Quality goes down as quantity goes up

Quality goes down  as quantity goes up

“When you know better you do better”

Maya Angelou, celebrated civil rights activist and poet.

Do we know what is going on with South Africa’s antiretroviral treatment (ART) programme? Can we do better?

Despite the fact that this is the world’s largest antiretroviral treatment (ART) programme, detailed data on its performance are hard to find, and evaluations on the quality of the existing data even harder. When attempted, use of external data sources to verify plausibility of reported statistics provided inconsistent results[1].

In short, in order to scale up without messing up, increased attention is needed towards quality monitoring and evaluation. At the very least we should know how many were started on ART, how many are retained, and how many of these have an undetectable viral load.

So what do and don’t we know about the largest antiretroviral programme in the world?

According to the District Health Barometer, at the end of March 2013, approximately 2.1 million (33% up from 1.4 million in 2012) adults and 148,000 children under the age of 15 were receiving ART in South Africa. Of the adults, more than 700,000 were in the five high burden districts of eThekwini, Johannesburg, Ekurhuleni, Tshwane and Ehlanzeni[2].

The report acknowledges that these indicators do not give clear indications of enrolment, coverage or adherence. It also acknowledges the potential for overestimation of retention in care, as it depends on reporting of patient outcomes (such as deaths and losses to follow-up) by health facilities, which is often incomplete.

Optimistic estimates, mostly from research cohorts, tells us that approximately 1 in 5 patients has left the programme (i.e. is lost to follow-up) three years after starting ART[3,4,5,6]. Data on the proportion of patients lost to follow-up in South Africa’s ART programme is not publicly available to date.

And maybe most importantly, we don’t know how many of the patients currently on ART have an undetectable viral load, the ultimate measure of treatment success. With a detectable viral load, patients not only deteriorate immunologically and clinically, but also develop drug resistance and have a higher risk of transmitting HIV to others.

An analysis of five South African public sector programmes with research support estimated that between 15% and 19% of patients were failing first line ART at five years on treatment[7]. This might be an underestimate given that in at least one district up to 50% of patients on ART were found to have a detectable viral load in 2012[8]. Again, no national data on virological outcomes is publicly available at the moment.

Why should we be concerned about retention in care?

Quoted figures on retention in care in South Africa often compare unfavorably with those of its less resourced neighbors. In a systematic review among Southern African HIV programmes overall retention was found to be 76.8% at 12 months, with 5 of the 7 South African sites included having lower retention than this. In addition, trends of retention over time often appear to be falling year on year.

Earlier initiation leads to lower mortality, thus we would expect retention in care to improve, yet we see declining retention due to increasing losses to follow-up. Whilst we would hope for programmes to gradually better serve the needs of patients as they mature, the reality seems to be that programmes are struggling to adapt to increasing patient burden. Quality goes down as quantity goes up. The reality might be even bleaker if we consider that programmes that do well may be more likely to publish their results.

Is the data to blame?

Conventionally we have looked at the patient path from diagnosis to retention on treatment as a straight path along which patients walk, with those walking off into the land of lost to follow up never to be seen again. The reality is more likely to be that while some patients do follow this linear journey, for many a multitude of psychological and social factors give rise to an erratic and unpredictable engagement and disengagement with multiple service providers in one or more locations over time, basically walking on and off the path.

Linear cohort monitoring systems were not designed to deal with such complexity, and although they may do so at a stretch, in many cases such patients will be lost to one system or counted as a new patient on their return, resulting in an overestimation of loss to follow up.

A second reason why systems may overestimate loss to follow up can be found in the pharmacy. Frequently many more patient supplies of drugs are dispensed in a clinic’s pharmacy than there are patients in the system. While a first reaction can be to assume that corruption or pharmacy errors are to blame, in many cases when we look in detail we find another explanation. Many patients simply get their drugs after no, or minimal contact with a nurse. This can result in their records never being recorded in the clinic’s data system.

The above issues will lead to an underestimation of patient numbers, but problems with data may also overestimate retention. Most importantly, without proper systems to remove those dead and lost to follow up from the cohort of actively followed patients, estimates of active patients in clinics without the more advanced computer systems may be closer in reality to estimates of all the patients who ever started ART in that facility. Perhaps this may even explain why some well-resourced programmes do not show greatly improved results (or even worse ones) than poorly resourced ones. They may just be better at counting who is lost.

It would be a mistake for us to look at the often disappointing data on retention in care in South Africa as “Just a data problem” though. While the gaps in the data mean that we can’t be sure what is going on, we do have enough information to make some educated guesses, and to begin to address issues identified.

Why might retention in care of patients on ART be falling?

Firstly, the health systems’ incapacity in the face of massive scale up of patients on ART. The addition of more than two million people on antiretroviral treatment in the South African public health system in the past few years is a public health intervention of unprecedented scale in history. This hasn’t been matched by equivalent increases in human and structural resources. The system has reached the limits of its capacity in its current form; it needs to transform.

The observation that rapidly growing programmes can have particularly rapid declines in retention gives some support to this hypothesis6. Pinpointing the principal limiting resource needs is challenging, but given the key role of counseling in promoting adherence and the time consuming nature of the activity, attention here is warranted.

In addition, with more patients starting ART earlier, fewer will have experienced severe disease. Without ever having such experience we can wonder whether the motivation to remain disease free and in care can be as strong.

Finally patient fatigue may be setting in. Some patients have spent a decade or more taking their ARVs, a decade of the often monthly trudge to the frequently overcrowded clinic, of queues and sometimes side effects. It might not be unexpected if some were to lose sight of the life saving power of these drugs.


In summary, the achievements of the South African antiretroviral programme are undeniable, yet much remains to be done in terms of monitoring its progress. Current numbers on treatment might be over- or underestimated, and national data on loss to follow-up and treatment failure are not publicly available. To improve outcomes, the health system has to adapt to patients’ needs and numbers. This will require a shift towards more community focused models of care. We can and should know better, but we can’t afford to wait before we start doing better.

[1] Johnson, LF. (2012). Access to antiretroviral treatment in South Africa, 2004-2011. The Southern African Journal of HIV Medicine, 13(1):22-27.

[2] Massyn N, Day C, Dombo M, Barron P, English R, Padarath A, editors. District Health Barometer 2012/13.

Durban: Health Systems Trust; October 2013.

[3] Nglazi MD, Lawn SD, Kaplan R, et al. (2011). Changes in programmatic outcomes during 7 years of scale-up at a community-based antiretroviral treatment service in South Africa. J Acquir Immune Defic Syndr. 56(1):e1-8.

[4] Boulle A, Van Cutsem G, Hilderbrand K, et al. (2010) Seven year experience of a primary care antiretroviral treatment programme in Khayelitsha, South Africa. AIDS, 24(14):563-572.

[5] Fox MP & Rosen S. (2010). Patient retention in antiretroviral therapy programs up to three years on treatment in sub-Saharan Africa, 2007-2009: systematic review. TM&IH, 15 Suppl 1, 1–15.

[6] Cornell M, Grimsrud A, Fairall L et al. (2010). Temporal changes in programme outcomes among adult patients initiating antiretroviral therapy across South Africa, 2002-2007. AIDS, 24(14), 2263–2270.

[7] Fox MP, Van Cutsem G, Giddy J et al. (2012). Rates and predictors of failure of first line antiretroviral therapy and switch to second line ART in South Africa. J Acquir Immune Defic Syndr, 60:428-437.

[8] Van Cutsem G. (2013). The next 10 years of ART scale up in Southern Africa: Scale up without messing up. Federation of Infectious Diseases Societies of Southern Africa Congress.