Table 2.
Logistic regression analyses for predicting nutritional risk among urban community-dwelling older adults†, Chicago, IL, USA, August–October 2017
| OR | 95 % CI | b | P | |
|---|---|---|---|---|
| Intercept | – | −1·507 | 0·16 | |
| Age, year | 0·97 | 0·94, 1·00 | −0·034 | 0·02 |
| Female (ref.=male) | 1·69 | 1·13, 2·51 | 0·522 | 0·01 |
| Race (ref.=White) | 0·0004 | |||
| Black | 2·41 | 1·55, 3·73 | 0·878 | <0·0001 |
| Hispanic | 1·85 | 1·09, 3·14 | 0·616 | 0·02 |
| Education (ref.=college graduate) | 0·09 | |||
| Not HS graduate | 1·68 | 0·94, 3·01 | 0·521 | 0·08 |
| HS graduate/GED | 1·17 | 0·69, 1·98 | 0·155 | 0·57 |
| Some college | 1·69 | 1·07, 2·66 | 0·523 | 0·02 |
| BMI, kg/m2 | 1·01 | 0·99, 1·04 | 0·011 | 0·41 |
| Dependent for ≥1 ADL (ref.=independent for all 6 ADL) | 1·36 | 0·72, 2·59 | 0·308 | 0·35 |
| ER visit (ref.=none) | 2·50 | 1·56, 4·00 | 0·915 | 0·0001 |
| Hospital admission (ref.=none) | 1·10 | 0·63, 1·90 | 0·091 | 0·74 |
| Home care services (ref.=none) | 1·22 | 0·69, 2·14 | 0·196 | 0·50 |
| Eat alone most of the time (ref.=no) | 2·36 | 1·66, 3·36 | 0·858 | <0·0001 |
| Take three or more different prescribed or OTC drugs per day (ref.=no) | 2·33 | 1·59, 3·42 | 0·845 | <0·0001 |
ref., reference category; HS, high school; GED, General Equivalency Diploma; ADL, activities of daily living; ER, emergency room; OTC, over-the-counter.
†
From a logistic regression model predicting the risk of an abridged Patient Generated Subjective Global Assessment (abPG-SGA) score of ≥6 points (high nutrition risk). Some observations were excluded from the model due to missing data for covariates: N 810 (210 were high risk).