Validating Electronic Frailty Index in National Health System
Dr. Fabienne Hershkowitz Sikron from the Meuhedet Health Maintenance Organization (HMO) in Tel-Aviv, discusses a research paper she co-authored that was published in Volume 16, Issue 20 of Aging, entitled “Development and validation of an electronic frailty index in a national health maintenance organization.”
DOI - https://doi.org/10.18632/aging.206141
Transcript
Listen to an audio version of this interview
My name is Fabienne Hershkowitz, an epidemiologist at Meuhedet Health Maintenance Organization, which provides community health services to over 1,300,000 people in Israel. This work is in collaboration with Joint-Eshel and the Geriatric Division of the Ministry of Health. The paper describes the development of an electronic frailty index based on the English electronic frailty index and the validation of this index.
Frailty is a geriatric syndrome characterized by a decline of physiological reserves, and it leads to an increased risk of negative outcome such as hospitalization and immortality. As a community healthcare provider, we are required to deliver optimal care under limited resources. Since it has long been clear that age is just one factor affecting health, and that people of the same age can be very different in terms of their physical and mental resources, our goal was to classify the adult population in terms of frailty so that we could intervene with each patient in the most suitable way.
During the development phase, we encountered many decision points, such as which deficits would we include in the index and from which health service providers we would retrieve the information for each deficit, such as doctors, nurses, paramedical health providers, pharmacists, social workers. The selection of deficits was based on several existing frailty indices with the guiding principle being that the index would be holistic and would reflect physical aspects such as chronic diseases, as well as functional aspects, such as activity limitation, or mental aspects, such as anxiety, or social aspects, such as loneliness.
During the validation phase, a necessary step to justify the use of the index, we examined the relationship between the index and mortality and hospitalization within a year. We also examined how well the index aligns with the nursing benefits awarded by National Social Security. We showed that the frailty index we developed converts with the nursing benefits and adds significant predictive value for hospitalization and mortality beyond age and beyond the Charleston Comorbidity Index after controlling for co-variants.
Since 2024, we are using the frailty index in practice at the Health Maintenance Organization when planning an intervention program. For example, we have launched projects that provides a support package for patients discharged from hospitalization, and we have chosen to focus on individuals with mild frailty. The reason was, that on the one hand, mild frailty patients are at greater risk of deterioration after hospitalization compared to fit patients. But on the other hand, they are more likely to recover with our support than the moderate and severe frailty patients. The frailty index has been integrated into our information system so that it is reflected down to each health clinic for proactive outreach to patients according to their frail status.
Currently, we are working on building a screen that will reflect to the primary physician the frailty status of the elderly patient coming for a visit, and the deficits he suffers from, so that it will be taken into account immediately during the diagnosis phase and in choosing the treatment program.
I want to thank our partners from the Joint-Eshel organization and the Ministry of Health, as well as our colleagues, especially the medical and paramedical managers, the geriatrics specialist, the information systems experts, the data analysis, and the senior citizen department. Thanks to Aging for the opportunity to publish the paper and present it in the current digital format. Thank you.
