Using linked healthcare records to identify the prevalence, characteristics and clinical outcomes of patients with Severe Eosinophilic Asthma in Northern Ireland

Problem

Severe Asthma (SA) is asthma that does not respond to standard treatment and is characterized by one or more exacerbations in the previous year despite high dose inhaled or oral corticosteroids. This patient subgroup is difficult to diagnose and associated with significant morbidity and preventable premature mortality. There is evidence of improved clinical outcomes if these patients can be identified and offered specialist assessment and management including novel monoclonal antibody treatment. However, there is significant inequality in access to specialist asthma care across the United Kingdom, with unacceptable variation in prevalence, frequency of exacerbations, and health outcomes across geography, age, ethnicity and socio-economic groups. Data linkage presents a unique opportunity to identify this high-risk subgroup. Asthma patients have healthcare records held in various databases throughout the Health and Social Care system. Primary Care healthcare records have many unique properties which make them essential to gain an accurate picture of SA clinical burden. These include the representativeness of the entire asthma population and the high quality of clinical data coding, in particular for Quality and Outcome Framework (QOF) data. An unrealised opportunity may exist via data linkage with a variety of other datasets (including those from secondary care) to identify those likely to have SA.

Approach

The General Practice Intelligence Platform (GPIP) is being established to extract primary care data from all GP practices in Northern Ireland under agreed governance processes. Data will be extracted and de-identified at source. A secured data governance and linkage framework is being developed to incorporate primary healthcare data into established systems for linkage of secondary care records.We plan to use this process to link primary healthcare record data from the Northern Ireland asthma population including QOF, prescribing and laboratory data to secondary healthcare record datasets including admissions and emergency department attendances. We aim to use this linked dataset to determine the prevalence, characteristics and clinical outcomes of patients with SA in Northern Ireland and identify potential candidates for monoclonal antibody treatment. Linkage of these records into a single anonymised dataset presents a unique opportunity to better understand this high-risk subgroup and to shed light on how inequalities in care and outcomes might be mitigated.

Findings

In preparation for this study we have conducted a scoping review on how primary care records linked to secondary and administrative data can help identify the prevalence and characteristics of SA. Results from this review have been used to guide the design of our study.

Consequences

This novel proof of concept study, which plans to link primary and secondary healthcare records of an entire asthma patient population, could lead to improved clinical outcomes and pave the way for similar linkage studies for other conditions.