Keywords
Women, health insurance, developing country
This article is included in the Sociology of Health gateway.
Women, health insurance, developing country
The need for Universal Health Coverage has been continually increasing across the world (Odokonyero, Mwesigye, Adong, & Mbowa, 2017). This has been driven by the third Sustainable Development Goal that advocates for healthy life as well as promoting a decent life for all individuals regardless of their age (Fleming, Wise, Hansen, & Sams, 2017). It is on the basis of this goal that countries are aiming to reduce death of mothers during childbirth, death of children (both infants and adolescents) and ending the spread of diseases like HIV/AIDS, tuberculosis, and malaria. But not all of these health objectives can be achieved by countries because some individuals don’t have access to basic health care (Odokonyero et al., 2017).
Absence of Medicaid for individuals and households globally is the major cause of lack of access to quality health care because individuals or household do not have financial protection during periods of illness (Aizawa, 2019; Ssempala, 2018). This is not so different for Uganda, according to Zikusooka et al. (2009) the insurance market in Uganda is very small with a few companies providing insurance services. The private sector contributes more to health expenditure in Uganda (82.6%) as compared to the government which contributes only 17.4%. Out of the private expenditure, 65.4 % comes from out-of-pocket payments making it the single largest component of total healthcare expenditure (Sempala, 2018). The absence of national health insurance has left the health insurance sector in Uganda to be dominated by private sector where the services of the private health insurance companies are costly explaining the low demand for different types of Medicaid.
The reliance on direct payments for health care financing amongst households increases catastrophic health expenditures and is often cited by Mulenga et al. (2016) as barrier to the achievement of universal health care. The problem is more severe amongst women in countries that are still developing. In underdeveloped countries many women are poor and don’t have access to affordable health care, yet they are faced with higher rates of morbidity and hospitalization (RamPrakash & Lingam 2021; Christiani et al., 2017). Despite the fact that countries are encouraged to provide universal health care for their citizens with more attention given to women’s specific needs because of gender inequalities, many women in underdeveloped countries do not have any form of Medicaid (RamPrakash & Lingam 2021; Christiani et al., 2017). Data from the 2016 Uganda Demographic Health survey by Uganda Bureau of Statistics & Inner City Fund, (2018) indicates that only one quarter (24%) of Ugandan women between 15-49 years had health insurance.
Several studies (Ssempala, 2018; Mpuuga et al., 2020; Basaza et al., 2007; Nshakira Rukundo et al., 2019) have looked at factors affecting the demand for health insurance, but no study has assessed the effect of demographic and media factors on women’s demand for different types of health insurance among women, with a focus on a developing country (in this case, specifically Uganda). This study aims to address this gap by focusing on the following objectives.
i) To ascertain the impact of demographic factors on women’s demand for different types of health insurance
ii) To investigate the impact of media factors on women’s demand for different types of health insurance
These two objectives were answered using the women recode data from the 2016 Uganda Demographic Health Survey (UBOS & ICF, 2018). Results of the study indicated that only 0.44% of women had health insurance provided through community groupings, 0.64% had health insurance provided by their prospective employers while 0.18% privately purchased their own health insurance. The women’s chance for demanding community health insurance significantly increased among women whose demographic factors included lack of education, those who attained only primary level of education and those who were either married or divorced. However, coming from the poorest income households reduced the likelihood demand for community health insurance. Similarly, demographic factors like having attained only primary level of education increased the women’s chances of acquiring their own private health insurance. For employer provided health insurance, demographic factors like being between 35 to 44 years old increased the likelihood demand for employer provided health insurance though coming from a poorer, middle-income household, having attained only primary or secondary level of education decreased the likelihood demand for employer provided health insurance. Media factors like not listening to radio reduced the likelihood demand for community health insurance among women while the likelihood demand for employer provided health insurance decreased with media factors like not reading newspapers at all and reading newspapers not more than one time a week though it increased with the once a week frequency of listening to radio.
The importance of this paper is treble; it will guide government on how to increase the uptake of health insurance and also be pivotal in the development of a country wide health insurance scheme. Academically, it adds on the existing but scarce literature about the demand for different types of health insurance in Uganda. Future researchers may benefit from such important literature. Practically, the study provides evidence on what can be based on to choose any type of health insurance in Uganda.
The other parts of this paper are outlined as follows, following the introduction we had the literature review, the methodology, results, discussion and conclusion.
According to Mulenga et al. (2016), The demand for health insurance is a based on the derived demand for health which is explained by the fact that individuals randomly fall ill. Different individuals attach different weights on the costs and effectiveness of several types of health insurance though the motivation for their restricted choice is possibly ideological (Bourne & Kerr-Campbell, 2010). Following (RamPrakash & Lingam 2021; Bourne & Kerr-Campbell, 2010; Hopkins & Kidd, 1996) the demand for different types of health insurance is attributed to several demographic and media factors that vary between men and women.
According to Boateng and Awunyor-Vitor (2013), the individual choice for any type of health insurance is influenced by two sets of determinants which are similar in nature though analytically separable. Boateng and Awunyor-Vitor (2013), further state that the nature of the health insurance plan and the personal demographic characteristics like age, marital status, education level, wealth and health status of individuals are important determinants when one is choosing a particular type of health insurance. Kimani et al. (2014) also points out the increasing role of demographic characteristics in the demand for health insurance among women of demographic characteristics like age, marital status, education level, wealth and health status are important when one is choosing a particular type of health insurance. Kimani et al. (2014) also argues that demographic factors like income, education, household wealth status, marital status, age, place of residence are important predictors of health insurance amongst women. Foubister et al. (2006) also concurs that demographic factors like age, gender, income, education and area of residence affect individual subscription to private health insurance. Foubister et al. (2006) also concur that demographic factors like age, gender, income, education and area of residence affect individual subscription to private health insurance.
Mulenga et al. (2016); Owusu-Sekyere and Chiaraah, A. (2014); Kirigia et al. (2005) further argue that being married, belonging to higher age category, higher level of education, health status had a positive influence on health insurance coverage. Shaikh, Woodward, Norton, and Jha (2018) note that, illiteracy and age greatly affected individual enrollment in community health insurance among nations with high gender inequality levels. This was supported by RamPrakash and Lingam (2021) who revealed that illiteracy, age, marital status, coupled with poor negotiating skills affected the demand for comprehensive health insurance amongst women. Chakrabarti and Shankar (2015) indicated that, tribe, wealth, region and age were important factors in the demand for community based, public and private health insurance schemes. In terms of enrollment and renewal of individual insurance membership Nshakira-Rukundo et al. (2019) revealed that the amount of wealth an individual had was a key factor in the enrollment and renewal of community health insurance.
According to Gao, Guan & Wang (2021), over the years, the role of media in relaying information to society cannot be ignored. Media not only relays information about entertainment, political matters but also relays health related information to society like information about the benefits of having health insurance and the risks of not having health insurance plus the different types of health insurance individuals can explore (Gao et al., 2021; Chakrabarti & Shankar, 2015).
Following Mathauer et al. (2008) and Nandi et al. (2016), the lack of information affects the demand of different types of health insurance. According to Mathauer et al. (2008), limited information not only affects the demand, but it also affects the ability of individuals who are already enrolled on different forms of health insurance schemes to pay.
Focusing on enrolling on each type of health insurance, media plays a more crucial role in private and community health insurance enrollment, and this is more common among individuals in urban areas as compared to those in rural areas (Chakrabarti & Shankar, 2015). This is supported by Maradze et al. (2020) who suggests that getting information from different media channels like from television, radio and newspapers informs individuals about the available health insurance plans which eventually leads to utilization of health insurance. Even in areas where the demand for health insurance is low, Mulenga et al. (2016) indicates that media improved the demand for health insurance amongst women. This is supported by Mathuer et al. (2008) who pointed out not listening to radio, not watching television and not reading newspapers were major barriers to enrollment in social health insurance.
The study used secondary data on women obtained from the 2016 Uganda Demographic and Health Survey collected by Uganda Bureau of Statistics (UBOS) and Inner City Fund (ICF) (2018). The 2016 Uganda Demographic Health Survey (UDHS) sampling frame was based on the frame of the 2014 Uganda National Population and Housing Census (NPHC) provided by the Uganda Bureau of Statistics (UBOS). The 2014 NPHC census frame was a complete list of all enumeration areas (EAs) whereby each EA covered 130 households and belonged to a sampling frame which had information about the location, residence type and the estimated number of residential areas (UBOS & ICF, 2018).
Furthermore, according to UBOS & ICF (2018), the 2016 UDHS sample was stratified and selected in two stages. In the first stage, 697 EAs were selected with 162 EAs in urban areas and 535 in rural areas. At the second stage a household list was compiled in each of the 696 accessible EAs from April to October 2016, with some listing overlapping with fieldwork. To minimize household listing, each large EA (one with more than 250 households) was segmented and only one segment, with probability proportional to the segment size, was selected for the survey. UBOS & ICF (2018), further state that in total, a representative sample of 20,880 households (30 from each EA segment) were randomly selected for the 2016 UDHS. Out of the 20,880 selected household, 18,506 women were successfully interviewed and the individual data that was generated from the interviews was used in the study with no exclusion of any interview data, with an average of 1,200 complete interviews carried out per domain. All women aged 15-49 who were either permanent residents of the selected households or visitors who stayed in the household the night before the survey were eligible to be interviewed. For younger women consent to be interviewed was sought from parents or guardians. Information on several demographic, health and media related information were gathered from the participants during interviews. This interview information was used to build datasets whose data formed the basis of the study analysis (see Underlying data for information on how to access the associated data).
Due to the use of secondary DHS data, no ethical approval was required. However, The DHS Program maintains strict standards for protecting the privacy of respondents and household members in all DHS surveys. An elaborative statement on ethical review, informed and voluntary participation, privacy and confidentiality during data collection and data processing, biomarker referral, treatment, and counselling can be accessed from the DHS website here.
The outcome and independent variables were selected from literature of studies (Mpuuga et al. (2020); Ayanore et al. (2019); Aregbeshola & Khan (2018); Ssempala (2018); Mulenga et al. (2016) and Kimani et al. (2014)) with focus on determinants of community, employer provided and privately purchased health insurance.
The dependent variable was whether a woman had or didn’t have any of community provided health insurance, employer provided health insurance and privately purchased health insurance. The explanatory variables included level of education, wealth, age, health status, residence and marital status which were categorized as demographic factors while the number of times a woman watches television, reads newspapers and listens to radio a week were categorized as media factors.
Data was analyzed at three stages using STATA (version 15, Stata Corp 2015). At stage one analysis of the data, a summary of all outcome variables and independent variables was done. Frequencies and percentages were used for the categorical variables. In stage two we looked at the bivariate analysis between the dependent variables and categorical independent variables using Pearson’s Chi-square test.
Explanatory variables that had significant associations i.e. at 1%, 5% and 10% level of significance with community, employer provided, and private health insurance were considered for the third stage analysis. At the third stage, the Firth logistic was adopted to identify significant demographic and media determinants for different types of health insurance. Choice of the Firth logistic was based on the fact that it provides better estimates as compared to the normal or ordinary logistic regression while in terms of maximum likelihood estimation the firth logistic uses penalized likelihood to reduce sample bias in maximum likelihood estimation while the normal or ordinary logistic regression suffers from small sample bias (Okyere et al., 2021; Njau, 2019).
Results of the of the descriptive analysis presented in Table 1 indicate that 0.44 % of women subscribed to community health insurance schemes, 0.64% had health insurance provided by their employers or place of work while 0.18% women privately purchased their own health insurance. This shows that the number of women who are having any type of health insurance is very low and is indeed a rare event. Results of the demographic factors indicated that 21% of the women came from the richest income households, had attained primary level of education (58.86%), were between 15-24 years of age (43.54%), married (31.41%), didn’t smoke (99.27%) and lived in rural areas (76.43%). For media factors, (70.51%) of the women claimed not to have watched television at all, (79.05%) had never read newspapers while (57.13%) got information from radio at least once a week.
Table 2 shows results of the association between women’s demand for different types of health insurance and categorical explanatory variables.
Variable | Community health insurance | Employer provided insurance | Private health insurance | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
No | Yes | Total | Probability value | No | Yes | Total | P value | No | Yes | Total | Probability value | |
Demographic factors | ||||||||||||
Women’s wealth status | ||||||||||||
Poorest Income household | 3,881 | 3 | 3,884 | 0.001* | 3,882 | 2 | 3,884 | 0.000*** | 3,882 | 2 | 3,884 | 0.002** |
Poorer Income household | 3,624 | 16 | 3,640 | 3,639 | 1 | 3,640 | 3,638 | 2 | 3,640 | |||
Middle Income household | 3,463 | 22 | 3,485 | 3,484 | 1 | 3,485 | 3,479 | 6 | 3,485 | |||
Richer Income household | 3,430 | 24 | 3,454 | 3,441 | 13 | 3,454 | 3,446 | 8 | 3,454 | |||
Richest Income household | 4,026 | 17 | 4,043 | 3,942 | 101 | 4,043 | 4,027 | 16 | 4,043 | |||
Women’s level of Education | ||||||||||||
No education | 2,054 | 17 | 2,071 | 0.035* | 2,071 | 0 | 2,071 | 0.000*** | 2,070 | 1 | 2,071 | 0.000*** |
Primary education | 10,847 | 46 | 10,893 | 10,888 | 5 | 10,893 | 10,883 | 10 | 10,893 | |||
Secondary level education | 4,197 | 16 | 4,213 | 4,182 | 31 | 4,213 | 4,197 | 16 | 4,213 | |||
Higher level education | 1,326 | 3 | 1,329 | 1,247 | 82 | 1,329 | 1,322 | 17 | 1,329 | |||
Age group | ||||||||||||
15-24 years | 8,040 | 18 | 8,058 | 0.001* | 8,028 | 30 | 8,058 | 0.000*** | 8,047 | 11 | 8,058 | 0.442 |
25-34 years | 5,580 | 34 | 5,614 | 5,561 | 53 | 5,614 | 5,600 | 14 | 5,614 | |||
35-44 years | 3,628 | 22 | 3,650 | 3,617 | 33 | 3,650 | 3,644 | 6 | 3,650 | |||
45+ years | 1,176 | 8 | 1,184 | 1,182 | 2 | 1,184 | 1,181 | 3 | 1,184 | |||
Women’s marital status | ||||||||||||
Never been married | 4,728 | 10 | 4,738 | 0.000*** | 4,704 | 34 | 4,738 | 0.058 | 4,730 | 8 | 4,738 | 0.901 |
Currently married | 5,764 | 49 | 5,813 | 5,764 | 49 | 5,813 | 5,802 | 11 | 5,813 | |||
Currently living with partner | 5,549 | 17 | 5,566 | 5,538 | 28 | 5,566 | 5,554 | 12 | 5,566 | |||
Lost partner (widowed) | 521 | 2 | 523 | 521 | 2 | 523 | 523 | 0 | 523 | |||
Currently divorced | 138 | 1 | 139 | 139 | 0 | 139 | 139 | 0 | 139 | |||
Separated no longer living together | 1,724 | 3 | 1,727 | 1,722 | 5 | 1,727 | 1,724 | 3 | 1,727 | |||
Women’s health status | ||||||||||||
does not smoke | 18,289 | 82 | 18,371 | 0.739 | 18,253 | 118 | 18,371 | 0.646 | 18,338 | 33 | 18,371 | 0.017* |
smokes every day | 55 | 0 | 55 | 55 | 0 | 55 | 54 | 1 | 54 | |||
Smokes some days | 80 | 0 | 80 | 80 | 0 | 80 | 80 | 0 | 80 | |||
Residence | ||||||||||||
Urban area | 4,367 | 12 | 4,379 | 0.054 | 4,295 | 84 | 4,379 | 0.000*** | 4,362 | 17 | 4,379 | 0.000*** |
Rural area | 14,057 | 70 | 14,127 | 14,093 | 34 | 14,127 | 14,110 | 17 | 14,127 | |||
Media factors | ||||||||||||
Frequency of watching television | ||||||||||||
Does not watch television at all | 12,995 | 53 | 13,048 | 0.493 | 4,965 | 9 | 4,974 | 0.000*** | 13,033 | 15 | 13,048 | 0.000*** |
Watches television less than once a week | 1,961 | 11 | 1,972 | 2,936 | 23 | 2,959 | 1,969 | 3 | 1,972 | |||
Watches television at least once a week | 3,468 | 18 | 3,486 | 10,487 | 86 | 10,573 | 3,470 | 16 | 3,486 | |||
Frequency of reading newspaper | ||||||||||||
Does not read newspapers at all | 14,563 | 66 | 14,629 | 0.907 | 14,602 | 27 | 14,629 | 0.000*** | 14,611 | 18 | 14,629 | 0.001** |
Reads newspapers less than once a week | 2,264 | 10 | 2,274 | 2,252 | 22 | 2,274 | 2,265 | 9 | 2,274 | |||
Reads newspapers at least once a week | 1,597 | 6 | 1,603 | 1,534 | 69 | 1,603 | 1,596 | 7 | 1,603 | |||
Frequency of listening to radio | ||||||||||||
Does not listen to radio at all | 4,965 | 9 | 4,974 | 0.003* | 4,965 | 9 | 4,974 | 0.000*** | 4,969 | 5 | 4,974 | 0.120 |
Listens to radio less than once a week | 2,946 | 13 | 2,959 | 2,936 | 23 | 2,959 | 2,950 | 9 | 2,959 | |||
Listens to radio at least once a week | 10,513 | 60 | 10,573 | 10,487 | 86 | 10,573 | 10,553 | 20 | 10,573 |
The demographic factors that were associated significantly (p≤0.05) with community health insurance included wealth, level of education, age and marital status of women. The number of women who had community health insurance where from richer income households (17), had attained primary level of education (46), were between 25-34 years (34) and married (49). Similarly, the demographic factors that were significantly (p≤0.05) related with employer provided health insurance included wealth, level of education, age and place of residence of women. The number of women respondents whose place of work provided health insurance were from richest income households (101), had higher level of education (82), aged between 25-34 years (53) and where from urban areas (84). However, the wealth of women, women education level, women health status and women place of residence were significantly (p≤0.05) associated with private health insurance. Women form the richest income households (16), attained higher level of education (17), didn’t smoke (33), came from urban (17) and rural (17) areas had private health insurance.
The only media factor that was significantly (p≤0.05) associated with community health insurance was the number of times women listened to the radio. Sixty (60) who women listened to radio at least once a week had community organized health insurance. The number of times women watched television, the number of times women read newspapers and the number of times women listened to radio were significantly (p≤0.05) related to employer provided health insurance. watching television at least once a week (86), reading newspapers at least once a week (69) and listening to radio at least once a week (86) was common among women who had employer provided health insurance. likewise, the number of times women watched television and read newspapers, was significantly related to possession of private health insurance among women. watching television at least once a week (16) and not reading newspapers at all (18) was common among women who didn’t have private health insurance.
Results of the Firth logistic regressions for factors affecting the demand of different types of health insurance are shown in Table 3.
Variables | (1) | (2) | (3) |
---|---|---|---|
Community health insurance | Employer provided health insurance | Private health insurance | |
Demographic factors | |||
Education level | |||
No education | 1.856*** (0.634) | -3.966*** (1.484) | -1.219 (1.001) |
Primary level of education | 0.991* (0.588) | -3.257*** (0.498) | -1.072* (0.578) |
Secondary level of education | 0.755 (0.600) | -1.397*** (0.227) | -0.127 (0.463) |
Higher level of education | Reference category | Reference category | Reference category |
Residence | |||
Urban area | 0.203 (0.235) | 0.455 (0.446) | |
Rural area | Reference category | Reference category | |
Women wealth status | |||
Poorest income household | -2.053*** (0.619) | -0.701 (0.738) | -0.394 (0.880) |
Poorer income household | -0.299 (0.375) | -1.524* (0.883) | -0.419 (0.848) |
Middle income household | 0.0151 (0.346) | -1.820** (0.871) | 0.472 (0.670) |
Richer income household | 0.181 (0.330) | -0.376 (0.354) | 0.440 (0.559) |
Richest income household | Reference category | Reference category | Reference category |
Age group | |||
15-24 Years | -0.385 (0.475) | 0.284 (0.672) | |
25-34 Years | 0.292 (0.400) | 0.828 (0.660) | |
35-44 Years | 0.0236 (0.407) | 1.324** (0.670) | |
Over 45 Years | Reference category | Reference category | |
Women marital status | |||
Never been married | 0.547 (0.655) | ||
Currently married | 1.441*** (0.556) | ||
Currently living with partner | 0.464 (0.588) | ||
Lost partner (widowed) | 0.832 (0.839) | ||
Currently divorced | 1.638* (0.983) | ||
Separated no longer living together | Reference category | ||
Women health status | |||
Doesn’t smoke | -1.484 (1.436) | ||
Smokes every day | 1.985 (1.663) | ||
Smokes some days | Reference category | ||
Media factors | |||
Frequency of watching television | |||
Doesn’t watch television at all | -0.576* (0.333) | -0.551 (0.522) | |
Watches television once a week | -0.520 (0.334) | -0.686 (0.632) | |
Watches television at least once a week | Reference category | Reference category | |
Frequency of reading newspapers | |||
Doesn’t read newspapers at all | -0.943*** (0.260) | -0.289 (0.504) | |
Reads newspapers less than once a week | -0.932*** (0.258) | 0.165 (0.504) | |
Reads newspapers at least once a week | Reference category | Reference category | |
Frequency of listening to radio | |||
Doesn’t listen to radio at all | -0.868** (0.357) | -0.190 (0.359) | |
Listens to radio once a week | -0.0205 (0.306) | 0.468* (0.252) | |
Listens to radio at least once a week | Reference category | Reference category | |
Constant | -6.832*** (0.847) | -2.917*** (0.677) | -3.757** (1.536) |
Observations | 18,506 | 18,506 | 18,506 |
Results in Table 3 indicate the Adjusted Odds Ratios (AOR) of several demographic and media determinants for community provided health insurance, employer provided health insurance and privately purchased health insurance.
Lack of education (AOR= 1.856, se=0.634), attainment of primary level of education (AOR=0.991, se=0.588), being married (AOR=1.441, se=0.556), being divorced (AOR=1.683, se=0.983) significantly (p≤0.05) increased the likelihood demand of community health insurance among women. However, coming from the poorest income household (AOR= -2.053, se=0.619) significantly decreased the likelihood demand for community health insurance among women.
The results further revealed that, having no education (AOR=-3.966, se=1.484), having attained primary level of education (AOR=-3.257, se=0.498), attained secondary level of education (AOR=-1.397, se=0.227), coming from middle income households (AOR=-1.820, se=0.871), significantly reduced the likelihood demand for employer provided health insurance among women. On the contrary, belonging to the 35-44 years’ age category (AOR=1.324, se=0.670) significantly increased the likelihood demand for employer provided health insurance among women. The likelihood demand for private health insurance significantly reduced among women who attained only primary level of education (AOR=-1.072, se=0.578).
Not listening to radio at all (AOR= -0.868, se=0.357) significantly decreased the likelihood demand for community health insurance among women. Our results also revealed that, not watching television at all (AOR=-0.576, se=0.333), not reading newspapers at all (AOR=-0.943, se=0.260), and reading newspapers less than once a week (AOR=-0.932, se=0.258) significantly reduced the likelihood demand for employer provided health insurance among women. However, listening to radio once a week (AOR=0.468, se=0.252) significantly increased the likelihood demand for employer provided health insurance among women.
Results of the study indicate that only 0.44 % of the women had community health insurance, 0.64% had health insurance that was provided by their places of work while 0.18% of the women privately purchased their own health insurance. This is consistent with the findings of earlier scholars (See Ssempala, 2018; Christiani et al., 2017; Kimani et al., 2014; Aregbeshola & Khan, 2018), who also observed low demand for health insurance amongst women. A possible explanation for this could be the high levels of poverty which is more rampant amongst Ugandan women (UBOS, 2017).
Our findings revealed an important relationship among wealth of women possession of community health insurance, health insurance provided by employers and privately purchased health insurance. Women from richer households had community health insurance while women from richest households had health insurance provided by their respective employers and also privately purchased their own health insurance. This can be attributed to the fact that women from poorest households do not have enough monetary resources and are also anticipated to face challenges in paying back the premium contributions. (Aregbeshola & Khan, 2018). This finding is in agreement with studies like Nshakira – Rukundo et al. (2019); Aregbeshola and Khan (2018); Ayanore et al. (2019) and Kimani et al. (2014) that showed the role played by wealth of individuals as an important predictor in the demand for different types of health insurance.
Similarly, significant relationships are observed between highest level of education and possession of any type of health insurance. Women with only primary level of education had community health insurance while women with higher level of education had health insurance provided by their respective employers but were also in position to purchase their own private health insurance. Such results are in line with Aregbeshola and Khan (2018) who argued that because of education women are in position to make informed choices like purchasing a type of health insurance which suites their needs and they are also aware about the advantages of having health insurance. This points out the increasing role of education in the demand for different types of health insurance as suggested in numerous studies like (see Aregbeshola & Khan, 2018; Kimani et al., 2014; Boateng & Awunyor-Vitor, 2013; Govender et al., 2013).
Our study results also revealed age of women was significantly related to community based health insurance and health insurance provided by the women place of work. Women aged 25 to 34 years had community based health insurance and health insurance provided by the women place of work. The results are consistent with Aregbeshola and Khan (2018) who argued as women grow old their risk of illness also increases but also they are more experienced and have access to finances which are some of the reasons why they may have increased investments in health including enrollment in community based health insurance schemes. Similarly, Ayanore et al. (2019), argues that the women willingness to obtain any type of Medicaid is strongly associated with the age of the woman more so if they are in the advanced age category. It is worth noting that women aged 25 to 34 years in Uganda are not in possession of privately purchased health insurance. A possible explanation for this is that most of the available private health insurance offers on the market are expensive and not affordable (Mulenga et al., 2016). Studies like Ayanore et al. (2019); Aregbeshola and Khan (2018) and Mulenga et al. (2016) present similar results.
In this study an important association is observed among married women and those have health insurance organized at community level. This is so because in most cases this type of health insurance covers your spouse i.e. married women benefit because the health insurance given to their husbands covers them as well. But also on the other hand married men may also convince their wives to enroll for community health insurance because of the benefits of having health insurance. This is consistent with the results of authors like Mulenga et al. (2016); Kimani et al. (2014) and Boateng and Awunyor-Vitor (2013) who established that the likelihood of having health insurance was high among married couples as compared to the unmarried couples. Given the nature of community health insurance as a health insurance that serves the core needs of community with voluntary membership targeting those outside the formal sector (Mulenga et al., 2016). It is also worth noting that married Ugandan women are not in formal employment which is known to provide health insurance to most individuals employed in the sector as compared to informal employment where most individuals are involved in businesses where the earnings are very few for the purchase private health insurance (Ikoja-Odongo, 2002).
Findings of the study also revealed compelling evidence of an association between women health status and the purchase of private health insurance. Following Yamada et al. (2014) a possible explanation for this would be individuals with poor health tend to acquire health insurance as compared to those in good health, this is done to mitigate and care for the side effects of poor health.
It was also established the women place of residence was a momentous determinant of health insurance provided by women’s place of work and privately purchased health insurance. Following Aregbeshola and Khan (2018); Mulenga et al. (2016), A possible explanation for this is that most women in urban areas are employed in formal sector and private health insurance companies are highly mostly found operating in urban areas than in rural areas. This is in agreement with results of studies like; Aregbeshola and Khan (2018); Ssempala (2018) and Mulenga et al. (2016).
Having no education at all and attaining only primary level of education significantly increased the likelihood demand for community health insurance by Ugandan women. following Odokonyero et al. (2017), a possible explanation for this is that community health insurance is that community health insurance is a community based and common in rural areas with a target of providing health insurance to communities which are not in formal employment with low levels of education.
Having no education at all, having attained on primary or secondary level of education significantly decreased the women’s likelihood of demanding employer provided health insurance. According to Ayanore et al. (2019) These results present the increasing role of education in the demand employer health insurance and further indicate that acquiring a degree or its equivalent creates opportunities for employment which would translate into enrollment into employer provided health insurance or privately purchasing a health insurance cover. Similarly, having attained only primary level of education significantly reduced women’s demand for private health insurance. Ssempala (2018) also argues that lack of education hinders women potential to earn an income thus stopping them from putting some money aside to purchase for health insurance. similar results are reported by authors like (Ayanore et al., 2019; Ssempala, 2018).
Belonging to poorest income households significantly reduced the women’s demand for community based health insurance and enrollment on health insurance provided by the women’s place of work. Similarly, coming from middle income households significantly reduced women’s demand for employer provided health insurance. According to Ayanore et al. (2019) a possible explanation for this is that women who don’t have enough wealth can’t afford health insurance and are less likely to enroll for any type of health insurance. A similar argument is put forward by Mpuuga et al. (2020) who argues that rich women are capable of purchasing a health insurance plan given the fact that they have enough resources at their exposure as compared to poor women who are financially constrained. This is in agreement with the findings of (Ayanore et al., 2019; Ssempla, 2018; Mpuuga et al., 2020).
The possibility of enrolling for employer provided Medicaid was powerful in women aged between 35-45 years. following Mulenga et al. (2016) A possible explanation for this is as one grows older their chances of getting ill are also high thus they need health insurance as protection to the risk of illness. Ssempla (2018) also argues that increase in age and being a female are positively related to enrolling into a health insurance scheme thus elderly household heads and household which are headed by females are more likely to enroll and acquire Medicaid. Mulenga et al. (2016) and Kimani et al. (2014) also agree with these findings.
Similarly, the likelihood demand for community health insurance significantly increased among married and divorced women. According to Ssempala (2018) a possible reason for this is that married individuals may take better care of themselves since they have people that depend on them. This is in agreement with results of Kirigia et al. (2005) however we could not find a plausible explanation as to the likelihood for community health insurance significantly increased among divorced women.
Watching television and reading newspapers was significantly associated with the women demand for employer health insurance and private purchased Medicaid. Also worth noting is that listening to radio was significantly associated with the women’s demand for community and employer provided health insurance. A possible explanation for this is that the government of Uganda and other stake holders have informed public about the benefits and of possessing Medicaid through different media outlets like television, newspapers and radio. This finding is in agreement with studies like Ssempala (2018); Mulenga et al. (2016) and Kimani et al. (2014).
Our results also show that the likelihood of demanding for employer provided health insurance decreased with not watching television, not reading newspapers and reading newspapers once a week. Similarly, not listening to radio significantly reduced the likelihood enrollment for community based health insurance amongst women. this is self-explanatory that not reading papers, not watching television and reading newspapers once a week will make women miss out on key information about different types of Medicaid. However, chances of enrolling for employer provided Medicaid significantly increased with listening to radio once a week. Authors like Nshakira-Rukundo et al. (2019) and Ssempala (2018) also agree with these findings.
The study concludes that enrollment for community provided health insurance, employer provided health insurance and privately purchased health insurance by women is still very low in Uganda. Additionally, demographic factors (wealth, level of education, age, current marital status) and media factors (number of times women watch television, number of times women read newspapers and number of times women listen to radio) were significant predictors in the demand for community provided health insurance, employer provided health insurance and privately purchased health insurance. The findings give an insight into important information that several stakeholders like government and other players like donors should focus on while formulating and implementing health insurance related policies. There is also need to tailor policies aimed at enhancing factors that will promote women’s enrollment into different types of Medicaid while working on issues that deter women’s enrollment into community provided health insurance, employer provided health insurance and privately purchased health insurance.
To understand the demographic and media factors affecting enrollment into community provided health insurance, employer provided health insurance and privately purchased health insurance amongst women from developed and underdeveloped countries, other researchers should undertake a comparative study to this effect. Other than focusing on the demand factors alone, it would be beneficial to also understand demographic and media determinants that affect supply community provided health insurance, employer provided health insurance and privately purchased health insurance.
Data used in this study are from the women recode dataset of the Uganda 2016 Demographic Health Survey, available from the Demographic and Health Survey (DHS) website https://dhsprogram.com/data/available-datasets.cfm. The data was collected by Uganda Bureau of Statistics (UBOS) and Inner City Fund (ICF) (2018). Access to the dataset requires registration and is granted only for legitimate research purposes. A guide for how to apply for dataset access is available at: https://dhsprogram.com/data/Access-Instruction.cfm
Once downloaded, the datasets must not be passed on to other researchers without the written consent of The DHS Program. All reports and publications based on the requested data must be sent to The DHS Program Data Archive in a Portable Document Format (pdf).
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Is the work clearly and accurately presented and does it cite the current literature?
Partly
Is the study design appropriate and is the work technically sound?
Partly
Are sufficient details of methods and analysis provided to allow replication by others?
Partly
If applicable, is the statistical analysis and its interpretation appropriate?
Partly
Are all the source data underlying the results available to ensure full reproducibility?
Partly
Are the conclusions drawn adequately supported by the results?
Partly
References
1. Basaza RK, Namyalo PK, Mutatina B: Health Financing Reforms in Uganda: Dispelling the Fears and Misconceptions Related to Introduction of a National Health Insurance Scheme Comment on "Health Coverage and Financial Protection in Uganda: A Political Economy Perspective".Int J Health Policy Manag. 2023; 12: 7348 PubMed Abstract | Publisher Full TextCompeting Interests: No competing interests were disclosed.
Reviewer Expertise: I am an economist with experience in Uganda but my experience in studying health insurance is limited to Ghana. I am also not very familiar with the specific questions about health insurance from the DHS in 2016.
Is the work clearly and accurately presented and does it cite the current literature?
Yes
Is the study design appropriate and is the work technically sound?
Yes
Are sufficient details of methods and analysis provided to allow replication by others?
Yes
If applicable, is the statistical analysis and its interpretation appropriate?
Yes
Are all the source data underlying the results available to ensure full reproducibility?
Yes
Are the conclusions drawn adequately supported by the results?
Yes
Competing Interests: No competing interests were disclosed.
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | ||
---|---|---|
1 | 2 | |
Version 1 24 Mar 22 |
read | read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)