Incentivising doctor attendance in rural Bangladesh: a latent class analysis of a discrete choice experiment

Objective Doctor absenteeism is widespread in Bangladesh, and the perspectives of actors involved are insufficiently understood. This paper sought to elicit preferences doctors over aspects jobs rural areas Bangladesh that can help inform development packages policy interventions may persuade them stay at their posts. Methods We conducted a discrete choice experiment with 308 across four hospitals Dhaka, Bangladesh. Four attributes postings were included based on literature review, qualitative research consensus-building workshop policymakers key health-system stakeholders: relationship community, security measures, attendance-based policies incentive payments. Respondents’ choices analysed mixed multinomial logistic latent class models used simulate likely uptake under different packages. Results All significantly impacted doctor (p<0.01). Doctors strongly preferred facilities where there was supportive community (?=0.93), considered good attendance education training (0.77) or promotion decisions (0.67), functional (0.67) higher payments (0.5 per 10% increase base salary). Jobs disciplinary action for poor disliked by respondents (?0.63). Latent analysis identified three groups who differed jobs. Scenario modelling intervention differentially behaviour combinations could feasibly improve doctors’ attendance. Conclusion Bangladeshi have strong but varied overcome absenteeism. generated evidence suggesting considering perspective themselves result substantial reductions Designing take account situations facing begin ability motivation be present job generate sustainable solutions