Министерство здравоохранения Южной Африки признает неточность данных о падении вакцинации из-за ошибок переписи

The health department has conceded that the precipitous decline in childhood vaccination rates it recently published may have been skewed by potentially inaccurate population figures provided by Stats SA. The issue goes to the heart of concerns raised by critics of Stats SA’s Census 2022, which it uses to calculate midyear population estimates. When the census was released in 2024, some of South Africa’s top demographers said it was so riddled with anomalies it was not fit for purpose and would compromise future efforts to gauge the impact of government programmes such as childhood vaccination. Last month the health department said South Africa faced an immunisation crisis, as more than a third of the country’s infants under the age of one had not received any vaccinations at all. The proportion of so-called “zero dose” babies had surged from 12% in 2017 to 36% by April this year, while coverage of the first dose of the six-in-one infant vaccine Hexaxim had plunged from 87% in 2021 to just 60.5% in 2021-25, it said. The situation put “hundreds of thousands” of babies at risk of severe and potentially life-threatening illnesses, it said, calling for urgent action to increase the vaccination rates. Several independent sources expressed scepticism about the scale of the declines described by the health department in a statement released on August 26. Wits professor of vaccinology Shabir Madhi said if the figures released by the health department were correct, it would imply a total collapse of the country’s immunisation programme. While the recent measles outbreaks were concerning, they were not significantly worse than those reported in previous years, he said. Additional data subsequently provided by the health department to Business Day, which shows the coverage rate for the BCG vaccine against tuberculosis had dropped to 66.5% in 2025, was equally questionable, he said. “BCG is given religiously before the baby is discharged [from hospital]. Unless there was a BCG stockout — which I’m unaware of — it’s pretty much impossible for BCG coverage to be 66%. There is a major data source issue,” he said. Almost all births in South Africa take place in hospitals. Immunisation coverage is calculated by dividing the number of doses provided to children in a specific age group (the numerator) divided by the size of the target population (the denominator). Experts raised concerns about the health department’s data sources for both the numerators and denominators used in its calculations. Rather than capturing the details of each vaccinated child on an electronic registry, the health department used the number of doses procured by provinces as a proxy, said Madhi. UCT epidemiologist Leigh Johnson said there was considerable uncertainty regarding the true trend in birth rates in recent years and that the denominators used in the vaccine coverage calculations might therefore be inaccurate. “It’s not just an issue with the census, as it’s not clear what the trend in births has been in the postcensus period,” he said. In response to Business Day’s questions, the health department said it was confident that its data for the number of vaccinations provided in the public sector was correct, but there was “some uncertainty regarding the population figures provided by Stats SA”. It’s not just an issue with the census, as it’s not clear what the trend in births has been in the postcensus period —   Leigh Johnson, UCT epidemiologist “The department has embarked on a process of reviewing the accuracy of these figures, as this may affect coverage estimates,” it said. It also said its figures did not reliably include vaccinations administered in the private sector. “In some instances, children are vaccinated in the private sector using public sector vaccines — these vaccinations are included. However, vaccinations funded through medical insurance or through out-of-pocket payments are not included,” it said. Given the uncertainty in the estimates of the vaccine-eligible population, the health department’s figures were not a particularly useful indicator of immunisation coverage, said Tom Moultrie, director of the Centre for Actuarial Research at UCT. He expressed concern about the lack of reliable independent data sources for the number of births each year, which could be used to estimate the vaccine-eligible population. The data was previously published in the demographic and health survey conducted by the health department and in the national community survey run by Stats SA, but the most recent time these surveys were done was in 2016, he said. “The failure of the 2022 census to produce useable data on fertility, which was never released by Stats SA, has compounded our uncertainty,” he said. The lack of empirical data also made it impossible to estimate the extent to which births were under- or overreported to the department of home affairs. “The considerable uncertainty in the number of births carries significant implications for national service delivery and improving the lives of all people in South Africa. “In addition to allowing us to better understand trends in infant, child and maternal mortality, the numbers of births directly determine the numbers of children enrolling in school some six years later … and (in the longer term) the future demographic trajectory of the South African population,” he said. At the time of publication, Stats SA had not responded to Business Day’s request for comment. Business Day