Markers of adhesion to tuberculosis treatment: validation of instrument

Objective: To validate an instrument containing markers of adhesion to tuberculosis treatment to identify the patients vulnerable to non-adhesion to treatment in primary health care. Methods: Prospective and of methodological development study was conducted. We utilized construct validation and reliability in an instrument composed of 31 markers, each with a score of 1 to 3, applied to individuals with tuberculosis in the first 30 days of starting treatment, in basic health care units of the municipality of São Paulo, Brazil. Predictions of the groups were done by the receiver operating characteristic (ROC) curve to identify cutoff points on the scale. Results: The instrument was applied to 89 individuals with tuberculosis, after subjecting it to validation and reliability, where it ended up being composed of seven markers grouped in three dimensions: Life and family context, Work and health-disease process, Support in the context of health care. Cronbach’s alpha coefficient was 0.706 and ROC was 79.8%, with 95% CI of 69.8-89.8%, demonstrating that the group that adhered to treatment showed a better score than those who did not. Conclusion: The instrument is adequate with respect to reliability and validity and its use is of great relevance to studying adhesion to tuberculosis treatment, in primary health care.