University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Nurse versus doctor management of HIV-infected patients receiving antiretroviral therapy (CIPRA-SA): a randomised non-inferiority trial
Global Health Sciences Literature Digest
Published November 30, 2010
Journal Article

Sanne I, Orrell C, Fox MP, Conradie F, Ive P, Zeinecker J, et al. Nurse versus doctor management of HIV-infected patients receiving antiretroviral therapy (CIPRA-SA): a randomised non-inferiority trial. Lancet. 2010 Jul 3;376(9734):33-40

In Context

Expanded access to ART care in resource-poor settings will require task shifting, particularly from doctors to other health care providers and even community workers, due to shortages in human resources.(2, 3) WHO has proposed a public health approach to ART to allow for scaling up access, with standardized and simplified treatment protocols and decentralized service delivery to allow lower level health care workers to provide care.(4) This is the only randomized controlled trial from sub-Saharan Africa (of which the reviewer is aware) that evaluates whether task shifting to nurses results in differences in clinical outcomes among patients on ART. However, observational data from South Africa, Rwanda and Lesotho,(5, 6, 7) and from the UK(8) have reported successful use of task shifting for HIV care.


To compare outcomes of nurse versus doctor management of ART care for HIV-infected patients.


Two South African primary-care clinics in townships of Cape Town and Johannesburg

Study Design

Randomized, non-blinded controlled trial


HIV-infected adults


This trial was performed as part of the Comprehensive International Program for Research in AIDS in South Africa (CIPRA-SA), between 2005-2009.(1) HIV positive adults >16 yrs with CD4 counts <350 cells/ µL, or WHO stage 3 or 4, and no prior AIDS defining illness were assigned by stratified block randomization to either primary health-care nurses or doctors for standardized first line ART care. There was no blinding. Patients who were pregnant, clinically unstable, on TB treatment for <2 weeks, on complicated medical regimens, who had significant peripheral neuropathy, or were abusing drugs or alcohol were excluded. Doctors and nurses with little or no previous HIV care experience were chosen as providers; both groups received similar structured training in HIV treatment and use of ART. Regimens consisted primarily of stavudine and lamivudine, with either efavirenz, nevirapine, or ritonavir-boosted lopinavir. The primary-care providers in each group were responsible for treatment initiation, adherence counseling, follow-up visits, and maintaining clinic records. Both teams were supported by clinic nurses, community adherence counselors, and pharmacists. The primary outcome was a composite endpoint of possible treatment-limiting events including all cause mortality, viral failure, treatment-limiting toxic effects, failure to return, and disease progression (not including TB). Single drug substitution was not considered failure if treatment interruption was <42 days. Analysis was by intention to treat. The trial was powered to detect non-inferiority of the nurse vs. doctor group, with a hazard ratio (HR) of 1.4. A data safely monitoring board (DSMB), an endpoint review committee, and a data team monitored the trial.


Of 917 patients who were screened, 812 met eligibility criteria, consented and were randomized (404 to the nurse group, and 408 to the doctor group); 70% were female. Among all patients, 35% had a previous CDC AIDS-defining event (32.9% in the nurse group and 36.5% in the doctor group, p=ns); 64% had a CD4 cell count <200 cells/µL, and 57% had a viral load >100,000 copies/mL. There were no significant differences between groups at baseline. Medium follow up was 120 weeks. Cumulative treatment failure occurred among 371 (46%) patients (48% in the nurse group and 44% in the doctor group (HR=1.09, CI: 0.89,1.33). Time to death, and failure due to viral failure, toxicity and loss to follow up were similar for both groups. Viral failure contributed to 22% of composite failures, toxicity failure 36%, loss to follow-up 36%, and deaths 6%. There were 10 deaths in the nurse group (2 of which were non HIV or study related), and 11 in the doctor group (2 of which were due to lactic acidosis). No participants met criteria for disease progression. The median increase in CD4 cell count was 155 cells in the nurse group and 158 in the doctor group. The most common grade 3 and 4 toxic effects were anemia and neutropenia, raised lactate and abnormal hepatic enzymes. Due to the high frequency of hyperlactatemia (9.8-10.5 cases per 100 person-years), the DSMB recommended additional training in the management of raised lactate.


This study showed that nurse-monitored ART was not inferior to doctor-managed treatment, based on clinical, virological and immunological endpoints. HIV management by nurses could be safe and effective even for those starting therapy with advanced HIV infection.

Study Quality

This was an excellent randomized controlled trial performed in a LDC setting. Data analyses were evaluated by a DSMB and were based on intention-to-treat. The study was not blinded; this could have resulted in some detection bias for secondary endpoints. Procedures were well described.

Programmatic Implications

The findings of this study are important, as they demonstrate that primary health care nurses may be equivalent to doctors in monitoring first line ART in a public health treatment programme. However, as the authors point out, this study may not necessarily replicate typical conditions under which treatment is given, particularly as the study was a closely monitored and highly structured randomized trial, and all clinical staff received protocol-specific training. The trial endpoints were treatment failure, death or loss to follow-up, so the ability of nurses to carry patients through regimen switching was not evaluated. Also, the study cannot be generalized to settings in which multiple first-line ART options might be used to individualize patient treatment. Finally, doctors in the comparison group were not experienced in HIV care. Because of the relatively high rate of potentially serious toxic side effects, particularly due to the use of stavudine, nurses will need increased training and a clinical support structure.


  1. Comprehensive International Program for Research on AIDS in South Africa.
  2. Muula AS, Chipeta J, Siziya S, et al. Human resources requirements for highly active antiretroviral therapy scale up in Malawi. BMC Health Serv Res 2007; 7: 208.
  3. Bedelu M, Ford N, Hilderbrand K, Reuter H. Implementing antiretroviral therapy in rural communities: the Lusikisiki model of decentralized HIV/AIDS care. J Infect Dis 2007; 196: S464-68.
  4. Gilks CF, Crowley S, Ekpini R, et al. The WHO public-health approach to antiretroviral treatment against HIV in resource limited settings. Lancet 2006; 368: 505-10.
  5. Charalambous S, Grant AD, Day JH, et al. Establishing a work place antiretroviral therapy programme in South Africa. AIDS Care 2007;19: 34-41.
  6. Shumbushe F, van Griesven J, Lowrance D, et al. Task shifting for scale-up of HIV care: evaluation of nurse-centered antiretroviral treatment at rural health centers in Rwanda. Plos Med 2009; 6: e1000163.
  7. Cohen R, Lynch S, Bygrave H, et al. Antiretroviral treatment outcomes from a nurse-driven, community-supported HIV/AIDS treatment programme in Lesotho: observational cohort assessment at two years. J IAS 2009; 12: 23.
  8. Keitz SA, Box TL, Homan RK, Bartlett JA, Oddone EZ. Primary care for patients infected with human immunodeficiency virus: a randomised controlled trial. J Gen Intern Med 2001; 16: 573-82. People Living with HIV/AIDS