Treatment Default amongst Patients with Tuberculosis in Urban Morocco: Predicting and Explaining Default and Post-Default Sputum Smear and Drug Susceptibility Results

Abstract

SettingPublic tuberculosis (TB) clinics in urban Morocco.ObjectiveExplore risk factors for TB treatment default and develop a prediction tool. Assess consequences of default, specifically risk for transmission or development of drug resistance.DesignCase-control study comparing patients who defaulted from TB treatment and patients who completed it using quantitative methods and open-ended questions. Results were interpreted in light of health professionals’ perspectives from a parallel study. A predictive model and simple tool to identify patients at high risk of default were developed. Sputum from cases with pulmonary TB was collected for smear and drug susceptibility testing.Results91 cases and 186 controls enrolled. Independent risk factors for default included current smoking, retreatment, work interference with adherence, daily directly observed therapy, side effects, quick symptom resolution, and not knowing one’s treatment duration. Age >50 years, never smoking, and having friends who knew one’s diagnosis were protective. A simple scoring tool incorporating these factors was 82.4% sensitive and 87.6% specific for predicting default in this population. Clinicians and patients described additional contributors to default and suggested locally-relevant intervention targets. Among 89 cases with pulmonary TB, 71% had sputum that was smear positive for TB. Drug resistance was rare.ConclusionThe causes of default from TB treatment were explored through synthesis of qualitative and quantitative data from patients and health professionals. A scoring tool with high sensitivity and specificity to predict default was developed. Prospective evaluation of this tool coupled with targeted interventions based on our findings is warranted. Of note, the risk of TB transmission from patients who default treatment to others is likely to be high. The commonly-feared risk of drug resistance, though, may be low; a larger study is required to confirm these findings.

Factors associated with non-adherence during tuberculosis treatment among patients treated with DOTS strategy in Jayapura, Papua Province, Indonesia

“…Similar to our study, two African studies found that younger age is associated with non-adherence [20,21]. This association was also observed in the Indian and Moroccan studies [7,19]. Other studies showed that older age was significantly related to the risk of LTFU [22,23].…”

“…Defaulting early in the treatment period has also been found in two studies from Bandung [13] and Mumbai [19]. In contrast, two studies conducted in Morocco and Southern Ethiopia observed that most LTFU occurred during the continuation phase [7,20]. The difference between these studies lies in the fact that in the two African studies drug-taking was observed directly by healthcare workers during the intensive phase, whereas in Indonesian settings, TB treatment was most frequently supervised by a family member and not by healthcare workers [10,15].…”

“…Previous studies examining risk factors for LTFU reported that low knowledge about the cause of TB, TB transmission and the consequences of incomplete treatment played a role in lower treatment compliance [14,19]. Many studies also demonstrated that inadequate knowledge of TB was associated with poor adherence [7,22–24,27,28]. In contrast to our finding, a similar case-control study conducted in urban Morocco found that TB knowledge (duration, causes, transmission, and consequence of stopping TB treatment) was not associated with compliance during TB treatment [7].…”

“…Many studies also demonstrated that inadequate knowledge of TB was associated with poor adherence [7,22–24,27,28]. In contrast to our finding, a similar case-control study conducted in urban Morocco found that TB knowledge (duration, causes, transmission, and consequence of stopping TB treatment) was not associated with compliance during TB treatment [7]. The fact that in this setting, patients had daily contact with a healthcare provider during observed treatment may explain this difference.…”

“…In 2016 the Papua province TB control program notified 9522 TB cases (case notification rate of 302.4/100,000 population) [6]. The proportion of TB cases with MDR-TB was estimated at 2% among primary cases in Papua [7,8]. The provincial TB program identified 42 MDR-TB cases among all cases notified in 2016.…”