/ 12 June 2015

Crisis? What crisis? Africa Check tests Free State health claims

Jamaica's Usain Bolt
Jamaica's Usain Bolt

When doctors working in Free State government hospitals went public with a damning whistle-blower’s account in February, the provincial government’s spokesperson, Mondli Mvambi, responded by saying that the health system was “not collapsing”.

In a statement issued by Mvambi – and Makalo Mohale, who is the spokesperson for the Free State’s health MEC, Benny Malakoane – several claims about the health department’s “significant progress” were made. The data on which most of these assertions were based is now publicly available, and Africa Check has investigated whether the Free State government’s claims match the available data.

Background to ‘health crisis’
The Free State is South Africa’s second-smallest province by population after the Northern Cape, with a 2014 midyear population estimate of almost 2.8?million – just over 5% of South Africa’s total population.

The province has historically “feature[d] poorer health outcomes than South Africa in general” and, as far back as 1996, has consistently [been] ranked as having the lowest life expectancy at birth of all the provinces.

In November 2008, a funding crisis caused the Free State to declare a moratorium on the provision of antiretroviral drugs. The complex causes and implications of this moratorium are dealt with in detail in a report by the University of Cape Town’s Centre for Social Sciences.

The Free State “health crisis” therefore needs to be understood as one that has continued for at least two decades. Is the province’s health department making progress?

Funding
Claim: “[The] department and the Free State have not been getting the right kinds of allocations as per the national formula of the equitable share.”

Our verdict: Nonsense

“Equitable share” involves how the national government divides the tax revenue pie among national, provincial and local government. This allows provincial and local authorities to provide services such as education, healthcare, social development, housing and infrastructure.

To ensure the allocation is fair, a standard formula is used for each province. This formula has six components, starting with population size and also factoring in poverty levels, economic output, existing services and infrastructure, education enrolment and health risk profile.

The latter takes into account aspects such as the share of the population without medical aid and the distribution of chronic diseases in the province.

The equitable share formula is not perfect, a health budget specialist at the Rural Health Advocacy Project, Daygan Eagar, told Africa Check: “The health need component is largely dependent on utilisation data and does not account for unmet needs. So in rural provinces where access is more limited, need will not be accurately reflected because people use services less.”

Eagar said the formula does not adequately factor in the higher costs of providing healthcare in outlying rural areas, people moving back to a poorer province when they retire or become ill, or infrastructural backlogs. “But these problems with the formula do not explain why the Free State [health department] is performing so poorly, and the Free State is certainly not at the biggest disadvantage,” he said.

Last year the provincial treasury took over the financial management of the Free State health department, which had received qualified audits for the previous four financial years.

National treasury spokesperson Phumza Macanda explained to Africa Check that it “is illogical for the financial troubles of one department to be blamed on a formula that makes allocations across the board with respect to the provincial sphere.

“It is the constitutional responsibility of each provincial government to make budget allocations for each department’s requirements and to, in turn, ensure that financial management by each department is undertaken in an effective way.”

Immunisation
Claim: “Immunisation coverage [for children] under one year improved from 86.8% in 2013-2014 to 91.2% in 2014-2015.”

Our verdict: Cherry-picking

According to the Free State health department’s annual report for 2013-2014, 86.8% of babies aged one or younger were immunised. But this was down from 95.1% and 91.7% in the previous two years. The department admitted in its last annual report that it had not achieved its target of 95%, citing vaccines being out of stock, new data policies and poor data management as reasons.

The Free State committed to reach 90% coverage in its 2014-2015 annual performance plan with the presidency’s department of performance evaluation and monitoring. The province exceeded this in the first quarter by achieving 91.2%, as was claimed in the press statement.

Coverage for the whole year was 90.1%, according to figures that the nonprofit organisation Health Systems Trust received from the national department.

However, there are ongoing questions about the reliability of immunisation data in South Africa in general, and actual coverage is likely to be significantly lower than claimed.

Vaccines
Claim: “[In 2014-2015] coverage for the rotavirus vaccine exceeded the 95% target and the pneumococcal vaccine reached 91.5%.”

Our verdict: Cherry-picking

In its annual performance plan the Free State aimed for 95% coverage for the rotavirus vaccine and 90% for the pneumococcal vaccine.

It had exceeded these targets in 2012-2013, when coverage for the rotavirus vaccine was at 99% and that of the pneumococcal vaccine at 95%, according to the Free State health department’s 2013-2014 annual report.

It then dropped to 94.4% and 87.8% the next year. For both of these vaccines, a shortage of stock “due to poor stock management” and “reduced rural services due to reduced mobile clinics” were cited as reasons for the decreases.

For the reporting year that has just ended, figures supplied by the Health Systems Trust show that the department’s performance is approaching 2012-2013 levels, with coverage of 98.6% (rotavirus vaccine) and 91.7% (pneumococcal vaccine) – and exceeding 100% coverage in the second quarter for the rotavirus vaccine.

This “implausible” figure shows that the vaccine data “really doesn’t tell you much at all”, said National Institute for Communicable Diseases executive director Shabir Madhi. He explained that vaccine coverage is calculated by counting the number of doses administered and dividing it by the number of children of a certain age expected to be in a district, using past census data.

“One of the major challenges with the vaccine coverage data – and this is not just for the Free State but for the whole of the country – is that they don’t have accurate population [data] down to the district level,” Madhi said. To determine actual vaccine immunisation coverage, he said, a survey is needed.

Pneumonia
Claim: “Pneumonia incidence decreased from 89 per 1?000 of the under-five population in 2012-2013 to 84.7 in 2013-2014.”

Our verdict: Yes, but …

Influenza and pneumonia constitute the second-largest cause of natural deaths – after intestinal infectious diseases – in South African children under the age of 14 years.

The Free State health department’s annual reports confirm the figures stated, but the rate of 84.7 per 1?000 children under five in 2013-2014 is higher than the target of 75.

And by comparison, the national average incidence for 2013-2014 was 53.2, as reported in the Health Systems Trust’s annual district health barometer.

Tuberculosis
Claim: “The tuberculosis (TB) new client success rate improved from 79.7% to 83.3% and the TB cure rate from 73% to 77.1%.”

Our verdict: Correct, according to available data

In 2013, the leading cause of natural death in the Free State was TB, followed by influenza and pneumonia, Statistics South Africa’s last report on causes of death showed.

The department’s 2013-2014 report stated that the TB new client treatment success rate – an indicator that measures how many newly diagnosed patients completed their treatment – was, indeed, 79.7%. The cure rate given for all cases was 73.1%.

For the past reporting year, the department had committed to reaching a new client success rate of 82% and a cure rate of 75%, which it exceeded in the third quarter of 2014-2015 – achieving the rates quoted in its press statement.

The implementation of a new automated diagnostic test in 2011 has aided the fight against TB immensely, said Christo Heunis, associate professor at the University of the Free State’s Centre for Health Systems Research and Development.

His analysis of a decade of TB data in the Free State – soon to be submitted for publication – has shown a significant decrease in patients dying of TB since peaking at 12.08% of cases in 2009. One of the reasons for this is that diagnosis within two hours is now possible using special TB testing machines, which means that patients can be started on treatment the same day they are diagnosed.

Drug-resistant TB
Claim: “All multidrug-resistant TB confirmed patients have been initiated on treatment.”

Our verdict: Yes, but …

The recorded proportions of patients diagnosed with multidrug-resistant TB in the Free State who were started on treatment in the first three quarters of 2014-2015 are 100%, 100% and 83% respectively. The fourth-quarter data (also showing 100%) still needs to be validated.

But Helen Cox, an epidemiologist at the University of Cape Town who specialises in drug-resistant TB, said she has some doubts that the Free State health department was able to start all patients diagnosed with multidrug-resistant TB on treatment.

One reason is that some patients, especially those who are also infected with HIV and are diagnosed late, die before they can be treated. Another is that diagnosed patients who do not start treatment are not entered into the national recording system – a “big failing”.

Yet the government must be commended for rolling out rapid diagnosis, Cox said, and for increasing the number of multidrug-resistant TB treatment sites – in the Free State’s case, from one to 15.

“But with the new test, the number of patients diagnosed has increased dramatically while treatment has not kept pace, and that’s why we see a treatment gap,” she explained.

HIV counselling and testing
Claim: “HIV counselling and testing is implemented in all health facilities, public places and gatherings. A total of 109?047 people have been tested for HIV so far in the 2014-2015 financial year.”

Our verdict: Cherry-picking

The Free State is one of South Africa’s top three provinces with the highest rates of HIV infection.

Although the province claims to have implemented HIV counselling and testing in “all health facilities”, the health department’s own 2013-2014 annual report suggests there was confusion about the indicator. This led to it having to be recalculated to reflect HIV testing offered to those aged 15 to 49.

It should be noted that the Free State had the highest prevalence of HIV among those aged 50 years and older in 2012, and this population would have been excluded from this indicator as it currently stands.

According to previous annual reports, 534?771 “clients” were tested for HIV in 2013-2014; in the previous year 488?535 people were tested. The figure has dropped to 364?576 people for the most recent reporting year, data supplied by the Health Systems Trust shows.

It is important to remember that testing for HIV does not automatically translate into treatment. Repeated stock shortages of antiretroviral medication are of concern in the Free State.

This is at the heart of ongoing disputes between healthcare workers and organisations such as the Treatment Action Campaign (TAC), and the provincial government, which have resulted in the arrest of more than 100 community healthcare workers and TAC members in July last year.

Mother-to-child transmission of HIV
Claim: “[The] prevention of mother-to-child transmission of HIV improved from 1.8% in 2013-2014 to 0.09% in the first quarter of 2014-2015.”

Our verdict: Incorrect

The presidency’s data showed that the Free State had a transmission rate of 1.3% in the first quarter, not 0.09%.

“Nevertheless, it is an excellent HIV transmission rate and less than the [national] target for 2015, which is 2%,” said Professor Gayle Sherman, head of the Early Infant Diagnosis Unit at the National Health Laboratory Service.

Anim van Wyk is Africa Check’s deputy editor and Nechama Brodie is its head of training, research and information