Objective Falls can be deadly for older adults, making accurate documentation essential for prevention and quality improvement. The Minimum Data Set (MDS) is used in nursing homes to document falls; however, it fails to capture fall dates and has been critiqued for underreporting and misclassification. We developed a natural language processing (NLP) algorithm to extract fall dates from electronic health records (EHRs). This study outlines the NLP algorithm development and compares its findings with the MDS to evaluate misclassification and underreporting. Design Retrospective cohort study. Setting and Participants All veterans residing in long-term care between January 7, 2012 and December 31, 2024. Methods A rule-based NLP algorithm was developed to identify fall dates in the EHR and validated with manual chart review. The NLP results (reference) were compared with the MDS v3.0 assessment to assess the concordance between the 2 data sources. Results We identified 38,852 long-term care residents with a median length of stay of 204 days. About half (49.0%) of residents experienced at least 1 fall during their stay, as identified by either data source. The NLP algorithm achieved 96% accuracy against manual chart review. Using the NLP output as the reference standard, the MDS correctly identified 86.1% of residents with a fall in the EHR (F1-score = 0.885, and specificity = 0.930). However, 2430 (6.3%) residents had falls documented in their medical records that were never reported in the MDS, and 3.9% of MDS falls were not identified in the EHR (misclassification). The median time between a fall and an MDS assessment documenting a fall was 17 days. Conclusions and Implications The MDS identified 86% of NH residents with EHR documentation of a fall, but it did not provide the specific date of each fall. By supplementing MDS fall reporting with NLP, we may improve the sensitivity of fall detection and gain more precise information regarding when a fall occurs.

OBJECTIVES:We previously developed a multi-outcome prognostic model for older adults admitted to skilled nursing facilities (SNFs) for short-term rehab using Medicare data. However, incorporating predictors from the Minimum Data Set (MDS), a mandated comprehensive assessment, may improve model performance. This study sought to develop an updated model with MDS elements for use on day 7 of SNF admission when clinical trajectories are more established. DESIGN:Retrospective cohort study. SETTING AND PARTICIPANTS:Twenty percent national sample of community-dwelling Medicare Fee-for-Service beneficiaries aged ≥66 admitted to an SNF for at least 7 days following a hospitalization between 2017 and 2019. METHODS:We predicted 2 outcomes: 6-month mortality and successful community discharge (community discharge without rehospitalization or death in the subsequent 30 days). For model development, we started with predictors from our published Medicare-based model (age, sex, Medicaid status, discharge diagnosis, hospital length of stay, admission type, comorbidities, prior hospitalizations), used Least Absolute Shrinkage and Selection Operator (LASSO) on MDS elements for variable selection, and performed logistic regression to determine predictor coefficients. Model performance was assessed by concordance statistics (c-statistics), calibration plots, and decision curve analysis. RESULTS:The cohort included 426,680 individuals [mean age 81.3 years (SD = 8.3), 62.7% female, 7.9% Black]. Overall, 19.9% died within 6 months, and 57.6% experienced a successful community discharge. The updated MDS model, which included Medicare predictors and 6 MDS items (activities of daily living score, cognitive status, urinary incontinence, bowel incontinence, oxygen use, walking balance), showed improvements over the Medicare model in discrimination [bootstrapped optimism-corrected c-statistic of 0.789 (95% CI, 0.787-0.790) vs 0.747 (95% CI, 0.745-0.749) for 6-month mortality and 0.730 (95% CI, 0.728-0.731) vs 0.685 (95% CI, 0.683-0.687) for successful community discharge, respectively], net benefit, and fraction of new information. Models showed good calibration. CONCLUSIONS AND IMPLICATIONS:Incorporating MDS data from the first 7 days of SNF admission improved the accuracy of predictions of 6-month mortality and successful community discharge.

INTRODUCTION:Cross-national evidence on undiagnosed dementia and its prognostic implications remains limited. We compared the proportion of undiagnosed dementia, associated factors, and mortality in the United States and Brazil. METHODS:This population-based cohort study included adults aged ≥ 65 years from the 2016 US Health and Retirement Study (n = 9,539) and the 2015-2016 Brazilian Longitudinal Study of Aging (n = 3603), followed for mortality through 2020. Dementia was classified as no dementia, undiagnosed dementia, or diagnosed dementia using harmonized cognitive, functional, and informant measures. RESULTS:The proportion of undiagnosed dementia was higher in Brazil (76.1%) than in the United States (45.1%). Undiagnosed dementia was associated with increased 4-year mortality compared with no dementia in both countries. It was linked to younger age and absence of memory complaints, with marked socioeconomic and healthcare disparities in Brazil. DISCUSSION:Undiagnosed dementia is common and associated with increased mortality, identifying a vulnerable population missed by current diagnostic pathways.

Objectives Distress behaviors in dementia (DBDs) are common, are a significant burden to patients, families, and caregivers, and decrease quality of life. Despite risks, many nursing home (NH) residents are prescribed ≥ 1 classes of psychoactive medications for DBD. The objective of this study was to examine daily use of 4 classes of psychoactive medication in NH residents with DBD. Design This was a retrospective longitudinal cohort study. Setting and Participants Participants included long-stay Veterans Affairs NH residents with dementia in 2019-2020, with and without DBD. Methods We identified daily use of antipsychotics, benzodiazepines, antidepressants, and antiepileptics, identified from bar-coded administration data, from 30 days prior to 30 days after a documented DBD incident. We estimated multivariate associations between resident characteristics and administration of selected classes of medication on the DBD incident day, accounting for prior class use. Results The DBD and comparison groups consisted of 974 and 1138 individuals, respectively. Overall, the groups were 76 years old on average, and 14% had severe cognitive impairment. Administration of antipsychotic and benzodiazepine medication rose slightly 3 to 7 days before the DBD incident day, followed by sharp rises in administration on the DBD incident day, and sharp declines the next day. Increases in both classes persisted 30 days after the DBD incident day. None of these changes were observed in the comparison group. There were associations between benzodiazepine and antipsychotic administration and mental health conditions, and an inverse association between benzodiazepine administration and recent past gabapentin receipt. Conclusions and Implications Patterns of psychoactive medication administration varied among drug classes, but were linked across drug classes. Efforts to reduce prescribing of these medications need to be designed to take into account multiple classes of medication and target early signs of distress behavior.

BACKGROUND:Delirium is common in hospitalised older adults and is associated with mortality. Whether this prognostic association varies by baseline cognition is uncertain. We evaluated the association between delirium and 90-day mortality and whether baseline cognitive status modified this relationship. METHODS:We conducted a prospective, multicentre cohort study of adults aged ≥65 years admitted to 43 hospitals in five countries (Brazil, Angola, Chile, Colombia and Portugal; June 2022-December 2023). Delirium was assessed using the Confusion Assessment Method; cognitive status was measured using an informant-based Clinical Dementia Rating (CDR). Mortality within 90 days of admission was ascertained from hospital records, structured telephone follow-up by blinded assessors and registry linkage. We used mixed-effects survival models with random intercepts (state/province and study centre) and sequential adjustment for sociodemographic, clinical and hospital-related factors. Effect modification by CDR was examined with stratified analyses. RESULTS:Among 2556 patients (mean age 79 ± 9 years; 56% women), delirium occurred in 957 (37%). Delirium frequency rose with worsening cognition (CDR 0: 16%; CDR 0.5: 27%; CDR 1: 59%; CDR 2-3: 77%; P < .001). Delirium was associated with higher 90-day mortality (adjusted HR = 3.45; 95% CI = 2.83-4.20). The relative association with mortality was greatest in no dementia and attenuated in moderate-severe dementia. At 90 days, cumulative mortality was 54% with delirium vs. 15% without in CDR 0 (HR = 4.40; 95% CI = 3.15-6.16) and 36% vs. 17% in CDR 2-3 (HR = 2.22; 95% CI = 1.34-3.66). Patients with delirium also experienced more in-hospital complications (nosocomial infection, functional decline and prolonged stay). CONCLUSIONS:Although delirium was more frequent among patients with dementia, its relative association with 90-day mortality was strongest in those with no baseline dementia. The results provide a strong rationale for intervention trials to determine whether delirium prevention and management strategies can reduce mortality, particularly among patients without known dementia.

Treating older people with diabetes is challenging due to multiple medical comorbidities that might interfere with patients' ability to perform self-care. Most diabetes guidelines focus on improving glycaemia through addition of medications, but few address strategies to reduce medication burden for older adults-a concept known as deprescribing. Strategies for deprescribing might include stopping high-risk medications, decreasing the dose, or substituting for less harmful agents. Accordingly, glycaemic management strategies for older adults with type 1 and type 2 diabetes not responding to their current regimen require an understanding of how and when to realign therapy to meet patient's current needs, which represents a major clinical practice gap. With the gap in guidance on how to deprescribe or otherwise adjust therapy in older adults with diabetes in mind, the International Geriatric Diabetes Society, an organisation dedicated to improving care of older individuals with diabetes, convened a Deprescribing Consensus Initiative in May, 2023, to discuss Optimization of diabetes treatment regimens in older adults: the role of de-prescribing, de-intensification and simplification of regimens. The recommendations from this group initiative are discussed and described in this Review.

BACKGROUND:To avoid potential harms from hypoglycemia, guidelines for diabetes management in nursing home residents recommend less intensive glycemic control. However, it is unknown how often hypoglycemia and hyperglycemia co-occur in the same resident, which may present challenges for deintensification of diabetes treatment. METHODS:We conducted a cross-sectional study of insulin-treated Veterans Affairs nursing home residents with diabetes aged ≥ 65 years from 1/1/2016 to 9/30/2019 with a nursing home stay ≥ 7 days. Residents missing fingerstick glucose measurements during the first 7 days were excluded. We classified insulin use as basal insulin only, bolus insulin only, or a combination of basal and bolus insulin. We examined the prevalence of fingerstick-detected hypoglycemia (< 54 mg/dL, 54-69 mg/dL) and hyperglycemia (250-299, 300-349, 350-399, ≥ 400 mg/dL) overall and stratified by type of insulin. RESULTS:Among 12,031 insulin-treated residents, the mean age was 74.4 years, 98% were male, and 22% were non-White. Most residents (n = 7176, 59.6%) were treated with a combination of basal and bolus insulin, 31.8% (n = 3829) used bolus insulin alone and 8.5% (n = 1026) used basal insulin alone. During the first 7 days of the nursing home stay, 5730 (48%) had hyperglycemia ≥ 250 mg/dL alone, 862 (7%) had hypoglycemia < 70 mg/dL alone, 1488 (12%) had both hyperglycemia and hypoglycemia, and 3951 (33%) had neither hypoglycemia nor hyperglycemia. Residents on a combination of basal and bolus insulin were more likely to have hyperglycemia ≥ 400 mg/dL (10.2% vs. 3.6% for bolus insulin alone and 1.6% for basal insulin alone, p < 0.001) and to have hypoglycemia < 54 mg/dL (8.4% vs. 2.9% for bolus alone vs. 5.9% for basal alone, p < 0.001). CONCLUSION:Nearly two-thirds of nursing home residents with hypoglycemia also had hyperglycemia. Efforts to de-intensify diabetes treatment in nursing homes will need to address the high burden of hyperglycemia by tailoring the timing and type of insulin to minimize hypoglycemia while also not worsening hyperglycemia.

The prevalence of diabetes is rising among older adults. While most diabetes cases in older adults are type 2 diabetes mellitus (T2D), advances in type 1 diabetes mellitus (T1D) management and rising rates of adult-onset T1D have translated into a growing number of individuals with T1D living into older adulthood. This narrative review integrates existing evidence on management of T1D in older adults with expert opinions to provide practical guidance for generalists increasingly encountering the unique challenges and complexities of this growing population. The profound heterogeneity in clinical presentation, pathobiology, and disease progression across the older adult population can make it challenging to differentiate older adults with T1D from those with insulin treated T2D, particularly in adult-onset cases. However, timely diagnosis is critical to minimize exposure to hyperglycemia and reduce the risk for complications, as individuals with T1D rely entirely on exogenous insulin and require intensive self-management to prevent acute complications like hypoglycemia and ketoacidosis. Self-management of T1D in older adults presents unique challenges, including a high risk of hypoglycemia that must be mitigated in the setting of a lifelong requirement for insulin and evolving mismatches between intensive self-management demands and an older person’s capacity for self-care. The growing number of older adults with T1D underscores a lack of access to specialized care and limited training and resources for evidence-based management in primary care and post-acute/long-term care settings, as well as the dearth of high-quality clinical evidence specific to this population to inform care. Research to support changes across healthcare systems and at the policy level, in combination with education and multi-specialty collaboration, will ensure that healthcare providers and health systems are equipped and prepared to better meet the needs of the growing population of older adults with T1D.

INTRODUCTION:Lung cancer screening with low-dose computed tomography reduces lung cancer mortality in the long term but carries immediate risks. Guidelines recommend screening persons whose life expectancy exceeds the screening test's time to benefit, defined as the time from screening initiation to first observed benefit. This study aimed to estimate the time to benefit for lung cancer screening to prevent lung cancer mortality. METHODS:Randomized controlled trials of lung cancer screening with low-dose computed tomography were identified from two prior systematic reviews and an updated search to December 3, 2023. Studies that reported lung cancer mortality were included. For each study, independent Weibull survival curves were fitted and Markov chain Monte Carlo simulations were generated to estimate the absolute risk reduction at different time points. Time to benefit was determined as the time at which absolute risk reduction thresholds (ARR=0.0005, 0.001, 0.002) were crossed. These estimates were pooled using a random-effects meta-analysis model. RESULTS:A total of eight randomized controlled trials comprising 88,526 participants were included. Enrollment age ranged from age 50 to 70 years; follow-up duration ranged from 7.3 to 12.3 years. For every 1,000 persons screened, 3.4 years (95%=CI 2.2, 5.1) passed before 1 death from lung cancer was prevented (ARR=0.001). The time to prevent one lung cancer death per 2,000 persons screened (ARR=0.0005) was 2.2 years (95% CI=1.4, 3.4); per 500 persons screened (ARR=0.002), it was 5.2 years (95%=CI 3.7, 7.3). DISCUSSION:Lung cancer screening is most appropriate for older adults at high risk of lung cancer with a life expectancy greater than 3.4 years.

OBJECTIVES:Nearly 20% of hospitalized older adults are discharged to a skilled nursing facility (SNF) for short-term rehabilitation. Many subsequently experience adverse outcomes, such as hospital readmissions, transitioning to long-term care rather than returning home, or death. To guide shared decision making, we developed a prognostic model for multiple outcomes for older adults admitted to SNFs. DESIGN:Retrospective cohort study. SETTING AND PARTICIPANTS:Twenty percent national Medicare sample of community-dwelling older adults aged ≥66 discharged to an SNF after a hospitalization between 2017 and 2019. METHODS:We predicted 2 outcomes: 6-month all-cause mortality and "successful community discharge" (discharge to the community without subsequent rehospitalization or death within 30 days). Model predictors were pre-specified as age, sex, Elixhauser comorbidity score, hospital length of stay, elective vs urgent/emergency hospitalization, Medicaid status, principal hospital discharge diagnosis, surgical procedures, and hospitalizations in the past year. We used LASSO to reduce the 38 Elixhauser comorbidities to 12 comorbidities and logistic regression to determine separate predictor coefficients for the mortality and successful discharge outcomes. Model performance was assessed by discrimination [concordance statistic (c-statistic)] and calibration (calibration plots). Internal validation was performed via bootstrapping. RESULTS:The cohort included 523,740 individuals (median age 81, 62% female, 8% Black). Overall, 22% died by 6 months and 54% experienced a successful community discharge. Adjusted odds ratios varied based on outcome (eg, hospital length of stay was a stronger predictor of community discharge than mortality). The optimism-corrected c-statistics for the final model were 0.753 (95% CI, 0.752-0.755) for 6-month mortality and 0.692 (95% CI, 0.691-0.694) for successful community discharge. Calibration plots showed that the model was well calibrated for both outcomes. CONCLUSIONS AND IMPLICATIONS:In a national sample of older adults, this multi-outcome SNF prognostic model showed good discrimination and calibration. Risk predictions can help guide shared decision making and future planning among SNF clinicians, patients, and caregivers.

Estimating mortality risk in incarcerated adults is important for identifying individuals who may benefit from palliative care and compassionate release referrals. To develop and internally validate a 2-year mortality prediction model in incarcerated adults. Cohort study (February 1, 2018–February 1, 2020). Incarcerated adults aged ≥ 18 years residing at a California Department of Corrections and Rehabilitation (CDCR) prison for ≥ 1 year. Model predictors included demographics (e.g., age, sex), housing status (general housing vs. higher acuity infirmary bed vs. lower acuity infirmary bed), functional assessment (level of mobility restriction), healthcare utilization (e.g., hospitalizations and intensive care unit admissions in the previous year), and chronic conditions. The primary outcome was natural death at 2 years, defined as death due to causes other than suicide, homicide, accidental injury, or drug overdose. Cox proportional hazards regression with LASSO for variable selection was used to develop the model. Model performance was assessed by discrimination (area under the receiver operating characteristic curve (AUC) at 2 years) and calibration (plots of predicted and observed mortality). Classification metrics were assessed at clinically relevant thresholds. The final cohort included 89,430 adults (median age 40 years (interquartile range = 20), 10.2

Importance:The Walter Index is a widely used prognostic tool for assessing 12-month mortality risk among hospitalized older adults. Developed in the US in 2001, its accuracy in contemporary non-US contexts is unclear. Objective:To evaluate the external validity of the Walter Index in predicting posthospitalization mortality risk in Brazilian older adult inpatients. Design, Setting, and Participants:This prognostic study used data from a cohort of adults aged 70 years or older admitted to the geriatric unit of a university hospital in Brazil from January 1, 2009, to February 28, 2020. Participants underwent comprehensive geriatric assessments at admission, were reevaluated at discharge, and were subsequently followed up for 48 months. Data were analyzed from March to July 2024. Main Outcomes and Measures:The Walter Index, a score based on 6 risk factors (male sex, dependent activities of daily living at discharge, heart failure, cancer, high creatinine level, and low albumin level), was calculated to assess its predictive accuracy for 12-month mortality as well as 6-, 24-, and 48-month mortality. The study investigated whether incorporating delirium, frailty, or C-reactive protein level enhanced accuracy. Performance was assessed using discrimination, calibration, and clinical utility measures. Results:In total, 2780 participants (mean [SD] age, 81 [7] years; 1795 [65%] female) were included, with 89 (3%) lost to follow-up. The 12-month posthospitalization mortality rate was 23% (646 participants). Mortality was 7% (47 of 634) in the lowest-risk group (0-1 point), 17% (111 of 668) for 2 to 3 points, 25% (198 of 803) for 4 to 6 points, and 43% (290 of 675) in the highest-risk group (≥7 points). The index demonstrated an area under the receiver operating characteristic curve (AUC) of 0.714 (95% CI, 0.691-0.736) for predicting 12-month posthospitalization mortality (AUCs were 0.75 and 0.80 in the original derivation and validation cohorts, respectively). Comparable results were observed for mortality at 6 months (AUC, 0.726; 95% CI, 0.700-0.752), 24 months (AUC, 0.711; 95% CI, 0.691-0.730), and 48 months (AUC, 0.719; 95% CI, 0.700-0.738). Adding delirium modestly increased the index's discrimination (AUC, 0.723; 95% CI, 0.702-0.749); additionally including frailty and C-reactive protein level did not improve discrimination further (AUC, 0.723; 95% CI, 0.701-0.744). Conclusions and Relevance:In this prognostic study of hospitalized older adults in Brazil, the Walter Index showed similar discrimination in predicting postdischarge mortality as it did 2 decades ago in the US. These findings highlight the need for continuous validation and potential modification of established prognostic tools to improve their applicability across settings.

Many hospitalized older adults are discharged to a skilled nursing facility (SNF) for short-term rehabilitation following a hip/femur fracture or stroke. Mortality rates are high, and many never return home (i.e., high risk of “rehabbed to death”). To inform shared decision-making, we developed prognostic models using a 20% Medicare sample from 2017-2019 of community-dwelling adults aged ≥66 admitted to a SNF for hip/femur fracture or stroke. Within each cohort, we developed a prognostic model to predict 2 outcomes: 6-month mortality and “successful community discharge” (discharge to the community without subsequent rehospitalization or death within 30 days). Model predictors were pre-specified based on literature review: age, sex, hospital length of stay, Medicaid status, comorbidities, and hospitalizations in the past year. Model performance was assessed by discrimination (c-statistic) and calibration (calibration plots). Internal validation was performed via bootstrapping. The hip/femur fracture cohort included 52,843 individuals (mean age 83.4 years, 73% female, 3.3% Black), and the stroke cohort included 20,599 individuals (mean age 81.7 years, 60.1% female, 11.4% Black). Overall, in the hip/femur fracture and stroke cohorts, 15.5% and 24.6% died within 6 months and 60.1% and 45.4% experienced a successful community discharge, respectively. For the hip/femur fracture and stroke cohorts, the optimism-corrected c-statistics were 0.74 and 0.70 for 6-month mortality and 0.68 and 0.67 for successful community discharge, respectively. Calibration plots showed that the models were well-calibrated for both outcomes in both cohorts. Risk predictions from these models can help guide shared decision-making and future planning between SNF clinicians, patients, and caregivers.

Objective Sodium-glucose cotransporter-2 inhibitors (SGLT2is) are recommended as first-line cardiorenal protective therapy in type 2 diabetes. Because SGLT2is cause glycosuria and increase urine volume, they may exacerbate incontinence symptoms among patients with pre-existing urinary incontinence. Our objective was to determine how many adults meeting guideline indications for SGLT2i have frequent urinary incontinence.Research design and methods We conducted a cross-sectional analysis of National Health And Nutrition Examination Survey (NHANES) participants aged ≥55 with type 2 diabetes in 2013–2020. We determined whether participants met American Diabetes Association guideline indications for an SGLT2i due to heart failure or chronic kidney disease, or due to atherosclerotic cardiovascular disease or high cardiovascular risk (for the latter two cardiovascular indications, GLP-1RAs are guideline-recommended alternative medications). Frequent urinary incontinence was defined by self-report of leaking urine daily/nightly or a few times per week.Results There were 1726 NHANES participants aged ≥55 with type 2 diabetes, representing 16.0 million US adults; 19.6% (95% CI 17.3% to 22.2%) (3.1 million) met indications for an SGLT2i specifically and 50.9% (95% CI 47.2% to 54.7%) (8.2 million) met indications for either an SGLT2i or a GLP-1RA. Among those with indications for an SGLT2i specifically, 32.4% (95% CI 25.9% to 39.8%) had frequent urinary incontinence, representing 333 000 men and 685 000 women. Among those with indications for either an SGLT2i or GLP-1RA, 25.5% (95% CI 20.8% to 30.8%) had frequent urinary incontinence, representing 630 000 men and 1413 000 women.Conclusions Frequent urinary incontinence affects >15% of men and >40% of women aged ≥55 years with guideline indications for SGLT2i. Studies are needed to determine if incontinence increases risk of genital infections when initiating SGLT2is.

Introduction and Objective: Hyperglycemia causes glycosuria and frequent urination, which may worsen existing urinary incontinence. Our objective was to determine whether higher blood glucose levels or high glucose variability were associated with worse urinary incontinence in nursing home residents. Methods: We conducted a case-crossover study of Veterans’ Affairs nursing home residents with diabetes from 1/1/2016 to 9/30/2019. For study inclusion, residents were required to have ≥2 different assessments of urinary incontinence from the Minimum Dataset 3.0 (MDS) and ≥7 fingerstick glucose measurements during the 7 days prior to MDS assessment. From 7 days of fingerstick glucoses, we calculated the average and coefficient of variation (CV) of blood glucose. We used conditional logistic regression to examine the unadjusted association of mean glucose and CV glucose with more vs. less urinary incontinence. Results: Among 3,474 nursing home residents, mean age was 75, 98% were male, and 18% were non-Hispanic Black and 77% were non-Hispanic White. 84% were on insulin. Over 7 days, the median number of fingerstick measurements was 22 [25th percentile, 15, 75th percentile, 26]. The mean of the average glucose was 174 (standard deviation (SD) 47.6), and the mean of the CV was 26.6 (SD 9.5). Each 50mg/dL higher average glucose was associated with 1.14 greater odds of worse urinary incontinence symptoms (95% confidence interval, 1.05-1.22). CV was not associated with worse urinary incontinence (odds ratio per 10-percentage point greater CV: 1.03 (95%CI 0.96-1.11). Conclusion: Higher mean glucose is associated with more frequent urinary incontinence. Further research should determine if there is a particular threshold of hyperglycemia that significantly worsens urinary incontinence symptoms. A.K. Lee: Stock/Shareholder; GRAIL. Y. Shi: None. K.J. Lipska: None. J. Boscardin: None. S.J. Lee: None. National Institutes on Health (K01AG073532); National Institutes on Health (K24AG066998)

ABSTRACTBackgroundDeprescribing antihypertensives is of growing interest in geriatric medicine, yet the impact on functional status is unknown. We emulated a target trial of deprescribing antihypertensive medications compared with continued use on functional status measured by activities of daily living (ADL) in a long‐term care population.MethodsWe included 12,238 Veteran Affairs long‐term care residents age 65+ who had a stay ≥ 12 weeks between 2006 and 2019. After 4+ weeks of stable antihypertensive medication use, residents were classified as either deprescribed antihypertensives (reduced ≥ 1 medication or ≥ 30% dose) or continued users. Residents were followed up for 2 years, or censored at discharge, admission to hospice, protocol deviation (per‐protocol analysis only), or Sept 30, 2019. The outcome was ADL dependencies (scored 0–28; higher score = worse functionality), assessed approximately every 3 months. Our primary approach was to estimate per‐protocol effects using linear mixed‐effects regressions with inverse probability of treatment and censoring weighting, overall and stratified by dementia status. We estimated intention‐to‐treat effects as a secondary analysis.ResultsIn long‐term care residents, ADL scores worsened by a mean of 0.29 points (95%CI = 0.27, 0.31) per 3 months and antihypertensive deprescribing did not impact this worsening (difference between groups −0.04 points every 3 months, 95%CI = −0.15, 0.06). In the non‐dementia subgroup, ADL worsened by 0.15 points (95%CI = 0.11, 0.19) every 3 months. However, residents who were deprescribed showed a slightly improved ADL score over time while the continued users showed ADL decline (difference between groups −0.23 points every 3 months, 95%CI = −0.43, −0.03). Deprescribing was not associated with ADL change in the dementia subgroup. The intention‐to‐treat results were not meaningfully different.ConclusionsAntihypertensive deprescribing did not have a deleterious effect on functional status in long‐term care residents with or without dementia. This may be reassuring to residents and clinicians who are considering antihypertensive medication reduction or discontinuation in long‐term care settings.