Menu
Menu
Menu
Menu
Introduction: Inequities in accessibility to, and utilisation of maternal healthcare services impede progress towards attainment of the maternal health-related Millennium Development Goals. The objective of this study is to examine the extent to which maternal health services are utilised in Ghana, and whether inequities in accessibility to and utilization of services have been eliminated following the implementation of a user-fee exemption policy, that aims to reduce financial barriers to access, reduce inequities in access, and improve access to and use of birthing services. Methods: We analyzed data from the 2007 Ghana Maternal Health Survey for inequities in access to and utilization of maternal health services. In measuring the inequities, frequency tables and cross-tabulations were used to compare rates of service utilization by region, residence and selected socio-demographic variables. Results: Findings show marginal increases in accessibility to and utilisation of skilled antenatal, delivery and postnatal care services following the policy implementation (2003-2007). However, large gradients of inequities exist between geographic regions, urban and rural areas, and different socio-demographic, religious and ethnic groupings. More urban women (40%) than rural, 53% more women in the highest wealth quintile than women in the lowest, 38% more women in the best performing region (Central Region) than the worst (Upper East Region), and 48% more women with at least secondary education than those with no formal education, accessed and used all components of skilled maternal health services in the five years preceding the survey. Our findings raise questions about the potential equity and distributional benefits of Ghana’s user-fee exemption policy, and the role of non-financial barriers or considerations. Conclusion: Exempting user-fees for maternal health services is a promising policy option for improving access to maternal health care, but might be insufficient on its own to secure equitable access to maternal health services in Ghana. Ensuring equity in access will require moving beyond user-fee exemption to addressing wider issues of supply and demand factors and the social determinants of health, including redistributing healthcare resources and services, and redressing the positional vulnerability of women in their communities.
The study reported in this paper forms part of a larger, original study that the authors conducted to examine the effects of Ghana’s user-fee exemption policy on women’s maternity care seeking experience, equity of access, and barriers to accessibility and utilization of maternal and newborn healthcare services. The design of this larger study followed a mixed methods approach; involving analysis of a nationally representative retrospective household survey data in combination with qualitative exploration using data generated from anthropological research techniques of focus group discussions, in-depth interviews and structured field observations. In this paper, we focus on and report findings from the quantitative component of the study, which assessed inequities in accessibility to, and utilisation of maternal health services in Ghana. Ghana is a lower middle-income West African country, with an estimated total population of 24,658, 823 [30]. Average life expectancy at birth is 60 (59 for male and 60.7 for females). Adult literacy – defined as the proportion of population aged 15 years or above who can read and write in English and a Ghanaian language – is 57.9%. Ghana has a human development index (HDI) of 0.526 and a multi-dimensional poverty index of 0.14. In 2005, about 30% of Ghana’s population was estimated to live on less than US$1 per day. Like many lower-income countries, communicable diseases account for about two-thirds of out patient department visits in Ghana, with malaria being the main cause of outpatient morbidity [31]. In addition to the fact that maternal health outcomes continue to be poor in Ghana, we chose Ghana for this research because it is one of only a handful of countries in Africa to have actively started implementing both universal maternity care and health insurance policies at the national level. Because of this, Ghana is often seen as ‘an example of global good practice’ [32]. Despite this, maternal, neonatal and infant mortality ratios have remained persistently high in Ghana. The data for this study were extracted from the Ghana Maternal Health Survey 2007. The GMHS is the first nationally representative, high-quality population-based survey to collect information specifically on maternal health services accessibility and utilization since the implementation of the fee-free maternal health policy. The survey is a retrospective five-year (2003–2007) nationally representative survey of 10,858 households and 10,370 individual women aged 15–49 years. The survey was carried out to collect data to assess the level of maternal mortality in Ghana; identify specific causes of maternal and non-maternal deaths; and measure indicators of access to and utilization of maternal health services in Ghana. The survey was conducted in two phases. In phase I, a short nationally representative household survey questionnaire was administered to 240,000 households from 1,600 clusters or primary sampling units within the 10 administrative regions of Ghana. The 1,600 clusters were selected from a pre-existing list created for Ghana’s 2000 Population and Housing Census. Out of the 240,000 households sampled in phase I, 226,209 households completed the questionnaire, with a 94.3% response rate. The purpose of the Phase I survey was to identify deaths to women aged 12–49 years in the 5 years preceding the survey. In Phase II, a verbal autopsy survey was conducted with households that reported one or more deaths of women aged 12–49 years. Apart from the verbal autopsy survey, Phase II also involved interviews with individual women aged 15–49 years from a total of 11,579 randomly selected households (independent of the households identified in Phase I as having experienced a female death). Of the 11,579 households, 10,994 were occupied at the time of the survey. However, 10,858 households were successfully interviewed, giving a response rate of 99%. From the 10,858 interviewed households, a total of 10,627 women were identified as eligible for individual interview (i.e. women aged 15–49 years). Interviews were however completed for 10,370 women – 98% response rate – using a questionnaire for individual women. The purpose of this Phase II survey was to collect information on key demographic and maternal and neonatal health indicators such as access and use of antenatal and emergency obstetric care in the event of a birth, abortion, or miscarriage. For the purposes of this paper, we used data from the interviews with individual women (i.e. data relating to access to and use of antenatal, maternity, and emergency obstetric care) generated in the second part of Phase II of the survey with the 10,370 individual women. Our analysis involved a total of 5,077 births – 4996 live births and 81 stillbirths – that were recorded in the five years preceding the survey. According to the International Society for Equity in Health, equity is the absence of potentially remediable, systematic differences in access and use of one or more aspects of maternal health services across socially, economically, demographically, or geographically defined population groups or subgroups [33]. This definition is useful for the discussion in this paper because it suggests that non-medical features of individuals or groups (such as their geographic location or ability to pay) should not determine their access to skilled maternity care services. It also implies a situation in which individuals or groups face equal or equivalent access and costs of utilization for equal or equivalent need [34]. In attempting to assess inequities in accessibility to, and utilisation of maternal health services, we used a three-step process outlined by Zere and colleagues [3,5,31]. These steps are: (i) identification of the care intervention whose distribution is to be measured; (ii) classification/grouping of the population into different strata by a selected equity stratifier; and (iii) measuring the degree of inequality. The first step in assessing inequities involved definition of the interventions whose distributions are to be measured. These interventions included antenatal check-ups, tetanus toxoid immunization, delivery at a health facility, skilled attendance at birth, and caesarean sections (CS) during delivery, and postnatal check-ups. Access and equity of access to antenatal care was assessed by the timing, number of visits and type of care provider, and measured by the percentage of women from different socio-demographic backgrounds receiving these types of services. We assessed inequities in protection against tetanus by comparing the percentage of women from different socio-demographic backgrounds receiving the WHO recommended doses of at least two tetanus toxoid injection during their last live or still birth in the five years preceding the survey. Inequities in access to, and use of delivery care was measured by skilled attendant at delivery (i.e. percentage of births delivered by skilled providers including doctor, nurse, midwife, auxiliary midwife and community health officer), delivery in a health facility (i.e. percentage of births delivered in a public or private sector health facilities), delivery at public facility (i.e. percentage of births delivered in public sector health facilities), and home delivery (i.e. percentage of births delivered at home). We assessed inequities in access to and use of CS by measuring the percentage of live births in the five years preceding the survey delivered by CS according to our variable stratifiers. Within the literature, there is debate about the acceptable level at which a given population should be receiving CS [35-38]. Recently however, it has been argued that the proportion of deliveries by CS in a geographical area is a measure of access to, and use of, obstetric emergency care for averting maternal and newborn mortality, and that a population-based rate of 5–15% is considered as the acceptable level of CS to ensure the best outcomes for mothers and newborns [3]. Finally, we assessed inequities in postnatal care access by comparing whether a woman sought care after delivery and from whom across our variable stratifiers. In the second step, we classified women by variable stratifiers against which accessibility to and utilisation of antenatal, delivery and postnatal care services was then assessed. These variables were mother’s education, mother’s age at birth, birth order, place of residence (urban/rural), geographical region of residence, wealth quintiles, religion and ethnicity. The survey data we used do not contain data on household income or consumption income. Therefore wealth index is used as a proxy. This wealth index is constructed from household ownership of assets and consumer goods (radio, television, telephone and refrigerator), dwelling characteristics, type of drinking water source, toilet facilities, electricity, wall and floor materials of house, cooking fuel, and means of transport. Each asset was assigned a weight (factor score) generated using the methods of principal component analysis [5], and the resulting asset scores standardized in relation to a normal distribution with a mean of zero and standard deviation of one (see [31]. From here, each household was given a score for each asset and these asset scores were then summed up for each household. Finally, individual women were ranked according to the total score of the household they came from; the sample was then divided into quintiles from lowest (one) to highest (five). Following this, a single asset index was developed for the whole sample, with no separate indices prepared for different regional or urban and rural populations. We acknowledge that gauging the wealth status of households based on assets may be flawed because ownership of consumer goods is partly a function of taste and choice, and may therefore be independent of wealth [39]. Research has however shown that household assets often approximate the long-run economic status of households [40]. In the third and final step, we assessed access patterns, and equity in utilisation of the interventions we defined in the first step by analysing and comparing accessibility and utilization rates across the variable stratifiers using descriptive statistical tools. Within the healthcare literature, there is still considerable debate regarding the development of appropriate methods for assessing inequities in health and differentials in access among social groups [41]. However, Gulliford’s recent work summarizes the different debates to suggest three main approaches, namely those depending on simple comparison of rates of access for different groups; those depending on the use of regression methods; and those that rely on the development of Gini-like coefficients [41]. Given that our study is mainly descriptive, we chose the first approach. Rates of access were compared for different population groups using both absolute measures (the difference in rates between the selected group and the reference group), and relative measures (the ratio of rates between selected and reference groups). We analysed all the data using the IBM SPSS Statistics data analysis software package (version 20), and MS Excel.
N/A
