Pearce, Fiona A.
(2018)
The epidemiology and natural history of ANCA-associated vasculitis in the UK: a response to The UK Strategy for Rare Diseases.
PhD thesis, University of Nottingham.
Abstract
Introduction:
Rare diseases have gained recognition over the past decade as an important area for health service improvement. They affect 1 in 17 people over a lifetime, consume 14% of the NHS budget, and are an important cause of illness and death[1]. In 2013, the Department of Health published for the first time a “UK Strategy for Rare Diseases” which identified 5 key recommendations for improving the lives of people with rare diseases, and included the need for epidemiological studies. 75% of rare diseases are genetic in origin, and have onset in childhood. However, rheumatologists provide care for people with a number of rare diseases, amounting to an important component of the 25% of rare diseases which are believed to be non-genetic and present in adulthood.
One of the main challenges in studying rare disease, is the difficulty in finding enough people with the disease to study. Recent linkage of databases of primary and secondary care health records in the UK provide an opportunity to study representative samples, large numbers of people, and the breadth of healthcare provision.
In this thesis, I report 4 epidemiological studies in routinely collected healthcare data, conducted in an exemplar rheumatic disease, ANCA-associated vasculitis.
Methods:
I used routinely collected healthcare data from local hospitals, a large database of UK general practice records, and linked hospital episode statistics to identify cases of ANCA-associated vasculitis.
The specific questions addressed in the studies were:
1. What are the current incidence, prevalence and mortality of ANCA-associated vasculitis in the UK, stratified by age, sex and ethnic group? I addressed this question in a local hospital cohort, and then in linked primary and secondary care records in England. I used projections of the population structure in 20 years’ time to predict the expected number of incident and prevalent cases.
2. What is the natural history of the most common type of ANCA-associated vasculitis (granulomatosis with polyangiitis) prior to diagnosis, and are there opportunities to diagnose it sooner in primary or secondary care? I conducted a case-control study in a large database of primary care records, and attempted to develop a model for GPs that predicted the risk of a person having granulomatosis with polyangiitis, such as is advocated by the UK Strategy for Rare Diseases.
3. What are the strongest aetiological factors in granulomatosis with polyangiitis in the UK? I compared the frequency of possible aetiological exposures between cases and population-based controls in a large database of prospectively collected primary care data.
Results:
1. There were about 1300 new cases of ANCA-associated vasculitis in the UK in 2015, more than previously thought. The incidence has been stable since 2000, however the disease is more common in older people, therefore due to predicted aging of the UK population there will be 34% more cases in 2035. We are not able to detect any differences in incidence rates between people from different ethnic groups, but our studies lacked power for this question.
2. People with granulomatosis with polyangiitis consulted their GP more than healthy people prior to their diagnosis. However, they consulted with a wide variety of symptoms, and none were highly predictive of the diagnosis. In addition, the lower the prevalence of the disease, the lower the positive predictive value of a diagnostic model. Granulomatosis with polyangiitis is rare, so even with an exceptionally well-performing model (with a sensitivity of 100% and a specificity of 90%) only 2 of every 10,000 people flagged as ‘at risk’ would have granulomatosis with polyangiitis, and the rest would have false positive results.
3. People with granulomatosis with polyangiitis were 5 times more likely than population-based controls to have a previous diagnosis of bronchiectasis, and 2-3 times more likely to have a previous diagnosis of an autoimmune disease or chronic renal impairment.
Conclusions:
1. Although incidence appears stable, commissioners need to expand services to diagnose and treat people with GPA, and other adult-onset rare diseases, over the next 20 years due to the predicted increase in the proportion of the population in the age-groups at highest risk.
2. In particular, over the next 20 years, the age structure of the Black and Minority ethnic population in the UK is predicted to change to have many more people in older age-groups, and the UK medical community need to be alert to an expected emergence of significant numbers of Black and Minority ethnic people with ANCA-associated vasculitis.
3. Computerised prompts to alert GPs to consider a diagnosis of rare disease are unlikely to work. Resources to improve early diagnosis and treatment of ANCA-associated vasculitis would be better targeted at secondary care where the majority of cases have contact in the year before diagnosis.
4. The novel association of bronchiectasis with developing granulomatosis with polyangiitis raises a new hypothesis for bronchiectasis as a possible aetiological factor in granulomatosis with polyangiitis.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of FPearce Thesis_final_after_viva&amendments.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "FPearce Thesis_final_after_viva&amendments.pdf")