Table 3.
Multiple linear regression model testing the association between volunteer activities and depression among urban and rural elderly.
| Model 1 | Model 2 | Model 3 | ||
| Type of volunteer activity | B (SE) | B (SE) | B (SE) | |
| Urban | Formal voluntary activities (ref. No participation) | |||
| Not regularly | −1.03 (0.68) | −0.19 (0.67) | −0.02 (0.59) | |
| Almost every week | −2.08 (1.21) | −1.70 (1.18) | −0.74 (1.04) | |
| Almost daily | −4.16 (1.21)† | −4.04 (1.19)† | −2.69 (1.05)∗ | |
| Informal voluntary activities | ||||
| Cared for a sick or disabled adult (ref. No participation) | ||||
| Not regularly | 0.67 (1.02) | 0.47 (0.99) | 0.91 (0.87) | |
| Almost every week | 0.19 (2.43) | 1.20 (2.38) | 1.66 (2.10) | |
| Almost daily | 2.66 (1.78) | 3.01 (1.75) | 3.13 (1.54)∗ | |
| Provided help to family, friends or neighbors (ref. No participation) | ||||
| Not regularly | −0.57 (0.48) | −0.29 (0.47) | 0.02 (0.42) | |
| Almost every week | −1.57 (1.03) | −0.77 (1.01) | −0.07 (0.89) | |
| Almost daily | −0.25 (1.08) | −0.32 (1.06) | −0.04 (0.93) | |
| Rural | Formal voluntary activities (ref. No participation) | |||
| Not regularly | −1.77 (0.70)∗ | −0.97 (0.68) | −0.46 (0.59) | |
| Almost every week | −2.83 (1.13)∗ | −2.11 (1.09) | −1.45 (0.96) | |
| Almost daily | 0.05 (1.54) | 0.55 (1.49) | 0.93 (1.31) | |
| Informal voluntary activities | ||||
| Cared for a sick or disabled adult (ref. No participation) | ||||
| Not regularly | 0.19 (0.67) | 0.11 (0.64) | 0.01 (0.56) | |
| Almost every week | 1.60 (1.52) | 1.39 (1.47) | −0.18 (1.28) | |
| Almost daily | 2.83 (1.40)∗ | 2.70 (1.36)∗ | 2.56 (1.18)∗ | |
| Provided help to family, friends or neighbors (ref. No participation) | ||||
| Not regularly | −0.43 (0.28) | −0.13 (0.28) | 0.26 (0.24) | |
| Almost every week | −0.76 (0.64) | −0.40 (0.62) | −0.37 (0.54) | |
| Almost daily | −1.99 (0.86)∗ | −2.02 (0.83)∗ | −1.10 (0.73) |
∗
P < .05.
†
P < .01.
B = coefficient, BADL = basic activities of daily living, IADL = instrumental activities of daily living, SE = standard error.
Model 1: crude model. Model 2: adjusted for age, gender, marital status, education level, smoking and drinking. Model 3: adjusted for age, gender, marital status, education level, smoking, alcohol consumption, self-rated health, chronic diseases, BADL, IADL and cognitive function.