Influenza in Madrid, Spain, 1991-92: validity of the sentinel network.
OBJECTIVES--The aim was, firstly, to study the validity of the sentinel network data by analysing the correlation between the weekly influenza cases detected by the network and the number of cases notified to the compulsory disease notification system and, secondly, to describe the epidemiology of the period 1991-92 in terms of the sentinel network data. DESIGN--The study design was descriptive. SUBJECTS--The population resident in the city of Madrid. MAIN RESULTS--There was a high correlation between the cases notified to the compulsory disease notification system and those listed by the sentinel network (r = 0.91, p < 0.001, r2 = 0.82). The epidemic activity during the period 1991-92 was centred approximately on week 48 in 1991 and week 6 in 1992 (24 November 1991-8 February 1992). Altogether 913 cases of influenza were notified through the sentinel network. The greatest number of cases occurred in people aged between 20 and 39 years. CONCLUSION--For the epidemiological surveillance of influenza it is necessary to develop systems that can quickly detect epidemic periods and provide information about populations at risk, in addition to systems that isolate and identify seasonal epidemic viruses. Both types of data will help the development of adequate public health policies. Sentinel networks provide these data and offer additional advantages such as lower cost.