Multiphase Growth Curve Models Predicting Changes in Older Adults’ Depressive Symptoms in Response to Spouses’ Dementia Onset
| . | Model 1 . | Model 2 . |
|---|---|---|
| Variable . | B (SE) . | B (SE) . |
| Fixed effects | ||
| Intercept (β 0i) | 1.20*** (0.12) | 1.39 (0.79) |
| Overall positive marital qualitya | — | −0.06 (0.17) |
| Overall negative marital qualitya | — | 0.49** (0.15) |
| Discrepant positive marital qualityb | — | −0.10 (0.11) |
| Discrepant negative marital qualityb | — | −0.07 (0.09) |
| Time-to/from-dementia (β 1i) | 0.00 (0.03) | 0.00 (0.02) |
| Reaction (β 2i) | 0.31* (0.12) | 1.48 (1.00) |
| × Overall positive marital qualitya | — | −0.24 (0.19) |
| × Overall negative marital qualitya | — | −0.35* (0.17) |
| × Discrepant positive marital qualityb | — | 0.08 (0.12) |
| × Discrepant negative marital qualityb | — | 0.12 (0.11) |
| Time-to/from-dementia × Reaction (β 3i) | 0.05 (0.03) | 0.02 (0.03) |
| Random effects | ||
| Intercept variance | 1.48*** (0.26) | 0.65** (0.23) |
| Time-to/from-dementia variance | 0.00 (0.01) | 0.01 (0.01) |
| Reaction variance | — | 0.39 (0.42) |
| Residual variance | 2.02*** (0.09) | 1.37*** (0.08) |
| −2 log likelihood | 7,090.6 | 4,897.2 |
| . | Model 1 . | Model 2 . |
|---|---|---|
| Variable . | B (SE) . | B (SE) . |
| Fixed effects | ||
| Intercept (β 0i) | 1.20*** (0.12) | 1.39 (0.79) |
| Overall positive marital qualitya | — | −0.06 (0.17) |
| Overall negative marital qualitya | — | 0.49** (0.15) |
| Discrepant positive marital qualityb | — | −0.10 (0.11) |
| Discrepant negative marital qualityb | — | −0.07 (0.09) |
| Time-to/from-dementia (β 1i) | 0.00 (0.03) | 0.00 (0.02) |
| Reaction (β 2i) | 0.31* (0.12) | 1.48 (1.00) |
| × Overall positive marital qualitya | — | −0.24 (0.19) |
| × Overall negative marital qualitya | — | −0.35* (0.17) |
| × Discrepant positive marital qualityb | — | 0.08 (0.12) |
| × Discrepant negative marital qualityb | — | 0.12 (0.11) |
| Time-to/from-dementia × Reaction (β 3i) | 0.05 (0.03) | 0.02 (0.03) |
| Random effects | ||
| Intercept variance | 1.48*** (0.26) | 0.65** (0.23) |
| Time-to/from-dementia variance | 0.00 (0.01) | 0.01 (0.01) |
| Reaction variance | — | 0.39 (0.42) |
| Residual variance | 2.02*** (0.09) | 1.37*** (0.08) |
| −2 log likelihood | 7,090.6 | 4,897.2 |
Notes: SE = standard error. Participant n = 410. Intercept and reaction were fully adjusted for covariates in Model 2 (not shown in table). The level-1 equation for individual i’s depressive symptoms at time t: Depressive symptomsti = β 0i + β 1i (time-to/from-spouse-dementiati) + β 2i (reactionti) + β 3i (time-to/from-spouse-dementiati × reactionti) + eti; the intercept β 0i refers to individual i’s depressive symptoms in the year of spousal dementia onset; β 1i is an individual-specific slope indicating rates of linear change in individual i’s depressive symptoms prior to spousal dementia; β 2i captures the effect of spousal dementia on individual i’s depressive symptoms; β 3i examines whether rates of change in individual i’s depressive symptoms differed prior to and post spousal dementia; eti indicates residual error.
aAveraged across partners.
bSelf-rating subtracted by partner rating.
*p < .05. **p < .01. ***p < .001.
Multiphase Growth Curve Models Predicting Changes in Older Adults’ Depressive Symptoms in Response to Spouses’ Dementia Onset
| . | Model 1 . | Model 2 . |
|---|---|---|
| Variable . | B (SE) . | B (SE) . |
| Fixed effects | ||
| Intercept (β 0i) | 1.20*** (0.12) | 1.39 (0.79) |
| Overall positive marital qualitya | — | −0.06 (0.17) |
| Overall negative marital qualitya | — | 0.49** (0.15) |
| Discrepant positive marital qualityb | — | −0.10 (0.11) |
| Discrepant negative marital qualityb | — | −0.07 (0.09) |
| Time-to/from-dementia (β 1i) | 0.00 (0.03) | 0.00 (0.02) |
| Reaction (β 2i) | 0.31* (0.12) | 1.48 (1.00) |
| × Overall positive marital qualitya | — | −0.24 (0.19) |
| × Overall negative marital qualitya | — | −0.35* (0.17) |
| × Discrepant positive marital qualityb | — | 0.08 (0.12) |
| × Discrepant negative marital qualityb | — | 0.12 (0.11) |
| Time-to/from-dementia × Reaction (β 3i) | 0.05 (0.03) | 0.02 (0.03) |
| Random effects | ||
| Intercept variance | 1.48*** (0.26) | 0.65** (0.23) |
| Time-to/from-dementia variance | 0.00 (0.01) | 0.01 (0.01) |
| Reaction variance | — | 0.39 (0.42) |
| Residual variance | 2.02*** (0.09) | 1.37*** (0.08) |
| −2 log likelihood | 7,090.6 | 4,897.2 |
| . | Model 1 . | Model 2 . |
|---|---|---|
| Variable . | B (SE) . | B (SE) . |
| Fixed effects | ||
| Intercept (β 0i) | 1.20*** (0.12) | 1.39 (0.79) |
| Overall positive marital qualitya | — | −0.06 (0.17) |
| Overall negative marital qualitya | — | 0.49** (0.15) |
| Discrepant positive marital qualityb | — | −0.10 (0.11) |
| Discrepant negative marital qualityb | — | −0.07 (0.09) |
| Time-to/from-dementia (β 1i) | 0.00 (0.03) | 0.00 (0.02) |
| Reaction (β 2i) | 0.31* (0.12) | 1.48 (1.00) |
| × Overall positive marital qualitya | — | −0.24 (0.19) |
| × Overall negative marital qualitya | — | −0.35* (0.17) |
| × Discrepant positive marital qualityb | — | 0.08 (0.12) |
| × Discrepant negative marital qualityb | — | 0.12 (0.11) |
| Time-to/from-dementia × Reaction (β 3i) | 0.05 (0.03) | 0.02 (0.03) |
| Random effects | ||
| Intercept variance | 1.48*** (0.26) | 0.65** (0.23) |
| Time-to/from-dementia variance | 0.00 (0.01) | 0.01 (0.01) |
| Reaction variance | — | 0.39 (0.42) |
| Residual variance | 2.02*** (0.09) | 1.37*** (0.08) |
| −2 log likelihood | 7,090.6 | 4,897.2 |
Notes: SE = standard error. Participant n = 410. Intercept and reaction were fully adjusted for covariates in Model 2 (not shown in table). The level-1 equation for individual i’s depressive symptoms at time t: Depressive symptomsti = β 0i + β 1i (time-to/from-spouse-dementiati) + β 2i (reactionti) + β 3i (time-to/from-spouse-dementiati × reactionti) + eti; the intercept β 0i refers to individual i’s depressive symptoms in the year of spousal dementia onset; β 1i is an individual-specific slope indicating rates of linear change in individual i’s depressive symptoms prior to spousal dementia; β 2i captures the effect of spousal dementia on individual i’s depressive symptoms; β 3i examines whether rates of change in individual i’s depressive symptoms differed prior to and post spousal dementia; eti indicates residual error.
aAveraged across partners.
bSelf-rating subtracted by partner rating.
*p < .05. **p < .01. ***p < .001.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
