Table 1.
Characteristics of Study Participants at Baseline (MIDUS 2 n=3,535; MIDUS Refresher n=2,591; HRS Coping n=511)
| Dataset | |||
|---|---|---|---|
| Characteristic | MIDUS 2 (n=3,535)a |
MIDUS Refresher Sample (n=2,591)b |
HRS Coping Module (n=511)c |
| Demographic Factors | |||
| Mean Age (SD) | 57.1 (13.4) | 52.8 (14.6) | 71.1 (8.9) |
| Female (%) | 55.5 | 54.2 | 60.5 |
| Race/Ethnicity (%) | |||
| White | 90.5 | 81.8 | 75.5 |
| African-American | 4.5 | 7.5 | 13.5 |
| Other | 5.0 | 10.7 | 11.0 |
| Married (%) | 69.4 | 64.8 | 59.9 |
| Education (%) | |||
| <High School | 8.2 | 5.7 | 20.9 |
| High School | 29.0 | 20.8 | 53.6 |
| ≥College | 62.7 | 73.4 | 25.4 |
| Income (MIDUS) or Total Wealth (HRS) (%) | |||
| 1st Quintile | 22.0 | 20.5 | 20.4 |
| 2nd Quintile | 18.4 | 20.0 | 19.8 |
| 3rd Quintile | 19.8 | 20.3 | 20.2 |
| 4th Quintile | 20.8 | 20.2 | 19.8 |
| 5th Quintile | 19.0 | 19.1 | 20.0 |
| Health Insurance (%) | 91.9 | 91.6 | 91.2 |
| Employment Status (%) | 60.3 | 60.1 | 31.1 |
| Psychological Distress | |||
| Depression (%) | 10.7 | 11.6 | 13.3 |
| Anxiety (%) | 1.9 | 2.9 | 14.1 |
| Mean Lifetime Stressors (SD) | 3.4 (2.6) | 3.0 (2.5) | 2.0 (1.7) |
| Baseline Health | |||
| Chronic Condition(s) (%) | 77.6 | 76.8 | 53.2 |
| Mean # of Nights Hospitalized (SD) | 1.0 (5.4) | 0.7 (4.7) | 2.9 (14.1) |
| Chronic Pain (%) | 38.2 | 33.4 | 33.7 |
| Health Behaviors | |||
| Current drinker (%) | 57.3 | 60.9 | 49.7 |
| Current smoker (%) | 15.7 | 11.1 | 11.6 |
MIDUS 2 participants who did not misuse drugs at baseline (longitudinal analyses)
MIDUS Refresher participants (cross-sectional analyses)
HRS coping module participants (cross-sectional analyses)
Generalized anxiety disorder was assessed in MIDUS using the CIDI-SF and defined as a score of ≥3 (47,49). Elevated anxiety was assessed in HRS using the Beck Anxiety Inventory and defined as a score of ≥12 (50,51). There is no established cutoff score for anxiety disorder in HRS, thus elevated anxiety was used—which does have a previously tested cutoff score in this cohort (50). The difference in assessed level of anxiety (e.g., anxiety disorder vs. elevated anxiety) and cutpoints helps explain the higher rate of anxiety in HRS participants when compared to MIDUS participants. As a sensitivity analysis, we also created a new HRS anxiety disorder cutoff score that matched MIDUS anxiety disorder prevalence rates (2%−2.5%). When using this new cutoff score, in lieu of the originally constructed elevated anxiety score, results were nearly identical. Thus, we continued using the validated anxiety cutoff score in our main analyses.