Regional variation in non-traumatic major lower limb amputation in England: observational study of linked primary and secondary care data

Abstract

Background: Peripheral artery disease and diabetes are the main primary risk factors for non-traumatic major lower limb amputation. Regional variation in incidence of major lower limb amputation has yet to be fully described in terms of these risk factors and explained. The aim of this study was to estimate yearly incidence of major lower limb amputation over a 10-year interval (2010-2019) across England, by related condition and by region and, additionally, to investigate reasons for regional variation.

Methods: This observational study utilized primary care (Clinical Practice Research Datalink Aurum), secondary care (Hospital Episode Statistics), death and demographic data in England. Adults registered with a practice using Clinical Practice Research Datalink Aurum and with Hospital Episode Statistics linkage were included. Patients with a record of major lower limb amputation during the interval 1 January 2010 to 31 December 2019 were identified and yearly incidence rates of major lower limb amputation were calculated. Co-morbidities analysed were cardiovascular disease (including coronary artery disease, peripheral artery disease and cerebrovascular disease), diabetes (of any type) and cancer. Demographic and socioeconomic covariates analysed were age, sex, ethnicity, deprivation level, region and urban/rural categorization.

Results: The study included 18 397 483 individuals, 8584 of which had a record of major lower limb amputation. The age-standardized yearly incidence rate of major lower limb amputation in England decreased by 30% from 11.2 per 100 000 person-years in 2010 to 7.8 in 2019. The incidence rate in those with diabetes fell by 30% over the 10-year interval, rose by 20% for those with both diabetes and cardiovascular disease, and changed little in those with cardiovascular disease. In 2019, the age-standardized incidence rate was highest in the North East (14.8 per 100 000 person-years) and lowest in the East of England (4.5 per 100 000 person-years). Between 2010 and 1019, incidence rates decreased across all regions, the largest decrease of 56% in the East Midlands and the smallest of 8% in the North East. Statistically significant regional variation remained after full adjustment for demographic, socioeconomic data and related conditions.

Conclusion: Whilst the incidence of major lower limb amputation is decreasing overall, significant regional variation in major lower limb amputation exists and is unexplained by demographic, socioeconomic and health data. Regional differences in service provision and accessibility should be investigated to provide further explanation.

Fig. 1

Crude and age-standardized yearly incidence rates of MLLA in England per 100 000 person-years 2010–2019

Fig. 2

a Yearly incidence rate of MLLA and b number of MLLA and number of person-years at risk by morbidity rate

Fig. 3

a Crude versus b age-standardized incidence rate by region and year

Fig. 4

a Incidence rate ratios and confidence interval of each model applied by region compared with London over the whole 10-year study interval b Model definitions