Article Text


Prevalence of TB in healthcare workers in south west London
  1. T B L Ho1,
  2. C F J Raymer1,
  3. T Lindfield2,
  4. Y Young2,
  5. R J Whitfield3
  1. 1Chest Unit, St George’s Hospital, London SW17 0QT, UK
  2. 2South West London Health Protection Unit, London SW17 7DJ, UK
  3. 3Chest Clinic, Mayday University Hospital, Croydon CR7 7YE, UK
  1. Correspondence to:
    Dr T B L Ho
    Chest Unit, St George’s Hospital, London SW17 0QT, UK;

Statistics from

In the UK, and London specifically, the rise in the incidence of tuberculosis (TB) has been ascribed to reactivation of latent disease and importation of infection from recent immigrants.1 The recent increase in the recruitment of healthcare workers from countries with a high prevalence of TB raises the possibility of healthcare workers being a significant source of disease. Previous estimates of TB infection among National Health Service (NHS) employees were calculated before the current levels of HIV infection and the mass migration of healthcare workers.2,3 The current number of healthcare workers with TB is unknown but an estimate of this would provide data on the risk that they pose for spreading TB infection.

We conducted a retrospective interrogation of the local TB database (Integrated Tuberculosis Surveillance System, ITSS) for all healthcare workers notified in 2002. Their medical notes were then reviewed and a basic dataset was collated. A healthcare worker was defined as doctor, nurse, healthcare assistant, physiotherapist, occupational therapist, radiographer, or student equivalent. The data collected included profession, age, sex, type of disease, HIV status, country of origin, length of time in the UK when diagnosed (if applicable), history of Bacillus Calmette Guérin (BCG) vaccination, and presence of accompanying scar.

372 patients were notified as having TB in 2002 within the south west London catchment area (as of April 2003). Of these, 25 were healthcare workers (6.7%). Four were doctors, 13 nurses, five healthcare assistants, and three healthcare students. 22 (88%) were originally of overseas origin with a median (range) of 3 (0.75–22) years residence in the UK before diagnosis. Three were originally from the Indian subcontinent, 18 came from Africa, and one from the Caribbean. 18 patients had evidence of BCG vaccination (14 had a scar, 13 born overseas) and 17 (68%) had pulmonary TB. Nine patients (36%) were diagnosed as being HIV antibody positive, although not all patients agreed to be tested (table 1).

Table 1

 Basic demographic data for healthcare workers with tuberculosis

Healthcare workers contribute significantly to the number of patients with TB. A large proportion (36%) were co-infected with HIV and this is consistent with previous estimates.4 The majority of patients identified were nurses which, in part, reflects the high proportion of nurses among healthcare workers. Over two thirds had pulmonary TB and would therefore be deemed a greater infection risk.

Previous estimates of TB infection among NHS workers were calculated more than a decade ago. The total number of cases reported annually ranged from 3 to 5 among nearly 22 000 NHS staff monitored.3 The NHS workforce in our sector was estimated at 26 273. In order to calculate a rate of tuberculosis infection in the population we assumed that, unless otherwise indicated, all these healthcare workers worked for the NHS and that the number of cases treated within our sector, but working outside were equivalent to the number of south west London workers treated outside the sector. The TB rate for the south west London population has been estimated at 25 per 100 000 population per year, notably lower than the rate estimated for our healthcare worker population.5

No patient was diagnosed as part of pre-employment screening but the median time of 3 years from arrival in the UK to presentation suggests that most were unlikely to have had clinically apparent disease at the time of entry. It is unclear, however, if these patients had evidence of latent disease at this time. Currently there is no uniform health screening procedure for NHS workers. The British Thoracic Society (BTS) has produced guidelines for screening immigrant employees5 which rely on questionnaire evaluation of suspicious symptoms and evidence of BCG vaccination to screen for high risk individuals. 18 out of 25 (72%) of our patients had evidence of BCG vaccination and may therefore have been considered low risk if they did not report suspicious symptoms. Accordingly, a chest radiograph would not have been deemed necessary, even though this could have picked up evidence of tuberculous infection. The Department of Health in the UK has recently produced draft guidelines regarding TB screening for NHS employees.6 Based on the BTS recommendations, they propose further screening manoeuvres for workers from areas of high TB prevalence (incidence levels greater than 40 per 100 000 population per year). These include universal tuberculin skin testing (TST), HIV testing for those with negative TST results, and a low threshold for chest radiography. We believe these new guidelines would increase the detection of both active and latent TB and accordingly reduce the risk represented by infected healthcare workers.


View Abstract
  • Online Erratum

    Please note that there is an error in the author listing, the correct list is shown here:

    T B L Ho, C F J Rayner, T Lindfield, Y Young, and R J Whitfield

    The error is much regretted.

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles