Without facts, nobody really knows

Wednesday, 21 December 2011

South Africa, like most other countries, has a routine notification system for reporting certain medical conditions. The 2003 Health Act lists the conditions that need to be notified to the National Department of Health, how soon after diagnosis, and the information required for each condition.

The notification system helps the Health Department and other relevant authorities to monitor disease trends over time, which in turn allows for evaluation of the effectiveness of promotional and intervention strategies.

Thirty-three medical conditions are notifiable – Anthrax, Brucellosis, Cholera, Diphtheria etc.

So, here’s a question for all of you: how many people in South Africa are infected with HIV?

The short answer is that no one really knows.

Continue Reading Below

This is because in South Africa, HIV infection is not regarded as a notifiable medical condition. Therefore, when any person gets tested for HIV infection, the results of that test cannot be disclosed to anyone other than the person being tested or the person conducting the test. So the results of such tests are never submitted to any central statistical repository.

What this means is that any estimate one might see as to infection rates in this country is a guesstimate – based primarily on the infection rates among expectant mothers passing through public health care facilities.

So, for example, if one out of four such mothers is HIV positive, statisticians use formulae to extrapolate that into figures for the rest of the population.

And so we end up with widely reported “facts” such as “South Africa has the highest rate of HIV infection in the world”.

Now for those of us who try to look for common sense solutions to problems, the obvious question we ask is “Why is HIV not a notifiable medical condition”?

The short answer is that nearly all notifiable conditions have the potential for very quickly spreading through the rest of the populace and can nearly always be contained by immediate action such as quarantine. This is clearly not the case with HIV as proximity to an infected person is not dangerous. (I, for one, hug my HIV-positive friends without the slightest concern.)

But there are many people who believe that HIV should be a notifiable medical condition because this allows for tracking both prevalence and incidence of the virus. (Prevalence is the number of people infected. Incidence tracks the number of new infections over time.) This is particularly important for assessing how effective strategies and policies are in combatting the spread of infection.

There are many countries that have made HIV notifiable including Australia and the United States.

In 1997, then Health Minister Nkosazana Dlamini-Zuma declared that Aids should be made an anonymously notifiable disease – “anonymous” meaning numbers would be collected but names of infected people would be kept confidential. The Minister said the purpose would be to collect information on how many people have AIDS or have died from AIDS, how AIDS manifests itself, or was distributed in the population.

The information would be used for “surveillance of the disease, identification of risk factors, planning of prevention, treatment, supply of medicines as well as monitoring the epidemic”.

Now this might seem very logical, but the proposal was never turned into policy. This is because Aids activists fought against its implementation.

To quote a 2000 UNAIDS report: “Central to the opposition were human rights issues. AIDS notification, it was stressed, would occur in a context characterized by continued discrimination and stigmatization.

There were doubts about whether the proposed anonymity of reports would in fact provide the confidentiality such systems are designed to provide. As a consequence, notification would discourage individuals from coming forward to clinical care.”

And so, we have the situation today where we as a nation are being called upon to spend billions on providing treatment for infected people without being able to accurately determine whether we are winning the war.

But consider this: the cost of treatment per person on anti-retrovirals runs to US $129 per month [Martinson N et al. Outpatient provider costs of treating adults with ARVs in Soweto, 2006]. If only 10 percent of our population is infected and needs treatment, that’s 5 million times R1 000 per patient per month or R60 billion per year. That’s one Arms Deal every six months.

And if we as a nation are investing that type of money, we had damned well better be sure it’s a good investment.