Exploring the use of social network analysis to measure communication between disease programme and district managers at sub-national level in South Africa

Highlights

• Assesses degree to which programme and district managers collaborate in South Africa.

• Makes novel use of social network analysis to explore communication on HIV monitoring.

• Shows weak communication regarding data use for HIV programme monitoring in districts.

• Provides quantifiable measures of communication that can be compared across districts.

• Identifies potential interventions to promote collaborative HIV programme monitoring.

Abstract

With increasing interest in maximising synergies between disease control programmes (DCP) and general health services (GHS), methods are needed to measure interactions between DCP and GHS actors. In South Africa, administrative integration reforms make GHS managers at decentralised level (district managers) responsible for the oversight of DCP operations within districts, with DCP managers (programme managers) providing specialist support. The reforms necessitate interdependence, but these actors work together ineffectively. Communication is crucial for joint working, but no research to assess communication between these actors has been done. This study explores the use of social network analysis (SNA) to measure the extent to which programme and district managers in South Africa communicate, using HIV monitoring and evaluation (M&E) as an exemplar. Data were collected from fifty one managers in two provinces during 2010–2011, to measure: a) one-on-one task-related communication – talking about the collation (verification, reporting) and use of HIV data for monitoring HIV interventions; and b) group communication through co-participating in management committees where HIV data are used for monitoring HIV interventions in districts. SNA measures were computed to describe actor centrality, network density (cohesion), and communication within and between respective manager groups. Block modelling was applied to identify management committees that connect respective manager groups. Results show HIV programme managers located at higher level communicated largely amongst themselves as a group (homophily), seldom talked to the district managers to whom they are supposed to provide specialist HIV M&E support, and rarely participated with them in management committees. This research demonstrates the utility of SNA as a tool for measuring the extent of communication between DCP and GHS actors at sub-national level. Actions are needed to bridge observed communication gaps in order to promote collaborative monitoring of HIV programme interventions within districts.

Introduction

Health systems in many low- and middle-income countries (LMICs) have been characterised by tensions between disease control programmes (DCPs) that address specific diseases (vertical approach) and general health services (GHS) that cater for a wider range of diseases (horizontal approach) (Mills, 2005). The number of DCPs in LMICs has increased significantly in the last two decades, in the wake of unprecedented increases in funding for the control of priority diseases such as HIV and tuberculosis by Global Health Initiatives (GHI) (Brugha, 2008). Tensions arise when dedicated systems established for DCPs – for example drug delivery, finance, or human resources – run parallel to GHS and cause fragmentation in the delivery and management of health services (Marchal et al., 2009, Travis et al., 2004). Various ways of addressing fragmentation are proposed – for example through better integration (Unger et al., 2003) or the so-called ‘diagonal’ approach (using disease interventions to drive system-wide improvements in GHS) (Sepúlveda et al., 2006).

Integration is sometimes viewed as integrating the delivery of two or more DCPs at the point of care (e.g. HIV within family planning services) (Spaulding et al., 2009). However it can also refer to providing DCP interventions within multi-function general services (operational integration) and integrating the management of DCP operations within GHS management (administrative integration) (Unger et al., 2003). DCPs can also be integrated within GHS governance, planning, financing, and monitoring and evaluation (Atun et al., 2010). In this paper we focus on administrative integration, which Unger et al. (2003) conceptualise as transferring authority for managing DCP operations to GHS middle managers, and re-defining DCP middle managers' roles to providing technical support rather than managing operations. This notion of integration aims to enhance coherence in service delivery without losing specialisation. Therefore it does not mean abolishing DCP actors – who are needed to ensure technical efficacy – but rather requires them to support and work collaboratively with GHS actors (Unger et al., 2003). The collaboration spectrum has been characterised as ranging from no integration, through communication, co-operation [joint work], formal collaboration, to full integration (Konrad, 1996). Using a case study of HIV monitoring and evaluation (M&E), we analyse communication between DCP and GHS managers in South Africa, where health reforms support administrative integration.