RESUMO
AIM: In recent years, the Kuchi-kara Taberu (KT) index has been used as a new tool for the evaluation of dysphagia; however, its predictive ability remains unknown. This study was conducted to examine the validity of oral intake recovery prediction using the KT index in patients with aspiration pneumonia. METHODS: The subjects were older inpatients admitted to an acute-care hospital for the treatment of aspiration pneumonia. A logistic regression analysis was conducted to clarify factors significantly associated with oral intake recovery. In addition, cut-off values of the predictors were calculated using a receiver operating characteristic curve analysis with the area under the curve (P < 0.05). RESULTS: A total of 124 patients (mean age 84.3 ± 7.9 years old, 70 males, 54 females) were enrolled in this study and divided into two groups at discharge: the oral intake group (62.1%) and the non-oral intake group (37.9%). The significant predictors of the recovery of oral intake were the pre-treatment period, treatment period, and the total score of the KT index; the respective odds ratios were 1.082 (95% confidence interval [CI] 1.013-1.156), 1.018 (1.003-1.033) and 0.850 (0.780-0.927), and the respective areas under the curve were 0.407 (95% CI 0.300-0.515), 0.304 (0.208-0.399), and 0.732 (0.640-0.824). The cut-off value of the total score of the KT index was 30. The model showed 85.7% sensitivity and 57.4% specificity. CONCLUSION: The KT index is suggested to be a valid variable for predicting whether or not a patient with aspiration pneumonia can recover their oral intake ability during acute-care hospitalization. Geriatr Gerontol Int 2023; 23: 221-226.