Women & Health, 55:485–504, 2015 Copyright © Taylor & Francis Group, LLC ISSN: 0363-0242 print/1541-0331 online DOI: 10.1080/03630242.2015.1022815

The Association Between Household Consumer Durable Assets and Maternal Health-Seeking Behavior in Ghana ERIC ANSONG, MPhil, PhD School of Law and Social Sciences, University of East London, London, United Kingdom

This article examined the association between household consumer durable assets and maternal health-seeking behavior. Several studies have suggested a relationship between households’ socioeconomic status (SES) and health outcomes. However, SES is a multidimensional concept that encompasses variables, such as wealth, education, and income. By grouping these variables together as one construct, prior studies have not provided enough insight into possible independent associations with health outcomes. This study used data from the 2008 Ghana Demographic and Health Survey of 2,065 women aged between 15 and 49 years to examine the association between household consumer durables (a component of SES) and maternal health-seeking behavior in Ghana. Results from a set of generalized linear models indicated that household consumer durable assets were positively associated with four measures of maternal health-seeking behaviors, namely, seeking prenatal care from skilled health personnel, delivery by skilled birth attendant, place of delivery, and the number of antenatal visits. Also, households with more assets whose residents lived in urban areas were more likely to use skilled health personnel before and during delivery, and at an approved health facility, compared those who lived in rural areas. Implications for health interventions and policies that focus on the most vulnerable households are discussed. KEYWORDS consumer durable asset, health-seeking behavior, household, maternal health, well-being

Received February 22, 2014; revised June 12, 2014; accepted June 24, 2014. Address correspondence to Eric Ansong, MPhil, PhD, School of Law and Social Sciences, University of East London, Docklands Campus, University Way, London E16 2RD, United Kingdom. E-mail: [email protected] 485

486

E. Ansong

BACKGROUND Human health and wellbeing are important ingredients in the economic and social development of societies and nations. It is argued that the ability to enjoy the highest and attainable standard of health is a fundamental right that every human being should have (Deneulin and Shahani 2009). It is, therefore, not surprising that the core of the eight Millennium Development Goals1 (MDGs) significantly focus on health. However, an estimated 880 women globally die daily from preventable pregnancy and childbirth related episodes (World Health Organization [WHO] 2012). Similarly, a WHO report in 2012 suggested that nearly all (99 percent) of these deaths occur in developing countries, especially in predominately rural and poor communities (WHO 2012). This is primarily due to the lack of access to skilled routine obstetric and emergency care before, during, and after childbirth (Erim, Kolapo, and Resch 2012). Arguably, accessing healthcare before, during, and after birth can positively impact the survival of both mothers and children. It is similarly important to point out that the high concentration of maternal mortality in some areas may be a manifestation of the gross inequalities, especially in access to health services, between the “haves” and “have-nots.” Like other sub-Saharan African countries, Ghana has managed to reduce the maternal mortality rate from 550 per 100,000 live births in 2002 to 350 per 100,000 live births in 2010. This feat has been achieved through the MDG Acceleration Framework (MAF), an action plan developed by Ghana’s Ministry of Health (MoH), the Ghana Health Service (GHS), and other domestic and global development partners. MAF was developed to intensify efforts to overcome bottlenecks in implementing evidenced-based interventions around family planning, skilled delivery, and emergency obstetric and newborn care in Ghana. Despite this achievement, it is feared that Ghana will still not achieve the MDG target (i.e., MDG goal 5A) of reducing by three-quarters the maternal mortality ratio between 1990 and 2015, which stood at 580 per 100,000 live births in 2007 (Andoh 2014). With private expenditure on health (i.e., out-of-pocket expenditure) in 2011 at 63 percent of total health expenditures in Ghana (World Bank 2013), it is feared that poorer households will continue to be burdened by a large share of healthcare cost. This is even more crucial for vulnerable members of households, especially pregnant women from households with fewer resources. For Ghana to achieve MDG target 5A it is vital that the health-seeking behavior of pregnant women are given the needed attention so as to help improve practices that engender healthy antenatal and child delivery. Prior research has provided insight into factors that could engender healthy antenatal and child delivery. Buor and Bream (2004) pointed out that these determinants of maternal health-seeking behavior could be categorized broadly under social, economic, and psychological factors. Much of

Household Assets and Maternal Health-Seeking Behavior

487

what we know about health-seeking behavior has focused on the socioeconomic determinants (Ahmed et al. 2005; Leon-Gonzalez and Tseng 2011). It should be acknowledged that socioeconomic factors are multi-dimensional and can include variables, such as income, education, and wealth. Wealth is often denoted by resource endowments, which is essentially asset ownership. Chowa, Ansong, and Masa (2010) suggested an association between households’ assets ownership and a range of health outcomes, including maternal health. To test Chowa, Ansong, and Masa’s (2010) postulation, the present study examined the association between household consumer durable assets and maternal health-seeking behavior in Ghana.

THEORETICAL AND EMPIRICAL FRAMEWORK Sen’s capability approach (1992) and the asset-effect framework (Lerman and McKerman 2008) support the pre-position that, personal and household resources and endowments are integral to the functioning and attainment of well-being. The capability approach to well-being essentially centers on “a person’s ability to do valuable acts or reach valuable state of being” (Sen 1993, p. 30). Hence, the pursuance of good health through the use of healthcare facilities and systems can be characterized as what people “value and have reason to value” (Sen 1999, p. 18) in the capability approach. For instance, an argument regarding maternal health is that pregnant women should be able to have the freedom otherwise known as real opportunities to pursue appropriate antenatal care, including suitable place of delivery, skilled or professional delivery assistance, and postnatal care. Clearly, the capability framework recognized the instrumental role of asset ownership (Ansong, Chowa, and Grinstein-Weiss 2013), in that it provides opportunities for individuals and households to meet health expenditure. Empirical support exists for the theoretical relationship between household consumer durable assets and maternal health-seeking behavior (MHSB). However, a major research gap in the knowledge base is that much of the evidence comes from developed countries (Bronte-Tinkew and DeJong 2004; Houweling, Kunst, and Mackenbach 2003). In Ghana, studies addressing the factors influencing MHSB are limited. One of the few empirical studies that have examined the connection between assets and health-seeking behavior in the Ghanaian context was by Leive and Ke (2008). Examining how households in fifteen African countries (including Ghana) cope with health expenditure, the study found that poorer households consistently sold household assets to meet health costs. Besides this, there was limited emphasis on MHSB in Leive and Ke’s study. Thus, as to whether the results will hold specifically for MHSB in Ghana is something research has yet to establish. Another gap in the state of knowledge is that many prior studies on the factors associated with MHSB have not clearly separated the components

488

E. Ansong

of socioeconomic status (SES) to establish their independent relations to MHSB. The majority of these studies have actually focused more on the non-economic dimensions of SES, such as education. For instance, the association between mothers’ education and maternal health outcomes is well documented in several studies (McTavish et al. 2010). Although these studies examined the association of SES with health outcomes, emphasis was on education, ignoring how household resources might also be associated with health outcomes. Studies that used SES as proxies for households’ wealth or economic status failed to account for households “long-running” economic status; they did not demonstrate how key indicators of SES, such as household consumer durable asset ownership, alone may be associated with MHSB. Similarly, income—used as the only proxy for the economic dimension of SES—has also been found in several studies to be associated with health outcomes (Leon-Gonzalez and Tseng 2011; Urbanos-Garrido 2012). Evidently, development researchers and economists in the past decade have increasingly began to embrace and advocate the use of assets as a complement to income and consumption-based measures of welfare and wealth in developing countries (Filmer and Pritchett 2001; Lerman and McKerman 2008). Although numerous studies find that health improves with income, some (Meer, Miller, and Rosen 2003) are of the opinion that wealth is a superior measure of an individual’s economic status over income because wealth more accurately reflects long-term economic conditions of individuals (Sherraden and Wallace 1992). Similarly, the lack of consistency, inaccuracies, and the unreliable nature of data on income in Ghana and other developing countries make it imperative to embark on a pragmatic way of measuring standards of living (Montgomery, Gragnolati, Burke, and Paredes 2000). With little consensus in the literature regarding how best to measure standard of living in the absence of income and consumption data, household consumer durables assets are considered a reasonable way of defining living-standards indicators (Montgomery and Hewett 2005). Consequently, some studies in developing countries have used household consumer durable assets as a measure of economic status (Grinstein-Weiss, Curley, and Charles 2007; Sherrraden, Guo, and Zhang 2008). This article focused on household consumer durables and their association with MHSB. With the few studies in Ghana focusing predominantly on the socio-demographic determinants (Van den Boom, Nsowah-Nuamah, and Overbosch 2004; Yakong et al. 2010), more research is needed to offer insights into other economic areas associated with MHSB. MHSB primarily centers on women’s interaction with healthcare systems during pregnancy and delivery. However, not much is known about the different types of behaviors of pregnant women in the Ghanaian context. The magnitude and direction of the relationship between assets and each of these measures of MHSB may be different, but research has yet to clarify the nature of these relationships. In other words, is asset ownership related to which health facility and personnel one goes to when she is pregnant or in labor,

Household Assets and Maternal Health-Seeking Behavior

489

and how frequently does one use services from these facilities and personnel? This study addressed these questions by examining the connections between assets and four MHSBs: type of health facility used for both prenatal care and delivery, type of health personnel used, and frequency of visits to health facility. In doing so, the following four hypotheses are tested: Hypothesis 1: Pregnant women with more household consumer durable assets are more likely to seek prenatal care from skilled health personnel (H1 ). Hypothesis 2: Pregnant women with more household consumer durable assets are more likely to have a skilled birth attendant during delivery of their new babies (H2 ). Hypothesis 3: Pregnant women with more household consumer durable assets are more likely to have a facility-based delivery of their newborn (H3 ). Hypothesis 4: Pregnant women with more household consumer durable assets are more likely to have frequent prenatal visits to a health facility (H4 ).

DATA AND METHOD OF ANALYSIS This study used cross-sectional data from the 2008 Ghana Demographic and Health Survey (GDHS). The GDHS survey collected information from a nationally representative sample of 4,916 women aged 15–49. The GDHS is a nationally-representative household survey that provides data on a wide range of indicators in the areas of population demographics, health, and nutrition for monitoring and evaluation. This study analyzed the responses from 2,065 women aged 15–49 years who had ever had children and data from their most recent pregnancy and birth before the data collection period. Because the study aimed to explore the nature of health-seeking behavior before and during pregnancy, the sampling inclusion criterion was women who have ever had children. The highest percentage of missing data in the variables of interest was 6 percent. Bivariate tests showed that these data were missing completely at random. Therefore, the study used a multiple imputation method to replace the missing data. Approval for the conduct of the GDHS survey was granted by the ethical committee of Ghana’s Ministry of Health and the Ghana Health Service (Accra, Ghana) and the Ethics Committee of the Opinion Research Corporation Macro International Incorporated (ORC Macro Inc., Calverton, Maryland, USA). All information for the survey was collected confidentially and anonymously. Written and signed informed consent was obtained from all participants (MOH, GSS and ICF Macro, 2009). All identifiers and personal information were removed prior to data analysis.

490

E. Ansong

Dependent Variables This study focused on four dependent variables related to MHSB during the most recent pregnancy. All but one of the dependent variables was dichotomized. The first dependent variable was use of skilled prenatal care. This variable assessed the type of health professional the respondent received prenatal care from during pregnancy. The original response options were doctor, nurse/midwife, auxiliary midwife, trained (traditional) birth attendant, and traditional birth attendant. These responses were dichotomized to reflect whether respondent received antenatal care from skilled birth attendant. Respondents are deemed to have received antenatal care from a skilled attendant if respondent received at least one prenatal care from a doctor or nurse/midwife. These categorizations were based on the level of training and know-how of health personnel, and their ability to deal with obstetric complications. The second dependent variable was delivery by skilled birth attendant. This variable assessed whether a respondent’s most recent child was delivered by a skilled birth attendant. The original response scale included doctor, nurse/midwife, auxiliary midwife, trained (traditional) birth attendant, traditional birth attendant, relative, no one, and other. These responses were recoded into dummy responses to indicate whether a respondent used a skilled birth attendant. In both the first and second dependent variables, the study compared skilled to unskilled attendants because the decision to use skilled rather than unskilled personnel may have implications on the quality of care and subsequent outcomes. The expectation was that women who used unskilled attendants might have been at a higher risk of prenatal complications and unsafe delivery as compared to those who used skilled attendants who are generally proficient in the specialized skills needed to manage pregnancy complications (Guliani, Sepehri, and Serieux 2012). The third dependent variable was place of delivery, which indicated whether respondents gave birth to their youngest child in an approved health facility. Because approved health facilities generally have the ability to deal with most pregnancy problems, the expectation was that delivery at such facilities would be comparatively safer (Hounton et al. 2008). The original response scale included home, government hospital, government health center, government health post, public mobile clinic, other public, private hospital/clinic, private mobile clinic, and other country specific responses. These were similarly dichotomized to indicate whether the respondent had a facility-based delivery (FBD) for the most recent delivery. This recoding was based on the health facility’s ability to deal with comprehensive emergency obstetric cases as well as performing fetal extractions, caesarean sections, and blood transfusions (United Nations Population Fund 2012). The fourth and last dependent variable was number of antenatal visits. This was a continuous variable, which measured the number of times

Household Assets and Maternal Health-Seeking Behavior

491

the respondent received antenatal care during her most recent pregnancy. The variable was log-transformed to meet the normality assumption of the statistical test, and results were back-transformed to ensure meaningful interpretation of findings.

Independent Variables The primary independent variable of interest was household consumer durable assets index. This index is a measure of households’ ownership of different types of household consumer durable assets, namely, television sets, refrigerators, radio, mobile phone, fix-telephone, freezer, generator, washing machine, computer, DVD, and video cassette recorder (VCR). The “simple count” approach was used to create a composite asset index to show the number of these types of household durables the household owned (Bollen, Glanville, and Stecklov 2002; Mukherjee 2006; Paxson and Schady 2005). The higher the number of household consumer durable assets, the more durable household assets the individual had.

DATA ANALYSIS Two sets of generalized linear models—general binary logistic (GBLM) and linear models (GLM)—were employed to address the four research hypotheses. The study used generalized linear models because they are able to handle most dependent variables regardless of their shape and distribution. GBLM was used to model the probability that respondents received prenatal care from skilled attendants (H1 ), delivered the baby with assistance of skilled birth attendant (H2 ), and delivered the baby at a facility-based health center (H3 ). Linear regression was used to model the number of antenatal visits during pregnancy (H4 ). The household durable asset index was included in all models because it was the main independent variable of interest. Also, because many factors might potentially affect the hypothesized relationship between household consumer durable assets and the dependent variables, the regression models controlled household and individual level factors, including respondents’ age, gender of household head, level of education, partner’s educational level, occupation, number of children in the household, marital status, rural/urban residence, and whether respondent had a health insurance policy. Each model also includes an interaction term between asset index and place of residence (i.e., urban versus rural). Inclusion of this interaction term is based on the differences in how rural and urban households relate to household assets (Chuma, Gilson, and Molyneux 2007). Akaike’s Information Criterion (AIC) was used to compare the final model with their corresponding null models to assess model fit (Chowa, Ansong, and Despard 2014).

492

E. Ansong

RESULTS Description of the Sample The sample consisted of 2,065 women who have had a child in the past year preceding the study. The average age of respondents was 30 years (SD = 7.29). The youngest respondent was 15 years and the oldest was 49 years. Most respondents (72.7 percent) lived in households headed by males. The sample was predominantly rural with as many as 64.8 percent respondents living in rural communities. On average, nearly six people resided per household (Mean = 5.70, SD = 2.74), ranging from one to twenty-two people. Most respondents (59.9 percent) had little or no formal education. The majority of respondents (58.8 percent) were covered by health insurance. The average number of children ever born to the respondent was 3.49 (SD = 2.26). Nearly two-thirds of respondents have had one child in the last 5 years. The average score on the household consumer durable assets index was 2.20 (SD = 2.04) with a range from 0 to 10 (Table 1). The typical respondents made almost six antenatal visits (M = 5.77, SD = 3.26) during pregnancy. Some never had antenatal visits, and others made as many as thirty-two visits. A little over one-third of respondents (38.7 percent) did not receive antenatal care from a recognized/registered healthcare facility. While pregnant with their last child, the overwhelming majority of respondents (86.2 percent) received prenatal care from skilled attendants, but just a little over half of respondents (57.5 percent) delivered the child with the assistance of a skilled birth attendant.

MULTIVARIATE RESULTS Relationship Between Household Consumer Durable Assets and Prenatal Care From Skilled Attendant Model 1 provides results of the relationship between the level of durable asset ownership and the likelihood that pregnant women sought prenatal care from a skilled birth attendant (Table 2). The level of ownership of household consumer durable assets was statistically significantly associated with the odds of seeking prenatal care from a skilled personnel (odds ratio [OR] = 1.25, 95% Confidence Interval [CI] = 1.15–1.35). Hence, the data supported the first hypothesis (H1 ) that “Pregnant women with more household consumer durable assets were more likely to seek prenatal care from skilled health personnel.” Thus, for every additional unit increase in household durable assets, the odds that a pregnant woman sought prenatal care from a skilled personnel increased by 25 percent, controlling for other variables in the models. Similarly, the results of analyses indicated that respondents living in urban communities and having more household consumer durable assets

493

Household Assets and Maternal Health-Seeking Behavior TABLE 1 Descriptive Characteristics of the Sample Variables Current age (years) of respondent Total children ever born Age (years) of household head Household size Household durables Number of antenatal visits Current marital status Married Not married Educational level Little or no education Some or high school education Births in the last 5 years 1 2 3 4 Place of residence Rural Urban Covered by health insurance No Yes Sex of household head Male Female Delivery assistance by skilled birth attendants No Yes Prenatal care from skilled attendant No Yes Place of delivery Not approved health facility Approved health facility

n (%)

Mean (SD) 30 3.49 40.42 5.70 2.20 5.77

(7.29) (2.26) (13.21) (2.74) (2.04) (3.26)

Min–Max 15–49 1–14 16–88 1–22 0–10 0–32

1, 476 (71.5) 589 (28.5) 1, 237 (59.9) 828 (40.1) 1, 338 647 73 7

(64.8) (31.3) (3.5) (.3)

1, 337 (64.7) 728 (35.3) 1, 215 (58.8) 847 (41.0) 1, 501 (72.7) 564 (27.3) 976 (47.3) 1, 089 (52.7) 286 (13.8) 1, 779 (86.2) 947 (45.9) 1, 118 (54.1)

were more likely to seek prenatal care from skilled personnel than those living in rural areas (OR = 1.04, 95% CI = 0.94–1.14) (Table 2). Thus, for every additional unit increase in household durable assets, the odds that a pregnant woman in an urban community sought prenatal care from skilled personnel increased by 4 percent compared to a pregnant woman in a rural community. Results also show that other control variables were statistically significantly associated with the propensity to seek prenatal care from skilled personnel.

Relationship Between Household Consumer Durable Assets and Delivery Assistance by Skilled Birth Attendants Model 2 (Table 2) provides results of the relationship between the level of durable asset ownership and the likelihood that a pregnant woman sought

494

Note. AIC: Akaike’s Information Criterion. + p < .10, ∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001.

Educational level Urban residency Number of children in the household Partner’s educational level Age per year Male-headed household Has insurance Married Employed Household durable assets Urban∗ durable assets Model fit indices Deviance Likelihood ratio χ 2 AIC

Independent variables

TABLE 2 Results of Multivariate Analyses

(2E-4 )∗∗∗ (2E-4 )∗∗∗ (3.80E-4 )∗∗∗ (1E-4 )∗∗∗ (1.66E-5 )∗∗∗ (2E-4 )∗∗∗ (2E-4 )∗∗∗ (2E-4 )∗∗∗ (2E-4 )∗∗∗ (4.58E-5 )∗∗∗ (6.38E-5 )∗∗∗

1053256670.43 187545754.14 1053257353.68

1.48 1.79 0.87 2.51 1.03 0.73 1.21 1.42 0.90 1.25 1.04

Model 1: Prenatal care from skilled personnel Odds (SE) (1E-4 )∗∗∗ (2E-4 )∗∗∗ (4.16E-5 )∗∗∗ (1E-4 )∗∗∗ (1.26E-5 )∗∗∗ (1E-4 )∗∗∗ (1E-4 )∗∗∗ (1E-4 )∗∗∗ (2E-4 )∗∗∗ (4.72E-5 )∗∗∗ (7.22E-5 )∗∗∗

1784218177.51 640751886.52 1784218752.70

1.87 2.60 0.87 1.96 1.10 0.91 1.90 0.87 0.86 1.35 1.02

Model 2: Delivery by skilled birth attendant Odds (SE)

1778705654.86 647066939.17 1778706141.04

1.78 (1E-4 )∗∗∗ 2.57 (2E-4 )∗∗∗ 0.86 (4.16E-4 )∗∗∗ 1.84(1E-4 )∗∗∗ 1.05 (1.26E-4 )∗∗∗ 0.75 (1E-4 )∗∗∗ 2.13 (1E-4 )∗∗∗ 0.86 (1E-4 )∗∗∗ 0.73 (2E-4 )∗∗∗ 1.40 (4.79E-5 )∗∗∗ 1.03 (7.22E-5 )∗∗∗

Model 3: Delivery at approved health facility Odds (SE)

(1.21E-2 )∗∗ (1.8E-2 )∗∗∗ (3.6E-3 )∗∗∗ (1.22E-2 )∗∗∗ (1.1E-3 )∗∗∗ (1.21E-2 )∗∗ (1E-2 )∗∗ (1.22E-2 )∗∗ (1.56E-2 ) (4.2E-3 )∗∗∗ (5.4E-3 ) 75721907.54 380.47 −290.30

0.03 0.07 −0.02 0.06 0.01 −0.03 0.02 0.03 0.002 0.03 −0.01

Model 4: Number of antenatal visits b (SE)

Household Assets and Maternal Health-Seeking Behavior

495

delivery assistance from a skilled birth attendant. Ownership of household consumer durable assets significantly increased the odds of seeking delivery assistance from a skilled birth attendant (OR = 1.35, 95% CI = 1.25–1.45). Therefore, the data analysis results supported the hypothesis (H2 ) that, “Pregnant women with more household consumer durable assets were more likely to have a skilled birth attendant during delivery of their new babies.” Holding all other variables constant, a pregnant woman was 35 percent more likely to seek assistance from skilled birth attendants when her household consumer durable assets increased by one unit. Similarly, results showed that the number of household consumer durable assets, along with living in an urban community, had a statistically significant association (OR = 1.02, 95% CI = 0.92–1.12) with the likelihood that an expectant woman would seek delivery assistance from a skilled birth attendant. This means that for every additional increase in household consumer durable assets of an expectant woman in an urban community, the likelihood to seek delivery assistance from a skilled birth attendant increased by 2 percent compared to an expectant woman in a rural area. Results also show that respondents’ age, living in an urban community, number of children, educational level of respondents and their partners, and having health insurance were statistically significantly related to the odds of seeking assistance from skilled birth attendants when in labor (Table 2).

Relationship Between Household Consumer Durable Assets and Place of Delivery Model 3 provides results indicating the probability of delivery at an approved health facility (Table 2). Results showed a statistically significant relationship between ownership of household consumer durable assets and delivery at an approved health facility (OR = 1.40, 95% CI = 1.27–1.47). Thus, the data provides evidence to support the hypothesis (H3 ) that, “pregnant women with more household consumer durable assets are more likely to have facility-based delivery of their new born.” Essentially, whenever household consumer durable assets increased by one unit, the expectant woman was 40 percent more likely to deliver at an approved health facility, holding all other variables in the model constant. The GBLM test also showed that age (OR = 1.05, 95% CI = 0.95–1.14), living in urban community (OR = 2.57, 95% CI = 2.47–2.68), higher educational level (OR = 1.78, 95% CI = 1.67–1.88), having health insurance (OR = 2.13, 95% CI = 2.02–2.23), and partner’s greater education (OR = 1.84, 95% CI = 1.74–1.94) were significantly positively associated with the likelihood that, one would deliver at an approved healthcare facility, but living in a male-head household (OR = 0.75, 95% CI = 0.65–0.84) and greater number of children (OR = 0.86, 95% CI = 0.76–0.96) were significantly negatively associated with delivering at an approved healthcare facility.

496

E. Ansong

Relationship Between Household Consumer Durable Assets and Frequent Antenatal Visits We run a GLM to test the hypothesis that “pregnant women with more household consumer durable assets are more likely to have frequent prenatal visits to a health facility.” Results of the test are labeled Model 4. Because the dependent variable was log transformed, results are back-transformed (i.e., b ∗ 100 percent) prior to interpretation. The main variable of interest, household consumer durable assets, was statistically significantly associated with number of antenatal visits during pregnancy (b = 0.03, 95% CI = 0.018–0.034). Hence, the analysis results supported the hypothesis (H4 ) that the more household consumer durable assets a household owned the more pregnant women in that household sought antenatal care. A pregnant woman’s antenatal visits increased by 30 percent (i.e., 0.03 ∗ 100 percent) for every one additional unit increase in durable assets, regardless of other control variables in the model. Results also showed that number of antenatal visits was positively associated with current age of the respondent (b = 0.01, 95% CI = 0.00–0.11), living in an urban community (b = 0.07, 95% CI = 0.03–0.10), partner’s education (b = 0.06, 95% CI = 0.03–0.08), and being married (b = 0.03, 95% CI = 0.00–0.05) but negatively associated with living in a male-headed household (b = −0.03, 95% CI = −0.05−0.01), total children ever born (b = −0.02, 95% CI = −0.24−0.01).

DISCUSSION OF FINDINGS The capability approach and asset effect framework, which posits that resources at the disposal of individuals affect their general well-being, were the basis for the examination of the association between household consumer durable assets and maternal health-seeking behavior in this study (Ransome 2010). Using these theoretical perspectives, this study examined four key elements of maternal health-seeking behavior (MHSB) in Ghana.

Relationship Between Household Consumer Durable Assets and Prenatal Care From Skilled Attendants The results of this study suggested that household assets in the form of consumer durables were statistically significantly associated with pregnant women seeking prenatal care from skilled birth attendants in Ghana. This is consistent with several other studies that point to the association between asset ownership and well-being (Leon-Gonzalez and Tseng 2011; Rathavuth 2007). This means ownerships of assets may facilitate WHO’s recommendation for pregnant women to use basic prenatal obstetric care. Receiving prenatal care from a skilled birth attendant enhances the likelihood that

Household Assets and Maternal Health-Seeking Behavior

497

both fetus and mother receive appropriate medical care, which increases the chance of safe childbirth and improved maternal health as suggested in Goal 5 of the MDGs. It is also important to point out that certain types of assets may have stronger associations with MHSB than others. For instance, having technology-based assets, such as a television set or a radio at home, can have a positive effect on MHSB through access to media broadcasts of maternal health education programs. Equally, having a mobile phone can influence MHSB if it can be used as a medium to contact skilled health personnel for help and advice. mHealth (mobile health technology) is currently on the upsurge in developing countries as information and communications technology (ICT) has become very integral to healthcare delivery (Krüger and Niemi 2012). This study like several studies (Dasgupta 2003) acknowledges the importance of other SES factors in evaluating wellbeing. However, it is worth disentangling the multidimensional nature of the SES concept to ascertain how individual components may be associated with MHSB. The model also indicated the disproportionate use of prenatal care from skilled attendants in urban areas (OR = 1.79, 95% CI = 1.68–1.89) as pregnant women living in urban communities were one-and-a-half times as likely to seek prenatal care from skilled personnel compared to those in rural communities. Although this could be a result of people in the urban centers having a wider range of household consumer durable assets, controlling for living in an urban community and having household assets still indicated that ownership of household durables was crucial to the tendency to seek prenatal care from skilled personnel.

Relationship Between Household Consumer Durable Assets and Delivery Assistance by Skilled Birth Attendants It is vital that all pregnant women receive assistance during delivery by appropriately trained health personnel with adequate equipment to reduce maternal deaths. This will go a long way to help achieve the MDG’s goal of improving maternal health. The results of this study showed that pregnant women with more household consumer durables were 35 percent more likely to receive assistance during delivery from skilled birth attendants. Similarly, women living in urban communities were more likely to have skilled birth assistance. Similarly, the results of analyses indicated that respondents living in urban communities and having more household consumer durable assets were more likely to seek assistance from skilled birth attendants compared to respondents living in rural communities (OR = 1.02, 95% CI = 0.92–1.12). This is consistent with findings by Ochako et al. (2011), which highlighted the widespread inequalities in accessing prenatal care due to the lack of resources to meet health costs.

498

E. Ansong

Relationship Between Household Consumer Durable Assets and Place of Delivery The importance of childbirth, especially place of delivery, cannot be overemphasized particularly in developing countries like Ghana. Delivering newborns in an approved health facility is a crucial aspect of ensuring safe delivery since complications that can arise can be dealt with when one opts for a facility-based delivery. Findings from this study suggest that household consumer durables are highly significant when it comes to choosing to deliver babies in an approved health facility. In other words, household wealth in the form of household consumer durable assets was highly associated with facility-based delivery. This study is consistent with findings by Khun and Manderson (2008) who argue that the lack of resources and endowments are crucial barriers for poorer members of society who have to battle with catastrophic health costs at the point of delivery. Although women with health insurance coverage were twice as likely to have facility-based delivery, 41 percent of women in the study did not have health insurance coverage. Studies (Marriott and Apoya 2011) have demonstrated that asset-poor households are less likely to have health insurance coverage. This study provides evidence on how household consumer durable assets may be associated with facility-based delivery. This is consistent with a study by Hounton et al. (2008) in Burkina Faso where household assets were found to be related to whether an expectant mother would have a facility-based delivery. Similarly household consumer durable assets have been shown by Van Damme et al. (2004) to be instrumental in meeting out of pocket health expenditures.

Relationship Between Household Consumer Durable Assets and Frequent Antenatal Visits One critical way of ensuring, maintaining, and enjoying a healthy pregnancy is through regular visits to health facilities to seek obstetric care (Pereira et al. 2007). Pregnancies that receive regular care and attention from skilled professionals are more likely to result in safe deliveries and general well-being of mothers and newborns (Carroli, Rooney, and Villar 2001). Ample literature points to the relationship between household wealth, mostly depicted by ones’ SES, and maternal health includes the propensity to seek regular antenatal care. This study similarly demonstrated that household consumer durables were statistically significantly related to the likelihood that a pregnant mother would seek frequent prenatal care. Interestingly, pregnant women from male-headed households were less likely to have frequent antenatal visits. The less likelihood for expectant mothers in maleheaded households to seek frequent antenatal visits could be a contextual issue. Although household consumer durable assets are generally seen as

Household Assets and Maternal Health-Seeking Behavior

499

possessions of the general household, it could well be that considering 72.7 percent of households were headed by males, female members of the house had limited control over household possessions. This could be due to the presence of domestic power imbalances and intra-household inequalities in terms of the bargaining power endemic in many developing countries (Robeyns 2003). Again, the GoG and stakeholder would have to do more by way of user exemptions for the asset-poor and effective educational campaign to inform expectant mothers on the needs for regular antenatal visits to ensure safe delivery of babies and the general health of mothers.

LIMITATIONS This study had several limitations that are worth pointing out. First, the study used a cross-sectional design; hence, the temporal sequence of variables could not be discerned, and thus claims of causality cannot be made. Second, the time lapse between when data for this study were collected and when the study was initiated requires that follow-up studies be made to confirm the current relationships. The data for this study were collected in 2008; thus, relationships might have evolved. Similarly, the nuance in measurement of assets may be a concern for reliability of results and other important control variables may not have been considered in the modeling. Also, bias due to omitted variables was possible because the study did not examine all possible characteristics that might affect the relationships tested. The exclusion of such variables was solely based on lack of information in the dataset. Notwithstanding these limitations, this study is an important step in building knowledge on how assets are associated with MHSB, especially in the Ghanaian context.

IMPLICATIONS OF FINDINGS This study highlights the association between household consumer durable assets and four elements of maternal health-seeking behavior. These results have research implications as they pave the way for further research on how household asset ownership can influence asset-based welfare policies. Given that this study’s results point to a close association between asset ownership and improved maternal health (MDG5), policy initiatives to improve maternal health outcomes may consider asset development opportunities to aid families’ ability to accumulate assets that supports families’ MHSB. Results from this study also have implications for practitioners. Results indicated that an increased number of children in households reduced the likelihood of seeking prenatal care from skilled personnel. This means that health education programs ought to incorporate more information about the benefits of

500

E. Ansong

smaller household sizes in making it easier for families to afford skilled care. Targeted programs for large families are also needed as they have a higher tendency not to use skilled care during pregnancy.

CONCLUSION This study found that ownership of household consumer durables was significantly associated with preventive and precautionary practice by expectant mothers in Ghana. Consequently, improving maternal health (MDG 5) in Ghana needs policy frameworks and initiatives that incorporate ways of mitigating the challenges posed by asset poverty and the lack of resources to meet health expenditure. It must be conceded that the widespread health inequalities precipitated by poverty and poor socioeconomic status (Chandola and Marmot 2007; Fotso 2007) are worse for individuals in asset-poor households. Findings of this study are consistent with other studies that point to the positive association between holding assets at the household level and health outcomes. This study does not seek to disregard the effects of other “traditional” determinants of MHSB but rather shed light on how household consumer durables can independently influence MHSB. Invariably, this study suggests that if people have the needed resources to meet the pregnancy-related healthcare costs, then they may be ultimately healthy enough to effectively participate in socio-economic development. It is therefore vital that a study of this nature be conducted to ascertain the determinants of MHSB in Ghanaian context. The study focuses on maternal and health-seeking behavior, which is a vital measure toward being healthy in order to be able to fully participate in sustainable development. With results in this study pointing to a positive association between household consumer durable assets and MHSB, government and stakeholders may have to consider innovative interventions the provide support for asset poor households.

ACKNOWLEDGMENTS The author thanks the MEASURE DHS for releasing the data for this study. The author also takes the opportunity to thank the Ghana Statistical Service (GSS) and Ghana Health Service (GHS) for their meticulous collection of data. Special thanks go to Seyoum Hameso of School of Law and Social Sciences, University of East London, and Clifford Amoako of School of Geography and Environmental Science, Monash University, for their insightful feedback on this study.

Household Assets and Maternal Health-Seeking Behavior

501

ORCID Eric Ansong

http://orcid.org/0000-0001-8134-4507

NOTE 1. Millennium development goals (MDGs) are eight development goals established by all 139 United Nations member countries during the Millennium Summit in 2000.

REFERENCES Ahmed, S. M., G. Tomson, M. Petzold, and Z. N. Kabir. 2005. Socioeconomic status overrides age and gender in determining health-seeking behaviour in rural Bangladesh. Bulletin of the World Health Organization 83(2): 109–17. Andoh, D. 2014. Ghana struggles to achieve MDGs 4&5. Daily Graphic. Accra, Ghana. http://graphic.com.gh/news/politics/16149-ghana-struggles-to-achievemdgs-4-5.html (accessed June 10, 2014). Ansong, D., G. A. N. Chowa, and M. Grinstein-Weiss. 2013. Future orientation as a mediator between assets and perceived household economic stability: A structural equation modeling approach. Social Work Research 37(2): 147–58. doi:10.1093/swr/svt012 Bollen, K. A., J. L. Glanville, and G. Stecklov. 2002. Economic status proxies in studies of fertility in developing countries: Does the measure matter? Population Studies 56(1): 81–96. doi:10.1080/00324720213796 Bronte-Tinkew, J., and G. F. DeJong. 2004. Children’s nutrition in Jamaica: Do household structure and household economic resources matter? Social Science & Medicine 58(3): 499–514. Buor, D., and K. Bream. 2004. An analysis of the determinants of maternal mortality in Sub-Saharan Africa. Journal of Women’s Health (15409996) 13(8): 926–38. doi:10.1089/jwh.2004.13.926 Carroli, G., C. Rooney, and J. Villar. 2001. How effective is antenatal care in preventing maternal mortality and serious morbidity? An overview of the evidence. Paediatric & Perinatal Epidemiology 15(s1): 1–42. doi:10.1046/j.1365-3016.2001.0150s1001.x Chandola, T., and M. Marmot. 2007. Health and socioeconomic status. In Encyclopedia of stress, George Fink Associate Editors: ed. M. Bruce, E. R. D. Kloet, R. Robert, C. George, S. Andrew, R. Noel, C. Ian, and F. Giora, 255–62. New York, NY: Academic Press. Chowa, G., D. Ansong, and M. Despard. 2014. Financial capabilities: Multilevel modeling of the impact of internal and external capabilities of rural households. Social Work Research 38(1): 19–35. doi:10.1093/swr/svu002 Chowa, G., D. Ansong, and R. Masa. 2010. Assets and child well-being in developing countries: A research review. Children and Youth Services Review 32(11): 1508–19. doi:10.1016/j.childyouth.2010.03.015

502

E. Ansong

Chuma, J., L. Gilson, and C. Molyneux. 2007. Treatment-seeking behaviour, cost burdens and coping strategies among rural and urban households in Coastal Kenya: An equity analysis. Tropical Medicine & International Health 12(5): 673–86. doi:10.1111/tmi.2007.12.issue-5 Dasgupta, P. 2003. Chapter 5: Population, poverty, and the natural environment. In Handbook of environmental economics, ed. K.-G. Mäler and R. V. Jeffrey, 191–247. Amsterdam: Elsevier. Deneulin, S., and L. Shahani. 2009. An introduction to the human development and capability approach: Freedom and agency. London, UK: Earthscan. Erim, D. O., U. M. Kolapo, and S. C. Resch. 2012. A rapid assessment of the availability and use of obstetric care in Nigerian healthcare facilities. PLoS One 7(6): e39555. doi:10.1371/journal.pone.0039555 Filmer, D., and L. Pritchett. 2001. Estimating wealth effects without expenditure data or tears: An application of educational enrolment in states of India. Demography 38(1): 115–32. Fotso, J.-C. 2007. Urban-rural differentials in child malnutrition: Trends and socioeconomic correlates in Sub-Saharan Africa. Health & Place 13(1): 205–23. doi:10.1016/j.healthplace.2006.01.004 Ghana Statistical Service, Ghana Health Services, and ICF Macro. 2009. Ghana demographic and health survey 2008: Key findings. Calverton, MD: ICF Macro. Grinstein-Weiss, M., J. Curley, and P. Charles. 2007. Asset building in rural communities: The experience of individual development accounts. Rural Sociology 72(1): 25–46. doi:10.1526/003601107781147383 Guliani, H., A. Sepehri, and J. Serieux. 2012. What impact does contact with the prenatal care system have on women’s use of facility delivery? Evidence from low-income countries. Social Science & Medicine 74(12): 1882–90. doi:10.1016/j.socscimed.2012.02.008 Hounton, S., G. Chapman, J. Menten, V. De Brouwere, T. Ensor, I. Sombié, N. Meda, and C. Ronsmans. 2008. Accessibility and utilisation of delivery care within a skilled care initiative in rural Burkina Faso. Tropical Medicine & International Health 13: 44–52. doi:10.1111/j.1365-3156.2008.02086.x Houweling, T. A. J., A. E. Kunst, and J. P. Mackenbach. 2003. Measuring health inequality among children in developing countries: Does the choice of the indicator of economic status matter? International Journal for Equity in Health 2: 1–12. Khun, S., and L. Manderson. 2008. Poverty, user fees and ability to pay for health care for children with suspected dengue in rural Cambodia. International Journal for Equity in Health 7: 1–8. doi:10.1186/1475-9276-7-10 Krüger, C., and M. Niemi. 2012. A telemedicine network to support paediatric care in small hospitals in rural Tanzania. Journal of Telemedicine and Telecare 18: 59–62. doi:10.1258/jtt.2011.110312 Leive, A., and X. Ke. 2008. Coping with out-of-pocket health payments: Empirical evidence from 15 African countries. Bulletin of the World Health Organization 86(11): 849–56C. doi:10.2471/blt.07.049403 Leon-Gonzalez, R., and F. M. Tseng. 2011. Socio-economic determinants of mortality in Taiwan: Combining individual and aggregate data. Health Policy 99(1): 23–36. doi:10.1016/j.healthpol.2010.07.005

Household Assets and Maternal Health-Seeking Behavior

503

Lerman, R. I., and S. McKerman. 2008. Benefits and consequences of holding assets. In Asset building and low-income families, ed. S. McKernan and M. Sherraden, 89–152. Washington, DC: The Urban Institute Press. Marriott, A., and P. Apoya. 2011. Achieving a shared goal: Free universal health care in Ghana, 68. Oxford: Oxfam International. McTavish, S., S. Moore, S. Harper, and J. Lynch. 2010. National female literacy, individual socio-economic status, and maternal health care use in Sub-Saharan Africa. Social Science & Medicine 71(11): 1958–63. doi:10.1016/j.socscimed.2010.09.007 Meer, J., D. L. Miller, and H. S. Rosen. 2003. Exploring the health–wealth nexus. Journal of Health Economics 22(5): 713–30. doi:10.1016/S0167-6296(03)00059-6 Montgomery, M., M. Gragnolati, K. Burke, and E. Paredes. 2000. Measuring living standards with proxy variables. Demography 37(2): 155–74. Montgomery, M. R., and P. C. Hewett. 2005. Urban poverty and health in developing countries: Household and neighborhood effects. Demography 42(3): 397–425. doi:10.1353/dem.2005.0020 Mukherjee, S. (2006). On asset indices. Unpublished manuscript, Department of Economics, Princeton University, Princeton, NJ. Ochako, R., J.-C. Fotso, L. Ikamari, and A. Khasakhala. 2011. Utilization of maternal health services among young women in Kenya: Insights from the Kenya demographic and health survey, 2003. BMC Pregnancy & Childbirth 11(1): 1–9. doi:10.1186/1471-2393-11-1 Paxson, C., and N. Schady. 2005. Cognitive development among young children in Ecuador: The roles of wealth, health, and parenting. World Bank Policy Research Working Paper No. 3605. Washington, DC: The World Bank. Pereira, C., A. Cumbi, R. Malalane, F. Vaz, C. McCord, A. Bacci, and S. Bergström. 2007. Meeting the need for emergency obstetric care in Mozambique: Work performance and histories of medical doctors and assistant medical officers trained for surgery. BJOG: An International Journal of Obstetrics & Gynaecology 114: 1530–1533. doi:10.1111/j.1471-0528.2007.01489.x Ransome, B. 2010. Sen and Aristotle on Wellbeing. Australian Journal of Social Issues 45(1): 41–52. Rathavuth, H. 2007. Effect of economic inequality on chronic childhood undernutrition in Ghana. Public Health Nutrition 10(4): 371–78. Robeyns, I. 2003. Sen’s capability approach and gender inequality: Selecting relevant capabilities. Feminist Economics 9(2–3): 61–92. doi:10.1080/135457002 2000078024 Sen, A. 1992. Inequality re-examined. Oxford, UK: Clarendon Press. Sen, A. 1993. Capability and well-being. In The quality of life, ed. N. Martha and S. Amartya, 30–53. New York, NY: Oxford University Press. Sen, A. 1999. Development as freedom, 1st ed. New York, NY: Knopf. Sherraden, M., B. Guo, and L. Zhang. 2008. Household assets and health in China: Evidence and policy implications. CSD Working Paper No. 08-28. St. Louis, MO: Washington University, Center for Social Development. Sherraden, M. S., and S. P. Wallace. 1992. Innovation in primary care: Community health services in Mexico and the United States. Social Science & Medicine 35(12): 1433–43. doi:10.1016/0277-9536(92)90046-s

504

E. Ansong

United Nations Population Fund. 2012. Providing emergency obstetric and newborn care. http://www.unfpa.org/resources/providing-emergency-obstetricand-newborn-care (accessed June 10, 2014). Urbanos-Garrido, R. M. 2012. Social inequalities in health: Measuring the contribution of housing deprivation and social interactions for Spain. International Journal for Equity in Health 11(1): 77–90. doi:10.1186/1475-9276-11-77 Van Damme, W., L. Van Leemput, I. Por, W. Hardeman, and B. Meessen. 2004. Out-of-pocket health expenditure and debt in poor households: Evidence from Cambodia. Tropical Medicine and International Health 9(2): 273–80. doi:10.1046/j.1365-3156.2003.01194.x Van den Boom, G. J. M., N. N. N. Nsowah-Nuamah, and G. B. Overbosch. 2004. Healthcare provision and self-medication in Ghana. http://www.saga.cornell. edu/images/vandenboom.pdf (accessed June 10, 2014). World Bank. 2013. Out-of-pocket health expenditure. http://data.worldbank.org/indi cator/SH.XPD.OOPC.ZS/countries/GH?display=default (accessed September 18, 2013). World Health Organization (WHO). 2012. Fact Sheet No. 348. http://www.who.int/ mediacentre/factsheets/fs348/en/ (accessed June 17, 2013). Yakong, V. N., K. L. Rush, J. Bassett-Smith, J. L. Bottorff, and C. Robinson. 2010. Women’s experiences of seeking reproductive health care in rural Ghana: Challenges for maternal health service utilization. Journal of Advanced Nursing 66(11): 2431–41. doi:10.1111/j.1365-2648.2010.05404.x

The Association Between Household Consumer Durable Assets and Maternal Health-Seeking Behavior in Ghana.

This article examined the association between household consumer durable assets and maternal health-seeking behavior. Several studies have suggested a...
242KB Sizes 0 Downloads 9 Views