1. and number with tuberculosis (England and Wales, excluding visitors),1991
2. Patients reported in the MRC tuberculosis notification surveys of 1983 and 1988 were compared with those reported to the Communicable Disease Surveillance Centre national AIDS reporting system to establish whether any patients were common to both lists. For reasons of confidentiality the AIDS reporting system does not require patients to be identified by name. In cases where the name is withheld a name code (Soundex)'3 is used instead. The matching process was carried out by computer to select patients with the same Soundex code, sex, and year of birth. Potential matches were then compared by inspection of the original report forms, with name if given, ethnic group, exact date of birth, district of residence, and marital status as matching criteria
3. Physicians notifying one or more unmarried white men aged 25-64 years with tuberculosis during the 1988 MRC tuberculosis survey were sent a questionnaire that asked (a) whether they had tested any of the patients they had notified during the survey for HIV infection and (b) if so the total number tested and the number positive. This group was selected because unmarried white men in this age group alone accounted for most of the cases of AIDS reported to the Communicable Disease Surveillance,1988
4. Tuberculosis and the acquired immunodeficiency syndrome;Rieder, H.L.; Snider, D.E.;Chest,1986
5. Centres for Disease Control. Tuberculosis United States, 1985 and the possible impact of human T-lymphotropic virus type III/lymphadenopathy-associated virus infecportion of patients with AIDS in these ethnic groups who had developed tuberculosis is not surprising; but 3-8% of those cases of "white" tion;Morbidity and Mortality Weekly Report