Inequalities in stroke patients' management in English public hospitals: a survey on 200,000 patients

PLoS One. 2011 Mar 2;6(3):e17219. doi: 10.1371/journal.pone.0017219.


Background: According to clinical guidelines, every patient affected by stroke should be given a brain-imaging scan (BIS) - Computerized Tomography or Magnetic Resonance Imaging - immediately after being admitted to hospital.

Aim of the study: To describe the variation in use of BIS among English public hospitals and identify any patient groups being excluded from appropriate care.

Methods: We collected hospital administrative data for all patients admitted to any English public hospital with a principal diagnosis of stroke from 2006 to 2009. We calculated the proportion of patients treated with BIS in the whole sample and after stratification by hospital. We compared hospitals' performance using funnel plots. We performed a multiple logistic regression analysis using BIS as outcome and age, gender, socio-economic deprivation, and comorbidity as covariates.

Results: In English public hospitals there are about 70,000 emergency admissions for stroke per year. Nationally, only 35% receive a BIS immediately, and only 84% receive it within the admission. There is large variation in the use of BIS for stroke patients among English public hospitals, with some of them approaching the recommended 100% and some having very low rates. Young (P<0.001), male (P = 0.012), and least socio-economically deprived patients (P = 0.001), as well as patients with fewer comorbidities (P<0.001) appear to have more chance of being selected for a brain scan.

Conclusion: Some English public hospitals appear to be falling well below the clinical guideline standards for scanning stroke patients and inappropriate patient selection criteria may be being applied, leading to health inequalities.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Delivery of Health Care / statistics & numerical data*
  • England / epidemiology
  • Female
  • Health Care Surveys*
  • Hospitalization
  • Hospitals, Public / statistics & numerical data*
  • Humans
  • Logistic Models
  • Male
  • Patient Selection
  • Practice Guidelines as Topic
  • Stroke / epidemiology*
  • Stroke / therapy*