Table 2.

Summarized Evidence of Included Studies

Author (year)	Title/objective	Design, setting, population, and sample	Results	Strengths/limitations	Crowe Score (out of 40)
Epstein-Lubrow et al. (2015)	Hospice referral after inpatient psychiatric treatment of individuals with advanced dementia from a nursing home To assess discharge disposition following inpatient psychiatric treatment for advanced dementia	Design: A descriptive analysis, brief report utilizing MDS Setting: Medicare minimum data set (MDS) data of nursing home (NH) residents Sample: 1,027 NH decedents	In the last 90 days of life, 1,027 (0.15%) persons received inpatient psychiatric treatment before death and place of death was determined. The discharge dispositions included hospital (n = 132, 12.9%), NH without hospice (n = 728, 70.9%), NH with hospice (n = 73, 7.1%), home without hospice (n = 32, 3.1%), and hospice at home (n = 16, 1.6%). The rate of transition as a result of referral to hospice services for advanced dementia was low	Strengths: Large sample size and robust analyses for examining transition to hospice and hospice referral after inpatient psychiatric treatment Limitations: There was limited information about factors that contribute to discharge disposition with or without hospice services	33
Ernecoff et al. (2018)	Concordance between Goals of Care (GOC) and treatment decisions for persons with dementia To describe family decision-makers’ preferred GOC and perceptions of goal-concordant care for NH residents with late-stage dementia	Design: Secondary data analysis of baseline and 9-month family decision-maker interviews and chart reviews. This study compared subsequent treatments when families chose a primary goal of comfort versus other goals Setting: 22 NHs in North Carolina Sample: 302 dyads of NH residents and family decision-makers	Families reported their GOC at baseline and in follow-up interviews. The treatment orders, hospice, and hospital transfers were compared after selecting comfort as the primary goal versus other goals of care. Families chose comfort for 66% of residents at baseline, and 80% at 9 months or death. 49% were in concordance with the nursing home staff on the primary goal at baseline and that increased to 69% at the follow-up	Strengths: Prospective data collection and validated outcome measures Limitations: Findings were exploratory with limited generalizability due to the study being focused in one state. The use of retrospective chart reviews could not capture care decisions and outcomes	34
Gozalo et al. (2011)	End-of-life transitions among nursing home residents with cognitive issues To examine healthcare transitions among Medicare decedents with advanced cognitive and functional impairment who were nursing home residents 120 days before death	Design: Retrospective study that linked Medicare Minimum Data Set (2000 through 2007) Setting: NH Sample: 474,829 NH decedents	19% of the sample population had one or more burdensome transitions. Black and Hispanic individuals (and those without an advance directive) were at a higher risk. Longer time in ICU, feeding tube, stage IV ulcer, and late enrollment in hospice were associated with nursing home residents within the highest quintile of transitions	Strengths: Study revealed predictors of burdensome transition and factors that led to later hospice transition Limitations: Lack of information on patient preferences; Unable to assess end-of-life quality measures and outcomes	35
Hanson et al. (2019)	Triggered palliative care for late-stage dementia: A pilot randomized trial To test dementia-specific specialty palliative care triggered by hospitalization	Design: Randomized controlled trial Setting: Hospital Sample: 62 dyads of persons with late-stage dementia and family decision-makers	Intervention was feasible and completion of the protocol ranging between 77% (family two-week call) and 93% from the initial consultation. There was no difference among the hospital and emergency department visits in the intervention versus control. The intervention group had more of the palliative care domains addressed and was more likely to get hospice. The intervention groups were more likely to discuss goals of care, prognosis, and decisions to avoid hospitalization. Consultations were found to result in improved decision-making and outcomes	Strengths: Strong addition to a limited field of research and promising results to foster research and intervention development for triggered palliative care Limitations: Short duration of specialty palliative care; enrollment was limited due to short hospital stays and caregiver stress	37
Lackraj et al. (2021)	Implementation of specialist palliative care and outcomes for hospitalized patients with dementia To determine whether implementation of hospital-based specialist palliative care was associated with differences in treatment intensity outcomes for hospitalized patients with dementia	Design: Retrospective cohort study Setting: 51 hospitals in New York that either did or did not implement a palliative care program between 2008 and 2014 Sample: 82,068 hospitalized persons with dementia	82,118 patients diagnosed with dementia experienced acute hospitalization. Among them, 41,227 individuals (equivalent to 50.27%) received treatment in hospitals that had integrated a palliative care program. In contrast to patients treated in hospitals lacking palliative care services, those with dementia who received care in hospitals after the introduction of palliative care were 35% more likely to be discharged to hospice	Strengths: Additional evidence that the utilization of hospice care is a movable outcome of specialist palliative care Limitations: Differences between hospitals and lack of information about the timing and specifics of palliative care delivery such as number and availability of palliative care providers; institutional cultural differences related to palliative care	36
Miller et al. (2017)	Specialty palliative care consultations for nursing home residents with dementia To examine the impact of expanded palliative care access for NH residents with moderate to severe dementia	Design: Retrospective cohort design; NH consultation data from two palliative care organizations Setting: North Carolina and Rhode Island Sample: 505 NH residents with moderate-to-very severe dementia who died between 2006 and 2010 across 31 NH	With earlier consultation (vs no consultation), hospitalization rates in the seven days before death were on average 13.2 percentage points lower and with later consultation 5.9 percentage points lower. The consultations appeared to decrease acute care and burdensome transitions without PLWD spending additional Medicare. Earlier consultations showed greater reductions	Strengths: Matched controls in a similar time; examination of outcomes during the last 30 days of life which minimized bias Limitations: Unable to see the association between palliative care consultations and use of acute care versus hospice; unable to match nursing homes and limited generalizability	36.5
Moyo et al. (2022)	Effect of a video-assisted advance care planning intervention on end-of-life health care transitions among long-stay nursing home residents To determine the relationship between an advance care planning (ACP) video intervention, Pragmatic Trial of Video Education in Nursing Homes (PROVEN), and end-of-life healthcare transitions among long-stay nursing home residents with advanced illness	Design: Pragmatic cluster randomized clinical trial Setting: February 2016 to May 2019 in 360 nursing homes (119 intervention, 241 control) Population: Long-stay NH residents 65 years who died during the trial who had advanced dementia	Video topics included goals of care, hospice transitions, hospitalization, and advance care planning Decedents with advanced illness in the intervention group has less hospital transfer in the last 90 days of life than the control arm. There was no difference between the two trial arms when comparing the number of decedents with multiple hospital transfers or late transitions to hospital or hospice	Strengths: Promising intervention that could inform future efforts to reduce late transitions and burdensome care Limitations: Trend of reduced hospital transfer rates during the trial; lack of information on resident or family preferences; low intervention fidelity	34.5
Oud (2017)	Predictors of transition to hospice care among hospitalized older adults with a diagnosis of dementia in Texas: A population-based study To identify factors associated with hospice use in the inpatient setting and patterns of escalation of care to ICU setting preceding discharge to hospice	Design: Retrospective, population-based cohort study. Setting: Texas, United States (Inpatient Public Use Data File TIPUDF (2001–2010)) Sample: 889,008 hospitalized persons with dementia	40,669 persons with dementia were discharged to hospice. Hospice transitions increased between 2001 and 2010, 908 to 7,398 respectively. These involved admission to the ICU prior to discharge in 45.2% of them. Organ failure development, number of organs in failure, and mechanical ventilation use were prevalent. Additionally, minoritized individuals, uninsured, or noncommercial primary insurance holders were less likely to discharge to hospice	Strengths: Robust data set with a diverse population; population-based approach to analysis Limitations: Dementia severity could not be identified; other clinical contributors to hospice transitions (procedures, severity of illness, etc.) could not be examined	34.5
Prater et al. (2023)	Hospice transitions from the perspective of the caregiver: A qualitative study and development of a preliminary hospice transition checklist To explore caregiver experiences and perceptions on the transition to hospice and to develop a preliminary checklist for hospice transition	Design: Qualitative descriptive study using semistructured interviews and descriptive inductive/deductive thematic analysis Setting: United States Sample: 19 adult caregivers of patients had enrolled in hospice and died in the year prior (January to December 2019)	Four themes that were identified for the framework: hospice intake, preparedness, burden of care, and hospice resources. A checklist was created that included factors that caregivers thought could contribute to successful hospice care transitions and how to improve the transitional process	Strengths: An initial checklist was created for hospice agencies to improve their transition process; can be modified for future intervention development Limitations: Lack of racial and ethnic diversity of sample	35
Sullivan, de Rosa, et al. (2022)	Dementia caregiver burdens predict overnight hospitalization and hospice utilization To determine sociodemographic and caregiver burdens associated with overnight hospitalization, hospice utilization, and hospitalization frequency among persons with dementia	Design: Cross-sectional analysis using National Health and Aging Trends Study (NHATS) and National Study of Caregiving (NSOC) data. Logistic and proportional odds regression Setting: United States Sample: 899 persons with dementia and caregivers in 2015	Caregivers financial difficulty or felt like their burden was too much to management, persons with dementia were 5.60 times more likely to have overnight hospitalizations. Caregivers that had emotionally difficulty resulted in the person with dementia being 5.89 times more likely to use hospice than caregivers that did not report having emotional difficulty	Strengths: Robust data; subjective care burden factors associated with use of hospice in the national data set linking their primary caregivers in dyads were studied Limitations: Overnight hospitalization and use of hospice could have been affected by the presence of other caregivers; unable to know other factors that contributed to hospice use as a result of caregiver burden (frequency of care responsibilities, behavioral symptoms of person with dementia, etc.)	35.5
Sullivan, Bo, et al. (2022)	Predicting hospice transitions in dementia caregiving dyads: An exploratory machine learning approach To identify quality of life and social determinants of health factors associated with hospice use among people living with dementia and their caregivers	Design: Secondary data analysis of the NHATS and NSOC data using multiple machine learning Setting: United States Sample: 117 PLWD and primary caregivers	Factors associated with hospice use included comorbid diabetes, a regular physician, a good memory rating, not relying on food stamps, not having chewing or swallowing problems, and whether health prevents them from enjoying life. Primary caregivers’ age, income, census, number of days of help provided by caregiver per month, and whether health prevents them from enjoying life was associated with hospice use	Strengths: Exploratory models created a starting point for future research; precision health approaches Limitations: Unable to assess for barriers to care; unclear applicability of research findings to practice	36
Toles et al. (2022)	Adapting the Connect-Home transitional care intervention for the unique needs of people with dementia and their caregivers: A feasibility study To assess fidelity, acceptability, preliminary outcomes, and mechanisms of the Connect-Home ADRD transitional care intervention	Design: Feasibility study Setting: Two skilled nursing facilities in the Southeastern United States Sample: 19 persons living with dementia and caregiver dyads	Intervention was shown as feasible and acceptable as dyads stated it was easy to use and helpful. The Connect-Home ADRD adaptations included home support for management of symptoms of dementia and support for the transition to hospice	Strengths: Acceptability and feasibility of the intervention; use of structured interviews with persons living with dementia and caregivers Limitations: Sample size; limited racial and ethnic diversity; unable to test for efficacy	34
Wiggins et al. (2019)	Understanding the factors associated with patients with dementia achieving their preferred place of death: A retrospective cohort study To understand preferences for place of death among people with dementia; to identify factors associated with achieving these preferences	Design: Retrospective cohort study Setting: Coordinate My Care (Electronic Palliative Care Coordination System) Setting: 1,047 persons with dementia who died between December 2015 and March 2017	803 decedents had information on the preferred and actual place of death. The care home or home was the most common preferred place of death at 58.8% (n = 474). 83.7% (n = 672) actually died in their preferred place and this was more likely for those that were most functionally impaired. When the goals of treatment focused on symptomatic relief, more people with dementia were able to die in their preferred place of death	Strengths: Preferences of the person with dementia were available; preferred place of death examined; large sample Limitations: Limited generalizability to other populations; other variables unknown such as place of residence and the presence of caregivers	33.5
Yeh et al. (2021)	Improving end-of-life care for persons living with dementia: Bereaved caregivers’ observations and recommendations To assess satisfaction with and elicit recommendations for improving end-of-life care experiences of persons with dementia from the perspective of bereaved caregivers	Design: Cross-sectional study (survey) Setting: California, United States Sample: 53 caregivers of decedent persons with dementia	Caregivers felt that there were gaps present in provider communication about progression of disease continuity of care and coordination at the end of life prior to transitions and preferred location of death. Most caregivers shared views that providers should talk about the end-of-life decisions and care with persons with dementia during the earlier stages of the disease	Strengths: Caregiver perspectives were garnered specific to end-of-life care and decisions; evidence that earlier conversations about transitions and end-of-life care may be desired Limitations: Potential for recall bias due to cross-sectional interviews of bereaved caregivers; limited racial/ethnic diversity	35

Notes: ICU = intensive care unit; PLWD = persons living with dementia.