Abstract
BACKGROUND: Frailty is a state of increased vulnerability to poor resolution of homeostasis following a stress. Early diagnosis and intervention of frailty are essential to prevent its adverse outcomes. However, simple diagnostic criteria have not been established. The Questionnaire for Medical Checkup of Old-Old (QMCOO) is widely used for medical checkups of older adults in Japan. In our previous report, we developed a method to score the QMCOO and showed that frailty can be diagnosed with the highest accuracy when the score cutoff was set at 3/4 points. We aimed to validate the criteria in a larger cohort. METHODS: Participants aged 65 years or over were recruited in the western region of Japan. They answered all the items of the Kihon Checklist (KCL) and the QMCOO. Based on the KCL score, they were diagnosed as robust (3 or lower), prefrail (4 to 7), or frail (8 or over). Then we tested the effectiveness to diagnose frailty using the QMCOO cutoff of 3/4 points. We also aimed to determine the score cutoff to separate robust and prefrail. RESULTS: 7,605 participants (3,458 males and 4,147 females, age 77.4 ± 6.9 years) were recruited. 3,665 participants were diagnosed as robust, 2,448 were prefrail, and 1,492 were frail based on the KCL score. The diagnosis of frailty had a sensitivity of 84.0%, specificity of 82.5%, and accuracy of 82.8% with a QMCOO score cutoff of 3/4 points, suggesting its validity. To separate robust and prefrail, both the accuracy and the Youden index were the highest with the QMCOO cutoff of 2/3 points (sensitivity, specificity, and accuracy were 63.9%, 83.4%, and 75.6%, respectively). All the questions of the QMCOO except Q12 (about smoking) were significantly related to prefrailty status after a logistic regression analysis. CONCLUSION: Diagnosis of frailty using the QMCOO score cutoff of 3/4 points was validated. Prefrailty could be diagnosed using the score cutoff of 2/3 points.