Identifying the optimum strategy for identifying adults and children with coeliac disease: systematic review and economic modelling

Talk Code: 
P1.1B.1
Presenter: 
Martha Elwenspoek
Twitter: 
Co-authors: 
Martha MC Elwenspoek, Howard Thom, Athena L Sheppard, Edna Keeney, Rachel O’Donnell, Joni Jackson, Cristina Roadevin, Sarah Dawson, Deborah Lane, Jo Stubbs, Hazel Everitt, Jessica C Watson, Alastair D Hay, Peter Gillett, Gerry Robins, Hayley E Jones, Sue Mallett, Penny Whiting
Author institutions: 
University of Bristol, NIHR ARC West, University of Leicester, Royal Hospital for Sick Children Edinburgh, University of Southampton, York Teaching Hospital NHS Foundation Trust, University College London

Problem

Coeliac disease (CD) is an autoimmune disorder that affects approximately 1% of the UK population and is triggered by ingesting dietary gluten. Only 1 in 3 are thought to have a diagnosis, meaning that the majority of cases are not aware they have the disease. Untreated CD damages the lining of the gut, which may lead to malnutrition, anaemia, and osteoporosis. Our main objective was to define at risk groups and determine the cost-effectiveness of active case finding in primary care.

Approach

We performed systematic reviews and meta-analyses on the accuracy of risk factors for CD, such as chronic conditions and symptoms, and on the accuracy of diagnostic tests for CD, including serological and genetic tests. We used the identified risk factors to develop prediction models for identification of people who may benefit from testing for CD in routine primary care data. We also conducted an online survey to identify how certain people want to be about their diagnosis before starting a gluten free diet. All this information was used to inform the development of economic models to identify the cost-effectiveness of different active case finding strategies.

Findings

People with dermatitis herpetiformis, family history of CD, migraine, anaemia, type 1 diabetes, osteoporosis, or chronic liver disease are 1.5 to 2 times more likely to have CD. The prediction models showed good discrimination between patients with and without CD but performed less well when externally validated. For children, the strongest predictors for having CD were type 1 diabetes, Turner syndrome, IgA deficiency, or a first-degree relative with CD; for adults these were a first-degree relative with CD or anaemia.

Serological tests have good diagnostic accuracy for CD. IgA tTG had the highest sensitivity and EMA had highest specificity. Genetic tests (HLA DQ2/8) had a very high sensitivity but low specificity, suggesting they would be useful tests to rule out CD.

Survey respondents wished to be 66% certain of the diagnosis from a blood test before starting a gluten-free diet if symptomatic, and 90% certain if asymptomatic.

Cost-effectiveness analyses found that, in adults, IgA tTG at a 1% pre-test probability (equivalent to population screening) was most cost-effective. For strategies that do not involve population screening, IgA EMA plus HLA was most cost effective at pre-test probabilities of 1.5% in adults and 5% in children. There was substantial uncertainty in economic model results and high value in conducting further research.

Consequences

The most cost-effective strategy in adults appears to be population based screening with IgA tTG. However, decisions to implement this cannot be made based on our economic analysis alone. Future work should consider whether population based screening for CD could meet the UK National Screening Committee criteria and requires a long-term RCT of screening strategies.

Submitted by: 
Martha Elwenspoek
Funding acknowledgement: 
Funding for this study was provided by the Health Technology Assessment programme of the National Institute for Health Research (NIHR129020).