Quality and Impact of Real-World Data in Chronic Disease Research: Insights From Retrospective Studies Using the Hospital Episode Statistics (HES) Database
Author(s)
Evelyn Gomez Espinosa, BSc, PhD1, Maria Koufopoulou, MSc2, Kimberly M. Ruiz, EdM3.
1Global Consulting Services, Cencora, Bishops Stortford, United Kingdom, 2Global Consulting Services, Cencora, Reading, United Kingdom, 3Global Consulting Services, Cencora, Vista, CA, USA.
1Global Consulting Services, Cencora, Bishops Stortford, United Kingdom, 2Global Consulting Services, Cencora, Reading, United Kingdom, 3Global Consulting Services, Cencora, Vista, CA, USA.
OBJECTIVES: Real-world data from administrative medical records is essential for assessing healthcare quality and outcomes. Using large databases such as the HES database, a primary repository for NHS hospitals, can be critical for researching and developing effective interventions, informing public health strategies, and optimizing healthcare resources for chronic diseases. This review evaluates the publication landscape and quality of retrospective studies using HES data.
METHODS: A comprehensive search was conducted in Embase, Emcare, and MEDLINE on June 2025 for retrospective cohort studies published in English since 2021 that utilized HES data. The review examined study characteristics, including publication type, journal, database linkage, statistical analysis, disease area, and reported outcomes.
RESULTS: Out of 541 reviewed abstracts, 143 met the inclusion criteria following review of full-text publications. The annual publication rate remained steady from 2021 to 2024, averaging 33 per year, nearly half of which were published only as conference abstracts. The average impact factor of journals publishing this research was significantly higher for conference abstracts compared to full-text articles (11.1 vs 4.4). A total of 64 conditions were evaluated, with a predominant focus on respiratory diseases (23%) followed by cardiovascular diseases (17%) and rheumatic diseases (12%). Most studies reported on epidemiology (90%), risk factors for morbidity or progression (87%), and healthcare-resource-utilization (83%). While most studies linked HES data with other sources—particularly the Clinical Practice Research Datalink (CPRD) in 85% of the studies— the lesser use of advanced statistical methods, such as multilevel hierarchical modeling or other confounder adjustments, raises concerns regarding the depth and validity of the analyses.
CONCLUSIONS: This review highlights the significant role of large administrative databases in advancing research on chronic diseases and their healthcare implications in recent years. Future research should prioritize dataset linkage and confounder adjustment to enhance the reliability and comprehensiveness of healthcare outcome assessments using RW data.
METHODS: A comprehensive search was conducted in Embase, Emcare, and MEDLINE on June 2025 for retrospective cohort studies published in English since 2021 that utilized HES data. The review examined study characteristics, including publication type, journal, database linkage, statistical analysis, disease area, and reported outcomes.
RESULTS: Out of 541 reviewed abstracts, 143 met the inclusion criteria following review of full-text publications. The annual publication rate remained steady from 2021 to 2024, averaging 33 per year, nearly half of which were published only as conference abstracts. The average impact factor of journals publishing this research was significantly higher for conference abstracts compared to full-text articles (11.1 vs 4.4). A total of 64 conditions were evaluated, with a predominant focus on respiratory diseases (23%) followed by cardiovascular diseases (17%) and rheumatic diseases (12%). Most studies reported on epidemiology (90%), risk factors for morbidity or progression (87%), and healthcare-resource-utilization (83%). While most studies linked HES data with other sources—particularly the Clinical Practice Research Datalink (CPRD) in 85% of the studies— the lesser use of advanced statistical methods, such as multilevel hierarchical modeling or other confounder adjustments, raises concerns regarding the depth and validity of the analyses.
CONCLUSIONS: This review highlights the significant role of large administrative databases in advancing research on chronic diseases and their healthcare implications in recent years. Future research should prioritize dataset linkage and confounder adjustment to enhance the reliability and comprehensiveness of healthcare outcome assessments using RW data.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
RWD139
Topic
Real World Data & Information Systems
Disease
Cardiovascular Disorders (including MI, Stroke, Circulatory), No Additional Disease & Conditions/Specialized Treatment Areas, Respiratory-Related Disorders (Allergy, Asthma, Smoking, Other Respiratory)