This paper examines the influence of health insurance status on healthcare use in rural Ghana using 286 sampled respondents from four rural communities in the Bekwai Municipality. Data were obtained using structured interview and Pearson’s Chi square and bivariate regressions were used to analyse data. The results show low healthcare utilization among study participants, with most respondents having irregular use (43.5%) or rare use (43.3%). Respondents with health insurance utilized healthcare more than those without health insurance, the results being statistically significant (df = 4;
Good health is a major contributor to quality of life, enabling people to fully participate in productive activities for wealth creation. It is also a key determinant of human development [
The extent of healthcare use is however related to several factors including (physical) access, cost of service, and a host of other socioeconomic and personal characteristics [
Increasing global attention is being drawn towards universal health coverage, and policies on health insurance play a strategic role in that drive. As a form of prepaid financing system, health insurance makes possible collective pooling of risks and the redistribution of financial resources in such a way that assures financial protection against the cost of illness [
Ghana, a lower middle income country with even more ambitious development targets, requires a healthy population as a key ingredient for sustained economic and social progress, hence a clear policy focus on health. Addressing financing is to touch on a key component of the healthcare challenges in Ghana. Before independence in 1957, healthcare in Ghana was mainly financed by out-of-pocket payments at point of service use [
A healthy rural population is needed in Ghana for several reasons. First, the rural sector contributes immensely to the growth and development of the country in several ways such as agriculture to produce food for local consumption as well as for exports for foreign exchange, which forms a bigger proportion of the national income. The rural sector also produces raw materials to feed local manufacturing industries. Also, a large percentage of the Ghanaian population lives in rural areas. The 2010 Ghana Population and Housing Census revealed that almost half (49.1%) of Ghanaians live in rural areas, and, in the Bekwai Municipality where this study was undertaken, out of total population of 118,024, about 97,277 representing 82.4% reside in rural areas [
Nevertheless, when it comes to health, rural areas in Ghana face difficulties with transport and communication, as well as challenges of shortages of doctors, nurses, and other health professionals and health service providers. The rural people in Ghana are also mostly low income people, which imply that they are least able to cater for the cost of accessing health services when the need arises. These challenges lead to low levels of healthcare utilization even in the face of ill health, with several negative consequences [
However, one way of improving access and use of healthcare by the rural people is to reduce or eliminate barriers of financing which includes the elimination of out-of-pocket expenditure, and this calls for health insurance. Though a subject of much research in the literature not much is known about the influence of health insurance status on the use of healthcare by rural people. This study thus examines the influence of health insurance status on healthcare use in rural Ghana, with selected rural communities in the Bekwai Municipal as case study. The Bekwai Municipality where the study was conducted has as many as 82.4% of the population residing in rural areas [
The study is guided by the following hypothesis: H0: There is no significant difference in the use of healthcare between insured and uninsured people. H0: Health insurance has no significant influence on the use of healthcare.
This study formed part of a larger research that investigated the determinants of healthcare use in rural Ghana within the Bekwai Municipality. The study adopted a population-based cross-sectional and quantitative survey design to investigate the influence of insurance status on the use of healthcare, using four (4) selected communities, namely, Senfi, Sehwi, Chiransah, and Huntado. The Bekwai Municipality was selected because of its rural character with its very few health facilities. These communities were selected because of their nearness to health facilities. Systematic random sampling was used to select the housing units from which one household representative was randomly chosen as a respondent.
A total of 286 respondents were selected from a combined 1,003 housing units in the four communities, with the selection based on the total number of housing units in each community. The respondents constituted only heads of households. However, in a situation where the head of a selected household is not available, any member of the household aged 18 years and above and willing to participate was selected. A list of housing units was collected for each community from the Municipal Assembly and was used as sample frame. The total numbers of housing units in each of the study communities were 373, 422, 149, and 68 for Senfi, Huntado, Sehwi, and Chiransah, respectively whereas the samples selected from each community were 103, 116, 41, and 26 for Senfi, Huntado, Sehwi, and Chiransah, respectively. A sampling interval of 4 was used to select the housing units in Senfi and Huntado. This sampling interval was calculated by dividing the total number of households in Senfi and Huntado (373 and 422, resp.) by the sample sizes of 103 and 116 to be, respectively, selected from the two communities. The sample interval of 3 for selection of housing units in Sehwi and Chiransah was also derived by dividing the total number of housing units of 149 and 68, respectively, by the sample sizes of 41 and 26. In every community, one house was randomly selected between the first house on the list and the sample interval and the sample selection continued using the sample interval for every community. In a situation where there were more than one household in a dwelling, only one household was randomly selected with the assumption that all the households had similar characteristics. Also, in a situation where nobody was found in a selected house or where no member of a particular house was willing to participate, the next house was selected and the same sample interval was used to select the subsequent houses from there.
Primary data was collected using structured questionnaire. The low literacy levels in the study communities necessitated adoption of the interviewer-administered approach. The interactions were also done in the local Twi language which is the most patronized dialect in the study area. Participation in the study was voluntary, and informed consent was obtained before interviewing began. All the respondents were adults aged 18 years and above. A total of 283 out of the 286 rural people sampled responded to the questionnaire giving a response rate of 99%.
The dependent variable for the study was healthcare use, and this was operationalised as the number of times a respondent visited a health facility in the last four (4) times of illness spells. Healthcare use comprised total doctor visits, general practitioner visits, and specialist visits and was ranked following Buor [
Continuous and ranked data were used in the quantitative analysis and bivariate regression was run to determine the effect of insurance status on use of healthcare. A 0.05 or less probability (
Table
Background characteristics of study participants by health insurance status.
Variables | Categories | National Health Insurance Status | Total |
|
||||
---|---|---|---|---|---|---|---|---|
Insured |
Uninsured |
|||||||
|
(%) |
|
(%) |
|
(%) | |||
Gender | Male | 60 | (45.1) | 66 | (44.0) | 126 | (44.5) | |
Female | 73 | (54.9) | 84 | (56.0) | 157 | (55.5) | 0.473 | |
|
||||||||
Age | 18–39 | 21 | (15.8) | 63 | (42.0) | 84 | (29.7) | |
40–59 | 49 | (36.8) | 54 | (36.0) | 103 | (36.4) | 0.052 | |
60 and above | 63 | (47.4) | 33 | (22.0) | 96 | (33.9) | ||
|
||||||||
Education | No education | 49 | (36.8) | 58 | (38.7) | 107 | (37.8) | |
Basic education | 42 | (31.6) | 80 | (53.3) | 102 | (36.0) | ||
Secondary | 14 | (10.5) | 12 | (8.0) | 26 | (9.1) | 0.000 | |
Tertiary education | 28 | (21.1) | 0 | (0.0) | 28 | (17.1) | ||
|
||||||||
Employment status | Unemployed | 31 | (24.8) | 19 | (12.7) | 50 | (17.7) | |
Employed | 100 | (75.2) | 131 | (87.3) | 231 | (82.3) | 0.012 | |
|
||||||||
Nature of occupation | Farming | 38 | (38.0) | 70 | (53.4) | 108 | (46.8) | |
Trading | 16 | (16.0) | 23 | (17.6) | 39 | (16.9) | ||
Artisan | 25 | (25.0) | 39 | (29.0) | 64 | (27.7) | 0.061 | |
Public/civil service | 21 | (21.0) | 0 | (0.0) | 21 | (8.6) | ||
|
||||||||
Marital status | Single | 34 | (25.6) | 44 | (29.3) | 78 | (27.6) | 0.020 |
Married | 99 | (74.4) | 106 | (70.7) | 205 | (72.4) | ||
|
||||||||
Household size | 1–5 | 63 | (47.4) | 15 | (10.0) | 78 | (27.6) | |
6–10 | 65 | (48.7) | 121 | (80.7) | 186 | (65.7) | 0.000 | |
>10 | 5 | (3.6) | 14 | (9.3) | 19 | (6.7) | ||
|
||||||||
Average monthly income (GH |
≤100 | 42 | (31.6) | 82 | (54.7) | 124 | (43.8) | |
101–400 | 57 | (42.9) | 47 | (31.3) | 104 | (36.7) | ||
401–600 | 21 | (15.8) | 11 | (7.3) | 32 | (11.2) | 0.012 | |
601–1000 | 9 | (6.8) | 6 | (4.0) | 15 | (5.3) | ||
>1000 | 4 | (2.9) | 4 | (2.7) | 8 | (3.0) |
Older people were more uninsured in terms of percentage though the difference was not significant (
Results on the extent of healthcare use by respondents are presented in Table
Selected questions regarding participants’ healthcare use.
Variables | Categories | Health Insurance Status | Total |
|
||||
---|---|---|---|---|---|---|---|---|
Insured ( |
Uninsured ( |
|||||||
|
(%) |
|
(%) |
|
(%) | |||
Quality of service | Very poor | 0 | (0.0) | 7 | (4.7) | 7 | (2.5) | |
Poor | 12 | (9.0) | 16 | (10.7) | 28 | (9.9) | ||
Satisfactory | 57 | (42.7) | 71 | (47.3) | 128 | (45.2) | ||
Good | 51 | (38.2) | 15 | (10.0) | 66 | (23.3) |
|
|
Very good | 10 | (7.5) | 0 | (0.0) | 10 | (3.6) | ||
|
3 | (2.6) | 41 | (27.3) | 44 | (15.5) | ||
|
||||||||
Attitude of staff | Very poor | 4 | (3.0) | 8 | (5.3) | 12 | (4.2) | |
Poor | 28 | (21.1) | 52 | (34.7) | 80 | (28.3) | ||
Satisfactory | 58 | (43.5) | 53 | (35.3) | 111 | (39.2) | ||
Good | 25 | (18.5) | 3 | (2.0) | 28 | (9.9) |
|
|
Very good | 15 | (11.3) | 0 | (0.0) | 15 | (5.3) | ||
Missing | 3 | (2.6) | 34 | (22.7) | 37 | (13.1) | ||
|
||||||||
Healthcare use | 0 (rarely) | 10 | (7.5) | 78 | (52.0) | 88 | (43.3) | |
1 (irregularly) | 56 | (42.2) | 67 | (44.7) | 123 | (43.5) | 0.000 | |
2 (moderately) | 32 | (24.1) | 5 | (3.3) | 37 | (13.1) | ||
3 (regularly) | 21 | (15.7) | 0 | (0.0) | 21 | (7.4) | ||
4 (very regularly) | 14 | (10.5) | 0 | (0.0) | 14 | (6.9) |
The bivariate analysis to compare the extent of healthcare use between the insured and uninsured respondents also resulted in a statistically significant difference (df = 4;
Table
Bivariate regression factor for influence of insurance status on use of healthcare.
Variable | Constant | Beta coefficient | Sig ( |
|
---|---|---|---|---|
Insurance status | 0.513 | 1.284 | 0.000 | 0.348 |
(a) Dependent variable: healthcare use.
The result from the bivariate regression analysis shows that health insurance status has a significant influence on use of healthcare (
An examination of the influence of health insurance status on use of healthcare services in rural Ghana was the focus of this study, using data gathered from the Bekwai Municipality. Specifically the study highlighted the differences between insured and uninsured people in terms of health seeking behaviour.
Use of healthcare by the rural folks was generally found to be low, and there was a significant difference between the insured and the uninsured. The study found that respondents with active health insurance cover generally tended to use healthcare more frequently than those with no active insurance cover. In other words, respondents who had actively registered under health insurance schemes tended to visit health facilities more frequently than those who had not. This result is worrying since low levels of healthcare services use has negative effects on the health conditions of the population, which is very vital for the socioeconomic development of the municipality and the nation as a whole. Reasons for the insured using healthcare services more than the uninsured may include weaker perceived health status by the former which makes them have a higher perceived need for healthcare. Another reason might be that the cost of accessing healthcare tends to be cheaper for the insured more than those not insured. Actively insured people find access and use of healthcare cheaper than the uninsured who have to pay for the use of healthcare services, and this prevents them from visiting health facilities when the need arises. Also, the insured were mostly found to be highly educated making them more likely to take advantage of health care services [
The study also found from the bivariate regression analysis that health insurance status has a positive and significant influence on use of healthcare services in the Bekwai Municipality. A number of factors may account for this finding. First, health insurance has a positive influence on the use of healthcare by way of making healthcare more accessible to the people in terms of reduced cost of service. Registered people no longer have to pay for healthcare services such as consulting, laboratory tests, drugs, and other health consumables. Also, health insurance reduces the financial burden of people since they no longer have to spend huge portion of their incomes on healthcare but rely on the insurance services. This finding implies that people with active health insurance cover are more likely to access and use healthcare services than those without active insurance cover when the need arises. Health insurance services promote the use of healthcare services for the rural dwellers leading to positive health outcomes, and this means that it is a policy that should be expanded to all sections of the population in order to promote good health which is very significant for development. This finding of a significant influence of people’s health insurance status on their use of healthcare services supports the findings of earlier studies in Israel [
Both null hypotheses for the study have been rejected. Firstly, the hypothesis that there is no significant difference in the use of healthcare between insured and uninsured people is rejected. The results revealed that people with active insurance cover utilize healthcare services more than those without insurance cover. The second hypothesis that health insurance has no significant influence on people’s use of healthcare is similarly rejected. The result revealed a positive and significant influence of health insurance status on use of healthcare services, such that the insured are more likely than the uninsured to use healthcare services when the need arises.
There were some limitations with regard to the methodology. There may be a possibility of recall bias in self-reported use of healthcare. There may also be some biases during the sampling procedure. For example, study region, district, and the rural communities were all selected purposively making it statistically unrepresentative. Also, the systematic random sampling technique used in selecting the households gives the tendency of losing some vital information from the target population that was skipped over. However, the research made use of a homogeneous sampling frame and could therefore report similar cases.
The study has shown that healthcare use among rural folks in the Bekwai Municipality is generally low and there is therefore the need to improve access and use of healthcare services. It has shown that rural people are mostly not registered under any health insurance scheme. In view of this, all interventions geared towards improving access to insurance services should be put in place by stakeholders such as the Ghana Health Service, Ministry of Health, National Health Insurance Authority, and other NGOs.
The study has also shown a significant difference between insured and uninsured participants with regard to healthcare services use, as well as the significant and positive influence of insurance status on rate of healthcare use. Therefore, given the importance of health insurance services in improvement of healthcare use and health outcomes, interventions should be made towards expanding its coverage to everyone particularly the low income people as well as those in the rural areas. The scheme’s management should highlight the communal nature of the scheme, especially targeting the relatively younger populations as well as well-informed educated individuals so as to discourage the notion that there are no incentives to enroll in health insurance schemes unless there is illness.
This study has successfully examined the essential difference in the use of healthcare services by people’s insurance status. The survey has used the bivariate regression technique to demonstrate the influence of health insurance status on the use of healthcare among rural dwellers.
The privacy and anonymity of all participants were protected, both during and after the research. Also, permission was obtained from the Bekwai Municipal Health Directorate and Municipal Assembly prior to the implementation of study methods.
This study was part of student’s project which was conducted under the general internal review of the substantive supervisors. Hence, all respondents were made to sign informed consent prior to participation, and benefits, intent, procedure, and expected risks were all explained to them.
The authors declare that there are no conflicts of interest regarding the publication of this paper.