Use of predicted vital status to improve survival analysis of multidrug-resistant tuberculosis cohorts.

BACKGROUND: Multidrug-resistant tuberculosis (MDR-TB) cohorts often lack long-term survival data, and are summarized instead by initial treatment outcomes. When using Cox proportional hazards models to analyze these cohorts, this leads to censoring subjects at the time of the initial treatment outcome, instead of them providing full survival data. This may violate the non-informative censoring assumption of the model and may produce biased effect estimates. To address this problem, we develop a tool to predict vital status at the end of a cohort period using the initial treatment outcome and assess its ability to reduce bias in treatment effect estimates. METHODS: We derive and apply a logistic regression model to predict vital status at the end of the cohort period and modify the unobserved survival outcomes to better match the predicted survival experience of study subjects. We compare hazard ratio estimates for effect of an aggressive treatment regimen from Cox proportional hazards models using time to initial treatment outcome, predicted vital status, and true vital status at the end of the cohort period. RESULTS: Models fit from initial treatment outcomes underestimate treatment effects by up to 22.1%, while using predicted vital status reduced this bias by 5.4%. Models utilizing the predicted vital status produce effect estimates consistently stronger and closer to the true treatment effect than estimates produced by models using the initial treatment outcome. CONCLUSIONS: Although studies often use initial treatment outcomes to estimate treatment effects, this may violate the non-informative censoring assumption of the Cox proportional hazards model and result in biased treatment effect estimates. Using predicted vital status at the end of the cohort period may reduce this bias in the analyses of MDR-TB treatment cohorts, yielding more accurate, and likely larger, treatment effect estimates. Further, these larger effect sizes can have downstream impacts on future study design by increasing power and reducing sample size needs.

Aggressive Regimens Reduce Risk of Recurrence After Successful Treatment of MDR-TB.

BACKGROUND: We sought to determine whether treatment with a "long aggressive regimen" was associated with lower rates of relapse among patients successfully treated for pulmonary multidrug-resistant tuberculosis (MDR-TB) in Tomsk, Russia. METHODS: We conducted a retrospective cohort study of adult patients that initiated MDR-TB treatment with individualized regimens between September 2000 and November 2004, and were successfully treated. Patients were classified as having received "aggressive regimens" if their intensive phase consisted of at least 5 likely effective drugs (including a second-line injectable and a fluoroquinolone) used for at least 6 months post culture conversion, and their continuation phase included at least 4 likely effective drugs. Patients that were treated with aggressive regimens for a minimum duration of 18 months post culture conversion were classified as having received "long aggressive regimens." We used recurrence as a proxy for relapse because genotyping was not performed. After treatment, patients were classified as having disease recurrence if cultures grew MDR-TB or they re-initiated MDR-TB therapy. Data were analyzed using Cox proportional hazard regression. RESULTS: Of 408 successfully treated patients, 399 (97.5%) with at least 1 follow-up visit were included. Median duration of follow-up was 42.4 months (interquartile range: 20.5-59.5), and there were 27 recurrence episodes. In a multivariable complete case analysis (n = 371 [92.9%]) adjusting for potential confounders, long aggressive regimens were associated with a lower rate of recurrence (adjusted hazard ratio: 0.22, 95% confidence interval, .05-.92). CONCLUSIONS: Long aggressive regimens for MDR-TB treatment are associated with lower risk of disease recurrence.