Pub Date : 2025-02-08DOI: 10.1016/j.jamda.2024.105427
Heeeun Jang PhD , Ariel Avgar PhD , Russell Weaver PhD , Daniel Spertus BA , Kiran Abraham-Aggarwal , Joanna Bryan Ringel MPH , Madeline Sterling MD, MPH, MS
Objectives
Our study investigates unionization trends among direct care workers (DCWs) in the United States and examines the association between unionization and their wealth outcomes.
Design
This is a cross-sectional study using data from the Current Population Survey and Annual Social and Economic Supplement from 2009 to 2023.
Setting and Participants
Our study is based on US representative household surveys. The sample includes 17,522 DCWs (eg, personal care aides, nursing assistants, home health workers).
Methods
The prevalence and trend in labor unionization among DCWs and the association between their union status and wealth outcomes (hourly wage, employer-sponsored health insurance, pension plans, and poverty status) were analyzed. Union status identified by formal union membership or coverage by a union without formal membership.
Results
The sample was composed of mostly women (87%); they had a mean age of 41.8 ± 14.5 years, 38.1% were non-Hispanic White, 30.4% were non-Hispanic Black, 21.6% were Hispanic, and 9.9% were Asian and other. Overall, 12% (n = 1850) of DCWs were unionized. The Cox trend test showed decreasing in unionization rates over the study years. In fully adjusted models, unionized DCWs earned more than nonunionized DCWs ($1.2; 95% CI, $0.8-$1.6; P < .001). They were also more likely to have employer-sponsored health care insurance (odds ratio, 2.0; 95% CI, 1.8-2.3; P < .001) and pension plan (odds ratio, 1.8; 95% CI, 1.6-2.0; P < .001), with higher employer's contribution to health insurance ($561.2; 95% CI, $303.1-$819.3; P < .001). Also, unionization was negatively associated with DCWs' poverty status (odds ratio, 0.7; 95% CI, 0.6-0.9; P < .001).
Conclusions and Implications
Unionization is an effective means of improving the financial well-being of DCWs, a workforce that is currently facing challenging working conditions and high levels of turnover.
{"title":"What Do Unions Do for Direct Care Workers? Assessing Employment and Economic Outcomes","authors":"Heeeun Jang PhD , Ariel Avgar PhD , Russell Weaver PhD , Daniel Spertus BA , Kiran Abraham-Aggarwal , Joanna Bryan Ringel MPH , Madeline Sterling MD, MPH, MS","doi":"10.1016/j.jamda.2024.105427","DOIUrl":"10.1016/j.jamda.2024.105427","url":null,"abstract":"<div><h3>Objectives</h3><div>Our study investigates unionization trends among direct care workers (DCWs) in the United States and examines the association between unionization and their wealth outcomes.</div></div><div><h3>Design</h3><div>This is a cross-sectional study using data from the Current Population Survey and Annual Social and Economic Supplement from 2009 to 2023.</div></div><div><h3>Setting and Participants</h3><div>Our study is based on US representative household surveys. The sample includes 17,522 DCWs (eg, personal care aides, nursing assistants, home health workers).</div></div><div><h3>Methods</h3><div>The prevalence and trend in labor unionization among DCWs and the association between their union status and wealth outcomes (hourly wage, employer-sponsored health insurance, pension plans, and poverty status) were analyzed. Union status identified by formal union membership or coverage by a union without formal membership.</div></div><div><h3>Results</h3><div>The sample was composed of mostly women (87%); they had a mean age of 41.8 ± 14.5 years, 38.1% were non-Hispanic White, 30.4% were non-Hispanic Black, 21.6% were Hispanic, and 9.9% were Asian and other. Overall, 12% (n = 1850) of DCWs were unionized. The Cox trend test showed decreasing in unionization rates over the study years. In fully adjusted models, unionized DCWs earned more than nonunionized DCWs ($1.2; 95% CI, $0.8-$1.6; <em>P</em> < .001). They were also more likely to have employer-sponsored health care insurance (odds ratio, 2.0; 95% CI, 1.8-2.3; <em>P</em> < .001) and pension plan (odds ratio, 1.8; 95% CI, 1.6-2.0; <em>P</em> < .001), with higher employer's contribution to health insurance ($561.2; 95% CI, $303.1-$819.3; <em>P</em> < .001). Also, unionization was negatively associated with DCWs' poverty status (odds ratio, 0.7; 95% CI, 0.6-0.9; <em>P</em> < .001).</div></div><div><h3>Conclusions and Implications</h3><div>Unionization is an effective means of improving the financial well-being of DCWs, a workforce that is currently facing challenging working conditions and high levels of turnover.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 3","pages":"Article 105427"},"PeriodicalIF":4.2,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142872427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-08DOI: 10.1016/j.jamda.2025.105488
Geetha Mukerji, Leahora Rotteau, Joanne Goldman, Amol A Verma, Kaveh G Shojania, Fahad Razak, Sid Feldman, Patricia Rios, Laura Pus, Pauline Pariser, Tara O'Brien, Andrea L Moser, Brian M Wong
Objectives: Transfers to acute care hospitals expose long-term care residents to potential harm. We implemented Long-Term Care Plus (LTC+) at the outset of the COVID-19 pandemic to reduce emergency department (ED) transfers and improve access to urgent medical services by providing virtual specialist consultation, system navigation, and diagnostic and laboratory testing to 54 long-term care homes (LTCHs).
Design: This mixed-methods study aimed to determine if LTC+ led to a decrease in avoidable acute care transfers and to explore participants' perceptions and contextual factors influencing uptake.
Setting and participants: LTC+ was implemented across 54 LTCHs and 3 hospital hubs in Toronto, Canada.
Methods: Statistical process control charts were created to detect changes in ED transfer rates, stratifying data into high- and low-uptake LTCHs to evaluate the effect of LTC+ on ED transfer rates across 54 LTCHs. Semistructured interviews were conducted with health care providers, administrators, residents, and caregivers across 6 LTCHs and 3 hospital hubs and analyzed thematically.
Results: There were 9658 ED transfers during the study period (April 2020 to March 2022), of which 3860 (40.0%) did not require admission. LTC+ delivered 534 virtual consultations, with 5 LTCHs accounting for 59% of program use. Compared with baseline (January 2019 to February 2020), transfer rates decreased by 40%, with no difference seen between LTCHs with high vs low uptake. Factors influencing uptake include program awareness, motivation, alignment of LTCH resources and program services, and commitment to ED avoidance.
Conclusions and implications: The LTC+ program did not reduce ED transfers beyond secular trends attributable to the broader effects of the COVID-19 pandemic. Participants that used LTC+ identified important benefits that extended beyond ED avoidance including building self-efficacy and capacity in LTCHs to provide client-centered care with cross-sectoral collaboration. Refinements to the LTC+ program design and delivery and structural changes are needed to increase impact.
{"title":"A Program to Reduce Emergency Department Transfers and Build Long-Term Care Home Capacity: A Mixed-Methods Study.","authors":"Geetha Mukerji, Leahora Rotteau, Joanne Goldman, Amol A Verma, Kaveh G Shojania, Fahad Razak, Sid Feldman, Patricia Rios, Laura Pus, Pauline Pariser, Tara O'Brien, Andrea L Moser, Brian M Wong","doi":"10.1016/j.jamda.2025.105488","DOIUrl":"https://doi.org/10.1016/j.jamda.2025.105488","url":null,"abstract":"<p><strong>Objectives: </strong>Transfers to acute care hospitals expose long-term care residents to potential harm. We implemented Long-Term Care Plus (LTC+) at the outset of the COVID-19 pandemic to reduce emergency department (ED) transfers and improve access to urgent medical services by providing virtual specialist consultation, system navigation, and diagnostic and laboratory testing to 54 long-term care homes (LTCHs).</p><p><strong>Design: </strong>This mixed-methods study aimed to determine if LTC+ led to a decrease in avoidable acute care transfers and to explore participants' perceptions and contextual factors influencing uptake.</p><p><strong>Setting and participants: </strong>LTC+ was implemented across 54 LTCHs and 3 hospital hubs in Toronto, Canada.</p><p><strong>Methods: </strong>Statistical process control charts were created to detect changes in ED transfer rates, stratifying data into high- and low-uptake LTCHs to evaluate the effect of LTC+ on ED transfer rates across 54 LTCHs. Semistructured interviews were conducted with health care providers, administrators, residents, and caregivers across 6 LTCHs and 3 hospital hubs and analyzed thematically.</p><p><strong>Results: </strong>There were 9658 ED transfers during the study period (April 2020 to March 2022), of which 3860 (40.0%) did not require admission. LTC+ delivered 534 virtual consultations, with 5 LTCHs accounting for 59% of program use. Compared with baseline (January 2019 to February 2020), transfer rates decreased by 40%, with no difference seen between LTCHs with high vs low uptake. Factors influencing uptake include program awareness, motivation, alignment of LTCH resources and program services, and commitment to ED avoidance.</p><p><strong>Conclusions and implications: </strong>The LTC+ program did not reduce ED transfers beyond secular trends attributable to the broader effects of the COVID-19 pandemic. Participants that used LTC+ identified important benefits that extended beyond ED avoidance including building self-efficacy and capacity in LTCHs to provide client-centered care with cross-sectoral collaboration. Refinements to the LTC+ program design and delivery and structural changes are needed to increase impact.</p>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":" ","pages":"105488"},"PeriodicalIF":4.2,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143399466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-07DOI: 10.1016/j.jamda.2024.105474
Joury van der Griend MD , Fenne Wouters PhD , Gary Y.C. Yeung MD , Ineke J. Gerridzen MD, PhD , Karlijn J. Joling PhD , Sytse U. Zuidema MD, PhD , Martin Smalbrugge MD, PhD , Eefje M. Sizoo MD, PhD
Objective
Psychotropic drugs are frequently prescribed for challenging behavior in residents with dementia in nursing homes. Recommendations on psychotropic drug use for challenging behavior are described in the Dutch multidisciplinary guideline “Problem behavior in dementia.” This study aimed to gain insight into the adherence to guideline recommendations on drug type and timing of evaluations of different types of psychotropic drugs for challenging behavior in a national sentinel network of Dutch nursing homes.
Design
Prospective observational study.
Setting and Participants
Data on psychotropic drug use of residents in a sentinel network of 22 nursing homes across the Netherlands were collected during a 3-month measurement period in 2021.
Methods
Physicians completed registration forms integrated into the electronic health record after reporting on psychotropic drug use in progress notes. Prescribed drug type(s), type of drug intervention (start/stop/change in dosage/evaluation), and drug indications were recorded. Adherence to guideline recommendations on drug type was achieved if prescribed psychotropic drugs were recommended for the specified indication. Adherence to guideline recommendations on timing of evaluations was achieved if evaluations were reported within 7 days after starting.
Results
A total of 1279 forms on psychotropic drug use for specified challenging behaviors in 599 residents were collected. Recommended psychotropic drugs were used in 57% of all forms. The highest rate of recommended psychotropic drugs was for psychotic behavior (80%), followed by agitation (48%). Adherence was lowest for nighttime restlessness (22%). Of all newly started prescriptions, 17% were evaluated within the recommended 7 days after starting.
Conclusions and Implications
Adherence to guideline recommendations on drug type was present in just over half of the cases. Frequently, evaluations were not reported or were performed after the recommended time frame. Further research is needed to determine the reasons for non-recommended psychotropic drug use, low evaluation reporting rates, and factors that influence adherence. The use of a sentinel network may increase awareness and adherence.
{"title":"Adherence to Guideline Recommendations on Psychotropic Drug Use for Challenging Behavior in Dementia","authors":"Joury van der Griend MD , Fenne Wouters PhD , Gary Y.C. Yeung MD , Ineke J. Gerridzen MD, PhD , Karlijn J. Joling PhD , Sytse U. Zuidema MD, PhD , Martin Smalbrugge MD, PhD , Eefje M. Sizoo MD, PhD","doi":"10.1016/j.jamda.2024.105474","DOIUrl":"10.1016/j.jamda.2024.105474","url":null,"abstract":"<div><h3>Objective</h3><div>Psychotropic drugs are frequently prescribed for challenging behavior in residents with dementia in nursing homes. Recommendations on psychotropic drug use for challenging behavior are described in the Dutch multidisciplinary guideline “Problem behavior in dementia.” This study aimed to gain insight into the adherence to guideline recommendations on drug type and timing of evaluations of different types of psychotropic drugs for challenging behavior in a national sentinel network of Dutch nursing homes.</div></div><div><h3>Design</h3><div>Prospective observational study.</div></div><div><h3>Setting and Participants</h3><div>Data on psychotropic drug use of residents in a sentinel network of 22 nursing homes across the Netherlands were collected during a 3-month measurement period in 2021.</div></div><div><h3>Methods</h3><div>Physicians completed registration forms integrated into the electronic health record after reporting on psychotropic drug use in progress notes. Prescribed drug type(s), type of drug intervention (start/stop/change in dosage/evaluation), and drug indications were recorded. Adherence to guideline recommendations on drug type was achieved if prescribed psychotropic drugs were recommended for the specified indication. Adherence to guideline recommendations on timing of evaluations was achieved if evaluations were reported within 7 days after starting.</div></div><div><h3>Results</h3><div>A total of 1279 forms on psychotropic drug use for specified challenging behaviors in 599 residents were collected. Recommended psychotropic drugs were used in 57% of all forms. The highest rate of recommended psychotropic drugs was for psychotic behavior (80%), followed by agitation (48%). Adherence was lowest for nighttime restlessness (22%). Of all newly started prescriptions, 17% were evaluated within the recommended 7 days after starting.</div></div><div><h3>Conclusions and Implications</h3><div>Adherence to guideline recommendations on drug type was present in just over half of the cases. Frequently, evaluations were not reported or were performed after the recommended time frame. Further research is needed to determine the reasons for non-recommended psychotropic drug use, low evaluation reporting rates, and factors that influence adherence. The use of a sentinel network may increase awareness and adherence.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 4","pages":"Article 105474"},"PeriodicalIF":4.2,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143059454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-07DOI: 10.1016/j.jamda.2024.105481
Joshua D. Niznik PharmD, PhD , Florentia E. Sileanu MS , Xinhua Zhao PhD , Kelvin Tran PharmD, MPS , Laura C. Hanson MD, MPH , Alan Kinlaw PhD , Thomas R. Radomski MD, MS , Alexa Ehlert MS , Sydney Springer PharmD, MS , Binxin Cao PharmD, MPH , Loren J. Schleiden MS , Carolyn T. Thorpe PhD, MPH
Objectives
Identifying people with possible dementia in health care systems is important to study outcomes and target improvements in care. This study sought to compare the performance of diagnostic codes and Minimum Data Set (MDS)-based measures for identifying dementia and cognitive impairment in older veteran nursing home residents.
Design
Retrospective, cross-sectional analysis.
Setting and Participants
We used real-world health care data from the Veterans Affairs (VA) Residential History File, VA Corporate Data Warehouse (CDW), Medicare claims, and the MDS to assemble a cohort of VA Community Living Center (CLC) admissions over 2015 to 2021 for veterans aged ≥ 65 with dual VA and Medicare enrollment (n = 54,234).
Methods
We defined 3 measures of possible dementia: (1) claims/CDW diagnoses using Chronic Conditions Warehouse (CCW) algorithms for Alzheimer’s disease or non-Alzheimer’s dementia; (2) MDS active diagnosis items for Alzheimer’s disease and non-Alzheimer’s dementia; and (3) MDS Cognitive Function Scale (CFS) assessment indicating at least mild cognitive impairment. We calculated proportions identified with each definition, and sensitivity, specificity, and positive predictive value of claims/CDW diagnoses and MDS indicators for dementia for identifying CFS impairment.
Results
Among VA CLC residents, 61.4% met at least 1 criterion for possible dementia (38.6% claims/CDW, 23.3% MDS active diagnosis, 50.8% CFS). Diagnoses from claims/CDW had 56.5% sensitivity and 80.0% specificity for identifying veterans with CFS cognitive impairment. Active diagnoses from the MDS exhibited poorer sensitivity (38.1%), but higher specificity (92.0%) identifying veterans with cognitive impairment on the CFS.
Conclusions and Implications
Consistent with what has been reported in Medicare nursing home residents, we observed only partial overlap between indicators of possible dementia across diagnosis codes and other indicators vs cognitive assessments in MDS. Our findings support the utility of these measures for identifying individuals with possible dementia across different systems, but further work is needed to understand implications when using diagnosis codes or cognitive assessments.
{"title":"A Comparison of Measures for Identifying Possible Dementia in Veterans Affairs Nursing Home Residents","authors":"Joshua D. Niznik PharmD, PhD , Florentia E. Sileanu MS , Xinhua Zhao PhD , Kelvin Tran PharmD, MPS , Laura C. Hanson MD, MPH , Alan Kinlaw PhD , Thomas R. Radomski MD, MS , Alexa Ehlert MS , Sydney Springer PharmD, MS , Binxin Cao PharmD, MPH , Loren J. Schleiden MS , Carolyn T. Thorpe PhD, MPH","doi":"10.1016/j.jamda.2024.105481","DOIUrl":"10.1016/j.jamda.2024.105481","url":null,"abstract":"<div><h3>Objectives</h3><div>Identifying people with possible dementia in health care systems is important to study outcomes and target improvements in care. This study sought to compare the performance of diagnostic codes and Minimum Data Set (MDS)-based measures for identifying dementia and cognitive impairment in older veteran nursing home residents.</div></div><div><h3>Design</h3><div>Retrospective, cross-sectional analysis.</div></div><div><h3>Setting and Participants</h3><div>We used real-world health care data from the Veterans Affairs (VA) Residential History File, VA Corporate Data Warehouse (CDW), Medicare claims, and the MDS to assemble a cohort of VA Community Living Center (CLC) admissions over 2015 to 2021 for veterans aged ≥ 65 with dual VA and Medicare enrollment (n = 54,234).</div></div><div><h3>Methods</h3><div>We defined 3 measures of possible dementia: (1) claims/CDW diagnoses using Chronic Conditions Warehouse (CCW) algorithms for Alzheimer’s disease or non-Alzheimer’s dementia; (2) MDS active diagnosis items for Alzheimer’s disease and non-Alzheimer’s dementia; and (3) MDS Cognitive Function Scale (CFS) assessment indicating at least mild cognitive impairment. We calculated proportions identified with each definition, and sensitivity, specificity, and positive predictive value of claims/CDW diagnoses and MDS indicators for dementia for identifying CFS impairment.</div></div><div><h3>Results</h3><div>Among VA CLC residents, 61.4% met at least 1 criterion for possible dementia (38.6% claims/CDW, 23.3% MDS active diagnosis, 50.8% CFS). Diagnoses from claims/CDW had 56.5% sensitivity and 80.0% specificity for identifying veterans with CFS cognitive impairment. Active diagnoses from the MDS exhibited poorer sensitivity (38.1%), but higher specificity (92.0%) identifying veterans with cognitive impairment on the CFS.</div></div><div><h3>Conclusions and Implications</h3><div>Consistent with what has been reported in Medicare nursing home residents, we observed only partial overlap between indicators of possible dementia across diagnosis codes and other indicators vs cognitive assessments in MDS. Our findings support the utility of these measures for identifying individuals with possible dementia across different systems, but further work is needed to understand implications when using diagnosis codes or cognitive assessments.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 4","pages":"Article 105481"},"PeriodicalIF":4.2,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143074902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-07DOI: 10.1016/j.jamda.2024.105480
Tingting Zhang MD, PhD , Christopher M. Santostefano MPH, RN , Daniel Harris MPH, PhD , Antoinette B. Coe PharmD, PhD , Andrew R. Zullo PharmD, PhD , Lauren B. Gerlach DO, MS , Richa Joshi MS , Julie P.W. Bynum MD, MPH , Theresa I. Shireman PhD
Objectives
To estimate the immediate and long-term effects of a visitor restriction policy on antipsychotic use in nursing home (NH) residents with Alzheimer's disease and related dementias (ADRD) during COVID-19.
Design
A repeated cross-sectional study time series analysis was conducted using NH electronic health records (EHRs) from January 1, 2020, to December 31, 2021.
Setting and Participants
A large, multistate sample of NH residents living with ADRD.
Methods
We calculated weekly changes in facility-level prevalence of antipsychotic use using interrupted time series (ITS) to compare level and slope changes in antipsychotic use before, during, and after NH visitation restrictions. Generalized linear models with generalized estimating equations, logit link, binomial distribution, and AR-1 correlation structure were used for all ITS analyses. Final models were stratified by long- and short-stay residents and adjusted for NH-level covariates including resident demographics, clinical diagnoses, and nurse staffing.
Results
We observed more than 8500 long-stay and 2700 short-stay NH residents with ADRD. Among long-stay residents, the weekly prevalence of antipsychotic use increased from 18.9% as of January 7, 2020, to 24.9% by December 31, 2021. For short-stay residents, antipsychotic use increased from 21.1% to 26.6% over this same window. The ITS analysis showed no meaningful changes in the relative rate of change in antipsychotic use during and after visitor restrictions, relative to pre-policy trends.
Conclusions and Implications
The Centers for Medicare and Medicaid Services visitation restriction policy had no meaningful impact on antipsychotic use among NH residents with ADRD. However, antipsychotic use increased over time for both long- and short-stay residents and remained above pre-pandemic levels by the end of 2021. Our findings emphasize the potential for increased reliance on pharmacotherapy to manage resident symptoms during public health emergencies. Future infection control efforts should prioritize safe interpersonal care delivery and consider policies that improve vigilance of medication utilization changes among high-risk populations.
{"title":"Effects of Visitation Restriction on Antipsychotic Drug Use among Nursing Home Residents with ADRD during the COVID-19 Pandemic","authors":"Tingting Zhang MD, PhD , Christopher M. Santostefano MPH, RN , Daniel Harris MPH, PhD , Antoinette B. Coe PharmD, PhD , Andrew R. Zullo PharmD, PhD , Lauren B. Gerlach DO, MS , Richa Joshi MS , Julie P.W. Bynum MD, MPH , Theresa I. Shireman PhD","doi":"10.1016/j.jamda.2024.105480","DOIUrl":"10.1016/j.jamda.2024.105480","url":null,"abstract":"<div><h3>Objectives</h3><div>To estimate the immediate and long-term effects of a visitor restriction policy on antipsychotic use in nursing home (NH) residents with Alzheimer's disease and related dementias (ADRD) during COVID-19.</div></div><div><h3>Design</h3><div>A repeated cross-sectional study time series analysis was conducted using NH electronic health records (EHRs) from January 1, 2020, to December 31, 2021.</div></div><div><h3>Setting and Participants</h3><div>A large, multistate sample of NH residents living with ADRD.</div></div><div><h3>Methods</h3><div>We calculated weekly changes in facility-level prevalence of antipsychotic use using interrupted time series (ITS) to compare level and slope changes in antipsychotic use before, during, and after NH visitation restrictions. Generalized linear models with generalized estimating equations, logit link, binomial distribution, and AR-1 correlation structure were used for all ITS analyses. Final models were stratified by long- and short-stay residents and adjusted for NH-level covariates including resident demographics, clinical diagnoses, and nurse staffing.</div></div><div><h3>Results</h3><div>We observed more than 8500 long-stay and 2700 short-stay NH residents with ADRD. Among long-stay residents, the weekly prevalence of antipsychotic use increased from 18.9% as of January 7, 2020, to 24.9% by December 31, 2021. For short-stay residents, antipsychotic use increased from 21.1% to 26.6% over this same window. The ITS analysis showed no meaningful changes in the relative rate of change in antipsychotic use during and after visitor restrictions, relative to pre-policy trends.</div></div><div><h3>Conclusions and Implications</h3><div>The Centers for Medicare and Medicaid Services visitation restriction policy had no meaningful impact on antipsychotic use among NH residents with ADRD. However, antipsychotic use increased over time for both long- and short-stay residents and remained above pre-pandemic levels by the end of 2021. Our findings emphasize the potential for increased reliance on pharmacotherapy to manage resident symptoms during public health emergencies. Future infection control efforts should prioritize safe interpersonal care delivery and consider policies that improve vigilance of medication utilization changes among high-risk populations.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 4","pages":"Article 105480"},"PeriodicalIF":4.2,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143074916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Outdoor falls can negatively impact the health and functional abilities of community-dwelling older adults. Although there are existing evidence-based programs for falls prevention, none specifically target outdoor falls. To fill this gap in research and practice, the Stroll Safe program was developed. Prior studies have examined outcomes for Stroll Safe related to knowledge of outdoor fall risks and safe strategy use. The purpose of this study was to examine outcomes related to outdoor falls self-efficacy.
Design
In this cluster randomized controlled trial, we examined the effects of the Stroll Safe program on outdoor falls self-efficacy as per scores on the Outdoor Falls Self-Efficacy Questionnaire (OFSQ).
Setting and Participants
The study was conducted in 8 naturally occurring retirement communities. Participants (N = 93) were community-dwelling older adults with a history of outdoor falls and/or who were fearful of falling outside.
Methods
Linear mixed-effects models were used for the analyses.
Results
The intervention group had a significantly higher OFSQ score post-intervention (P < .001), which persisted at the 6-week follow-up (P < .001). Four of the 5 items in the OFSQ showed the same behavior.
Conclusions and Implications
Results reveal that the Stroll Safe program is effective in increasing outdoor falls self-efficacy building on previously established program benefits.
{"title":"The Stroll Safe Randomized Controlled Trial: Program Effects on Falls Self-Efficacy","authors":"Tracy Chippendale PhD, OTR/L , Lijing Wei BS , Alex Dahlen PhD","doi":"10.1016/j.jamda.2024.105478","DOIUrl":"10.1016/j.jamda.2024.105478","url":null,"abstract":"<div><h3>Objective</h3><div>Outdoor falls can negatively impact the health and functional abilities of community-dwelling older adults. Although there are existing evidence-based programs for falls prevention, none specifically target outdoor falls. To fill this gap in research and practice, the Stroll Safe program was developed. Prior studies have examined outcomes for Stroll Safe related to knowledge of outdoor fall risks and safe strategy use. The purpose of this study was to examine outcomes related to outdoor falls self-efficacy.</div></div><div><h3>Design</h3><div>In this cluster randomized controlled trial, we examined the effects of the Stroll Safe program on outdoor falls self-efficacy as per scores on the Outdoor Falls Self-Efficacy Questionnaire (OFSQ).</div></div><div><h3>Setting and Participants</h3><div>The study was conducted in 8 naturally occurring retirement communities. Participants (N = 93) were community-dwelling older adults with a history of outdoor falls and/or who were fearful of falling outside.</div></div><div><h3>Methods</h3><div>Linear mixed-effects models were used for the analyses.</div></div><div><h3>Results</h3><div>The intervention group had a significantly higher OFSQ score post-intervention (<em>P</em> < .001), which persisted at the 6-week follow-up (<em>P</em> < .001). Four of the 5 items in the OFSQ showed the same behavior.</div></div><div><h3>Conclusions and Implications</h3><div>Results reveal that the Stroll Safe program is effective in increasing outdoor falls self-efficacy building on previously established program benefits.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 4","pages":"Article 105478"},"PeriodicalIF":4.2,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143066531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-06DOI: 10.1016/j.jamda.2024.105463
Youngran Kim PhD , Trudy M. Krause DrPH , Rafael Samper-Ternent MD, PhD , Antonio L. Teixeira MD, PhD
Objectives
To assess recent trends in antipsychotic use among older adults with Alzheimer's disease and related dementias (ADRDs) according to their residential status and determine the factors associated with the use of antipsychotics.
Design
Population-based, cross-sectional study using Texas Medicare Fee-for-Service data.
Setting and Participants
Individuals aged ≥ 65 years with ADRDs who had at least 3 months of Medicare Part A and B, and Part D for prescription drug coverage, in any year between 2015 and 2020.
Methods
Temporal trends for antipsychotic use were reported by calendar year, and the associations between antipsychotic use and potential predictors were assessed overall and by residential status.
Results
Among an annual average of 161,848 older adults with ADRDs (median age, 82 years; 64.8% female), overall antipsychotic use decreased by 25.8%, from 14.5% in 2015 to 10.8% in 2020. The decline was primarily observed among those with any nursing facility (NF) residence, where use dropped from 22.1% to 12.4%, whereas community-dwelling individuals maintained a steady rate of approximately 10%. Factors associated with increased antipsychotic use included male sex, Black and Hispanic individuals, dual eligibility, Alzheimer's disease (non-Alzheimer's disease), emergency department visits, hospitalization, depression, and anxiety disorders. However, these associations varied across residential statuses. Older age was more strongly associated with decreased antipsychotic use among those with NF residence than those in the community. Compared with white individuals, Black individuals were more likely to receive antipsychotics in the community, whereas Hispanic and Asian individuals were more likely to receive antipsychotics among those with NF residence.
Conclusions and Implications
Although antipsychotic use substantially decreased among those with NF residence, it remained steady among community-dwelling individuals. Given that two-thirds of individuals with dementia reside in the community, more attention is needed to understand antipsychotic use in this population.
{"title":"Antipsychotic Use in Older Adults with Dementia: Community and Nursing Facility Trends in Texas, 2015-2020","authors":"Youngran Kim PhD , Trudy M. Krause DrPH , Rafael Samper-Ternent MD, PhD , Antonio L. Teixeira MD, PhD","doi":"10.1016/j.jamda.2024.105463","DOIUrl":"10.1016/j.jamda.2024.105463","url":null,"abstract":"<div><h3>Objectives</h3><div>To assess recent trends in antipsychotic use among older adults with Alzheimer's disease and related dementias (ADRDs) according to their residential status and determine the factors associated with the use of antipsychotics.</div></div><div><h3>Design</h3><div>Population-based, cross-sectional study using Texas Medicare Fee-for-Service data.</div></div><div><h3>Setting and Participants</h3><div>Individuals aged ≥ 65 years with ADRDs who had at least 3 months of Medicare Part A and B, and Part D for prescription drug coverage, in any year between 2015 and 2020.</div></div><div><h3>Methods</h3><div>Temporal trends for antipsychotic use were reported by calendar year, and the associations between antipsychotic use and potential predictors were assessed overall and by residential status.</div></div><div><h3>Results</h3><div>Among an annual average of 161,848 older adults with ADRDs (median age, 82 years; 64.8% female), overall antipsychotic use decreased by 25.8%, from 14.5% in 2015 to 10.8% in 2020. The decline was primarily observed among those with any nursing facility (NF) residence, where use dropped from 22.1% to 12.4%, whereas community-dwelling individuals maintained a steady rate of approximately 10%. Factors associated with increased antipsychotic use included male sex, Black and Hispanic individuals, dual eligibility, Alzheimer's disease (non-Alzheimer's disease), emergency department visits, hospitalization, depression, and anxiety disorders. However, these associations varied across residential statuses. Older age was more strongly associated with decreased antipsychotic use among those with NF residence than those in the community. Compared with white individuals, Black individuals were more likely to receive antipsychotics in the community, whereas Hispanic and Asian individuals were more likely to receive antipsychotics among those with NF residence.</div></div><div><h3>Conclusions and Implications</h3><div>Although antipsychotic use substantially decreased among those with NF residence, it remained steady among community-dwelling individuals. Given that two-thirds of individuals with dementia reside in the community, more attention is needed to understand antipsychotic use in this population.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 3","pages":"Article 105463"},"PeriodicalIF":4.2,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143007482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-06DOI: 10.1016/j.jamda.2024.105477
Ali Kapan PhD , Milos Ristic MSc , Richard Felsinger MD , Thomas Waldhoer PhD, Professor
Objectives
Fatigue and sleep disorders are common geriatric conditions and are associated with an increased risk of cognitive decline. This study aimed to examine the relationships among self-perceived fatigue, objective muscle fatigue, sleep apnea risk, insomnia, and cognitive function, focusing on their associations with mild cognitive impairment (MCI).
Design
Cross-sectional study.
Setting and Participants
The study involved 217 older adults (median age: 80 years, 65.6% women) from 2 residential care facilities in Vienna, Austria.
Methods
Cognitive function was assessed with the Mini-Mental State Examination. Sleep quality and disturbances were evaluated using the Athens Insomnia Scale (AIS) and STOP-BANG questionnaire. Subjective fatigue was measured with the Multidimensional Fatigue Inventory, and objective muscle fatigue through 2 rounds of 10 maximal handgrip strength tests, separated by a 1-hour rest. Multivariable regression models adjusted for age, sex, comorbidities, sleep medication, and total number of medications were used to analyze associations among fatigue, sleep disturbance, and MCI.
Results
Participants with MCI (n = 72) had significantly lower handgrip strength (median Fmax1: 18.75 kg vs 23.44 kg), higher muscle fatigue ratios (median: 1.40 vs 1.19), and poorer recovery ratios (median: 0.85 vs 1.01) compared with those without MCI (n = 145). Sleep disorders and fatigue measures were more common in those with MCI, with 80.6% reporting insomnia (AIS ≥6) compared with 21.4% in the cognitively intact group. Logistic regression showed that muscle fatigue combined with sleep apnea risk was associated with the highest odds of cognitive impairment [odds ratio (OR), 6.12; 95% CI, 2.77−13.51; adjusted OR, 3.02; 95% CI, 1.15−7.92]. Recovery capacity showed a positive association with cognitive function (β = 0.150, P < .05).
Conclusions and Implications
Sleep apnea, muscle fatigue, and reduced recovery capacity are significantly associated with cognitive decline in older adults. Assessing these factors may help in identifying individuals at risk of cognitive impairment.
目的：疲劳和睡眠障碍是常见的老年病，与认知能力下降的风险增加有关。本研究旨在探讨自我感觉疲劳、客观肌肉疲劳、睡眠呼吸暂停风险、失眠和认知功能之间的关系，重点关注它们与轻度认知障碍（MCI）之间的关联：设计：横断面研究：研究对象：奥地利维也纳两家养老机构的 217 名老年人（中位年龄：80 岁，65.6% 为女性）：认知功能通过小型精神状态检查进行评估。采用雅典失眠量表（AIS）和 STOP-BANG 问卷评估睡眠质量和睡眠障碍。主观疲劳通过多维疲劳量表进行测量，客观肌肉疲劳通过2轮10次最大手握力测试进行测量，测试间歇1小时。使用调整了年龄、性别、合并症、睡眠药物和药物总数的多变量回归模型来分析疲劳、睡眠障碍和 MCI 之间的关联：与未患有 MCI 的参与者（n = 145）相比，患有 MCI 的参与者（n = 72）手握力明显较低（Fmax1 中位数：18.75 kg vs 23.44 kg），肌肉疲劳比率较高（中位数：1.40 vs 1.19），恢复比率较低（中位数：0.85 vs 1.01）。睡眠障碍和疲劳测量在 MCI 患者中更为常见，80.6% 的患者报告失眠（AIS ≥6），而在认知功能完好组中仅为 21.4%。逻辑回归显示，肌肉疲劳合并睡眠呼吸暂停风险与认知障碍的最高几率相关（几率比 [OR]，6.12；95% CI，2.77-13.51；调整后的几率比，3.02；95% CI，1.15-7.92）。恢复能力与认知功能呈正相关（β = 0.150，P < .05）：睡眠呼吸暂停、肌肉疲劳和恢复能力下降与老年人认知能力下降有显著关联。对这些因素进行评估有助于识别有认知功能障碍风险的人群。
{"title":"Association of Self-Perceived Fatigue, Muscle Fatigue, and Sleep Disorders with Cognitive Function in Older Adults: A Cross-Sectional Study","authors":"Ali Kapan PhD , Milos Ristic MSc , Richard Felsinger MD , Thomas Waldhoer PhD, Professor","doi":"10.1016/j.jamda.2024.105477","DOIUrl":"10.1016/j.jamda.2024.105477","url":null,"abstract":"<div><h3>Objectives</h3><div>Fatigue and sleep disorders are common geriatric conditions and are associated with an increased risk of cognitive decline. This study aimed to examine the relationships among self-perceived fatigue, objective muscle fatigue, sleep apnea risk, insomnia, and cognitive function, focusing on their associations with mild cognitive impairment (MCI).</div></div><div><h3>Design</h3><div>Cross-sectional study.</div></div><div><h3>Setting and Participants</h3><div>The study involved 217 older adults (median age: 80 years, 65.6% women) from 2 residential care facilities in Vienna, Austria.</div></div><div><h3>Methods</h3><div>Cognitive function was assessed with the Mini-Mental State Examination. Sleep quality and disturbances were evaluated using the Athens Insomnia Scale (AIS) and STOP-BANG questionnaire. Subjective fatigue was measured with the Multidimensional Fatigue Inventory, and objective muscle fatigue through 2 rounds of 10 maximal handgrip strength tests, separated by a 1-hour rest. Multivariable regression models adjusted for age, sex, comorbidities, sleep medication, and total number of medications were used to analyze associations among fatigue, sleep disturbance, and MCI.</div></div><div><h3>Results</h3><div>Participants with MCI (n = 72) had significantly lower handgrip strength (median Fmax1: 18.75 kg vs 23.44 kg), higher muscle fatigue ratios (median: 1.40 vs 1.19), and poorer recovery ratios (median: 0.85 vs 1.01) compared with those without MCI (n = 145). Sleep disorders and fatigue measures were more common in those with MCI, with 80.6% reporting insomnia (AIS ≥6) compared with 21.4% in the cognitively intact group. Logistic regression showed that muscle fatigue combined with sleep apnea risk was associated with the highest odds of cognitive impairment [odds ratio (OR), 6.12; 95% CI, 2.77−13.51; adjusted OR, 3.02; 95% CI, 1.15−7.92]. Recovery capacity showed a positive association with cognitive function (β = 0.150, <em>P</em> < .05).</div></div><div><h3>Conclusions and Implications</h3><div>Sleep apnea, muscle fatigue, and reduced recovery capacity are significantly associated with cognitive decline in older adults. Assessing these factors may help in identifying individuals at risk of cognitive impairment.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 4","pages":"Article 105477"},"PeriodicalIF":4.2,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143066525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-02DOI: 10.1016/j.jamda.2025.105485
Kathy C Richards, Liam M Fry, Alicia J Lozano, Wenyan Ji, Janet D Morrison, Katherine C Britt, Donald L Bliwise, Nalaka S Gooneratne, Alexandra L Hanlon
Objectives: Restless legs syndrome (RLS), a common, treatable, sensorimotor disorder of nighttime uncomfortable leg sensations that interfere with sleep, may prompt nighttime agitation in persons with dementia.
Design: This randomized trial was double-blind and placebo-controlled. Participants received a Food and Drug Administration-approved drug for RLS, gabapentin enacarbil (GEn) (Horizant) or placebo.
Setting and participants: Older adults (N = 147) with dementia due to Alzheimer's disease, nighttime agitation, and RLS, residing in long-term care or at home, participated.
Methods: The primary outcome was change from baseline to 8 weeks in nighttime agitation between 5 pm and 7 am on the Cohen-Mansfield Agitation Inventory Index, Direct Observation. Multivariable linear mixed effects regression models based on multiply imputed data were estimated on nighttime agitation and sleep, with treatment group, week, the 2-way interaction of group and week as predictors, and mean arterial pressure as a covariate based on baseline group imbalances.
Results: Mean age ± SD was 83.4 ± 9.1 years. Most were female (72.0%), White (92.3%), non-Hispanic (84.6%), and lived in nursing homes (76.9%). Nighttime agitation, by group over time, was significant at 8 weeks (estimate, -1.67; P = .003) and 2 weeks. Total sleep time (actigraphy) by group over time was significant at 8 weeks (estimate, 48.45; P = .026). Observed nighttime wake by group over time was significant at 2 (estimate, -12.54; P = .006) and 8 weeks (estimate, -11.12; P = .015). The number having ≥1 adverse events was 60 in the GEn group (81.1%) and 50 in the placebo group (68.5%); with 12 serious adverse events in placebo and 10 in the GEn group. The GEn group had a trend toward more falls (P = .066).
Conclusions and implications: Our findings suggest a novel approach for nighttime agitation in persons with dementia: assessing for RLS and initiating interventions. Larger and longer trials are needed.
{"title":"Treatment of Restless Legs Syndrome Improves Agitation and Sleep in Persons With Dementia: A Randomized Trial.","authors":"Kathy C Richards, Liam M Fry, Alicia J Lozano, Wenyan Ji, Janet D Morrison, Katherine C Britt, Donald L Bliwise, Nalaka S Gooneratne, Alexandra L Hanlon","doi":"10.1016/j.jamda.2025.105485","DOIUrl":"https://doi.org/10.1016/j.jamda.2025.105485","url":null,"abstract":"<p><strong>Objectives: </strong>Restless legs syndrome (RLS), a common, treatable, sensorimotor disorder of nighttime uncomfortable leg sensations that interfere with sleep, may prompt nighttime agitation in persons with dementia.</p><p><strong>Design: </strong>This randomized trial was double-blind and placebo-controlled. Participants received a Food and Drug Administration-approved drug for RLS, gabapentin enacarbil (GEn) (Horizant) or placebo.</p><p><strong>Setting and participants: </strong>Older adults (N = 147) with dementia due to Alzheimer's disease, nighttime agitation, and RLS, residing in long-term care or at home, participated.</p><p><strong>Methods: </strong>The primary outcome was change from baseline to 8 weeks in nighttime agitation between 5 pm and 7 am on the Cohen-Mansfield Agitation Inventory Index, Direct Observation. Multivariable linear mixed effects regression models based on multiply imputed data were estimated on nighttime agitation and sleep, with treatment group, week, the 2-way interaction of group and week as predictors, and mean arterial pressure as a covariate based on baseline group imbalances.</p><p><strong>Results: </strong>Mean age ± SD was 83.4 ± 9.1 years. Most were female (72.0%), White (92.3%), non-Hispanic (84.6%), and lived in nursing homes (76.9%). Nighttime agitation, by group over time, was significant at 8 weeks (estimate, -1.67; P = .003) and 2 weeks. Total sleep time (actigraphy) by group over time was significant at 8 weeks (estimate, 48.45; P = .026). Observed nighttime wake by group over time was significant at 2 (estimate, -12.54; P = .006) and 8 weeks (estimate, -11.12; P = .015). The number having ≥1 adverse events was 60 in the GEn group (81.1%) and 50 in the placebo group (68.5%); with 12 serious adverse events in placebo and 10 in the GEn group. The GEn group had a trend toward more falls (P = .066).</p><p><strong>Conclusions and implications: </strong>Our findings suggest a novel approach for nighttime agitation in persons with dementia: assessing for RLS and initiating interventions. Larger and longer trials are needed.</p>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":" ","pages":"105485"},"PeriodicalIF":4.2,"publicationDate":"2025-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143256108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The study aimed to develop a machine learning (ML) model to predict early postdischarge falls in older adults using data that are easy to collect in acute care hospitals. This may reduce the burden imposed by complex measures on patients and health care staff.
Design
This prospective multicenter study included patients admitted to and discharged from geriatric wards at 3 university hospitals and 1 national medical center in Japan between October 2019 and July 2023.
Setting and Participants
The participants were individuals aged ≥65 years. Of the 1307 individuals enrolled during the study period, 684 were excluded, leaving 706 for inclusion in the analysis.
Methods
We extracted 19 variables from admission and discharge data, including physical, mental, psychological, and social aspects and in-hospital events, to assess the main outcome measure: falls occurring within 3 months postdischarge. We developed a prediction model using 4 major classifiers, Extra Trees, Bernoulli Naive Bayes, AdaBoost, and Random Forest, which were evaluated using a 5-fold cross-validation. The area under the receiver operating characteristic curve (AUC) was used to evaluate predictive performance.
Results
Among the 706 patients, 114 (16.1%) reported a fall within 3 months postdischarge. The Extra Trees classifier demonstrated the best predictive performance, with an AUC of 0.73 on the test data. Important features included the Lawton Instrumental Activities of Daily Living scale, Clinical Frailty Scale (≥4 points), presence of urinary incontinence, 15-item Geriatric Depression Scale (≥5 points), and preadmission residence, all assessed at admission.
Conclusions and Implications
To our knowledge, this is the first study to develop an ML model for predicting early postdischarge falls among older patients in acute care hospitals. The findings suggest that this model could assist in developing fall-prevention strategies to ensure seamless transition of care from hospitals to communities.
{"title":"Machine Learning Prediction for Postdischarge Falls in Older Adults: A Multicenter Prospective Study","authors":"Yuko Takeshita RN, MSN , Mai Onishi RN, MSN , Hirotada Masuda MD, PhD , Mizuki Katsuhisa RN, MSN , Kasumi Ikuta RN, MSN , Yuichiro Saizen RN, MSN , Misaki Fujii RN , Misaki Kasamatsu RN, MSN , Nobuyuki Inaizumi RN , Yuzuki Maeizumi RN , Yoshinobu Kishino MD , Tsuneo Nakajima MD, PhD , Eriko Koujiya RN, PhD , Miyae Yamakawa RN, PhD , Yoichi Takami MD, PhD , Koichi Yamamoto MD, PhD , Yumi Umeda-Kameyama MD, PhD , Shosuke Satake MD, PhD , Hiroyuki Umegaki MD, PhD , Yasushi Takeya MD, PhD","doi":"10.1016/j.jamda.2024.105414","DOIUrl":"10.1016/j.jamda.2024.105414","url":null,"abstract":"<div><h3>Objectives</h3><div>The study aimed to develop a machine learning (ML) model to predict early postdischarge falls in older adults using data that are easy to collect in acute care hospitals. This may reduce the burden imposed by complex measures on patients and health care staff.</div></div><div><h3>Design</h3><div>This prospective multicenter study included patients admitted to and discharged from geriatric wards at 3 university hospitals and 1 national medical center in Japan between October 2019 and July 2023.</div></div><div><h3>Setting and Participants</h3><div>The participants were individuals aged ≥65 years. Of the 1307 individuals enrolled during the study period, 684 were excluded, leaving 706 for inclusion in the analysis.</div></div><div><h3>Methods</h3><div>We extracted 19 variables from admission and discharge data, including physical, mental, psychological, and social aspects and in-hospital events, to assess the main outcome measure: falls occurring within 3 months postdischarge. We developed a prediction model using 4 major classifiers, Extra Trees, Bernoulli Naive Bayes, AdaBoost, and Random Forest, which were evaluated using a 5-fold cross-validation. The area under the receiver operating characteristic curve (AUC) was used to evaluate predictive performance.</div></div><div><h3>Results</h3><div>Among the 706 patients, 114 (16.1%) reported a fall within 3 months postdischarge. The Extra Trees classifier demonstrated the best predictive performance, with an AUC of 0.73 on the test data. Important features included the Lawton Instrumental Activities of Daily Living scale, Clinical Frailty Scale (≥4 points), presence of urinary incontinence, 15-item Geriatric Depression Scale (≥5 points), and preadmission residence, all assessed at admission.</div></div><div><h3>Conclusions and Implications</h3><div>To our knowledge, this is the first study to develop an ML model for predicting early postdischarge falls among older patients in acute care hospitals. The findings suggest that this model could assist in developing fall-prevention strategies to ensure seamless transition of care from hospitals to communities.</div></div>","PeriodicalId":17180,"journal":{"name":"Journal of the American Medical Directors Association","volume":"26 2","pages":"Article 105414"},"PeriodicalIF":4.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142864686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号