Communities in Kenya, Tanzania, Ghana, Botswana, Namibia. Than Wi-Fi, a TVWS deployment would cost less, and would. Our system with Wi-Fi as well as WiMax based PHY and MAC. Ernment of Kenya's Ministry of Information and Communica. For commerce, education, healthcare and delivery of govern. This paper describes the frame work for deploying Wi-Fi and WiMAX. Request Full-text Paper PDF. Request the article directly. WiFi and WiMAX Deployment at the Ghana Ministry of Food and Agriculture. PREDICTORS OF HEALTH WORKERS' WILLINGNESS TO WORK IN RURAL COMMUNITIES OF SOKOTO STATE, NORTHWEST NI.
![](/uploads/1/2/7/6/127625823/240666636.jpg)
PMID: 22942455
This article has been cited by other articles in PMC.
Summary
Objectives
The study investigates the effect of Ghana's National Health Insurance Scheme (NHIS) on health care utilisation.
Methods
We provide a short history of health insurance in Ghana, and briefly discuss general patterns of enrolment in Ghana as well as in Accra in a first step. In a second step, we use data from the Women's Health Study of Accra wave II to evaluate the effect of insurance on health seeking behaviour using propensity score matching.
Results
We find that on average individuals enrolled in the insurance scheme are significantly more likely to obtain prescriptions, visit clinics and seek formal health care when sick.
Conclusion
These results suggest that the government's objective to increase access to the formal health care sector through health insurance has at least partially been achieved.
Keywords: Health insurance, health care utilisation, treatment differentials
Introduction
Health care financing in Ghana began with a tax-funded system that provided free public health care services to all after independence. As this system gradually became financially unsustainable with economic stagnation in the 1970s, initially low user fees were established for hospital services to discourage unnecessary use, locally recover some costs and generate provider performance incentives. Continued declines in government spending on health through the 1970s and 1980s led to shortages of medicines and supplies and deteriorating quality of care., 2 Following adoption of structural adjustment reforms in 1983, the Rawlings administration raised and expanded user fees for public health care services in a system that became known as “cash and carry.” The user fee system improved operating revenues for some facilities, but it was poorly regulated, inconsistently implemented, and found to have worsened access to care for the poor.–5
Starting from the early 1990s, Ghana began to seek other ways of financing health care, including NGO-initiated community-based health insurance schemes (CBHIS). While popular among members and international donors at the time, the schemes were only targeted to specific areas, failed to address key social insurance issues, and were not supported by general government revenue to allow them to cater for the poor. Most importantly, with CBHIS covering only about 1% of the population with limited benefit packages,6 the system of user fees remained the predominant means of paying for health care. The highly unpopular “cash and carry” system became a salient political issue and the main opposition party, the National Patriotic Party (NPP), began to call for its abolishment in its manifestos and campaigns—a promise that may have helped the NPP win the 2000 presidential and parliamentary elections.7 Ultimately, the National Health Insurance Scheme (NHIS) was established under Act 650 of 2003 by the Government of Ghana to provide a broad range of health care services to Ghanaians through district mutual and private health insurance schemes.8
Coming as a consequence of the deleterious effects of user fees, one of the primary goals of Ghana's NHIS was to increase affordability and utilisation of drugs and health services in general, and among the poor and most vulnerable populations in particular. Few studies have tried to investigate whether the NHIS achieved this objective. Witter and Garshong note that the number of outpatient visits per capita in Ghana increased sharply after 2005, the same year NHIS operations began. Mensah et al find that pregnant women enrolled in the NHIS are more likely to receive prenatal care, give birth in a hospital, and have skilled attendants present at birth. Similar results were found in a recent study tracking health seeking behaviour in two districts before and after NHIS rollout.11
In this paper, we use data collected as part of the Women's Health Study of Accra to evaluate the degree to which enrolment in health insurance is associated with increased utilisation of drugs and health services in the Accra Metropolitan Area.
We analyse six measures of health care use collected as part of the WHSA-II interview to investigate whether NHIS enrolment is associated with changes in health behaviour in general, and increased use of formal health care in particular. Our main hypothesis is that, conditional on several observable characteristics, being enrolled in NHIS should increase the probability of women seeking care. Women who fall ill without insurance are likely to self-treat themselves if possible in order to avoid incurring external treatment cost. With insurance, the cost of seeking professional treatment decreases, so that increased health care utilisation should be observed.
The Ghana National Health Insurance Scheme
Ghana's National Health Insurance Scheme (NHIS) was created by the National Health Insurance Act of August 2003, and is one of very few attempts by a sub-Saharan African country to implement a national-level, universal health insurance program. A newly-created National Health Insurance Authority (NHIA) was commissioned “to secure the implementation of a national health insurance policy that ensures access to basic healthcare services to all residents.”13 The NHIA licenses and regulates district-level mutual health insurance schemes (DMHISs) as well as other schemes allowed under the Act, accredits providers, determines—in consultation with DMHISs—premium levels, and generally oversees and reports on NHIS operations. There are currently 145 district schemes, including ten that operated in the Greater Accra area during the study period.14
The NHIS is financed from four main sources: a value-added tax on goods and services, an earmarked portion of social security taxes from formal sector workers, individual premiums, and miscellaneous other funds from investment returns, Parliament, or donors. The 2.5 per cent tax on goods and services, called the National Health Insurance Levy (NHIL), is by far the largest source, comprising about 70 per cent of revenues. Social security taxes account for an additional 23 per cent, premiums for about 5 per cent, and other funds for the remaining two per cent.15
The NHIS (including all DMHISs) has a single benefit package that is set by Legislative Instrument 1809 and described by the NHIA as covering “95% of disease conditions” that afflict Ghanaians., 16 The NHIS covers outpatient services, including diagnostic testing and operations such as hernia repair; most in-patient services, including specialist care, most surgeries, and hospital accommodation (general ward); oral health treatments; all maternity care services, including Caesarean deliveries; emergency care; and, finally, all drugs on the centrally-established NHIA Medicines List.14
The NHIS package excludes some very expensive procedures such as certain surgeries, cancer treatments (other than breast and cervical cancer), organ transplants, and dialysis; non-vital services such as cosmetic surgery; and some high profile items such as HIV antiretroviral drugs (which are heavily subsidized by the separate National AIDS Program). Other than the excluded services, there are few formal limits placed on NHIS members' consumption of benefits—there is no cost-sharing beyond premiums (i.e., no co-payments, coinsurance, or deductibles), no annual or lifetime limits and little effective gate-keeping. Benefits were intended to be “portable” from district to district, but actual portability has been mixed and is one reason for the recent introduction of a single, national NHIS identification card to replace district-level cards.
Act 650 technically requires all Ghanaians to enrol in the NHIS or in another health insurance plan. Specifically, Section 31 of Act 650 reads: “(1) A person resident in Ghana other than a member of the Armed Forces of Ghana and the Police Service shall belong to a health insurance scheme licensed under this Act. (2) A person resident in a district, who is not a member of a private health insurance scheme or any other district scheme registered under this Act, shall apply to be enrolled as a member of the district mutual health insurance scheme in the relevant district.” However, enrolment is de facto voluntary because there is no penalty for failing to enrol, and individuals or households are not automatically enrolled. Ghanaians generally must go in-person to a DMHIS office, complete registration paperwork (often after waiting a substantial amount of time), and pay a small registration fee meant to cover the photo ID and administrative expenses of registration. Even workers who contribute to the NHIS through social security contributions (SSNIT) have to personally enrol and pay registration fees in order to obtain the insurance cards. In addition to SSNIT contributors, broad swaths of the population are exempt from paying premiums (but not registration fees), including:
-
People over age 70;
-
Children under 18 whose parents both enrol;
-
The “core poor,” defined as being unemployed with no visible source of income, no fixed residence, and not living with someone employed and with a fixed residence16; and
-
Since July 2008, all pregnant women.
Ghanaians who are not exempt must pay an annual insurance premium in addition to the registration fees. The official NHIA guidelines call for a range of premiums to be charged according to a person's income or wealth, ranging from 7.2 GhC for the “very poor” to 48 GhC for the “very rich”. However, given that accurate income measures are not generally available, many DMHISs have moved to charging a constant premium to all, typically in the GhC 8–10 range. In the Accra Metropolitan Area analysed in this paper, the average premium reported by study respondents was GhC 21.
Official statistics on NHIS registration provided by the National Health Insurance Authority show the increase in enrolment since operations began in late 2005. As Table 1 shows, the total number of active members reportedly increased from 2.4 million in 2006 to 11.1 million in 2009, suggesting that close to 50% of the population was covered by the insurance by 2009. More recently, however, the NHIA changed its methodology for calculating active members and estimated in its 2010 annual report that about 34% of Ghanaians were active enrolees at the end of 2010.18
Table 1
Members of Ghana's NHIS and percentages of the population enrolled 2006–2009
Year | Estimated population of Ghana |
Total registered members |
Total active members |
Percentage of population registered |
Percentage of population active |
2006 | 21,876,031 | 3,867,862 | 2,422,097 | 18 | 11 |
2007 | 22,387,911 | 8,184,294 | 6,674,270 | 37 | 30 |
2008 | 22,876,031 | 12,518,560 | 9,969,846 | 55 | 44 |
2009 | 23,416,518 | 14,511,777 | 11,132,981 | 62 | 48 |
Note: Reproduced from NHIA, p.3.18
Theoretical background and empirical hypothesis
A large literature has investigated the behavioural effects of insurance in general, and the effects of health insurance on health seeking behaviour in particular. From a theoretical perspective, being covered by insurance can be expected to affect individual behaviour through several distinct mechanisms. First, individuals may feel safer with insurance, and thus take on more risk in the presence of insurance, a problem commonly referred to as the moral hazard problem in the economics literature.19 Second, insurance may change the choices individuals make once health problems arise. From the perspective of a rational decision maker, insurance makes formal treatment cheaper in comparison with other treatment modalities, and thus induces a shift towards increased service utilisation. Last, individuals may want to use the insurance to recover parts of the insurance premium already paid, a behaviour which is inconsistent with models of rational choice, but well explained in models of loss aversion as developed by Kahneman and Tversky.20
In the context of health, moral hazard is generally considered of lesser importance, since insurance covers the financial consequences of ill health, but offers only limited protection against the physical and emotional burden of ill health generated by risky behaviour. Changes in relative cost, on the other hand, appear to be of central importance for the rollout of health insurance in general, and in a developing country setting in particular. In the presence of user fees and transaction costs, poor populations may refrain from seeking professional health care from doctors or hospitals. Health insurance lowers this barrier to access, and is thus designed to increase use of professional medical care in general, and particularly among the poor who may be deterred by often steep user fees. This effect may be even stronger if the poor feel the need to recover some of the initial expenditure on insurance.
The hypothesis we wish to test in this paper is whether insurance affiliation is associated with increased utilisation of formal care. To do so, we look at six different health utilisation variables as described in detail below, and investigate the degree to which each of them is affected by insurance affiliation.
Data and Methods
Data for this analysis are from the second wave of the Women's Health Survey of Accra (WHSA II), conducted from September 2008 to June 2009. As described in other chapters of this volume, the WHSA-II is a prospective, community-based study intended primarily to measure the burden of disease among a representative sample of adult women (age 18 and over in 2003) in the Accra Metropolitan Area.
In order to investigate the effects of NHIS affiliation on health service utilisation, our analysis proceeds in two steps. In the first step, we briefly analyse the patterns of NHIS affiliation. Despite Act 650's intention of universal coverage, only a minority of women reported to be enrolled in the NHIS at the time of the interview (roughly three years after NHIS started operating). We investigate the relation between NHIS affiliation and socio-demographic characteristics in the first part of our analysis. In the second part, we analyse six variables describing health seeking behaviour. Five variables measure health care utilisation over the 12 months preceding the interview, including: (1) whether the woman received a general health examination, (2) whether the women received a breast examination, (3) whether the woman visited a health clinic or hospital, (4) whether the woman stayed overnight as a patient in a hospital, and (5) the number of drug prescriptions she obtained. The sixth variable is whether the woman's usual reference point when she is sick is in the formal or informal health sector. Formal sector options include: clinic/health centre, doctor's office, hospital emergency room, hospital outpatient department, and maternity home. Informal sector options include: pharmacist/chemical shop, self-medicate, church/pastor, spiritualist, or nowhere.
We first examine the unconditional health behaviour differential by simply comparing group averages of women currently enrolled in the NHIS and women not currently enrolled in the NHIS. In the second step, we use propensity score estimation to remove bias generated through differences in observable variables, and to identify conditional differences attributable to NHIS affiliation. All analysis was conducted using Stata© 10.
![Ministry Ministry](/uploads/1/2/7/6/127625823/192490503.png)
Results
Demographic characteristics
The 2008 WHSA data was based on household interviews with 2,814 women aged 18 and over in 2003. We restricted our analysis to those women where all key variables are available, which leaves us with a sample of 2,543 women. The average woman in our sample was 41.8 years old at the time of the interview, consistent with the average age of 35 among women 18 and older in 2003 when the original enrolment of women was conducted.
About fifty-five per cent of women (55.6%) are currently married, with an average number of 3.6 pregnancies. A little over 70 per cent of women (72%) are working. The majority of women in the sample have attained some secondary schooling, while 15 per cent report no schooling and 12.3 per cent report tertiary schooling. Fewer than forty per cent (38%) of respondents are Ga, approximately 35 per cent belong to Asante, Fante or other Akan ethnicities, and about 14 per cent are Ewe. Less than half of women (48%) consider their health average, 13 per cent consider their health below average, and 39 per cent consider their health to be good or excellent (Table 2).
Table 2
Variable | N | Mean | Std. dev |
Age | . | 41.791 | 14.877 |
Married | 1413 | 0.556 | 0.497 |
Pregnancies | . | 3.601 | 2.718 |
Working | 1834 | 0.721 | 0.448 |
No education | 387 | 0.152 | 0.359 |
Primary | 287 | 0.113 | 0.316 |
JSS | 1030 | 0.405 | 0.491 |
SSS | 519 | 0.204 | 0.403 |
Higher | 313 | 0.123 | 0.329 |
Qur'anic | 7 | 0.003 | 0.052 |
Asante | 165 | 0.065 | 0.246 |
Akwapim | 143 | 0.056 | 0.231 |
Fante | 316 | 0.124 | 0.330 |
Other Akan | 231 | 0.091 | 0.287 |
Ga | 968 | 0.381 | 0.486 |
Ewe | 367 | 0.144 | 0.352 |
Excellent health | 311 | 0.122 | 0.328 |
Good health | 680 | 0.267 | 0.443 |
Average health | 1226 | 0.482 | 0.500 |
Fair health | 284 | 0.112 | 0.315 |
Poor health | 43 | 0.017 | 0.129 |
Notes: Based on 2,543 observations with complete information. All descriptive statistics are based on survey-specific sampling weights.
Patterns of NHIS enrolment
Figures 1–3 show NHIS enrolment patterns across educational attainment, self-assessed health status, and age groups. While there is some education gradient in enrolment, the likelihood of NHIS enrolment is nearly equal across the main four attainment groups (i.e. up to senior secondary), and only slightly higher for women with tertiary education. As the confidence intervals in Figure 1 show, the null hypothesis that enrolment is equal across educational groups cannot be rejected at the 95% level.
NHIS enrolment by educational attainment
NHIS enrolment by age group
NHIS enrolment by self-assessed health
Even more surprising is the weak relation between enrolment in health insurance and self-assessed health, with women who assess their own health as “poor” only marginally more likely to be enrolled in the NHIS than women who rate their own health as “excellent”. Enrolment is highest among the “good” and “fair” health groups. From the three variables analysed, age appears to be by far the best predictor of NHIS enrolment. While less than 25 per cent of women under the age of 30 are enrolled in the NHIS, the same is true for more than 45 per cent of women over the age of 60.
Health care utilisation variables
We analyse six variables capturing respondents' health seeking behaviour. The first two items we analyse capture preventive exams: 15.4 per cent of women indicated having had a general health exam over the last 12 months, and 8 per cent of women indicated that they had had breast examinations (Table 3). The remaining four items capture typical health seeking behaviour as well as the individual experience over the last year: when asked about where they usually go when they get sick, 79.7 per cent of women indicated that they seek care at a “formal” facility (clinic, doctor, hospital, maternity home). When asked whether they had visited a health clinic/centre or hospital over the past 12 months, 57.4 per cent of women stated this to be the case; a remarkably high 9.9 per cent of women reported to have been hospitalized overnight over the previous 12 months. Last, women were also asked whether they had been prescribed medicines by a doctor, such as medication for cholesterol, de-worming pills, antibiotics, vitamins, weight loss or weight gain supplements, malaria treatment, pain killers, birth control items, hormone replacement therapy, or other health treatments. About a fifth (21 per cent) of women received prescriptions from a doctor, and the average number of prescriptions was 0.31 (Table 3).
Table 3
Variable | Number of cases | Mean | Std. dev | Min | Max |
Had a general exam last year | 392 | 0.154 | 0.361 | 0 | 1 |
Had a breast exam last year | 205 | 0.08 | 0.272 | 0 | 1 |
Seeks formal care when sick | 2028 | 0.797 | 0.402 | 0 | 1 |
Visited clinic/hospital last year | 1461 | 0.574 | 0.495 | 0 | 1 |
Hospitalized overnight last year | 252 | 0.099 | 0.299 | 0 | 1 |
Number of prescriptions last year | 784 | 0.308 | 0.698 | 0 | 6 |
Notes: Based on 2,543 observations with complete information. Reported means and standard deviations represent sampling weight adjusted averages. The mean for the first five variables represents the proportion of respondents who reported having/seeking the given health care variable.
Health care utilisation by NHIS enrolment
Table 4 shows the six health care seeking variables for non-enrolled and enrolled women, respectively. The mean value of all six variables is higher in the NHIS-enrolled group, and the difference between the two groups is significant at the 99 per cent confidence level in all variables except for the breast exam, where the absolute likelihood is below 10 per cent for both groups. The difference between the two groups is largest for the clinic visits and prescriptions: 76.3 per cent of women enrolled in NHIS visited a health clinic or hospital during the twelve months preceding the interview, compared to just 50.2 per cent of women without insurance. The relative differences appear to be particularly large for hospitalization and prescriptions. While only 7.4 per cent of women not enrolled in the NHIS report a hospitalization during the 12 months preceding the interview, the same was true for 12.7 per cent of women enrolled in NHIS. Similarly, women with NHIS insurance report on average 0.576 prescriptions, which is more than twice as many as women who were not enrolled at the time of the interview (0.274 prescriptions on average).
Table 4
Not Enrolled | Enrolled | Difference | p-value | |
Had a general exam last year | 14.7% | 20.7% | 6.0% | 0.000 |
Had a breast exam last year | 7.5% | 9.3% | 1.8% | 0.115 |
Seeks formal care when sick | 75.5% | 90.5% | 15.0% | 0.000 |
Visited clinic/hospital last year | 50.2% | 76.3% | 26.1% | 0.000 |
Hospitalized overnight last year | 7.4% | 12.7% | 5.3% | 0.000 |
Number of prescriptions last year | 0.274 | 0.576 | 0.302 | 0.000 |
Notes: Based on a sample of 2543 women. P-values based on a simple two-sided t-test.
Given that women with NHIS enrolment on average appear to be more educated and older than women without health insurance, the increased use of health services among insured women reported in Table 4 may reflect differences in respondent characteristics rather than true differences generated by health insurance. In order to reduce the risk of confounding, we use propensity score matching to compare enrolled women to non-enrolled women conditional on all observable characteristics. The matching algorithm we use takes into account all socio-economic characteristics reported in Table 2, which means that the estimated differences reflect observed differences keeping age, marital status, number of pregnancies, work status, ethnicity, education, wealth quintile, and self-assessed health the same.
The results of the propensity score matching estimation are displayed in Table 5. The elimination of observable differences attenuates the observed differentials a bit, but does not substantially change the picture obtained from Table 4.
Table 5
Group Assignment |
Not enrolled Matched control |
Enrolled Treatment |
Differences | ||
Absolute | Relative | p-value | |||
Had a general exam last year | 18.7% | 20.7% | 2.0% | 10.6% | 0.238 |
Had a breast exam last year | 7.9% | 9.3% | 1.4% | 18.0% | 0.242 |
Seeks formal care when sick | 78.6% | 90.5% | 11.9% | 15.1% | 0.000 |
Visited clinic/hospital last year | 54.5% | 76.3% | 21.8% | 40.0% | 0.000 |
Hospitalized overnight last year | 6.9% | 12.7% | 5.8% | 83.1% | 0.000 |
Number of prescriptions last year | 0.366 | 0.576 | 0.210 | 57.5% | 0.000 |
Notes: Based on a sample of 2,543 women. All estimation results control for age, marital status, number of pregnancies, employment status, education category, wealth quintile, ethnicity and self-rated health.
Conditional on all observable characteristics, women enrolled in the NHIS appear to be substantially more likely to seek formal care and to go to a clinic. Keeping all other factors constant, NHIS-enrolled women are 40 per cent more likely to have attended a clinic over the past year, and they have about 57 per cent more prescriptions. Most remarkably, NHIS-enrolled women appear nearly twice as likely (83 per cent more likely) to have stayed overnight at a hospital than non-enrolled women.
Discussion
In this paper, we analysed data from wave II of the Women's Health Study of Accra to investigate the association between NHIS insurance enrolment and health seeking behaviour among women in the Accra Metropolitan Area. We demonstrated that on average women enrolled in the insurance scheme are more likely to seek formal care when sick, have a larger number of prescriptions, are more likely to have visited a clinic or hospital in the year prior to the interview, and are substantially more likely to have experienced an overnight stay at a hospital in the last year. These differences are partially explained by the fact that women enrolled in insurance are on average older, more educated, and in slightly poorer health. However, even when all observable characteristics are controlled for, insurance affiliation appears to be positively associated with health care usage in this sample.
The notion that insurance increases usage is consistent with our theoretical priors as well as a few previous attempts to quantify the effects of the NHIS on access to health care., However, while the empirical associations between NHIS enrolment and health service usage appear strong, the cross-sectional nature of our analysis does not allow us to directly interpret the observed differences as causal effects of the insurance itself. Given that NHIS enrolment is voluntary, women enrolled may differ from those not-enrolled with respect to their health needs or health seeking preferences. Even though we use a large number of control variables to adjust for individual differences, we cannot rule out that some of the observed differences in behaviour may reflect differences in unobservable characteristics rather than a true causal effect of insurance on health seeking behaviour. Given the consistently positive effects found across studies and the substantial size of the effects for outpatient visits, prescriptions, and hospitalizations, it appears unlikely, however, that such selection effects would entirely account for the associations found here. Still, further research will be needed to more closely identify the true causal impact of the NHIS on health seeking behaviour.
Another limitation of the current study is that the results are not necessarily generalizable outside of the study population. The Accra Metropolitan Area is likely to differ from other regions with respect to the number of potential treatment locations as well as the locally available health resources, so that populations outside of Accra may be differentially affected by NHIS enrolment. On the one hand, one could argue that a larger absolute number of facilities and medical professionals locally available should make it easier for individuals to take advantage of the benefits provided by the NHIS. On the other hand, one could argue that the relatively high incomes of Accra make the local population particularly sensitive to the time cost associated with using accredited facilities, so that the insurance impact may be smaller than in other locations. Similarly, children and men's use of health care may respond differently to insurance than did the adult women studied here.
Beyond the question of causal effect size, the limited reach of the NHIS to date has raised concerns regarding equity and the sustainability of the scheme. 21 The main equity concern is that wealthier Ghanaians are currently better able and more likely to enrol than poorer Ghanaians, despite premiums that were supposed to vary by income, but in practice are fixed at one level for all in general, and premium exemptions for the “core poor”.
Sustainability concerns stem from self-selection of less healthy individuals into the scheme, which tends to raise average costs as long as coverage remains less than universal; and from increasing utilization that must be paid for from the mostly tax-based financing sources, which may grow more slowly than enrolment and consequent use grow.
Our results served to underline both of these concerns—enrolees were older and therefore very likely costlier to insure than average; and enrolment did appear to increase utilization substantially as originally intended. Despite limitations of generalizability of our results, it is relatively clear from this and prior studies that adjustments to the NHIS will be needed over the coming years to reach comprehensive and continued health care coverage for the entire population in the medium to long run.
Conclusion
This study analysed cross-sectional data and employed propensity score matching to estimate the effect that NHIS enrolment had on six measures of health care use among a sample of urban, adult Ghanaian women. The overall results suggested that the NHIS's goal of improving access to health care has been achieved at least among adult women living in the Accra Metropolitan Area.
References
1. Agyepong IA, Adjei S. Public social policy development and implementation: A case study of the Ghana national health insurance scheme. Health Policy Plan. 2008;23:150–160. [PubMed] [Google Scholar]
2. Assensoh AB, Wahab H. A historical-cumpolitical overview of Ghana's national health insurance law. African & Asian Studies. 2008;7(2/3):289–306.[Google Scholar]
3. Nyonator F, Kutzin J. Health for some? The effects of user fees in the Volta region of Ghana. Health Policy Plan. 1999;14(4):329. [PubMed] [Google Scholar]
4. Waddington CJ, Enyimayew K. A price to pay: The impact of user charges in Ashanti-Akim district, Ghana. Int J Health Plann Manage. 1989;4(1):17–47.[Google Scholar]
5. Waddington C, Enyimayew K. A price to pay, part 2: The impact of user charges in the Volta region of Ghana. Int J Health Plann Manage. 1990;5(4):287–312.[Google Scholar]
6. Atim C, Grey S, Apoya P, Anie SJ, Aikins M. A survey of health financing schemes in Ghana. Bethesda, Maryland: Partners for Health Reformplus, Abt Associates; 2001. [Google Scholar]
7. Rajkotia Y. National health insurance in Ghana: Politics, adverse selection, and the use of child health services [PhD] Baltimore: Johns Hopkins University; 2009. [Google Scholar]
8. Ghana National Health Insurance Act (Act 650) 2003[Google Scholar]
9. Witter S, Garshong B. Something old or something new? Social health insurance in Ghana. BMC Int Health Hum Rights. 2009;9:20.[PMC free article] [PubMed] [Google Scholar]
10. Mensah J, Oppong JR, Schmidt CM. Ghana's National Health Insurance Scheme in the context of the health MDGs: An empirical evaluation using propensity score matching. Health Econ. 2010;19(S1):95–106. [PubMed] [Google Scholar]
11. Health Systems 20/20 Project and Research and Development Division of the Ghana Health Service, author. An evaluation of the effects of the National Health Insurance Scheme in Ghana. Bethesda, Maryland: Health Systems 20/20 project, Abt Associates Inc; 2009. [Google Scholar]
12. Kirigia JM, Preker A, Carrin G, Mwikisa C, Diarra-Name AJ. An overview of health financing patterns and the way forward in the WHO African region. East Afr Med J. 2006;83(9):S1–S28. [PubMed] [Google Scholar]
13. National Health Insurance Act, Act 650. 2003 Aug[Google Scholar]
14. NHIA, author. NHIS website. Available at: http://www.nhis.gov.gh/
15. Yankah B. Financial sustainability of NHIS based on recent financial assessment of the scheme; Presentation at Health Summit; 2009 November 16–20. [Google Scholar]
16. National Health Insurance Regulations, LI 1809. 2004[Google Scholar]
17. Sarpong N, Loag W, Fobil J, et al. National health insurance coverage and socio-economic status in a rural district of Ghana. Trop Med Int Health. 2010;15(2):191–197. [PubMed] [Google Scholar]
18. NHIA, author. National Health Insurance Scheme annual report. Accra, Ghana: National Health Insurance Authority; 2010. [Google Scholar]
19. Arrow K. Uncertainty and the welfare economics of medical care. Am Econ Rev. 1963;53(5):941–973.[Google Scholar]
20. Kahneman D, Tversky A. Prospect theory: An analysis of decision under risk. Econometrica. 1979:263–291.[Google Scholar]
21. Apoya P, Marriott A. Achieving a shared goal: Free universal health care in Ghana. Oxford: Oxfam International; 2011. [Google Scholar]
22. Jehu-Appiah C, Aryeetey G, Spaan E, de Hoop T, Agyepong I, Baltussen R. Equity aspects of the National Health Insurance Scheme in Ghana: Who is enrolling, who is not and why? Soc Sci Med. 2011;72(2):157–165. [PubMed] [Google Scholar]
Articles from Ghana Medical Journal are provided here courtesy of Ghana Medical Association
Published online 2014 Sep 3.
PMID: 25191147
Focus: Global Health and Development
This article has been cited by other articles in PMC.
Abstract
Ghana is a developing country in West Africa with a population of about 25 million. Medical illnesses in Ghana overlap with those in developed countries, but infection, trauma, and women’s health problems are much more prominent. Medical practice in rural Africa faces extremely limited resources, a multiplicity of languages (hundreds in Ghana), and presentation of severe illnesses at later stages than seen elsewhere. Despite these limitations, Ghana has established a relatively successful national medical insurance system, and the quality of medical practice is high, at least where it is available. Ghana also has a well-established and sophisticated administrative structure for the supervision of medical education and accreditation, but it has proven very difficult to extend medical training to rural areas, where health care facilities are particularly short of personnel. Physicians are sorely needed in rural areas, but there are few because of the working conditions and financial limitations. Hospital wards and clinics are crowded; time per patient is limited. This article details some of the differences between medical practice in Ghana and that in wealthier countries and how it functions with very limited resources. It also introduces the medical education and training system in Ghana. The following article describes an attempt to establish and maintain a residency training program in General Medicine in a rural area of Ghana.
Keywords: West Africa, Ghana, tropical medicine, medical education
Introduction
Ghana is a developing country in West Africa with a population of about 25 million. Although economically poor by Western standards, it has a proud history, including as the first sub-Saharan state to regain independence from European colonial powers (in 1957), universal education through primary school, and a relatively high literacy rate for developing countries [1]. Life expectancy at birth is 60 years [2]. Well-endowed with natural resources (including gold, timber, and cocoa), Ghana has a much higher per capita economic output than most West African countries [3] but remains partially dependent on international financial and technical assistance.
Ghana’s medical system is centered heavily in the two largest cities: Accra, the capital on the coast with about 2 million inhabitants, and Kumasi, with about 1 million citizens in the center of the country. Nevertheless, Ghana needs physicians and medical facilities in smaller towns where most of the population lives, but as elsewhere, it is hard to retain physicians in remote rural areas. Twenty-five years ago, most physicians at mid-sized hospitals in rural Ghana were Europeans on tours of a few years, but this is no longer common. Ghanaian and European authorities have decided to end dependence on visitors and shift resources to programs that can be sustained over a longer period by Ghanaians.
The article following this one describes an attempt to address the scarcity of physicians in rural Ghana by development of a General Medicine residency program in Techiman, a small city in central Ghana. Here, we detail three aspects of practicing medicine in Ghana pertinent to a young physician in training: the different range of illnesses found in West Africa, especially outside the cities; the substantial financial limitations on the practice of medicine in Ghana and attempts to ameliorate them; and the very well-developed organization and administration of medical education and training in Ghana.
The Illnesses
While the practice of General Medicine is similar in principle to that in wealthier countries and the total range of illnesses is similar, some illnesses are seen in far greater measure than in developed countries. They include infections, particularly those common to tropical countries, trauma, and women’s health problems such as complications of pregnancy (specific examples of all these problems are detailed in the following article on the development of a particular General Medicine residency program in rural Ghana).
Infections
Infections are a primary reason for hospital visits and admissions. Malaria is rampant, and its burden is tremendous []. Most malaria is treated easily with 3 days of pills distributed by the government [,], but many people acquire malaria repeatedly, and if untreated, it can be fatal. Typhus infection is common and is a major cause of intestinal perforation and urgent surgery. There are frequent, fulminant cases of tuberculosis, as well as filiariasis and schistosomiasis. Tuberculosis and HIV infection are common (the latter infecting 1.8 percent of the population [7] ― much less than in Southern Africa), and about half of the patients in Ghana who die with HIV do so from consequences of tuberculosis [].
Trauma
Trauma comes from the same sources as in other countries, including home and work-related accidents, and violence, domestic and other. Much more trauma, however, comes from motor vehicle accidents, which are far more common in most developing countries than in the developed world [9]. Ghana is no exception. The cause is a combination of poor road conditions, vehicle malfunction, and, mostly, less well-trained drivers with dangerous habits. Rules of the road are not well adhered to or enforced. Motor vehicle accidents generate the large majority of severe trauma cases in Ghana.
Women's Health
A disturbingly large percentage of emergency and critically ill patients are young women who are pregnant — and many from complications of botched abortions. Safe abortions cost approximately $25 in some private clinics, but this is prohibitive for many young women and their families. Various herbal concoctions may or may not produce miscarriages, but they also cause fulminant infections or even sepsis and subsequent scarring, obstructions, and child-bearing trauma. Early life mutilation produces similar anatomic barriers to healthy deliveries. Toxemia and eclampsia are common. Maternal mortality is high in West Africa, approximately 560/100,000 births in Ghana [10], but Ghana’s is one of the best rates in West Africa. Infant mortality is 61/1000 live births [11], but reliable statistics are difficult to determine and just beginning to be gathered, in part because most births occur at home.
Other Illnesses
Diabetes may be less common than it is in the overweight population of the United States, but it is almost certainly under-diagnosed in Ghana. It also takes a terrible toll when not attended to appropriately, leading to the spread of devastating infections (including perinatal) and gangrene of the limbs after minor cuts. A patient may wait for a few weeks at home before going to a hospital with a nearly necrotic limb. Often, it is too late, and many septic patients die.
Heart disease and cancer, so prominent in the developed world, certainly exist in West Africa, but they are not the major focus of the medical system. The acute presentation of coronary disease is relatively rare in rural hospitals, but the disease appears to be increasing rapidly and is being recognized much more nationwide. Asthma exists but is much less common than infectious illnesses of the lung, especially tuberculosis. Cancers tend to present late, often with gigantic masses, leaving little to offer.
Sickle cell disease causes some cases of severe anemia, including in very young children. In rural Ghanaian medical facilities, a hemoglobin of 5 to 6 (hematocrit 15 to 18) is considered a non-urgent, outpatient evaluation. Osteomyelitis is a very frequent complication of sickle cell disease, often requiring surgery to remove infected bone (a “sequestrectomy”), with long courses of antibiotics, or leading to crippling bone deformities to the point where it is common to see hobbled, indigent people on the streets of towns of any size.
Medical Technology
While equipment and other facilities are generally modern in the university hospitals, this is not the case in most small towns. Nationwide, there are relatively few ventilators, except in operating rooms and in the thoracic surgery ICU at the university in Accra, with smaller units at large teaching and military hospitals. Cardiopulmonary resuscitation is seldom attempted outside major urban centers. Neither patients with coma from severe head injury nor those with severe respiratory diseases are intubated. There are hemodialysis units at only the two primary teaching hospitals in the large cities. It can be difficult to get a cardiogram done, sometimes even in the teaching hospital in Kumasi. With the development of a small but growing middle class, especially in the cities, some private clinics are springing up, and some have sophisticated equipment.
Specialization
Specialist physicians are relatively few in Ghana as most acquire their formal specialty training abroad, and most stay there. A few years ago, three of Ghana’s five urologists were killed in a single car crash after a medical meeting.
Neurology, the field of one of the authors (AA), for example, does not have its own section in the universities. There are only a couple of neurologists in the country, and none in the Ashanti or Brong-Ahafo regions. There is one clinically active neurologist in Ghana, at the Korle Bu Teaching Hospital in Accra, who runs the only EEG and EMG laboratory. The two main teaching hospitals have CT scanners, as do a few medium-sized cities and private clinics. There are just four or five neurosurgeons in the country, so most intracranial hemorrhages take their courses without assisted ventilation or surgical drainage. The first MRI scanner was installed at Korle Bu in 2006. Still, throughout Ghana, lumbar radiculopathies from disks, peripheral neuropathies, seizures, and strokes are common problems. Spinal cord trauma is common, and some neurologic function improves with immobilization and traction; decompression surgery is generally not available. Spastic paraparesis and Guillain-Barré syndrome are not rare, and cerebral malaria is commonplace.
Finances
The Ghana Health Service consists of government (including university) hospitals and clinics, hospitals and clinics within the Christian Health Association of Ghana (CHAG), and private facilities. The Ghana government pays most of the salaries of the staff physicians, house officers, nurses, and some others in the government and CHAG hospitals. For other funding, the CHAG hospitals must be self-sufficient, and much of the operating expenses depend on income from patient charges. For special projects, expansion, and development of new projects, most CHAG hospitals are largely dependent on foreign assistance and non-governmental organizations.
The Ghana Ministry of Health has introduced a national universal medical insurance system, the National Health Insurance Scheme (NHIS), which pays for hospitalizations and outpatient doctor visits, as well as basic laboratory testing and certain medications []. All hospitals bill the NHIS monthly for the patients who have joined the program.
The NHIS, however, does not cover HIV medications, thoracic, neuro- or plastic surgery (except after trauma), other elective surgery, infertility evaluations, transplant medication or surgery, and many other expensive items, such as hemodialysis. Among cancer treatments, only those for cervical and breast cancer are included. Nevertheless, payment for physician visits and long-established medications is of tremendous benefit.
Medications are a major expense, as elsewhere []. In Ghana, a patient or patient’s family typically buys medications individually for the patient, including for hospitalized patients. A week’s outpatient amoxicillin might cost just $1, but for a hospitalized patient, intravenous ceftriaxone (often necessary because of antibiotic resistance) is about $6 a day. For perspective, GDP/person in Ghana is $1,605 per year (2012), compared to approximately $51,700 in the United States [13]. Families without insurance may decide that they cannot afford certain antibiotics, and sometimes they are stopped in mid-course for lack of funds.
The NHIS has a premium of about $10 a year, but ironically, the poorest and sickest people may be least likely to sign up []. The plan will take years to implement fully, in part because of its complexity and the lack of patient education (not completely unlike some U.S. Medicare benefits and some stipulations of the Affordable Care Act), but by 2010, over half the population of Ghana was enrolled.
Medical Education
Even more than physicians and medical facilities in general, Ghana’s medical education is heavily concentrated in the cities, where the two university medical centers employ about 100 physicians each. Korle Bu is a 1,500 bed teaching hospital connected to the University of Ghana in Accra. That university and a second in Kumasi produce about 200 graduate physicians each year, while a third, new university at Tamale in the North graduates a few dozen more. Still, the number of physicians in Ghana has held steady at about 1,400 for many years because most physicians emigrate [,]. Ghana is working hard to educate and retain locally trained physicians, but it is difficult to get them to work in rural settings, where the need is greatest.
As part of the trend toward local, i.e., Ghanaian, training, and self-sufficiency, physician training programs have been organized under the aegis of the West African College of Physicians (WACP) and the West African College of Surgeons (WACS), founded in 1976, and the Ghana College of Physicians and Surgeons (GCPS), founded by an act of the Ghanaian parliament in 2002. The Ghana Postgraduate Medical College (GPMC), founded in 2003, promotes the development of residency training programs. In coordination with the Ghana Medical and Dental Council, the GPMC also promotes (more junior) house officer training, including, more recently, outside of the cities.
The GCPS was founded “to promote specialist education in Medicine, Surgery, and related disciplines.” It accredits training centers and supervises the medical education of about 300 residents currently in training [18]. The WACS accredits surgery training programs, runs postgraduate courses, and determines eligibility for and sponsors a board examination. Fellowships for these specialties are sponsored by the WACP or GCPS, but it is at that point that many trainees choose to leave Ghana, most never to return [,].
A recent graduate (author AA) estimates that at least 70 percent of his classmates have left Ghana for the United Kingdom, United States, Canada, Australia, and the Gulf states, almost all to remain abroad permanently. Often, their parents have spent substantial portions of their income on earlier school fees and consider this an investment. Beyond the loss of trained personnel to the country, some of the emigrants send money home to their families, and some do not. Better income and the chance for more specialty training are the major motivations [].
Conclusion
Despite facing many limitations, Ghana has made tremendous strides in promoting medical care in a moderately large and very complicated country. It is a leader in West Africa and beyond. The practice of medicine and the illnesses faced are similar in many ways to those in developed countries, but with infections particularly common in the tropics, more trauma, and more extensive problems in women’s health as the most prominent burdens. As with other developing countries, however, Ghana has great difficulty extending modern medical care to smaller towns and rural areas. Its universities and medical training programs are heavily concentrated in the larger urban centers, where most of the physicians stay ― unless they, as most, emigrate for better jobs and for other reasons. Still, Ghana has many government-sponsored health programs and even a reasonably successful national health insurance system, as well as a well-developed and sophisticated administration of medical education, now focusing more on rural areas. It has begun to bring General Medicine, and even specialty training, to smaller communities, in part with the hope that physicians will stay in rural areas, where there are many sick patients to treat and learn from. The following article describes in detail the establishment and functioning of a General Medicine training program in rural Ghana.
Abbreviations
CHAG | Christian Health Association of Ghana |
NHIS | National Health Insurance Scheme |
WACP | West African College of Physicians |
WACS | West African College of Surgeons |
GCPS | Ghana College of Physicians and Surgeons |
GPMC | Ghana Postgraduate Medical College |
HIV | human immunodeficiency virus |
ICU | intensive care unit |
GDP | gross domestic product |
CT | computed tomography |
EEG | electroencephalogram |
EMG | electromyogram |
MRI | magnetic resonance imaging |
Notes
Author contributions
Dr. Drislane wrote most of the paper based on visits to the University of Ghana and the General Medicine Training Program at Techiman, with the perspective of a residency program director in the United States. Dr. Akpalu is a neurologist and MD graduate of the University of Ghana and provided much of the information about recent Ghanaian physician graduates, medical practice at Korle Bu Teaching Hospital at the University of Ghana, and the administrative structure of medical education and accreditation in Ghana. Dr. Wegdam is a Dutch trauma surgeon who worked in Ghana for more than 12 years, co-founded the residency in Techiman, and provided information about how it developed and functions.
References
- United Nations Educational, Scientific and Cultural Organization Institute for Statistics. Adult and youth literacy. April 2007 Assessment
- United Nations, Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2006 Revision, Highlights, Working Paper No. ESA/P/WP.202. 2007
- International Monetary Fund. World Economic Outlook Database — September 2010
- Bawah AA, Binka FN. How many years of life could be saved if malaria were eliminated from a hyperendemic area of northern Ghana? Am J Trop Med Hyg. 2007;77(6 Suppl):145–152. [PubMed] [Google Scholar]
- Yeboah-Antwi K, Gyapong JO, Asare IK, Barnish G, Evans DB, Adjei S. Impact of prepackaging antimalarial drugs on cost to patients and compliance with treatment. Bull World Health Organ. 2001;79:394–399.[PMC free article] [PubMed] [Google Scholar]
- Biritwum RB. Impact of health care financing reforms on the management of malaria in Ghana. East Afr Med J. 2001;78:636–640. [PubMed] [Google Scholar]
- Country Comparison: HIV/AIDS ― Adult Prevalence Rate. The World Factbook. Central Intelligence Agency [Internet] 2010. Available from:https://www.cia.gov/library/publications/the-world-factbook/rankorder/2155rank.html .
- Harries AD, Hargreaves NJ, Chimiziz R, Salaniponi FM. Highly active antiretroviral therapy and tuberculosis control on Africa: synergies and potential. Bull World Health Organ. 2002;80:464–469.[PMC free article] [PubMed] [Google Scholar]
- Global status report on road safety: time for action. Geneva: World Health Organization; 2007. [Google Scholar]
- Maternal mortality in 2005. Estimates developed by WHO, UNICEF, UN FPA and the World Bank. Geneva: World Health Organization; 2007. [Google Scholar]
- United Nations, Department of Economic and Social Affairs. United Nations World Population Prospects, 2011 revision
- Biritwum RB. The cost of sustaining the Ghana’s “Cash and Carry” system of health care financing at a rural health centre. West Afr J Med. 1994;13:124–127. [PubMed] [Google Scholar]
- GDP per capita (US$) The World Bank [Internet] Available from:http://data.worldbank.org/indicator/NY.GDP.PCAP.CD .
- Mensah J, Oppong JR, Schmidt CM. Ghana’s National Health Insurance Scheme in the context of the health MDGs: an empirical evaluation using propensity score matching. Health Econ. 2010;19 Suppl:95–106. [PubMed] [Google Scholar]
- Jehu-Appiah C, Aryeetey G, Spaan E, de Hoop T, Agyepong I, Baltussen R. Equity aspects of the National Health Insurance Scheme in Ghana: who is enrolling, who is not and why? Soc Sci Med. 2011;72:157–165. [PubMed] [Google Scholar]
- Mullan F. The metrics of the physician brain drain. N Engl J Med. 2005;353(17):1810–1818. [PubMed] [Google Scholar]
- Hagopian A, Ofusu A, Fatusi A, Biritwum R, Essel A, Gary Hart L. et al. The flight of physicians from West Africa: views of African physicians and implications for policy. Soc Sci Med. 2005;61:1750–1760. [PubMed] [Google Scholar]
- Residency programme. Ghana College of Physicians and Surgeons [Internet] Available from:www.ghcps.org .
- Ojo KO. International migration of health manpower in sub-Saharan Africa. Soc Sci Med. 1990;31:631–637. [PubMed] [Google Scholar]
Articles from The Yale Journal of Biology and Medicine are provided here courtesy of Yale Journal of Biology and Medicine
![](/uploads/1/2/7/6/127625823/240666636.jpg)