Oshadi Jayakody

OBJECTIVE: Adults with type 1 diabetes experience short-term changes in cognitive performance after glycemic events; however, the duration of such cognitive effects remains unclear. We examined whether hypoglycemic and hyperglycemic events and rapid glucose changes were associated with changes in processing speed or sustained attention in adults with type 1 diabetes and whether these associations varied by time since the event. RESEARCH DESIGN AND METHODS: Participants (N = 190) completed 15 days of smartphone-based ecologic momentary assessment (EMA; 3 prompts/day) paired with continuous glucose monitoring. Cognitive performance (median reaction time and accuracy) was tested using digit symbol matching (DSM) and gradual-onset continuous performance testing (GCPT; sustained attention) by TestMyBrain (processing speed). Glycemic exposures were defined as hypoglycemia or hyperglycemia lasting ≥15 min and rapid (5-min) increases or decreases in glucose >15 mg/dL, occurring during nonoverlapping intervals: 0–1, 1–2, and 2–3 h before EMA. Multilevel models estimated associations between glycemic exposures and cognitive performance, adjusting for age, EMA count, and interruptions. RESULTS: Hypoglycemic events occurring within 0–1 h of EMA were associated with slower reaction time and poorer performance on DSM and GCPT. No associations were observed for events occurring 1–2 or 2–3 h before EMA. Hyperglycemic events and rapid glucose increases or decreases were not associated with cognition across most outcomes or time windows. CONCLUSIONS: Hypoglycemia was associated with a transient period of cognitive vulnerability largely confined to the hour after an event, with little evidence of delayed effects. Findings support avoiding or exercising caution during cognitively demanding tasks for up to an hour after a hypoglycemic event. ARTICLE HIGHLIGHTS: Why did we undertake this study?: Short-term cognitive fluctuations are common among adults with type 1 diabetes; however, whether they reflect immediate or lingering effects of prior glycemic events remains unclear. What is the specific question(s) we wanted to answer?: We examined whether glycemic events and rapid glucose changes were associated with processing speed or sustained attention and whether the associations varied by time since the event. What did we find?: Hypoglycemic events occurring within the preceding hour of assessment were associated with slower reaction time and poorer cognitive performance. No associations were observed for events occurring 1–3 h before assessment. What are the implications of our findings?: Cognitive vulnerability after hypoglycemic events seems transient and limited largely to the first hour after an event, supporting caution when engaging in cognitively demanding tasks during that time.

Background: Apathy is an increasingly recognized predictor of cognitive decline. Although both poor glycemic status and apathy have been independently linked to cognitive decline, whether apathy mediates the relationship between poor glycemic status and cognitive decline remains unclear. Methods: We used data from 4571 adults aged ≥60 years without baseline memory problems from the U.S. Health and Retirement Study. Glycemic status was measured using glycosylated hemoglobin (HbA1c). Apathy-like symptoms were derived from CES-D items, reflecting diminished positive affect and motivation. Outcomes included incident cognitive impairment and incident self-reported Alzheimer’s disease and related dementias (ADRD). Accelerated Failure Time models, adjusted for age, sex, and education, estimated associations between HbA1c and time to each outcome. Mediation was tested using a product-of-coefficients approach that incorporated survey weights. Results: Over a mean 8.89 (SD 4.7) years, 938 participants developed cognitive impairment. Higher HbA1c was associated with shorter time to cognitive impairment (TR = 0.85; 95% CI 0.79–0.91). After adjustment for apathy-like symptoms, the association was attenuated (TR = 0.87; 95% CI 0.81–0.92), indicating approximately 11% mediation. Over 9.2 (SD 4.6) years, 583 participants reported incident ADRD. Higher HbA1c was associated with shorter time to ADRD (TR= 0.91; 95% CI 0.85–0.97) that was attenuated (TR = 0.93; 95% CI 0.86–0.99), suggesting 20% mediation through apathy-like symptoms. However, the mediation effects attenuated after further adjustment for baseline cognition and vascular/lifestyle factors. Conclusions: Overall, poorer glycemic status was associated with accelerated onset of adverse cognitive events, with apathy-like symptoms as a potential mediator requiring confirmation in future studies.

Apathy is an increasingly recognized neuropsychiatric syndrome and predictor of cognitive decline, distinct from depression. Although type 2 diabetes mellitus is a well-established risk factor for cognitive impairment, longitudinal evidence examining whether apathy links diabetes risk to adverse cognitive outcomes remains limited. We used data from 4,571 U.S. adults aged ≥60 years without baseline memory problems enrolled in the U.S.Health and Retirement Study. Diabetes risk was measured using glycosylated hemoglobin (HbA1c), treated continuously. Apathy was derived from four CES-D items reflecting diminished positive affect and motivation. Outcomes included incident self-reported Alzheimer’s disease and related dementias (ADRD) and incident cognitive impairment. Accelerated time-to-failure Weibull models were used to estimate associations between HbA1c and time to each outcome. Mediation was tested using a product-of-coefficients approach incorporating survey weights. Higher HbA1c was associated with shorter time to ADRD (coefficient −0.09; 95% CI −0.16, −0.02) and cognitive impairment (−0.16; 95% CI −0.23, −0.09), as well as greater apathy (β = 0.01; 95% CI 0.007, 0.03). After including apathy in Weibull models, associations with ADRD (−0.07; 95% CI −0.14, −0.006; p = 0.032) and cognitive impairment (−0.14; 95% CI −0.21, −0.07; p < 0.001) were changed but remained significant. Indirect effects through apathy were statistically significant for both outcomes, indicating partial mediation. Overall, elevated diabetes risk was associated with accelerated onset of ADRD and cognitive impairment, with apathy partially mediating these relationships, highlighting apathy as a potential target for behavioral interventions in individuals with diabetes.

Background: Recruitment challenges in ADRD clinical trials, especially when enrolling underrepresented populations, lead to significant delays, increased costs and reduced scientific merit. We aimed to examine participant‐ and study‐related barriers to recruitment and identify modifiable barriers that can be addressed in future efforts. Methods: We analyzed screen failure data from two ADRD clinical trials at the Department of Medicine, Albert Einstein College that recruited older adults from diverse populations. Two independent, blinded raters systematically reviewed screen failure cases to categorize reasons for non‐enrollment. Participant‐related barriers included lack of interest, scheduling conflicts (e.g., time/travel/weather constraints), family‐related issues, dislike of certain aspects of the study, mistrust and illness. Study‐related factors included exclusion due to eligibility criteria, study burden, caregiver‐related issues and concurrent enrollment in other studies. Disagreement between raters were resolved through discussion with a third team member. Chi‐square and Fisher's exact tests were used to determine whether participant characteristics (sex, education, ethnicity) influenced the likelihood of screen failure. Results: Data from 311 participants (mean age 78.2 (SD); 74.9% women) were analyzed. Table 1 reports the number of participants associated with each screen failure reason. Sixty two percent of screen failures were due to participant‐related factors including; no interest (40%), conflicting schedules (32%), illness (27%) and dislike of certain aspects of the study (24%). Study‐related reasons included ineligibility (57%), study burden (19%) and concurrent enrollment (14%). Notably, women reported significantly more mistrust. Among those who were screen‐failed due to eligibility criteria, 68.7% were minority older adults (Hispanic White/Black, Non‐Hispanic Black, and Asian), while 71% of those with scheduling conflicts were White older adults. Conclusions: Our preliminary findings highlight several modifiable participant‐ and study‐related barriers to target in future recruitment efforts. Participant‐related strategies could include increasing awareness of clinical trials, fostering partnerships with older adults through community programming and identifying trusted peer‐messengers. Study‐related strategies might involve carefully reconsidering eligibility criteria that may disproportionately impact diverse groups and considering additional incentives to compensate for participants’/caregivers’ time. Examining the intersectionality of social identities in relation to screen failures can provide further insights into how to improve recruitment efforts and design more inclusive ADRD clinical trials.

Background: Alzheimer's disease and related dementias (ADRD) develop over 20 years with subtle impairments in Instrumental Activities of Daily Living (IADLs). This study aimed to determine whether specific IADLs are associated with risks for cognitive impairment in pre‐ADRD stages. Methods: We used Gateway to Global Aging Data (g2aging.org) from the US Health and Retirement Study. In adults ≥65 years without an ADRD diagnosis ( n = 10, 346), IADL impairments were self‐reported as difficulty in managing money, medication, reading maps, using telephone, shopping for groceries and preparing meals. Memory‐related cognitive impairment was defined as ≥1.5 SD below the study‐specific mean for total recall combined with poor self‐reported memory. Non‐memory‐related cognitive impairment was ≥1.5 SD below the mean in ≥2 cognitive tests including orientation, serial 7s and backward counting. Latent class analysis was performed to identify different groups of IADLs at baseline. Cox proportional hazards models adjusted for age, sex, education, were used to determine the associations between group membership in IADLs and the risk for cognitive impairment. Models were adjusted for post‐stratification sample weights. Results: Weighted mean age of participants was 74.3 years (SD 6.8), and 58% were women. Three IADL groups were identified: A high functioning group (86%), with low probabilities of difficulty in any IADLs ; a group with moderate difficulty in specific tasks such as managing money, shopping, reading maps and making meals (10%) ; and a low functioning group (4%), with high difficulty across all IADLs. The low functioning group showed increased risk (hazard ratio [HR] 2‐2.5; p p = 0.001) and non‐memory‐related impairment (HR 1.6, p <0.001) compared to the high functioning group (Table 1). Conclusions: Our findings suggest that even moderate difficulty in performing specific IADLs (i.e., managing money) is associated with an increased risk for cognitive impairment in older adults. These IADLs may serve as early risk indicators and have the potential to be developed into clinical trial endpoints in pre‐clinical ADRD populations.

Background: Slow gait is a robust functional marker for dementia that identifies individuals’ risk for cognitive impairment up to a decade before a clinical onset. But, what about the other end of the spectrum? This abstract presents the results on whether “super movers” (a novel phenotype of individuals aged ≥80 years without dementia and with gait speeds ≥1.5 SD of sex‐adjusted cutoffs) have lower risks of cognitive impairment and/or slower cognitive decline. Methods: Data from five Health and Retirement Study International Network of Studies (HRS‐INS) were used to assess the risk of cognitive impairment in super‐versus non‐super movers. Cognitive impairment was defined as >1.5 SD below age‐adjusted cutoffs on at least one out of three tests, combined with impaired Instrumental Activities of Daily Living (Table 1). Age‐ and sex‐adjusted hazard ratios (HR) from cox proportional hazards models were pooled in a meta‐analysis to determine the overall risk for incident cognitive impairment. Data from the LonGenity study (a prospective longitudinal study in the Bronx) were used to examine the rates of decline in cognitive tests in super‐ versus non‐super movers using linear mixed effect models adjusted for age, sex, education and parental longevity,. Results: In individual HRS‐INS, super movers showed lower risks of cognitive impairment in the ELSA (English Longitudinal Ageing Study; HR 0.31, 95% CI 0.16‐0.60)) and SHARE studies (the Survey of Health, Ageing and Retirement in Europe HR 0.49, 95% CI 0.26‐0.89)). In the meta‐analysis, among 3,999 participants selected from 15,911=aged ≥ 65 ( n = 358 super movers; baseline age 83.6‐84.4 years) followed for 3.8‐6.1 years, super movers had significantly lower risk of cognitive impairment (HR 0.49, 95% CI 0.28‐0.71) (Figure 1). In the LonGenity study, ( n = 197, baseline age 84.6 years (SD 3.3)) super movers showed a slower decline in global cognition, total recall, digit symbol substitution test and Trial Making Tests A and B (Table 2). Conclusions: Super movers—a novel aging phenotype—show lower risks of cognitive impairment and slower cognitive decline, in memory and non‐memory related tests. Investigating the behavioral and biological factors contributing to this phenotype may reveal protective mechanisms against cognitive decline and dementia risk.

This pilot study offers preliminary evidence that a virtual meal-preparation task is feasible for older adults and highlights that the community engagement studios are an effective approach to generate community-informed strategies to enhance intervention designs and reach.

Background: Although gait speed typically declines with age, some individuals manage to maintain higher walking speeds well into older age. To better understand healthy longevity, we propose a novel exceptional aging phenotype of super movers, individuals 80 years or older with walking speeds comparable to individuals 3 decades younger. Methods: We conducted a retrospective cohort study to examine the epidemiology and determinants of super mover status as well as mortality, biological aging and gait speed trajectories of super movers compared to age peers identified from 15 aging cohort studies across 24 countries. Results: In a pooled cohort of data from multicountry population-based surveys, prevalence of super movers ranged from 4.9% to 11.1% in individuals 80 and older, with an average overall prevalence of 7.3%. Compared to age peers, super movers exhibited lower disease prevalence, healthier lifestyles and had lower mortality rates (hazard ratios: 0.38, 95% CI: 0.09–0.66). In 2 smaller samples with gait and biological aging data and longitudinal gait assessments, compared to age-matched peers, super movers had younger biological age and showed faster gait speed from their seventh to tenth decades, respectively. Conclusions: Super movers are an exceptional aging phenotype enriched with unique biological and clinical profiles that can provide valuable insights into healthy longevity.

UNSTRUCTURED This pilot study offers preliminary evidence that a virtual meal-preparation task is feasible for older adults and highlights that the community engagement studios are an effective approach to generate community-informed strategies to enhance intervention designs and reach.