Jessica R. Marden

A Simulation Platform for Quantifying Survival Bias: An Application to Research on Determinants of Cognitive Decline

Bias due to selective mortality is a potential concern in many studies and is especially relevant in cognitive aging research because cognitive impairment strongly predicts subsequent mortality. Biased estimation of the effect of an exposure on rate of cognitive decline can occur when mortality is a common effect of exposure and an unmeasured determinant of cognitive decline and in similar settings. This potential is often represented as collider-stratification bias in directed acyclic graphs, but it is difficult to anticipate the magnitude of bias. In this paper, we present a flexible simulation platform with which to quantify the expected bias in longitudinal studies of determinants of cognitive decline. We evaluated potential survival bias in naive analyses under several selective survival scenarios, assuming that exposure had no effect on cognitive decline for anyone in the population. Compared with the situation with no collider bias, the magnitude of bias was higher when exposure and an unmeasured determinant of cognitive decline interacted on the hazard ratio scale to influence mortality or when both exposure and rate of cognitive decline influenced mortality. Bias was, as expected, larger in high-mortality situations. This simulation platform provides a flexible tool for evaluating biases in studies with high mortality, as is common in cognitive aging research.

Validation of a polygenic risk score for dementia in black and white individuals.

To determine whether a polygenic risk score for Alzheimers disease (AD) predicts dementia probability and memory functioning in non‐Hispanic black (NHB) and non‐Hispanic white (NHW) participants from a sample not used in previous genome‐wide association studies.

Changes in Depressive Symptoms and Incidence of First Stroke Among Middle-Aged and Older US Adults

Background Although research has demonstrated that depressive symptoms predict stroke incidence, depressive symptoms are dynamic. It is unclear whether stroke risk persists if depressive symptoms remit. Methods and Results Health and Retirement Study participants (n=16 178, stroke free and noninstitutionalized at baseline) were interviewed biennially from 1998 to 2010. Stroke and depressive symptoms were assessed through self-report of doctors’ diagnoses and a modified Center for Epidemiologic Studies - Depression scale (high was ≥3 symptoms), respectively. We examined whether depressive symptom patterns, characterized across 2 successive interviews (stable low/no, onset, remitted, or stable high depressive symptoms) predicted incident stroke (1192 events) during the subsequent 2 years. We used marginal structural Cox proportional hazards models adjusted for demographics, health behaviors, chronic conditions, and attrition. We also estimated effects stratified by age (≥65 years), race or ethnicity (non-Hispanic white, non-Hispanic black, Hispanic), and sex. Stroke hazard was elevated among participants with stable high (adjusted hazard ratio 2.14, 95% CI 1.69 to 2.71) or remitted (adjusted hazard ratio 1.66, 95% CI 1.22 to 2.26) depressive symptoms compared with participants with stable low/no depressive symptoms. Stable high depressive symptom predicted stroke among all subgroups. Remitted depressive symptoms predicted increased stroke hazard among women (adjusted hazard ratio 1.86, 95% CI 1.30 to 2.66) and non-Hispanic white participants (adjusted hazard ratio 1.66, 95% CI 1.18 to 2.33) and was marginally associated among Hispanics (adjusted hazard ratio 2.36, 95% CI 0.98 to 5.67). Conclusions In this cohort, persistently high depressive symptoms were associated with increased stroke risk. Risk remained elevated even if depressive symptoms remitted over a 2-year period, suggesting cumulative etiologic mechanisms linking depression and stroke.

Dementia and dependence Do modifiable risk factors delay disability

Objective: To identify modifying factors that preserve functional independence among individuals at high dementia risk. Methods: Health and Retirement Study participants aged 65 years or older without baseline activities of daily living (ADL) limitations (n = 4,922) were interviewed biennially for up to 12 years. Dementia probability, estimated from direct and proxy cognitive assessments, was categorized as low (i.e., normal cognitive function), mild, moderate, or high risk (i.e., very impaired) and used to predict incident ADL limitations (censoring after limitation onset). We assessed multiplicative and additive interactions of dementia category with modifiers (previously self-reported physical activity, smoking, alcohol consumption, depression, and income) in predicting incident limitations. Results: Smoking, not drinking, and income predicted incident ADL limitations and had larger absolute effects on ADL onset among individuals with high dementia probability than among cognitively normal individuals. Smoking increased the 2-year risk of ADL limitations onset from 9.9% to 14.9% among the lowest dementia probability category and from 32.6% to 42.7% among the highest dementia probability category. Not drinking increased the 2-year risk of ADL limitations onset by 2.1 percentage points among the lowest dementia probability category and 13.2 percentage points among the highest dementia probability category. Low income increased the 2-year risk of ADL limitations onset by 0.4% among the lowest dementia probability category and 12.9% among the highest dementia probability category. Conclusions: Smoking, not drinking, and low income predict incident dependence even in the context of cognitive impairment. Regardless of cognitive status, reducing these risk factors may improve functional outcomes and delay institutionalization.

Dementia and dependence

Objective: To identify modifying factors that preserve functional independence among individuals at high dementia risk. Methods: Health and Retirement Study participants aged 65 years or older without baseline activities of daily living (ADL) limitations (n = 4,922) were interviewed biennially for up to 12 years. Dementia probability, estimated from direct and proxy cognitive assessments, was categorized as low (i.e., normal cognitive function), mild, moderate, or high risk (i.e., very impaired) and used to predict incident ADL limitations (censoring after limitation onset). We assessed multiplicative and additive interactions of dementia category with modifiers (previously self-reported physical activity, smoking, alcohol consumption, depression, and income) in predicting incident limitations. Results: Smoking, not drinking, and income predicted incident ADL limitations and had larger absolute effects on ADL onset among individuals with high dementia probability than among cognitively normal individuals. Smoking increased the 2-year risk of ADL limitations onset from 9.9% to 14.9% among the lowest dementia probability category and from 32.6% to 42.7% among the highest dementia probability category. Not drinking increased the 2-year risk of ADL limitations onset by 2.1 percentage points among the lowest dementia probability category and 13.2 percentage points among the highest dementia probability category. Low income increased the 2-year risk of ADL limitations onset by 0.4% among the lowest dementia probability category and 12.9% among the highest dementia probability category. Conclusions: Smoking, not drinking, and low income predict incident dependence even in the context of cognitive impairment. Regardless of cognitive status, reducing these risk factors may improve functional outcomes and delay institutionalization.

Do physical activity, smoking, drinking, or depression modify transitions from cognitive impairment to functional disability?

BACKGROUND Individual-level modifiers can delay onset of limitations in basic activities of daily living (ADLs) among cognitively impaired individuals. We assessed whether these modifiers also delayed onset of limitations in instrumental ADLs (IADLs) among individuals at elevated dementia risk. OBJECTIVES To determine whether modifiable individual-level factors delay incident IADL limitations among adults stratified by dementia risk. METHODS Health and Retirement Study participants aged 65+ without activity limitations in 1998 or 2000 (n = 5,219) were interviewed biennially through 2010. Dementia probability, categorized in quartiles, was used to predict incident IADL limitations with Poisson regression. We estimated relative (risk ratio) and absolute (number of limitations) effects from models including dementia, individual-level modifiers (physical inactivity, smoking, no alcohol consumption, and depression) and interaction terms between dementia and individual-level modifiers. RESULTS Dementia probability quartile predicted incident IADL limitations (relative risk for highest versus lowest quartile = 0.44; 95% CI: 0.28-0.70). Most modifiers did not significantly increase risk of IADL limitations among the cognitively impaired. Physical inactivity (RR = 1.60; 95% CI: 1.16, 2.19) increased the risk of IADL limitations among the cognitively impaired. The interaction between physical inactivity and low dementia probability was statistically significant (p = 0.009) indicating that physical inactivity had significantly larger effects on incident IADLs among cognitively normal than among those with high dementia probability. CONCLUSION Physical activity may protect against IADL limitations while not smoking, alcohol consumption, and not being depressed do not afford substantial protection among the cognitively impaired. RESULTS highlight the need for extra support for IADLs among individuals with cognitive losses.

Diabetic Phenotypes and Late-Life Dementia Risk: A Mechanism-specific Mendelian Randomization Study.

Background:Mendelian Randomization (MR) studies have reported that type 2 diabetes (T2D) was not associated with Alzheimer disease (AD). We adopted a modified, mechanism-specific MR design to explore this surprising result. Methods:Using inverse-variance weighted MR analysis, we evaluated the association between T2D and AD using data from 39 single nucleotide polymorphisms (SNPs) significantly associated with T2D in DIAbetes Genetics Replication And Meta-analysis (DIAGRAM) and the corresponding associations of each SNP with AD risk obtained from the International Genomics of Alzheimer’s Project (IGAP, n=17,008 AD cases and n=37,154 controls). We evaluated mechanism-specific genetic subscores, including &bgr;-cell function, insulin sensitivity, and adiposity, and repeated analyses in 8501 Health and Retirement Study participants for replication and model validation. Results:In IGAP, the overall T2D polygenic score did not predict AD [odds ratio (OR) for the T2D polygenic score=1.01; 95% confidence interval (CI), 0.96, 1.06] but the insulin sensitivity polygenic score predicted higher AD risk (OR=1.17; 95% CI, 1.02, 1.34). In the Health and Retirement Study, polygenic scores were associated with T2D risk; the associations between insulin sensitivity genetic polygenic score and cognitive phenotypes were not statistically significant. Conclusions:Evidence from polygenic scores suggests that insulin sensitivity specifically may affect AD risk, more than T2D overall.

High Hemoglobin A1c and Diabetes Predict Memory Decline in the Health and Retirement Study

Background: Type 2 diabetes (T2D) is an established risk factor for dementia, but evidence for T2D and memory decline is less consistent. Understanding how T2D and blood glucose relate to memory decline is crucial to elucidating the mechanisms linking T2D and dementia. Materials and Methods: For 8888 Health and Retirement Study participants aged 50+, glycosylated hemoglobin (HbA1c) was measured in either 2006 or 2008 and physician’s diagnosis of diabetes was self-reported in the same year. Composite memory (z scored) was assessed biennially through 2012 using immediate and delayed word list recall or the Informant Questionnaire for Cognitive Decline. Marginal mean regression models for repeated outcomes were specified to predict memory decline as a function of diabetes or HbA1c, using age as the timescale and adjusting for health and social confounders. Results: Diabetes was associated with a 10% faster rate of memory decline [&bgr;=−0.04 per decade; 95% confidence interval (CI), −0.06 to −0.01). A 1 U increase in HbA1c corresponded with a 0.05 SD decrease in memory score per decade (95% CI, −0.08 to −0.03). Even among individuals with HbA1c<6.5% (threshold for diabetes), higher HbA1c was associated with memory decline (&bgr;=−0.05 per decade; 95% CI, −0.08 to −0.03). Discussion: Diabetes accelerated memory loss and higher HbA1c predicted memory decline even in nondiabetics.

African Ancestry, Social Factors, and Hypertension Among Non-Hispanic Blacks in the Health and Retirement Study

The biomedical literature contains much speculation about possible genetic explanations for the large and persistent black–white disparities in hypertension, but profound social inequalities are also hypothesized to contribute to this outcome. Our goal is to evaluate whether socioeconomic status (SES) differences provide a plausible mechanism for associations between African ancestry and hypertension in a U.S. cohort of older non-Hispanic blacks. We included only non-Hispanic black participants (N = 998) from the Health and Retirement Study who provided genetic data. We estimated percent African ancestry based on 84,075 independent single nucleotide polymorphisms using ADMIXTURE V1.23, imposing K = 4 ancestral populations, and categorized into quartiles. Hypertension status was self-reported in the year 2000. We used linear probability models (adjusted for age, sex, and southern birth) to predict prevalent hypertension with African ancestry quartile, before and after accounting for a small set of SES measures. Respondents with the highest quartile of African ancestry had 8 percentage points’ (RD = 0.081; 95% CI: −0.001, 0.164) higher prevalence of hypertension compared to the lowest quartile. Adjustment for childhood disadvantage, education, income, and wealth explained over one-third (RD = 0.050; 95% CI: −0.034, 0.135) of the disparity. Explanations for the residual disparity remain unspecified and may include other indicators of SES or diet, lifestyle, and psychosocial mechanisms.

Mother’s education and late-life disparities in memory and dementia risk among US military veterans and non-veterans

Background Adverse childhood socioeconomic status (cSES) predicts higher late-life risk of memory loss and dementia. Veterans of U.S. wars are eligible for educational and economic benefits that may offset cSES disadvantage. We test whether cSES disparities in late-life memory and dementia are smaller among veterans than non-veterans. Methods Data came from US-born men in the 1995–2014 biennial surveys of the Health and Retirement Study (n=7916 born 1928–1956, contributing n=38 381 cognitive assessments). Childhood SES was represented by maternal education. Memory and dementia risk were assessed with brief neuropsychological assessments and proxy reports. Military service (veteran/non-veteran) was evaluated as a modifier of the effect of maternal education on memory and dementia risk. We employed linear or logistic regression models to test whether military service modified the effect of maternal education on memory or dementia risk, adjusted for age, race, birthplace and childhood health. Results Low maternal education was associated with worse memory than high maternal education (β = −0.07 SD, 95% CI −0.08 to –0.05), while veterans had better memory than non-veterans (β = 0.03 SD, 95% CI 0.02 to 0.04). In interaction analyses, maternal education disparities in memory were smaller among veterans than non-veterans (difference in disparities = 0.04 SD, 95% CI 0.01 to 0.08, p = 0.006). Patterns were similar for dementia risk. Conclusions Disparities in memory by maternal education were smaller among veterans than non-veterans, suggesting military service and benefits partially offset the deleterious effects of low maternal education on late-life cognitive outcomes.