Stages of change: A qualitative study on the implementation of a perinatal audit programme in South Africa

Background: Audit and feedback is an established strategy for improving maternal, neonatal and child health. The Perinatal Problem Identification Programme (PPIP), implemented in South African public hospitals in the late 1990s, measures perinatal mortality rates and identifies avoidable factors associated with each death. The aim of this study was to elucidate the processes involved in the implementation and sustainability of this programme. Methods: Clinicians' experiences of the implementation and maintenance of PPIP were explored qualitatively in two workshop sessions. An analytical framework comprising six stages of change, divided into three phases, was used: pre-implementation (create awareness, commit to implementation); implementation (prepare to implement, implement) and institutionalisation (integrate into routine practice, sustain new practices). Results: Four essential factors emerged as important for the successful implementation and sustainability of an audit system throughout the different stages of change: 1) drivers (agents of change) and team work, 2) clinical outreach visits and supervisory activities, 3) institutional perinatal review and feedback meetings, and 4) communication and networking between health system levels, health care facilities and different role-players. During the pre-implementation phase high perinatal mortality rates highlighted the problem and indicated the need to implement an audit programme (stage 1). Commitment to implementing the programme was achieved by obtaining buy-in from management, administration and health care practitioners (stage 2). Preparations in the implementation phase included the procurement and installation of software and training in its use (stage 3). Implementation began with the collection of data, followed by feedback at perinatal review meetings (stage 4). The institutionalisation phase was reached when the results of the audit were integrated into routine practice (stage 5) and when data collection had been sustained for a longer period (stage 6). Conclusion: Insights into the factors necessary for the successful implementation and maintenance of an audit programme and the process of change involved may also be transferable to similar low-and middle-income public health settings where the reduction of the neonatal mortality rate is a key objective in reaching Millennium Development Goal 4. A tool for reflecting on the implementation and maintenance of an audit programme is also proposed.