TABLE 3.
Robustness to unmeasured confounding (E‐values) for the associations between candidate predictors and subsequent sense of control (N = 13,771) a .
| Effect estimate b | Confidence interval limit c | |
|---|---|---|
| Health behaviors | ||
| Frequent physical activity | 1.11 | 1.00 |
| Smoking | 1.23 | 1.00 |
| Heavy drinking | 1.11 | 1.00 |
| Sleep problems | 1.27 | 1.09 |
| Physical health | ||
| Number of physical conditions | 1.23 | 1.00 |
| Diabetes | 1.23 | 1.00 |
| Hypertension | 1.00 | 1.00 |
| Stroke | 1.23 | 1.00 |
| Cancer | 1.20 | 1.00 |
| Heart disease | 1.00 | 1.00 |
| Lung disease | 1.16 | 1.00 |
| Arthritis | 1.33 | 1.00 |
| Overweight/obese | 1.33 | 1.00 |
| Physical functioning limitations | 1.39 | 1.16 |
| Cognitive impairment | 1.20 | 1.00 |
| Chronic pain | 1.27 | 1.00 |
| Self‐rated health | 1.36 | 1.24 |
| Hearing | 1.11 | 1.00 |
| Eyesight | 1.23 | 1.11 |
| Psychological well‐being | ||
| Positive affect | 1.56 | 1.47 |
| Life satisfaction | 1.42 | 1.33 |
| Optimism | 1.50 | 1.42 |
| Purpose in life | 1.53 | 1.40 |
| Health mastery | 1.42 | 1.34 |
| Financial mastery | 1.42 | 1.34 |
| Psychological distress | ||
| Depression | 1.42 | 1.12 |
| Depressive symptoms | 1.27 | 1.14 |
| Hopelessness | 1.53 | 1.45 |
| Negative affect | 1.45 | 1.39 |
| Anxiety | 1.42 | 1.33 |
| Trait anger | 1.27 | 1.18 |
| State anger | 1.20 | 1.10 |
| Cynical hostility | 1.30 | 1.22 |
| Stressful life events | 1.16 | 1.00 |
| Financial strain | 1.27 | 1.14 |
| Daily discrimination | 1.30 | 1.22 |
| Major discrimination | 1.11 | 1.00 |
| Social factors | ||
| Loneliness | 1.36 | 1.28 |
| Living with spouse | 1.36 | 1.02 |
| Contact children | ||
| <Every few months | Reference | Reference |
| 1–2×/month | 1.30 | 1.00 |
| 1–2×/week | 1.33 | 1.00 |
| >3×/week | 1.39 | 1.00 |
| Contact other family | ||
| <Every few months | Reference | Reference |
| 1–2×/month | 1.20 | 1.00 |
| 1–2×/week | 1.16 | 1.00 |
| >3×/week | 1.16 | 1.00 |
| Contact friends | ||
| <Every few months | Reference | Reference |
| 1–2×/month | 1.27 | 1.00 |
| 1–2×/week | 1.30 | 1.00 |
| >3×/week | 1.39 | 1.15 |
| Closeness with spouse | 1.11 | 1.00 |
| Number of close children | 1.11 | 1.00 |
| Number of close other family | 1.11 | 1.00 |
| Number of close friends | 1.16 | 1.00 |
| Positive social support from spouse | 1.23 | 1.00 |
| Positive social support from children | 1.30 | 1.19 |
| Positive social support from other family | 1.00 | 1.00 |
| Positive social support from friends | 1.11 | 1.00 |
| Social strain from spouse | 1.30 | 1.15 |
| Social strain from children | 1.33 | 1.23 |
| Social strain from other family | 1.16 | 1.00 |
| Social strain from friends | 1.10 | 1.00 |
| Volunteer | ||
| 0 h/year | Reference | Reference |
| 0–49 h/year | 1.27 | 1.00 |
| 50–99 h/year | 1.16 | 1.00 |
| 100–199 h/year | 1.42 | 1.07 |
| >200 h/year | 1.45 | 1.14 |
| Helping friends/neighbors/relatives | ||
| 0 h/year | Reference | Reference |
| 1–49 h/year | 1.23 | 1.00 |
| 50–99 h/year | 1.20 | 1.00 |
| 100–199 h/year | 1.27 | 1.00 |
| >200 h | 1.47 | 1.29 |
| Religious service attendance | ||
| Not at All | Reference | Reference |
| <1×/week | 1.20 | 1.00 |
| >1×/week | 1.11 | 1.00 |
| Social status ladder | 1.33 | 1.21 |
| Change in social status ladder | ||
| Moved down | Reference | Reference |
| No change | 1.36 | 1.09 |
| Moved up | 1.47 | 1.23 |
| In labor force | 1.33 | 1.05 |
See VanderWeele and Ding (2017) for the formula for calculating E‐values.
The E‐values for effect estimates are the minimum strength of association on the risk ratio scale that an unmeasured confounder would need to have with both the exposure and the outcome to fully explain away the observed association between the exposure and outcome, conditional on the measured covariates.
The E‐values for the limit of the 95% confidence interval (CI) closest to the null denote the minimum strength of association on the risk ratio scale that an unmeasured confounder would need to have with both the exposure and the outcome to shift the confidence interval to include the null value, conditional on the measured covariates.