Potential inaccuracy in recording the presence of diabetes in people with amputations of the lower limb in hospitals in England
- First Online:
- Cite this article as:
- Ahmed, M., Moulik, P. & Macleod, A. Diabetologia (2012) 55: 3396. doi:10.1007/s00125-012-2735-6
- 399 Views
KeywordsAmputationsDiabetesLower limbPrimary care trustsRecordsStatistics
Hospital episode statistics
Primary care trust
Shropshire County PCT
Telford and Wrekin PCT
To the Editor: We applaud Holman et al  for gathering data regarding the incidence of major lower limb amputations in people with diabetes. However, we believe that comparison of areas covered by primary care trusts (PCTs) is not sufficiently accurate because of poor quality data collection in hospitals in England. Data regarding prevalence of diabetes were gleaned from general practice systems, generated during clinical care. Data recording co-existence of diabetes are taken from hospital episode statistics (HES), usually collected by non-clinical coders, after perusal of hospital notes.
We have examined the case records of all patients coded as having major lower limb amputations in two hospitals that are the major catchment of our two PCTs, for the same time period. There were 21 patients in Shropshire County PCT (SCPCT), and 15 in Telford and Wrekin PCT (TWPCT), a prevalence of 0.58 and 0.71 amputations per year per 1,000 people with diabetes respectively (as opposed to 0.6–0.9 and >1.2, quoted by Holman et al ). However, seven of those patients in SCPCT and one from TWPCT were incorrectly coded as not having diabetes, which would have resulted in a false prevalence of 0.36 for SCPCT and 0.66 for TWPCT. We suspect that similar inaccuracies occur in other hospitals in England.
We therefore believe that it is premature to compare areas with small numbers of amputations, such as our PCTs, given that coding for diabetes in relevant patients from HES, as opposed to GP systems, is so inaccurate. Some areas may be unfairly criticised, whilst others become complacent.
The authors thank the audit departments of both hospitals for their help.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
The authors were responsible for the conception, design and drafting of the manuscript, and approved the final version for publication.