Health-adjusted life expectancy (HALE) in Chongqing, China, 2017: An artificial intelligence and big data method estimating the burden of disease at city level

Abstract

Background A universally applicable approach that provides standard HALE measurements for different regions has yet to be developed because of the difficulties of health information collection. In this study, we developed a natural language processing (NLP) based HALE estimation approach by using individual-level electronic medical records (EMRs), which made it possible to calculate HALE timely in different temporal or spatial granularities. Methods We performed diagnostic concept extraction and normalisation on 13•99 million EMRs with NLP to estimate the prevalence of 254 diseases in WHO Global Burden of Disease Study (GBD). Then, we calculated HALE in Chongqing, 2017, by using the life table technique and Sullivan's method, and analysed the contribution of diseases to the expected years “lost” due to disability (DLE). Findings Our method identified a life expectancy at birth (LE 0 ) of 77•9 years and health-adjusted life expectancy at birth (HALE 0 ) of 71•7 years for the general Chongqing population of 2017. In particular, the male LE 0 and HALE 0 were 76•3 years and 68•9 years, respectively, while the female LE 0 and HALE 0 were 80•0 years and 74•4 years, respectively. Cerebrovascular diseases, cancers, and injuries were the top three deterioration factors, which reduced HALE by 2•67, 2•15, and 1•19 years, respectively. Interpretation The results demonstrated the feasibility and effectiveness of EMRs-based HALE estimation. Moreover, the method allowed for a potentially transferable framework that facilitated a more convenient comparison of cross-sectional and longitudinal studies on HALE between regions. In summary, this study provided insightful solutions to the global ageing and health problems that the world is facing. Funding (2018YFC2000400).

Applying natural language processing to electronic medical records for estimating healthy life expectancy

“…The main sources of ill health were found to be cancer, injuries-particularly for the male population-and cerebrovascular disease. The HALE estimations, 68 • 9 and 74 • 4 years for the male and female population respectively, were found to be in line with previous estimations of HALE for Chinese populations, showing the feasibility of the overall approach [2] .…”

“…This method made it possible to create a joint representation of both the structured and the free text data in the included health records. Evaluation confirmed the importance of careful analysis of free text information in EMRs, as a considerable portion of the diseases mentioned in the free text notes were never recorded in the structured parts of the EMRs, and would be missed in the final analysis if only the structured data from the EMRs had been included [2] .…”

“…In health-adjusted life expectancy (HALE) in Chongqing, China, 2017: an artificial intelligence and big data method estimating the burden of disease at city level , Liang Xu and colleagues propose an alternative method for estimating prevalence of injury and disease, by leveraging information extracted from electronic medical records (EMRs) [2] . In their study, health records of close to a million patients were collected from a wide range of hospitals and health centres in Chongqing.…”

“…In their study, health records of close to a million patients were collected from a wide range of hospitals and health centres in Chongqing. These records included both structured data containing ICD-10 codes, and free text notes [2] .…”

“…Liang Xu and colleagues set a threshold for this similarity measure; if an entity from the health records and an ICD-10 descriptor were found to have a similarity score over this threshold, they were considered as representing the same concept [2] . This method made it possible to create a joint representation of both the structured and the free text data in the included health records.…”