A formula to calculate budgetary allocations to health districts in a South African Province

C. FERDI BLOK, MONIKA ZWEYGARTH, ROBERT SUMMERS*

* Dr Ferdi Blok is Academic Coordinator MEDUNSA/Mpumalanga Province, South Africa; Ms Monika Zweygarth is a Technical Writer, School of Pharmacy, MEDUNSA and Professor Robert S. Summers is Head, School of Pharmacy, MEDUNSA, P.O. Box 218, MEDUNSA 0204, South Africa.

Budgets allocated to health districts in Mpumalanga, one of South Africa's nine provinces, still reflect the inequitable pattern of past years, with a lack of guidelines and statistical data to calculate actual requirements. As the main instrument for health care delivery in the Province, the 16 health districts receive financial allocations from the Provincial Health Department, including for their pharmaceuticals¹. Currently, these allocations do not reflect the districts' needs, as they are based on historical allocations. Three main problems prevent the allocation of equitable budgets to health districts in Mpumalanga. They may reflect a typical situation in South Africa and in many other developing countries.

Historical budgeting

The allocation of pharmaceutical budgets is mainly based on the previous year's budgets. Hence, in South Africa's case, it reflects the results of apartheid, whereby certain districts were advantaged, while others were severely disadvantaged. Health service infrastructure is still based on this outdated principle and cannot be changed overnight. As all the allocated budgets are fully used, no additional funds are available to cater for the actual needs of disadvantaged districts. Equity can only be attained if a proper basis for resource allocation is developed and applied.

Right to service

According to Government policy, no patient can be turned away at the institution of his or her choice. As patients disregard district, provincial and national boundaries, population figures used at the moment do not reflect the actual number of users of services. This problem, although recognised, is not reflected at all in the allocated budgets. The lack of up-to-date and accurate statistics aggravates the present situation.

Lack of budgeting guidelines

No clear guidelines exist at the provincial Department of Health for the allocation of a fixed amount per capita (or a proportion of the budget) to pharmaceutical services.

Developing a formula

Methods to redress imbalances in the allocation of health care resources in South Africa have been described previously². In consultation with the Provincial Department of Health of Mpumalanga³, the authors developed a formula to be applied to the calculation of health care budgets for districts. The formula can be adapted to suit the allocation process for different services at provincial and national levels.

Initially this formula will not address the allocation of a budget to the provincial pharmaceutical service as a whole, but applying the formula to the current budgeted amount will give a good indication of the validity of this process. The formula is based on the area, population and patient visits of each district. A percentage score is calculated for each district in terms of its share in these three elements in the provincial total.

Each score is weighted. For pharmaceutical services, the scores were weighted as follows: area x 1, population x 2 and consultations x 5. This weighting reflects the view that for pharmaceutical services the most important factor is the number of consultations in a district, as it may directly determine the quantity of medicines used. The number of people living in the district is also important as it reflects potential need. The area of the district is relatively unimportant for pharmaceutical services, but might play a role, for example, for transport allocation.

The three weighted scores for each district are added up to give an aggregate score. This total is divided by the sum of the weighting factors to give an aggregate percentage, which represents the district's share in the total provincial budget.

This approach was applied in Mpumalanga Province. The Provincial Department of Health supplied data on the population of the health districts as at July 1999, their area and the number of consultations which took place within their boundaries in 1998. An out-patient visit to a primary health care clinic, a community health centre or a hospital was taken as one consultation, a hospital in-patient day was counted as three consultations. Table 1 shows the model calculation.

Table 1
Calculation of allocations based on area, population and consultations of health districts in Mpumalanga

 

Fields shaded grey contain numbers to be entered by the user. All other fields are calculated automatically on the basis of the values in the shaded fields.

Allowance must be made for previously disadvantaged districts' upgrading needs. Depending on the department's policy, the amount to be allocated for this purpose can be either a specific allocation for development or part of the budgeted amount for pharmaceuticals. In this model, the latter approach has been adopted. For each financial year the calculated proportion can be decreased or increased by a certain percentage. The resulting calculated proportions, which will not necessarily add up to 100, are reconverted to percentages and are applied to the total available budget to determine each district's calculated allocation.

This factor should be used in a gradual, long-term approach. Care should be taken to prevent a collapse of existing services if decreasing a district's resources, and wastage or non-usage of funds if increasing them.

Comparison with actual budget and expenditure

The School of Pharmacy at the Medical University of Southern Africa (MEDUNSA) has tested the proposed formula for pharmaceutical budgets. Data on past budgetary amounts and expenditure on pharmaceuticals for the period April 1998 - March 1999 were obtained from the Chief Pharmacist, Department of Health, Mpumalanga¹.

Overall, the provincial budget was overspent by 3%. However, there were considerable differences between amounts budgeted and spent for each health district, ranging between 36% under expenditure and 71% over expenditure, indicating an inadequate distribution of funds between districts. No controls exist to improve adherence to the budget, i.e. incentives to stay within the limit or penalties for exceeding it. Figure 1 compares amounts budgeted and spent with theoretical allowances calculated with the approach described here.

In the majority of cases, the calculated amounts matched actual expenditure more closely than the budgeted allocation. The tendency of the model to increase allocations where the budget and expenditure are low, and to decrease them where budget and expenditure are high, is clearly shown.

It must be noted that the Provincial data on area, population and consultations may not consistently reflect the same health district boundaries, which have frequently been changed in the past. Figure 1 therefore shows only a tentative result.

Figure 1 - Pharmaceuticals budgets 1998/99, expenditure 1998/99 and calculated allocation in Mpumalanga by health districts Conference calls for action on neglected diseases

A versatile model

The model we have designed and used to analyse pharmaceutical expenditure across districts in Mpumalanga clearly shows maldistribution and inequity. It also allows remedial action to be based on specific needs and parameters, by offering guidelines for equitable allocations. Hence, a relatively simple formula will enable management to allocate budgets to the districts based on actual needs, and to redress the imbalances of the past. The use of the formula will also stimulate districts to devote more effort to accurate record-keeping and enhance interaction between health professionals at different levels. As it could also tempt officials to submit inflated statistics in order to obtain more funds for the district, a control mechanism should be introduced at the same time.

Clearly, the model can be used in other environments. It is likely to be particularly useful in South Africa's provinces, as they struggle to escape the legacy of misresourcing, and aim to bring proper health care to all their people. The model may also be of use at national level, where the proposed National Health Authority may determine policy on equitable financial mechanisms for the funding of health services, inter alia⁵. As the allocation of a budget is directly related to the district's statistics, this formula will stimulate record keeping and reporting.

References

1. MEDUNSA. The impact of rationalised pharmaceutical procurement and distribution and of essential drugs programme implementation: Mpumalanga Province. Report: baseline study. MEDUNSA: School of Pharmacy; 1998.

2. Bourne DE, Pick WM, Taylor SP, McIntyre DE, Klopper JML. A methodology for resource allocation in health care for South Africa. Part III. A South African health resource allocation formula. SAMJ 1990;77:456 - 459.

3. Report. Joint Pharmaceutical Services Committee Meeting: Mpumalanga Province/School of Pharmacy, MEDUNSA. Witbank, 14 July 1999.

4. Department of Health, Mpumalanga. 4/98 to 3/99, Budget and Expenditure Analysis Report 1998/99.

5. Draft National Health Bill. Chapter 3, para 30. (2) (vi). 24 May 1998.