Table 2. The Association of 24-Year Blood Pressure Patterns with Dementia, Mild Cognitive Impairment, and Cognitive Changea.
| Incident Dementia (Full Sample) | Incident Dementia (Attended Visit 6) | Visit 6 Mild Cognitive Impairment | Cognitive Change | |||||
|---|---|---|---|---|---|---|---|---|
| Hazard Ratio (95% CI) | No./Total No. (%) | Hazard Ratio (95% CI) | No./Total No. (%) | Odds Ratio (95% CI) | No./Total No. (%) | β (95% CI)b | No. | |
| 24-y Blood Pressure Patterns Defined Using the Standard Hypertension Criteriac | ||||||||
| No. | 4738 | 3264 | 2584 | 4403 | ||||
| Midlife and late-life normotension | 1 [Reference] | 53/833 (6) | 1 [Reference] | 25/616 (4) | 1 [Reference] | 68/509 (13) | 0 [Reference] | 811 |
| Midlife normotension, late-life hypertension | 1.16 (0.84 to 1.61) | 149/1559 (10) | 0.95 (0.58 to 1.56) | 51/1096 (5) | 1.32 (0.96 to 1.82) | 153/874 (18) | −0.01 (−0.06 to 0.03) | 1462 |
| Midlife and late-life hypertension | 1.49 (1.06 to 2.08) | 138/1030 (13) | 1.10 (0.64 to 1.89) | 47/690 (7) | 1.18 (0.80 to 1.73) | 84/530 (16) | −0.01 (−0.06 to 0.03) | 905 |
| Midlife normotension, late-life hypotension | 1.11 (0.78 to 1.57) | 91/927 (10) | 0.93 (0.54 to 1.61) | 32/647 (5) | 1.19 (0.83 to 1.70) | 93/519 (18) | −0.01 (−0.05 to 0.04) | 884 |
| Midlife hypertension, late-life hypotension | 1.62 (1.11 to 2.37) | 77/389 (20) | 1.45 (0.76 to 2.75) | 21/215 (10) | 1.65 (1.01 to 2.69) | 37/152 (24) | −0.05 (−0.13 to 0.02) | 341 |
| 24-y Blood Pressure Patterns Defined Using the Measured Hypertension Criteriad | ||||||||
| No. | 4752 | 3273 | 2590 | 4409 | ||||
| Midlife and late-life normotension | 1 [Reference] | 181/2116 (9) | 1 [Reference] | 68/1525 (4) | 1 [Reference] | 182/1234 (15) | 0 [Reference] | 1998 |
| Midlife normotension, late-life hypertension | 1.13 (0.89 to 1.44) | 113/938 (12) | 1.19 (0.80 to 1.78) | 39/635 (6) | 1.34 (1.01 to 1.77) | 96/494 (19) | −0.01 (−0.06 to 0.03) | 856 |
| Midlife and late-life hypertension | 1.39 (0.95 2.04) | 33/202 (16) | 1.37 (0.69 to 2.70) | 10/123 (8) | 0.94 (0.50 to 1.75) | 14/90 (16) | −0.01 (−0.10 to 0.08) | 176 |
| Midlife normotension, late-life hypotension | 1.06 (0.84 to 1.33) | 135/1186 (11) | 1.08 (0.72 to 1.61) | 44/789 (6) | 1.14 (0.87 to 1.50) | 117/621 (19) | 0.01 (−0.03 to 0.05) | 1111 |
| Midlife hypertension, late-life hypotension | 1.37 (0.93 to 2.02) | 33/130 (25) | 1.73 (0.83 to 3.60) | 9/73 (12) | 1.48 (0.75 to 2.93) | 13/50 (26) | −0.05 (−0.16 to 0.06) | 111 |
| Midlife hypertension, late-life normotension | 1.12 (0.69 to 1.80) | 19/180 (11) | 1.02 (0.46 to 2.25) | 7/128 (5) | 1.14 (0.64 to 2.02) | 16/101 (16) | −0.05 (−0.14 to 0.04) | 157 |
Abbreviations: APOE, apolipoprotein E; GEE, generalized estimating equation.
Cox proportional hazard, logistic regression, and GEE regression (with a Gaussian distribution to model the continuous outcome of cognitive performance) models were adjusted for baseline age, sex, race+center (a combined variable was used because center [study site], in many cases, determined race; this method was used instead of coding participants for race and center separately for analytic reasons), education, and APOEε4 status, and visit 5 body mass index, total cholesterol, high-density lipoprotein cholesterol, cigarette smoking status, alcohol use status, prevalent diabetes, coronary heart disease, heart failure, and previous stroke, as defined at visit 5.
Unstandardized estimates of additional 5-year cognitive change (z score) associated with blood pressure pattern. These estimates were derived from GEE models adjusted for baseline age, an age × time interaction term, sex, race+center (see footnote a for definition), education, APOEε4 status, and visit 5 body mass index, total cholesterol, high-density lipoprotein cholesterol, cigarette smoking status, alcohol use status, prevalent diabetes, coronary heart disease, heart failure, and previous stroke, as defined at visit 5.
The standard hypertension definition classified a participant as hypertensive if the participant met hypertension criteria based on measured blood pressure or if the participant was taking blood pressure medication at that visit, even if the blood pressure was in the normal range. Participants with midlife hypertension and late-life normotension were not included in the standard hypertension definition analyses because of small sample size (n = 23).
The measured hypertension definition classified a participant as hypertensive if the measured blood pressure value was greater than the threshold for hypertension at that visit. Participants with missing systolic or diastolic blood pressure at visit 5 (n = 9) were not included in the measured hypertension definition analyses.