Table 4.
Odds ratios for BMI and prevalence rates of frailty defined by the FP model and the KCL, calculated using multivariate logistic regression a..
| BMI Categories (kg/m2) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| <18.5 (n = 562) | 18.5–19.9 (n = 787) | 20.0–22.4 (n = 2195) | 22.5–24.9 (n = 2218) | 25.0–27.4 (n = 1006) | ≥27.5 (n = 423) | |||||||
| BMI (kg/m2) | 17.3 | (1.0) | 19.3 | (0.4) | 21.3 | (0.7) | 23.6 | (0.7) | 26.0 | (0.7) | 29.5 | (2.1) |
| FP model | ||||||||||||
| Case (n (%)) | 142 | (25.3) | 154 | (19.6) | 313 | (14.3) | 276 | (12.4) | 127 | (12.6) | 82 | (19.4) |
| Crude model | 2.34 | (1.84 to 2.97) | 1.69 | (1.34 to 2.12) | 1.15 | (0.96 to 1.38) | 1.00 | (Ref) | 0.98 | (0.78 to 1.25) | 1.66 | (1.25 to 2.20) |
| Model 1 b | 2.07 | (1.62 to 2.65) | 1.63 | (1.29 to 2.05) | 1.13 | (0.94 to 1.36) | 1.00 | (Ref) | 1.01 | (0.80 to 1.29) | 1.73 | (1.30 to 2.31) |
| Model 2 c | 2.04 | (1.58 to 2.63) | 1.69 | (1.33 to 2.14) | 1.16 | (0.96 to 1.41) | 1.00 | (Ref) | 1.00 | (0.78 to 1.27) | 1.54 | (1.15 to 2.07) |
| KCL | ||||||||||||
| Case (n (%)) | 312 | (55.5) | 297 | (37.7) | 750 | (34.2) | 723 | (32.6) | 345 | (34.3) | 208 | (49.2) |
| Crude model | 2.58 | (2.14 to 3.12) | 1.25 | (1.06 to 1.48) | 1.07 | (0.95 to 1.22) | 1.00 | (Ref) | 1.08 | (0.92 to 1.26) | 2.00 | (1.62 to 2.47) |
| Model 1 b | 2.08 | (1.70 to 2.54) | 1.13 | (0.94 to 1.35) | 1.01 | (0.88 to 1.15) | 1.00 | (Ref) | 1.10 | (0.93 to 1.30) | 2.13 | (1.71 to 2.66) |
| Model 2 c | 2.04 | (1.65 to 2.54) | 1.16 | (0.96 to 1.41) | 1.03 | (0.90 to 1.19) | 1.00 | (Ref) | 1.07 | (0.90 to 1.28) | 1.86 | (1.47 to 2.35) |
BMI, body mass index; FP, Fried phenotype; KCL, Kihon Checklist; Ref, reference. a The prevalence of frailty defined by FP model and the KCL in the Kyoto–Kameoka Study were 15.2% (1094 people) and 36.6% (2635 people), respectively. BMI values are shown as means and standard deviations. The prevalence rates of frailty are shown as numbers of people and percentages. Statistical values regarding the relationship between BMI and the prevalence of frailty are shown as odds ratio and 95% confidence intervals. b Model 1 was adjusted for age, sex, and region. c Model 2 was adjusted for age, sex, region, smoking habit, alcohol consumption, education history, number of drugs taken, family composition, economic status, physical activity, presence/absence of dentures, and history of hypertension, stroke, heart disease, diabetes, and dyslipidemia.