Table 3.
The association of frailty with different SP types based on fixed-effects model and time-varying Cox regression
| Fixed-effects model (n = 9422)a | Time-varying Cox model (n = 6073)b | |||
|---|---|---|---|---|
| Coefficient (95%CI) | P value | HR (95%CI) | P value | |
| IWF | ||||
| Occasional | 0.002 (-0.001, 0.004) | 0.270 | 0.96 (0.86, 1.08) | 0.496 |
| Weekly | -0.004 (-0.008, -0.001) | 0.009 | 0.99 (0.87, 1.13) | 0.902 |
| Daily | -0.008 (-0.010, -0.005) | < 0.001 | 0.89 (0.80, 0.99) | 0.031 |
| MCCC | ||||
| Occasional | -0.007 (-0.011, -0.003) | < 0.001 | 0.86 (0.74, 0.98) | 0.028 |
| Weekly | -0.010 (-0.014, -0.006) | < 0.001 | 0.86 (0.74, 1.00) | 0.056 |
| Daily | -0.014 (-0.019, -0.010) | < 0.001 | 0.68 (0.57, 0.80) | < 0.001 |
| DFQ | ||||
| Occasional | -0.008 (-0.016, 0.000) | 0.049 | 0.80 (0.56, 1.13) | 0.200 |
| Weekly | -0.011 (-0.019, -0.002) | 0.015 | 0.59 (0.37, 0.92) | 0.020 |
| Daily | -0.015 (-0.020, -0.010) | < 0.001 | 0.72 (0.60, 0.86) | < 0.001 |
| CRO | ||||
| Occasional | -0.003 (-0.012, 0.005) | 0.423 | 1.04 (0.76, 1.42) | 0.823 |
| Weekly | 0.000 (-0.012, 0.012) | 0.992 | 0.74 (0.48, 1.16) | 0.192 |
| Daily | 0.002 (-0.016, 0.020) | 0.811 | 0.55 (0.22, 1.36) | 0.193 |
| VOC | ||||
| Occasional | 0.002 (-0.001, 0.005) | 0.197 | 0.97 (0.86, 1.09) | 0.641 |
| Weekly | -0.001 (-0.006, 0.005) | 0.854 | 0.97 (0.77, 1.23) | 0.824 |
| Daily | 0.001 (-0.006, 0.008) | 0.823 | 0.78 (0.57, 1.09) | 0.143 |
| INT | ||||
| Occasional | -0.001 (-0.012, 0.011) | 0.910 | 0.70 (0.42, 1.17) | 0.173 |
| Weekly | 0.003 (-0.010, 0.016) | 0.699 | 0.57 (0.30, 1.12) | 0.102 |
| Daily | 0.011 (0.005, 0.017) | < 0.001 | 0.90 (0.75, 1.08) | 0.245 |
IWF Interacting with friends, MCCC Playing mah-jong, chess, cards or visiting community clubs, DFQ Going to community-organized dancing, fitness, qigong and so on; CRO, participating in community-related organizations, VOC Voluntary or charitable work, INT Using the Internet, HR Hazard ratio, CI Confidence Interval
a“None” group was set as the reference in each type analysis. Controlled covariates included age, gender, marital status, hukou status, education levels, rural/urban residence, public health insurance coverage, current work status, alcohol intake, smoking status and household per capita consumption
bThe intensity of each SP type was set as time-variant exposure. “None” group was set as the reference. Age, marital status, hukou status, public health insurance coverage, current work status, alcohol intake, smoking status and household per capita consumption were controlled as time-variant covariates, and gender, education level and rural/urban residence were controlled as fixed covariates