Table 3. Final multinomial logistic regression model including the variables associated with the condition of frailty in a population of community-dwelling elderly people. Uberaba (MG), Brazil, 2014-2016 (n = 353).
| Variables | Frailty syndrome | |||||
|---|---|---|---|---|---|---|
| Pre-frail | Frail | |||||
| OR | 95% CI | P | OR | 95% CI | P | |
| Age range (in years) | ||||||
| 60-69 | 1 | 1 | ||||
| 70-79 | - | - | ||||
| 80 or over | 4.92 | 1.57-15.38 | 0.006 | 8.64 | 2.05-36.35 | 0.003 |
| BADL | ||||||
| Dependent | - | 3.66 | 1.22-11.02 | 0.021 | ||
| Independent | 1 | 1 | ||||
| AADL | - | 0.85 | 0.74-0.99 | 0.037 | ||
| Physical performance (SPPB) | ||||||
| Very poor | - | - | ||||
| Poor | - | 7.87 | 1.97-31.39 | 0.003 | ||
| Moderate | - | - | ||||
| Good | 1 | 1 | ||||
OR = odds ratio; 95% CI = 95% confidence interval; P < 0.05; 1 = reference category - non-frail group; BADL = basic activities of daily living; AADL = advanced activities of daily living; SPPB = Short Physical Performance Battery.