Abstract

BACKGROUND: Notification of tuberculosis is essential for local contact tracing and for assessing the national incidence of tuberculosis. The accuracy of notification figures is uncertain. This study examined the notification rates of all patients diagnosed as having tuberculosis at two hospitals in the East End of London over five years. METHODS: In a retrospective survey of all patients aged 16 years or more presenting with tuberculosis to the London Chest Hospital or the Royal London Hospital from 1 January 1985 to 31 December 1989, cases of tuberculosis were identified from microbiology and histology records, statutory notifications, necropsy reports, coroners' records, hospital activity data, and death certificates. Clinical data were obtained from case notes and notification was determined from the local authority notification lists. RESULTS: Six hundred and nine adult patients with tuberculosis were identified. Notes were available for 580 cases (95%), of which 426 (73%) had been notified. The proportion of cases notified varied according to the specialty of the clinician in charge of the patient at diagnosis. Patients with a past history of tuberculosis and those who died within one year were less likely to have had their tuberculosis notified. Age, race, and lack of microbial or histological confirmation of diagnosis did not influence the proportion of cases notified. One hundred and eighty five patients had smear positive sputum, but 25 of these cases (14%) were not notified. Eighty five patients who had presented with pulmonary tuberculosis did not have their disease notified; 20 (24%) had smear positive sputum. CONCLUSIONS: Many cases of tuberculosis are not notified (27%). Fourteen per cent of all sputum smear positive cases of tuberculosis were not notified, and these patients are a considerable public health risk. The true incidence of tuberculosis in the area studied is at least one third higher than current notification figures suggest.