Routinely collected data estimate mortality in South Africa at five years on ARV treatment to be 10%, however it is generally accepted that routine data under-estimate mortality after starting treatment.

Using the death registry and by following up on a cohort (group) of patients who had been lost to follow-up and could be linked to the death registry via their identity numbers, researchers were able to statistically adjust the estimates, reaching a more reliable mortality estimate of 18% five years after starting ARV treatment.

‘€œThese are figures we would have happily settled on when we started out on the treatment programme,’€ said Dr Andrew Boulle, one of the members of the research team.

Boulle said the good outcomes would also have been influenced by the fact that patients were presenting to health facilities in a less severe state than when the treatment programme started.

The data shows that in 2001/2 the average CD4 count (measure of immunity) of a patient presenting for treatment was 55. In 2008/9 this figure had increased to 125.

South Africa’€™s HIV treatment guidelines stipulate that all patients presenting with a CD4 count below 200 should be placed on ARV treatment. Indications are that the country will soon bring it in line with the World Health Organisation guidelines indicating treatment initiation at a CD4 count of 350.

‘€œAmid the anxiety around the programme and the perceived chaos because of the high number of patients it is re-assuring to know that by going to the death registry we can ascertain that although we are missing   a lot of the deaths, that the programme has resulted in substantial and durable benefits to those accessing it,’€ said Boulle.

A comparable cohort in Entebbe, Uganda who had not had access to ART showed a survival rate of 8%, five years down the line.

The South African research included 48 124 adults who started on ART in 8 cohorts between 2001 and 2009.

The average age was just over 34 years, with 65% of those in the cohort women.