Authors (including presenting author) :
CHAN NY (1)(2), LEUNG E(3)(4), LAW CB (1) (2), WONG YF B (2), LEE A (4)(6)(8), GUAN JL (5), CHEN YHF(3), CHING CC (4), WONG CSM (4), LAM O(4), HE Y(4), YAU TYS (4), LIU Y(4), ng CT (4), Yeoh EK (4), Hector TSANG H(6),CHOW YH (7)
Affiliation :
1. Department of Medicine & Geriatric, Princess Margaret Hospital, Kowloon West Cluster
2. Department of Medicine & Geriatric, North Lantau Hospital, Kowloon West Cluster
3. Department of Management Sciences, City University of Hong Kong
4. JC School of Public Health and Primary Care, The Chinese University of Hong Kong
5. EpitelligenceHK, Hong Kong
6. Department of Rehabilitation Science, Hong Kong Polytechnic University
7. Kwai Tsing Safe and Healthy City Association, Hong Kong
8. Centre for Health Education and Health Promotion, JC School of Public Health and Primary Care, The Chinese University of Hong Kong
Introduction :
MSK pain is responsible for the majority of ED visits. Yet, most diseases and disorders of the MSK system do not require inpatient care when appropriate outpatient care is provided (i.e. ambulatory care sensitive conditions). Nevertheless, MSK conditions such as OA and diagnoses that are grouped collectively under inflammatory arthropathy are of high prevalence among inpatient populations. For inflammatory arthropathy in particular, its high prevalence among patients is also on the rise the past few years. Recent studies have also shown that hospitalizations among those in the general population who suffered from inflammatory arthropathy were more likely found among those who were aged 50-64, which were younger than those who were hospitalized due to other MSK conditions. In addition, multi-chronic comorbidity and low SES have been separately linked to increased hospitalizations among those in the general population who suffer from inflammatory arthropathy, even more so than those suffering from other MSK conditions. However, while it has been shown that the heterogenous clinical and sociodemographic profiles are responsible for the diverse hospitalization outcome among those in the general population who suffer from inflammatory arthropathy, little is known about factors responsible for the heterogeneity in acute care needs and post-discharge outcome among inflammatory arthropathy patients at the inpatient population level. This gap is of note given that if the outcome-critical heterogeneity observed in samples of inflammatory arthropathy patients reported is representative of the integral aspect of their corresponding inpatient populations' heterogeneity in case-mixing and acute care utilization parameters, then the observed consistency and heterogeneity are of great implication to the targeted designs of acute care pathway to precision-driven discharge planning that optimize treatment outcomes.
Objectives :
A hybrid machine learning methodology was applied to identify the heterogeneity and consistency in clinical and acute and post acute service profiles, as well as post-discharge outcomes, among inflammatory arthropathy patients across the hospital networks (clusters) in Hong Kong.
Methodology :
To verify study 1's finding that inflammatory arthropathy is the diagnostic group to which patients whose case-mixing and acute care utilization parameters are representative of their aged 50-64 inpatient population fall, we applied the hybrid machine leaning methodology advanced in Study 1 to other municipalities where the flagship hospitals of all HK hospital networks (i.e. HK Hospital Clusters) are located. Hospital networks that include rural areas or remote island were excluded from the current analysis, resulting in four regional medical system that centered around one flagship hospital from each hospital network. In total, 276,552 admissions during 2019 were included in the current analysis.
Result & Outcome :
1) inflammatory arthropathy are the diagnostic case mix group of patients whose case-mixing and acute care utilization parameters configure one of the typical clinical and utilization profiles that are representative of aged 50-64 inpatient populations across all geographically distinct hospital clusters in HK whose catchment does not includes rural areas or remote islands; 2) notwithstanding the homogeneity among inflammatory arthropathy patients and patients who shared similar case-mixing and acute care utilization parameters with them, their 28-day rehospitalization outcomes differ as a function of the different post acute services patients in this homogeneous segment received - with patients who received none of the machine-selected service exhibited the greatest likelihood of rehospitalization; 3) Multi-chronicity, while not included as the case-mixing parameters that partitioned from the inpatient population the inflammatory arthropathy-representative patient segment, it was consistently greater in proportion among those aged 50-64 who rehospitalized within 28-days of discharge before having received any effective post acute services compared to their 65+ counterparts; and 4) among the aged 50-64 patients of inflammatory arthropathy and those who shared with them similar case-mixing and acute care utilization parameters, the rates of Multi-chronicity and 28-day rehospitalization across the hospital clusters (networks) were different in accordance with the studied clusters/networks' geographical municipalities' SES
To conclude, the results were discussed in terms of the clinical value added of identifying from population-level EHRs a consistent configuration of case-mixing and acute care utilization among aged 50-64 inpatient populations across different hospital networks, who share inflammatory arthropathy as their diagnostic group but present systematic heterogeneity in multichronicities, postacute services and 28-day rehospitalization outcomes.
