Tuberculosis treatment default in a large tertiary care hospital in urban Nigeria: prevalence, trend, timing and predictors

Objectives: Few studies have investigated tuberculosis treatment default in tertiary care settings. We aimed to determine the prevalence, trend, timing and predictors of defaulting from tuberculosis treatment in a Nigerian tertiary hospital.

Methods: Data entered from 2006 to 2010 in the Federal Medical Centre, Abakaliki, tuberculosis treatment register were sorted into six treatment outcomes. Five outcomes were combined into one variable called 'non-defaulters' and were compared with "defaulters". The statistical analysis was conducted using SPSS.

Results: Of 671 tuberculosis patients, 192 (28.6%) defaulted. Of these, 126 (66%) were ≥30 years old, and 115 (60%) had pulmonary tuberculosis. Furthermore, 106 (55%) were males, and 125 (65%) lived in a rural area. The annual proportion of defaulters dropped from 34.8% to 20.6%, but the decreasing trend was not statistically significant (P=0.132 for trend). Of the defaulters, 148 (77.1%) defaulted during their intensive phase of treatment. The median default time was 7 (IQR 5-8) weeks. The independent predictors of treatment default were older age (aOR 1.5), rural residence (aOR 2.3), and HIV seropositivity (aOR, 2.8).

Conclusion: TB treatment default is high and must be reduced. This may be achieved through improved rural DOT, further patient education, and enhanced coordination of TB/HIV care.