Table 4. Gender-stratified Poisson regression analysis of the associations of internet addiction with poor sleep quality, self-rated health, and psychological distress.
| Poor sleep quality | ||||
| Exposure/s | Female (n = 335) | Male (n = 290) | ||
| No. of cases | RR (95% CI) & | No. of cases | RR (95% CI) & | |
| Problematic internet use | ||||
| No | 93 | 1.00 | 57 | 1.00 |
| Yes | 34 | 1.17 (0.86, 1.58) | 36 | 1.63 (1.17, 2.26) |
| P for interaction! | .122 | |||
| Level of internet addiction | ||||
| Lowest quintile (ref) | 23 | 1.00 | 11 | 1.00 |
| 2nd quintile | 24 | 1.08 (0.69, 1.71) | 18 | 1.56 (0.80, 3.02) |
| 3rd quintile | 26 | 1.10 (0.70, 1.72) | 18 | 1.53 (0.79, 2.97) |
| 4th quintile | 25 | 1.08 (0.69, 1.70) | 17 | 1.86 (0.96, 3.59) |
| Highest quintile | 29 | 1.40 (0.91, 2.15) | 29 | 2.83 (1.51, 4.95) |
| P for heterogeneity* | .542 | .004 | ||
| P for trend# | .176 | .001 | ||
| P for interaction! | .478 | |||
| Internet addiction z-scores (per std. deviation) | 1.14 (1.01, 1.30) | 1.32 (1.14, 1.54) | ||
| P for interaction! | 0.115 | |||
| Psychological distress | ||||
| Problematic internet use | ||||
| No | 132 | 1.00 | 62 | 1.00 |
| Yes | 56 | 1.32 (1.09, 1.59) | 47 | 1.93 (1.46, 2.56) |
| P for interaction! | .025 | |||
| Level of Internet addiction | ||||
| Lowest quintile (ref) | 25 | 1.00 | 9 | 1.00 |
| 2nd quintile | 34 | 1.39 (0.91, 3.62) | 19 | 197 (0.97, 4.02) |
| 3rd quintile | 39 | 1.55 (1.15, 4.55) | 19 | 1.94 (0.96, 3.94) |
| 4th quintile | 47 | 1.85 (1.81, 7.44) | 28 | 3.72 (1.94, 7.12) |
| Highest quintile | 43 | 1.83 (1.74, 7.44) | 34 | 3.81 (2.01, 7.21) |
| P for heterogeneity* | .004 | .001 | ||
| P for trend# | .001 | .001 | ||
| P for interaction! | .153 | |||
| Internet addiction z-scores (per std. deviation) | 1.20 (1.11, 1.30) | 1.41 (1.24, 1.61) | ||
| P for interaction# | .040 | |||
| Poor/fair self-rated health | ||||
| Problematic internet use | ||||
| No | 143 | 1.00 | 86 | 1.00 |
| Yes | 46 | 1.01 (0.81, 1.25) | 37 | 1.09 (0.82, 1.45) |
| P for interaction! | .607 | |||
| Level of Internet addiction | ||||
| Lowest quintile (ref) | 27 | 1.00 | 16 | 1.00 |
| 2nd quintile | 37 | 1.41 (0.98, 2.03) | 28 | 1.62 (0.98, 2.68) |
| 3rd quintile | 43 | 1.58 (1.12, 2.23) | 30 | 1.71 (1.05, 2.80) |
| 4th quintile | 49 | 1.79 (1.28, 2.48) | 23 | 1.71 (1.02, 2.87) |
| Highest quintile | 33 | 1.31 (0.90, 1.90) | 26 | 1.62 (0.98, 2.68) |
| P for heterogeneity* | .006 | .263 | ||
| P for trend# | .035 | .076 | ||
| P for interaction! | .684 | |||
| Internet addiction z-scores (per std. deviation) | 1.07 (0.98, 1.18) | 1.09 (0.96, 1.24) | ||
| P for interaction! | .720 | |||
Note: RR: Risk Ratio, CI: Confidence Interval.
&The estimates were minimally adjusted for age.
!P-value for interaction was obtained from the pooled sample by fitting an exposure*gender interaction term in the age-adjusted Poisson model.
*P-value for heterogeneity to test the overall association between quintiles of IA and the outcome.
#P-value for trend to test whether the quintiles of IA are linearly associated with the outcome.