Educational outreach to promote screening for tuberculosis in primary care: a cluster randomised controlled trial

Lancet. 2007 May 5;369(9572):1528-1534. doi: 10.1016/S0140-6736(07)60707-7.

Abstract

Background: Tuberculosis is re-emerging as an important health problem in industrialised countries. Uncertainty surrounds the effect of public-health control options. We therefore aimed to assess a programme to promote screening for tuberculosis in a UK primary health care district.

Methods: In a cluster randomised controlled trial, we randomised 50 of 52 (96%) eligible general practices in Hackney, London, UK, to receive an outreach programme that promoted screening for tuberculosis in people registering in primary care, or to continue with usual care. Screening was verbal, and proceeded to tuberculin skin testing, if appropriate. The primary outcome was the proportion of new cases of active tuberculosis identified in primary care. Analyses were done on an intention-to-treat basis. This study was registered at clinicaltrials.gov, number NCT00214708.

Findings: Between June 1, 2002, and Oct 1, 2004, 44,986 and 48,984 patients registered with intervention and control practices, respectively. In intervention practices 57% (13,478 of 23,573) of people attending a registration health check were screened for tuberculosis compared with 0.4% (84 of 23 051) in control practices. Intervention practices showed increases in the diagnosis of active tuberculosis cases in primary care compared with control practices (66/141 [47%] vs 54/157 [34%], odds ratio (OR) 1.68, 95% CI 1.05-2.68, p=0.03). Intervention practices also had increases in diagnosis of latent tuberculosis (11/59 [19%] vs 5/68 [9%], OR 3.00, 0.98-9.20, p=0.055) and BCG coverage (mean BCG rate 26.8/1000 vs 3.8/1000, intervention rate ratio 9.52, 4.0-22.7, p<0.001).

Interpretation: Our educational intervention for promotion of screening for tuberculosis in primary care improved identification of active and latent tuberculosis, and increased BCG coverage. Yield from screening was low, but was augmented by improved case-finding. Screening programmes in primary care should be considered as part of tuberculosis control initiatives in industrialised countries.

Publication types

  • Multicenter Study
  • Randomized Controlled Trial

MeSH terms

  • Cluster Analysis
  • Health Promotion / methods*
  • Humans
  • London / epidemiology
  • Mass Screening / methods*
  • Tuberculin Test / statistics & numerical data*
  • Tuberculosis / diagnosis*
  • Tuberculosis / epidemiology

Associated data

  • ClinicalTrials.gov/NCT00214708