J. of Health and Development, Vol.1, Nos. 2+3, April + September, 2005.
I. Introduction
II. A sorry diagnosis?
III. Fee for service-financed health care
IV. Medicines: How much of a culprit?
V. A basket of potential solutions?
VI. ‘Righting the wrong’ in the obsolete geographic allocation of funding for health
VII. And the winner is…Community-based health insurance (CBHI)
The world cannot remain half healthy and half sick and still maintain its economic, moral and spiritual equilibrium. (World Health Assembly President, T. Scheel, 1951).
I. Introduction
In covering this topic, several words of caution are called for upfront:
1. Ultimately, the name of the game is NOT to become efficient at giving health care to the poor, BUT to reduce poverty!…and NOT much is happening in this front yet these days. We have declared war against poverty and poverty has won! [Certainly, if poverty was an infectious disease, the rich would do much more to eradicate it…].
2. In health care financing (HCF), the task is NOT about adapting services to poverty; that would be merely coping – not resolving. Moreover, just increasing the health care delivery system’s efficiency will not lead to increased access as much as very rapid economic growth will NOT automatically result in improved health! Decreasing the level of care to make it cheaper – e.g., ‘basic health packages’ – is NOT an option either.
3. For the years to come, local resources in rural areas of LDCs can NEVER finance 100% of the needed health care.
4. Cost-effectiveness analyses do NOT have an equity dimension and, therefore, must be adjusted if they are to be useful at all for achieving equity-oriented health targets.
5. The key issues to center the discussion on HCF are:
– How to finance the health care system;
– How to best allocate available resources;
– How to increase the efficiency AND the equity effect of the allocative actions taken.
6. Finally, if we are to succeed, sentences starting with “The Government should” are out! Stop complaining: DO!
II. A sorry diagnosis?
7. A quick look at the Third World countries shows that, still today, income maldistribution is very much with us: Comparing the top 20% income group with the bottom 20% (1995) shows that in China the gap was 7 times higher, in Thailand it was 9x; in Malaysia 11x; in The Philippines 7x, and in Vietnam 5x. (These ratios may have grown since).
8. The numbers show that rich countries use around 7% of their income on health yet poor countries are using around 10-11% (around 25% of the income of the poorest quintile families). In 2003, the average spending on health care in all developing countries worldwide was US$11/capita. (It is deemed this amount must be increased to $60).
[In Vietnam. poor households are already allocating around 13% of their expenditures to all forms of health care, with the most affluent only spending around 9%].
9. Households are the main source financing health care services altogether (close to 80% in many countries in Asia) and, through open and hidden fees, also finance up to 60% of the public sector already.
[In Vietnam, the national health insurance finances 3%; foreign aid 2% and the Government 15%].
10. Government allocations to the health sector have been declining in real per capita terms in many developing countries. Increasing government financing for health has historically often been below population growth and inflation rates. It has been easy for the central government to set targets and then…not to pay.
11. Against this background, the disease burden of the poorest quintile is roughly threefold that of the richest, their life expectancy is around 10 years shorter, and their infant mortality rate is twice as high.
12. Close to one third of the patients who need hospitalization fail to be hospitalized: around 80% due to financial constraints. There is also evidence that poor patients abscond due to lack of money to pay hospital bills.
13. In many countries, unofficial payments to (mostly) doctors are estimated to be about 15 times higher than official payments for hospital services. People wrongly think these (“envelope”) payments – formal as well as informal – are necessary for receiving proper treatment. [We need to challenge this assumption that one can only receive good care if one is able to pay for it].
14. But keep in mind that health care seeking contacts with private providers are not primarily with doctors, but overwhelmingly with drug sellers (…close to 60% of the people self-medicate!!)
15. Ergo, the poor face cost-related barriers to accessing health care. Medical expenditures are now one of the main causes of why people become poor (the ‘medical poverty trap’).
16. Additionally, the world is experiencing a breakdown of the commune-level (or equivalent) health care system; the village health workers network, and traditional medicine have also decreased significantly in many countries.
17. Many countries have national compulsory health insurance – which only covers the more affluent quintiles (often civil servants) and, if voluntary health insurance cards can be purchased, percentually, the poor pay more for them as a share of their income.
18. It is not exaggerated to estimate that a good 80% of private practitioners are primarily employed in the pubic health care sector. But they are not making a decent living, so, now, even a big part of their monthly take-home wages have to be generated locally through fee-for-service regimes. The problem is sometimes so acute that even traditionally preventive health care facilities have had to start providing curative services to make money.
19. Furthermore, hospitals all over have started to operate private wards. Most have ended up being subsidized by funds from regular hospital operations – rather than the other way around – primarily because a) the more affluent users using these wards are charged below full cost, and b) VIPs do not pay when hospitalized there.
III. Fee for service-financed health care
20. The charging of (explicit and implicit) fees in public health facilities is inconsistent with equity objectives; revenues from it only cover about 10% of all health expenditures (although up to 50% of hospital expenditures). These fees thus only marginally contribute in terms of overall HCF – and they grossly increase inequality.
21. In the user fee system (a regressive tax), the sick pay more; the poor pay more; children and the elderly pay more.
22. The fee for service system increases the burden on the sick, especially the rural. It has weakened the structure of preventive care, because it has decreased funding for it causing a loss of preventive care staff with the ones left having poor skills.
23. If and when fees have been raised, they have NOT cross-subsidized the poor, i.e., the reallocation of funds from user fees to the poor has, so far, been rather limited or non-existent. In all truth, user fees can hardly be used for subsidizing the care of the poor.
24. User fees are increasingly replacing government recurrent expenditures in health, including salaries; user fees are now the main source of income for many health workers. We are thus witnessing an “under-the-table” commercialization of public health services.
25. On the other hand, there is NO positive relationship between high user charges and good quality of care – for either the rich or the poor. And, if the fee for service system does not have a financial ceiling, i.e. a budget, it becomes very difficult to control total costs.
26. Under the fee for service regime, providers tend to over-provide costly services (e.g., extra diagnostic tests and costly treatments) neglecting the needs for regular services of the rest of the population. This clearly gives priority to the more profitable, paying segment of the health care market.
27. With facilities competing for fees, the referral system has also been weakened; there has been a veritable change from cooperation to competition between providers.
28. User fees are unlikely to improve the current situation. [It is a too costly alternative that, in Vietnam, has cost close to 8% of GDP].
29. It is thus obvious that market-oriented strategies for HCF with high user fees have generated more negative than positive effects on equitable access.
IV. Medicines: How much of a culprit?
30. Drugs account for up to 70% of overall health expenditures. They, therefore, ARE of key importance to HCF.
31. Drug vendors in the private sector are the relief providers of choice and they are NOT health providers (nor, as a fait accompli, do we give them any minimum training!). Why do people choose them? Because of the limited (mostly economic) access they have to professional health care services: the poor often have no choice; self-medication is thus a major item of household expenditure.
32. Otherwise, private health care providers earn a large share of their income selling drugs in their private offices. This is as unethical as it is dangerous, causing losses to society that end up wasting money on unnecessary, useless or dangerous drugs. (Note that expensive drugs and poly-pharmacy contribute proportionately more to the income of the provider selling the drugs…and these are the medicines pushed by the veritable enemies of poor people’s health – the pharmaceutical houses representatives).
33. The regulation of the commercial drug market thus is of critical importance. But the control of the cost of drugs is difficult. The government simply has to develop (and/or expand) its capacity to monitor and control this market – at the same time strengthening the overall drug management system.
34. National Essential Drug Programs (EDPs) are in desperate need of enforcement (with these drugs only having low mark-ups).
35. Moreover, prescription audits are needed (using carbon copy pads that include the diagnosis and drugs prescribed). The focus is to be on decreasing poly-pharmacy and the use of antibiotics, injections, steroids and vitamins, i.e., controlling the irrational use of drugs.
36, An option to also consider is the replacement of the mark-up-per-drug-system by a fixed charge per prescription.
V. A basket of potential solutions?
I see some general ‘golden rules’ that I think do apply to HCF almost everywhere in the Third World. Among them are the following:
37. In HCF, one has to measure any planned intervention against the cost of resulting ill-health if nothing is done. The horizon should be to progressively plan to increase access to health services by the poor over a 3-5 years period.
38. A larger population of users allows for economies of scale. Ideally, in that population, the healthy should subsidize the sick, the rich should subsidize the poor*, the working should subsidize the dependent (mostly children and the elderly).
(*: i.e., a progressive taxation with higher rates of taxation of the higher income groups).
39. Also, a greater cross-subsidization across income groups is needed. (One option to keep in mind is for private for profit hospitals to be taxed at levels that will secure cross-subsidization to public hospitals).
40. Any future HCF scheme will continue to have to primarily draw from traditional third party payers other that the patients themselves, i.e., the government (national and local), different insurance schemes, donors and community health care funds.
41. To universalize access to health services, one needs to subsidize the poor using both local and outside funds coming both from inside and outside the health sector. [In Vietnam, it has been proposed that, in rural areas: The government should end up paying 75% of all health care costs; the community 10%; foreign aid 10%; and user fees 5%. In urban areas: taxes should cover 50-60% of the costs; health insurance 20-25%; fees 10%; and foreign aid 5%].
42. Insurance schemes cannot replace a national publicly financed health care system; they can only contribute to it. Subsidies will thus be required even for well-designed and efficiently run public insurance schemes.
43. In general, our choices in HCF will drive the health sector either towards:
– a market-oriented privatized system;
– a hybrid system where the government and the private sector play equal roles;
– a fully tax subsidized system as in the (far) past; or
– a direction in which the government takes the key role and the private sector is considered a complementary component, but shares the same responsibilities and tasks (including preventive care).
The latter seems to be the most appropriate option, i.e. the public sector takes the lead and exerts an effective controlling role over the private sector.
44. There is no escape: In one way or another, the government has to increase its health spending. But it has to gear it to pro-poor interventions, e.g., preventive services and services directed at the diseases of poverty; it is these that have to be fully financed.
45. Making resources available is not enough; they can be mis-used, under- utilized and wasted in many different ways. As part of any HCF strategy, measures have to be taken to have all health resources used more effectively.
46. Also, assessing medical technologies in use (or proposed for use) is of critical importance from the HCF point of view as this makes it possible to reduce or rule out costly diagnoses and treatments without proven benefits (e.g. ultrasound).
47. Exemption schemes are infinitely more difficult to implement than having across-the-board low or no user fees for the more vulnerable populations. Alternatively, the government can grant tax exemptions to poor districts. Exemptions are nevertheless needed for all cases of public health importance such as TB, malaria, STDs, HIV/AIDS.
48. Even if painful, another avenue that needs to be pursued is adjusting personnel deployment to current levels of services provided. Mobilizing the abundant under-utilized manpower is thus also a must; facilities are often overstaffed relative to patient load; much time is wasted waiting for patients.
49. Furthermore, patients tend to seek care at too high levels of the health system (hospitals over health centers) and this is very costly. Efforts are needed to ensure more appropriate entry levels of consultation. Consequently, patients have to be seen first at commune health center level (or equivalent) where it is definitively cheaper. Incentives will have to gear consultations to this entry point. In practical terms, this means having to strengthen health centers and the referral system!…and that is easier said than done.
50. For any HCF scheme to succeed, there also is a need for management training.
VI. ‘Righting the wrong’ in the obsolete geographic allocation of funding for health
51. Financial resources for health should be driven according to need (with resources going to clearly identified priority areas), and not be based on a per-capita basis (where paying does not take into account the more vulnerable populations and those with lower ability to pay).
52. The HCF we advocate-for calls for a needs-based health care planning, and this means reallocating tax revenues from better off to poor regions. Why? Because only tax funds are amenable to be reallocated from rich to poor regions.
53. Currently, tax funds (the funding for health included) often favor better-off provinces, for instance, allocating funds according to the number of hospital beds: this simply reinforces existing differences!
54. Allocation according to need considers size of the population, disease burden of different groups and their ability to pay.
55. To accomplish this, needs-based resource allocation, indexes do exist already. There is a need to further develop and adopt country-specific needs-based indexes though, for resources to be more equitably allocated between regions, taking into full account total resources availability and deciding “how to cut the cake”. [Note that a special index for allocating funds for preventive services according to need is also needed].
56. For all the above reasons, it is here contended that by far the more appropriate choice for HCF in developing countries is to focus on possibilities of risk pooling and pre-payment, ensuring that both healthy and sick people contribute via taxes and social health insurance.
VII. And the winner is…Community-based health insurance (CBHI)
Some general attributes and preconditions:
57. Social insurance – in this case CBHI – is a household responsibility system; it is meant for people in the informal sector, mostly the rural and urban poor; as any insurance, it is based on risk sharing and resources pooling; it is more viable where the community is already organized for other purposes.
58. A CBHI system assures a highly equitable coverage, a potentially more egalitarian geographical access, a greater equity in the financing of its operations, a potential for health care cost controls and an increased efficiency in the use of resources, as well as good financial sustainability and a chance for improved quality of care. It provides a greater latitude compared with pure tax financing of health care (and for sure compared with the user fee system). [The alternative of going back to full direct tax financing of health services would require considerable budgetary reallocations that are not as politically viable as implementing a sound CBHI].
59. Implementing a CBHI scheme needs a clear government policy and legal framework. To begin with, three major policy changes are deemed necessary:
– legislation for a community tax of 1.5-2% of income;
– an additional central government tax support of 0.2-0.5 of GDP; and
– changes in the personnel compensation system allowing for bonus payments.
60. The option here proposed is to gradually develop a subsidized community-based health insurance system that covers an increasing proportion of the curative care needs of the rural population and is jointly financed by community contributions, public funds and, initially, also foreign aid, i.e., community contributions are matched by public funds. (The public sector continues to pay 100% for traditional preventive services as before). [China has, among other, considered a CBHI system in which the pre-payment only covers a minimum package of a) peri-natal services: 4 antenatal care check-ups, delivery costs and 3 post-natal check-ups; b) EPI: 13 vaccines given to <7 year olds; and c) basic medical care: 3 ambulatory visits/yr and ½ in-patient day/yr. (So, it is to be noted here that they also include the financing of some preventive services under the CBHI scheme.)]. 61. A fully voluntary CBHI system will inevitably lead to an adverse selection problem (with mostly the sicker and the elderly wanting to join). A compulsory system will also not work. People at the community/village level will have to make a local voluntary-but-collective decision to join the scheme. Piloting is recommended. 62. Note that a subsidized rural health insurance system also helps the allocation of resources to different provinces according to need. A special fund to support less developed areas (remote, minorities) will have to be created as needed. 63. The main challenge will ultimately be to convince people that what is proposed is for them to reduce their ‘envelope (under-the-table) payments’, and, instead, start paying a clearly identified local health tax. Getting started (a non-exhaustive listing): 64. A village can only join the CBHI scheme once 60% of its members have agreed to subscribe; membership can be individual or household-based. (Note that, to keep premia low, members can choose to adopt a user co-payment of 10-20%). 65. Once the community accepts to join, people get together and make a pre-payment (an investment, really, since they will not have a financial burden at the time of illness). At this time, communities can also determine exemption mechanisms to be applied. 66. Getting the CBHI started will require social mobilization to make people understand its long-term benefits. For instance, the membership will be educated on the philosophy and mechanics of the new health care financing scheme and on how the service delivery system will work from now on. They will also be trained in cost-saving measures in the provision of health care services. Members will further need education on insurance principles and the principles of democratic supervision. 67. Some ‘benefits’ from ambulatory services (in part from a moderate mark-up on drugs?) can be used to help (pre)pay needed in-patient care and accidents/injury care. 68. Increasing members’ awareness of personal and community health issues is very important as well in an effort to reduce morbidity and ensuing consultations. In a related matter, an additional important benefit to be discussed with members is that, if the quality of health services is below standard and causes health problems - such as EPI-related infections still being prevalent - beneficiaries will be able to claim compensation. 69. For those who will need subsidies, communities themselves would need to identify poor households, define the support they need and where this support is to come from; this means that, as part of the scheme, a community health care fund has to be established for the indigent. 70. Once a CBHI is in place, it is foreseen that a large share of resources now going to the private providers will return to the public providers. Caveats to keep in mind when setting up a CBHI scheme: 71. Most importantly, the government has to enjoy sufficient popular credibility upfront for the successful organization and management of a nationwide CBHI system. 72. No less important is our understanding that, unless quality health care is guaranteed, there will be few takers. In general, CBHI must offer benefits such that the common person will want to become part of it, i.e., the challenge is to ‘create-a-demand’, especially in rural areas. 73. As regards the payment of this insurance premia, a process of consultation has to precede the drawing up of policy plans for any future CBHI. Premia are best set with the community according to their objectively assessed purchasing power; in this same manner, premia are to be (re)set annually. 74. At the end, premia should be attractive as relates to the coverage offered to members, as well as competitive and fair in the reimbursements paid to affiliated provider institutions. 75. Premia are paid annually in cash or in kind using a sliding scale (or percentage of income). The challenge is for payments to be made on time - and to be 100% accounted for. One has to be flexible and tailor premia collection dates to the periods when the community has the means to pay. They can be collected either as a deductible or as a door-to-door contribution. 76. Irregular accounting or embezzlement in any CBHI scheme will make matters worse. 77. As regards participation in CBHI, we said the local community has to set up a committee to oversee the whole pre-payment system. This participation is particularly important in the management of the finances of the health center to provide members entry and referral services; it represents a real (and indispensable) shift-in-the-balance-of-influence in favor of the users of services. 78. In a CBHI scheme, users are thus empowered to audit accounts and to oversee the use of funds, as well as to monitor health workers, i.e., one should look at CBHI as an instrument of community empowerment where members who join really have a voice, if for no other reason than because they are directly contributing to shoulder the costs. [Note that CBHI also promotes gender equality and ensures women have direct and easy access to health centers]. 79. Finally, the complexity of organizing these schemes is not to be underestimated; it limits our ability to scale them up rapidly. The case of Vietnam: 80. Vietnam is considering six financing sources for its CBHI (*): - the National Health Insurance, - provincial budgets (using per-capita-based allocations); - international donor support; - a new community tax of 1.5-2% (**); - a reallocation of the government recurrent budget of 0.2%-0.5% towards the health budget; and - a change in the personnel compensation system (introducing productivity bonuses based on services actually rendered). (*): These allocations would not go directly to the facilities providing the services, but to the insurance entity that would then reimburse facilities as per the respective volume and type of services they provide. (**): It is estimated that around an additional one half of one percent (0.5 %) of GDP allocated by the national budget to health could finance significant increases in the equity of health care when used to subsidize a CBHI scheme. The government is also distributing health insurance cards to about four million poor. 50-75% of the cost of these cards is borne by the government (special item line in the national budget) and the rest is borne by users; the very poor are getting them free or at nominal cost. Claudio Schuftn, Ho Chi Minh City schuftan@gmail.com Acknowledgements: Concepts mostly taken from Goran Dahlgren’s and David Dunlop’s writings on the topic.