A reliable and record-based frailty assessment method for older medical inpatients

Abstract

Purpose: The comprehensive geriatric assessment (CGA) including frailty assessment is considered the gold standard of assessment in geriatric patients. The Multidimensional Prognostic Index (MPI) is a CGA-based bedside assessment tool. Older medical inpatients' medical records comprehensively describe the MPI-featured components. Consequently, MPI-based frailty assessment may be accomplished retrospectively. We found no previous studies concerning record-based MPI. We studied the reproducibility and diagnostic accuracy of a record-based MPI.

Methods: The study was designed as a fully crossed, prospective, and cross-sectional study. A total of 50 inpatients aged ≥ 75 years were included from two medical wards. Record-based MPI was assessed by two independent raters in patients who required personal assistance on a daily basis or had a Charlson Comorbidity Index (CCI) ≥ 1. In the same patients, a bedside MPI rating was performed. Inter-rater and inter-method reproducibility and diagnostic accuracy measures were calculated.

Results: Evaluating the inter-rater reproducibility; the mean difference was -0.02 points [95% confidence interval (CI) - 0.06 to 0.01, p = 0.20]. Intraclass correlation coefficient (ICC) was 0.71. Evaluating inter-method reproducibility; the mean difference was -0.02 (95% CI - 0.04 to 0.01, p = 0.18); ICC = 0.83. Sensitivity was 100% and specificity 80%. The areas under the receiver operating curves (ROC) was 0.92 (95% CI 0.75-1.00) and 0.77 (95% CI 0.52-1.00).

Conclusion: The record-based MPI rating method has an acceptable inter-rater reliability, good inter-method reliability, and high agreement as compared to the bedside-rated MPI. The diagnostic accuracy seems considerable. The record-based MPI seems useful in retrospective frailty assessment among older medical inpatients.

Keywords: Comprehensive geriatric assessment; Diagnostic accuracy; Frailty; Medical record review; Multidimensional Prognostic Index; Reproducibility.