Background Detection delay is defined as the period between onset of first signs and symptoms of leprosy and the time of diagnosis, comprising of a ‘patient delay’ and a ‘health-system delay’ and reliability refers to the consistency of a measure. Three types of consistency are considered: over time (test-retest reliability), across items (internal consistency), and across different researchers (interrater reliability). For the case detection delay (CDD) two are applicable: test-retest reliability and interrater reliability.

Methods The study was conducted in Ethiopia, Mozambique and Tanzania. The CDD questionnaire was administered to 79 leprosy patients. One month later, another researcher re-administered the CDD questionnaire with these same patients. Interrater reliability was assessed using the intra class correlation coefficient (ICC). The test-retest reliability was assessed among 69 leprosy patients by determining the CDD at one month difference, both times using the same rater, and then we determined the reliability by looking at the Pearson correlation between the two sets of CDD data.

Results Results from 79 leprosy patients show that, 3 (3.8%) were children under 15 years of age and 25 (31.6%) were women. Interrater reliability: the first interviews led to a mean CDD of 24.0 months (95% CI =17.1 - 30.9). The second interviews, also led to a mean CDD of 24.0 months (95% CI =17.7 - 30.3). For the test-retest reliability, the mean CDD of the first and second interviews were 16.5 (95% CI =13.6 - 19.5) and 16.9 (95% CI =13.8 - 20.1) respectively: the interrater reliability measured with ICC was 0.89 (95% CI of 0.84 - 0.93). Test-retest repeatability coefficient was 0.90 (p = 0.01).

Conclusion The PEP4LEP CDD tool to determine the case detection delay of newly diagnosed leprosy patients, was validated in the three African countries showing that both test-retest and interrater reliability measurements demonstrated good reliability of the instrument.