Related Experiment Videos

Identification of stroke in the community: a comparison of three methods

J Mant1, R J McManus, R Hare

  • 1Department of Primary Care and General Practice, University of Birmingham, Primary Care Clinical Sciences Building, Edgbaston, Birmingham B15 2TT. j.w.mant@bham.ac.uk

Insights

Identifying prevalent stroke cases is crucial for secondary prevention. General practice computer systems offer a reasonably accurate method for creating stroke registers, with patient surveys as a viable alternative.

Area of Science:

  • Public Health
  • Epidemiology
  • Health Informatics

Background:

  • Secondary prevention of cerebrovascular disease is underutilized in clinical practice.
  • Accurate identification of eligible patients is a prerequisite for effective secondary prevention strategies.
  • Establishing a register of prevalent stroke cases in primary care is essential.

Purpose of the Study:

  • To compare the effectiveness of three methods for identifying prevalent cerebrovascular disease cases in the community.
  • Methods compared include general practice computer systems, population surveys, and hospital information systems.

Main Methods:

  • A comparative study involving 5801 individuals aged 65+ in South Birmingham.
  • Sensitivity, specificity, and positive predictive value (PPV) of each identification method were calculated against reference criteria.
  • Reference criteria were applied independently by two investigators.

Main Results:

  • The prevalence of validated cerebrovascular disease in the target population was 8.2%.
  • General practice computer systems showed 81.0% sensitivity and 97.2% specificity, with a 71.8% PPV, but with significant inter-practice variability.
  • Patient surveys (75.7% sensitivity, 77.5% PPV) and hospital systems (28.4% sensitivity, 89.2% PPV) had different performance profiles. Notably, 39% of patients with cerebrovascular disease had no hospital admission history.

Conclusions:

  • General practice-based computer systems are capable of generating accurate prevalent stroke registers.
  • Patient surveys serve as a practical alternative in settings with underdeveloped primary care IT systems.
Abstract

Related Concept Videos