Scientists say the current tests designed to detect even very low levels of HIV present in the body are simply not sensitive enough.
Scientists say the current tests designed to detect even very low levels of HIV present in the body are simply not sensitive enough.

Earlier this month we published some of the damning results from a leaked report on the state of HIV testing in South Africa. Here’s the health department’s deputy director general for communicable and non-communicable diseases’ response.

A comment from the researcher who wrote the evaluation on which the story was based also appears below: “There is an urgent need to unpack the political and donor pressure to reach HIV testing service targets, which almost every site assessed cited as the primary goal.”


COMMENT

Bhekisisa’s article ‘Expired materials and fly away tents’ published on 7 October requires a response. 

In the health department’s view the article, rather than being an honest reflection on the report commissioned by the Global Fund to Fight Aids, TB and Malaria to review the quality of HIV testing services provided by the non-governmental organisations it funds, it is unnecessarily sensational. 

The article reported on an assessment of HIV testing sites in five provinces: KwaZulu-Natal (KZN), Limpopo, Mpumalanga, North West and Free State. 

While the report does not provide any information on the total number of tests that were performed in the 47 sites assessed, nationally more than 14 million tests are conducted annually. We totally concur on the need for quality testing services being provided, it is important to note the scale of the testing services in the country. It is also important to point out that every person is tested using test kits from two different manufacturers, with the second test being the confirmatory test. 

While the headline of the article, “Expired materials and fly away tents: What researchers found when they looked at HIV testing” suggested a chaotic situation with expired material and tents flying away, the report noted that of the 47 sites only one had expired stock and two sites — one in KZN and another in Mpumalanga experienced disruptions caused by the weather. The latter should not be surprising given that non-governmental organisations were required to set up facilities in various locales to take these services to communities that usually don’t receive them.  

Additionally, it is a mystery why the report had to be leaked as suggested in the article given that the report was presented to stakeholders in a dissemination workshop in June 2019.

How did the majority of sites fare? Here are the facts:

  • The majority of the sites were outside of public sector health facilities, as the services funded by the Global Fund aimed to reach people that typically do not attend clinics — men and young people.
  • The authors of the report noted in their executive summary that: “Facilities were, in the main, well run. A noticeable increase in quality of services and supervision of HIV testing services was observed in correctional centres, likely due to the highly regulated environment already in place.”
  • 45 of 46 sites were found to be clean and organised;
  • The authors noted that: “In general, facility-based services that participated in the assessment were centrally located, easy to access and had a wide range of information and materials available.”
  • In 45 of 47 sites test kits were stored in a secure lock-up;
  • All sites used nationally approved kits;
  • In 43 of 45 sites the quality control results were correctly recorded;
  • In 42 of 45 sites clients were provided with information post testing;
  • 44 of 45 testing registers were found to be up to date;
  • In 45 of 46 sites it was found that client information was kept securely;
  • Clean water and soap was available in 46 or 47 sites;
  • Personal protective equipment was available in 45 of 47 sites but used in 30 of the sites (so whilst available they were not always used), which led the authors to note that “personal protective equipment use remained significantly inconsistent”.
  • Medical waste management followed national guidelines in 45 of 47 sites;
  •  Quality control processes were carried out at the vast majority of sites, with the exception of a few sites that obtained their test kits from public health facility partners or a centralised office which conducted the quality controls themselves.

With very few exceptions, the majority of facilities performed adequately. Two areas stand out in particular that do require attention: the lack of standardised training and ensuring that sufficient time is allocated to provide information to clients that test positive and those that test negative (despite the finding that 97% of sites did provide information after conducting the HIV test).

These observations also apply to an additional, second report on the three provinces, the Eastern Cape, Gauteng and Western Cape. Bhekisisa says it did not have this report at the time of publication.

In summary this second report on sites assessed in these three provinces found the following:

  • All sites had clear evidence that testers were trained…All of the sites kept a record of initial training done in a log and in the majority of cases this log was available and seen.
  • All sites assessed had a visible and ready reference for the HTS test, either in the form of a job aide of the wall, a copy of the testing procedure pictures in a file, or the illustrated instructions from the test kit box.
  • Supervision of testers was generally done by a qualified nurse.
  •  All sites stored test kits in a locked space, with limited access to keys.
  • In most cases, test kits were only taken out of the cool box when needed, and the tester kept the cool boxes closed between access points.
  • Stock outs did not appear to be a problem, and in cases where stocks are running low it appears that sub-recipients (SR) have been able to make alternative arrangements for stock, generally from another clinic.
  • Stock management: 75%, or 30 sites, did have standard operating procedures (SOP) for stock management, although at least five sites reported that this function sat with the clinic, so the SOP is the one used by the clinic.
  • The majority of SR sites demonstrated adequate to good management of their test kit stocks.
  • In the majority of cases testers talked the client through the finger prick job aid photographs, explaining how the test would be done, how long it would take and what the positive and negative results look like.
  • The majority of sites do quality control (QC) weekly, with a minority undertaking the process less frequently…In the majority of cases the QC tester demonstrated that the test kits accurately showed the results for positive and negative serums.
  • Documentation at SR sites was found to be mainly adequate.
  • At the majority of fixed sites, the testing venue seen during the assessment was suitable for testing and counselling activities. Testing rooms had doors that could be closed to ensure privacy, test kits and PPE were available and accessible, and medical waste containers were on hand.
  • At outreach sites the assessors observed good practice in maintaining a clean testing area in some challenging circumstances.

The report’s summary is instructive in its overall finding: “Based on the analysis, the functioning of the majority of test sites and delivery of HTS services was found to be adequate across the three provinces: that is, the majority of sites scored between 5 and 8 points. This indicates that most sites comply with policy and guidelines requirements but there is room for improvement.”

Any assessment report can be read as either all bad if the reader of the report is only looking for shortcomings. But an objective reader of these reports will find both good practices as well as some shortcomings – which is usually why assessments are commissioned in the first place. Our reading of these reports, unlike that of Bhekisisa, is that much good work is being done by NGOs in reaching people who usually don’t come to health facilities to be tested — these are people that are likely to be HIV positive and who require treatment for their own health and that of their loved ones.

We always welcome criticism so that we can do better but the criticism needs to be fair and constructive rather than sensational. In this instance sensationalism may lead to people, who need to know their status, being reluctant to get tested. 

Yogan Pillay is the deputy director general of communicable and non-communicable diseases at the national health department. Follow him on @ygpillay.

Tian Johnson, author of the report on which the Bhekisisa story was based, responds:

Tian Johnson, who led the five-province evaluation detailed in the report entitled “Quality vs Quantity” on which the Bhekisisa article is based says:

The assessment reminded us that accessing HIV testing services in South Africa is a tale of two worlds with some sites performing adequately in terms of clinical quality and others performing adequately at pre and post counselling.

There is an urgent need to unpack the political and donor pressure to reach HIV testing service targets, which almost every site assessed cited as the primary goal. Similarly, there is an urgent need to address both the clinical quality of HIV testing services provided at community sites and the capacity of frontline staff to provide an HIV testing service that centres the rights and respects the individual choice of each and every client.

It should be noted that the “provision” of information to the client does not equate to rights-affirming pre or post-test counselling – something that has been proven time and time again to directly impact on repeat testing and adherence to prevention or treatment options. There is a need to move beyond the mechanical function of HIV testing services and towards a highly individualised rights-based counselling session pre and post-test. This remains a significant challenge across all sites.

The reality is that while there are centres of excellence that should be supported to share and strengthen their work, not a single site in the five provinces conducted informed consent to the satisfaction of the assessor. 

Beyond the drawing of blood at the time of the test, the harder — and admittedly more time-intensive and expensive work — is getting frontline staff to a place where HIV testing can be offered without prejudice, coercion or pressure to meet targets. 

Donors like the Global Fund should take concrete steps to ensure that quality (and the rights of clients) are not jeopardised in our quest for quantity. Both the Department of Health and the South African National Aids Council should stand firmly on the side of communities in this regard. It is affirming that the recommendations contained in the — nearly 9 months old — report have already been incorporated into the current granting cycle and that all agencies will be closely monitoring the implementation of these. 

We do not have a choice but to get HIV testing provision right and even as we rapidly scale in our quest to meet targets we need to guard against the loss of the fundamental human rights principles that have increasingly shaped South Africa’s response to the epidemic.

Tian Johnson is the founder of the African Alliance for HIV Prevention. Follow them on Twitter @tianjohnson.

Bhekisisa’s reply to Yogan Pillay’s response:  

In his response to our article, “Expired materials and fly away tents”, the director general of the national health department, Yogan Pillay takes issue with our coverage of one of two reports commissioned to evaluate the quality of HIV testing services at sites supported by the Global Fund to Fight Aids, TB and Malaria. 

Particularly, he criticises our story’s headline, which itself stems from real-world examples described in the report itself. Pillay does not dispute that these events occurred, but rather that they were isolated incidents. 

In our article, we described how the research was done, including that researchers only reported what happened to them on the day that they went to test. We stressed that this is not a large enough sample to reflect what is happening nationally. 

Did we pick shocking examples to illustrate what poor quality HIV testing looks like? Yes, but we did not invent them. 

Pillay adds that the report in question notes that “facilities, in the main, were largely well run”. Also, that “a noticeable increase in quality of services and supervision of HIV testing services was observed in correctional centres, likely due to the highly regulated environment already in place”.

Our article mentions this. It also alludes to the sentence that precedes this extract: “The assessment revealed a tale of two worlds with HIV testing services in the same district displaying a significant variation in the overall quality of HIV testing services provided.” 

And arguably, we did not use the worst examples contained in the report. 

Pillay also argues that the report used for our story was made public to stakeholders in June 2019. 

However, when the health department responded on 3 October to questions regarding this report, days before it was published on the 7th, it maintained it had not seen the document. 

We submitted questions to the department on 6 September, almost a full month before responses were returned. 

Pillay’s response also quotes from a second similarly commissioned report on three other provinces — the Eastern Cape, the Western Cape and Gauteng. 

Bhekisisa was told about the existence of this second report as referenced by Yogan Pillay during the course of our reporting. We repeatedly asked researchers for this report ahead of publication, but without success — despite repeated assertions that both these documents have already been made public. 

Last week, we were told to contact the South African National Aids Council for access to the second research document that dealt with the three provinces Pillay references.

We know that health stories can help to create demand for services and also destroy them as quickly. We were, of course, concerned that the story might negatively impact some people’s willingness to test for HIV. To try to counteract this, we stressed that this was not a representative sample and that these were indeed only instances. In follow-up media interviews, we reiterated that people should continue to have confidence in their test results because of confirmatory testing. 

We weighed this risk against concerns of some people involved in the research that there had been inaction on the report’s findings since the June meeting at which it was released to some organisations. 

Our questions to the department and others prompted a 14 October dissemination meeting. At this event, we’re told, a commitment to incorporate learnings from both reports into future Global Fund grants was undertaken. Beginning in October, regular meetings will be held to discuss progress made in improving the quality of HIV testing. 

Following the publication of our article, the national health department, the Global Fund and the national Aids council have committed to having regular meetings with journalists to ensure future studies are communicated to the public. 

Finally, we have been told that another meeting is expected in coming weeks with provincial officials to reiterate the health department’s commitment to quality, human rights-based testing as it continues to strive to ensure that people’s experience with HIV testing sets a respectful and empathetic tone for a lifetime of regular testing and — for those who need it — treatment.

In the meantime, we’ve continued to reach out to relevant researchers and organisations, as we did prior to the publication of our article, and we welcome additional feedback. 

[9:34 am 22 October 2019. This post was updated to include additional comment from the author of the report on which the story was based.]