CoChat
Sign Up
Cited by 8Emergency and Acute Care StudiesDementia and Cognitive Impairment ResearchGeriatric Care and Nursing Homes

Factors associated with recurrent emergency department visits among people living with dementia: A retrospective cohort study

Open access
Jones, Aaron, Watt, Jennifer A., Maclagan, Laura C., Swayze, Sarah, Jaakkimainen, Liisa, Schull, Michael J., Bronskill, Susan E.·2023·Journal of the American Geriatrics Society
DOI 10.1111/jgs.18457
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Research on factors associated with recurrent emergency department (ED) visits and their implications for improving dementia care is lacking. The objective of this study was to examine associations between the individual characteristics of older adults living with dementia and recurrent ED visits. Methods: We used health administrative databases to conduct a population‐based retrospective cohort study among older adults with dementia in Ontario, Canada. We included community‐dwelling adults 66 years and older who visited the ED between April 1, 2010, and March 31, 2019 and were discharged home. We recorded all ED visits within one year after the baseline visit. We used recurrent event Cox regression to examine associations between repeat ED visits and individual clinical, demographic, and health service use characteristics. We fit conditional inference trees to identify the most important factors and define subgroups of varying risk. Results: Our cohort included 175,863 older adults with dementia. ED use in the year prior to baseline had the strongest association with recurrent visits (3+ vs.0 adjusted hazard ratio (aHR): 1.92 (1.89, 1.94), 2vs.0 aHR: 1.45 (1.43, 1.47), 1vs.0 aHR: 1.23 (1.21, 1.24)). The conditional inference tree utilized history of ED visits and comorbidity count to define 12 subgroups with ED revisit rates ranging from 0.79 to 7.27 per year. Older adults in higher risk groups were more likely to live in rural and low‐income areas and had higher use of anticonvulsants, antipsychotics, and benzodiazepines. Conclusions: History of ED visits may be a useful measure to identify older adults with dementia who would benefit from additional interventions and supports. A substantial proportion of older adults with dementia have a pattern of recurrent visits and may benefit from dementia‐friendly and geriatric‐focused EDs. Collaborative medication review in the ED and closer follow‐up and engagement with community supports could improve patient care and experience.

AI snapshot

Written from the paper’s full text.

TL;DR

In a population-based retrospective cohort of 175,863 community-dwelling older adults with dementia in Ontario, Canada, 62% had at least one recurrent emergency department (ED) visit within one year of an index ED visit. History of prior ED use was the strongest factor associated with recurrent ED visits, and tree-based analysis identified subgroups with one-year revisit risk ranging from 42.5% to 94.7%.

Key findings

  • During one-year follow-up, 108,906 (62%) had at least one recurrent ED visit; 25,722 (15%) had two recurrent visits and 33,381 (19%) had three or more.
  • In adjusted recurrent-event Cox regression, prior ED visits had the strongest association: 3+ vs. 0 visits aHR 1.92 (1.89, 1.94); 2 vs. 0 aHR 1.45 (1.43, 1.47); 1 vs. 0 aHR 1.23 (1.21, 1.24). Congestive heart failure had the highest chronic-condition aHR (1.13 [1.12, 1.14]), antipsychotic use aHR 1.06 (1.05, 1.07), benzodiazepine use aHR 1.05 (1.04, 1.07), and cholinesterase inhibitor use was protective (aHR 0.91 [0.90, 0.92]).
  • The conditional inference tree selected history of ED visits and comorbidity count as the most powerful splitting variables, with age used once in the branch with no history of ED visits. The highest-risk terminal node (L) versus lowest-risk node (A) had HR 4.03 (3.94, 4.13); estimated one-year ED revisit risk was 94.7% vs. 42.5%, and estimated mean one-year ED revisits were 7.27 vs. 0.79.
  • High-risk groups had higher proportions of rural residents and lowest-income-quintile residents, and higher anticonvulsant, antipsychotic, and benzodiazepine use; cholinesterase inhibitor use decreased from 42.4% in group A to 19.6% in group L.

Methods

  • Population-based retrospective cohort using linked Ontario health administrative databases, including community-dwelling adults aged 66+ with dementia who visited the ED and were discharged home between April 1, 2010, and March 31, 2019; follow-up was one year or until death or nursing home placement.
  • Dementia was identified with a validated administrative algorithm (sensitivity 79.3%, specificity 99.1%, positive predictive value 80.4%).
  • Associations with recurrent ED visits were examined using gap-time Prentice-Williams-Peterson Cox regression; conditional inference trees identified important factors and risk subgroups, with sensitivity analyses using only the first recurrent ED visit and a subset with home care clinical assessment data.

Limitations

  • Stated in the paper: the validated administrative algorithm does not capture all older adults with dementia; detailed functional and cognitive data were available only for a nonrandom subset and recorded up to six months before baseline; race and ethnicity data were unavailable; individuals younger than 66 were excluded; findings are most generalizable to health systems with universal, single-payer coverage.
  • Evident from the text: this is an observational administrative-data study, so residual confounding is possible; the authors note that all effects were statistically significant given the cohort size, though most were modest; language data were available for only 32% of the cohort.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1111/jgs.18457

Cites

38 of 38 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 30 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    During the one-year follow-up period, 108,906 (62%) older adults had at least one recurrent emergency department visit.

    “During the one-year follow-up period, 108,906 (62%) older adults had at least one recurrent ED visit.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Two thirds (66%) of cohort members had an emergency department visit in the year preceding the baseline emergency department visit.

    “Two thirds (66%) of cohort members had an ED visit in the year preceding the baseline ED visit, and one third (33%) had been admitted to hospital in that period.”
    Quote in the paperNot a substantial claim
  • Findingown data

    In adjusted analysis, 2 versus 0 emergency department visits in the past year was associated with an adjusted hazard ratio of 1.45 (1.43, 1.47) for recurrent emergency department visits.

    “In the adjusted analysis, history of ED visits in the past year had by far the strongest association (3+ vs.0 aHR: 1.92 (1.89, 1.94), 2vs.0 aHR: 1.45 (1.43, 1.47), and 1vs.0 aHR: 1.23 (1.21, 1.24)).”
    Quote in the paperSubstantial claim
  • Findingown data

    There were 33,381 (19%) older adults with three or more recurrent emergency department visits.

    “There were 25,722 (15%) older adults with two recurrent visits and 33,381 (19%) with three or more.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Congestive heart failure had the highest adjusted hazard ratio among chronic conditions for recurrent emergency department visits (aHR 1.13 (1.12, 1.14)).

    “Congestive heart failure had the highest adjusted hazard ratio (aHR 1.13 (1.12, 1.14)) among chronic conditions, followed by chronic obstructive pulmonary disease (aHR 1.10 (1.09, 1.11)).”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. 19
  • May combine two claims. 8
  • The quote only partly states this claim. 2

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Cohort members had a median of two chronic conditions in addition to dementia.

    “In addition to dementia, cohort members had a median of two chronic conditions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    One third (33%) of cohort members had been admitted to hospital in the year preceding the baseline emergency department visit.

    “Two thirds (66%) of cohort members had an ED visit in the year preceding the baseline ED visit, and one third (33%) had been admitted to hospital in that period.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There was a gradient of increasing family physician use, specialty physician use, and home care clinical assessments across risk groups.

    “There was a gradient of increasing family physician use, specialty physician use, and home care clinical assessments across risk groups.”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    Among community-dwelling older adults with dementia, there was high use of antipsychotics, anticonvulsants, and benzodiazepines.

    “our finding on the high use of antipsychotics, anticonvulsants, and benzodiazepines”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The conditional inference tree fit on the first recurrent emergency department visit was similar to the main tree while additionally using receipt of a home care clinical assessment to split a single node.

    “The conditional inference tree fit on the first recurrent ED visit was similar to the main tree while additionally using receipt of a home care clinical assessment to split a single node (Figure S1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    No other variables were selected except age, which was used once in the branch of the tree with no history of emergency department visits.

    “No other variables were selected except age, which was used once in the branch of the tree with no history of ED visits.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Findings were reported according to STROBE guidelines.

    “We reported our findings according to STROBE guidelines.28”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [28] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Conditional inference trees select splitting variables and cutpoints using global and local statistical tests to provide unbiased variable selection.

    “However, the conditional inference tree selects splitting variables and cutpoints using a series of global and local statistical tests to provide unbiased variable selection.32”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [32] source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductcitation

    The dementia algorithm achieved sensitivity 79.3%, specificity 99.1%, and positive predictive value 80.4% in Ontario.

    “In Ontario, the algorithm was found to outperform other claims-based formulations and achieved a sensitivity of 79.3%, a specificity of 99.1% and a positive predictive value of 80.4%.26”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [26] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The dementia algorithm outperformed other claims-based formulations in Ontario.

    “In Ontario, the algorithm was found to outperform other claims-based formulations and achieved a sensitivity of 79.3%, a specificity of 99.1% and a positive predictive value of 80.4%.26”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [26] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The minimum terminal node size was restricted to 5% of the cohort to ensure stability of estimates within each terminal node.

    “To ensure stability of estimates within each terminal node, we restricted the minimum terminal node size to 5% of the cohort.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Prentice-Williams-Peterson regression is recommended when occurrence of the first event increases the likelihood of subsequent events.

    “This stratified approach enables each event recurrence to have a distinct baseline hazard and is recommended when occurrence of the first event increases the likelihood of subsequent events.31”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [31] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Existing research found higher emergency department visit rates among people living with dementia with higher comorbidity burden.

    “Existing research on associations between individual characteristics of people living with dementia and general use of the ED have found higher visit rates among those with individuals with more severe neuropsychiatric symptoms, higher comorbidity burden, and those living in rural locations.18,19”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [18] source

      Not checked in the free run.

    • cites [19] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Medication reviews with pharmacist and/or geriatrician involvement can improve medication appropriateness measures.

    “Medication reviews with pharmacist and/or geriatrician involvement can improve medication appropriateness measures, 49”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [49] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The benefits of medication reviews to future health service use are unclear.

    “though benefits to future health service use are unclear.50”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [50] source

      Not checked in the free run.

  • May combine two claims. The quote only partly states this claim.

    Interpretationown data

    The results of this study suggest that lack of access to primary care and/or continuity with primary care contributes to high emergency department visitation among older adults with dementia.

    “and there is some suggestion of that in our results.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Some older adults prefer visiting the emergency department even when they have a primary care provider.

    “there is existing evidence that some older adults prefer visiting the ED, even when they have a primary care provider.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [42] source

      Not checked in the free run.

    • cites [43] source

      Not checked in the free run.

    • cites [44] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The association between high risk of recurrent ED use and high family physician use has been noted previously in other populations.

    “This has been noted previously in other populations”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [40] source

      Not checked in the free run.

    • cites [41] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    There is heterogeneity in visit reasons across serial emergency department visits.

    “Given our previous research demonstrating heterogeneity in visit reasons across serial ED visits, 13”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [13] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationown data

    Past emergency department use would be a useful measure to identify patients who would benefit from closer follow-up, geriatric referral, emergency department-to-community transition support, and closer engagement with primary care and home care.

    “The relative strength of the association we observed with past ED visits is informative, suggesting that past ED use would be useful measure to identify patients who would benefit from closer follow-up, geriatric referral, ED-to-community transition support, and closer engagement with primary care and home care.36”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [36] source

      Not checked in the free run.

    • Rests on an assumption: Recurrent emergency department visits may be a result of unmet healthcare needs and indicate opportunities to improve care.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    Individuals may be less likely to visit the emergency department frequently if they do not have health insurance or their insurance requires deductibles or copayments.

    “Individuals may be less likely to visit the ED frequently if they do not have health insurance or their insurance requires deductibles or copayments.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Detailed functional and cognitive data were available only for a nonrandom subset of the cohort and were recorded up to six months before the baseline event.

    “Detailed functional and cognitive data were only available for a nonrandom subset of our cohort and was recorded up to six months prior to the baseline event.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Rural residents and residents of lower income areas are populations known to have lower access to primary care.

    “populations known to have lower access to primary care, 39”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [39]

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Race and ethnicity data were unavailable in the study data sources.

    “Race and ethnicity data were unavailable.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    The factors that underlie recurrent emergency department use likely go beyond medical problems, involving formal and informal support networks, living arrangements, and caregiver burnout.

    “However, the factors that underlie recurrent ED use likely go beyond medical problems, involving formal and informal support networks, living arrangements, and caregiver burnout.”
    Quote in the paperNot a substantial claim
25 of 25

  • The paper relies on this without support.

    Research focused on the factors associated with recurrent emergency department visits and their implications for improving dementia care is lacking.

    “However, research focused on the factors associated with recurrent ED visits and their implications for improving dementia care is lacking.”

  • The paper relies on this without support.

    Recurrent emergency department visits may be a result of unmet healthcare needs and indicate opportunities to improve care.

    “Recurrent ED visits may be a result of unmet healthcare needs and indicate that there are opportunities to improve care.”
2 of 2

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • In practice, admission and discharge procedures prioritise the admission diagnosis and do not adequately account for dementia symptoms (e.g. cognitive decline, BPSD).

    “Our findings suggest that in practice however, admission and discharge procedures prioritise the admission diagnosis, and do not adequately account for dementia symptoms (e.g. cognitive decline, BPSD) [47, 49, 56, 60, 76, 77, 90, 93].”

    From Do hospital-to-home transitions work for older adults with multiple long-term conditions including dementia? A realist review (2025) · cites this paper as [93]

    The Discussion (page unavailable) identifies opportunities to improve detection of cognitive impairment in the ED, tailor care plans, and coordinate follow-up after discharge. It does not say that admission and discharge procedures prioritize the admission diagnosis or specifically fail to account for behavioral and psychological symptoms of dementia (BPSD). (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • Fragmentation between health and social care contributed to reduced responsibility for healthcare professionals in emergency departments to support social and non-urgent needs, likely increasing the patient's risk of under-triage.

    “Additionally, fragmentation between health and social care contributed to reduced responsibility for HCPs in emergency departments to support social and non-urgent needs, likely increasing the patient’s risk of under-triage (CMOC13) [47, 49, 56, 60, 76, 77, 82, 90, 93, 94].”

    From Do hospital-to-home transitions work for older adults with multiple long-term conditions including dementia? A realist review (2025) · cites this paper as [93]

    The Discussion (page not provided) recommends better coordination between the ED and community care to improve transitions, but does not say that fragmentation reduced ED professionals’ responsibility for social or non-urgent needs or increased under-triage. Searches for fragmentation, responsibility, and under-triage found no passage addressing that proposed chain of effects. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

  • Elevated BPSD often masks the patient's care needs, leading to misrecognition and inaccuracies in the care provided by healthcare professionals.

    “Elevated BPSD often masks the patient’s care needs, leading to misrecognition and inaccuracies in the care provided by HCPs (CMOC18) [45, 72, 77, 93].”

    From Do hospital-to-home transitions work for older adults with multiple long-term conditions including dementia? A realist review (2025) · cites this paper as [93]

    The Introduction (page unavailable) says recurrent ED visits may reflect unmet healthcare needs, and the Discussion (page unavailable) recommends better detection of cognitive impairment and care tailored to patients’ priorities. Neither passage says that elevated BPSD masks care needs or causes healthcare professionals to misrecognize them and provide inaccurate care; searches for BPSD and that proposed mechanism found no such passage. (LLM, conf=0.74)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Terminal node is smaller than the stated 5% minimum

    Minor

    The methods state that each terminal node had to contain at least 5% of the cohort, but Node L contains 4.2%.

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Report summarys: Prevalence and monetary costs of dementia in Canada, 2016, A report by the Alzheimer Society of Canada Alzheimer Society of Canada, Health Promot Chronic Dis Prev Can, 2016, 231

Cited by

Sorted by citations

Related

Authors
Jones, A., Watt, J. A., Maclagan, L. C., Swayze, S., Jaakkimainen, L., Schull, M. J., & Bronskill, S. E.
Affiliations
  • Institute for Clinical Evaluative Sciences (CA)
  • St. Michael's Hospital (CA)
  • Sunnybrook Health Science Centre (CA)
Corresponding author
Jones, Aaron
ORCID
Venue
Journal of the American Geriatrics Society
Publisher
Wiley
Published
2023-06-14
Volume
71
Issue
12
Pages
3731-3743
ISSN
  • 0002-8614 (ISSN-L)
  • 1532-5415
DOI
10.1111/jgs.18457
OpenAlex
W4380576786
Semantic Scholar
f2df92e1bdeae892a67dfe05c39ea430382936d4
Cited by
8
References
38
Citations by year
2025: 6 · 2024: 1 · 2023: 1
FWCI
2.46
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of the American Geriatrics Society (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
Funders
  • Institute for Clinical Evaluative Sciences
  • Ontario Brain Institute
Topics
  • Health Sciences › Medicine › Emergency Medicine › Emergency and Acute Care Studies
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Health Sciences › Health Professions › General Health Professions › Geriatric Care and Nursing Homes
Keywords
dementia, emergency department utilization, older adults, comorbidities, antipsychotic drugs, benzodiazepines, antiepileptic drugs, rural residence, low-income areas, geriatric emergency care, medication review, risk subgroups
MeSH terms
Aged, Dementia, Emergency Service, Hospital, Humans, Ontario, Retrospective Studies

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review