sample size calculation for cluster-randomized trials. International Journal of Epidemiology  28: 319– 26. Google Scholar CrossRef Search ADS PubMed  Institut National de la Statistique, 2012. Enquète Demographique et de Santé et à Indicateurs Multiples. République du Cameroun, Yaoundé. Institut National de la Statistique, 2008. Conditions de Vie des Populations et Profil de Pauvreté au Cameroun en 2007: Rapport Principal de l’ECAM3. République du Cameroun, Yaoundé. Institut National de la Statistique du Cameroun, 2014. La quatrième Enquête Camerounaise Auprès des Ménages (ECAM 4). République du Cameroun, Yaoundé. International Organization for Migration (Ed.), 2013. Migrant well-being and development, World Migration Report. IOM, Geneva. Jehu-Appiah C, Aryeetey G, Agyepong I, Spaan E, Baltussen R. 2012. Household perceptions and their implications for enrolment in the National Health Insurance Scheme in Ghana. Health Policy Planning  27: 222– 33. Google Scholar CrossRef Search ADS PubMed  Jehu-Appiah C, Aryeetey G, Spaan E et al.   2011. Equity aspects of the National Health Insurance Scheme in Ghana: who is enrolling, who is not and why? Social Science & Medicine  72: 157– 65. Google Scholar CrossRef Search ADS   Kimani JK, Ettarh R, Kyobutungi C, Mberu B, Muindi K. 2012. Determinants for participation in a public health insurance program among residents of urban slums in Nairobi, Kenya: results from a cross-sectional survey. BMC Health Services Research  12: 66. Google Scholar CrossRef Search ADS PubMed  Kimani JK, Ettarh R, Warren C, Bellows B. 2014. Determinants of health insurance ownership among women in Kenya: evidence from the 2008–09 Kenya Demographic and Health Survey. International Journal for Equity in Health  13: 27. Google Scholar CrossRef Search ADS PubMed  Kishor S, Nietzel K. 1997. The Status of Women: Indicators for Twenty-Five Countries - Gender and Headship, DHS Comparative Studies. Macro International Inc., Calverton, Maryland. Lagomarsino G, Garabrant A, Adyas A, Muga R, Otoo N. 2012. Moving towards universal health coverage: health insurance reforms in nine developing countries in Africa and Asia. The Lancet  380: 933– 43. Google Scholar CrossRef Search ADS   Leive A, Xu K. 2008. Coping with out-of-pocket health payments: empirical evidence from 15 African countries. Bulletin of the World Health Organization  86: 849– 56. Google Scholar CrossRef Search ADS PubMed  Macha J, Kuwawenaruwa A, Makawia S, Mtei G, Borghi J. 2014. Determinants of community health fund membership in Tanzania: a mixed methods analysis. BMC Health Services Research  14: 538. Google Scholar CrossRef Search ADS PubMed  Mebratie AD, Sparrow R, Yilma Z, Alemu G, Bedi AS. 2015. Enrollment in Ethiopia‘s community-based health insurance scheme. World Development  74: 58– 76. Google Scholar CrossRef Search ADS   Milazzo A, Van de Walle DP. 2015. Women left behind? Poverty and headship in Africa. Poverty Headship Afr. June 23 2015 World Bank Policy Research Working Paper. Mladovsky P, Ndiaye P, Ndiaye A, Criel B. 2015. The impact of stakeholder values and power relations on community-based health insurance coverage: qualitative evidence from three Senegalese case studies. Health Policy Planning  30: 768– 81. Google Scholar CrossRef Search ADS PubMed  Mulupi S, Kirigia D, Chuma J. 2013. Community perceptions of health insurance and their preferred design features: implications for the design of universal health coverage reforms in Kenya. BMC Health Services Research  13: 474. Google Scholar CrossRef Search ADS PubMed  Musango L, Dujardin B, Dramaix M, Criel B. 2004. Profile of members and non members of mutual health insurance system in Rwanda: the case of the health district of Kabutare. Tropical Medicine & International Health TMIH  9: 1222– 7. Google Scholar CrossRef Search ADS   Onwujekwe O, Onoka C, Uguru N et al.   2011. Socio-economic and geographic differences in acceptability of community-based health insurance. Public Health  125: 806– 8. Google Scholar CrossRef Search ADS PubMed  Parmar D, De Allegri M, Savadogo G, Sauerborn R. 2014. Do community-based health insurance schemes fulfil the promise of equity? A study from Burkina Faso. Health Policy Planning  29: 76– 84. Google Scholar CrossRef Search ADS PubMed  Parmar D, Reinhold S, Souares A, Savadogo G, Sauerborn R. 2012. Does community-based health insurance protect household assets? Evidence from rural Africa. Health Services Research  47: 819– 39. Google Scholar CrossRef Search ADS PubMed  République du Cameroun, Agence Régionale du Nord-Ouest, 2015. Annuaire Statistique de la Région du Nord-Ouest 2014. Richards E, Theobald S, George A et al.   2013. Going beyond the surface: gendered intra-household bargaining as a social determinant of child health and nutrition in low and middle income countries. Social Science Medicine  95: 24– 33. Google Scholar CrossRef Search ADS PubMed  Schneider P. 2004. Why should the poor insure? Theories of decision-making in the context of health insurance. Health Policy Planning  19: 349– 55. Google Scholar CrossRef Search ADS PubMed  Soors W, Devadasan N, Durairaj V, Criel B. 2010. Community health insurance and universal coverage: multiple paths, many rivers to cross. World Health Report. United Nations General Assembly. 2012. Global Health and Foreign Policy. Agenda Item 123. Wagstaff A. 2008. Measuring financial protection in health, Policy Research Working Papers. The World Bank. doi:10.1596/1813-9450-4554 World Bank, JICA, The Global Fund, African Development Bank, WHO. 2016. Universal health coverage in Africa: a framework for action, in: Joint 2016 TICAD Conference Working Paper. World Health Assembly. 2005. Resolution WHA58.33: sustainable health financing, universal coverage, and social health insurance. Presented at the Fifty-Seventh World Health Assembly. World Health Organization, 1992. International Statistical Classification of Diseases and Related Health Problems, 10th ed. World Health Organization, Geneva. World Health Organization (Ed.). 2014. The health of the people: what works: the African Regional Health Report 2014. World Health Organization, Regional Office for Africa, Brazzaville, Republic of Congo. World Health Organization. 2010. The world health report: health systems financing: the path to universal coverage. World Health Organization, Geneva. © The Author 2017. Published by Oxford University Press in association with The London School of Hygiene and Tropical Medicine. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Health Policy and PlanningOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off