Faculty Bibliography

Background: Up to one third of older adults with diabetes (DM) have co-occurring cognitive impairment and/or Alzheimer's disease and Related Dementias (ADRD). These patients are more likely to experience episodes of hypo and hyperglycemia. The American Geriatric Society (AGS) and American Diabetic Association (ADA) recommend liberalizing hemoglobin (Hb) A1c targets for patients with multiple comorbidities, but the impact of ADRD on glycemic management of patients with DM-ADRD is unknown.
Method(s): Within the primary care and endocrine clinics in the NYU Langone Health System, we collected characteristics of DM-ADRD patients participating in a DM-ADRD clinical quality improvement program. We administratively collected patients' most recent (within18 months) HbA1c from the Electronic Medical Record. We also surveyed the English and Spanish-speaking caregivers (CG) of these DM-ADRD patients. The CG survey included a measure of CG-reported patient dementia severity using the Dementia Severity Rating Scale (DSRS). We examined the relationship between the DSRS score and HbA1c.
Result(s): Patients (n=173) had a mean age of 79.7 (+/-7.18) and a mean HbA1c of 7.08%. 63% (n=106) were female, 63% (n=106) white; 37% (n=64) identified as being Latino/Hispanic. The mean DSRS score was 25 (+/-12.7) (range: 0-54), within the range of moderate cognitive impairment (18-36). Those older than 75 and those who were Spanish speaking had higher DSRS scores (26.1, p=.02; and 26.7, p=.04, respectively). Mean HbA1c of patients in the severe DSRS range (scores 37-54) was 6.81 (N=35) and was lower than in patients with moderate and mild dementia severity (mean 7.15 and 7.24, respectively); however, this difference was not statistically significant.
Conclusion(s): While the data does not confirm a statistically significant relationship between dementia severity and lower A1c, this finding is worrisome for DM-ADRD patients. Our data suggests possible overtreatment and if confirmed, there is a clear need for increased family and provider education and quality improvement programs for this vulnerable population

Background: Caregivers (CGs) of older-adults with Alzheimer's disease and related dementias (ADRD) and CGs of older-adults with diabetes (DM) report substantial CG burden. CG burden is known to be linked to patients' behavioral problems, poor cognition, and increased dependency. There is no literature addressing CG burden in CGs of individuals with co-occurring diabetes and dementia (DM-ADRD). The aim of this study was to identify CG and care-recipient (CR) factors associated with high levels of CG burden in CGs of DM-ADRD patients.
Method(s): This study used bivariate and descriptive statistics to analyze surveys collected as part of a quality improvement intervention being conducted at NYU Langone Health primary care and endocrine Faculty Group Practices and Family Health Centers. Inclusion criteria for patients were age >= 65, cognitive impairment, and DM with recent HbA1c > 6.4 or ever prescribed hyperglyemic medication. Telephonic surveys were conducted with CGs of eligible patients. The Treatment Burden Questionnaire (TBQ) was used to measure CG burden. TBQ results were analyzed for association with CG factors including age, sex, race, relationship to patient, education level, residence status, and level of social support, as well as CR factors including age, sex, race, dementia severity, Charlson comorbidity score, and recent HbA1c values.
Result(s): CGs that completed surveys (n=58) had a mean age of 54.3 years, 74% (n=43) female, 46% (n=27) white, 84% (n=49) were children of CRs, 70% (n=41) had education beyond 12th grade, and 55% (n=32) lived separately from CR. CRs of CGs that completed surveys (n=58) had a mean age of 80.5 years, 67% (n=39) female, 67% (n=37) white. We found CGs who were male, Asian, co-resident, with low level of social support, of CRs with more-advanced dementia, and of CRs with recent out-of-range HbA1c had significantly higher levels of CG burden (p<0.1).
Conclusion(s): Our study demonstrates there are several CG and CR factors that are associated with increased levels of CG burden in this population. Findings may assist in identification of CGs at risk for increased burden. If these results are found to be replicable, future studies should focus on the development of prevention and treatment plans consistent with these findings

Background We developed a risk model to predict hospitalization for bleeding within 6 months of discharge in older adults hospitalized for acute MI (AMI) and discharged on dual antiplatelet therapy (DAPT). Methods SILVER-AMI is a cohort study of 3041 patients age >=75 hospitalized with AMI at 96 U.S. hospitals. Participants underwent in-hospital functional assessment (cognition, vision, hearing, unintentional weight loss, ADLs, grip strength, functional mobility, falls). These analyses focused on participants discharged on DAPT (N=1858). Our outcome was rehospitalization for bleeding within 6 months. We used Bayesian model averaging to develop a risk model with split sample validation. Results Mean age was 81.5 years. Compared with participants not prescribed DAPT, those prescribed DAPT had slightly better functional mobility and lower cognitive impairment. Overall, 150 (8.1%) participants on DAPT experienced hospitalization for bleeding within 6 months; nearly half (48.7%) were gastrointestinal. Rates of functional impairments were similar among participants who did and did not experience bleeding. The final risk model included 8 predictors (Table), had moderate discrimination (C-statistic = 0.66), and good calibration (Hosmer-Lemeshow P value > 0.05). Conclusion Hospitalization for bleeding within 6 months of discharge on DAPT among older AMI patients was not predicted by aging-related functional impairments, but 8 other clinically plausible predictors were identified. [Figure presented]
Copyright

Background We sought to examine whether people with a diagnosis of cardiovascular disease (CVD) experienced a greater incidence of subsequent cognitive impairment (CI) compared to people without CVD, as suggested by prior studies, using a large longitudinal cohort. Methods We used biennial data collected on adults age >=50 from the Health and Retirement Study (HRS) to compare the incidence of CI over 8 years in 1,931 participants newly diagnosed with CVD vs. 3,862 age- and gender-matched controls. Diagnosis of CVD was adjudicated with an established HRS methodology. CI was defined as <=11 on the 27-point Telephone Interview for Cognitive Status, based on a previously accepted clinical cutpoint. To examine the incidence of CI, we used a cumulative incidence function accounting for competing risk of death. Results Mean age at study entry was 70 years, and 55% were female. CI developed in 1,335 participants over 8 years. Death was more common among participants with incident CVD (20.4% vs. 13.4%, p <.001). Cumulative incidence analysis for CI, after adjusting for death, showed no significant difference in incidence of cognitive impairment between the CVD and control groups at the end of the study period (Figure). Conclusion We found no increased risk of subsequent cognitive impairment among participants with CVD (compared with no CVD), despite previous research indicating that CVD accelerates cognitive decline. This finding may be due to appropriately accounting for the competing risk of death. [Figure presented]
Copyright