Table 3.
Association of informal caregiving with increased loneliness, self-reported deterioration in mental health, and incidence of new suicidal ideation, by intensity of caregiving.
| No. of participants | Incidence, n (%) | Adjusted rate,% (95% CI) | Adjusted OR (95% CI) | P-value | P-value for trend* | |
|---|---|---|---|---|---|---|
| Increased loneliness | <0.001 | |||||
| Caregiving, High intensity | 1326 | 841 (63.4) | 46.1 (34.4, 57.8) | 2.56 (1.52, 4.31) | <0.001 | |
| Caregiving, Low intensity | 1174 | 460 (39.2) | 39.6 (34.8, 44.3) | 1.89 (1.51, 2.37) | <0.001 | |
| Caregiving, No | 22,982 | 6007 (26.1) | 27.1 (25.7, 28.5) | Reference | ||
| Self-reported deterioration in mental health | 0.022 | |||||
| Caregiving, High intensity | 1326 | 358 (27.0) | 25.6 (15.4, 35.7) | 1.66 (0.91,3.02) | 0.100 | |
| Caregiving, Low intensity | 1174 | 283 (24.1) | 23.4 (19.8, 27.0) | 1.45 (1.15, 1.83) | 0.002 | |
| Caregiving, No | 22,982 | 4079 (17.8) | 17.8 (16.3, 19.4) | Reference | ||
| Incidence of new suicidal ideation | <0.001 | |||||
| Caregiving, High intensity | 990 | 235 (23.7) | 15.0 (6.9, 23.1) | 6.39 (2.73, 14.94) | <0.001 | |
| Caregiving, Low intensity | 1043 | 40 (3.8) | 5.2 (2.7, 7.7) | 1.56 (0.83, 2.93) | 0.17 | |
| Caregiving, No | 21,389 | 760 (3.6) | 3.6 (2.9, 4.4) | Reference |
OR, odds ratio; CI, confidence interval.
We used multivariable logistic regression models using inverse probability weighting. The analyses were weighted to account for selection in an internet survey. Standard errors were clustered at the prefectural level to account for the potential correlation of participants in the same prefecture. We adjusted for caregiver gender, age, marital status, educational attainment, self-rated health, income, employment status, and indicators of comorbidities. Adjusted rates were calculated using marginal standardization. According to the hours of caregiving, individuals providing informal care were categorized to “high intensity (2–3 h per day or more)” and “low intensity (Support when needed).” In the analysis for the incidence of new suicidal ideation, we excluded 2060 participants who had suicidal ideation before the COVID-19 pandemic. *P values for trend tests assessed the trends in the incidence of outcomes.