Skip to main content

Table 2 Analysis of quantitative outcomes

From: Impact of a mobile phone-based interactive voice response software on tuberculosis treatment outcomes in Uganda (CFL-TB): a protocol for a randomized controlled trial

Outcomes

Measure of outcomes (intervention versus control arm)

Statistical analysis

Treatment success

Proportion of patients with treatment success at six months. Odds ratio of treatment success

Logistic regression, odds ratios

Cure rates

Proportion of patients cured of TB; odds ratio of TB cure

Logistic regression, odds ratios

Treatment completion

Treatment completion rate; odds ratio of treatment completion

Logistic regression, odds ratios

Retention in care

Hazards of attrition in the intervention versus control; Cumulative attrition at six months

Cox’s proportional hazards model, log-rank test

Adherence to TB medicines

Mean adherence rates (proportion of TB medicines)

Mann-Whitney U test and Poisson regression

Appointment keeping and punctuality

Proportion of patients who keep an appointment; Proportion of patients with early, prompt, and late visits

Ordered logistic regression

Drug-resistant TB rates

Proportion of patients with rifampicin-resistant TB

Logistic regression, odds ratios

Call success rates (2,4, and 6 months)

Proportion of successful calls

t test