In this study, we sought to determine the risk factors for mortality among patients started on MDR-TB treatment under programmatic resource limited settings. We employed a case-control study nested within the 2016 national MDR-TB cohort. We found that being co-infected with HIV and being non-adherent to treatment doubled the risk of death while on MDR-TB treatment while age above 50 years tripled the risk of mortality from HIV infection. Having no education was the greatest risk factor for mortality from MDR-TB, increasing the risk of death by almost four times.
The risk factors elicited in this study have been found in other evaluations of TB associated mortality4,7−10. HIV infection has been shown to significantly increase the risk of mortality from MDR-TB with this effect increasing with advancing disease7,8. Patients with CD4 cell counts < 200 cells/mm3 are four times more likely to die from MDR-TB than those with CD4 cell counts > 200 cells/mm3 7. In our study, patients co-infected with HIV were twice as likely to die during treatment than patients without HIV infection. Although more than half of HIV + patients enrolled in this study had a CD4 cell count < 200 cells/mm3, the increase in mortality was less than has been previously documented, probably due to the widespread use of antiretroviral therapy. All patients in this study received antiretroviral therapy within the first month of MDR-TB treatment.
The standard second line therapy for MDR-TB used in this study was a minimum of 20 months and consisted of six months of injectable medicine6,11. Adherence to this regimen has been shown to be suboptimal globally with over one third of patients started on treatment being nonadherent to therapy8. In our study, 55% of patients were nonadherent to the TB treatment. In our context, reasons for nonadherence to this treatment regimen include long duration of injectable medicine, lack of transport to health facilities for daily DOT, and adverse drug reactions to some of the drugs used in the regimen.
Older age has been associated with increased mortality from TB due to atypical presentations, increasing co-morbidities and frequent drug related adverse events12,13. In our setting, older age has also been shown to be associated with decreased access to healthcare services. The recently completed national prevalence survey found that one of the largest prevalence to notification gaps was among persons 50 years and older2. Older persons are also less likely to be able to have the afford transport fares for daily DOT to the treatment initiation sites or the nearest health facility making them susceptible to suboptimal adherence to MDR TB treatment.
In our study, having no education was the strongest risk factor for mortality during MDR-TB treatment. The association between education and good health is well established14,15. Globally, well-educated persons are less likely to be unemployed, more likely to have higher incomes, and more likely to have healthy lifestyles 14,15. In our setting, lower education levels are associated with unemployment, poorly paid work, and low social economic status16. Low social economic status has been associated with an increased likelihood of TB and HIV infection and with poorer outcomes from both diseases17,18. In addition, accessing diagnosis and treatment for MDR-TB has been associated with catastrophic costs to patients and their families19,20 which are likely to severely affect patients with low socioeconomic status making it difficult for them to adhere to daily DOT and refill visits.
Our study had several limitations. The use of routinely collected data resulted in missing information. However, this was minimized by triaging several data sources. The study also collected some variables from patient files which were self-reported, e.g., the alcohol use and use of recreational drugs. It is likely that some of these variables were prone to information bias as patients would have been reluctant to report undesirable behavior to their healthcare providers. Finally, the study population was chosen from a national MDR-TB cohort which had a relatively significant proportion of patients who were lost to follow-up during treatment. It is likely that a proportion of these patients could have died and that therefore the cases were underrepresented in the study. However, the national programmatic management of drug-resistant TB (PMDT) team made a team to interview relatives of patients lost to follow-up and record any known additional deaths that may have occurred. However, this was an evaluation of a national MDR-TB treatment cohort, and the findings of this study therefore reflect of the risk factors for mortality in patients managed for MDR-TB in routine care settings.