Table 3.
(a) Associations between child maltreatment (CM) and each mediator and (b) counterfactual disparity measures between CM and midlife HRQoL after accounting for each mediator in separate analyses. (N=301).
| Mediation models (analyzed separately) | (a) Associations between CM and each mediator |
(b) Counterfactual disparity measures between CM and HRQoL after separately accounting for each mediatorc |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mediators N (column %) | OR (95% CI)a of each mediator (Yes) predicted by any CM | Mental Component Score (MCS) | Physical Component Score (PCS) | |||||||
| No | Yes | ψx (95% CI) b | Proportion explained by the mediatord | ψx (95% CI) b | Proportion explained by the mediatord | |||||
| Total association of any CM and HRQoL | −2.43 (−4.52, −0.35)* | −2.36 (−4.37, −0.35)* | ||||||||
| Model 1 | Lifetime psychiatric history | |||||||||
| CM | No | 84 (68) | 103 (58) | reference | reference | |||||
| Yes | 40 (32) | 74 (42) | 1.52 (0.91, 2.54) | −2.14 (−4.17, −0.10)* | 12% | −2.46 (−4.47, −0.44)* | −4% | |||
| Model 2 | Depressive symptoms | |||||||||
| CM | No | 168 (64) | 19 (48) | reference | reference | |||||
| Yes | 93 (36) | 21 (53) | 1.93 (0.92, 4.02) | −1.59 (−3.40, 0.21) | 35% | −2.48 (−4.50, −0.45)* | −5% | |||
| Model 3 | Sleep problems | |||||||||
| CM | No | 115 (66) | 72 (57) | reference | reference | |||||
| Yes | 58 (34) | 55 (43) | 1.70 (1.02, 2.84)* | −2.35 (−4.38, −0.33)* | 3% | −2.10 (−4.08, −0.12)* | 11% | |||
| Model 4 | Very upsetting life events | |||||||||
| CM | No | 111 (68) | 76 (55) | reference | reference | |||||
| Yes | 52 (32) | 62 (45) | 1.75 (1.05, 2.92)* | −1.92 (−3.92, 0.09) | 21% | −2.29 (−4.28, −0.29)* | 3% | |||
| Model 5 | Low social support | |||||||||
| CM | No | 167 (67) | 20 (39) | reference | reference | |||||
| Yes | 83 (33) | 31 (61) | 2.76 (1.40, 5.44)* | −1.74 (−3.73, 0.26) | 28% | −2.34 (−4.34, −0.33)* | 1% | |||
ORs for each potential mediator predicted by at least one CM were estimated by logistic regression adjusted for age, race, and childhood socioeconomic circumstances. If a OR 95% confidence interval does not include 1, then there is a statistically significant difference between the groups. (*p<0.05)
If a 95% confidence interval does not include 0, then there is a statistically significant difference between the groups. (*p<0.05)
Counterfactual disparity measures between CM and HRQoL accounting for each potential mediator as 0, respectively, were examined by doubly-robust structural nested mean model after adjusting for age, race, childhood socioeconomic circumstances, education attainment, marital status, adulthood financial strain, smoking, alcohol consumption, total number of medical conditions, menopausal status, hormone therapy use, vasomotor symptoms, physical activity, body mass index, trait anxiety, trait anger, and ever treated for emotional problems.
Proportion explained for MCS = (−2.43 - ψx)/ −2.43 × 100%; proportion explained for PCS = (−2.36 - ψx)/ −2.36 × 100%.