Validity of UK electronic health records to study migrant health: a population-based cohort study. (30th September 2020)

• Record Type: Journal Article
• Authors: Pathak, N; Patel, P; Mathur, R; Burns, R; Gonzalez-Izquierdo, A; Denaxas, S; Sonnenberg, P; Hayward, A; Aldridge, R
• Is Part Of: European journal of public health. Volume 30(2020)Supplement 5
• Publication Date: 2020-09-30
• Subjects: Epidemiology -- Europe -- Periodicals; Public health -- Europe -- Periodicals; 362.109405
• DOI: 10.1093/eurpub/ckaa166.768
• Languages: English
• ISSNs: 1101-1262

Abstract

Abstract: Background: An estimated 14.3% (9.4 million people) of people living in the UK in 2019 were international migrants. Despite this, little is known about how migrants access and use healthcare services. To use electronic healthcare records (EHRs) to study migration health, a valid migration phenotype is necessary: a transparent reproducible algorithm using clinical terminology codes to determine migration status. We have previously described the validity of a migration phenotype in CALIBER data using the Clinical Practice Research Datalink (CPRD), the largest UK primary care EHR. This study further evaluates the phenotype by examining certainty of migration status. Methods: This is a population-based cohort study of individuals in CPRD Gold (1997-2018) with a Read term indicating migration to the UK. We describe completeness of recording of migration over time: percentage of individuals recorded as migrants. We also describe cohort size based on certainty of migration status: "definite" (country of birth or visa status terms), "probable" (non-English first/main language terms), and "possible" (non-UK origin terms). Results: Overall, 2.5% (403, 768/16, 071, 111) of CPRD had ≥1 of 434 terms indicating migration to the UK. The percentage of recorded migrants per year increased from 0.2% (4, 417/2, 210, 551) in 1997 to 3.64% (100, 626/2, 761, 397) in 2018, following a similar trend to national migration data. 44.27% (178, 749/403, 768) were "definite" migrants and 53.68% (216, 731/403, 768) were "probable" migrants. Only 2.05%(8, 288/16, 071, 111) were "possible" migrants. Conclusions: We have created a large cohort of international migrants in the UK and certainty of migration status is high. This cohort can be used to study migration health in UK primary care EHR. The large contribution of language terms make this phenotype particularly suitable for understanding healthcare access and use by non-English speaking migrants who may face additional barriers to care. Key messages: We have developed a way to study migration health in UK primary care electronic health records. Our method is particularly useful to study healthcare for non-English speaking migrants who may face additional barriers to care.