Risk Factors and Leprosy Incidence among Contacts in Bangladesh: A Multilevel Analysis

Unnati Saha, Department of Public Health, Erasmus Medical Center Rotterdam
Abu Sufian Chowdhury, The Leprosy Mission International, Bangladesh
Johan Chandra Roy, The Leprosy Mission International, Bangladesh
Khorshed Alam, The Leprosy Mission International, Bangladesh
Daan Nieboer, Department of Public Health, Erasmus MC, University Medical Center Rotterdam
Renate Verbiest-Richardus, Department of Infectious Diseases, Leiden University Medical Center
Annemieke Geluk, Department of Infectious Diseases, Leiden University Medical Center
Jan Hendrik Richardus , Department of Public Health, Erasmus MC, University Medical Center Rotterdam

The single center cluster randomized controlled Maltalep trial in Bangladesh assessed whether single-dose rifampicin (SDR) given after bacillus Calmette–Guérin (BCG) vaccination was able to prevent possible excess leprosy cases due to BCG in contacts of newly diagnosed leprosy patients. Simultaneously, a (non-randomized) non-intervention cohort of new patients was followed to establish incident cases among their contacts. The Maltalep trial included 1,553 index patients. Of these, 14,986 eligible contacts were randomized into two arms of the trial. The non-intervention cohort included 554 index patients and 4,216 eligible contacts. We confirmed the protective effect of BCG vaccination of contacts in preventing leprosy. The SDR intervention was effective for the contacts of MB patients, smear positive index patients, and contacts of index patients that are blood related in the second degree (e.g. cousins, etc.). Genetic relationship is a more profound risk factor for leprosy in contacts than being a household contact only. Unobserved heterogeneity in leprosy outcome influences the effect of risk factors, such as the age of contacts and genetic relationship of contact to the index case. Therefore, to address the omitted variable problem, it is important to adjust for unobserved heterogeneity in the outcome analysis.

Keywords: Health and Morbidity, Randomized controlled experiments , Econometrics , Multi-level modeling

See extended abstract.