Examination of attendance patterns before and after introduction of south africa's policy of free health care for children aged under 6 years and pregnant women

BMJ 1997; 314 doi: (Published 29 March 1997) Cite this as: BMJ 1997;314:940
  1. David Wilkinson, specialist scientista,
  2. Marlene E Sach, nursing sisterb,
  3. Salim S Abdool Karim, directora
  1. a Centre for Epidemiological Research in Southern Africa Medical Research Council PO Box 658 Hlabisa 3937 South Africa
  2. b Hlabisa Hospital, P/Bag X5001, Hlabisa
  • Accepted 29 November 1996


President Nelson Mandela's first major policy announcement after election was that all health care for children under 6 years and for pregnant and lactating women would be free in the South African government's health service. Financing mechanisms that balance equity and efficiency in health care are under vigorous debate.1 To contribute to this debate we evaluated the impact of this new, free care policy on rural mobile clinic services in a health district with a relatively well developed public health service.

Subjects, methods, and results

A mobile clinic team in the Hlabisa health district of KwaZulu/Natal was chosen because the communities served are representative of the whole district and record keeping is excellent. The team visits most clinic points monthly and sees 100-150 patients each day. Most patients are pregnant women receiving antenatal care, children for immunisation and growth monitoring (“under 6 clinics”), and those requiring treatment for illnesses. Around 80% of patients seeking treatment are children. Under 6 clinics have always been free and are included here as a control group. The new policy on free care was implemented in July 1994; until then pregnant women paid 5 rand (£1; $1.60) per antenatal visit and treatment services cost 3 rand.

Quarterly data for clinic attendance were extracted from registers. To determine the impact of free care we compared the mean number of new clients registering for each service (antenatal care, under 6 clinic, treatment services) and the total number of attendances during the 30 month period before the new policy on free care (January 1992 to June 1994) with the corresponding numbers during the 18 month period after free care (July 1994 to December 1995). We also compared proportions of children referred to hospital during these two periods and compared 100 consecutive women registering for antenatal care in early 1994 with the same number in early 1995 for age, gravidity, and gestation at registration.

We found no important changes in attendance patterns before the implementation of the new policy on free care. The number of both new registrations and total visits to the under 6 clinics did not change significantly after the new policy was implemented (table 1). Although the average number of women registering for antenatal care each quarter was similar before and after care became free, the total number of visits each quarter increased. At the same time mean gestation at registering decreased from 28 weeks in 1994 to 26 weeks in 1995 (P=0.018), while age and gravidity did not change. Most significantly, the number of new patients registering for treatment services and the total number of such visits increased substantially after the new policy on free care was implemented. The proportion of children referred to hospital decreased.

Table 1

Changes in attendance patterns after implementation of new policy on free health care. Values are mean (SD) numbers of visits per quarter unless stated otherwise

View this table:


Free care substantially increased the use of treatment services by children but not the use of preventive services (under 6 clinics), which have always been free. The number of pregnant women registering for antenatal care did not increase, probably because most already received it.2 A reduction in the gestational age at registration is to be welcomed. The new policy on free care was widely popular with communities and achieved some equity by improving access to the treatment services. The reduced referral rate to hospital suggests that the increased workload was due to presentation of either earlier or milder forms of illness. If the decrease is due to earlier presentation of serious illness it is to be welcomed; if it is due to presentation of self limiting illness–previously discouraged by a fee–then the policy is potentially counterproductive. Staff in clinics and hospitals were overworked and stressed by the increased workload–a workload widely perceived by them as largely unnecessary. The demand for treatment services is probably almost limitless and may, if not controlled, gradually steer important healthcare resources away from preventive services and health promotion.


Funding: South African Medical Research Council.

Conflict of interest: None.


  1. 1.
  2. 2.
View Abstract