Feasibility and reliability of health-related quality of life measurements among tuberculosis patients.

The dramatic global impact of tuberculosis on mortality has been well documented, but its impact on morbidity has not been well described. The emphasis on treatment of latent tuberculosis (TB) infection highlights the tradeoff between short-term decrements in health status from 'preventive' therapy, and long-term gains related to fewer cases of active TB. However, these changes in health status have not been characterized. As a first step, we examined the feasibility and reliability of administering two health status questionnaires, in a multicultural TB clinic setting. The Medical Outcomes Study SF-36 and the EuroQOL EQ-5D were self-administered during 3 weekly interviews. One hundred and eighty-six potentially eligible patients were identified, of whom 112 could be evaluated; 106 (57%) were confirmed eligible. Sixty-seven (63%) agreed to participate; 24 (36%) were women. Fifty-three participants (79%) were foreign-born, with median residence in Canada of 3.5 years. Fifty (75%) of the participants completed all study measurements: 25 were treated for latent TB, 17 for active TB, and eight had previous active TB. Cronbach's alpha coefficients ranged from 0.73 to 0.94 for the SF-36 domain scores. Intraclass correlation coefficients were 0.66 for the SF-36 physical component summary, 0.79 for the mental component summary, and 0.73 for the EQ-5D. These instruments appeared reliable in a highly selected group of TB patients.