Improving surgical patients' safety in African hospitals: exploring healthcare professionals' experiences of Phase 1 of the '5Rs to Rescue' programme.

backgroundThe mortality rate of surgical patients in Africa is twice the global average. Pan-African studies have identified that this excess mortality is driven by deaths arising from postoperative complications. Preventable deaths can be reduced with enhanced monitoring in the postoperative period. This study reports on the first of a two-phase Quality Improvement (QI) programme, '5 Rs to Rescue', designed to reduce failure to rescue for surgical patients in Africa.

methodsUsing a QI approach, we co-produced a failure to rescue intervention with stakeholders across four African countries (Ethiopia, South Africa, Tanzania, and Uganda). These initial interventions were as follows: (i) Risk assessment using a validated tool, (ii) Recognition of patient deterioration using an early warning system (EWS), (iii) Response to patient deterioration (using locally defined protocols, (iv) Reassessment using an EWS, and (v) Reflection on care provided. These were tested and refined in clinical practice at nine hospitals through QI approaches. We interviewed frontline staff, project leaders, and QI advisors at the end of Phase 1 to learn from their experience to refine the intervention prior to Phase 2. Deductive thematic analysis was used, sensitized by constructs from the Consolidated Framework for Implementation Research (CFIR).

resultsWe interviewed 11 frontline staff, seven national project leads and five QI advisors. From the CFIR-based coding, two cross-cutting, higher-order themes were developed to describe the participant's experience. Theme 1 centred around the substantial relational work involved in leading the 5Rs project. Surgeons reported the recurrent communication required to engage, train, and motivate all the necessary stakeholders. Surgical and nursing hierarchies were poorly linked in many settings, initially hampering these relational efforts. Theme 2 centred on the complexity of the 5Rs tools being introduced into complex, and already overstretched systems. A long period was required for nurses and doctors to become familiarized with the proposed '5Rs' tools. Although their value was broadly accepted, not all tools were a good fit for all the project settings. Potential solutions to overcome barriers identified in Phase 2 included a longer QI programme, more emphasis on involving nurse leaders and selection of a less complicated 'EWS' for detecting patient deterioration.

conclusionDeveloping a multifaceted intervention that fits within complex, resource-limited contexts is a major challenge for efforts to improve patient care. This Phase 1 evaluation has provided actionable learning, targeting specific barriers, to be incorporated into an improved failure to rescue QI programme being developed in Phase 2.