Objectives: tuberculosis is a global heath priority and the surveillance systems are essential tools to control the spread of the disease. The objectives are to estimate: the incidence tuberculosis in Piedmont for the period 2001-03, using multiple sources of information; the sensitivity of the regional system for the surveillance of tuberculosis; the impact ofimmigration from areas with high prevalence of tuberculosis.
Design: deterministic linkage between electronic databases and estimate of non-observed cases of microbiologically confirmed tuberculosis, using the 'capture-recapture method'.
Methods: the Piedmonti population was the study base. Tuberculosis cases were identified from the integrated Regional Tuberculosis Surveillance System (SSR- TB), the Tuberculosis Laboratory Registry (RL-TB), and the Regional Registry of Inpatient Records (SDO).
Results: 1790 cases of tuberculosis were recorded, in 28.6% (512) of the cases the disease was observed among individuals from countries with high prevalence of tuberculosis. Microbiologically confirmed cases of tuberculosis were 1254 (70%), while pulmonary cases were 1233 (69%). The annual crude incidence of tuberculosis (per 100,000 inhabitants) was 14.1 cases, 31.5% of whom were attributable to immigration from countries with high prevalence of tuberculosis. The incidence among immigrants was 122.4 cases, while that of microbiologically confirmed cases was 9.8 cases. The overall estimated sensitivity, for the microbiologically confirmed cases, was 97%, while the sensitivity of SSR-TB was 90%.
Conclusions: Our estimate of tuberculosis incidence in Piedmont is about two times as great as that given by the WHO for Italy (8 cases per 100,000 inhabitants). The fraction of incident cases attributable to immigration from countries with high prevalence of tuberculosis is also consistent with that estimated in other industrialized countries undergoing high immigration flows. The integration of surveillance systems is essential to obtain accurate and timely estimates of the tuberculosis epidemiology.