Endgames Statistical Question

Screening tests and indices of performance: effects of prevalence

BMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d6483 (Published 12 October 2011) Cite this as: BMJ 2011;343:d6483
  1. Philip Sedgwick, senior lecturer in medical statistics
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, Tooting, London, UK
  1. p.sedgwick{at}sgul.ac.uk

The accuracy of a screening test for depression was evaluated in primary care. The screening test involved asking patients two questions about depression and pleasure in the previous month. In total, 421 consecutive general practice patients who weren’t taking psychotropic drugs agreed to participate. The two questions were asked at any time during the patient’s consultation. If the patient responded positively to either question, the screening test result was deemed “positive” and the patient at “high risk” of depression—otherwise, screening was considered “negative” and the patient at “low risk” of depression. A self completed, computerised international diagnostic interview was used to diagnose depression. The table shows the results of the depression screening test in relation to the clinical diagnostic interview. 1

View this table:

Comparison of results of screening test for depression with clinical diagnosis

As a screening test for depression, the two questions showed a sensitivity of 96.6%, a specificity of 67.1%, a positive predictive value of 17.8%, and a negative predictive value of 99.6%. The sample estimate of the population prevalence of depression was 6.9%.

If the screening test …

View Full Text

Log in

Log in through your institution


* For online subscription