Testing a data-based definition of severity in atopic dermatitis in a linked primary and secondary care dataset in England: Implications on clinical outcomes and healthcare resource use

Study type
Protocol
Date of Approval
Study reference ID
18_318
Lay Summary

Atopic dermatitis (AD) is a common condition where individuals develop itchiness, redness, and swelling of the skin. AD may flare up due to reasons that vary from person to person, such as changes in weather, exposure to pollen, pet dander, certain types of food or food components, or even for no apparent reason at all. The cause of this sensitivity is not yet fully understood, but it appears that your genes may play a role, as the condition seems to be passed on from one generation to another.

This study aims to determine whether a definition of how severe AD is can be created using just data from health records. This study will first explore whether this data-based definition will successfully reflect a difference in outcomes. For example, we would like to see whether severe AD as defined using data will result in those patients using up larger healthcare resources, such as more admissions to hospital, or more outpatient appointments.

With a reliable definition of AD severity, we can utilise this definition in further studies on AD using datasets, which present a more practical, faster, and less expensive way of performing such research. Assessment of severity would also be useful for policy change and resource allocation in England.

Technical Summary

Atopic dermatitis is a highly prevalent condition in the United Kingdom. Datasets have become very useful in assessing the impact of various diseases, including AD, on the National Health Service. This effort is particularly important in the face of tightening budgets for healthcare in the country.

The definition of conditions in various datasets is an important task to improve the reliability and accuracy of estimates of healthcare resource use and clinical outcomes. A poor definition would result in an unreliable picture on which to base health policy and intervention in the health system.

This study aims to determine whether severity of AD as defined through codes in datasets have the increased cost and clinical outcome implications that are expected with clinical definitions. We plan to use a definition combining the use of immunosuppressants, systemic corticosteroids, HRG tariffs with a certain complication score, and atopic dermatitis as a primary inpatient admission diagnosis as criteria defining a cohort of patients with moderate to severe AD. All other AD shall be classified as mild.

We shall then measure prevalence, patient demographics and clinical profiles, co-morbidities, clinical outcomes, healthcare resource use and costs for each of these cohorts. The goal is to determine if there is a significant difference in these outcomes between the mild AD, and moderate to severe AD. A significantly higher healthcare resource use, and worse clinical outcomes would indicate that the data-based definition is functionally viable and may be used for further research endeavours on these datasets.

Health Outcomes to be Measured

Prevalence (prevalence of AD, prevalence of mild AD, prevalence of moderate to severe AD, prevalence of co-morbidities including asthma, allergic conjunctivitis, allergic rhinitis, allergic urticaria, anxiety, cutaneous bacterial infection, depression, eosinophilic oesophagitis, food allergies), demographics (total patients, age, percent males, family history of atopy, time in cohort, follow-up), clinical outcomes (remission rate, absolute refractory rate, relapse rate, skin infection rate), healthcare resource use outcomes (inpatient admissions, inpatient length of stay, 30-day readmission rate, inpatient HRG tariffs, outpatient appointments, outpatient HRG tariffs, A&E attendances, GP appointments, nursing appointments in primary care, diagnostic tests in primary care, referrals in primary care, medications in primary care, consultation costs in primary care, prescription costs in primary care)

Collaborators

Adrian Paul J. Rabe - Chief Investigator - Health iQ Ltd ( UK ) t/a CorEvitas
Adrian Paul J. Rabe - Corresponding Applicant - Health iQ Ltd ( UK ) t/a CorEvitas
Hassan Chaudhury - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas

Former Collaborators

Austen El-Osta - Collaborator - Imperial College London

Linkages

HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data