A Covid-19 front that really counts: collecting the data

The government, in its fight against the coronavirus, is turning scores of state and private hospitals into early warning systems to collect data critical for the battle.

A survey by the National Institute for Communicable Diseases (NICD), which ran from March 5, when SA's first coronavirus infection was recorded, to April 29, reveals that only 21% of patients hospitalised for Covid-19 had underlying medical conditions, including hypertension, diabetes, HIV and TB.

It surveyed 116 private and 29 state hospitals. Of the 686 Covid-19 admissions, 462 patients (67%) were between the ages of 20 and 59.

In total, 27 of the 71 ventilated patients (38%) had a co-morbid condition, and 13 of the 27 patients (48%) with co-morbid conditions died.

The government has said that the majority of those who die have underlying health conditions and are elderly.

However, Professor Lucille Blumberg of the NICD, whose mandate is to conduct surveillance on communicable diseases of public health importance, said it would be premature to draw conclusions from the research so far.

She said the survey was developed to provide information on infections in hospitalised patients, to determine factors that may contribute to the severity of illness.

Asked about differences in the survey's findings and the numbers the government has released, she said because it is a snapshot surveillance programme, the survey is limited to a number of facilities where in-depth collection of data and analysis is being done.

"At this stage, reporting is preliminary. It would be premature to draw any conclusions. A number of large public hospitals admitting Covid-19 patients were enrolled in the system in the past week. We can expect to see the reports evolving," she said.

Infection discrepancies

Blumberg said it would appear that, to date, similar reasons for severe illness are present in South African Covid-19 patients as those identified in other countries.

She said going forward the survey's focus will include Covid-19 infections in health-care workers and the interaction of HIV and the coronavirus.

Health department spokesperson Popo Maja said the NICD report data represents a small segment of the total hospital data.

"Inferences for the general population should not be made as the number of public hospitals are rather small, [with] more than 80% of South Africans using public hospitals."

Paula Armstrong, FTI Consulting health economist, said for SA to get to grips with the outbreak it is vital that standard protocols exist when it comes to not only screening, testing and contact tracing, but infections reporting.

"Currently our data could be wrong as provinces have different hospital admissions, screenings and testing reporting strategies," said Armstrong.

"It can be wrong because it's limited, with only a view existing on what has been tested to date."

She said huge infection discrepancies exist between the Western Cape and other provinces, but that could be because the province is following more advanced health protocols.

Professor Martin Veller, dean of the Wits University's School of Health, said SA has "patchy data because of very patchy interventions". He said to understand the pandemic one needs to look into populations and the background of cases.

Veller said there are too many unknowns that SA is dealing with in fighting the virus.

"Any early judgment on current data is potentially dangerous. It could see resources sent to areas while there is a greater need elsewhere," he said.

He warned that although SA, with its world recognised HIV and TB data monitoring systems, is well set up for the current situation, "what we don't have, and where the problem arises, is our battle to do testing and contract tracing for an acute epidemic like this".