AARTI KISHUNA*

Divided into nine provinces, South Africa has a population of approximately 45 million¹, about 80% of whom rely on public sector health services. The rest use the private sector, which includes a very small private-not-for profit element. In 1996 South Africa introduced its Na- tional Drug Policy,² and also adopted the Essential Drugs Programme, currently only implemented in the public sector.³ Only drugs registered by the South African Medicines Control Council can be used in the country.

Patients who attend public sector hospitals are charged a fee that is inclusive of the drugs dispensed, while at primary level patients are treated free of charge. The majority of patients attending private sector facilities have medical insurance. Most health care providers are contracted to these insurance companies and make direct claims to them for services provided, including the drugs that are dispensed.

The pilot study on drug pricing took place in the Province of KwaZulu-Natal, which has a population of approximately 8 million.¹ In the public health sector there are approximately 300 clinics⁴ and 60 hospitals.¹ There are also 28 private hospitals¹, and the majority of private sector dispensers are either retail pharmacists or dispensing doctors. The flow-chart below shows the main drug pricing survey activities undertaken in KwaZulu-Natal.

Data were collected at selected facilities from the public and private sectors in four geographical areas. As the private-not-for profit sector is small it was excluded from the study. Public sector hospitals were chosen from a list of facilities, according to the area. The private sector included the nearest retail pharmacy and dispensing doctor within a five kilometre radius from the public facility. The study also included a private hospital in each geographical area.

Drug procurement in the public sector is through a closed tender system.

The tender process is managed nationally with input from the provinces. Once the tenders are awarded, provinces order directly from the recommended suppliers who deliver to the provincial depots. Health care facilities are charged the price of the drugs plus a fixed levy by the depot. The prices paid by all health care facilities are the same, so data were not collected at individual public health facilities but from the provincial depot. Table 1 shows the number of facilities sampled per geographical area.

Drug prices

In analysing drug prices, ratios were used, with South African prices compared with international index prices (reference prices) which had been converted into local currency. The international index prices were taken from Management Sciences for Health’s International Drug Price Indicator Guide 2000,⁴ and the Australian Pharmaceutical Benefits Scheme.⁵ Survey results are given in Tables 2-5.

From Table 2 it is clear that the branded products, in the private sector are the most expensive, with one product 119 times more expensive than the reference price. In the public sector prices are approximately 1.64 times higher than the reference price, which is to be expected as public sector prices are much cheaper than those in the private sector.

KwaZulu-Natal Table 1
Number of facilities sampled per area

 

A total of 43 private sector facilities were sampled. Data were collected from 29. The numbers in brackets indicate the number of facilities, from those sampled, that actually provided data. All the facility managers were asked for permission to carry out the survey.

The information from the private hospitals was included with the data from the retail pharmacies, for analysis, since each private hospital has a private retail pharmacy.

For the public sector data were collected from the depot.

Time constraint was the reason most often given by those facilities that did not want to participate in the survey.

Data were collected by four data collectors, all based in the selected geographical area. Data collection, in the field, was undertaken in a week.

KwaZulu-Natal Table 2
Average price difference for the three sectors compared with the reference price

The figures in brackets indicate the range of ratios that exist from the lowest to the highest. The public sector procures just one product.

KwaZulu-Natal Table 3
Public sector drug price components of amoxicillin, 250 mg capsule, 500s

KwaZulu-Natal Table 4
Private sector drug price components of amoxicillin, 250 mg capsules, 500s

 

The ex-manufacturer’s price was collected from the wholesaler. The prices quoted above in the cost column, apart from the last row, do not include VAT. However, each supplier, from manufacturer to wholesaler to retailer, charges 14% VAT on their mark-ups. These are then claimed back from the receiver of revenue. (8 Rand = US$1 approximately).

KwaZulu-Natal Table 5
Number of days required to work to afford a course of amoxicillin treatment

Price composition

One product, amoxicillin 250 mg capsules, pack size 500, was followed through the distribution chain, in both sectors, to determine the components that make up the final price.

Table 4 shows that the patient ends up paying approximately 2.6 times the manufacturer’s selling price of the drug. Information on how the manufacturer makes up the price of the product, before it leaves the factory, was unavailable.

Affordability

Table 5 shows the number of days the lowest paid government worker needs to work in order to afford a course of amoxicillin treatment in the private sector. The annual wage of the lowest paid government worker is US$2793.91 (Rand 24036),⁶ and the daily wage is US$ 7.78.

Results indicate that the branded product costs the lowest paid government worker the most, while the cheapest generic costs the equivalent of half a day’s pay for a course of treatment.

Need for change

The survey results show that South Africans are paying high prices for their drugs, especially in the private sector.

The following recommendations are made in the light of these findings:

• All sectors need to improve their procurement practices. A standard reference price should be agreed upon to assist this process.

• A review of the mark-ups levied by wholesalers and retailers is needed. Also manufacturers need to be more transparent with their exit prices. Fortunately this process is presently underway in South Africa as a result of the National Drug Policy’s pricing component.

• The government needs to review its levying of VAT on essential medicines, as it is the patient who ultimately carries the final burden of this tax.

* Aarti Kishuna is a consultant on public health and a member of the Advisory Board of the WHO/HAI drug pricing project.

References

1. Day C, Gray A. Health and related indicators. South African Health Review 2001. Durban: Health Systems; 2001. Available at URL: http://www.hst.org.za/sahr/

2. National Department of Health. National drug policy for South Africa. Pretoria: National Department of Health, South Africa; 1996.

3. National Department of Health. Essential Drugs Programme, South Africa, 1998. Pretoria: National Department of Health 1998. Available at URL: www.sadap.org.za/edl/

4. MSH. International drug price indicator guide. Boston: Management Sciences for Health; 2000. Available at URL: http://erc.msh.org

5. Australian Department of Health and Ageing. Australian Pharmaceutical Benefits Scheme 010801. Woden: Australian Department of Health and Ageing. Available at URL: www.health.gov.au/pbs/

6. Pick W, Conway S, Fisher B, Kgosidintsi N, Kowo H, Weiner R. Measuring quality of care in South African clinics and hospitals. Technical Report to Chapter 14 of the 1998 South African Health Review. Durban: Health Systems Trust; 1998.

SOME PRELIMINARY FINDINGS FROM OTHER COUNTRIES

South Africa was one of a number of countries that conducted field tests as part of the WHO/Health Action International project to develop a drug pricing survey methodology. This will be a standardised method to collect and analyse medicine prices and price composition within a country at a point in time and over time, and between countries. It is felt that armed with reliable analyses of medicine prices governments, NGOs and others can better evaluate the impact of pricing policies and be in a stronger position to negotiate for more equitable prices. Below we present some preliminary findings from other countries participating in the study. In Monitor No.33 we will have a longer report on the project and on the accompanying manual, Medicine Prices: A New Approach to Measurement, which will be published in May 2003.

Measuring drug prices in Sri Lanka

The survey was carried out in four districts, with 15 pharmacies randomly sampled in each district. Price information was sought on 30 drugs, for both their innovator brand and the generic version most commonly sold in the country. There were 10 drugs on the survey list that were found in both brand and generic versions, in the targeted strength, in at least 3 or more surveyed facilities. Results in Sri Lanka reveal that the Median Price Ratio (MPR) for the 10 innovator brand name products was 4.41, while the MPR for the same 10 most sold generic drugs was 1.19, just 19% over the international price. The lowest MPR among the 10 generics was 0.14 (meaning that the local price was 14% of the international reference price), while the lowest MPR for brands was 1.46 (local price 46% over reference price). The highest brand MPR suggests that brand prices reach as high as six and a half times the international generic reference prices.

The table shows price variation across regions within Sri Lanka. These are all the targeted drug products that were found at several facilities in each of the four survey regions. Median MPRs within regions are presented. Several drug products are precisely consistent in terms of price across regions. However, for other drugs, such as cotrimoxazole, ibuprofen and nife-dipine, price differences between regions as high as 32% were found for the same product.

R. Wickremasinghe, K. Balasubramanium, U. Jayarathna, S. Jayarathna, C. Abeywardena, S. Ranwella, A. De Silva, B. Hettiarachchi.

Sri Lanka Table
Drug price variations across Sri Lanka: medians, for different geographic regions, of private outlet prices (expressed as a ratio over international generic supplier price)

 

* This column gives the relative increase going from the minimum to the maximum from the other 4 columns.

Some results from Armenia

The pilot study took place in Armenia in October and November 2001, with 40 private for-profit pharmacies sampled, 65% of them in the capital, Yerevan, and the rest in three regions (known as Marzes). Table 1 gives a regional level analysis of drug prices.

Armenia Table 1
A regional analysis of drug price ratios in the Armenian pricing survey

Relating drug prices to people’s earnings, the study found that the highest price for an average length course of a branded version of aciclovir was US$154 (see Table 2), meaning that the lowest paid Government worker in the country would have to work for 148 days to pay for the course. The same money would buy 318 kg of rice or 344 kg of sugar, enough for 10 years. In the case of the cheapest generic drug the price of a course of treatment comes down 60% not enough to solve the problem for most Armenians.

Armenia Table 2

Affordability of monitored drugs in Armenia

 

ARD = Armenian dram (national currency).

M. Aristakesyan.

The survey in Kazakhstan

In Kazakhstan the monitoring survey of prices and availability of 85 medicines in 21 pharmacies in Karaganda City was conducted from December 2000 to May 2001.

A cost calculation per unit was made, with a unit defined as a tablet/capsule, an ampule, a vial, or gram of ointment.

The Table on the following page shows that there were only four drugs with prices lower than the international median, all of which are available generically and have been used in Kazakhstan for some time. Fourteen out of 49 drugs (29%) were 100-199% higher than the international median; 21 out of 49 drugs (43%) were 200-499% higher; and 10 of the 49 drugs were 500-700% higher.

Kazakhstan Table
Median drug prices in Karaganda, compared to the international median - Average of December 2000 and May 2001

 

* Of 60 main list drugs, only 49 were in the International Drug Price Indicator Guide.

T. Nurghozian.