Millions of people are being treated for HIV – why are so many still dying?

Twenty years ago, treatment for HIV was a rare luxury in South Africa. Excessive costs and the fierce opposition of President Thabo Mbeki to antiretroviral treatment (ART) have kept it out of the public sector.

Those were terrible days. Many lives have been lost.

The environment has changed remarkably since then. The turning point came in 2004 when, after four years of struggle, led by the Treatment Action Campaign, the government reluctantly agreed to start providing ART.

Antiretroviral coverage of people with HIV in South Africa has increased from 0% in 2000 to 71% in 2019. The South African antiretroviral program is now the largest in the world, with more than five million people being treated and increasing. HIV-linked deaths have decreased from 150,000 in 2000 reached a peak of around 300 000 in 2006 – up to 72 000 in 2019.

But deaths have not decreased as much as hoped. HIV remains a major cause of death in South Africa. Many people are still in health care facilities with advanced HIV disease. And AIDS remains a major contributor to hospitalizations and deaths in Africa.

Worldwide, 690,000 people died of HIV in 2019.

Médecins Sans Frontières (MSF) supports hospitals in South Africa, Guinea, the Democratic Republic of the Congo (DRC), Malawi and the Central African Republic which still treat large numbers of people with AIDS. Because people with very advanced HIV disease occur, up to one in three die during their hospital stay.

One of the biggest challenges remains that diagnostics and medicine are not readily available to people suffering from advanced HIV. This group of people is very vulnerable to deadly opportunistic infections such as tuberculosis (TB), meningitis and serious bacterial infections.

It all shows that the world is very far from the end of AIDS.


For the past ten years, the focus has been on diagnosing people with HIV and starting them with treatment. Efforts around the test-and-treat approach have been mobilized around the UNAIDS 90-90-90 targets: 90% of people with HIV know their status; 90% of those known to have antiretroviral therapy; and 90% of those on antiretroviral drugs have an undetectable viral load.

It is necessary, but it is not enough to address HIV-related deaths. Lifelong treatment requires lifelong support. Some people will interrupt the treatment; some struggle to take their tablets every day, and are at risk of developing drug resistance and failing to treat.

Today, most people with advanced HIV do not succeed or have interrupted treatment. In two MSF-supported studies in the DRC and Kenya only 20% -35% of inpatients with advanced HIV were ART naive (have never had access to treatment) and more than half of those on ART failure of treatment.

The reality of treatment interruption and treatment failure requires a new approach.

This is why MSF launches Welcome Back Services in Khayelitsha, Cape Town. The services focus on the needs of patients returning to care and those not being treated. Stigmatization and accusing patients of interrupting or failing treatment is common. This leads to delays in seeking care, and patients who present themselves as false-naive – patients who re-test for HIV and hide the fact that they have been in treatment before.

This in turn leads to patients in more advanced stages of the disease occurring or being inadequately treated.

This is one of the reasons why HIV is still demanding too many lives. Very late-onset patients often have severe immunosuppression, multiple concomitant life-threatening diseases, and significant organ damage due to HIV itself. Treatment is complicated by the need for many different medications, with a higher risk of drug interactions and serious side effects. Even with intensive care, which is not available in most institutions, many patients die.

TB is the leading cause of death among people living with HIV in confined spaces. It is estimated that TB is responsible for approx. 50% of deaths. Two other main causes are cryptococcal meningitis, which is responsible for one in five HIV deaths, and serious bacterial infections.

Together, these infectious diseases cause more than two-thirds of HIV-related deaths. All three are preventable and treatable – if detected early enough.

No time to lose

There are immediate steps that can be taken.

There are more options than ever before to prevent TB disease. New testimony shows that shorter treatments of rifapentine and isoniazid, weekly for three months or daily for one month, are equally effective in treating latent TB and reducing mortality compared to the older regimen of isoniazid for six to 36 months. And a recent trial showed that a four-month treatment with a new treatment was just as effective as the current six-month treatment for the treatment of active TB disease.

If left untreated, the chances of cryptococcal meningitis surviving are zero. But cryptococcal meningitis can be prevented and there is progress in treatment. In some countries, daily fluconazole is recommended for the prevention of a first episode, and everywhere as a secondary prophylaxis to prevent recurrent diseases. Treatment with flucytosine and amphotericin B reduces mortality by 40%. This medicine is still lacking in many – if not most – health structures in Africa.

Steps can be taken to prevent the death of advanced HIV. This includes earlier detection at the primary care level – before patients are so serious that they are admitted to hospital. The longer the delay with diagnosis and treatment, the smaller the chance of survival.

This is where CD4 tests and rapid tests for TB and cryptococcal meningitis are life-saving.

What is urgently needed to save lives is accelerated access to a package of care for the prevention, diagnosis and treatment of advanced HIV at primary and hospital level, along with clear-cut strategies to reduce AIDS deaths.

Gilles van Cutsem, Honorary Research Fellow, Center for Epidemiology and Research on Infectious Diseases, University of Cape Town


Leave a Reply

Your email address will not be published. Required fields are marked *