INTRODUCTION:
Health policies inevitably produce unintended consequences when introduced into a health system. These consequences can be linked to policy process factors which occur during a mandate’s development and implementation. Understanding unintended consequences in relation to policy process factors can help mitigate its unintended health system impact. However, current literature reveals a paucity of information exploring this interaction. This research sought to contribute to filling this gap through the examination of the 2007 Occupation-Specific Dispensation (OSD) strategy in South Africa.

METHODS:
We conducted a health policy analysis exploring the relationship between unintended consequences and policy process factors using a qualitative systematic review approach, and applying a longitudinal lens. We included literature relating to the 2007 OSD in nursing and human resources for health (HRH) from 1994 onwards. In data analysis, we applied the Walt and Gilson (1994) Health Policy Analysis (HPA) Triangle to 1) give an overall description of the OSD policy process; 2) identify the main unintended health system consequences of the OSD policy; and 3) to investigate the relationship between the unintended health system consequences and key factors occurring throughout the policy process.

RESULTS:
Four main unintended consequences were identified: nursing demoralization; nursing division; a breakdown in the relationship between nursing managers and frontline staff and a further maldistribution of nurses. Our findings suggest that the relationship between unintended consequences and policy process factors are rooted in the historical and socio-political realities of the health system.

DISCUSSION:
Frontline-level actors can help mitigate potential unintended consequences through sharing their historical and socio-political insight. We therefore urge policymakers to include these actors in policy development and implementation processes. Future research should explore the inclusion of frontline knowledge in policy processes as a method to reduce the potential severity and/or number of unintended consequences experienced.