HIV patients of less experienced physicians die faster according to Canadian cohort study

People with HIV whose doctor sees few other HIV-positive patients may survive for a shorter period than people with more experienced doctors, despite the availability of HAART. This is the finding of the latest analysis of participants in the British Columbia HIV cohort, reported Wednesday at the Ninth Retroviruses Conference in Seattle.

An association between physician experience and survival in people with HIV was reported by Kitahata and colleagues in the pre-HAART era, but whether this observation held true in the context of more effective therapies has been unclear.

To answer this question, Julio Montaner’s group at the British Columbia Centre for Excellence in HIV/AIDS, Vancouver, analysed 1,416 people who began three-drug antiretroviral therapy in the period between August 1996 and July 1999. An experienced physician was defined as one who had enrolled more than five people into an HIV treatment programme.

By September 30th 2000, there had been 110 deaths in the cohort which were from HIV-related causes; a crude mortality rate of 7.8% overall. However, this rate masks significant differences between subgroups. Unsurprisingly, and as the group have previously reported in a separate analysis of this dataset, there were more deaths in people who initiated HAART at lower CD4 counts.

The impact of physician experience was also substantial, particularly amongst the most immunocompromised patients. Whilst there was no difference observed according to physician experience in patients who began treatment with a CD4 count above 200 (within the follow-up period), an advantage was detected at lower levels of immunity. After twelve months of follow-up, the mortality rate in individuals with a CD4 count below 50 at HAART initiation was 2.9% in those with an experienced physician, compared to 17.6% where the physician was less experienced.

In a multivariate analysis, after adjusting for the effect of AIDS diagnosis at baseline, age, baseline viral load and CD4 count, people with an experienced physician were almost seven times less likely to die than those with an inexperienced physician, within the follow-up period.