Non‐adherence to tuberculosis treatment

Marais et al¹ reported on a prospective community‐based study into adherence of mono‐chemotherapy with isoniazid in preventing tuberculosis in the suburbs of Cape Town, South Africa. They studied children who were less than 5 years of age and who had household contact with an adult pulmonary tuberculosis index case. The main evaluation was the children's prescription collection at various intervals from the start of chemotherapy to the end at 6 months. They found that only 20% of the children completed ⩽5 months of unsupervised chemotherapy. The ethical underpinning and structure of this study is questioned. It was clear that during the early stages—that is, at 1 and 2 months of the study—a number of children were not compliant with the therapy. At this stage, the guardians were not questioned regarding the reasons for non‐attendance, nor educated on the importance of compliance. Furthermore, no arrangements were made for ensuring subsequent compliance. Instead, the children were simply monitored for further non‐compliance, and were ”allowed” to develop tuberculosis; which is what six of them did. One would assume that normal professional practice would be that whenever any non‐compliers are located, a strict form of supervision or directly observed therapy be instigated—that is, before any disease progression occurred. Also, the authors did not investigate the reasons for non‐compliance. It is not possible that the guardians were too sick with their pulmonary tuberculosis or HIV (not mentioned in the study, but assumed to be endemic in the suburbs of South Africa) to bring their children regularly to the clinics. On the basis of anecdotal experience of non‐compliance, we use dual therapy with rifampicin and isoniazid. As soon as any non‐compliers are located, intensive supervised treatment or directly observed therapy is instigated.