Resource Requirements for Community-Based Care in Rural, Deep-rural and Peri-urban Sub-districts.  A comparative analysis in 2 districts in Gauteng and KwaZulu-Natal

Resource Requirements for Community-Based Care in Rural, Deep-rural and Peri-urban Sub-districts. A comparative analysis in 2 districts in Gauteng and KwaZulu-Natal

This study was conducted by the MRC and RHAP to compare the resource implications (both financial and human resources) of the CHW components of WBOT between rural and urban communities. The study focuses on different community-based service models implemented in two provinces to highlight potential differences between urban and rural sub-districts. This is done with a view to informing various WBOT policy options currently under consideration.

Download a copy of the findings here: ReportCHWCosting_MRC_RHAP_1010-17


To understand different models of community-based care being implemented in South Africa currently

To compare time requirements for CBS delivery between rural, deep-rural, and urban/peri-urban sub-districts.

To compare profiles and level of activity between categories of sub-districts.

To measure current expenditure and HR deployment for the delivery of Community-Based Services (CBS) in different categories of sub-districts.

To determine the adequacy of the observed level of activity by comparing to modelled target level of activity taking into consideration population structure and burden of disease.

To identify resources implications of the target level of activity per category of sub-district.

To establish a formula for resource requirements tailored for each type of sub-district.


Community health workers (CHWs) traditionally played a role in supporting the delivery of high priority interventions relating to HIV, TB and maternal and child health. While they are essential for these programmes and improving access to care, there has been significant inconsistency in the management and functioning of the programmes across South Africa. This inconsistency is partly because the role of CHWs articulated poorly in National or Provincial health policy and thus limited guidance on the qualification requirements, training, employment conditions, the scope of practice or the basic role of CHWS within the health system. In addition, very little attention has been paid to the differences in needs of CHWs’ services between urban and rural areas.

This study was conducted to compare the resource requirements (both financial and human resources) of the CHW component of Ward Based Outreach Teams (WBOTs) between rural, deep-rural and urban/peri-urban communities in order to inform policy. Two districts were selected from with a sample of WBOT teams covering different types of areas. CHWs filled detailed diaries for two consecutive weeks (10 days) reporting on travel time, and activity time per type of activity, recipients visited and conditions covered. This recording resulted in 2,511 entries in uMzinyathi (16% in urban/peri-urban areas) and 4,140 in Sedibeng (68% in urban/peri-urban areas).

While travel time across rural and urban settings within each district did differ between types of sites it was not the most significant predictor of differences in time utilization. In Sedibeng, median travel time to clinics/CHCs was twice as long in rural than urban areas, with no difference in the travel time for home visits, while in uMzinyathi, median travel time for home visits was a third longer in rural areas and 20% longer to travel to clinics/CHCs. Time on activities themselves showed more significant differences with overall median time by unit of activity being 15% longer in rural than urban areas in Sedibeng and 10% longer in uMzinyathi respectively. While the median time per home visit in uMzinyathi was 50% longer in deep-rural areas than urban areas, home visits median time in Sedibeng was 20% longer in rural than urban areas.

Referral rates following home visits were alarmingly low in both districts, around 4% in Sedibeng in both types of areas, and 2% in uMzinyathi’s urban areas, but significantly higher in deep-rural areas, at 12%.   This low referral rate raises questions about the quality of screening by CHWs and the quality of supervision.

Expressed in cost per capita total population, expenditure on WBOTs amounted to R47 per capita per year in Sedibeng and R28 in uMzinyathi, translating to under 4% of the PHC expenditure per capita of around R1, 200 in both districts—a clear sign of underfunding of this platform of service. If the CHWs monthly stipend was increased to the level of the national minimum wage (R3,500), based on the existing number of CHWS, WBOTs cost per capita in Sedibeng would stand at R59 (up from the current R47). uMzinyathi shows very similar figures with WBOTs per capita expenditure amounting to R48, up from R28. Increasing stipends to the level of the national minimum wage would still amount to a small proportion of PHC expenditure per capita: 4.9% of PHC expenditure per capita in Sedibeng and 4,2% in uMzinyathi.

The number of home visits per capita, a pointer to the coverage level, was low in both districts, taking into account their specific demographics and burden of disease, compared to the number expected as per the 2012 report to the national Department of Health. In addition it was lower in rural areas at 0.4 home visit per capita per year in both districts, compared to 0.7 and over in urban/peri-urban areas The expected number with full coverage would have been 1.2 home visits per capita in Sedibeng and 1.5 in uMzinyathi.

Modelling the number of CHWs required if each of the two districts was 100% urban or 100% rural and using each district’s travel and activity time data, we found that, based on the Sedibeng profile, a rural (farms) district would need 33% more CHWs than an urban district. Using the profile of uMzinyathi, a deep-rural district would require 62% more CHWs than an urban district. If a CHW covers 250 households in an urban/peri-urban site, a CHW in a rural site should be allocated 169 households and 96 in a deep-rural site.

To maximise the potential of this layer of service delivery, there is a need to address the low level of resourcing of this platform as evidenced by the very small share of PHC expenditure it represents. Investments need to be made in the supportive supervision to ensure improved referral processes and good working morale. Improved staff morale will require much more than adequate lines of supervision though. There is a need to properly budget for the formal integration of this cadre in the health system that will allow for benefits (such as a pension), job security and opportunities for job progression.

Importantly, effective budget planning will need to recognise that different geographical areas have different needs regarding community-based services. With travel and home visits in rural and deep rural areas being significantly longer than in urban areas in addition to CHWs spending a smaller share of their time in home visits, there is a need for increased human resource allocations in rural areas compared to urban areas.

The following recommendations emanate from this report

  • Type of sub-districts (urban/peri-urban, rural, deep-rural) must be taken into account for work and resource allocation:
    1. Compared to urban/peri-urban areas, 33% more CHWs are required in rural areas to cover the same size of population
    2. Compared to urban/peri-urban areas, 62% more CHWs are required in deep-rural areas to cover the same size of population
  • The number of households per CHW must be adjusted per type of sub district:
    1. In urban/peri-urban areas 250 households per CHW
    2. In rural areas: 169 households per CHW
    3. In deep-rural areas: 96 households per CHW
  • Overall the level of resourcing of CBS should be reassessed as it represents a very small proportion of PHC expenditure (under 4%), given the role the CBS platform is expected to play in the continuum of services
  • CHWs’ stipends need to be standardised across the country, and should as a first step be aligned to the national minimum wage.
  • Further research should be conducted to assess the reasons behind differences in duration of home visits and possibly differences in content between types of sub-districts.