The impact of drug side effects on retention in HIV care is probably being underestimated, according to an interview-based study of people taking antiretroviral treatment and health care workers in six countries in sub-Saharan Africa, published in the journal Sexually Transmitted Infections.

Although antiretroviral treatment has become easier to tolerate over the past decade as drugs with fewer serious side effects have become more widely available, many people report some side-effects of treatment.

Headache, dizziness, nausea, vomiting, tiredness, lack of energy, rash and difficulties in sleeping are common side effects of many of the drugs now used in antiretroviral drug combinations.

Although these side effects are usually classified as mild or moderate in clinical studies, health care workers and patients frequently differ in their assessments of the impact of these side effects on quality of life and on adherence to antiretroviral treatment.

To find out more about the differing perceptions of side-effects among people taking antiretroviral drugs and health care providers, and the impact of side effects – and differing perceptions of side effects –  on retention in HIV care, researchers in the Bottlenecks study group carried out individual face-to-face interviews at study sites in Kenya, Malawi, Tanzania, Uganda, South Africa and Zimbabwe.

The Bottlenecks study is a larger qualitative study of the health-seeking experiences of people living with HIV in sub-Saharan Africa. It is designed to inform the development of services by identifying how the delivery of current HIV services affects linkage to care, retention in care and adherence to treatment.

The side-effects study interviewed 168 people who had taken antiretroviral treatment at some point, and 53 health care workers involved in providing HIV care. Of those who had taken ART, 38 had been lost to follow-up and had interrupted treatment.

Patients reported that they were informed about side effects before starting treatment, while health workers said that they described side effects to patients as normal in the first few weeks of treatment.

For some patients interviewed in the study, these explanations and the ongoing support offered by the clinic helped to overcome side effects:

I took these pills and they made me feel dizzy. I could not walk and would crawl to bed and I would see the house going in circles but I came back here they encouraged me to continue taking the pills. (Female, PLHIVon ART 5 years; Zimbabwe)

But for others, side-effects led to individual decisions to modify or discontinue treatment:

I started feeling dizzy. I felt it for some time but I just kept it to myself. I told nobody I realised one tablet is the one which made me have that feeling and I stopped taking it. (Male, PLHIVLTFU, Kenya)

After I had taken the ARVs, my hands had peeled and my feet I visited the doctor twice and they would say I should continue to take my medication. So I just wondered if I was safe because of the continuous reaction, so I stayed at my home and I stopped taking the medication. (Female, PLHIVLTFU, Zimbabwe)

Side effects also undermined the sense of having been restored to health by ART, or of being able to live a normal life in those who had no HIV-related symptoms prior to starting treatment.

Although health care workers were aware of the challenges that people living with HIV face in taking medication that causes side effects, they rarely mentioned the possibility of changing treatment to reduce side effects. Health care workers saw their primary role as ensuring that the patient started treatment and achieved viral suppression, in line with national and facility-level targets.

The researchers found that people living with HIV placed value on biomedical indicators of their health status, such as CD4 counts or viral load. These indicators of improving health reassured people and provided a measure of progress. The study authors question whether the downgrading of CD4 cell testing in the era of `test and treat` might affect engagement in care and say that information for patients on viral load results will be crucial as a replacement indicator.

The Bottlenecks study authors say that their findings illustrate a continuing tension between the need of health care workers “to comply with public health policies dictating the prescription of ART against the individual needs of each PLHIV […} Adopting a `one-size-fits-all` practice [to which drugs can be prescribed] could carry negative and unintended consequences as individual PLHIV miss opportunities for more comprehensive assessments of their clinical and social needs.”