BACKGROUND: The integration of HIV care into primary care services is one of the strategies proposed to increase access to treatment for people living with HIV/AIDS in high HIV burden countries. However, how best to do this is poorly understood. This study documents different factors influencing models of integration within clinics. METHODS: Using methods based on the meta-ethnographic approach, we synthesised the findings from three qualitative studies of the factors that influenced integration of HIV care into all consultations in primary care. The studies were conducted amongst staff and patients in South Africa during a randomised trial of nurse initiation of antiretroviral therapy (ART) and integration of HIV care into primary care services - the Streamlining Tasks and Roles to Expand Treatment and Care for HIV (STRETCH) trial. Themes from each study were identified and translated into each other to develop categories and sub-categories and then to inform higher level interpretations of the synthesised data. RESULTS: Clinics varied as to how HIV care was integrated. Existing administration systems, workload and support staff shortages tended to hinder integration. Nurses' wanted to be involved in providing HIV care and yet also expressed preferences for developing expertise in certain areas and for establishing good nurse patient relationships by specialising in certain services. Patients, in turn, were concerned about the stigma of separate HIV services and yet preferred to be seen by nurses with expertise in HIV care. These factors had conflicting effects on efforts to integrate HIV care. CONCLUSION: Local clinic factors and nurse and patient preferences in relation to care delivery should be taken into account in programmes to integrate HIV care into primary care services. The integration of medical records, monitoring and reporting systems would support clinic based efforts to integrate HIV care into primary care services.