“What Matters” in the Emergency Department

Abstract

Objective:

To describe “What Matters” to older adults seeking emergency department (ED) care and to identify patient characteristics associated with meeting desired outcomes.

Background:

As part of the 4Ms framework, identifying “What Matters” has been captured across healthcare settings, yet limited attention has been directed to older adults in the ED.

Methods:

We performed a secondary analysis of a multicenter prospective observational study. The study enrolled 1013 patients aged 65 and older during an ED encounter and performed 90-day follow-up interviews. The primary outcome was the quantification of “What Matters” (concerns and desired outcomes) to older adults during emergency care. As secondary outcomes, we assessed concerns at day 90, if desired outcomes were met at follow-up, and patient characteristics associated with outcome achievement by estimating multivariable logistic regression models.

Results:

Older adults reported specific concerns at the time of the ED visit including: (1) symptom identification and/or persistence (31.4%), (2) ability to take care of oneself (19.4%), and (3) end-of-life (17.8%). Desired outcomes expressed by participants included: (1) getting well and symptom resolution (72.0%), (2) obtaining a diagnosis (25.7%), and (3) functional independence (19.5%). At day 90 follow-up, concerns remained similar to the initial ED visit, and the majority of participants (66.2%) achieved their desired outcome. Frailty (adjusted odds ratio = 0.56, 95% CI: 0.38–0.83, P < 0.01) was associated with older adults not meeting their desired outcomes.

Conclusions:

Older adults identified a variety of concerns during emergency care, and two-thirds reported that desired outcomes were met at longitudinal day 90 follow-up.

America’s population is rapidly aging and commonly seeking acute unscheduled care. Older adults, individuals ages 65 years and older, comprise 16.8% of the U.S. population and account for ∼20% of all emergency care, amounting to over 29 million emergency department (ED) visits annually in the United States.¹ Due to their complex care needs, older adults have been noted to face unique challenges related to emergency care, including the potential receipt of goal-discordant care and decreased attention to patient-centered care.^2–5 Much of the available ED research regarding older adults’ patient-centered goals has focused on treating clinicians performing end-of-life goals of care conversations.^6–8 However, ED treating clinicians are tasked with navigating older adult priorities and desired outcomes not just during critical illness or end-of-life.⁹

As a solution to address the underlying problem of clinicians potentially providing goal-discordant care to older adults, the John A. Hartford Foundation and the Institute for Healthcare Improvement founded the Age-Friendly Health Systems initiative in 2017.^10–13 As of July 2023,¹⁴ there were over 3000 age-friendly health system participants employing the “4Ms” Framework– Mobility, Mentation, Medication, and “What Matters” – to ensure patient-centered and evidence-based care for older adults across health care settings.^15–17 The 4Ms Framework aims to support patients and clinicians when developing care plans by focusing more on the needs, goals, and priorities of the patient, rather than solely treating their acute and chronic illnesses.¹⁵ This holistic view of care is intended to be person-specific and adaptable to an individual’s needs, home situation, and support. Identifying and acting on “What Matters” most to older adults has been the most challenging of the 4Ms, and a critical knowledge gap particularly exists in the ED setting.^18–20 Pilot work has shown the feasibility of incorporating “What Matters” questions addressing older adults’ concerns and desired outcomes within the ED setting,^9,21–23 yet large-scale quantitative data are needed to confirm these preliminary findings and to investigate possible existing concerns that health inequities are present among certain sociodemographic (eg, race, ethnicity, and insurance status) populations.

The objectives of this study were to describe “What Matters” to older adults seeking ED care and to identify patient characteristics associated with older adults meeting their desired outcomes. Emphasizing patient-centered, goal-concordant care for older adults visiting the ED through a focus on “What Matters” has the potential to ensure that the patients’ needs are prioritized during ED-to-community care transitions.

METHODS

Study Design and Setting

We conducted a secondary analysis of data prospectively collected as part of the Geriatric Emergency care Applied Research Standardization Study (GEARSS) funded by the National Institute on Aging. The GEARSS was a multicenter observational cohort study that enrolled older adults (≥65 y of age) seen within 5 EDs, including the Grady Memorial Hospital (Atlanta, GA), the Northwestern Memorial Hospital (Chicago, IL), and 3 EDs within the Yale New Haven Health System (New Haven, CT). Enrollment occurred between March 2021 and June 2022. Enrollees included patients admitted to inpatient or observation status as well as those discharged from the ED. Institutional Review Boards at each of the recruiting sites approved the study protocol. Our study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.²⁴

Eligible ED patients were English-speaking, were older than 65 years of age, and could be contacted by phone for follow-up. GEARSS excluded participants who were incarcerated, had decisional impairment without a legally authorized representative, were on a psychiatric hold, were clinically unstable for study recruitment as determined by the treating ED clinicians, or were in respiratory isolation. Legally authorized representatives, as defined by local and state statutes, were sought for eligible participants whose impaired decision-making capacity could compromise the consent process. In addition, legally authorized representatives were able to assist the participant in answering questions (during initial and follow-up interviews).

Data Collection and Outcomes

GEARSS research staff from all sites received identical training from the Yale Program on Aging to prospectively collect subsequent data on a wide array of prioritized topics. The GEARRS study data collection forms included questions regarding “What Matters,” care transitions, cognitive impairment, falls/mobility, frailty, elder abuse, and medication prescription safety. Local research staff performed in-person assessments during the index ED visit (day 0) and recorded follow-up by telephone on day 90.

Before incorporation within the GEARSS data collection forms, the “What Matters” questions for the ED were developed and tested by a group of clinicians, patients, caregivers, health system leaders, payers, and funders to address the “What Matters” element of the 4Ms framework as part of a Patient Priorities Care initiative.²⁵ Two of the questions from the “Condensed Conversation Guide for Identifying Patient Priorities (The One Thing) Emergency Department version” were selected (concerns and desired outcomes), and standardized answers for patients to choose from were added by the GEARSS study team, plus an “Other” answer option (Table 1). At 90 days, the question about health concerns was repeated and participants were reminded of their index ED visit desired outcome. The patients were then asked if their desired outcome(s) had been achieved. If patients chose “Other,” study staff would prompt patients to provide a specific answer which the staff recorded as free text. In total, 751 (74.1%) and 432 (42.6%) participants selected “Other” and provided free text information, respectively, for day 0 health concerns and desired outcomes.

TABLE 1.

“What Matters” Questions Asked at Days 0 and 90

“What Matters” Questions – day 0

1. What concerns you most when you think about your health and about being in the ED today/tonight? Choose all that apply:  a. Ability to take care of myself  b. Getting my medication  c. End of life  d. Finances  e. Appointments  f. Other 2. What outcomes are you most hoping for after this ED visit? Choose all that apply:  a. Getting well  b. Understanding my diagnosis  c. Self-care  d. What’s next  e. Other

“What Matters” Questions – day 90 follow-up

1. When you started this study, you stated (baseline response) concerned you the most about your health and being in the ED at that time (90 d ago). What concerns you most now about your health? Choose all that apply.  a. Ability to take care of myself  b. Getting my medication  c. End of life  d. Finance  e. Appointments  f. Other 2. When you started this study, you stated that (the baseline response) was the outcome you were hoping for after your ED visit. Do you feel that outcome happened (was achieved)?  a. Yes  b. No

ED indicates emergency department.

The primary outcome was the quantification of “What Matters” (concerns and desired outcomes) to older adult patients during emergency care. As secondary outcomes, we, in addition, assessed whether concerns remained stable or changed at 90-day follow-up, whether desired outcomes were met at 90 days, and whether outcome achievement was associated with available risk factors.

Data Analyses

We first performed descriptive analyses, identifying the proportion of older adults selecting concerns and desired outcomes from the categorical choices (Table 1). Responses were not required to be mutually exclusive, and respondents could provide free-text responses. Two members of the team (T.C. and C.J.G.) independently reviewed all free-text responses and categorized them within existing domains or identified new thematic response domains if several respondents provided similar concerns or desired outcomes under the “Other” free-text response. Once all free-text responses were reviewed, a second review was conducted to group similar domains with preexisting ones. Data management decisions regarding the categorization of concerns and desired outcomes are included (Supplemental Table, Supplemental Digital Content, http://links.lww.com/MLR/C876). We then identified associations between outcome achievement at 90-day follow-up and risk factors by estimating multivariable logistic regression models. The selection of variables was based on available data, existing literature, and consensus agreement. The patient-level variables included were sex, race, ethnicity, insurance payor, marital status, cognitive impairment, primary care provider status, frailty, and baseline activities of daily living (ADL) and instrumental ADL independence.^9,26,27 The 6-item Katz ADL Index was used to document a patient’s self-reported ability to complete daily activities.²⁸ The 8-item Lawton Instrumental ADL Index was used to document more complex self-reported activities of independent living.²⁹ Frailty was defined by the Clinical Frailty Score (≥5).³⁰

Statistical analysis was completed using Stata Version 16.1 (StataCorp).

RESULTS

Sample Characteristics

GEARRS study staff enrolled 1013 older adult ED patients, 667 (65.8%) of whom completed 90-day follow-up phone calls. Those who completed follow-up were similar regarding key characteristics to those who did not complete follow-up. Participants were primarily female (57.0%), White (56.9%), and had a mean age of 75.1 years (Table 2). Most (68.3%) reported having a primary care provider. Few (3.3%) had a formal diagnosis of dementia. More than half (57.2%) screened positive for frailty.

TABLE 2.

Characteristics of Study Population

Characteristic	Study population (N = 1013)	Population with 90 d follow-up (N = 667)
Age, mean (SD)	75.1 (7.7)	74.2 (6.9)
Sex (M), n (%)	436 (43.0)	288 (43.2)
Race
White	576 (56.9)	406 (60.9)
Black or African American	398 (39.3)	234 (35.1)
Asian American	14 (1.4)	11 (1.7)
Other	25 (2.5)	16 (2.4)
Ethnicity
Hispanic/Latino	37 (3.7)	28 (4.2)
Non-Hispanic/Latino	967 (95.5)	632 (94.8)
Not answered	9 (0.9)	7 (1.0)
Medicaid insurance, n (%)	228 (22.5)	138 (20.7)
Married, n (%)	384 (37.9)	285 (42.7)
Dementia diagnosis, n (%)	33 (3.3)	13 (2.0)
Has PCP, n (%)	692 (68.3)	545 (81.7)
Frail, n (%)	579 (57.2)	329 (49.3)
Baseline ADL independence score (0–6), mean (SD)	5.46 (1.19)	5.57 (1.00)
Baseline IADL independence score (0–8), mean (SD)	6.38 (2.19)	6.77 (1.87)

ADL indicates activities of daily living; IADL, instrumental activities of daily living; PCP, primary care provider.

“What Matters” – Concerns

While seeking ED care, the 3 most common concerns about their health and being in the ED selected by older adults were: (1) symptom identification and/or persistence (31.4%), (2) the ability to take care of oneself (19.4%), and (3) end-of-life concerns (17.8%). At 90 days after the ED visit, the 3 most common concerns about their health expressed by older adults were: (1) the ability to take care of oneself (32.5%), (2) obtaining medications and managing chronic conditions (24.0%), and symptom identification and/or persistence (22.2%; Table 3).

TABLE 3.

Patients’ Concerns and Desired Outcomes During and After Emergency Care

Concern number	Concerns at day 0 (ED visit)	N (%)	Desired outcomes at day 0 (ED visit)	N (%)	Concerns at day 90 follow-up	N (%)
1.	Symptom identification and/or persistence	318 (31.4)	Getting well and symptom resolution	729 (72.0)	Ability to take care of myself	217 (32.5)
2.	Ability to take care of myself	197 (19.4)	Obtaining a diagnosis	260 (25.7)	Getting my medication and chronic conditions	160 (24.0)
3.	End of life	180 (17.8)	Functional independence	198 (19.5)	Not answered or no concerns	148 (22.2)
4.	Getting my medication and chronic conditions	112 (11.1)	Going home	180 (17.8)	Symptom identification and/or persistence	144 (21.6)
5.	Appointments	94 (9.3)	Self-care	142 (14.0)	End of life	111 (16.6)
6.	Not answered or no concerns	76 (7.5)	Expectation setting	113 (11.2)	Appointments	33 (4.9)
7.	Wait times and ED experience	57 (5.6)	Reassurance	77 (7.6)	Family/friends/pets	20 (3.0)
8.	Finances	37 (3.7)	Missing	32 (3.2)	COVID-19	14 (2.1)
9.	Family/friends/pets	35 (3.5)	Other responses not categorized above	31 (3.1)	Wait times and ED experience	9 (1.3)
10.	Return home	28 (2.8)	Not answered or no concerns	12 (1.2)	Finances	8 (1.2)
11.	Missing	28 (2.8)	—	—	Return home	5 (0.7)
12.	COVID-19	20 (2.0)	—	—	Other responses not categorized above	4 (0.6)
13.	Other responses not categorized above	8 (0.8)	—	—	—	—

Each desired outcome category is a combination of survey categorical selections as well as qualitative synthesis of notes documented in the “Other” response field.

1013 participants responded at day 0 and 667 participants responded at day 90.

Columns may not add to 100% as responses were not mutually exclusive.

ED indicates emergency department.

Participant responses to “What Matters” questions at day 90 had changed from their identified concerns during the index ED visit (Fig. 1). At day 0, the main concern was symptom identification for 318 participants, with 66 older adults (20.8%) identifying this as their primary concern at 90-day follow-up. The majority of these older adults also identified additional concerns that were of higher priority at day 90, including obtaining medications and managing chronic conditions (20.8%), ability to take care of oneself (16.4%), or end-of-life concerns (11.0%).

FIGURE 1.

Concerns at baseline ED visit and at 90-day follow-up. Note–An example interpretation is as follows: Of respondents that identified a baseline ED visit concern as the “ability to take care of oneself” (#1 on the y-axis), 70 of those participants remained with that same concern at longitudinal 90-day follow-up, whereas 25 participants were concerned about “obtaining medications and managing chronic conditions,” 22 participants were concerned about “end of life,” etc. The concern at the 90-day follow-up could be either old (present on day 0) or new (absent on day 0). ED indicates emergency department.

“What Matters” – Desired Outcomes

While seeking ED care, the 3 most common desired outcomes selected by older adults were: (1) getting well and symptom resolution (72.0%), (2) obtaining a diagnosis (25.7%), and (3) functional independence (19.5%). Of the 667 older adults who completed the 90-day follow-up “What Matters” questions, 66.2% (n = 441) agreed that their desired outcome(s) after the ED visit was successfully achieved after their ED visit. Respectively, the 3 most common desired outcomes mentioned previously were met by 75.1%, 58.7%, and 70.7% at 90-day follow-up. Logistic regression demonstrated that having a patient meet their desired outcome was less common in those with frailty (odds ratio = 0.60, 95% CI: 0.40–0.88; P < 0.01); all other predictor variables were not associated with having a desired outcome met (Fig. 2).

FIGURE 2.

Forest plot of exposures and their association with an older adult meeting their desired outcome by 90 days after emergency care. ADL indicates activities of daily living; IADL, instrumental activities of daily living; OR, odds ratio; PCP, primary care provider.

DISCUSSION

This study is the first to longitudinally assess “What Matters” to older adults after emergency care. This study uniquely identified barriers and risk factors that limit older adults from achieving their desired outcomes. In review, this work has 4 primary findings. First, older adults were most concerned with symptom identification and/or persistence during their index ED visit, but by the 90-day follow-up, the majority of patients’ concerns and priorities had shifted to ensuring they had the ability to take care of themselves. At the index ED visit, more than 10 categories of prominent concerns were identified among older adults, highlighting the need for clinicians to avoid generalizing or developing preconceived bias as to what patients are concerned about and instead ask each patient every time. Second, both at the ED visit and during follow-up, the ability to perform self-care and functional independence were identified as patient priorities, underscoring the importance of the “Mobility” M of the 4Ms framework and the need for clinicians to consider older adults’ functional status at each ED visit. Third, 2 in 3 older adults achieved their desired “What Matters” outcome at 90-day follow-up. Finally, frail older adults were less likely to longitudinally achieve their desired health outcomes. These findings will be useful in guiding future research and encouraging clinicians to use an evidence-based approach when providing care that is patient-centered and goal-oriented to older adults who seek emergency care.

This work builds upon the literature in several ways. Prior research on “What Matters” in the emergency care setting has primarily been assessed through pilot studies.^9,21 This study is the first to utilize the “What Matters” framework within a larger population, which may overcome some of the potential limitations surrounding generalizability noted in prior work.^9,21 These findings further validate the feasibility of using “What Matters” in emergency care settings. Within the past decade, there have been increased efforts to incorporate evidence-based approaches to the care of older adults, which can serve as a foundation for the integration of the 4Ms framework in the ED. With the establishment of the “Geriatric ED Accreditation” program³¹ and efforts to use the 4Ms framework in Veteran Affairs EDs,³² there is a considerable opportunity for the 4Ms framework and “What Matters” to be implemented within EDs nationally. Of note, a considerable portion of older adults (Supplemental Digital Content, http://links.lww.com/MLR/C876) did identify concerns and desired outcomes aside from the predetermined responses provided within the studied questions within the “Condensed Conversation Guide for Identifying Patient Priorities (The One Thing) Emergency Department version.” This may reflect a more heterogeneous response base or concerns (eg, pets, ED wait times) and desired outcomes that were deemed pertinent by older adults in the ED. Once optimized to ensure attention to the key concerns and desired outcomes of older adults, integration of the 4Ms framework within electronic health records³³ offers the potential to improve communication among providers across various health care settings and ultimately improve the care patients receive. Further investigation is needed to address the ways in which the 4Ms framework and “What Matters” can improve ED-to-community care transitions for older adults.³⁴

The findings of this study indicate important takeaways for ED clinicians and researchers. During index ED visits, a plurality of participants highlighted “symptom identification and/or persistence” as a top concern. As seen in our study, older adults’ concerns and desired outcomes at 90 days evolved to emphasize the importance of maintaining independence and self-care, as the “ability to take care of oneself” was the participants’ main concern during follow-up. Interestingly, this focus on self-care and maintaining independence was also the second priority concern mentioned by older adults during their initial ED visit. These findings also suggest that ED clinicians and researchers should be mindful of populations at increased risk of not having their desired outcomes achieved. Our study identified frailty as a key barrier to successful outcome achievement. The population of frail older adults has been shown to be at increased risk for goal-discordant care and have worse outcomes post-ED visit.^35,36 For frail older adults, ED clinicians are in a unique position to address and mitigate potential adverse outcomes by providing care (eg, resources, home health support) that utilizes the “What Matters” framework and focuses on patient concerns, priorities, and desired outcomes. Furthermore, stratification of the desired outcomes revealed that the commonly desired outcomes of “getting well and symptom resolution” as well as “functional independence” were met at 90-day follow-up by ∼75% of participants, yet just over half of participants reported having their desired outcome met if they included “obtaining a diagnosis” as a priority outcome. These findings stratified by desired outcome exhibit face validity, as it is possible and even common for older adults to not obtain the desired outcome of reaching a definitive diagnosis after seeking emergency care, particularly given the common responsibility of the ED to serve as a “rule out pathology” setting. These findings highlight the critical need of ED clinicians regarding expectant management and discharge instructions for continued evaluation of the cause of their symptoms in the outpatient setting if a definitive etiology is not identified during the initial ED workup.

There are several limitations to consider with the findings of our research study. This was a longitudinal study with a fairly large sample size with modest attrition by 90 days. Responses at the 90-day follow-up may not have fully described the care older adults received in or after the ED, as respondents may have been inclined to respond to the follow-up if they had a particularly negative or positive experience. However, we do not believe there to be a risk of systematic bias with this limitation regarding specifically identified concerns or desired outcomes, and therefore, we believe our findings to remain valid. Second, data for this study were collected from the original GEARSS in 3 health care systems invested in geriatric ED care initiatives, with several EDs being accredited as geriatric EDs or in the process of becoming accredited at the time of data collection. We do not believe this poses a risk to the generalizability of our findings as we largely performed descriptive analyses of concerns and desired outcomes as opposed to an intervention assessment, which may be expected to have differing effectiveness depending on an ED’s accreditation status and associated age-friendly infrastructure. Third, telephone interview follow-up for outcome assessment may have introduced biases such as selection bias, nonresponse bias, or recall bias; the original GEARSS study team attempted to overcome these limitations by contacting participants several times if there was no initial response and also reminding participants of their initial desired outcome(s). Fourth, a participant may have identified multiple desired outcomes at day 0, whereas the day 90 follow-up question asked if cumulatively the outcome(s) had been met, as opposed to an individual assessment of each. Finally, participant responses may have been constrained to the options listed on the data collection forms derived by the GEARSS study team members. Free-text responses supplementing the “Other” option for each question were evaluated by study investigators, with a potential risk of misclassification. However, we believe this combination of predetermined categorical response options as well as free-text responses proposed anticipated common responses while also allowing more unique entries.

CONCLUSION

Older adults in the ED identifying “What Matters” were most concerned with symptom identification and/or persistence, the ability to take care of oneself, and end-of-life concerns. Older adults most commonly desired getting well and symptom resolution, obtaining a diagnosis, and functional independence, with ∼2 in 3 older adults stating that their initial desired outcome was met 90 days later. This work urges further research and development of interventions that incorporate “What Matters” conversations with older adults, specifically focusing on populations that are at increased risk of adverse outcomes during ED-to-community care transitions and goal-discordant care.