JMIR Publications

ABSTRACT

Background:

The theory of cumulative deficits using big data to develop the multimorbidity frailty index (mFI) has become a widely accepted approach in public health and healthcare services. However, constructing the mFI using the most critical determinants and stratifying different risk groups with dose-response relationships remain major challenges in clinical practice.

Objective:

The final ML-mFI model contained 38 diseases/deficits in this study. Compared with conventional mFI, both indices had similar distribution patterns by age and sex; however, among people aged 65-69, the mean mFI and ML-mFI were 0.037 (standard deviation (SD) 0.048) and 0.0070 (SD 0.0254), respectively. The difference may result from discrepancies in the diseases/deficits selected in the mFI and the ML-mFI. A total of 86,133 subjects aged 65 to 100 years were included in this study and were categorized into 4 groups according to the level of the ML-mFI. Both the Kaplan-Meier survival curves and Cox models showed that the ML-mFI significantly predicted all outcomes of interest, including all-cause mortality, unplanned hospitalizations and all-cause ICU admissions, at 1, 5 and 8 years of follow-up (P<0.01). In particular, a dose-response relationship was revealed between the four ML-mFI groups and adverse outcomes.

Methods:

In this study, we used Taiwan’s National Health Insurance Research Database to develop machine learning multimorbidity frailty index (ML-mFI) using the theory of cumulative diseases/deficits of individual older person. Compared to conventional mFI by selecting diseases/deficits based on expert opinions, we adopted random forest to select most influential diseases/deficits to predict adverse outcomes for older people. To ensure the survival curves showed dose-responsive relationship with overlapping during the follow-up, we developed Distance Index and Coverage Index at any time point to categorize the ML-mFI of all subjects into robust, mild frailty, moderate frailty and severe frailty. Survival analysis was conducted to evaluate the ability of the ML-mFI in predicting adverse outcomes, such as unplanned hospitalizations, intensive care unit (ICU) admissions and mortality.

Results:

The final ML-mFI model contained 38 diseases/deficits in this study. Compared with conventional mFI, both indices had similar distribution patterns by age and sex; however, among people aged 65-69, the mean mFI and ML-mFI were 0.037 (standard deviation (SD) 0.048) and 0.0070 (SD 0.0254), respectively. The difference may be resulted from discrepancies of diseases/deficit selected in mFI and ML-mFI. A total of 86,133 subjects aged 65 to 100 years were included in this study and categorized into 4 groups according to the level of ML-mFI. Both the Kaplan-Meier survival curves and Cox models showed that ML-mFI significantly predicted all outcomes of interests, including all-cause mortality, unplanned hospitalization and all-cause ICU admission, among 1-, 5- and 8- years of follow-up. Particularly, a dose-responsive relationship were revealed between the four ML-mFI groups and adverse outcomes.

Conclusions:

The ML-mFI consists of 38 diseases/deficits that can successfully stratify risk groups associated with all-cause mortality, unplanned hospitalizations and all-cause ICU admissions in older people, which indicates that precise, patient-centered medical care can be a reality in an aging society.