Factors affecting delay in initiation of treatment of tuberculosis in the Thames Valley, UK

To quantify and determine factors associated with delay in initiation of tuberculosis (TB) treatment in the Thames Valley area, South East England, and the proportion of this delay that could be attributed to patient care-seeking or to delay within the National Health Service (NHS).

Retrospective analysis study reviewing medical notes and enhanced TB surveillance data.

Demographic and clinical information was collected from medical notes and the Enhanced TB Surveillance database for patients who were diagnosed with TB and resident in the Thames Valley. Treatment delay was defined as the period between the onset of symptoms and the start of treatment. Patient delay was defined as the period between the onset of symptoms and the first presentation to the NHS. Health service delay was defined as the period between the first contact with the NHS and the start of treatment. Univariate and multivariate linear regression analyses were used to assess the association between delays and explanatory variables (age, gender, place of birth, ethnicity, disease site, sputum smear, culture, primary care trust of residence).

The study included 273 patients with TB. The median time between symptom onset and initiation of treatment was 73 days [95% confidence interval (CI) 65-89], of which the contributions of health service, patient and referral delays were 39 (95% CI 34-55), 29 (95% CI 22-36) and 16 (95% CI 12-24) days, respectively. On multivariate analysis, extrapulmonary TB (P = 0.010), female (P = 0.003) and UK-born (P = 0.008) patients were associated with longer health service delay. Age (P = 0.001) and extrapulmonary TB (P = 0.010) were associated with longer overall treatment delay.

Treatment delay for TB, especially delay after first presentation to the NHS, remains a public health concern. Differences in health service delay, for example by gender and country of birth, highlight that some of this should be open to health service intervention.