Get access

Reliability of pressure ulcer classification and diagnosis

Authors

  • Jane Nixon BSc MA PhD RGN,

  • Helen Thorpe BSc MSc,

  • Helen Barrow RGN,

  • Angela Phillips,

  • E. Andrea Nelson BSc PhD RN,

  • Susan A. Mason PhD RGN DN HV,

  • Nicky Cullum PhD RGN


Jane Nixon,
Clinical Trials Research Unit,
University of Leeds,
17 Springfield Mount,
Leeds,
LS2 9NG,
UK.
E-mail: j.e.nixon@leeds.ac.uk

Abstract

Aim.  To assess the inter-rater reliability of the PRESSURE Trial pressure ulcer diagnosis (≥Grade 2) and skin classification for all grades between the clinical research nurse (CRN) team leader and CRNs working across different hospital sites; and CRNs and ward-based nurses.

Background.  The United Kingdom National Health Service Health Technology Assessment Programme has funded a multi-centre, randomized controlled trial to compare the clinical and cost-effectiveness of alternating pressure mattress overlays and mattress replacements – PRESSURE Trial. Outcome skin assessments were recorded by qualified ward-based nurses daily, and expert CRNs twice weekly.

Method.  Paired assessments were undertaken and skin assessed on seven body sites. The per cent agreement between nurses in the diagnosis of a pressure ulcer was determined and the Kappa statistic and confidence intervals calculated. Per cent agreement between nurses in classifying skin for all grades was also determined.

Results.  Assessments were undertaken by 378 pairs: 16 paired patient assessments (107 site comparisons) by the CRN team leader and CRNs, and 362 paired patient assessments (2396 site comparisons) between CRNs and ward-based nurses. There was 100% agreement between the CRN team leader and CRNs in the diagnosis of a pressure ulcer, and the Kappa statistics indicated ‘very good’ agreement. There were only two (1·9%) disagreements in classifying skin for all grades between these nurses. The agreement in the diagnosis of a pressure ulcer between CRNs and ward-based nurses varied by skin site, ranging from 93·6% to 100%, with the Kappa statistics indicating ‘good’ and ‘very good’ agreement. However, there were 508 (21·2%) disagreements in classifying skin for all grades.

Conclusions.  Overall agreement and Kappa statistics indicated ‘very good’ and ‘good’ agreement between expert nurses, and between expert nurses and qualified ward-based staff, respectively. However, the high prevalence of normal skin concealed clinically important disagreements in both the diagnosis of pressure ulcers and skin classification for all grades.

Ancillary