BACKGROUND: Having nurses take on tasks that are typically conducted by doctors (doctor-nurse substitution, a form of 'task-shifting') may help to address doctor shortages and reduce doctors' workload and human resource costs. A Cochrane Review of effectiveness studies suggested that nurse-led care probably leads to similar healthcare outcomes as care delivered by doctors. This finding highlights the need to explore the factors that affect the implementation of strategies to substitute doctors with nurses in primary care. In our qualitative evidence synthesis (QES), we focused on studies of nurses taking on tasks that are typically conducted by doctors working in primary care, including substituting doctors with nurses or expanding nurses' roles.

OBJECTIVES: (1) To identify factors influencing implementation of interventions to substitute doctors with nurses in primary care. (2) To explore how our synthesis findings related to, and helped to explain, the findings of the Cochrane intervention review of the effectiveness of substituting doctors with nurses. (3) To identify hypotheses for subgroup analyses for future updates of the Cochrane intervention review.

SEARCH METHODS: We searched CINAHL and PubMed, contacted experts in the field, scanned the reference lists of relevant studies and conducted forward citation searches for key articles in the Social Science Citation Index and Science Citation Index databases, and 'related article' searches in PubMed.

SELECTION CRITERIA: We constructed a maximum variation sample (exploring variables such as country level of development, aspects of care covered and the types of participants) from studies that had collected and analysed qualitative data related to the factors influencing implementation of doctor-nurse substitution and the expansion of nurses' tasks in community or primary care worldwide. We included perspectives of doctors, nurses, patients and their families/carers, policymakers, programme managers, other health workers and any others directly involved in or affected by the substitution. We excluded studies that collected data using qualitative methods but did not analyse the data qualitatively.

DATA COLLECTION AND ANALYSIS: We identified factors influencing implementation of doctor-nurse substitution strategies using a framework thematic synthesis approach. Two review authors independently assessed the methodological strengths and limitations of included studies using a modified Critical Appraisal Skills Programme (CASP) tool. We assessed confidence in the evidence for the QES findings using the GRADE-CERQual approach. We integrated our findings with the evidence from the effectiveness review of doctor-nurse substitution using a matrix model. Finally, we identified hypotheses for subgroup analyses for updates of the review of effectiveness.

MAIN RESULTS: We included 66 studies (69 papers), 11 from low- or middle-income countries and 55 from high-income countries. These studies found several factors that appeared to influence the implementation of doctor-nurse substitution strategies. The following factors were based on findings that we assessed as moderate or high confidence.Patients in many studies knew little about nurses' roles and the difference between nurse-led and doctor-led care. They also had mixed views about the type of tasks that nurses should deliver. They preferred doctors when the tasks were more 'medical' but accepted nurses for preventive care and follow-ups. Doctors in most studies also preferred that nurses performed only 'non-medical' tasks. Nurses were comfortable with, and believed they were competent to deliver a wide range of tasks, but particularly emphasised tasks that were more health promotive/preventive in nature.Patients in most studies thought that nurses were more easily accessible than doctors. Doctors and nurses also saw nurse-doctor substitution and collaboration as a way of increasing people's access to care, and improving the quality and continuity of care.Nurses thought that close doctor-nurse relationships and doctor's trust in and acceptance of nurses was important for shaping their roles. But nurses working alone sometimes found it difficult to communicate with doctors.Nurses felt they had gained new skills when taking on new tasks. But nurses wanted more and better training. They thought this would increase their skills, job satisfaction and motivation, and would make them more independent.Nurses taking on doctors' tasks saw this as an opportunity to develop personally, to gain more respect and to improve the quality of care they could offer to patients. Better working conditions and financial incentives also motivated nurses to take on new tasks. Doctors valued collaborating with nurses when this reduced their own workload.Doctors and nurses pointed to the importance of having access to resources, such as enough staff, equipment and supplies; good referral systems; experienced leaders; clear roles; and adequate training and supervision. But they often had problems with these issues. They also pointed to the huge number of documents they needed to complete when tasks were moved from doctors to nurses.

AUTHORS' CONCLUSIONS: Patients, doctors and nurses may accept the use of nurses to deliver services that are usually delivered by doctors. But this is likely to depend on the type of services. Nurses taking on extra tasks want respect and collaboration from doctors; as well as proper resources; good referral systems; experienced leaders; clear roles; and adequate incentives, training and supervision. However, these needs are not always met.