Abstract
Background
The contemporary prevalence of ANCA-associated vasculitis (AAV) in England is unknown. Hospital Episode Statistics (HES) contain data on every hospital and day case NHS admission in England since 1997. In collaboration with the National Congenital Anomaly and Rare Disease Registration Service (NCARDRS) we validated the diagnosis of AAV using ICD codes in HES. The positive predictive value of these codes was 86%, which compares favourably to a median coding accuracy of 80% in a recent systematic review of NHS coding studies. This justifies using this novel dataset for population-based epidemiology with coverage of the whole population of England.
Methods
We worked within NCARDRS enabled by their Section 251 legal permissions (CAG 10-02(d)/2015). We extracted all cases of AAV from HES 2011/12 to 2016/17 using ICD-10 codes M313 Granulomatosis with polyangiitis (GPA), M317 Microscopic polyangiitis (MPA), and M301 Eosinophilic granulomatosis with polyangiitis (EGPA). We used the Summary Care Record to check vital status and record date of death where appropriate. We estimated point prevalence on 1 July 2016 using ONS mid-year population estimates for England in 2016 as the denominator. Standardised mortality ratio (SMR) was calculated using the Office for National Statistics death summary tables 2016 to provide expected number of deaths for each 5-year age-band and sex.
Results
We identified 9,890 patients who were coded as having AAV during a hospital admission 2011-2017. This included 6,856 (69.3%) with GPA, 964 (9.8%) with MPA and 2,070 (20.9%) with EGPA. On 1 July 2016, our dataset found 8,040 people in England were living with ANCA associated vasculitis. We estimate the prevalence was 14.55 (95% CI: 14.23-14.87)/100,000 adult population. The median age of these patients was 65.3 years (interquartile range 52.3-74.2). 47% were female. The prevalence of GPA was 9.97/100.000 (95% CI: 9.71-10.24), MPA was 1.40/100,000 (95% CI: 1.30-1.50), and EGPA was 3.18/100,000 (95% CI: 3.03-3.33). People with AAV were 4.6 times more likely to die than the background population of the same age and sex (Standardised Mortality Ratio = 4.58).
Conclusion
There are no recent UK prevalence estimates for all types of ANCA-associated vasculitis. Studies in Australia, Germany, Southern Sweden and the USA have found estimated prevalence to be between 4.6-18.4 cases per 100,000 individuals. Our estimate of 14.6/100,000 in England is consistent with this, and towards the higher end of the range. However, our estimates underestimate the prevalence of MPA compared to other studies, and further work is needed to increase the routine identification of cases of MPA. Further work within NCARDRS using their unique data linkages will enable more specific AAV case ascertainment as well as nationwide population-based studies on cause of death and studies using the database of English prescriptions dispensed in the community.
Disclosures
F.A. Pearce None. B. Griffiths None. C. Mukhtyar None. R. Al-Jayoussi None. R.A. Watts None. J. Aston None. M. Bythell None. S. Stevens None. P.C. Lanyon None.
The Dutch Hospital Standardised Mortality Ratio (HSMR) method and cardiac surgery: benchmarking in a national cohort using hospital administration data versus a clinical database
