Table 3:
Odds ratios of reporting being told of dementia by a doctor among model-predicted dementia cases1
| Model 1 | Model 2 | |
|---|---|---|
| Race and Ethnicity | ||
| Non-Hispanic White | Reference | Reference |
| Non-Hispanic Black | 0.66 (0.53, 0.83) | 0.74 (0.58, 0.93) |
| Hispanic | 0.61 (0.46, 0.80) | 0.60 (0.43, 0.85) |
| Age | ||
| 70-74 | 1.48 (1.02, 2.14) | 1.46 (0.89, 2.38) |
| 75-79 | 1.53 (1.22, 1.92) | 1.55 (1.17, 2.06) |
| 80-84 | 1.15 (0.98, 1.36) | 1.33 (1.10, 1.60) |
| 85+ | Reference | Reference |
| Female vs. Male | 1.29 (1.06, 1.57) | 1.12 (0.91, 1.40) |
| HRS Survey Year | 1.09 (1.07, 1.11) | 1.05 (1.03, 1.07) |
| Cognitive Impairment2 | -- | 1.31 (0.98, 1.76) |
| Number of ADL Limitations3 | -- | 1.00 (0.96, 1.03) |
| Number of IADL Limitations4 | -- | 1.65 (1.56, 1.75) |
| Number of Comorbidities | -- | 1.03 (0.97, 1.09) |
| Living in a Nursing Home | -- | 2.77 (2.32, 3.32) |
Sample used for the analyses corresponds with Box 5 of the consort diagram in Supplemental Figure S1 (n=2,367 respondents). The analyses used a weighted logit-link binomial distribution generalized estimating equations assuming an unstructured correlation structure. We used average sample weighs of each HRS respondent, following NHANES guidelines on combing survey cycles. The results had little to no change when using participants’ combined, first, last, or first year predicted to have dementia HRS wave-specific sample weights.
Cognitive function combined normalized TICS scores and IQCODES scores based on whether participants had a proxy respondent. 0: No impairment; 1: High impairment
Activities of Daily Living. Numbers are the reported number of activities (total of six) participants has difficulty performing; Lower scores indicate higher functional ability
Instrumental Activities of Daily Living. Numbers are the reported number of activities (total of five) participants has difficulty performing; Lower scores indicate higher functional ability