Pablo Millares-Martin 1 Show Affiliations »
Abstract
BACKGROUND: Primary care in UK is expected to use tools such as the electronic Frailty Index (eFI) to identify patients with frailty, which should be then validated and coded accordingly. AIM: To assess the influence of organisation and software on how eFI score and direct clinical validation occurs across practices in Leeds. METHOD: The 'minimum necessary' anonymised patient data required for the study (recorded eFI scores and frailty codes - mild, moderate or severe - with their dates of entry) was requested to the Health and Care Hub of the NHS Leeds Clinical Commissioning Group. Data from 44 185 patients from 104 practices using two different clinical software were collected. Descriptive statistics was carried out using SPSS software. RESULTS: 42 593 patients had a frailty code, 8881 had an eFI code. 7341 had both types of entry, and correlation between eFI and coded level of frailty was as expected high (85.3%), but there was statistically significant variation depending on practice and software used. When results did not match, there was a tendency to overstate, to code a level of frailty above the value to be assigned based on the numeric value of eFI, and it was more so on those practices using SystmOne software compared with those using EMIS Web. CONCLUSIONS: Although correlation was generally good, the variability encountered would indicate the need for training and also for software improvements to reduce current disparity and facilitate validation, so frailty level is adequately recorded. © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.
BACKGROUND: Primary care in UK is expected to use tools such as the electronic Frailty Index (eFI) to identify patients with frailty, which should be then validated and coded accordingly. AIM: To assess the influence of organisation and software on how eFI score and direct clinical validation occurs across practices in Leeds. METHOD: The 'minimum necessary' anonymised patient data required for the study (recorded eFI scores and frailty codes - mild, moderate or severe - with their dates of entry) was requested to the Health and Care Hub of the NHS Leeds Clinical Commissioning Group. Data from 44 185 patients from 104 practices using two different clinical software were collected. Descriptive statistics was carried out using SPSS software. RESULTS: 42 593 patients had a frailty code, 8881 had an eFI code. 7341 had both types of entry, and correlation between eFI and coded level of frailty was as expected high (85.3%), but there was statistically significant variation depending on practice and software used. When results did not match, there was a tendency to overstate, to code a level of frailty above the value to be assigned based on the numeric value of eFI, and it was more so on those practices using SystmOne software compared with those using EMIS Web. CONCLUSIONS: Although correlation was generally good, the variability encountered would indicate the need for training and also for software improvements to reduce current disparity and facilitate validation, so frailty level is adequately recorded. © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.
Entities: Disease
Species
Keywords:
electronic frailty index; electronic health record; frailty; primary care
Mesh： See more »
Year: 2019
PMID： 31039123 DOI： 10.1136/bmjhci-2019-000024
Source DB: PubMed Journal: BMJ Health Care Inform ISSN： 2632-1009