Article Text

G449(P) To what extent did peer review scoring reflect on patient outcomes in paediatric diabetes units in England
  1. I Aldred,
  2. HP Satish
  1. Child Health, Warrington and Halton Hospitals NHS Foundation Trust, Warrington, UK



  • To identify any association between the Diabetes Peer review (PR) scoring and clinical outcomes measured in this instance as median HbA1c.

  • To assess whether the PR scores correlate with the percentage of patients with good control (<58 mmol/mol) and poor control (>80 mmol/mol).

Methods/Design Of the hospitals included in the National Paediatric Diabetes Audit (NPDA) 138 underwent the PR process completed between November 2013–2014. Individual hospital and MDT measures taken from PR were matched with the unadjusted median HbA1c for the corresponding year taken from the National Paediatric Diabetes Audit (NPDA) 2013–2014. PR scores were also analysed against the number of patients with poor control (>80 mmol/mol) and good control (<58 mmol/mol) as defined in NPDA 2013–2014. Correlation analysis and students t testing were the tools used. The median HbA1c of the hospitals in the PR report with immediate risk (IR), serious concern (SC) and no concerns (NC) were also analysed and compared.

Results Of the 138 results included the PR MDT scores produced a correlation with median HbA1c (r=-0.157, p = 0.034), but hospital scores did not (r=-0.138, p > 0.05). A stronger correlation is produced when combining both MDT and hospital measures (r=-0163, p = 0.024). The percentage of good and poor controlled patients had no significant correlations with PR scores (r = 0.0448, p = 0.301, r=-0.0530, p = 0.268 respectively). Hospitals with IR had a mean score 51.3%, SC 67.5% and NC 77.3%, these subgroups mean of median HbA1c (mmol/mol) were 71.1, 69.6 and 69.4 respectively.

Conclusion Higher PR scores are indicative of improved compliance towards the measures, which have an impact on quality of care. The PR scores do reflect on clinical outcomes measured on a patient level as HbA1c. The PR scores did not correlate with the percentage of good and poor control patients.

Statistics from

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.