Joint Models for Estimating Determinants of Cognitive Decline in the Presence of Survival Bias.

BACKGROUND: Identifying determinants of cognitive decline is crucial for developing strategies to prevent Alzheimer's disease and related dementias. However, determinants of cognitive decline remain elusive, with inconsistent results across studies. One reason could be differential survival. Cognitive decline and many exposures of interest are associated with mortality making survival a collider. Not accounting for informative attrition can result in survival bias. Generalized estimating equations (GEE) and linear mixed-effects model (LME) are commonly used to estimate effects of exposures on cognitive decline, but both assume mortality is not informative. Joint models combine LME with Cox proportional hazards models to simultaneously estimate cognitive decline and the hazard of mortality. METHODS: Using simulations, we compared estimates of the effect of a binary exposure on rate of cognitive decline from GEE, weighted GEE using inverse-probability-of-attrition weights, and LME to joint models under several causal structures of survival bias. RESULTS: We found that joint models with correctly specified relationship between survival and cognition performed best, producing unbiased estimates and appropriate coverage. Even those with misspecified relationship between survival and cognition showed advantage under causal structures consistent with survival bias. We also compared these models in estimating the effect of education on cognitive decline after dementia diagnosis using Framingham Heart Study data. Estimates of the effect of education on cognitive decline from joint models were slightly attenuated with similar precision compared with LME. CONCLUSIONS: In our study, joint models were more robust than LME, GEE, and weighted GEE models when evaluating determinants of cognitive decline.

Twenty-seven-year time trends in dementia incidence in Europe and the United States: The Alzheimer Cohorts Consortium.

OBJECTIVE: To determine changes in the incidence of dementia between 1988 and 2015. METHODS: This analysis was performed in aggregated data from individuals >65 years of age in 7 population-based cohort studies in the United States and Europe from the Alzheimer Cohort Consortium. First, we calculated age- and sex-specific incidence rates for all-cause dementia, and then defined nonoverlapping 5-year epochs within each study to determine trends in incidence. Estimates of change per 10-year interval were pooled and results are presented combined and stratified by sex. RESULTS: Of 49,202 individuals, 4,253 (8.6%) developed dementia. The incidence rate of dementia increased with age, similarly for women and men, ranging from about 4 per 1,000 person-years in individuals aged 65-69 years to 65 per 1,000 person-years for those aged 85-89 years. The incidence rate of dementia declined by 13% per calendar decade (95% confidence interval [CI], 7%-19%), consistently across studies, and somewhat more pronouncedly in men than in women (24% [95% CI 14%-32%] vs 8% [0%-15%]). CONCLUSION: The incidence rate of dementia in Europe and North America has declined by 13% per decade over the past 25 years, consistently across studies. Incidence is similar for men and women, although declines were somewhat more profound in men. These observations call for sustained efforts to finding the causes for this decline, as well as determining their validity in geographically and ethnically diverse populations.

Antihypertensive medications and risk for incident dementia and Alzheimer's disease: a meta-analysis of individual participant data from prospective cohort studies.

BACKGROUND: Dementia is a major health concern for which prevention and treatment strategies remain elusive. Lowering high blood pressure with specific antihypertensive medications (AHMs) could reduce the burden of disease. We investigated whether specific AHM classes reduced the risk for dementia. METHODS: We did a meta-analysis of individual participant data from eligible observational studies published between Jan 1, 1980, and Jan 1, 2019. Cohorts were eligible for inclusion if they prospectively recruited community-dwelling adults; included more than 2000 participants; collected data for dementia events over at least 5 years; had measured blood pressure and verified use of AHMs; included in-person exams, supplemented with additional data, to capture dementia events; and had followed up cases for mortality. We assessed the association of incident dementia and clinical Alzheimer's disease with use of five AHM classes, within strata of baseline high (systolic blood pressure [SBP] ≥140 mm Hg or diastolic blood pressure [DBP] ≥90 mm Hg) and normal (SBP <140 mm Hg and DBP <90 mm Hg) blood pressure. We used a propensity score to control for confounding factors related to the probability of receiving AHM. Study-specific effect estimates were pooled using random-effects meta-analyses. RESULTS: Six prospective community-based studies (n=31 090 well phenotyped dementia-free adults older than 55 years) with median follow-ups across cohorts of 7-22 years were eligible for analysis. There were 3728 incident cases of dementia and 1741 incident Alzheimer's disease diagnoses. In the high blood pressure stratum (n=15 537), those using any AHM had a reduced risk for developing dementia (hazard ratio [HR] 0·88, 95% CI 0·79-0·98; p=0·019) and Alzheimer's disease (HR 0·84, 0·73-0·97; p=0·021) compared with those not using AHM. We did not find any significant differences between one drug class versus all others on risk of dementia. In the normal blood pressure stratum (n=15 553), there was no association between AHM use and incident dementia or Alzheimer's disease. INTERPRETATION: Over a long period of observation, no evidence was found that a specific AHM drug class was more effective than others in lowering risk of dementia. Among people with hypertensive levels of blood pressure, use of any AHM with efficacy to lower blood pressure might reduce the risk for dementia. These findings suggest future clinical guidelines for hypertension management should also consider the beneficial effect of AHM on the risk for dementia. FUNDING: The Alzheimer's Drug Discovery Foundation and the National Institute on Aging Intramural Research Program.

Association of Accelerometer-Measured Light-Intensity Physical Activity With Brain Volume: The Framingham Heart Study.

Importance: Dementia risk may be attenuated by physical activity (PA); however, the specific activity levels optimal for dementia prevention are unclear. Moreover, most older adults are unable to meet the nationally recommended PA guidelines, set at 150 minutes of moderate to vigorous PA per week. Objective: To assess the association of total steps walked per day and total dose (intensity × duration) of PA with brain volumes on magnetic resonance imaging (MRI) among Framingham Heart Study participants. Design, Setting, and Participants: This cross-sectional, community-based cohort study of the association of accelerometry-determined PA with brain MRI measures in Framingham, Massachusetts, included the Framingham Heart Study third-generation (examination 2, 2008-2011) and offspring (examination 9, 2011-2014) cohorts. Of 4021 participants who agreed to wear an accelerometer and had valid data (≥10 hours/day for ≥3 days), 1667 participants who did not undergo brain MRI (n = 1604) or had prevalent dementia or stroke (n = 63) were excluded. Data analysis began in 2016 and was completed in February 2019. Exposures: Physical activity achieved using accelerometry-derived total activity (steps per day) and 2 intensity levels (light intensity and moderate to vigorous intensity). Main Outcomes and Measures: Differences in total brain volume and other MRI markers of brain aging. Results: The study sample of 2354 participants had a mean (SD) age of 53 (13) years, 1276 (54.2%) were women, and 1099 (46.7%) met the PA guidelines. Incremental light-intensity PA was associated with higher total brain volume; each additional hour of light-intensity PA was associated with approximately 1.1 years less brain aging (ß estimate, 0.22; SD, 0.07; P = .003). Among individuals not meeting the PA guidelines, each hour of light-intensity PA (ß estimate, 0.28; SD, 0.11; P = .01) and achieving 7500 steps or more per day (ß estimate, 0.44; SD, 0.18; P = .02) were associated with higher total brain volume, equivalent to approximately 1.4 to 2.2 years less brain aging. After adjusting for light-intensity PA, neither increasing moderate to vigorous PA levels nor meeting the threshold moderate to vigorous PA level recommended by the PA guidelines were significantly associated with total brain volume. Conclusions and Relevance: Every additional hour of light-intensity PA was associated with higher brain volumes, even among individuals not meeting current PA guidelines. These data are consistent with the notion that the potential benefits of PA on brain aging may accrue at a lower, more achievable level of intensity or duration.

Self-Reported Physical Activity and Relations to Growth and Neurotrophic Factors in Diabetes Mellitus: The Framingham Offspring Study.

AIMS: Circulating insulin-like growth factor- (IGF-) 1, vascular endothelial growth factor (VEGF), and brain-derived neurotrophic factor (BDNF) levels are often lower in individuals with diabetes mellitus (DM) and are important for repairing vascular and neuronal dysfunction. The purpose of this investigation was to determine the cross-sectional relations of physical activity to circulating concentrations of IGF-1, VEGF, and BDNF in individuals with and without DM. METHODS: In 1730 participants from the Framingham Offspring Study examination cycle 7, including those with DM (n = 179, mean age 64 years, 39% women) and without DM (n = 1551, mean age 60 years, 46% women), we related self-reported physical activity variables to circulating concentrations of IGF-1, VEGF, and BDNF using linear multivariable regression models. We also tested for interactions by age. Participants with prevalent cardiovascular disease, stroke, and dementia or taking hormone replacement therapy were excluded. RESULTS: In participants with DM, more ambulatory physical activity was associated with higher IGF-1 levels (ß ± standard error (SE) = 0.22 ± 0.08, p = 0.009), and more total physical activity was related to higher BDNF levels (ß ± SE = 0.18 ± 0.08, p = 0.035), but physical activity was not significantly related to circulating VEGF. In participants without DM, no associations were observed. Moreover, in the examination of interactions by age, the association of ambulatory physical activity with IGF-1 levels was only observed in older adults with DM (age ≥ 60 years, ß ± SE = 0.23 ± 0.11, p = 0.042) but not in middle-aged adults with DM (age < 60 years, ß ± SE = 0.06 ± 0.13, p = 0.645). CONCLUSION: Our results suggest that more physical activity is associated with higher circulating IGF-1 and BDNF in participants with DM. These results, dissecting interactions by both age and DM status, may also help to explain some of the inconsistent results in studies relating physical activity to growth and neurotrophic factors.

Association of metformin, sulfonylurea and insulin use with brain structure and function and risk of dementia and Alzheimer's disease: Pooled analysis from 5 cohorts.

OBJECTIVE: To determine whether classes of diabetes medications are associated with cognitive health and dementia risk, above and beyond their glycemic control properties. RESEARCH DESIGN AND METHODS: Findings were pooled from 5 population-based cohorts: the Framingham Heart Study, the Rotterdam Study, the Atherosclerosis Risk in Communities (ARIC) Study, the Aging Gene-Environment Susceptibility-Reykjavik Study (AGES) and the Sacramento Area Latino Study on Aging (SALSA). Differences between users and non-users of insulin, metformin and sulfonylurea were assessed in each cohort for cognitive and brain MRI measures using linear regression models, and cognitive decline and dementia/AD risk using mixed effect models and Cox regression analyses, respectively. Findings were then pooled using meta-analytic techniques, including 3,590 individuals with diabetes for the prospective analysis. RESULTS: After adjusting for potential confounders including indices of glycemic control, insulin use was associated with increased risk of new-onset dementia (pooled HR (95% CI) = 1.58 (1.18, 2.12);p = 0.002) and with a greater decline in global cognitive function (ß = -0.014±0.007;p = 0.045). The associations with incident dementia remained similar after further adjustment for renal function and excluding persons with diabetes whose treatment was life-style change only. Insulin use was not related to cognitive function nor to brain MRI measures. No significant associations were found between metformin or sulfonylurea use and outcomes of brain function and structure. There was no evidence of significant between-study heterogeneity. CONCLUSIONS: Despite its advantages in controlling glycemic dysregulation and preventing complications, insulin treatment may be associated with increased adverse cognitive outcomes possibly due to a greater risk of hypoglycemia.

APOE and the Association of Fatty Acids With the Risk of Stroke, Coronary Heart Disease, and Mortality.

Background and Purpose- The role of dietary fat on cardiovascular health and mortality remains under debate. Because the APOE is central to the transport and metabolism of lipids, we examined associations between plasma fatty acids and the risk of stroke, coronary heart disease, and mortality by APOE-ε4 genotype. Methods- We included 943 FHS (Framingham Heart Study) and 1406 3C (Three-City) Bordeaux Study participants. Plasma docosahexaenoic, linoleic, arachidonic, and palmitic fatty acids were measured at baseline by gas chromatography. All-cause stroke, ischemic stroke, coronary heart disease, and all-cause mortality events were identified prospectively using standardized protocols. Each cohort used Cox models to separately relate fatty acid levels to the risk of developing each event during ≤10 years of follow-up adjusting for potential confounders and stratifying by APOE genotype (ε4 carriers versus noncarriers). We then meta-analyzed summary statistics using random-effects models. Results- On average, participants had a mean age of 74 years, 61% were women, and 21% (n=483) were APOE-ε4 carriers. Meta-analysis results showed that, only among APOE-ε4 carriers, every SD unit increase in linoleic acid was associated with a reduced risk of all-cause stroke (hazard ratio [HR], 0.54 [95% CI, 0.38-0.78]), ischemic stroke (HR, 0.48 [95% CI, 0.33-0.71]), and all-cause mortality (HR, 0.70 [95% CI, 0.57-0.85]). In contrast, every SD unit increase in palmitic acid was related to an increased risk of all-cause stroke (HR, 1.58 [95% CI, 1.16-2.17]), ischemic stroke (HR, 1.76 [95% CI, 1.26-2.45]), and coronary heart disease (HR, 1.48 [95% CI, 1.09-2.01]), also in APOE-ε4 carriers only. Results for docosahexaenoic acid and arachidonic acid were heterogeneous between cohorts. Conclusions- These exploratory results suggest that APOE-ε4 carriers may be more susceptible to the beneficial or adverse impact of fatty acids on cardiovascular disease and mortality. In this subgroup, higher linoleic acid was protective for stroke and mortality, whereas palmitic acid was a risk factor for stroke and coronary heart disease. The mechanisms underlying these novel findings warrant further investigation.

Atrial fibrillation and cognitive decline in the Framingham Heart Study.

BACKGROUND: There is a paucity of longitudinal research investigating the relations between atrial fibrillation (AF) and domain-specific cognitive performance. OBJECTIVE: The purpose of this study was to investigate the association between AF and cognitive performance cross-sectionally and longitudinally. METHODS: Eligible participants were dementia- and stroke-free at the time of baseline neuropsychological (NP) assessment and underwent at least 1 additional NP assessment with at least 1-year inter-test interval. AF status was examined as a 2-level variable (prevalent AF, no AF) in cross-sectional analyses and then separately as a 3-level variable (prevalent AF, interim AF, no AF) in longitudinal analyses. We examined the association between AF status and cognitive performance with linear regression. We first adjusted models for age and sex and then for vascular risk factors and apolipoprotein ε4 (APOE4) status. RESULTS: We studied 2682 participants of the Framingham Heart Study original and offspring cohorts. At the baseline NP assessment, 112 participants (4%) had AF (mean age 72 ± 9 years; 32% women). After adjustment for vascular risk factors and APOE4 status, prevalent AF was significantly associated with poorer attention; sex differences were also noted with men performing worse on tests of abstract reasoning and executive function, while women did better on a measure of executive function. Prevalent AF was significantly associated with longitudinal decline in executive function in the original cohort, and interim AF was significantly associated with longitudinal decline in executive function in the offspring cohort. CONCLUSION: After accounting for vascular risk factor burden and APOE4 status, AF was associated with a vascular profile of change in cognitive function.

Trends in the incidence of dementia: design and methods in the Alzheimer Cohorts Consortium.

Several studies have reported a decline in incidence of dementia which may have large implications for the projected burden of disease, and provide important guidance to preventive efforts. However, reports are conflicting or inconclusive with regard to the impact of gender and education with underlying causes of a presumed declining trend remaining largely unidentified. The Alzheimer Cohorts Consortium aggregates data from nine international population-based cohorts to determine changes in the incidence of dementia since 1990. We will employ Poisson regression models to calculate incidence rates in each cohort and Cox proportional hazard regression to compare 5-year cumulative hazards across study-specific epochs. Finally, we will meta-analyse changes per decade across cohorts, and repeat all analysis stratified by sex, education and APOE genotype. In all cohorts combined, there are data on almost 69,000 people at risk of dementia with the range of follow-up years between 2 and 27. The average age at baseline is similar across cohorts ranging between 72 and 77. Uniting a wide range of disease-specific and methodological expertise in research teams, the first analyses within the Alzheimer Cohorts Consortium are underway to tackle outstanding challenges in the assessment of time-trends in dementia occurrence.

Overweight, Obesity, and Survival After Stroke in the Framingham Heart Study.

BACKGROUND: We investigated how body weight affects survival after stroke, leveraging the availability of multiple prestroke body mass index (BMI) measurements and using a nested case-control design in a community-based sample. METHODS AND RESULTS: We compared all-cause mortality in participants stratified by prestroke weight. Separate analyses were performed for ischemic stroke and all stroke and for age-, sex-, and BMI category-matched stroke-free controls. Participants were grouped into BMI categories and followed for up to 10 years. Differences in survival were tested for interaction by case status. In sensitivity analysis, to exclude those with prestroke weight loss, we restricted the reference group to participants with 2 consistently normal BMI measurements within 10 years before stroke/matching. There were 782 stroke cases (age 71±9, 51% female participants, 87% ischemic stroke) and 2346 controls (age 72±9, 51% female participants). Overweight participants with ischemic stroke had a lower mortality compared with those with normal weight (hazard ratio [HR]=0.70, 95%CI 0.55-0.90, P=0.005). The association of reduced mortality with BMI ≥25, compared with normal-weight BMI 18.5 to <25, was pronounced among ischemic stroke cases but diminished with inclusion of hemorrhagic strokes (case-control interaction P=0.051 and P=0.130, respectively). Compared with participants with stable normal weight, moderately increased weight was protective after ischemic stroke (overweight HR=0.72, 95%CI 0.53-0.99, P=0.041). CONCLUSIONS: Overweight and mildly obese participants had better 10-year survival after ischemic stroke compared with normal-weight participants, even after excluding persons with recent prestroke weight loss. There may be unknown protective factors associated with a moderately increased body weight before stroke.