Meet the Judges: Dr Tom Boyles, Specialist in Infectious Diseases and HIV Medicine

South Africa is in a perilous position with respect to bacterial resistance, particularly hospital acquired Gram-negative infections. Rates of Extended Spectrum Beta-Lactamases (ESBL) and carbapenem-resistant Enterobacterales (CRE) infections are very high, and prescribing practices, infection prevention and control measures are below standard in many settings.

Currently, both HIV and TB are in a relative grace period with respect to drug resistance. The introduction of new and repurposed drugs means that for the first time in many years, there is likely to be an effective regimen for the vast majority of patients. 

This will not last though, as resistance to all these drugs is inevitable or already present at low levels – we must do everything possible to prolong the lifespan of these drugs. This must not be done by restrictive prescribing, but by large scale and well-managed lay health-worker programmes to support patient adherence, the major driver of resistance in HIV and TB.

The biggest impact for a hospital clinician is undoubtedly resistant Gram-negative infections in hospitalised patients. General wisdom states that poor prescribing is a major driver of resistance, and while this is a factor, the main reason for patients becoming colonised and infected with these organisms is poor Infection Prevention and Control (IPC) procedures. I am in no doubt that while antibiotic prescribing receives a great deal of attention, it is widespread implementation of IPC that is the key to reducing morbidity and mortality from hospital acquired infections in South Africa.

It has been fascinating to watch the development and roll out of COVID tests in such a short time. For the first time I can remember, terms like PCR, lateral flow, sensitivity and specificity are openly discussed in mainstream media and around the dinner table. A particularly interesting area has been the introduction of lateral flow tests. While cheaper and faster than PCR, they have reduced accuracy. They also tend to be used on people with lower pre-test probabilities, all of which has challenged many to develop guidelines for their appropriate use. 

There are many challenges; the most important is likely to be behaviour change amongst healthcare workers and patients. As was seen with the introduction of rapid diagnostic tests (RDTs) for malaria, even if the test is accurate, clinicians in high incidence settings are often reluctant to withhold treatment for malaria for a sick patient, even when the test is negative. Patient expectations in low resource settings are also often different; it is almost unheard of for a patient to be satisfied with a consultation if they do not receive some kind of medicine. This is likely one reason for the over-prescribing of antibiotics at primary care level. This ties in with healthcare worker behaviour change, as negative tests may not be sufficiently powerful for the healthcare worker to withhold treatment if they feel pressure from the patient. 

Other challenges include the possibility that test accuracy may deteriorate if diagnostic tests that work well in high-resource settings are brought to low-resource settings. Important factors include differences in the patient population – for example high HIV prevalence. Sample collection may also not be as straightforward in low-resource settings, with the obvious example being the supply of clean containers for mid-stream urine samples.Lastly there is the challenge of human resources required to perform the test. Laboratory-based tests can be adopted for facilities with on-site laboratories but this often introduces unacceptable turnaround time in primary care. Point-of-care tests are therefore preferred in primary care but it is important not to assume that nurses have spare time to perform them. Typically, these tasks will be shifted to lay healthcare workers with lower skill levels. Currently, basic lateral flow assays such as for pregnancy (urine) and HIV (capillary blood) are feasible, but more complex tests such as urine dipsticks or Xpert MTB/RIF are not.

I would think that accuracy is the most challenging. Based on my comments above, any test aimed at primary care in low-resource settings must be as simple as a basic lateral flow assay, for it to meet requirements such as affordability and time to result (TTR). It will however be challenging to develop such tests with appropriate accuracy. Hospital level tests can be more complex and accuracy may be easier to achieve, but affordability and TTR may be more challenging.

They can’t be motivated by profit. For any test to make a difference in sub-Saharan Africa it has to be cheap to produce with a very low profit margin. Test developers must therefore be motivated in similar ways to healthcare workers, who are driven by and committed to improving the lives of others. There may be formal ways to reward them, such as monetary or non-monetary prizes, but largely, developers need to be motivated by the goal of improving other people’s lives.