Table 5. Observational studies of mid-level health workers’ effectiveness,1996–2010.

Study	Country	Study design	Health workers	Objective	Outcomes
Chilopora, 200770	Malawi	Prospective cohort study	Health workers: clinical officers compared with medical officers (i.e. doctors). Training: clinical officers were trained locally for 3 years. After a 1-year internship, they were licensed to practise independently. Responsibilities: performing major emergency and elective surgery. The Government of Malawi has been training clinical officers since 1974.	To determine the extent of major surgical work carried out by clinical officers and medical officers in Malawi and to assess the quality of surgical care from postoperative outcomes.	Health-care outcomes, including morbidity
McGuire, 200871	Malawi	Cohort study	Health workers: clinical officers compared with nurses and medical officers. Training: not reported. Responsibilities: not reported.	To identify innovative, alternative and complementary means of delivering ART that can be used to scale up access to treatment.	Health behaviour, such as adherence to treatment, and health-care outcomes, such as mortality
McCord, 200973	United Republic of Tanzania	Retrospective cohort study	Health workers: assistant medical officers compared with medical officers. Training: assistant medical officers were selected from practising clinical officers on the basis of recommendations and examination results. They received another 2 years of clinical training, including 3 months on surgery and 3 months on obstetrics, during which they were expected to have carried out at least five caesarean sections. After graduation, they were licensed to practise medicine and surgery. Responsibilities: practising medicine and surgery. The United Republic of Tanzania started training assistant medical officers to perform caesarean sections and other forms of emergency surgery in 1963.	To assess the quality of care provided by assistant medical officers (i.e. non-physician clinicians); a prospective review was carried out of all patients admitted with obstetrical complications to district-level hospitals in two regions.	Health-care outcomes, including mortality
Gimble-Sherr, 200874	Mozambique	Retrospective cohort study	Health workers: clinical officers versus medical officers. Training: not reported. Responsibilities: not reported.	To evaluate the quality of care provided by clinical officers (tecnicos de medicina) who initiated ART and followed up patients.	Health behaviour, such as adherence to treatment
Labhardt, 201075	Cameroon	Uncontrolled before-and-after study	Health workers: non-physician clinicians compared with medical officers. Training: non-physician clinicians were trained in the same way as medical officers and took on many of their diagnostic and therapeutic functions.	To assess the feasibility and effectiveness of systematically integrating hypertension and diabetes care into the primary health-care services provided by 75 facilities staffed by non-physician clinicians in eight rural districts of Cameroon.	Change in systolic and diastolic blood pressure
Pereira, 199672	Mozambique	Prospective cohort study	Health workers: surgical technicians compared with obstetricians. Training: surgical technicians underwent a 3-year course on the principles of surgery and anaesthesiology and surgical techniques and methods. In Mozambique, government training of surgical technicians began in 1984.	To evaluate the outcomes of caesarean delivery, with particular attention to postoperative complications.	Health-care outcomes, including morbidity

ART, antiretroviral therapy.