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

Characterizing brain tau and cognitive decline along the amyloid timeline in Alzheimer's disease

Recent longitudinal PET imaging studies have established methods to estimate the age at which amyloid becomes abnormal at the level of the individual. Here we recontextualized amyloid levels into the temporal domain to better understand the downstream Alzheimer's disease processes of tau neurofibrillary tangle (NFT) accumulation and cognitive decline. This cohort study included a total of 601 individuals from the Wisconsin Registry for Alzheimer's Prevention and Wisconsin Alzheimer's Disease Research Center that underwent amyloid and tau PET, longitudinal neuropsychological assessments and met clinical criteria for three clinical diagnosis groups: cognitively unimpaired (n = 537); mild cognitive impairment (n = 48); or dementia (n = 16). Cortical 11C-Pittsburgh compound B (PiB) distribution volume ratio (DVR) and sampled iterative local approximation were used to estimate amyloid positive (A+; global PiB DVR > 1.16 equivalent to 17.1 centiloids) onset age and years of A+ duration at tau PET (i.e. amyloid chronicity). Tau PET burden was quantified using 18F-MK-6240 standardized uptake value ratios (70-90 min, inferior cerebellar grey matter reference region). Whole-brain and region-specific approaches were used to examine tau PET binding along the amyloid timeline and across the Alzheimer's disease clinical continuum. Voxel-wise 18F-MK-6240 analyses revealed that with each decade of A+, the spatial extent of measurable tau spread (i.e. progressed) from regions associated with early to late NFT tau stages. Regional analyses indicated that tau burden in the entorhinal cortex was detectable, on average, within 10 years of A+ onset. Additionally, the entorhinal cortex was the region most sensitive to early amyloid pathology and clinical impairment in this predominantly preclinical sample. Among initially cognitively unimpaired (n = 472) individuals with longitudinal cognitive follow-up, mixed effects models showed significant linear and non-linear interactions of A+ duration and entorhinal tau on cognitive decline, suggesting a synergistic effect whereby greater A+ duration, together with a higher entorhinal tau burden, increases the likelihood of cognitive decline beyond their separable effects. Overall, the amyloid time framework enabled a spatiotemporal characterization of tau deposition patterns across the Alzheimer's disease continuum. This approach, which examined cross-sectional tau PET data along the amyloid timeline to make longitudinal disease course inferences, demonstrated that A+ duration explains a considerable amount of variability in the magnitude and topography of tau spread, which largely recapitulated NFT staging observed in human neuropathological studies. By anchoring disease progression to the onset of amyloid, this study provides a temporal disease context, which may help inform disease prognosis and timing windows for anti-amyloid therapies.

Socioeconomic Status Transition Throughout Life and Risk of Dementia

Importance

Socioeconomic status (SES) is associated with dementia. However, the role of SES transitions in dementia is less explored; such evidence would be useful to understand whether social mobility is associated with healthy longevity at older ages.

Objective

To investigate the association of lifetime SES transition with risk of dementia.

Design, Setting, and Participants

This prospective cohort study, conducted from August 2010 to December 2016, used data from the Japan Gerontological Evaluation Study for participants aged 65 years or older from 31 different areas in Japan. Individuals with missing SES values, loss of follow-up, or new dementia onset 1 year or less from baseline were excluded. Data analysis was performed from April 2022 to April 2023.

Exposure

Transitions in SES across the life course.

Main Outcomes and Measures

The main outcome was risk of dementia incidence and corresponding loss or gain of dementia-free periods in a lifespan. The incidence of dementia was identified with a national registry of long-term nursing care services.

Results

A total of 9186 participants (4703 men [51.2%]) were included. The mean (SD) age at baseline was 74.2 (6.0) years. Six SES transitions were identified: upward, stable-high, upper-middle, lower-middle, downward, and stable-low. During the follow-up period, 800 cases of dementia were identified. Many dementia risk factors, including lifestyle behaviors, comorbidities, and social factors, were associated with SES transition patterns. Compared with lower-middle SES, the lowest risk of dementia was observed for upward transition (hazard ratio [HR], 0.66; 95% CI, 0.57-0.74) followed by stable-high (HR, 0.77; 95% CI, 0.69-0.86), downward (HR, 1.15; 95% CI, 1.09-1.23), and stable-low (HR 1.45; 95% CI, 1.31-1.61) transition (P <  .001 for linearity); there was no association of upper-middle transition with risk of dementia (HR, 0.91; 95% CI, 0.79-1.03). The greatest increases in dementia-free years in the lifespan were also associated with upward SES transition (eg, 1.8 years [95% CI, 1.4-2.2 years] at age 65 years), while the downward transition was associated with the largest loss in lifetime dementia-free years at 75 years or older (eg, -1.4 years [95% CI, -2.4 to -0.4 years] at age 85 years).

Conclusions and Relevance

This cohort study of Japanese older adults identified that upward and downward SES transitions were associated with risk of dementia and the length of dementia-free periods over the lifespan. The results may be useful to understand the association between social mobility and healthy longevity.

Associations between circulating cardiovascular disease risk factors and cognitive performance in cognitively healthy older adults from the NuAge study

Introduction

Cardiovascular disease risk factors (CVRFs) contribute to the development of cognitive impairment and dementia.

Methods

This study examined the associations between circulating CVRF biomarkers and cognition in 386 cognitively healthy older adults (mean age = 78 ± 4 years, 53% females) selected from the Quebec Longitudinal Study on Nutrition and Successful Aging (NuAge). Memory, executive function, and processing speed were assessed at baseline and 2-year follow-up. CVRF biomarkers included total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C), triglycerides, glucose, insulin, high sensitivity C-reactive protein (hs-CRP), homocysteine, protein carbonyls, and cortisol. Linear mixed models were used to determine associations between individual CVRF biomarkers and cognition at both time points.

Results

HDL-C was most consistently associated with cognition with higher values related to better performance across several domains. Overall, stronger and more consistent relationships between CVRF biomarkers and cognition were observed in females relative to males.

Discussion

Findings suggest that increases in the majority of circulating CVRFs are not associated with worse cognition in cognitively healthy older adults.

Causal Knowledge as a Prerequisite for Interrogating Bias: Reflections on Hernán et al. 20 Years Later

In their seminal 2002 paper, "Causal Knowledge as a Prerequisite for Confounding Evaluation: An Application to Birth Defects Epidemiology," Hernán et al. (Am J Epidemiol. 2002;155(2):176-184) emphasized the importance of using theory rather than data to guide confounding control, focusing on colliders as variables that share characteristics with confounders but whose control may actually introduce bias into analyses. In this commentary, we propose that the importance of this paper stems from the connection the authors made between nonexchangeability as the ultimate source of bias and structural representations of bias using directed acyclic graphs. This provided both a unified approach to conceptualizing bias and a means of distinguishing between different sources of bias, particularly confounding and selection bias. Drawing on examples from the paper, we also highlight unresolved questions about the relationship between collider bias, selection bias, and generalizability and argue that causal knowledge is a prerequisite not only for identifying confounders but also for developing any hypothesis about potential sources of bias.

Functional aging trajectories of older cancer survivors: a latent growth analysis of the US Health and Retirement Study

PURPOSE

We aimed to identify prototypical functional aging trajectories of US cancer survivors aged 50 and older, overall and stratified by sociodemographic and health-related characteristics.

METHODS

Data were from 2986 survivors of a first incident cancer diagnosis (except non-melanoma skin cancer) after age 50 in the population representative U.S. Health and Retirement Study from 1998-2016. Cancer diagnoses, episodic memory function, and activity of daily living (ADL) limitations were assessed at biennial study interviews. Using time of cancer diagnosis as the baseline, we used group-based trajectory modeling to identify trajectories of memory function and ADL limitations following diagnosis.

RESULTS

We identified five memory loss trajectories (high: 8.4%; medium-high: 18.3%; medium-low: 21.5%; low: 25.5%; and, very low: 26.2%), and four ADL limitation trajectories (high/increasing limitations: 18.7%; medium limitations: 18.7%; low limitations: 8.14%; no limitations: 60.0). The high memory loss and high/increasing ADL limitation trajectories were both characterized by older age, being female (52% for memory, 58.9% for ADL), having lower pre-cancer memory scores, and a higher prevalence of pre-cancer comorbidities including stroke (30.9% for memory and 29.7% for ADL), hypertension (64.7% for memory and 69.8 for ADL), and depressive symptoms. In joint analyses, we found that generally those with higher memory were more likely to have fewer ADL limitations and vice versa.

CONCLUSION

Older cancer survivors experience heterogeneous trajectories of functional aging that are largely characterized by comorbidities prior to diagnosis.

IMPLICATION FOR CANCER SURVIVORS

Results can help identify older cancer survivors at increased risk for accelerated functional decline.

Assessing the impacts of smoking cessation and resumption on back pain risk in later life

BACKGROUND

This study aimed to assess the impacts of smoking cessation and resumption over 4 years on the risk of back pain at the 6-year follow-up among older adults in England.

METHODS

We analysed 6467 men and women aged ≥50 years in the English Longitudinal Study of Aging. Self-reported smoking status, assessed in waves 4 (2008-2009) and 6 (2012-2013), was used as exposure for the study, whereas self-reported back pain of moderate or severe intensity, assessed in wave 7 (2014-2015), was used as the outcome. A targeted minimum loss-based estimator was used with longitudinal modified treatment policies to adjust for baseline and time-varying covariates.

RESULTS

Regarding the estimation of the effects of changes in smoking status on the risk of back pain, during the follow-up, individuals who resumed smoking within 4 years had a higher risk of back pain than those who avoided smoking for over 4 years, and the relative risk (RR) (95% confidence interval [CI]) was 1.536 (1.214-1.942). Regarding the estimation of effects of smoking cessation on the risk of back pain, smoking cessation over 4 years was associated with a significantly lower risk of back pain, as indicated by the originally observed data, and the RR (95% CI) was 0.955 (0.912-0.999).

CONCLUSIONS

Older adults who avoided smoking for over 4 years had a lower risk of back pain. However, those who resumed smoking within 4 years had a higher risk of back pain.

SIGNIFICANCE

Older adults who avoided smoking for over 4 years had a lower risk of back pain. However, those who resumed smoking within 4 years had a higher risk of back pain. Our study data suggest the importance of maintaining smoking cessation to reduce the risk of back pain in the older population.

Attrition in the Gothenburg H70 birth cohort studies, an 18-year follow-up of the 1930 cohort

Background

Longitudinal studies are essential to understand the ageing process, and risk factors and consequences for disorders, but attrition may cause selection bias and impact generalizability. We describe the 1930 cohort of the Gothenburg H70 Birth Cohort Studies, followed from age 70 to 88, and compare baseline characteristics for those who continue participation with those who die, refuse, and drop out for any reason during follow-up.

Methods

A population-based sample born 1930 was examined with comprehensive assessments at age 70 (N = 524). The sample was followed up and extended to increase sample size at age 75 (N = 767). Subsequent follow-ups were conducted at ages 79, 85, and 88. Logistic regression was used to analyze baseline characteristics in relation to participation status at follow-up.

Results

Refusal to participate in subsequent examinations was related to lower educational level, higher blood pressure, and lower scores on cognitive tests. Both attrition due to death and total attrition were associated with male sex, lower educational level, smoking, ADL dependency, several diseases, poorer lung function, slower gait speed, lower scores on cognitive tests, depressive symptoms, and a larger number of medications. Attrition due to death was also associated with not having a partner.

Conclusions

It is important to consider different types of attrition when interpreting results from longitudinal studies, as representativeness and results may be differently affected by different types of attrition. Besides reducing barriers to participation, methods such as imputation and weighted analyses can be used to handle selection bias.

Generational and Social Forces in the Life Events and Experiences of Lesbian and Gay Midlife and Older Adults Across the Iridescent Life Course

ObjectivesUtilizing Iridescent Life Course, we examine life events among three generations of lesbian and gay adults: Invisible (born 1920-1934), Silenced (born 1935-1949), and Pride (born 1950-1964) Generations. Methods: We utilized a subsample (n = 2079) from the 2014 wave of Aging with Pride: National Health, Aging, and Sexuality/Gender Study (NHAS). Demographic characteristics, life events, and gender and generational interactions were compared. Results: Compared to other generations, the Invisible Generation disclosed their identity at older ages, were more likely to be retired, served in the military, and survived a partner's death. Compared to the other generations, the Pride Generation was more likely to have disclosed their identities earlier and experienced higher levels of victimization/discrimination. Discussion: This paper is the first to examine the lived experiences of the oldest lesbians and gay men and compare them to other generations. The findings illustrate the heteronormative nature of most life course research.

The impact of repeated vaccination on relative influenza vaccine effectiveness among vaccinated adults in the United Kingdom

Annual seasonal influenza vaccination is recommended for individuals at high risk of developing post-infection complications in many locations. However, reduced vaccine immunogenicity and effectiveness have been observed among repeat vaccinees in some influenza seasons. We investigated the impact of repeated influenza vaccination on relative vaccine effectiveness (VE) among individuals who were recommended for influenza vaccination in the United Kingdom with a retrospective cohort study using primary healthcare data from the Clinical Practice Research Datalink, a primary care database in the United Kingdom. Relative VE was estimated against general practitioner-diagnosed influenza-like illnesses (GP-ILI) and medically attended acute respiratory illnesses (MAARI) among participants who have been repeatedly vaccinated compared with first-time vaccinees using proportional hazards models. Relative VE against MAARI may be reduced for individuals above 65 years old who were vaccinated in the current and previous influenza seasons for some influenza seasons. However, these findings were not conclusive as we could not exclude the possibility of residual confounding in our dataset. The use of routinely collected data from electronic health records to examine the effects of repeated vaccination needs to be complemented with sufficient efforts to include negative control outcomes to rule out residual confounding.

Quantifying the impact of environment factors on the risk of medical responders' stress-related absenteeism

Medical emergency response staff are exposed to incidents which may involve high-acuity patients or some intractable or traumatic situations. Previous studies on emergency response staff stress-related absence have focused on perceived factors and their impacts on absence leave. To date, analytical models on absenteeism risk prediction use past absenteeism to predict risk of future absenteeism. We show that these approaches ignore environment data, such as stress factors. The increased use of digital systems in emergency services allows us to gather data that were not available in the past and to apply a data-driven approach to quantify the effect of environment variables on the risk of stress-related absenteeism. We propose a two-stage data-driven framework to identify the variables of importance and to quantify their impact on medical staff stress-related risk of absenteeism. First, machine learning techniques are applied to identify the importance of different stressors on staff stress-related risk of absenteeism. Second, the Cox proportional-hazards model is applied to estimate the relative risk of each stressor. Four significant stressors are identified, these are the average night shift, past stress leave, the squared term of death confirmed by the Emergency Services and completion of the safeguarding form. We discuss counterintuitive results and implications to policy.

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.

Estimating the Impact of Sustained Social Participation on Depressive Symptoms in Older Adults

BACKGROUND

Social participation has been suggested as a means to prevent depressive symptoms. However, it remains unclear whether a one-time boost suffices or whether participation needs to be sustained over time for long-term prevention. We estimated the impacts of alternative hypothetical interventions in social participation on subsequent depressive symptoms among older adults.

METHODS

Data were from a nationwide prospective cohort study of Japanese older adults ≥65 years of age (n = 32,748). We analyzed social participation (1) as a baseline exposure from 2010 (approximating a one-time boost intervention) and (2) as a time-varying exposure from 2010 and 2013 (approximating a sustained intervention). We defined binary depressive symptoms in 2016 using the Geriatric Depression Scale. We used the doubly robust targeted maximum likelihood estimation to address time-dependent confounding.

RESULTS

The magnitude of the association between sustained participation and the lower prevalence of depressive symptoms was larger than the association observed for baseline participation only (e.g., prevalence ratio [PR] for participation in any activity = 0.83 [95% confidence interval = 0.79, 0.88] vs. 0.90 [0.87, 0.94]). For activities with a lower proportion of consistent participation over time (e.g., senior clubs), there was little evidence of an association between baseline participation and subsequent depressive symptoms, while an association for sustained participation was evident (e.g., PR for senior clubs = 0.96 [0.90, 1.02] vs. 0.88 [0.79, 0.97]). Participation at baseline but withholding participation in 2013 was not associated with subsequent depressive symptoms.

CONCLUSIONS

Sustained social participation may be more strongly associated with fewer depressive symptoms among older adults.

Monte Carlo Simulation Approaches for Quantitative Bias Analysis: A Tutorial

Quantitative bias analysis can be used to empirically assess how far study estimates are from the truth (i.e., an estimate that is free of bias). These methods can be used to explore the potential impact of confounding bias, selection bias (collider stratification bias), and information bias. Quantitative bias analysis includes methods that can be used to check the robustness of study findings to multiple types of bias and methods that use simulation studies to generate data and understand the hypothetical impact of specific types of bias in a simulated dataset. In this article, we review two strategies for quantitative bias analysis: 1) traditional probabilistic quantitative bias analysis and 2) quantitative bias analysis with generated data. An important difference between the two strategies relates to the type of data (real vs. generated data) used in the analysis. As we describe, Monte Carlo simulations are used in both approaches, but the simulation process is used for different purposes in each. For both approaches, we outline and describe the steps required to carry out the quantitative bias analysis and also present a bias analysis tutorial demonstrating how both approaches can be applied in the context of an analysis for selection bias. Our goal is to highlight the utility of quantitative bias analysis for practicing epidemiologists and increase the use of these methods in the epidemiologic literature.

The Influence of Socioeconomic Deprivation on Dementia Mortality, Age at Death, and Quality of Diagnosis: A Nationwide Death Records Study in England and Wales 2001-2017

BACKGROUND

Socioeconomic deprivation may be an important determinant of dementia risk, mortality, and access to diagnostic services. Premature mortality from other causes and under-representation of deprived individuals in research may lead to this effect being overlooked.

OBJECTIVE

We assessed the relationship between deprivation and dementia mortality using comprehensive death certificate data for England and Wales from 2001 to 2017.

METHODS

We used standardized mortality ratios (SMR) and a Poisson model to compare likelihood of dying from dementia in each deprivation decile. We also examined the associations of deprivation with age at death from dementia, and with likelihood of receiving a diagnosis of unspecified dementia.

RESULTS

Risk of dying from dementia was higher in more deprived deciles (Mean SMR [95% CI] in decile 1 : 0.528 [0.506 to 0.550], decile 10:0.369 [0.338 to 0.400]). In 2017, 14,837 excess dementia deaths were attributable to deprivation (21.5% of all dementia deaths that year). There were dose-response associations of deprivation with likelihood of being older at death with dementia (odds ratio [95% CI] for decile 10 (least deprived): 1.31 [1.28 to 1.33] relative to decile 1), and with likelihood of receiving a diagnosis of unspecified dementia (odds ratio [95% CI] for decile 10:0.78 [0.76 to 0.80] relative to decile 1).

CONCLUSION

Socioeconomic deprivation in England and Wales is associated with increased dementia mortality, younger age at death with dementia, and poorer access to specialist diagnosis. Reducing social inequality may have a role in the prevention of dementia mortality.

Causal Inference in Studying the Long-Term Health Effects of Disasters: Challenges and Potential Solutions

Two frequently encountered but underrecognized challenges for causal inference in studying the long-term health effects of disasters among survivors include 1) time-varying effects of disasters on a time-to-event outcome and 2) selection bias due to selective attrition. In this paper, we review approaches for overcoming these challenges and demonstrate application of the approaches to a real-world longitudinal data set of older adults who were directly affected by the 2011 Great East Japan Earthquake and Tsunami (n = 4,857). To illustrate the problem of time-varying effects of disasters, we examined the association between degree of damage due to the tsunami and all-cause mortality. We compared results from Cox regression analysis assuming proportional hazards with those derived using adjusted parametric survival curves allowing for time-varying hazard ratios. To illustrate the problem of selection bias, we examined the association between proximity to the coast (a proxy for housing damage from the tsunami) and depressive symptoms. We corrected for selection bias due to attrition in the 2 postdisaster follow-up surveys (conducted in 2013 and 2016) using multivariable adjustment, inverse probability of censoring weighting, and survivor average causal effect estimation. Our results demonstrate that analytical approaches which ignore time-varying effects on mortality and selection bias due to selective attrition may underestimate the long-term health effects of disasters.

Lifetime post-traumatic stress disorder in older individuals with a history of institutional upbringing in childhood: the role of social acknowledgement and stressful life events

Background: Child maltreatment (CM), particularly in institutional contexts, can affect the development of post-traumatic stress disorder (PTSD). Research suggests that factors during CM (e.g. severity, variety, duration) and in the aftermath of CM (e.g. stressful life events, and social acknowledgement, i.e. the degree to which an individual feels validated and supported following a traumatic event) can explain some of the heterogeneity in PTSD development. However, there is a lack of research on long-term correlates of CM and mitigating factors, with only a few studies having been conducted with older survivors of institutional upbringing. Such research is relevant, given the long-term associations between CM and the older age status of many survivors. Objective: The current study examined the link between CM and PTSD in older individuals with a history of institutional upbringing (risk group; RG) and a matched control group (CG). Differences in stressful life events and social acknowledgement were also investigated. Method: Participants were n = 116 RG (M_age = 70.25 years, 41% female) and n = 122 CG (M_age = 70.71 years, 51% female). Data was assessed using self-report questionnaires and a clinical interview. Results: The RG reported higher levels of exposure to CM. Lifetime PTSD showed a bigger association with the level of exposure to CM, compared to having an institutional upbringing. Participants with higher CM levels reported more stressful life events. High levels of social acknowledgement mediated the relationship between CM and PTSD in the CG. Conclusions: Exposure to CM had a stronger association with PTSD than a history of institutional upbringing. In the CG, the survivors' perception of social acknowledgement ameliorated lifetime PTSD to a small extent. A critical issue for policy makers should be to enhance safeguarding measures against CM exposure, not only in institutional contexts, but also more generally, given the link to PTSD.

Uncovering survivorship bias in longitudinal mental health surveys during the COVID-19 pandemic

AIMS

Markedly elevated adverse mental health symptoms were widely observed early in the coronavirus disease-2019 (COVID-19) pandemic. Unlike the U.S., where cross-sectional data indicate anxiety and depression symptoms have remained elevated, such symptoms reportedly declined in the U.K., according to analysis of repeated measures from a large-scale longitudinal study. However, nearly 40% of U.K. respondents (those who did not complete multiple follow-up surveys) were excluded from analysis, suggesting that survivorship bias might partially explain this discrepancy. We therefore sought to assess survivorship bias among participants in our longitudinal survey study as part of The COVID-19 Outbreak Public Evaluation (COPE) Initiative.

METHODS

Survivorship bias was assessed in 4039 U.S. respondents who completed surveys including the assessment of mental health as part of The COPE Initiative in April 2020 and were invited to complete follow-up surveys. Participants completed validated screening instruments for symptoms of anxiety, depression and insomnia. Survivorship bias was assessed for (1) demographic differences in follow-up survey participation, (2) differences in initial adverse mental health symptom prevalence adjusted for demographic factors and (3) differences in follow-up survey participation based on mental health experiences adjusted for demographic factors.

RESULTS

Adjusting for demographics, individuals who completed only one or two out of four surveys had significantly higher prevalence of anxiety and depression symptoms in April 2020 (e.g. one-survey v. four-survey, anxiety symptoms, adjusted prevalence ratio [aPR]: 1.30, 95% confidence interval [CI]: 1.08-1.55, p = 0.0045; depression symptoms, aPR: 1.43, 95% CI: 1.17-1.75, p = 0.00052). Moreover, individuals who experienced incident anxiety or depression symptoms had significantly higher adjusted odds of not completing follow-up surveys (adjusted odds ratio [aOR]: 1.68, 95% CI: 1.22-2.31, p = 0.0015, aOR: 1.56, 95% CI: 1.15-2.12, p = 0.0046, respectively).

CONCLUSIONS

Our findings reveal significant survivorship bias among longitudinal survey respondents, indicating that restricting analytic samples to only respondents who provide repeated assessments in longitudinal survey studies could lead to overly optimistic interpretations of mental health trends over time. Cross-sectional or planned missing data designs may provide more accurate estimates of population-level adverse mental health symptom prevalence than longitudinal surveys.

Uncovering Survivorship Bias in Longitudinal Mental Health Surveys During the COVID-19 Pandemic

Aims

Markedly elevated adverse mental health symptoms were widely observed early in the coronavirus disease 2019 (COVID-19) pandemic. Unlike the U.S., where cross-sectional data indicate anxiety and depression symptoms have remained elevated, such symptoms reportedly declined in the U.K., according to analysis of repeated measures from a largescale longitudinal study. However, nearly 40% of U.K. respondents (those who did not complete multiple follow-up surveys) were excluded from analysis, suggesting that survivorship bias might partially explain this discrepancy. We therefore sought to assess survivorship bias among participants in our longitudinal survey study as part of The COVID-19 Outbreak Public Evaluation (COPE) Initiative.

Methods

Survivorship bias was assessed 4,039 U.S. respondents who completed surveys including the assessment of mental health as part of The COPE Initiative in April 2020 and were invited to complete follow-up surveys. Participants completed validated screening instruments for symptoms of anxiety, depression, and insomnia. Survivorship bias was assessed for (1) demographic differences in follow-up survey participation, (2) differences in initial adverse mental health symptom prevalences adjusted for demographic factors, and (3) differences in follow-up survey participation based on mental health experiences adjusted for demographic factors.

Results

Adjusting for demographics, individuals who completed only one or two out of four surveys had higher prevalences of anxiety and depression symptoms in April 2020 (e.g., one-survey versus four-survey, anxiety symptoms, adjusted prevalence ratio [aPR]: 1.30, 95% confidence interval [CI]: 1.08–1.55, P=0.0045; depression symptoms, aPR: 1.43, 95% CI: 1.17–1.75, P=0.00052). Moreover, individuals who experienced incident anxiety or depression symptoms had higher odds of not completing follow-up surveys (adjusted odds ratio [aOR]: 1.68, 95% CI: 1.22–2.31, P=0.0015, aOR: 1.56, 95% CI: 1.15–2.12, P=0.0046, respectively).

Conclusions

Our findings revealed significant survivorship bias among longitudinal survey respondents, indicating that restricting analytic samples to only respondents who provide repeated assessments in longitudinal survey studies could lead to overly optimistic interpretations of mental health trends over time. Cross-sectional or planned missing data designs may provide more accurate estimates of population-level adverse mental health symptom prevalences than longitudinal surveys.

Environmental hazards, social inequality, and fetal loss: Implications of live-birth bias for estimation of disparities in birth outcomes

Restricting to live births can induce bias in studies of pregnancy and developmental outcomes, but whether this live-birth bias results in underestimating disparities is unknown. Bias may arise from collider stratification due to an unmeasured common cause of fetal loss and the outcome of interest, or depletion of susceptibles, where exposure differentially causes fetal loss among those with underlying susceptibility.

METHODS

We conducted a simulation study to examine the magnitude of live-birth bias in a population parameterized to resemble one year of conceptions in California (N = 625,000). We simulated exposure to a non-time-varying environmental hazard, risk of spontaneous abortion, and time to live birth using 1000 Monte Carlo simulations. Our outcome of interest was preterm birth. We included a social vulnerability factor to represent social disadvantage, and estimated overall risk differences for exposure and preterm birth using linear probability models and stratified by the social vulnerability factor. We calculated how often confidence intervals included the true point estimate (CI coverage probabilities) to illustrate whether effect estimates differed qualitatively from the truth.

RESULTS

Depletion of susceptibles resulted in a larger magnitude of bias compared with collider stratification, with larger bias among the socially vulnerable group. Coverage probabilities were not adversely affected by bias due to collider stratification. Depletion of susceptibles reduced coverage, especially among the socially vulnerable (coverage among socially vulnerable = 46%, coverage among nonsocially vulnerable = 91% in the most extreme scenario).

CONCLUSIONS

In simulations, hazardous environmental exposures induced live-birth bias and the bias was larger for socially vulnerable women.

Evaluation of Selective Survival and Sex/Gender Differences in Dementia Incidence Using a Simulation Model

IMPORTANCE

Dementia research is susceptible to bias arising from selective survival, a process that results in individuals with certain characteristics disproportionately surviving to old age. Spurious associations between risk factors and dementia may be induced when factors associated with longer survival also influence dementia incidence.

OBJECTIVE

To assess the role of selective survival in explaining reported sex/gender differences in dementia incidence.

DESIGN, SETTING, AND PARTICIPANTS

This decision analytical model used a simulated cohort of US participants aged 50 years and without dementia at baseline followed up for incident dementia through age 95 years. Selective survival was induced by a selection characteristic (eg, childhood social disadvantage or Alzheimer genetic risk) that influenced both mortality and dementia incidence at varying magnitudes. Data analysis was performed from April 2018 to May 2020.

EXPOSURE

Sex/gender, conceptualized as the combination of biological sex and social consequences of gender.

MAIN OUTCOMES AND MEASURES

Dementia incidence rate ratios (IRRs) for women compared with men. In all simulations, it was assumed that there would be no true effect of sex/gender on dementia incidence; all observed sex/gender differences were due to selective survival.

RESULTS

At baseline, the simulation included 100 000 participants aged 50 years (51 000 [51%] women, mirroring the 1919-1921 US birth cohort of non-Latino White individuals at age 50 years); distributions of the selection characteristic were standard normal (mean [SD], 0.0 [1.0]). Observed sex/gender differences in dementia incidence in individuals aged 85 years or older ranged from insignificant (IRR, 1.00; 95% CI, 0.91-1.11) to consistent with sex/gender differences (20% higher risk for women [IRR, 1.20; 95% CI, 1.08-1.32]) reported in an extant study. Simulations in which bias was large enough to explain prior findings required moderate to large differential effects of selective survival (eg, hazard ratio for selection characteristic on mortality at least 2.0 among men, no effect among women).

CONCLUSIONS AND RELEVANCE

These results suggest that selective survival may contribute to observed sex/gender differences in dementia incidence but do not preclude potential contributions of sex/gender-specific mechanisms. Further research on plausibility of selection characteristics with outcomes of the magnitude required for selective survival to explain sex/gender differences in dementia incidence and sex/gender-specific mechanisms represent an opportunity to understand prevention and treatment of dementia.

Cognitive ageing is premature among a community sample of optimally treated people living with HIV

OBJECTIVES

Evidence of premature cognitive ageing amongst people living with HIV (PLHIV) remains controversial due to previous research limitations including underpowered studies, samples with suboptimal antiretroviral access, varying rate of virological control, high rate of AIDS, over-representation of non-community samples, and inclusion of inappropriate controls. The current study addresses these limitations, while also considering mental health and non-HIV comorbidity burden to determine whether PLHIV showed premature cognitive ageing compared with closely comparable HIV-negative controls.

METHODS

This study enrolled 254 PLHIV [92% on antiretroviral therapy; 84% with HIV RNA < 50 copies/mL; 15% with AIDS) and 72 HIV-negative gay and bisexual men [mean (SD) age = 49 (10.2) years] from a single primary care clinic in Sydney, Australia. Neurocognitive function was evaluated with the Cogstate Computerized Battery (CCB) at baseline and 6 months after. Linear mixed-effects (LME) models examined main and interaction effects of HIV status and chronological age on the CCB demographically uncorrected global neurocognitive z-score (GZS), adjusting for repeated testing, and then adjusting sequentially for HIV disease markers, mental health and comorbidities.

RESULTS

HIV status and age interacted with a lower GZS (β = -0.43, P < 0.05). Higher level of anxiety symptoms (β = -0.11, P < 0.01), historical AIDS (β = -0.12, P < 0.05) and historical HIV brain involvement (β = -0.12, P < 0.05) were associated with lower GZS.

CONCLUSIONS

We found a robust medium-sized premature ageing effect on cognition in a community sample with optimal HIV care. Our study supports routine screening of cognitive and mental health among PLHIV aged ≥ 50 years.

The impact of adjusting for pure predictors of exposure, mediator, and outcome on the variance of natural direct and indirect effect estimators

It is now well established that adjusting for pure predictors of the outcome, in addition to confounders, allows unbiased estimation of the total exposure effect on an outcome with generally reduced standard errors (SEs). However, no analogous results have been derived for mediation analysis. Considering the simplest linear regression setting and the ordinary least square estimator, we obtained theoretical results showing that adjusting for pure predictors of the outcome, in addition to confounders, allows unbiased estimation of the natural indirect effect (NIE) and the natural direct effect (NDE) on the difference scale with reduced SEs. Adjusting for pure predictors of the mediator increases the SE of the NDE's estimator, but may increase or decrease the variance of the NIE's estimator. Adjusting for pure predictors of the exposure increases the variance of estimators of the NIE and NDE. Simulation studies were used to confirm and extend these results to the case where the mediator or the outcome is binary. Additional simulations were conducted to explore scenarios featuring an exposure-mediator interaction as well as the relative risk and odds ratio scales for the case of binary mediator and outcome. Both a regression approach and an inverse probability weighting approach were considered in the simulation study. A real-data illustration employing data from the Canadian Study of Health and Aging is provided. This analysis is concerned with the mediating effect of vitamin D in the effect of physical activity on dementia and its results are overall consistent with the theoretical and empirical findings.

Education and Cognitive Functioning Across the Life Span

Cognitive abilities are important predictors of educational and occupational performance, socioeconomic attainment, health, and longevity. Declines in cognitive abilities are linked to impairments in older adults' everyday functions, but people differ from one another in their rates of cognitive decline over the course of adulthood and old age. Hence, identifying factors that protect against compromised late-life cognition is of great societal interest. The number of years of formal education completed by individuals is positively correlated with their cognitive function throughout adulthood and predicts lower risk of dementia late in life. These observations have led to the propositions that prolonging education might (a) affect cognitive ability and (b) attenuate aging-associated declines in cognition. We evaluate these propositions by reviewing the literature on educational attainment and cognitive aging, including recent analyses of data harmonized across multiple longitudinal cohort studies and related meta-analyses. In line with the first proposition, the evidence indicates that educational attainment has positive effects on cognitive function. We also find evidence that cognitive abilities are associated with selection into longer durations of education and that there are common factors (e.g., parental socioeconomic resources) that affect both educational attainment and cognitive development. There is likely reciprocal interplay among these factors, and among cognitive abilities, during development. Education-cognitive ability associations are apparent across the entire adult life span and across the full range of education levels, including (to some degree) tertiary education. However, contrary to the second proposition, we find that associations between education and aging-associated cognitive declines are negligible and that a threshold model of dementia can account for the association between educational attainment and late-life dementia risk. We conclude that educational attainment exerts its influences on late-life cognitive function primarily by contributing to individual differences in cognitive skills that emerge in early adulthood but persist into older age. We also note that the widespread absence of educational influences on rates of cognitive decline puts constraints on theoretical notions of cognitive aging, such as the concepts of cognitive reserve and brain maintenance. Improving the conditions that shape development during the first decades of life carries great potential for improving cognitive ability in early adulthood and for reducing public-health burdens related to cognitive aging and dementia.

Education and age-related decline in cognitive performance: Systematic review and meta-analysis of longitudinal cohort studies

Central theories of cognitive aging propose that education is an important protective factor for decline in cognitive performance in older age. We conducted a systematic review and meta-analysis of reported estimates of an association between educational attainment and change in performance in six cognitive domains (episodic memory, processing speed, verbal fluency, crystallized intelligence, fluid intelligence, and global ability) in the general population of older individuals. The systematic search (11^th of October 2019) identified 92 eligible articles. The episodic memory domain had the highest number of estimates (37 estimates from 18 articles, n = 109,281) included in the meta-analysis. The fewest estimates (6 estimates from 6 articles, n = 5263) were included for fluid intelligence. Pooled mean estimates from an inverse-variance weighted random effects analysis were not statistically significant and indicated that any association between education and change in cognitive performance is likely of a negligible magnitude. The estimates for education's role (one additional year) for change in cognitive performance ranged from -0.019 (95 % confidence interval, CI = -0.047, 0.010) to 0.004SD (CI = -0.003, 0.012) per decade. Even if the larger positive point estimates (i.e., protective effects) are selectively considered, the influence of education on change is still at least 12 times less important for the cognitive functioning of an older individual than the association between education and level of cognitive performance. Sensitivity analyses did not substantially alter these results. However, heterogeneity was substantial, and remained largely unexplained by mean age, mean educational attainment, Gini coefficient, GDP per capita, maximum follow-up period, and publication year. Overall, education is an important factor in aging due to its robust association with level of performance, but the current base of empirical evidence is not revealing a consistent and substantial association between educational attainment and changes in cognitive performance in the general population. Theories of cognitive aging must be updated to incorporate this pattern of findings.

Vitamin D status, cognitive decline and incident dementia: the Canadian Study of Health and Aging

OBJECTIVE

Vitamin D could prevent cognitive decline because of its neuroprotective, anti-inflammatory and antioxidant properties. This study aimed to evaluate the associations of plasma 25-hydroxyvitamin D (25(OH)D) concentrations with global cognitive function and incident dementia, including Alzheimer's disease (AD).

METHODS

The Canadian Study of Health and Aging is a 10-year cohort study of a representative sample of individuals aged 65 years or older. A total of 661 subjects initially without dementia with frozen blood samples and follow-up data were included. Global cognitive function was measured using the validated Modified Mini-Mental State (3MS) examination. A consensus diagnosis of all-cause dementia and AD was made between the physician and the neuropsychologist according to published criteria. Cognitive decline for a 5-year increase in age at specific 25(OH)D concentrations was obtained using linear mixed models with repeated measures. Hazard ratios of incident dementia and AD were obtained using semi-parametric proportional hazards models with age as time scale.

RESULTS

Over a mean follow-up of 5.4 years, 141 subjects developed dementia of which 100 were AD. Overall, no significant association was found between 25(OH)D and cognitive decline, dementia or AD. Higher 25(OH)D concentrations were associated with an increased risk of dementia and AD in women, but not in men.

CONCLUSION

This study does not support a protective effect of vitamin D status on cognitive function. Further research is needed to clarify the relation by sex.

Body Shape and Alzheimer's Disease: A Mendelian Randomization Analysis

Obesity has been reported to be related to memory impairment and decline in cognitive function, possibly further leading to the development of Alzheimer’s disease (AD). However, observational studies revealed both negative and positive associations between body shape (BS) and AD, thereby making it difficult to confirm causality due to residual confounds and reverse causation. Thus, using genome-wide association study summary data, two-sample Mendelian randomization (MR) analyses were applied to identify whether there exists a causal association between BS and AD. BS was measured using anthropometric traits (ATs) in this study, including body mass index (BMI), waist-to-hip ratio (WHR), waist-to-hip ratio adjusted by body mass index (WHRadjBMI), and waist circumference (WC). The associations of single nucleotide polymorphisms (SNP) with each AT and AD were obtained separately from aggregated data from the Genetic Investigation of Anthropometric Traits (GIANT) consortium and International Genomics of Alzheimer’s Project (IGAP) summary data (17,008 cases with AD and 37,154 controls). An inverse-variance weighted method was applied to obtain the overall causal estimate for multiple instrumental SNPs. The odds ratio (OR) [95% confidence interval (CI)] for AD risk per 1-SD difference in BMI was 1.04 (0.88, 1.23), in WHR was 1.01 (0.77, 1.33), in WHRadjBMI was 1.12 (0.89, 1.41), and in WC was 1.02 (0.82, 1.27). Furthermore, simulation analyses of survivor bias indicated the overall causal effect of BMI on risk of AD was not biased. In conclusion, the evidence from MR analyses showed no casual effect of BS on AD risk, which is inconsistent with the results from previous observational studies. The biological mechanism underlying the findings warrants further study.

The Effects of Reverse Causality and Selective Attrition on the Relationship Between Body Mass Index and Mortality in Postmenopausal Women

Concerns about reverse causality and selection bias complicate the interpretation of studies of body mass index (BMI, calculated as weight (kg)/height (m)2) and mortality in older adults. The objective of this study was to investigate methodological explanations for the apparent attenuation of obesity-related risks in older adults. We used data from 68,132 participants in the Women's Health Initiative (WHI) clinical trial for this analysis. All of the participants were postmenopausal women aged 50-79 years at baseline (1993-1998). To examine reverse causality and selective attrition, we compared rate ratios from inverse probability of treatment- and censoring-weighted Poisson marginal structural models with results from an unweighted adjusted Poisson regression model. The estimated mortality rate ratios and 95% confidence intervals for BMIs of 30.0-34.9, 35.0-39.9 and ≥40.0 were 0.86 (95% confidence interval (CI): 0.77, 0.96), 0.85 (95% CI: 0.72, 0.99), and 0.88 (95% CI: 0.72, 1.07), respectively, in the unweighted model. The corresponding mortality rate ratios were 0.96 (95% CI: 0.86, 1.07), 1.12 (95% CI: 0.97, 1.29), and 1.31 95% CI: (1.08, 1.57), respectively, in the marginal structural model. Results from the inverse probability of treatment- and censoring-weighted marginal structural model were attenuated in low BMI categories and increased in high BMI categories. The results demonstrate the importance of accounting for reverse causality and selective attrition in studies of older adults.

You Can't Drive a Car With Only Three Wheels

Authors aiming to estimate causal effects from observational data frequently discuss 3 fundamental identifiability assumptions for causal inference: exchangeability, consistency, and positivity. However, too often, studies fail to acknowledge the importance of measurement bias in causal inference. In the presence of measurement bias, the aforementioned identifiability conditions are not sufficient to estimate a causal effect. The most fundamental requirement for estimating a causal effect is knowing who is truly exposed and unexposed. In this issue of the Journal, Caniglia et al. (Am J Epidemiol. 2019;000(00):000-000) present a thorough discussion of methodological challenges when estimating causal effects in the context of research on distance to obstetrical care. Their article highlights empirical strategies for examining nonexchangeability due to unmeasured confounding and selection bias and potential violations of the consistency assumption. In addition to the important considerations outlined by Caniglia et al., authors interested in estimating causal effects from observational data should also consider implementing quantitative strategies to examine the impact of misclassification. The objective of this commentary is to emphasize that you can't drive a car with only three wheels, and you also cannot estimate a causal effect in the presence of exposure misclassification bias.

Investigating and Remediating Selection Bias in Geriatrics Research: The Selection Bias Toolkit

OBJECTIVES

Selection bias is a well-known concern in research on older adults. We discuss two common forms of selection bias in aging research: (1) survivor bias and (2) bias due to loss to follow-up. Our objective was to review these two forms of selection bias in geriatrics research. In clinical aging research, selection bias is a particular concern because all participants must have survived to old age, and be healthy enough, to take part in a research study in geriatrics.

DESIGN

We demonstrate the key issues related to selection bias using three case studies focused on obesity, a common clinical risk factor in older adults. We also created a Selection Bias Toolkit that includes strategies to prevent selection bias when designing a research study in older adults and analytic techniques that can be used to examine, and correct for, the influence of selection bias in geriatrics research.

RESULTS

Survivor bias and bias due to loss to follow-up can distort study results in geriatric populations. Key steps to avoid selection bias at the study design stage include creating causal diagrams, minimizing barriers to participation, and measuring variables that predict loss to follow-up. The Selection Bias Toolkit details several analytic strategies available to geriatrics researchers to examine and correct for selection bias (eg, regression modeling and sensitivity analysis).

CONCLUSION

The toolkit is designed to provide a broad overview of methods available to examine and correct for selection bias. It is specifically intended for use in the context of aging research. J Am Geriatr Soc 67:1970-1976, 2019.

Dementia and Cognitive Impairment Among U.S. Veterans With a History of MDD or PTSD: A Retrospective Cohort Study Based on Sex and Race

Objective: The aim of this study was to examine major depressive disorder (MDD) and posttraumatic stress disorder (PTSD) diagnosed at age < 55 as predictors, and sex and race as potential moderators, of dementia and other forms of cognitive impairment. Method: Veterans (N = 4,800) aged ⩾ 56 years were grouped by psychiatric history, sex, and race. Hierarchical and stepwise regression were employed to determine significant predictors. Results: MDD and PTSD were associated with almost double the risk for developing dementia or cognitive impairment at age ⩾ 56. Sex, as a moderator, had small effects whereas race increased the risk almost twofold for Black veterans, given the presence of MDD history. Discussion: MDD and PTSD act as significant risk factors for dementia and other forms of cognitive impairment, and Black veterans, given a history of MDD, may be at an increased risk. An important endeavor for future research is to examine how this risk may vary across dementia subtypes and related conditions.

An Investigation of Selection Bias in Estimating Racial Disparity in Stroke Risk Factors

Selection due to survival or attrition might bias estimates of racial disparities in health, but few studies quantify the likely magnitude of such bias. In a large national cohort with moderate loss to follow-up, we contrasted racial differences in 2 stroke risk factors, incident hypertension and incident left ventricular hypertrophy, estimated by complete-case analyses, inverse probability of attrition weighting, and the survivor average causal effect. We used data on 12,497 black and 17,660 white participants enrolled in the United States (2003-2007) and collected incident risk factor data approximately 10 years after baseline. At follow-up, 21.0% of white participants and 23.0% of black participants had died; additionally 22.0% of white participants and 28.4% of black participants had withdrawn. Individual probabilities of completing the follow-up visit were estimated using baseline demographic and health characteristics. Adjusted risk ratio estimates of racial disparities from complete-case analyses in both incident hypertension (1.11, 95% confidence interval: 1.02, 1.21) and incident left ventricular hypertrophy (1.02, 95% confidence interval: 0.84, 1.24) were virtually identical to estimates from inverse probability of attrition weighting and survivor average causal effect. Despite racial differences in mortality and attrition, we found little evidence of selection bias in the estimation of racial differences for these incident risk factors.

Sex Differences in the Genetic Architecture of Alzheimer's Disease

PURPOSE OF REVIEW

Summarize sex-specific contributors to the genetic architecture of Alzheimer's disease (AD).

RECENT FINDINGS

There are sex differences in the effects of Apolipoprotein E (APOE), genes along the APOE pathway, and genes along the neurotrophic signaling pathway in predicting AD. Reported sex differences are largely driven by stronger associations among females. Evidence also suggests that genetic predictors of amyloidosis are largely shared across sexes, while sex-specific genetic effects emerge downstream of amyloidosis and drive the clinical manifestation of AD.

SUMMARY

There is a lack of comprehensive assessments of sex differences in genome-wide analyses of AD and a need for more systematic reporting a sex-stratified genetic effects. The emerging emphasis on sex as a biological variable provides an opportunity for transdisciplinary collaborations aimed at addressing major analytical challenges that have hampered advancements in the field. Ultimately, sex-specific genetic association studies represent a logical first step towards precision medicine.

The puzzle of sex, gender and Alzheimer’s disease: Why are women more often affected than men?

Objective:

There are impressive differences in the incidence, prevalence and experience of women and men with Alzheimer’s Disease (AD). Notably, two-thirds of those with AD, the most common form of dementia, are women. Our objective was to provide a literature-based framework to understand these sex and gender differences in AD.

Methods:

We conducted a narrative review to examine sex and gender influences on AD.

Results:

We present a framework to understanding why these sex and gender differences exist in AD. This includes the influence of longevity (women live longer than men), biological differences (hormonal differences, epigenetics and frailty), differences in cognitive performance (women and men tend to perform differently on some cognitive tests), and gendered social roles and opportunities (educational and occupational opportunities, functional roles post-retirement). Our review clearly indicates the complex interaction of these sex and gender differences and variability within each.

Conclusions:

Given these important sex and gender differences in AD, we provide recommendations and steps forward describing how both sex and gender should be considered in dementia diagnosis and management and in the design and implementation of dementia research, including studies of caregiving interventions and models of dementia care.

Education and Cognitive Aging: Accounting for Selection and Confounding in Linkage of Data From the Danish Registry and Survey of Health, Ageing and Retirement in Europe

Earlier studies report inconsistent associations between education and cognitive aging. We assessed the association, accounting for selective dropout due to death or dementia, and, in a subsample, accounting for confounding by early-life intelligence. Data from the Danish component of the Survey of Health, Ageing and Retirement in Europe (n = 3,400) were linked to registry data (education records, dementia diagnoses, and mortality) and the Danish Conscription Database (youth intelligence measurements for 854 men). Word recall and verbal fluency were assessed up to 4 times over 10 years (2004-2013) and combined by averaging the z scores. We fitted a joint model linking a time-to-event model for dementia or death to a linear mixed-effects model for cognitive change. Rate of cognitive decline was slower among people with high education compared with low education (β = 0.112, 95% confidence interval (CI): 0.056, 0.170). Adjusting for youth intelligence did not attenuate the association between education and cognitive decline (crude β = 0.136, 95% CI: 0.028, 0.244 vs. adjusted β = 0.145, 95% CI: 0.022, 0.269). The results suggest that higher education may slow cognitive decline in later life. In this sample, results changed little when accounting for selective attrition and confounding by intelligence.

Early MoCA predicts long-term cognitive and functional outcome and mortality after stroke

OBJECTIVE

To examine whether the Montreal Cognitive Assessment (MoCA) administered within 7 days after stroke predicts long-term cognitive impairment, functional impairment, and mortality.

METHODS

MoCA was administered to 274 patients from 2 prospective hospital-based cohort studies in Germany (n = 125) and France (n = 149). Cognitive and functional outcomes were assessed at 6, 12, and 36 months after stroke by comprehensive neuropsychological testing, the Clinical Dementia Rating (CDR) scale, the modified Rankin Scale (mRS), and Instrumental Activities of Daily Living (IADL) and analyzed with generalized estimating equations. All-cause mortality was investigated by Cox proportional hazard models. Analyses were adjusted for demographic variables, education, vascular risk factors, premorbid cognitive status, and NIH Stroke Scale scores. The additive predictive value of MoCA was examined with receiver operating characteristic curves.

RESULTS

In pooled analyses, a baseline MoCA score <26 was associated with cognitive impairment, defined by neuropsychological testing (odds ratio [OR] 5.30, 95% confidence interval [CI] 2.75-10.22) and by CDR score ≥0.5 (OR 2.53, 95% CI 1.53-4.18); functional impairment, defined by mRS score >2 (OR 5.03, 95% CI 2.20-11.51) and by IADL score <8 (OR 2.48, 95% CI 1.40-4.38); and mortality (hazard ratio 7.24, 95% CI 1.99-26.35) across the 3-year follow-up. Patients with MoCA score <26 performed worse across all prespecified cognitive domains (executive function/attention, memory, language, visuospatial ability). MoCA increased the area under the curve for predicting cognitive impairment (neuropsychological testing 0.81 vs 0.72, p = 0.01) and functional impairment (mRS score >2, 0.88 vs 0.84, p = 0.047).

CONCLUSION

Early cognitive testing by MoCA predicts long-term cognitive outcome, functional outcome, and mortality after stroke. Our results support routine use of the MoCA in stroke patients.

Does selective survival before study enrolment attenuate estimated effects of education on rate of cognitive decline in older adults? A simulation approach for quantifying survival bias in life course epidemiology

Background

The relationship between education and late-life cognitive decline is controversial. Selective survival between early life, when education is typically completed, and late life, when cognitive ageing studies take place, could attenuate effect estimates.

Methods

We quantified potential survival bias (collider-stratification bias) in estimation of the effect of education on late-life cognitive decline by simulating hypothetical cohorts of 20-year-olds and applying cumulative mortality from US life tables. For each of four causal scenarios (2000 replications each), we compared the estimated versus causal effect of education on cognitive decline over 9 years, starting at age 60, 75 or 90 in random samples of n = 2000 people who survived to each age.

Results

Effects of education on cognitive decline were underestimated when both education and U, another determinant of cognitive decline, influenced mortality (collider-stratification bias). The magnitude of bias was sensitive to the magnitude of the effect of U on cognitive decline and whether there was a multiplicative interaction between education and U on mortality. For example, when there was a multiplicative interaction between education and U on mortality, 95% confidence interval coverage of the causal effect ranged from 83.4% to 50.4% at age 60 and 25.8% to 0.2% at age 90.

Conclusions

Selective survival could lead to underestimation of effects of education on late-life cognitive decline. Our simulations map survival bias to testable assumptions about underlying causal structures.

The curse of the perinatal epidemiologist: inferring causation amidst selection

Purpose of review

Human reproduction is a common process and one that unfolds over a relatively short time, but pregnancy and birth processes are challenging to study. Selection occurs at every step of this process (e.g., infertility, early pregnancy loss, and stillbirth), adding substantial bias to estimated exposure-outcome associations. Here we focus on selection in perinatal epidemiology, specifically, how it affects research question formulation, feasible study designs, and interpretation of results.

Recent findings

Approaches have recently been proposed to address selection issues in perinatal epidemiology. One such approach is the ongoing pregnancies denominator for gestation-stratified analyses of infant outcomes. Similarly, bias resulting from left truncation has recently been termed "live birth bias," and a proposed solution is to control for common causes of selection variables (e.g., fecundity, fetal loss) and birth outcomes. However, these approaches have theoretical shortcomings, conflicting with the foundational epidemiologic concept of populations at risk for a given outcome.

Summary

We engage with epidemiologic theory and employ thought experiments to demonstrate the problems of using denominators that include units not "at risk" of the outcome. Fundamental (and commonsense) concerns of outcome definition and analysis (e.g., ensuring that all study participants are at risk for the outcome) should take precedence in formulating questions and analysis approach, as should choosing questions that stakeholders care about. Selection and resulting biases in human reproductive processes complicate estimation of unbiased exposure- outcome associations, but we should not focus solely (or even mostly) on minimizing such biases.

Survival-related Selection Bias in Studies of Racial Health Disparities: The Importance of the Target Population and Study Design

The impact of survival-related selection bias has not always been discussed in relevant studies of racial health disparities. Moreover, the analytic approaches most frequently employed in the epidemiologic literature to minimize selection bias are difficult to implement appropriately in racial disparities research. This difficulty stems from the fact that frequently employed analytic techniques require that common causes of survival and the outcome are accurately measured. Unfortunately, such common causes are often unmeasured or poorly measured in racial health disparities studies. In the absence of accurate measures of the aforementioned common causes, redefining the target population or changing the study design represents a useful approach for reducing the extent of survival-related selection bias. To help researchers recognize and minimize survival-related selection bias in racial health disparities studies, we illustrate the aforementioned selection bias and how redefining the target population or changing the study design can be useful.

Can Survival Bias Explain the Age Attenuation of Racial Inequalities in Stroke Incidence?: A Simulation Study

BACKGROUND

In middle age, stroke incidence is higher among black than white Americans. For unknown reasons, this inequality decreases and reverses with age. We conducted simulations to evaluate whether selective survival could account for observed age patterning of black-white stroke inequalities.

METHODS

We simulated birth cohorts of 20,000 blacks and 20,000 whites with survival distributions based on US life tables for the 1919-1921 birth cohort. We generated stroke incidence rates for ages 45-94 years using Reasons for Geographic and Racial Disparities in Stroke (REGARDS) study rates for whites and setting the effect of black race on stroke to incidence rate difference (IRD) = 20/10,000 person-years at all ages, the inequality observed at younger ages in REGARDS. We compared observed age-specific stroke incidence across scenarios, varying effects of U, representing unobserved factors influencing mortality and stroke risk.

RESULTS

Despite a constant adverse effect of black race on stroke risk, the observed black-white inequality in stroke incidence attenuated at older age. When the hazard ratio for U on stroke was 1.5 for both blacks and whites, but U only directly influenced mortality for blacks (hazard ratio for U on mortality =1.5 for blacks; 1.0 for whites), stroke incidence rates in late life were lower among blacks (average observed IRD = -43/10,000 person-years at ages 85-94 years versus causal IRD = 20/10,000 person-years) and mirrored patterns observed in REGARDS.

CONCLUSIONS

A relatively moderate unmeasured common cause of stroke and survival could fully account for observed age attenuation of racial inequalities in stroke.

Causal inference in studies of preterm babies: a simulation study

OBJECTIVE

Using a simple simulation, we illustrate why associations estimated from studies restricted to preterm births cannot be interpreted causally.

DESIGN, SETTING AND POPULATION

Data simulation involving a hypothetical cohort of fetuses who may be healthy or have one or more of four pathological factors (termed A through D, increasing in severity) with known effects on gestational length and risk of mortality. We focus on babies born at ≤32 weeks of gestation.

METHODS

We visually represent the simulated population and compare the association between A (which may represent pre-eclampsia) and neonatal death. We then repeat the exercise with D (standing in for chorioamnionitis) as the exposure of interest.

MAIN OUTCOME MEASURES

Odds ratios of neonatal death in the simulated data.

RESULTS

In most weeks, and for both A and D, the calculated odds ratios are substantially biased and underestimate the true risk of neonatal death associated with each pathology. For example, factor A has a true causal odds ratio of 1.50, yet it appears protective among births ≤32 weeks (estimated crude odds ratio 0.39; gestational age-adjusted odds ratio 0.71).

CONCLUSIONS

Among very preterm births, virtually all babies are born with pathologies that increase the risk of adverse outcomes. Hence, babies exposed to one factor (e.g. pre-eclampsia) are compared with babies who have a mix of other pathologies. Such selection bias affects studies carried out among very preterm births (e.g. where pre-eclampsia appears to reduce risk of adverse neonatal outcomes).

TWEETABLE ABSTRACT

Selection bias affects studies of preterm births, complicating interpretation.

Collider scope: when selection bias can substantially influence observed associations

Large-scale cross-sectional and cohort studies have transformed our understanding of the genetic and environmental determinants of health outcomes. However, the representativeness of these samples may be limited-either through selection into studies, or by attrition from studies over time. Here we explore the potential impact of this selection bias on results obtained from these studies, from the perspective that this amounts to conditioning on a collider (i.e. a form of collider bias). Whereas it is acknowledged that selection bias will have a strong effect on representativeness and prevalence estimates, it is often assumed that it should not have a strong impact on estimates of associations. We argue that because selection can induce collider bias (which occurs when two variables independently influence a third variable, and that third variable is conditioned upon), selection can lead to substantially biased estimates of associations. In particular, selection related to phenotypes can bias associations with genetic variants associated with those phenotypes. In simulations, we show that even modest influences on selection into, or attrition from, a study can generate biased and potentially misleading estimates of both phenotypic and genotypic associations. Our results highlight the value of knowing which population your study sample is representative of. If the factors influencing selection and attrition are known, they can be adjusted for. For example, having DNA available on most participants in a birth cohort study offers the possibility of investigating the extent to which polygenic scores predict subsequent participation, which in turn would enable sensitivity analyses of the extent to which bias might distort estimates.

Joint mixed-effects models for causal inference with longitudinal data

Causal inference with observational longitudinal data and time-varying exposures is complicated due to the potential for time-dependent confounding and unmeasured confounding. Most causal inference methods that handle time-dependent confounding rely on either the assumption of no unmeasured confounders or the availability of an unconfounded variable that is associated with the exposure (eg, an instrumental variable). Furthermore, when data are incomplete, validity of many methods often depends on the assumption of missing at random. We propose an approach that combines a parametric joint mixed-effects model for the study outcome and the exposure with g-computation to identify and estimate causal effects in the presence of time-dependent confounding and unmeasured confounding. G-computation can estimate participant-specific or population-average causal effects using parameters of the joint model. The joint model is a type of shared parameter model where the outcome and exposure-selection models share common random effect(s). We also extend the joint model to handle missing data and truncation by death when missingness is possibly not at random. We evaluate the performance of the proposed method using simulation studies and compare the method to both linear mixed- and fixed-effects models combined with g-computation as well as to targeted maximum likelihood estimation. We apply the method to an epidemiologic study of vitamin D and depressive symptoms in older adults and include code using SAS PROC NLMIXED software to enhance the accessibility of the method to applied researchers.

Compartmental Model Diagrams as Causal Representations in Relation to DAGs

Compartmental model diagrams have been used for nearly a century to depict causal relationships in infectious disease epidemiology. Causal directed acyclic graphs (DAGs) have been used more broadly in epidemiology since the 1990s to guide analyses of a variety of public health problems. Using an example from chronic disease epidemiology, the effect of type 2 diabetes on dementia incidence, we illustrate how compartmental model diagrams can represent the same concepts as causal DAGs, including causation, mediation, confounding, and collider bias. We show how to use compartmental model diagrams to explicitly depict interaction and feedback cycles. While DAGs imply a set of conditional independencies, they do not define conditional distributions parametrically. Compartmental model diagrams parametrically (or semiparametrically) describe state changes based on known biological processes or mechanisms. Compartmental model diagrams are part of a long-term tradition of causal thinking in epidemiology and can parametrically express the same concepts as DAGs, as well as explicitly depict feedback cycles and interactions. As causal inference efforts in epidemiology increasingly draw on simulations and quantitative sensitivity analyses, compartmental model diagrams may be of use to a wider audience. Recognizing simple links between these two common approaches to representing causal processes may facilitate communication between researchers from different traditions.

Association of midlife lipids with 20-year cognitive change: A cohort study

INTRODUCTION

Existing studies predominantly consider the association of late-life lipid levels and subsequent cognitive change. However, midlife rather than late-life risk factors are often most relevant to cognitive health.

METHODS

We quantified the association between measured serum lipids in midlife and subsequent 20-year change in performance on three cognitive tests in 13,997 participants of the Atherosclerosis Risk in Communities study.

RESULTS

Elevated total cholesterol, low-density lipoprotein cholesterol, and triglycerides were associated with greater 20-year decline on a test of executive function, sustained attention, and processing speed. Higher total cholesterol and triglycerides were also associated with greater 20-year decline in memory scores and a measure summarizing performance on all three tests. High-density lipoprotein cholesterol was not associated with cognitive change. Results were materially unchanged in sensitivity analyses addressing informative missingness.

DISCUSSION

Elevated total cholesterol, low-density lipoprotein cholesterol, and triglycerides in midlife were associated with greater 20-year cognitive decline.

Alzheimer's disease genetic risk variants beyond APOE ε4 predict mortality

INTRODUCTION

We hypothesized that, like apolipoprotein E (APOE), other late-onset Alzheimer's disease (LOAD) genetic susceptibility loci predict mortality.

METHODS

We used a weighted genetic risk score (GRS) from 21 non-APOE LOAD risk variants to predict survival in the Adult Changes in Thought and the Health and Retirement Studies. We meta-analyzed hazard ratios and examined models adjusted for cognitive performance or limited to participants with dementia. For replication, we assessed the GRS-longevity association in the Cohorts for Heart and Aging Research in Genomic Epidemiology, comparing cases surviving to age ≥90 years with controls who died between ages 55 and 80 years.

RESULTS

Higher GRS predicted mortality (hazard ratio = 1.05; 95% confidence interval: 1.00-1.10, P = .04). After adjusting for cognitive performance or restricting to participants with dementia, the relationship was attenuated and no longer significant. In case-control analysis, the GRS was associated with reduced longevity (odds ratio = 0.64; 95% confidence interval: 0.41-1.00, P = .05).

DISCUSSION

Non-APOE LOAD susceptibility loci confer risk for mortality, likely through effects on dementia incidence.