I spent some time attending sessions that looked at the changes in mortality since HAART has become widely available. The first study presented looked at rates and causes of mortality among platinum miners in South Africa in a workplace HIV program. Testing, care, and ART is free and available for all workers and their families. This study focused on miners and semiskilled workers. Time was divided into pre-HIV (1992-40), HIV/pre-ART (1995-2003), and post ART (2003-2008).
The 1995-2003 HIV, pre ART era saw an increase in all cause mortality driven entirely by HIV.
The 2003-2006 era saw a drop in all cause mortality related to an increase in HIV AIDS-specific mortality, but did not come back down to pre-1995 rates.
The 2006-2008 era saw some rise again in HIV (and therefore all-cause) mortality.
In the post ART era, the higher percentage of deaths among HIV-positive individuals should be examined with the data related to TB in that era. It was not clear why this occurred, although the presented hypothesis may have been related in part to improved diagnostics.
Among the HIV patients who died:
• 54% never registered for treatment
• 30% were treated and died
• Remainder died before initiating ART
The study shows that using ART at a population level resulted in reduction in mortality largely driven by drop in HIV-related causes of mortality, but even in the setting of free workplace treatment, there is still a need to address barriers to individuals coming into treatment at all and coming into treatment earlier.
A second study in the US was through UCSF’s public clinic. Data presented covered in the study looked specifically at socially disadvantaged patients eligible for ART. The study examined rates and causes of mortality; deaths were determined through social security death registry (only 10% did not have a social security number). Findings showed that 41% of deaths were still due to AIDS, despite individuals eligible for ART being linked to care. Among the socially disadvantaged, there was higher mortality significance for IDU in 2005-2009 and no difference in 2000-2004. A surprisingly small % proportion died due to liver disease.
Among those overall who died, only 42% EVER achieved viral suppression (vs 69% among survivors); however researchers were not able to determine if the patients who died had ever been started on ART, or if their deaths were due to poor adherence to treatment from art initiation. These data highlight a need for ongoing work to strengthen multidisciplinary approaches to ensure that patients remain in care and are supported to start and to adhere to treatment even in settings where care and treatment is free.
The question of when to start ART was a direct or indirect focus of a number of sessions. The IAS-USA Treatment Guidelines were released during the conference, supporting earlier initiation of treatment with ART recommended for all patients with CD4 counts < 500, selected clinical conditions, and all symptomatic patients (provided patients are ready to start therapy: “The patient must be ready and willing to adhere to lifelong therapy,” the document explicitly states. Additional populations for ART initiation included pregnant women, and those with acute primary infections. The document is well worth a good read for its careful review of the data behind the pendulum swinging to earlier initiation in 2010.