Time to connect the dots

In his Budget vote speech in July, Health Minister Aaron Motsoaledi said that only 50 percent of people in the public sector antiretroviral programme had received a viral load test in the past year. Of these, about 75 percent were virally suppressed. Putting these two figures together suggests that only in the region of 37 percent of patients on treatment in the public sector are known to be virally suppressed. It should be noted, though, that we understand very little about where these figures come from and how reliable they actually are.

According to Professor Wendy Stevens of the National Health Laboratory Service (NHLS), the NHLS has sufficient capacity to provide an annual viral load test for everyone receiving HIV treatment in the public sector – and are in fact doing well over two million tests per year. If the population number was to rise significantly in the coming years, as envisaged, she is confident that the NHLS would be able to scale up to meet the demand.
The low figures reported by the Minister of Health, it seems, are not caused by lab capacity, but more likely
reflect poor demand for viral load tests, or are a product of poor information systems that are not fully and accurately capturing all the viral load tests that are being done.

Even though annual tests are recommended in World Health Organisation (WHO) and South African treatment guidelines, anecdotal evidence suggests that many healthcare workers do not request these tests.

At an NSP Review session at the recent SA HIV Clinicians Society Conference in Cape Town, Dr Francesca Conradie argued that we need to build greater awareness about the importance of viral load tests. While CD4 count is an important indicator of when to start treatment, patients and healthcare workers need to understand that a viral load test is the most important test needed once someone is on treatment.

Having an electronic system with unique patient identifiers would make it easier for patients to move from one facility to another, or from a prison clinic to a clinic on the outside, without losing all their data or having to start treatment all over again.

Useful data not shared
In the previous issue of the NSP Review we argued that viral suppression rates should form a key part of how we measure the success of our AIDS response and that this data should be shared routinely. We have since learned that the NHLS does, in fact, keep track of viral load coverage and viral suppression rates by province, district and facility. We understand that this information is regularly shared with the National Department of Health. This is reassuring news.

It is, however, a concern that more of this information is not publically available. While some anonomisation of the data is essential to protect confidential patient information, this is a technical difficulty that can easily be overcome. Anonymised viral load coverage and viral suppression rates broken down by province, district and facility must be published online at regular intervals so that civil society and researchers can form a clearer picture of the epidemic. Whether or not the Department of Health makes this data accessible will be a clear test of its commitment to transparency and inclusivity.

The third 90
The third of the ‘three 90s’, which make up the the new UNAIDS treatment target, refers to 90 percent of all people on antiretroviral treatment being virally suppressed by 2030. The inclusion of a viral load target in this call is welcome, since this will help drive the rollout and uptake of viral load testing. Reporting of viral load suppression rates will give us a much better picture of how well people on treatment are actually doing – as opposed to just knowing the number of people on treatment.

While the UNAIDS viral suppression target is worthwhile in itself, it should be read in the context of the other two targets. Having 90 percent of HIV people aware of their status is an excellent target, but the aim to have 90 percent of HIV positive people on treatment is problematic. This target assumes – based on very poor evidence – that 90 percent of people who know their status need to be on treatment. (See page X for more on this issue.)

Why no unique patient identifier?
Whether it is for viral load tracking or other health records, the lack of a unique patient identifier for patients in the public sector appears to constitute a major bottleneck.

This identifier would be a number, code, or other indicator unique to a single individual – much like an ID number, and is essential to building an effective electronic health records system.

Having an electronic system with unique patient identifiers would make it easier for patients to move from one facility to another, or from a prison clinic to a clinic on the outside, without losing all their data or having to start treatment all over again. It would also make it much easier to track what is happening in the healthcare system since it would, for example, reduce double-counting of the same person at two different facilities.
Building an improved electronic record keeping system that makes use of unique patient identifiers is not a trivial undertaking. That said, with the technology at our disposal many such systems can and have been built. While the initial cost may be substantial, it seems likely that such a project will be cost-saving in the long term through increased efficiencies and facilitating improved patient management.

We need unique patient identifiers now!