Can we use routinely collected primary care medical records to assess and explain variations in major lower extremity amputation rates across the England?

Date of Approval: 
2021-03-17 00:00:00
Lay Summary: 
Individuals with poor circulation to their lower limbs (known as ‘chronic limb-threatening ischemia’) or diabetes mellitus can end up needing to have part, or even all, of their leg amputated. Having an amputation can have a devastating impact on a person’s quality of life and can be a life changing event. Despite this, there is uncertainty over the number of amputation procedures performed in the England. It is also unclear whether the numbers are increasing or decreasing over time. Having a greater understanding of how many amputations procedures are being performed, and whether numbers vary between the regions of England, is vital help reduce the need for patients to undergo such a procedure. We want to see if data on amputation procedures are being recorded correctly in individuals’ GP (primary care) health records, by comparing them to their hospital (secondary care) records. GP records contain more detailed information on a person’s health and therefore may help explain why the numbers of amputation procedures potentially varies between regions of the country. We also want to accurately estimate how many amputation procedures are being performed in England so we can see if the numbers are changing over time and if the numbers vary by region then investigate the reasons for any change. We hope the results of this study will ultimately help improve the care for patients with poor circulation and reduce the need for patients to undergo an amputation in the future.
Technical Summary: 
This study comprises two work packages which are outlined below. Work Package One: This work package aims to assess the agreement in case ascertainment of major lower extremity amputation (MLEA) between primary care (Clinical Practice Research Datalink (CPRD)) and secondary care (Hospital Episode Statistics (HES)) data in order to establish if MLEA are well recorded in primary care (CPRD). Cases of MLEA fitting the inclusion criteria within the study period of 01/01/2009 – 31/12/2019 will be ascertained in two ways: by using OPCS-4 codes in HES and also by using READ(SNOMED) codes in CPRD). Agreement in MLEA cases between HES and CPRD will be assessed using specificity, sensitivity, positive and negative predictive value and Cohen’s Kappa coefficient. Logistic regression analysis will be performed to investigate reasons for disagreement between databases. Work Package Two: The aims of this work package are to establish the incidence proportion of MLEA in England, investigate whether this differs over time and regionally and explain any regional differences. Using data from the created cohort in work package one, the overall and regional incidence proportion of MLEA will be calculated giving both a crude and standardised outcome measure. Time trends in incidence will be investigated using plots Poisson regression models. Regional variation in incidence will be assessed by calculating the coefficient of variation with reasons for any observed differences explored using Poisson regression models. CPRD population data needed to calculate case agreement, incidence and regional risk factor/demographic proportions and incidence denominators will be gathered using CPRD denominator files and a large random sample of one million patients for risk factor/demographic data.
Health Outcomes to be Measured: 
Major lower extremity amputation (MLEA) defined as a lower limb amputation above the ankle. See Appendix 1 for codes).
Application Number: 
21_000364
Collaborators: 

Laura Gray - Chief Investigator - University of Leicester
- Corresponding Applicant -
Andrew Nickinson - Collaborator - University of Leicester
Clare Gillies - Collaborator - University of Leicester
John Houghton - Collaborator - University of Leicester
Kamlesh Khunti - Collaborator - University of Leicester
Rob Sayers - Collaborator - University of Leicester
Tanya Payne - Collaborator - University of Leicester

Linkages: 
2011 Rural-Urban Classification at LSOA level;HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation