Monitoring of antiretroviral treatment by nurses resulted in outcomes just as good as those seen when patients are monitored by doctors in South Africa, a large randomised trial has shown.
The findings, presented on Wednesday at the Fifth International AIDS Society conference in Cape Town, are likely to reinforce calls to expand treatment capacity in South Africa by allowing nurses a much greater role in the management of patients receiving antiretroviral therapy.
Nurses and other cadres of medical staff below physician level are already taking substantial roles in staging patients for treatment and monitoring patients on treatment in several African countries, including Malawi and Ethiopia.
The policy, called task-shifting by the World Health Organization, is designed to maximise healthcare worker resources in settings where doctors are scarce.
The study was carried out by the Comprehensive International Programme for Research on AIDS in South Africa (CIPRA) an international partnership led by Professor Robin Wood of the University of Cape Town, and recruited patients at primary healthcare sites in two townships, Masiphumelele in Cape Town and Soweto in Johannesburg.
Patients were eligible to join the study if they had a CD4 count below 350 cells/mm3 or a prior AIDS-defining illness (excluding a single episode of bacterial sepsis or herpes zoster). Patients with newly diagnosed opportunistic infections were excluded, as were patients with more than six weeks of prior antiretroviral experience.
Treatment was initiated by a doctor before randomisation. The study enrolled and randomised 812 patients to receive care from either a doctor or a primary healthcare nurse after starting treatment with a first-line regimen of d4T, 3TC and either efavirenz or nevirapine.
Monitoring consisted of viral load and CD4 measurements, monitoring of clinical state for grade 3 and 4 adverse events associated with the study medications, assessment of clinical status, and measurement of height, weight and blood pressure.
Since neither doctors nor nurses had much experience of delivering antiretroviral therapy, all providers received lecture training in HIV management including toxicity and management of opportunistic infections, and instruction on the protocols in South African national treatment guidelines for monitoring and management of patients.
Patients received adherence and psychosocial support from lay community counsellors.
The primary outcome of the study was the emergence of treatment-limiting events: death, loss to follow-up, defaults on clinic appointments (included as a proxy for treatment adherence, since patients collected drugs at each scheduled visit), grade 3 or 4 toxicity that resulted in more than 42 days off treatment, or virologic failure, defined as either failure to achieve a viral load reduction of at least 1.5 log by week 12, or two consecutive viral load measurements above 1000 copies/ml after at least 24 weeks of treatment.
All the deaths and toxicity events were reviewed by an independent panel to ensure that they had been correctly assigned.
Results
There were no significant baseline differences between the two groups, although a slightly larger proportion of the group randomised to nurse care were women (73 vs 67%). Approximately two-thirds of patients had a CD4 count below 200 cells/mm3 at baseline and around 60% had symptomatic HIV disease or an AIDS-defining illness. Twenty per cent in each arm had been exposed to single-dose nevirapine for prevention of mother-to-child transmission.
After 96 weeks of follow-up there was no significant difference in outcomes between the two groups, whether analysed by all treatment-limiting events, by virologic failure, by time to loss to follow-up, by time to death or by time to treatment-limiting toxicity.
Overall the failure rate was 45%, but a large proportion of the events defined as failures were either missed clinic appointments or loss to follow-up (16.4%), or treatment-limiting toxicity (15.2%).
An analysis stratified by baseline CD4 count below 200 cells/mm3 and by viral load above 100,000 copies/ml similarly found no significant difference in outcome between the two modes of care, suggesting that more seriously ill patients did not do worse when monitored by a nurse rather than a doctor.
The only difference that emerged was a higher rate of reporting of grade 3 and 4 toxicities that resulted in dose alterations or drug switches by doctors. These were driven almost entirely by d4T (stavudine) and nevirapine (liver toxicity, hyperlactatemia and lipodystrophy).