Tuberculosis in England and Wales in 1993: results of a national survey. Public Health Laboratory Service/British Thoracic Society/Department of Health Collaborative Group.

backgroundA national survey of tuberculosis notifications in England and Wales was carried out in 1993 to determine the notification rate of tuberculosis and the trends in the occurrence of disease by ethnic group in comparison with the findings of similar surveys in 1978/79, 1983, and 1988. The prevalence of HIV infection in adults notified with tuberculosis in the survey period was also estimated.

methodsClinical, bacteriological, and sociodemographic information was obtained on all newly notified cases of tuberculosis in England and Wales during the six months from 2 January to 2 July 1993. The prevalence of HIV infection in 16-54 year old patients with tuberculosis notified throughout 1993 was assessed using "unlinked anonymous" testing supplemented by matching of the register of patients with tuberculosis with that of patients with AIDS reported to the PHLS AIDS centre. Annual notification rates were calculated using population estimates from the 1993 Labour Force Survey.

resultsA total of 2706 newly notified patients was eligible for inclusion in the survey of whom 2458 were previously untreated the comparable figures for 1988 were 2408 and 2163. The number of patients of white ethnic origin decreased from 1142 (53%) in 1988 to 1088 (44%) in 1993 whereas those of patients of Indian, Pakistani, or Bangladeshi (Indian subcontinent (ISC)) ethnic origin increased from 843 (39%) in 1988 to 1014 (41%) and those of "other" (non-white, non-ISC) ethnic origins increased from 178 (8%) to 356 (14%). The largest increase was seen in the black African ethnic group from 37 in 1988 to 171 in 1993. Forty nine per cent of patients had been born abroad and the highest rates were seen in those who had recently arrived in this country. The overall annual notification rate for previously untreated tuberculosis in England and Wales increased between 1988 and 1993 from 8.4 to 9.2 per 100,000 population. The rate declined in the white, Indian, and black Caribbean ethnic groups and increased in all other groups. In the white group the rate of decline has slowed since the last survey: in several age groups the rates were higher in 1993 than 1988 but the numbers in these groups were small. Thirty six (4.1%) of the 882 previously untreated respiratory cases were resistant to isoniazid and three (0.3%) to isoniazid and rifampicin. Sixty two (2.3%) adults aged 16-54 years were estimated to be HIV-infected. Evidence of under-reporting of HIV positive tuberculosis patients was found.

conclusionsThe number of cases and annual notification rate for previously untreated tuberculosis increased between 1988 and 1993. Although the decline in rates in the white population has continued, the rate of decline has slowed. The high rates in the ISC ethnic group population have continued to decline since 1988 whereas rates in the black African group have increased. An increased proportion of cases were found among people born abroad, particularly those recently arrived in this country. In previously untreated cases the level of drug resistance remains low and multi-drug resistance is rare. A small proportion of adults with tuberculosis were infected with HIV but there may be selective undernotification of tuberculosis in these patients.