SEARCH

SEARCH BY CITATION

Keywords:

  • Sudan;
  • exemption policies;
  • user fees;
  • emergency obstetric care;
  • child health;
  • health financing

SUMMARY

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

In 2008, the Government of Sudan launched a policy of free curative care for under-fives and caesareans. This paper presents the findings of a review of this policy, on the basis of research conducted in five focal states of northern Sudan in 2010. Policy implementation was assessed using four research tools: key informant interviews, exit interviews, a facility survey, and analysis of facility finances and the cost of the package of care.

The findings point to important weaknesses in implementation, such as unclear specification of the exact target group and package of care and inadequate funding. Despite this, service utilisation appears to have responded, at least in the short term. The findings also highlight the urgent need for improved access to basic health care and financial protection against health care costs in northern Sudan (for those with and without national health insurance membership).

This review contributes to the growing literature on the selective removal of user fees for priority services. It indicates the range of challenges to effective implementation (strategic, financial and organisational). Some of these are particular to Sudan, but many are shared, and indicate important lessons for improving access to and quality of care for women and children in Africa.

Copyright © 2012 John Wiley & Sons, Ltd.


INTRODUCTION

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

A policy of providing free care for caesareans and under-fives was launched in 2008 in northern Sudan. In 2010, a study was undertaken by a research team based in the Federal Ministry of Health (FMoH) Khartoum to review its implementation and impact after 2 years of operation. The aim of the study was to provide information to the government to enhance the policy's sustainability and impact. This work formed part of a series of studies that intended to feed into the development of a national health financing strategy.

The policy in Sudan took place against a backdrop of similar initiatives in many countries in the region (Witter, 2009; Ridde and Morestin, 2010; Meessen et al., 2011). The number of countries that have recently reformed their user fee regime is not exactly established, as many have recently introduced reforms that have not been fully documented at local or national level. However, a recent mapping report on user fees (Witter, 2010), based on a survey of key informants in 49 low-income and middle-income countries, found that more than half of the countries in the selection (28 out of 49) had, in recent years, introduced reforms to their user fee regimes. In many cases, the focus has been on priority groups, particularly pregnant women and young children. There is a clear regional focus for this in West Africa, but the approach extends beyond. Of the 28 countries, exactly half focussed on delivery care (either exclusively or together with curative services for young children). There is a growing body of evidence on some of the difficulties that are presented and on how to make the reforms work more effectively (Witter et al., 2008; Meessen, 2009; Ministry of Health, 2009; Witter et al., 2009; Meessen et al., 2011).

This article presents a summary analysis of the findings.1 It provides a significant regional case study of some of the issues facing implementation of fee exemption policies for priority groups and adds to the growing body of evidence in sub-Saharan Africa looking at preconditions for effectiveness of fee exemption policies.

BACKGROUND

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

Political and health system context

Since the south became independent in July 2011, Sudan comprises the 15 northern states (out of the previous total of 25). Although the study was conducted prior to southern independence, the health systems were already separate, and the study focussed on the northern states. Sudan has a three-tier government system—federal, state and local government. Under the decentralised system that has been rolled out in Sudan since 1994, the states take primary responsibility for health. States set overall policies and manage the hospitals. Localities are meant to be responsible for health centres and basic health units although in practice, their capacity to manage the health sector is limited in many areas (especially postconflict areas). In financial terms, block grants from the federal level are sent to the States, which have scope to allocate them to different sectors (apart from for salaries).

Health centres are the lowest level affected by the policy: they provide child health interventions and antenatal care/family planning but not (in most cases) deliveries. At community level, deliveries are conducted by midwives at home, and these are not covered by the policy. At the next level, the rural hospital provides deliveries and caesarean sections (CS), as well as other curative care (Federal Ministry of Health, 2009). The general hospitals at state level and the specialist centres in Khartoum are also participating in the policy.

The government's 2007 health sector strategy aimed to achieve progress towards the health and nutrition-related Millennium Development Goals with particular emphasis on improving the health of poor and vulnerable populations. An objective is to broaden coverage of a basic package of health services, including both facility-based services and interventions that can be delivered at the community level.

Health situation of northern Sudan

Sudan continues to confront major challenges in making progress towards the health and nutrition-related Millennium Development Goals. Health and nutrition indicators seem to have stagnated since the 1990s. The 2007 Sudan Household Health Survey estimated that under-five mortality in the 15 northern states was approximately 99 per 1000, little changed from the estimate of 104 from the 1999 Safe Motherhood Survey, which showed an improvement compared with the estimate of 135 from the 1989 to 1990 Demographic and Health Survey. Estimated maternal mortality ratios remained around 500 per 100,000 live births, with little change between the 1980s and the 1990s, although more recent modelling suggests a decline since 2000 (Hogan et al., 2010).

Although coverage of child and maternal health services has improved overall in recent years, utilisation remains low and significant disparities remain between states. For example, in 2007, 73% of one-year-old children in Northern state were fully vaccinated compared with just 23% in West Darfur (Federal Ministry of Health and Central Bureau of Statistics, 2006). Similarly, access to curative care, such as caesarean sections, has improved in all states but remains much lower in the poorer states of the Kordofan and Darfur regions, as well as Blue Nile and Gedarif. In 2007, although 14.2% of deliveries in River Nile state were by CS, the proportion was only 0.8% in West Darfur.

The poor also have significantly less access to health care services, with very large disparities in relation to a number of services across wealth quintiles. In the poorest quintile, for example, 53% of pregnant women received antenatal care compared with 91% in the highest quintile. An even lower proportion of pregnant women in the poorest quintile received qualified delivery care (38%) compared with 89% in the highest quintile. Disparities in access to emergency obstetric care, an important determinant of maternal mortality, are striking, as only 1% of deliveries among the poorest quintile were by CS compared with 19% among the highest quintile.

Origins of free care policy for caesareans and under-fives

From independence until 1991, health services were offered free of charge in Sudan. User fees for services and drugs were then as part of the wider economic reforms and adjustment policies. The impact of the introduction of user fees in public health facilities was not well documented; however, anecdotal evidence suggests that the introduction of user fees significantly affected access and utilization of health services with little or no significant improvement on the availability and quality of care (Federal Ministry of Health, 2007).

The access problems that this generated led to the development of health insurance, which has been implemented since 1995. The National Health Insurance Fund (NHI) is a semi-autonomous body sitting under the Ministry of Social Welfare. Contributions come from employees (4% of salary), employers (6% of salary), the government, charitable donations and company investments. Membership is compulsory for the formal sector and voluntary for the informal. Membership covers the family not just the individual. The monthly premium for the informal sector is SDG 15 ($5.6) outside Khartoum and SDG 20 ($7.5) in Khartoum. Benefits include most treatment costs and 75% of the cost of drugs. There is internal cross subsidy from the formal sector to the informal members. Overall coverage of the population in the northern states is estimated at 40% by the NHI.

To ameliorate the negative impact of the introduction of user fees on access to health services, emergency cases at hospitals (including any procedures and expenses incurred within the first 24 h in hospital) have been exempted from fees since 2006. Other exemptions include renal dialysis, immune suppressant drugs for renal implantation, chemotherapy, radiotherapy and treatment of haemophilia. The annual expenditure on free treatment amounted to some 15–22% of the total federal health budget in 2005 (Federal Ministry of Health, 2007).

The 2007 Interim Constitution of the Republic of the Sudan, article 46, stated the commitment of the Government to provide universal and free-of-charge basic health services. According to the 2008 National Health Accounts (NHA), total public spending on health is now close to $34 per capita. Of the total health care expenditure of $122 per capita, just under 67% comes from households (66% out of pocket) and 28% from public sources. Less than 4% of health finance was external (donor funds). The decision to make care for under-fives and pregnant women free in 2008 was the latest in a series of measures to shift the financing burden back from the population to the state.

RESEARCH METHODS

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

The research framework (presented in Figure 1) identifies the main preconditions for effectiveness of the exemptions policy, the intermediate outcomes that need to be achieved and the ultimate goals of the policy. The study set out to track these, although within limitations, such as lack of baseline data and limited time and resources. Given those, the study was not able to assess change to health outcomes.

image

Figure 1. Conceptual framework for study

Download figure to PowerPoint

Research tools

Four main research tools were used to examine these questions, in addition to secondary data sources:

  1. Key informant interviews (KIIs), to examine stakeholder perceptions of the policy and its impact, as well as to collect important secondary records, such as reports on free care
  2. Exit interviews (EIs), to collect information on patient characteristics, health-seeking behaviour, payments and satisfaction with quality of care (for two main groups—women who had recently delivered and carers of sick children under five)
  3. Facility survey, to collect data on inputs to care (such as finance, staffing, drugs, equipment and infrastructure) and also outputs
  4. Costing of package, to understand the cost to facilities of providing the core services covered by this policy. This costing focussed on marginal costs of services at four levels of facility—health centre, rural hospital, state and federal level hospitals. The costed services included normal deliveries, caesarean sections and the 3–4 main conditions for which children are treated, such as diarrhoeal diseases, respiratory infections including pneumonia, malaria and malnutrition.

The link between research questions and tools is set out in Table 1.

Table 1. Research questions and sources used
Research questionsSource and tools used to answer questions
Preconditions for success
1. Was the policy well specified and communicated?Key informant interviews
2. Was implementation carried out according to guidelines?Document review
3. Was the policy well managed and monitored?
4. Was the funding adequate and timely?Key informant interviews
Document review
Facility survey
Costing
5. Was there demand for services and awareness of the exemptions?Exit interviews
Outcomes (intermediate and ultimate)
6. Were facilities and health workers enabled and incentivised to provide appropriate services?Key informant interviews
Document review
Facility survey
Costing
7. Was quality of care maintained or improved?Facility survey
Costing
Key informant interviews
Exit interviews
8. Did families face reduced costs?Exit interviews
9. Did utilisation of care increase?Secondary data
Interactions with wider health system
10. How was the policy affected by, and how did it affect, other parts of the health system?Key informant interviews
Document review
Facility survey

Piloting was conducted in River Nile State. The tools were then used in five focal states—Khartoum, Red Sea, Kassala, Blue Nile and South Kordofan (see Figure 2). Fieldwork took place in March–April 2010. Findings were presented and discussed at a stakeholder meeting in September 2010.

image

Figure 2. Map of northern Sudan and study areas

Download figure to PowerPoint

All interviews were treated confidentially, and respondents were given the option to participate or not. Ethical approval was granted by the FMOH research ethics committee prior to fieldwork.

Sampling and analysis

The choice of states was determined by a prior project funded by the multidonor trust fund, which was operating in four of the five focal states. Khartoum was added during initial discussions as the capital state is considered an important comparator. The States presented a range of conditions, from the relatively wealthy eastern states of Red Sea and Kassala to the less developed Blue Nile and South Kordofan.

Within states, the facilities were selected from amongst a list of participating institutions obtained from the States' focal persons. Facilities were chosen within four institutional categories—initially randomly but modified if major problems of access or functionality were identified. Thirty facilities were included in the end: two national hospitals, seven state hospitals, five rural hospitals and 16 health centres. The aim was to obtain a snapshot of the situation on the ground in the focal areas rather than any kind of regional or national representativeness. These facilities represent the main levels of curative care in Sudan. Within these states, the sampled hospitals constitute about a tenth of the total number; for health centres, the proportion was smaller (2%).

A total of 117 key informants were interviewed. The bulk came from State Ministries of Health (43) and from facilities (39), but they also represented the FMoH, Ministries of Finance, the NHI, the Zakat (the body charged with collection and distribution of Muslim charitable donations), the Central Medical Stores (CMS), Revolving Drug Funds (RDFs), the development partners and researchers. The total for the national level was 24, for Khartoum State 26, for Red Sea 19, Kassala 17, for South Kordofan 17 and for Blue Nile 14.

For the EIs, 138 women who had recently delivered were interviewed, along with 248 carers of sick children under five. For the facility survey, 30 facilities (from national down to health centre level) were included. For the costing, 24 facilities were included although for reasons of missing data in two states, only 16 were analysed in the end.

Analysis of the KII used manual coding of themes from interviews. EIs and the facility survey were entered using EpiData and analysed using the statistical package stata. The financial data analysis and costing was undertaken in Excel.

FINDINGS

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

Preconditions for success

Policy specification

The policy was announced by the President in January 2008 and subsequently by the FMoH. In the states, a ministerial decree was issued, and the policy communicated via the media. However, key informants at all levels reported that no detailed written specification of the policy had been received.

Interpretation and implementation

Linked presumably to problems of funding and lack of clear guidelines, modifications had taken place at state and facility levels, which undermined the effectiveness of the policy. These modifications varied by area but included deciding to exclude whole categories of facilities, such as health centres, and also categories of patients (e.g. focussing on child inpatients only). Normal deliveries were excluded in all areas although in Khartoum, a 50% subsidy has been extended to them. In one state, Blue Nile, the policy has been largely suspended, with very few resources reaching any facilities. The state that has been able to implement the policy most thoroughly is Khartoum, which has called on more state-level resources to fund free care.

In general, when rationing occurred, the bias has been towards hospitals and urban facilities—in most states, it was the rural areas and health centres that lost out. In Red Sea, for example, 45% of the funding went to the emergency hospital, with only 13% going to the specialist hospitals for women and children in Port Sudan, 21% going to the rural hospitals as a group and only 6% to health centres. This is a pattern that raises concerns about efficiency, as well as equity. Although this may reflect utilisation patterns and access, it clearly threatens to aggravate inequities.

Looking at the very crude indicator of the proportion of facilities in each state participating in the policy (even theoretically), the coverage was very varied, ranging from 81% in Red Sea to 37% in Khartoum (presumably because of the large number of private facilities), 16% in Blue Nile, 11% in Kassala and 6% in South Kordofan (Table 2).

Table 2. Facilities participating in the policy, focal states
StateType of facilityNumber of facilitiesProportion of total facilities
Khartoum StateAll state hospitals and health centres28 hospitals and 159 health centres37%
KassalaAll state hospitals and some urban health centres13 hospitals11%
Red SeaAll state hospitals (urban and rural) and health centres in seven localities17 hospitals; 43 health centres81%
Blue NileAll state ministry of health hospitals and some health centres (ones run by doctors)13 hospital and 19 health centres16%
South KordofanAll Ministry of Health hospitals15 hospitals6%

There was considerable variation in interpretation of the services that were covered by the policy, even by different stakeholder groups at national level (see Table 3), which reflects a lack of clear specification but also practical difficulties with full implementation.

Table 3. National key informants' views of contents of the free care package
Key informantInterpretation of the package of care on offer
Health economics unitAll care for under-fives
All delivery costs
Reproductive health departmentAll caesareans (but not normal deliveries)
Child health departmentChild health services but only applied in practice at tertiary level, not lower facilities
Free health care directorate FMoHReports on care for under-fives, caesarean sections and also normal deliveries
Hospital directorCare for all children (not just under-fives); all deliveries (including normal, but these are only 50% subsidised, not fully free)
Gezira MoHEmergency caesarean sections are free; elective are charged; care for under-fives offered free but not in all hospitals
MoH finance departmentAll women, and under-fives
Khartoum MoHElective caesarean sections, D&C and all care for under-fives (but excluding some drugs for out-patients); also some subsidies for teenagers and normal deliveries

This confusion was mirrored at state level, where there were differences in the package that was theoretically on offer varies across states. Moreover, different key informants within each state gave different accounts of the package of care that was available. For Blue Nile and South Kordofan, the policy appeared to no longer be in effect in reality. Red Sea and Khartoum were continuing, albeit with some alterations to cope with the limited funding of the policy. Kassala was continuing in part, with debts mounting up to the facilities.

Management and monitoring

Management and monitoring was found to be fragmented. At the national level, the free care department held information on drugs sent and services delivered but no financial data. The FMoH finance department had information on budget and expenditure but only for the drugs that have been distributed to the states and not for all of the tertiary institutions. The Ministry of Finance (MoF) free care department was in a similar position, holding information on funds disbursed but not including tertiary institutions and with no reports on outputs or outcomes of the policy. None of the federal bodies held information on how much the States were committing, in terms of resources. Information was not combined across these different centres to allow for analysis of important aspects, such as unit costs per free patient, comparing resources and utilisation across the states, comparing the drugs used with the number and type of patients treated and so on. Supervision was limited, and there were no standard checklists or budgets for supervision, either from the federal level or below.

Financing of policy

The policy is supported by funds from the federal level. Only in Khartoum was there evidence of state support. Other potential sources—localities, Zakat and health insurance—were reported not to be contributing, at least in these focal states.

According to key informants, cash and drugs were inadequate to cover the needs of the intended beneficiary groups. Gaps of 60% to 100% were estimated by different stakeholders. At national level, the free care department was more optimistic, estimating that drugs covered roughly three-quarters of needs. Using our cost estimates to retrospectively assess funding adequacy, if the assumption is made that the package includes caesarean sections (CS) and all in-patient and out-patient child treatments, then funding would have covered 7% of the total needs.

Overall expenditure

On the basis of national estimates,2 overall expenditure for the free care programme for under fives and pregnant women in 2009 was SDG 18 million ($6.74 million) or SDG 0.58 ($0.223) per person (total population) in the northern states as a whole. The cash that was sent to the states varied from SDG 0.4 ($0.15) per person for South Kordofan per year to SDG 0.061 ($0.02) per person for Khartoum. In addition, monthly drugs were sent whose value ranged from SDG 1.3 ($0.49) per person per year in Blue Nile to SDG 0.66 ($0.25) per person in Khartoum. However, these drugs were for emergency care as well as under-fives and pregnant women.

Of total free care spending in 2009, this group (under fives and pregnant women) absorbed 13%. In relation to drugs, the proportion is lower—6%, according to figures for 2008. The National Renal Centre received more than four times the value of drugs for 2008 compared with the free care programme for women and children in the whole country.

According to the annual report from the Free Care Department in the FMoH, 3.4 million children were ‘covered’ in 2009, and 66,000 CS. As financial reports combine expenditure on these two categories, however, it is not possible to calculate average unit costs.

Where figures were available for total spending on health care at the state level (for Red Sea and Blue Nile), expenditure on this free care programme amounted to 1% of overall public health expenditure. Total per capita spending in Red Sea was SDG 27 ($10), whereas for Blue Nile, it was SDG 30 ($11).

In relation to total public spending on health in 2008, according to NHA figures, the free care policy for under-fives and pregnant women used 0.005% of overall expenditure.

Budgeting

Resources to support the policy flow through three main channels (all originating in the Ministry of Finance):

  1. Funds are sent to the Central Medical Store, via the FMoH, for free drugs to be supplied to the States (and some of the tertiary institutions).
  2. Funds are sent direct to the tertiary institutions to pay for the service costs of free care.
  3. Funds are sent monthly to the States for services costs.

However, the basis for budget setting at the national level was not clear—there is no evidence that this was based on cost and need estimates derived from the states. The division of resources between states was also not clear—certainly, it did not follow population patterns, but there may have been underlying calculations relating to poverty, as the funding was found to be higher per capita for the poorer states.

Allocation and use of funds

Within states, the system of distributing resources to the different areas and facilities also lacked transparency. None of the states used a fixed payment per episode to reimburse facilities. For the facilities, the resources for the policy were both inadequate and unpredictable, according to KIIs (the cash funds were especially unpredictable—drugs were more predictable but still well below the reported needs).

The reporting formats of the national free care programme were not used in practice, and states reported on use of funds in a wide range of formats. Khartoum and South Kordofan were unable to provide any breakdown for how funds were used. For Blue Nile and Kassala, the range of uses was wide, and it was not clear whether the funds were sent to facilities to make up for lost fee revenues or they were used for general investments by the State Ministries of Health.

Key informants were not clear on the extent to which free care funds were ring-fenced or not. It seems that funds are in practice sometimes used for other expenditures, such as rehabilitation, equipment and incentives. Conversely, some KI pointed out that hospitals are covering free care costs from other budgets.

Community awareness

Community awareness varied according to the area and correlated with implementation (in those rural areas where the policy was in effect not being applied, awareness was low). In EIs in focal states, awareness of exemptions was low for the carers of sick children under five—40% were aware of them. In the EIs with women who had delivered, awareness of the existence of exemptions was higher—54%. However, it should be noted that respondents, being interviewed in facilities, are likely to be more aware than the wider population.

OUTCOMES

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

Impact on staff and facilities

Facility KI reported increased workload, especially at the higher levels. There were also rumours of increased corruption as a result of the policy. For staff, the policy erodes the incentives for health workers that came from user fees, while offering opportunities to make money in other ways, according to KI, including sale of free drugs. KI reported that incentives were not paid under this policy, but the reports on the use of free care funds included for incentives, which is either contradictory or means that free care funds support a more general incentive system.

Overall, KI reported that staff had mixed feelings—they are relieved that patients are able to be treated but feel dissatisfied with the policy on a personal level. This applies in particular to surgical staff.

The main concern expressed by facility KI was that they are not being fully reimbursed for the services they are providing. Some facilities, especially in Kassala, reported growing debts to suppliers. Health centres and rural hospitals depend for 70% of their nonwage recurrent costs on user fees, according to a national KI. The effective operating of the policy is therefore particularly important for them. Reimbursements by the health insurance are also subject to delays and short falls.

Under the current system, however, revenues are not necessarily retained at the facility level—typically, they keep a proportion, sending the rest to the State Ministry of Finance, although the proportions vary by state. One KI in Red Sea estimated that the state MoF provided about 50% of estimated overall financial needs. In this context, the user fees were contributing not just to facility costs and activities but also Ministry of Health (MoH) overheads at state level (vehicles, running costs, management, etc.). In Red Sea, negotiations were underway for user fees to be returned to the MoH, to be shared between staff (40%), facilities for running costs (45%) and the MoH (15%). In Blue Nile, the MoH retained some of the user fees that it generated at state level, passing a ‘cut’ to the MoF. Any balance held at the end of the financial year by the MoH was also to be returned to the MoF. The balance might be returned or not to the MoH. For the facilities, the MoH had no records of their user fees, which were retained. There were no regular financial flows from the MoH to the facilities.

The facility survey showed the proportion of total revenue coming from different sources over 2007–2009. The patterns were markedly different. The federal hospitals reported the highest dependence on state funding, followed by funding for the free care policy, whereas state hospitals have much larger source in patient payments. Health insurance contributed only a small amount—from 0% to 5% of overall revenues (compared with 6–17% from free care). Looking at how their overall revenues have changed since 2007, revenues have grown most strongly at state hospitals, whereas rural health centres have lost almost a third.

Looking at the balance of revenues and expenditures, which indicate the overall financial health of facilities, the picture seemed to vary considerably year to year. However, most facilities types improved over the period, with the state hospitals' surplus growing and the federal hospitals' deficit diminishing over the 3 years. There was no indication from this that the free care policy was causing financial distress at facility level.

Quality of care

Key informant reports on quality of care varied—some saw improvements due to improved access and the federal resources, some saw no change and others again believed that diminished revenues and increased workload had reduced the quality of care given to patients.

Our facility survey highlights preexisting inequalities in the inputs to quality of care—staffing ratios are much higher in Khartoum than elsewhere, for example. Staff numbers have been increasing, especially in understaffed areas, such as South Kordofan. However, inequalities in 2009 remained high. Khartoum, for example, has more than six times as many consultants per facility than South Kordofan (although this is influenced by the inclusion of tertiary hospitals for Khartoum, making it less comparable than the other four states). Only for the lower cadres is the relationship ever reversed (e.g. medical assistants—here, there are more than four times as many in South Kordofan as in Khartoum).

Infrastructure (functional water and electricity) also follows a clear gradient of functionality, ranging from 100% access to functioning water and electricity in Khartoum to only 20% access to both for facilities in South Kordofan. Although these are not linked to the free care policy, they directly influence its effectiveness and ability to meet its goals. Facilities were also asked about a range of equipment that is important for MCH services. Although the level of equipment tends to link to facility type, it is interesting that for some items, the health centres are better equipped than state hospitals (e.g. 94% had a stethoscope compared with 86% for state hospitals). Although the sample numbers are small, this indicates the patchiness of quality in some institutions—these are items that all state hospitals should clearly have. Similarly, the fact that only 44% of health centres have sterilising equipment is worrying.

In the costing exercise, clinicians were asked about treatment for common conditions for children and for deliveries. Considerable variation was found in clinical and prescribing practices, with even the referral hospitals not following the national guidelines. This raises a concern, although one that is wider than the free care policy.

For households, and especially for deliveries, quality emerged as one of the most important factors motivating the choice of facility in our EIs. For children, proximity was the most important factor, as reported by carers, and price was ranked higher than for deliveries.

Reported waiting time was not very long—61% were seen in less than 30  minutes, according to our interviews with women, although there was a worrying tail to the distribution, with 8% being made to wait 2–4  hours and another 8% over 4  hours.

For children, more than half (51%) had been to the same facility more than twice before for treatment of the same episode, which seems high for these relatively simple conditions and may suggest poor treatment or wrong prescribing.

In relation to questions on user satisfaction, in the EIs with women who had delivered, 80% were very or quite satisfied with waiting times (and 20% dissatisfied or very dissatisfied). The responses were similar for satisfaction with health workers and quality of care in general, with higher level facilities tending to have slightly higher satisfaction scores. For pricing structure, however, the picture was very different, with 62% stating that they were dissatisfied or very dissatisfied—a reflection presumably of the lack of affordability and clarity of prices in most facilities. For medicines, the majority (51.5%) were dissatisfied—with, as usual, higher levels of dissatisfaction at state and rural hospitals compared with national ones. Similarly, for carers of children, the only two indicators where client unhappiness outweighed happiness were cost of treatment and availability of medicines—62.5% are dissatisfied or very dissatisfied with both of these.

Financial benefits for households

Key informant report that many charges are still falling on households because of the shortfall in funding. Some payments are formally mandated—for example, in Khartoum, where out-patient drugs for children are not covered. Others are ad hoc—when drugs run out, people pay. In addition, there are some reports of the loss of user fees being offset by higher charges for other services, such as visiting fees.

There is very little predictability of costs for households. Very few facilities visited have fixed charges per episode of care, and where they do, they are not made public. Similarly, no facility posted notices about exemptions, despite the fact that the exemptions on offer differed across facilities.

In the EIs with carers of sick children under five, less than 2% of respondents reported total free care (Figure 3). This suggests that the policy is not being implemented effectively.

image

Figure 3. Proportion of respondents reporting free services for sick under-fives (by cost item)

Download figure to PowerPoint

Affordability of care
Children's services

According to the children's EIs, the mean total expenditure on fees for the episode of care that their child had just received was SDG 62 ($23.2) per child, with the lowest mean reported in Blue Nile (SDG 23; $8.6) and the highest in Kassala (SDG 129; $48.3). There is also a clear gradient by facility, rising from SDG 20 ($7.5) at health centres to SDG 99 ($37) on average at national hospitals.

Comparing the expenditure on an episode of care for a child with the household's average monthly nonfood expenditure, the average proportion is 44.5%, although the range is very wide (from 0% to more than 600% of nonfood expenditure). Given the frequency of illness in a family of several members, this proportion is high.

By states, the proportion is lowest in Red Sea (mean of 27.5%) and highest in Kassala (77%). By facility visited, the proportion is lowest at health centres, as you would expect (14%). That it is highest at state, rather than national, hospitals (65%, as opposed to 58%) is more unexpected.

Looking at the different payment components, payments outside the facility account for 49% of the total payments, and within facilities, drugs and medicines are the single largest component. It is possible that other facility payments were higher before the free care policy—we have no ‘before’ data with which to compare our findings. It is however also clear that care for children is not free (in relation to facility costs) now.

The average number of prescriptions for children was found to be 2.1. However, only 55% of the drugs prescribed were in stock, and of these, 39% of families had to pay. Whereas charging for drugs that are not in stock is understandable, charging for those that are in stock is really a travesty of the free care policy.

Fifty-three percent could not afford to pay for their care, and this figure was higher at national hospital level. Twenty-eight percent of households had health insurance cover. Those with health insurance paid more on average than the noninsured, which was unexpected and worrying (Figure 4).

image

Figure 4. Mean costs, deliveries and child treatment episodes (SDG, by state), insured and noninsured

Download figure to PowerPoint

The most common coping strategy reported is sale of assets (27%), followed by borrowing from money lenders (18.5%), borrowing from family and friends (10%), assistance from family and friends (10%), not getting full treatment (6%), using family savings (3%), having a debt to the facility (1%) and receiving free services (0.8%). It is interesting that for a service that is meant to be free, such a low proportion report this as a way of dealing with the unaffordability of services. This correlates with the costs reported.

Deliveries

For women who had delivered, the mean payment was SDG 248 ($92.9), which is 208% of the monthly household expenditure after food. The cost in rural hospitals was twice that of national hospitals on average. Those with health insurance paid more on average (and this held true across all states and facility types, apart from rural hospitals) (Figure 4). Of the payments made, 33% were made out of the facility. Consultation charges, drugs and transport are the largest items, all at 22–23% of the total cost (Figure 5).

image

Figure 5. Breakdown of costs for deliveries (mean, SDG)

Download figure to PowerPoint

Women who had delivered had an average of 4.4 prescriptions per person. Only 42% of these were in stock. Even more worryingly, just under 54% of women reported paying for drugs that were in stock. For those not in stock, the vast majority (92%) said they would purchase them elsewhere. This was expected to add an average of SDG 45 ($16.8) to the cost of the episode.

Just under a quarter overall—24%—had health insurance, with the highest proportion being amongst those who used state hospitals (28%). However, ability to pay for their deliveries remained low—66% reported that they could not afford to pay. This was highest at lower level facilities—87.5% at rural hospitals, for instance.

Thirty percent were unable to explain how they coped with the costs in this case, but for those who did respond, getting assistance from family and friends was the most common strategy (20%), followed by borrowing from friends (16%), borrowing from a lender (16%), using family savings (6%) and selling family assets (4%). Only 1.4% could not obtain the full treatment and only 1.4% was offered free care by the facility.

The average cost per CS delivery was higher than other delivery types—SDG 362 ($135.6), as opposed to SDG 201 ($75.3), although the difference was less than you would expect, given that the bulk of other deliveries are normal, manually assisted cases. The costs of CS vary considerably, indicating a lack of predictability—the range was from SDG 54 ($20.2) to SDG 1056 ($395.5). The rural hospital was the most expensive location for CS, whereas the state hospital was for other deliveries.

We would expect that those with CS might be more likely to be unable to pay, given that this is a more expensive service. However, there was no difference in the responses to the question of ability to pay by women with CS and those with non-CS deliveries. There was a difference for the insured versus the uninsured, however—42% of those with health insurance could afford to pay, as against only 30% without. However, given that the mean payment does not differ across these groups, this may reflect ability to pay rather than cost reductions gained through health insurance membership.

Utilisation response

National free care reports indicated that between the first and second years of operation of the policy, child care cases in hospitals increased by 45% and normal deliveries by 14% (Tables 4 and 5). The increase occurred at state hospital level, according to these figures. For CS, there was an increase in both ‘cold’ (elective) and ‘hot’ (emergency) cases. The overall proportion of elective CS is worrying though—at two-thirds of the total, it looks very high for the region, suggesting either supply-induced demand and/or strong consumer preferences for CS.4 Ultrasound, for deliveries, and operations, for children, appeared to have grown disproportionately, which also raised questions about whether providers are trying to maximise profits.

Table 4. Services provided under free care for under-fives, 20082009
 20082009 
 Federal hospitalsState hospitalsTotalFederal hospitalsState hospitalsTotal% change in totals
  1. Source: adapted from Free Care Department, Federal Ministry of Health, report

Patients98,4011,607,1071,705,50833,3562,441,0022,474,35845
Surgical operations2,72112,33115,0521,80368,15569,958365
Lab tests54,6942,773,1352,827,82949,0341,894,0231,943,057−31

The completeness of Health Management Information System (HMIS) data varies by state. However, for states where the records were fairly well maintained, such as Red Sea, the trends are compatible with these national results. Under-five inpatient visits have risen considerably (by 44%) over 2007–2009 for Red Sea as a whole, although child out-patient department cases have fallen by 11%. This may in part be linked to the policy, which covers inpatient care for children (in theory), but not out-patient. Meanwhile, facility deliveries have increased by 13% over 2 years—only 10% for normal deliveries but 46% for CS. This again may link to the fact that the free care policy is focussed on CS. Forceps deliveries have reduced by 66%.

Table 5. Services provided under free care for pregnant women, 2008-2009
 20082009 
ServicesFederal hospitalsState hospitalsTotalFederal hospitalsState hospitalsTotal% change in totals
  1. Source: adapted from Free Care Department, Federal Ministry of Health, report

Normal delivery24,97985,641110,62021,024104,546125,57014
Elective CS579724,98430,781525231,70536,95720
Emergency CS425416,37720,631292023,56126,48128
Laboratories64,821981,1671,045,98875401,789,1591,796,69972
Ultrasound20,58267,28587,86783,7131,054,1091,137,8221195
Deaths2818821628528532
Ultrasound per delivery0.590.530.542.876.66.021010
Lab test per delivery1.857.736.460.2611.29.5147
Death rate0.0010.0010.0010.0020.00213

National HMIS data for CS are presented in Table 6. Although there is an overall increase of 93% in CS over 2006–2009 for northern Sudan as a whole (if the HMIS data are reliable), an analysis of year-by-year changes suggests that the increase predated the policy (with CS increasing by around 25% per year, even in the year before the policy came into place). It also shows considerable variations in pattern across the states.

Table 6. Trends in caesarean sections, northern Sudan, 2006-9
StateYear   % Change 2006–2009
2006200720082009
  1. Source: National Health Management Information System data

Red Sea672103214761667148
Gezira6753846710,98112,31782
Khartoum13,78214,58517,18519,99145
Northern State1260147023472666112
Gedaref164222042451228839
White Nile1591314338145330235
Blue Nile420574576395−6
N. Darfur2181296188037521621
S. Darfur745472291565−24
W. Darfur46853039468546
S. Kordofan102115385617505
N. Kordofan1376206026963917185
River Nile1155156422403504203
Kassala1063176124092513136
Sinar1591189118283198101
Total32,83841,16450,95363,40593
Annual change (%) 252424 

The facility survey revealed different patterns for different facilities,5 with the rural health centres, for example, increasing their child cases by more than 300% over 2007–2009 (much larger than the general increase seen in their out-patient department), whereas urban health centres appeared to have lost customers to hospitals. For the rural hospitals, there was a modest increase in deliveries (15% overall), apart from CS, which had grown strongly (99%). Overall, child care cases were also up by 73% between 2007 and 2009. The picture was similar for state hospitals, which had however seen a reduction in complicated (non-caesarean) deliveries, as had the referral hospitals. This reduction in complicated deliveries may be a recording issue and/or may signal some gaming by providers.

The national paediatric referral hospital reported a spectacular increase in child health cases, both in-patient and, more especially, out-patients. This amounts to a growth of nearly 3000% over 2007–2009. This compared with a very modest increase of 9% for deliveries at the national maternity referral hospital.

Overall perceptions of the policy

The KI were positive about the aims of the policy, while also noting considerable practical constraints and some negative effects on the health system. Only one person recommended stopping the policy; the rest had practical suggestions for how to strengthen it and also to address important related issues, such as the drug supply system.

INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

Decentralisation

The decentralisation of health to the states level not only is an important factor in explaining the local variations in implementation of the free care policy but also makes the process of funding and reporting on results more opaque. There is no integrated information at the national level on funding for health, including for free care, which includes state contributions. There are also very limited horizontal linkages with other related programmes, such as the Global Fund, which supplies first line treatment for malaria to the public facilities, and with primary health care departments.

The different forms of free care

The free care for pregnant women and under-fives has to be considered in the context of the other free care strands, especially that covering emergency care. Of the total free care budget, it absorbs only a small proportion (13% in 2009, according to our calculations, and this is probably an overestimate, as there are many other pots for free care of various types). Many of the life-saving situations, such as emergency CS and other maternal health complications, such as postpartum haemorrhage, are already covered by emergency care. In addition, there are funds for chronic illnesses, operations, treatment abroad, ‘high risk’ treatment in private facilities and rehabilitation, and investigations for the poor. The army and police have separate funds and facilities. Information on expenditure on all of these different forms of subsidy was not available.

Links with vertical programmes

At present, the free care programme maintains a list of drugs to be provided free to facilities. In addition, the Global Funds operates a vertical system for supplying TB, HIV and malaria drugs. A third system supplies free primary health care commodities, including family planning items and nutrition. The Expanded Program of Immunization is separate again.

The vertical programmes take the load off free care to some extent, especially the malaria programme, which distributes drugs through MoH facilities. The first line of treatment is provided for free to patients, although not the second line treatment for malaria. TB and HIV programmes tend to operate their own treatment centres, so in states like Khartoum, the interaction is nonexistent. Only in Red Sea was there a reported strong interaction, with free patients being treated through the HIV and TB centres.

In most states, KI reported that the free care policy and vertical programmes have no direct interaction—they operate in parallel. With other PHC and preventive health care programmes, there was no relationship, and PHC directors typically were not well informed about the free care programme.

In addition to vertical programmes, there are specific donor-funded initiatives, often in areas not covered by MoH facilities. It is not clear how information on these activities is collected—this may be through the Ministry of Economic Affairs, which takes responsibility for planning and budgeting. As the funds are spent independently, the MoF does not have information on disbursements.

Health insurance

Health insurance coverage is growing in Sudan. Although figures vary considerably, the national coverage is now estimated at 40% (Sudan NHA team, 2010). Those with health insurance still benefit from the free care programme, which is used to treat them where resources are available. This does subsidise the health insurance funds but only to the limited extent that funds and drugs are still available. For the NHI, free care could create a disincentive for people in the informal sector to join.

According to our EIs with both groups, overall the insured pay more than the uninsured, which is unexpected. This may reflect localised implementation difficulties with the NHI. In a number of states visited, relationships between the facilities, MoH and NHI were complicated, with different payment mechanisms in operation. In some cases, NHI reimbursements were not reaching the facilities reliably and regularly. Managers also complained about the low rates of pay of the NHI.

The drug supply system

Within Khartoum, the RDF provides drugs for public facilities and other outlets (e.g. People's Pharmacies). Outside Khartoum, the CMS is the main supplier, providing 60% of all drugs in Sudan. Both benefit from tax privileges and bulk-buying abroad that allow them to undercut the private sector. They aim to provide cheaper drugs, although the differences in price are reported to be declining compared with earlier periods, which may reflect reduced internal efficiency.

For free care, the CMS supplies the list that is prepared by the MoH. There are 235 items on the free drugs list. The efficiency of the drug supply system is critical for the free care policy as drugs are the single largest cost item for service users. This is confirmed by our EIs, by previous household surveys (such as Witter and Babiker (2005)) and by the current NHA, which found that drugs absorbed just under 58% of total spending on health care in Sudan in 2008. In relation to free care (for emergencies, pregnant women and children), drugs absorb about half of the overall expenditure. At the national level, the CMS claimed to be owed a very substantial amount for free drugs (about $95.5 million for all programmes) supplied in 2008.

The supply chain runs from the CMS to the state RDFs for most drugs, but in the case of free drugs, these are usually managed separately at state level rather than by the local RDF. Within the facilities then, there are multiple outlets—some free and others charged, which may be a source of risk as well as of organisational inefficiencies. Facilities have to transport the free drugs themselves, which also contributes to the difficulties of more peripheral facilities to participate in the free care policy.

There are reports by KI that the RDF does not always have essential drugs in stock and that higher cost ones have to be purchased from private outlets. In turn, the free care is reported to be undercutting RDF sales and profits, in some areas (through reduced demand), although where the CMS stocks run out, the opposite may be true as facilities (or indeed clients) cover their increased drug needs from the RDFs (with their 20% mark-up).

In our facility survey, facilities were asked about the availability, on the day of visit of the research team, of a list of drugs deemed essential for mother and child health care. The overall results confirm the findings of the EIs that there are problems with maintaining appropriate free drugs and supplies. This holds even at higher level facilities (e.g. note the absence of gloves at the national hospital).

Facilities were also asked about their purchase and sale price for a list of essential drugs for free care. What was interesting was both that they were purchasing and selling free drugs and the varying mark-up by state and drug. Many facilities in Khartoum were selling drugs at a loss, whereas in the other states, positive mark-ups were made. Analysed by facility type, there were also interesting results. Lower level facilities faced higher purchase prices (on average SDG 2.3—just under one USD—more per item, for those where prices were available, comparing health centres with national hospitals). These are passed on to the clients to some extent. On average, drug items were SDG 1.4 more for clients at health centres compared with national hospitals (although it varies considerably across items). This is bad for equity, as clients at more peripheral units are likely to be poorer on average.

The Zakat funds

The Zakat is funded from a mandatory payroll tax for all Muslim employees. There is no significant support from Zakat for the free care policy. It provides a range of support in relation to health care—mainly paying for poor families to receive health insurance cards but also for tertiary and overseas treatments.

The private sector

Purely private sector facilities do not participate in the free care policy. However, charitable hospitals and those with teaching responsibilities do offer free care, funded by the state. Potentially, the private sector could be undermined by the availability of free care in the public facilities. However, the improved amenities that private facilities are reported to have are likely to protect it from loss of business, at least in Khartoum.

The private sector mainly intersects with the public through the labour market for staff. Staff may receive two to three times the level of pay compared with the public sector, as well as other benefits, such as housing, transport and training abroad. Almost all public staff undertake dual practice, which can mitigate the negative effects for them of policies such as the fee exemptions.

DISCUSSION

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

The free care policy for pregnant women and under-fives, which was launched in 2008, has clearly suffered from a number of constraints that have led to patchy and poorly executed implementation. Chief amongst these were inadequate funding and a lack of clear specification of how the policy should be implemented. Despite this, service utilisation appears to have responded—an increase of around 45% for children's services and 14% for deliveries, according to national data. These are consistent with experience elsewhere when fee exemptions are introduced (Mwabu et al., 1995; Fafchamps and Minten, 2003; Gilson, 2003; Burnham et al., 2004; Penfold et al., 2007; Witter et al., 2010a, 2010b) and indeed with the responses in our EIs, which suggested that health care seeking is more price-sensitive for children than for deliveries. In addition, the virtual exclusion of normal deliveries from the package of care would explain the small rise. However, longer term trend analysis would be needed to attempt to attribute this robustly to the free care policy as such. It may be fair to assume that there has been some increased short-term demand due to the announcement and initial start-up of the free care policy, and debts incurred by the CMS and health facilities as they sustained implementation of the policy beyond available funding.

In general terms, the study confirms findings from other countries that exemptions policies targeted at vulnerable groups are often poorly specified, funded, implemented and monitored (Bitran and Giedion, 2003; Witter, 2009; Ridde et al., 2012). Northern Sudan seems to present an extreme case of the scenario of announcing free care without putting the relevant policies and resources in place to achieve it. As ever, though context matters—in this case, the impact of the policy is linked to the functioning of the federal system, the NHI, the drug supply (revolving drugs) system, the multiplicity of free care and vertical programmes, the imbalances in resources in the system and the mixed practice on financial autonomy of public facilities. It demonstrates the need to consider the integration of new programmes into the health system and their possible interactions. In the case of Sudan, free care has added another vertical programme into an already highly fragmented environment.

The review presented a range of options for improving the design and implementation of the free care policy. Based on this, the decision has been taken to integrate the free care with the health insurance scheme by offering free cards to poorer women and children. This will be piloted in 2012. The quick adoption of research into revised policy demonstrates the strength of an ‘embedded’ research approach, with researchers and ministry staff working in close partnership.

A recent literature review concluded that the scientific literature on fees abolition is still quite sparse and leaves a number of questions unanswered (Ridde and Morestin, 2010). More research was said to be needed to:

  • further analyse the implementation processes for abolition policies,
  • describe and compare different approaches for managing abolition,
  • calculate the real cost of abolition policies,
  • study the health expenses that households still have to assume,
  • understand the policies' effects on professional practice and the provider-patient relationship,
  • describe the effects and linkages between fees abolition and community financing systems (Bamako Initiative and health mutual organizations),
  • verify whether it really is the poorest who benefit from fees abolition and
  • study the longer-term effects.

Despite its limitations, this study has shed light on many of these questions, with the exception of the longer term effects, which cannot be assessed 2 years into a policy.

The main limitation faced by the study was that it relied on retrospective data for a policy that had been introduced 2 years earlier and on a national scale, thus making any assessment of before/after changes or counterfactuals impossible.

In relation to data collection, a number of constraints were also noted:

  • For KII, getting written reports on the free care programme was the main challenge.
  • For the EIs, gaining an adequate sample was difficult (especially for deliveries), given the low utilisation in some facilities. We therefore adopted a pragmatic approach of widening the group of facilities visited for this tool to conduct the number of interviews targeted.
  • For the EIs, some of the questions were hard for respondents to answer, notably the one on monthly expenditures by the household. The team discussed how to assist respondents to make this calculation, so that results would be more robust. Some women also needed to consult their husbands on specific cost issues, as it is often the men who pay for services. In some areas, there were language issues, with the team needing to translate questions into local languages.
  • We obtained a particularly low number of delivery EIs for Blue Nile because of low levels of facility deliveries in that state, and this has to be taken into account in the interpretation of data.
  • For facility financial analysis and costing, the gaps in financial records presented a challenge—analysis had to be based on a smaller number of facilities in the end.
  • Some methodological short-cuts were also necessitated by absence of data during the analysis of the costing data. For example, where first and second lines of treatment are reported by clinicians, there is no evidence of the balance of use of each one. We have therefore had to average the costs across both categories, which may not be fully accurate. The attempt to have expert opinion on how commonly one treatment was used versus the other was not successful, which meant that simple average costs for the drugs had to be used.
  • Secondary data were very fragmented and sometimes revealed gaps. The quality of the HMIS varied considerably across sites. At the national level, less than 40% of primary care facilities are said to be reporting, compared with 90% for hospitals. This limited our ability to rely on these secondary data sources for analysis of trends in service utilisation.

CONCLUSIONS

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

The study confirms that removal of user fees is a policy that appears to be simple but in fact is highly complex, as it involves tackling systemic issues about how resources are allocated and flow within a health system. User fees were introduced in part as a response to systemic weaknesses in the 1980s—many of those weaknesses remain unresolved. Removal of fees should be combined with—and may help to trigger—a wider set of health sector reforms.

The policy in Sudan was launched in a dynamic health sector context, with limited coherence across different initiatives. Decentralisation presents particular opportunities and challenges, as does the developing role of the NHI. Addressing equity and access issues requires ‘systems thinking’ to deal with not only the charges levied from patients but also the wider questions of how resources (including staff and infrastructure) are allocated, how drug supply systems operate, how to rebalance the health system to decrease the pressure on higher level health facilities and how to improve the quality of care (including more consistent treatment of common conditions). These go beyond free care and yet are critical to it meeting its goals.

The overall need for greater access to basic health care and greater financial protection against health care costs is clear. The substantial costs documented in the EIs raise concerns—44% of household monthly disposable income spent on just one average episode of curative care for a child under five and 208% for average delivery costs. These costs are incurred in facilities and outside: 49% of costs for children and 33% for women who delivered were external, showing that even with an effective exemptions, policy households would continue to bear significant costs. Whatever the situation prior to the policy, basic health care for these priority groups is not affordable for the majority in northern Sudan now: 55% of carers of sick under-fives and 66% of women who had delivered could not afford to pay for their care. The evidence on financial protection through the health insurance system is also not encouraging. This highlights the need to take corrective action on policy implementation and health system weaknesses identified in this study.

ACKNOWLEDGEMENTS

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES

We would like to acknowledge the work of the whole research team, which included Sally Hassan Gassim, Elsadig Eltigani, Fatima Elzahra Ismail, Hiba Nasser Eldain, Asrar Faddulelsied, Afraa Hamid Elbasha, Isra Abdemagid, Fatima Abderhamn, Mohamed Yahia, Ahmed Khalil and Khadiga Mohamed Bader. We would also like to acknowledge the guidance and support of Dr Iman Mustafa, Dr Mustafa Salih, Dr Khalid Habbani, Dr Ehssanallah Tarin and Isabel Soares, as well as the comments provided by reviewers and the time provided by all research participants. Financial support for technical assistance and fieldwork was provided by the Multi-Donor Trust Fund.

The authors have no competing interests.

  1. 1

    For further details, see (Witter et al., 2010b).

  2. 2

    It should be noted that there were considerable discrepancies between estimates provided at the national level and those reported by key informants (focal persons) and financial staff within the States. Even the latter two groups did not provide consistent figures.

  3. 3

    Conversion rate of $1 = 2.67 in April 2011.

  4. 4

    Unfortunately, we do not have trend information for elective versus emergency caesareans, so it is hard to say whether the breakdown is connected with the national exemption policy.

  5. 5

    Note however that the overall sample size was small for the facility survey—utilisation data were collected with financial data, and given the problems with data collection in Blue Nile and South Kordofan, the final number analysed was only 16.

REFERENCES

  1. Top of page
  2. SUMMARY
  3. INTRODUCTION
  4. BACKGROUND
  5. RESEARCH METHODS
  6. FINDINGS
  7. OUTCOMES
  8. INTERACTION OF FREE CARE POLICY WITH THE HEALTH SYSTEM
  9. DISCUSSION
  10. CONCLUSIONS
  11. ACKNOWLEDGEMENTS
  12. REFERENCES
  • Bitran R, Giedion U. 2003. Waivers and exemptions for health services in developing countries. World Bank: Washington, DC, Social Protection Discussion Paper No. 0308.
  • Burnham G, Pariyo G, Galiwango E, Wabwire-Mangen F. 2004. Discontinuation of cost sharing in Uganda. Bull World Health Organ 82(3): 187195.
  • Fafchamps M, Minten B. 2003. Public Service Provision, User Fees and Political Turmoil. Centre for the Study of African Economies: Oxford.
  • Federal Ministry of Health. 2007. 15-year Health Sector Strategy: Investing in Health and Achieving the MDGs (2007–2011). FMoH: Khartoum.
  • Federal Ministry of Health. 2009. Final Report on the Health Mapping Survey for the Northern States, 2008. FMoH, WHO, UNICEF, GAVI, Khartoum.
  • Federal Ministry of Health, Central Bureau of Statistics. 2006. Sudan Demographic and Health Survey. FMoH/CBS: Khartoum.
  • Gilson L. 2003. User Fees: The South African Experience. www.fic.nih.gov/dcpp/ppts/gilson1.ppt.
  • Hogan M, Foreman K, Ahn S, et al. 2010. Maternal mortality for 181 countries, 1980—2008: a systematic analysis of progress towards Millennium Development Goal 5. Lancet 375(9726): 16091623.
  • Meessen B. 2009. Removing User Fees in the Health Sector in Low-income Countries: A Policy Guidance Note for Programme Managers. UNICEF: New York.
  • Meessen B, Hercot D, Noirhomme M, et al. 2011. Removing user fees in the health sector: a review of policy processes in six sub-Saharan African countries. Health Policy Plan Supplement 2(26): ii16ii29.
  • Ministry of Health. 2009. Process and Impact Evaluation of the Removal of User Fees in the Health Sector in Zambia: Preliminary Findings. University of Zambia; Ministry of Health; LSHTM; UCT: Lusaka.
  • Molestin F, Ridde V. 2009. The Abolition of User fees for Health Services in Africa: Lessons from the Literature. University of Montreal: Montreal.
  • Mwabu G, Mwanzia J, Liambila W. 1995. User charges in government health facilities in Kenya: effect on attendance and revenue. Health Policy Plan 10(2): 164170.
  • Penfold S, Harrison E, Bell J, Fitzmaurice A. 2007. Evaluation of the delivery-fee-exemption policy in Ghana: population estimates of changes in delivery service utilisation in two regions. Ghana Med J 41(3): 100109.
  • Ridde V, Morestin F. 2010. A scoping review of the literature on the abolition of user fees in health care services in Africa. Health Policy Plan 26(1): 111.
  • Ridde V, Robert E, Meessen B. 2012. A literature review of the disruptive effects of user fee exemption policies on health systems. BMC Public Health 12(289).
  • Sudan NHA team. 2010. Sudan National Health Accounts 2008. Federal Ministry of Health: Khartoum.
  • Witter S. 2009. Service- and population-based exemptions: are these the way forward for equity and efficiency in health financing in low income countries? Adv Health Econ Health Serv Res 21: 249286.
  • Witter S. 2010. Mapping User Fees for Health Care in Low-income Countries—Evidence from a Recent Survey. HLSP Institute: London.
  • Witter S, Babiker M. 2005. Costs and Affordability of Health care in Khartoum State: Results of a Household Survey. Save the Children (UK): Khartoum.
  • Witter S, Richard F, De Brouwere V. 2008. Learning lessons and moving forward: how to reduce financial barriers to obstetric care in low-income contexts. Stud Health Serv Organ Policy 24: 277304.
  • Witter S, Adjei S, Armar-Klemesu M, Graham W. 2009. Providing free maternal health care: ten lessons from an evaluation of the national delivery exemption policy in Ghana. Glob Health Action 2.
  • Witter S, Dieng T, Mbengue D, Moreira I, De Brouwere V. 2010a. The free delivery and caesarean policy in Senegal—how effective and cost-effective has it been? Health Policy Plan 25(5): 384392.
  • Witter S, Khalid K, Hussein R, et al. 2010b. Free Health Care for Under-five Children and Pregnant Women in Northern Sudan: Progress So Far and Recommendations for the Future. Federal Ministry of Health: Khartoum.