Characterising patients with Alzheimer's disease in England: a cohort study using data from Clinical Practice Research Datalink linked to Hospital Episode Statistics.

Abstract

To characterise demographics, comorbidities, co-medications and healthcare resource utilisation (HCRU) in patients newly diagnosed with Alzheimer's disease (AD) in England using published and validated AD algorithms. Observational cohort study. Real-world data in England from the Clinical Practice Research Datalink Aurum primary care database linked to the Hospital Episode Statistics secondary care database. Two disease-based algorithms (Imfeld <i>et al</i> and Douros <i>et al</i>) and one medication-based algorithm (Schroeder <i>et al</i>) were selected from the literature to identify patients with AD. The index period for patient selection was from 1 January to 31 December 2019. Patients were grouped into three non-mutually exclusive cohorts reflecting these algorithms (cohorts A-C), then patients aged ≥60 years were further grouped into three subcohorts (subcohorts 1-3). Subcohort 1 included 9826 patients, subcohort 2 included 10 265 patients and subcohort 3 included 7355 patients. Demographics, comorbidities, co-medications and HCRU present up to and including the date of cohort qualification were described. Across subcohorts 1-3, mean age at index ranged from 81 to 83 years, and most patients were female (59%-63%) and White (93%). Common comorbidities were hypertension (67%-71%), asthma and chronic obstructive pulmonary disease (60%-63%) and arthritis and osteoarthritis (51%-53%). Common co-medications were analgesics (87%-89%), systemic corticosteroids (83%-84%) and anti-inflammatory and anti-rheumatic agents (81%-82%). Up to 34% of patients had a specialist referral or visit, with up to 18% involving a neurologist. In the 12 months prior to and including the index date, up to 46% of patients had an emergency department visit or inpatient visit for any cause. In England, most newly diagnosed patients with AD were White females in their early 80s with common comorbidities including hypertension, asthma, chronic obstructive pulmonary disease, arthritis and osteoarthritis. This study addresses a critical evidence gap by quantifying the England-specific burden of comorbidity, polypharmacy, and HCRU among patients with AD using multiple validated and published AD algorithms. Findings were broadly consistent across the three subcohorts identified using disease-based and medication-based algorithms, strengthening internal validity of the study and demonstrating that the findings are robust to the method of AD case definition. This study provides a novel insight into the real-world AD population in England, and findings may help healthcare professionals to identify patients living with AD, understand patients' needs and tailor treatment strategies.