A decision analytic tree model was used to assess the costs and outcomes associated with the interventions under study. The time horizon of the study was the lifetime of the patient. The authors reported that the perspective was that of the healthcare system incorporating costs borne by patients.
Clinical and effectiveness data were derived from previously published studies. The main effectiveness measures used in the model were the success rate, failure rate, transfer out rate and death rate for each intervention. Treatment effectiveness for directly observed treatment interventions and self-administered treatment were derived from a Thai randomised controlled trial (Kamolratanakul et al. 1999, see Other Publications of Related Interest). Treatment effectiveness for the mobile phone intervention were derived from a monitoring care study (Visrutaratna et al. 2008, see Other Publications of Related Interest).
Monetary benefit and utility valuations:
Disability weights associated with tuberculosis were derived from a Dutch disability weights study. Disability weights for all other causes were derived from the Thai Burden of Disease and Injuries study.
Measure of benefit:
The measure of benefit was disability-adjusted life-years (DALYs) averted. Expected health-adjusted life expectancy was combined with the disability weights to estimate DALYs. Benefits could be generated over the lifetime of the patient. Future benefits were discounted using an annual rate of 3%.
Included costs were for government expenditure (including health staff supervision and monitoring time, travel costs to monitor patients at home, mobile phone monitoring care, laboratory costs, pharmaceutical costs, radiology and health facilities) and costs incurred by patients and/or their families (including travel and time costs of supervision, follow-up or treatment and out-of-pocket expenses). Resource use information was derived from previously published studies and the authors’ own assumptions. Unit costs were derived from published sources. The price year was 2005. Costs could be incurred over the lifetime of the patient. Future costs were discounted using an annual rate of 3%. All costs were reported in international dollars (I$), using an exchange rate of I$=12.12 Thai Baht.
Analysis of uncertainty:
Uncertainty distributions were fitted alongside model parameters. Ninety-five per cent uncertainty intervals (95% UI) were determined for all outcome measures using 2,000 Monte Carlo simulations.