Background and Objective Leprosy continues to be an important public health problem. The magnitude of the disease is often expressed by the number of cases registered by the system. Estimation of leprosy by conventional sampling is difficult due to large sample size requirement. A pilot study was conducted using inverse sampling in Bareilly district, Uttar Pradesh, India with an aim to evolve inverse sampling procedure vis-à-vis conventional sampling.

Method Two community development blocks one with low endemic area namely Fatehganj and other high endemic namely Ramnagar in Bareilly district of Uttar Pradesh, India were selected. Inverse sampling was adopted in Fatehganj while conventional cluster sampling was used in Ramnagar. Under inverse sampling, 25 new cases of leprosy were predetermined while a sample of 44 000 subjects (population) was targeted for conventional sampling. Under inverse sampling the total population to be covered was not known in advance.

Findings A total of 63 new cases of leprosy were found in Ramnagar after covering a sample of 44 686 population while a sample of 14 734 population was covered in Fatehganj to detect 25 new cases of leprosy. Both the techniques provide similar estimates. The precision obtained under inverse sampling was though less than that under conventional sampling but found to be more feasible and suitable for estimation of leprosy due to less population covered, time and cost.

Conclusion Study reveals that inverse sampling was found to be advantageous over conventional sampling and could be adopted for the large scale survey at National level.