How effective are population health surveys for estimating prevalence of chronic conditions compared to anonymised clinical data?

• Main Authors: Tony Whiffen, Ashley Akbari, Tony Paget, Sarah Lowe, Ronan Lyons
• Format: Article
• Language: English
• Published: Swansea University 2020-06-01
• Series: International Journal of Population Data Science
• Online Access: https://ijpds.org/article/view/1151

Population health surveys are used to record person-reported outcome measures for chronic health conditions and provide a useful source of data when evaluating potential disease burdens. The reliability of survey-based prevalence estimates for chronic diseases is unclear nonetheless. This study applied methodological triangulation to validate prevalence of selected chronic conditions (angina, myocardial infarction, heart failure, and asthma) using data for a combined cohort of 11,323 adults from the 2013 and 2014 sweeps of the Welsh Health Survey (WHS). This approach utilised consented survey data linked to primary and secondary care electronic health record (EHR) data within the Secure Anonymised Information Linkage (SAIL) Databank.  Validation of self-reported chronic conditions using data linkage and clinical codes is demonstrated with varied success. Point prevalence based on survey data was shown to be slightly under-reported for all four conditions when compared with clinical data. Case identification for separate cardiovascular conditions was problematic without use of specific medication codes for each condition, but was more straightforward for asthma, where there was an extensive list of medications available. Whilst the results provide external validity for the WHS as an instrument for assessing disease burdens for chronic conditions, they also indicate that comparable prevalence estimates can be produced using clinical data if a condition-specific set of clinical codes are available.