Objective: To measure patients' and providers' delays in the presentation and treatment of newly diagnosed pulmonary tuberculosis.
Design: Retrospective study using structured questionnaires.
Setting: Tuberculosis and Chest Service, Centre of Health Protection, Department of Health.
Participants: Tuberculosis patients notified to the Department of Health, selected by systematic sampling of all notifications in the first 2 weeks of every even month in the year 2004.
Main outcome measures: Health-seeking behaviour of pulmonary tuberculosis patients, including respective demographic, clinical, and disease factors.
Results: Of a total of 6262 notified tuberculosis patients in 2004, 1662 (26.5%) were recruited into the study; of these, 42.6% first presented to private doctors, and 57.4% to the public sector. The diagnosis of tuberculosis was made in 13.7% of these patients by the former and 86.3% by the latter. The median patient delay (elapsed time from symptoms to medical consultation) and provider delay (elapsed time from medical consultation to treatment) were both 20 days; 25th to 75th percentiles being 7-37 and 6-55 days, respectively. Longer patient delay was associated with positive sputum smear and culture, and more extensive radiological disease. On multiple regression analysis, unemployment independently predicted longer patient delay, while haemoptysis predicted shorter patient and total delay. Patients older than 60 years, with no initial sputum and chest X-ray examination predicted longer provider and total delays.
Conclusions: Our patient and provider delays compared favourably with those of other countries, and very likely reflect easy service access. Adverse social factors and non-specific presentations prolong patient delay, whilst older age and unavailable bacteriological/radiological evidence delay diagnosis and treatment.