A Scoping Review of Geriatric Emergency Medicine Research Transparency in Diversity, Equity, and Inclusion Reporting

Abstract

Introduction:

The intersection of ageism and racism is underexplored in geriatric emergency medicine (GEM) research.

Methods:

We performed a scoping review of research published between January 2016-December 2021. We included original emergency department-based research focused on: falls, delirium/dementia, medication safety, and elder abuse. We excluded manuscripts that did not include (1) original research data pertaining to the 4 core topics, (2) older adults, (3) subjects from the United States, and for which full text publication could not be obtained. The primary objective was to qualitatively describe reporting about older adults’ social identities in GEM research. Secondary objectives were to describe (1) the extent of inclusion of minoritized older adults in GEM research, (2) GEM research about health equity, and (3) feasible approaches to improve the status quo of geriatric emergency medicine research reporting.

Results:

After duplicates were removed, 3277 citations remained and 883 full-text articles were reviewed, of which 222 met inclusion criteria. Five findings emerged. First, race and ethnicity reporting was inconsistent. Second, research rarely provided a rationale for an age threshold used to define geriatric patients. Third, GEM research more commonly reported sex than gender. Fourth, research commonly excluded older adults with cognitive impairment and speakers of non-English primary languages.

Conclusion:

Meaningful assessment of GEM research inclusivity is limited by inconsistent reporting of sociodemographic characteristics, specifically race and ethnicity. Reporting of sociodemographic characteristics should be standardized across different study designs. Strategies are needed to include in GEM research older adults with cognitive impairment and non-English primary languages.

Background

In 2021 the American Geriatrics Society (AGS) initiated an effort to enhance diversity, equity, and inclusion within the Journal of the American Geriatrics Society (JAGS) by explicitly highlighting the intersection of ageism and structural racism.¹ The AGS efforts included webinars addressing implicit bias awareness from various perspectives, the creation of public education materials, and an investment in research exploring diversity and health equity presented at the annual meeting. Racially and ethnically minoritized older persons too often experience delayed care, exclusion from research, and higher rates of adverse health outcomes.² The intersection of ageism and racism also overlaps with vulnerability associated with social determinants of health, as became increasingly apparent in emergency medicine during the COVID-19 pandemic.³

Healthcare disparities were the focus of the 2003 Academic Emergency Medicine Consensus Conference, though none of the sessions concentrated on the intersection with ageism.⁴ In addition, age-related disparities in emergency medicine may overlap not only with structural racism, but also sex, gender, primary language, cognitive status, sexual orientation, and rurality–none of which were explored during the prior Consensus Conference. Recognizing that AGS aspires to ensure that 100% of research presented at the Annual Scientific Meeting and published in JAGS reflects the diversity of the population studied by 2031, a deeper understanding of the status quo in geriatric emergency medicine (GEM) sociodemographic reporting is needed. Ideas from journal editorial boards, researchers, and funders regarding opportunities to improve that reporting is also needed. Since the AGS objective is to ensure that all presented and published research reflects principles of equity and inclusiveness, focusing solely on published research with a health equity objective is inadequate.

The primary objective of this scoping review was to qualitatively describe the current state of reporting about older adults’ diverse social identities in GEM research. The scoping review focused on falls, delirium/dementia, medication safety, and elder abuse, which are core topics in GEM research. Secondary objectives were: (1) to describe the extent of inclusion of older adults from minoritized racial and ethnic groups in GEM research, (2) to determine whether GEM research includes topics related to health equity, and (3) to outline feasible approaches to improve the status quo of GEM research reporting with attention to the interface of ageism and structural racism, even when the primary goal of researchers is not focusing on health equity in their investigations.

Methods

Study Design

Participating authors were identified by their leadership roles within the Society for Academic Emergency Medicine’s (SAEM’s) Academy for Diversity and Inclusion in Emergency Medicine as well as SAEM’s Academy of Geriatric Emergency Medicine and included eight emergency physicians, one medical librarian, one emergency nurse, and two public health professionals. The subtopics of geriatric falls, dementia/delirium, medication safety, and elder abuse were ultimately selected in order to focus this scoping review on specific topics that are frequently encountered in emergency department (ED) settings and for which a reasonable volume of original research currently exists. The physician co-authors were split into dyads for each topic: falls (CRC, MS), dementia/delirium (ANC, LCC), elder abuse (SA, AL), and medication safety (EO, AZ).

The authors performed a scoping review that adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) reporting guidelines.⁵ A medical librarian (AH) created a single electronic search strategy that included OVID Medline, EMBASE, CINAHL, and CENTRAL for the period 2016–2021. Full details of the search strategy are available in Supplementary text S1.

Study Selection

Each dyad independently reviewed an equal portion of the search findings in two stages using Covidence software. The first stage consisted of reviewing titles for potential inclusion. The second stage consisted of full text review for those titles deemed potentially relevant to the a priori inclusion and exclusion criteria from the first stage. During the second stage, each included article was tagged as applicable to falls, delirium/dementia, medication safety, or elder abuse. The only inclusion criteria was original research data in an ED setting. Exclusion criteria were (1) no original research data pertaining to falls, delirium/dementia, medication safety, or elder abuse; (2) non-geriatric patient population, which we did not define a priori; (3) duplicate articles or abstracts subsequently published as manuscripts; and (4) inability to obtain full text articles. Abstracts, registry data, and studies that included medical teams as the study subjects were not excluded. Later, in consultation with the editors of the JAGS Diversity in Research on Aging Series, the decision was made to also exclude research not based in the United States. Racial and ethnic categories, terms, and definitions of what constitutes a minoritized or minority group vary in different countries, as do conventions for collecting and reporting demographic patient data.⁶ Our team felt that extracting components related to patients’ social identities from individual studies outside of the United States would require an international consensus approach beyond the scope of this study. As such, we focused our efforts on providing a snapshot of the status quo within one nation.

Data Abstraction

The authors decided a priori to create two abstraction tables for each dyad. The first table provides the details of studies meeting inclusion criteria, including the timeframe of subject enrollment, number and mean/median age of subjects, research design, outcomes assessed, limitations, and key points. The second table details research subjects’ reported social identities and abilities. We specifically chose subject race, ethnicity, sex, gender, primary language, and cognitive impairment, as these were the most commonly reported. We considered abstracting information about research subject literacy, immigration status, incarceration, and veteran status, but found that these were not commonly reported. For studies that provided data about subjects’ race and/or ethnicity, we recorded whether information was provided about how race and/or ethnicity were ascertained (i.e. through chart review, research subject self-identification). For studies that included individuals with non-English primary languages or with cognitive impairment, we recorded information about methods used for informed consent. We also abstracted the age threshold that researchers used to define “older adult” and whether they cited any rationale for that threshold. The authors used the definitions and considerations described in Table 1 in abstracting individual study results for this second table.

Table 1.

Categories of Social Identities and Abilities Abstracted from GEM Research

Category	Definition	Examples
Race	Socially constructed way of categorizing humans into distinct groups based on apparent physical features such as skin color, facial shape, hair texture. Race has no biological or genetic basis as categories vary across societies.	In the US, standard racial categories are: Black or African American Asian American Indian or Alaskan Native Native Hawaiian or Other Pacific Islander White Other Mixed
Ethnicity	Socially constructed way of categorizing humans based on shared nationality, heritage, language, or cultural and religious practices	Lebanese Indian Irish Hispanic Latino/a Latine Latinx (gender-inclusive term)
Sex	Biological construct based on genetics, anatomy, and endogenous hormones	Male Female Intersex
Gender	Social category based on attitudes, roles, and behaviors related to a person’s self-identity	Cis-woman Cis-man Trans-woman Trans-man Non-binary
Age	The length of time in years that a person has lived or existed	60 65
Primary language	The language that someone uses most frequently and comfortably to communicate with. This may or may not be a person’s native language. For people who speak multiple languages, this is the language in which they prefer to speak in a given context.	English Spanish
Cognitive Impairment	Having difficulty with memory, recall, concentration, learning new things, and executive function and decision-making	Delirium Dementia Traumatic brain injury

The research team held iterative discussions to identify sociodemographic reporting patterns and arrive at consensus about health equity-related knowledge gaps related to GEM research on falls, delirium/dementia, medication safety, and elder abuse. The authors presented findings at national geriatrics and emergency medicine conferences and incorporated audience feedback to identify the knowledge gaps presented herein.

Results

A total of 3312 abstracts and papers were screened for inclusion, and 35 duplicates were removed. Of the remaining 3277 abstracts and papers, 2394 were deemed irrelevant, leaving 883 for full-text review. Six hundred sixty-four were excluded for not meeting inclusion criteria or being identified as duplicates, leaving 219 full-text studies (see Figure 1). Some of the included manuscripts evaluated more than one of our predefined topics (falls, delirium/dementia, medication safety, elder abuse), so the totals below exceed 219. Table 2 summarizes the paper types by topic and Supplement 1 provides details of the medical librarian search strategy.

Figure 1:

Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) reporting guidelines flow diagram

Table 2.

Summary of Research

	Falls	Delirium/Dementia	Medication Safety	Elder Abuse
Abstracts	23	16	13	4
Original Research	96	44	23	18
Systematic Review or Meta-analysis	3	4	0	0
Total*	122	64	36	22

^*There were 219 unique titles included in the scoping review, but some spanned more than one topic area. As such, total exceeds 219.

Several general trends emerged in research reporting of social identities and abilities across the four core topics, which are summarized in the following paragraphs.

Defining older adults.

In general, studies did not explicitly define “older adults” or “geriatric” patients. While most either used an age threshold of 65 and older as an inclusion criterion or to define a geriatric subset of participants compared to other age groups, few provided a specific rationale for selecting this age.

Race and ethnicity.

The extent to which minoritized racial and ethnic groups are represented in GEM research is unclear. The five major racial categories in the United States census are American Indian or Alaska Native, Asian, Black or African American, native Hawaiian or other Pacific Islander, or White. Ethnicity in the United States census is designated as Hispanic or Latino or non-Hispanic or non-Latino, and medical research tends to report along these lines even though many other ethnicities exist.⁷ GEM research tended to report participant race as White or non-White, rather than delineating specific racial groups included. Studies reporting ethnicity did so to distinguish Hispanic from non-Hispanic individuals, but did not provide information about inclusion of individuals belonging to any other specific ethnic groups. Manuscript texts and demographic tables often group race and ethnicity as one demographic characteristic and report them together. Researchers tend to omit how race and ethnicity of participants was determined.

Sex and gender.

GEM research tends to report sex, rather than gender. Researchers tend to use the terms sex and gender interchangeably or offer a demographic category without a heading, e.g. “Men, 50%,” rather than “Sex: Male, 50%.” Based on current standards for equitable language,⁸ misuse of sex and gender as terms is frequent, e.g. referring to subjects’ gender as “female” rather than “women.” Researchers tend to omit how the gender of participants is determined. No studies reported including non-binary or transgender individuals.

Primary language and cognitive impairment.

GEM research in the United States often excludes speakers of non-English primary languages and individuals with cognitive impairment, but tends to lack explicit statements about their exclusion. Descriptions of informed consent/assent procedures for research subjects with cognitive impairment were rare.

Detailed descriptions of these findings by topic are outlined below, stratified by the research themes, focus on health equity, and a synthesis of reporting on race, ethnicity, sex, gender, cognitive impairment, primary language, and age criteria. Systematic reviews and meta-analyses did not include detailed demographic data about pooled participant populations, and as such are excluded in the main text below where percentages are given about studies’ reporting of sociodemographic characteristics.

Falls

Main Themes and Content of Studies.

In total, 122 titles were abstracted, with research methods for those studies summarized in Supplement S2. As summarized in Table 2, 22 were research abstracts, 96 were original research manuscripts, and 3 were systematic reviews or meta-analyses. Publications focused on injury patterns, trauma activation protocols, screening for fall risk and interventions to reduce injurious falls. Supplement S3 summarizes each study’s clarity in reporting sociodemographic characteristics of their study population.

Health Equity Topics.

One paper examined women, fractures, frailty, and fall risk​.⁹ Some articles looked at falls across ages and found specific injury patterns/severity in older adults. None examined falls as related to race, ethnicity, or other sociodemographic categories.

Race and Ethnicity.

Forty-three abstracts and original research publications (36%) reported participant race. None described how participant race was determined. Twenty-two studies (18%) reported participant ethnicity. Outside of designating participants as Hispanic versus non-Hispanic, none named specific ethnicities of participants.

Sex and gender.

Most studies reported sex (85%). Few reported gender (21%).

Cognitive Impairment.

Among abstracts and original research manuscripts in which ED patients, rather than health care providers, were research subjects, over two-thirds (68%) did not report whether they excluded patients with cognitive impairment. Of 25 studies that included patients with cognitive impairment, 9 (36%) detailed their approach to subject consent/assent.^10–18

Primary language.

Only twenty-four abstracts and original research publications (20%) included an explicit statement about inclusion or exclusion of speakers of non-English languages. Of studies that offered this information, 13 included non-English speaking patients, and in all cases, the primary language of these patients was Spanish.

Age.

Very few studies explicitly defined “older adults” or “geriatric patients.” Of 106 titles that provided age thresholds for study inclusion, the threshold ranged from 50 to 75, and 87 studies (82%) used age 65 and older. Only two studies cited a rationale for the threshold used.^19,20

The unit of analysis was generally patient-level. Some studies explored interventions with healthcare team members, and those studies did not report subjects’ sociodemographic characteristics with greater clarity than patient-level analyses. Large database analyses were frequently observed in falls research, including local trauma registries, National Emergency Department Sample (NEDS), National Electronic Injury Surveillance System (NEISS), National Hospital Ambulatory Care Survey (NHAMCS), and Medicare claims data. Retrospective database studies generally failed to report on sex, gender, race, ethnicity, or primary language and did not describe inclusion of those with cognitive impairment.

Delirium and Dementia

Main Themes and Content of Studies.

Of 64 titles, 44 were original research, 17 were abstracts, 3 were systematic reviews. Supplement S4 synthesizes the research methods for those studies. Studies about delirium focused on characteristics, risk factors, and incidence of delirium as well as screening and assessment tools associated with Geriatric EDs. ED dementia literature focused on healthcare and/or ED utilization, caregiver burden, descriptions of programs for Persons Living With Dementia (PLWD), and neuropsychiatric symptoms and management. Five studies examined healthcare providers’ perceptions or behaviors.^21–25 Supplement S5 summarizes each study’s clarity in reporting sociodemographics of their study population.

Health Equity Topics.

One study about ED delirium focused specifically on a special population, i.e. veteran patients; this study examined caregiver burden and its association with ED and hospital disposition and utilization, and found delirium to be a risk factor for high caregiver burden.²⁶ Two ED dementia studies focused on minoritized older adults. The first studied Mexican American beneficiaries with and without Alzheimer’s Disease and Related Dementias (ADRD).²⁷ The second evaluated racial differences in healthcare utilization and self-assessed caregiver burden.²⁸ Notably, the second study asked participants to self-identify their race rather than abstracting from the chart and excluded anyone who did not identify as Black or White.

Race and Ethnicity.

Race was reported in 52% of studies and ethnicity in 22%. Ethnicity reporting was typically described as Hispanic or non-Hispanic, without accounting for alternate ethnic identities. The one meta-analysis studied delirium detection, prevention, and management, and described false positive screenings for Black Americans with dementia.²⁹ Additionally, some studies described PLWD who were female and with specific racial identities having higher ED utilization.^30–37

Sex and Gender.

Sex was reported in 28% of abstracts and original research and gender in 23%.

Cognitive Impairment.

Information regarding inclusion/exclusion of patients with cognitive impairment was explicitly included in 51 studies. Only 8 of these studies described informed consent procedures for individuals with cognitive impairment.^38–45

Primary Language.

Of the 65 publications, only 5 offered explicit information that they included speakers of non-English primary languages. Research on delirium and dementia using data from Center for Medicare and Medicaid Services (CMS) is common, and research subjects’ primary languages were not reported in these studies.

Age.

In two-thirds (67%) of abstracts and original research publications, the age threshold used in inclusion criteria for geriatric patients was age 65+. Two studies included patients aged 18+, one of which found higher ED utilization among older adults after natural disaster³⁴ and the other of which focused on delirium incidence in ED hallway care and had a predominantly geriatric cohort whose median age was 71.⁴⁶ Otherwise, age thresholds for older adults ranged from 45+ to 80+. Rationales for using specific age thresholds were not included, and there was no consistent definition of the term geriatric.

Medication Safety

Main Themes and Content of Studies.

Thirty-six publications about medication safety were identified, which included 13 abstracts, one systematic review, and 22 original research publications. Supplement S6 synthesizes the research methods for those studies. Most data were extracted from retrospective chart reviews or registries and performed at academic medical centers. Studies focused on assessing (1) the impact of quality improvement initiatives, such as protocol development, screening tools, best practice alerts, and the use of geriatric specialists or pharmacists, (2) adverse events, including mortality, delirium, falls, ED revisits, and hospitalizations, and (3) sedating medications, blood thinners, and other potentially inappropriate medications (PIMs). Common topics included medication safety related to psychotropic medications; sedatives, and analgesics; and the use of specialists or pharmacists was the primary investigation in five papers. Supplement S7 summarizes each study’s clarity in reporting sociodemographic characteristics of their study population.

Health Equity Topics.

While all included articles addressed older patients as a vulnerable group to polypharmacy, other sociodemographic dimensions were not a specific focus in most publications. One study at two urban academic centers examined medication safety in Black patients with diabetes mellitus and briefly described potential access barriers in this patient population.⁴⁷ A small study at one academic ED evaluated a prediction tool for which more than 50% of participants were listed as African American.⁴⁸ Another vulnerable population addressed was veterans, which was studied in three separate papers.^49–51 All three studies were conducted at Veteran’s Administration Hospitals and focused on use of PIMs. One was a retrospective chart review examining high-risk medications on the active outpatient medication list.⁵⁰ The other two studies with veterans described quality improvement initiatives to decrease prescribing of PIMs in veterans discharged from the ED.^49,51

Race and Ethnicity.

Of the publications reviewed, 17 of the 36 (47%) explicitly described that they included non-White participants. However, many studies reported race as White vs. non-White and did not provide specific descriptors of non-White races. They also incorrectly used the terms race and ethnicity or collapsed them into one category. Only one publication described how race of participants was determined.⁵² None described how ethnicity was determined.

Sex and Gender.

Twenty-three studies reported participant sex and 3 reported gender. Ten publications did not provide information about participant sex or gender.

Cognitive Impairment.

About half of studies (n=20, 56%) included patients with cognitive impairment. One study explicitly described excluding patients with cognitive impairment.⁵³ The remaining 15 studies did not provide explicit information about inclusion of patients with cognitive impairment.

Primary Language.

Only one study explicitly described including non-English speaking patients,⁵⁴ and another study described excluding non-English speaking patients.⁵⁵ Inclusion of speakers of non-English primary languages was unknown in the remaining 34 studies.

Age.

Twenty-six publications (72%) used age ≥65 as the threshold for older adults. The age threshold for geriatric patients ranged from ages 50+ to 80+. One abstract did not specify an age threshold for their geriatric population.⁵⁶ Another study pooled data that included various ages for participants in their review.⁵⁷ Only two studies provided a rationale for using a specific age threshold to define geriatric patients.^53,58

Elder Abuse

Main Themes and Content of Studies.

Twenty-two publications were identified about elder abuse and included 4 abstracts and 18 original research documents. No eligible systematic reviews or meta-analyses were identified. Supplement S8 synthesizes the research methods for those studies. Studies focused on patient characteristics and injuries, screening validation, qualitative feasibility studies, ED elder abuse protocol development, and barriers to identifying elder abuse. Supplement S9 summarizes each study’s clarity in reporting diversity and inclusivity of their study population.

Health Equity Topics.

No studies specifically focused on health equity or minoritized populations. Two studies sought to identify association of elder abuse with victims’ sex and race.^59,60

Race and Ethnicity.

Eleven studies reported participant race, five of which described how race was determined. Seven original research articles reported ethnicity, although only three described how ethnicity was determined.^61–63

Sex and Gender.

One abstract⁵⁹ and seven original research articles reported research subject sex. Four original research articles reported gender.

Cognitive Impairment.

Eight studies explicitly described inclusion or exclusion of subjects with cognitive impairment. Of these, one excluded and six included subjects with cognitive impairment.

Primary Language.

Only one study included speakers of non-English primary languages.⁶¹ Twelve studies explicitly reported whether they included speakers of non-English primary languages, and ten did not.

Age.

Fifteen studies reported an age threshold for older adults. Eight used age 65+ and 7 used age 60+ to define their geriatric study population.

Discussion

With respect to JAGS’ effort to promote uniformly transparent reporting to facilitate a clearer understanding of health inequities, our scoping review revealed several problematic trends in research reporting. There is a noticeable lack of GEM research focused on health equity-related topics. Furthermore, GEM research that is not explicitly focused on health equity fails to consistently report sociodemographic characteristics of the populations studied, and as such, meaningful post-hoc analyses are difficult. Based on our findings, we argue for an increased level of explicitness in GEM research reporting on race, ethnicity, sex, gender, cognitive impairment, and primary language in prospective and retrospective research (Figure 2).

Figure 2:

Summary of sociodemographic characteristics prioritized in the present review and key points

Health equity bears further exploration as a topic of GEM research. As of now, it is unclear what health disparities exist in the four core GEM areas on which this scoping review centered as related to race, ethnicity, sex, gender, primary language, and cognitive impairment. For these four core areas of GEM research, we have proposed a future research agenda along with examples of possible areas of investigation (Table 3). Areas of future research include but are not limited to racial and ethnic disparities, and future research should strive to include patients who speak non-English primary languages and those with cognitive impairment. While not specifically investigated in this scoping review, inclusion of geriatric patients with visual and/or hearing impairment should also be encouraged. We acknowledge the difficulty in including certain patient populations, such as navigating the issue of informed consent when a patient has severe dementia or funding interpreters to recruit participants and create non-English language materials.^64–66 However, our scoping review shows that failure to do so has led to a body of work that seemingly excludes patient populations with cognitive impairment and non-English primary languages. Identification of disparities is a critical first step to designing interventions that can provide equitable care to all geriatric patients in the ED.

Table 3.

Potential Topics for Future Health Equity-Focused Research in Core Areas of Geriatric Emergency Medicine

Subject Area	Potential Topics for Future Research
Falls	Racial and Ethnic Disparities • Are there disparities in pre-hospital selection of level 1 vs. 2 vs. 3 trauma centers for minoritized older adults with falls? • Are there disparities in triage Emergency Severity Index assignment or activation of trauma for minoritized older adults with falls? • Do minoritized older adults with acute injuries have delays to analgesia? Language and Culture • Do older adults with acute injuries who speak non-English primary languages experience delays to analgesia? • Does use of ED imaging differ for older adults with falls and/or acute injuries who speak non-English primary languages? • How does a patient’s primary language and social support network impact transitions of care after an ED visit for fall? Intersectional Identities • How do factors such as level of cognitive impairment and housing status/conditions impact frequency of falls and severity of associated injuries? • How do emergency care teams assess social support networks for older adults with falls, and how does this impact disposition decisions?
Delirium/dementia	Racial and Ethnic Disparities • Are there disparities in ED triage of minoritized older adults with chief complaints of “confusion” or “altered mental status”? • How has ED boarding impacted minoritized older adults with cognitive impairment? • Do clinicians obtain advanced imaging (CT head) more frequently for minoritized than non-minoritized older adults? • Are there disparities in ED use of chemical sedation and physical restraints for minoritized vs. non-minoritized older adults? Language and Culture • What screening tool is the most feasible to administer to patients who speak a non-English primary language? • Can and should ED delirium screening tools be validated in non-English languages that are widely spoken in the United States? • Are interpreters regularly used when staff screen patients for delirium? How does interpreter use impact delirium assessments? Intersectional Identities • Are there differences in how clinicians screen patients with hearing impairment for cognitive impairment?
Polypharmacy	Racial and Ethnic Disparities • How do minoritized older adults understand medication safety? Does (poor) access to medication impact their experiences of polypharmacy? • Do charts of minoritized older adults more frequently contain stigmatizing language about medication non-compliance? • Are there disparities in ED pharmacy consultation for minoritized older adults? Language and Culture • How does the ED medication reconciliation process differ for older adults who primarily speak non-English languages • How can discharge instructions about medication use be improved for older adults who primarily speak non-English languages • How are cultural practices of “gifting” medications, particularly those from overseas, taken into account in ED medication reconciliations? Intersectional Identities • Does medication prescribing from the ED vary based on older adults’ health insurance coverage?
Elder abuse	Racial and Ethnic Disparities • Are there differences in rates of elder abuse, neglect, or exploitation in older adults who do vs. do not belong to a minoritized racial or ethnic group? How does minoritized racial or ethnic group overlap with low socioeconomic status and its association with elder abuse, neglect, or exploitation? • Are there higher rates of consulting Adult Protective Services and/or legal reporting of elder abuse for minoritized older adults? Language and Culture • What are best practices for screening for elder abuse, neglect, or exploitation cross-culturally? • How do patients of diverse racial and ethnic identities interpret screening questions for elder abuse, neglect, or exploitation? Intersectional Identities • How does LGBTQIA+ identity impact risk for elder abuse, neglect, or exploitation?

Additionally, we call for a reconsideration of the 65 years age threshold for inclusion in a geriatric cohort given the phenomenon of accelerated aging. Accelerated aging describes the development of early-onset age-associated disease due to chronic environmental stressors.⁶⁷ This phenomenon disproportionately impacts Black, Hispanic, and American Indian people, who on average have a greater burden of diabetes, hypertension, and cardiovascular disease compared to White and non-Hispanic people of the same age.^67,68 In this context, the widespread use of 65 years as an age threshold for GEM research may have the unintended consequence of excluding minoritized subjects who are experiencing accelerated aging. As a result, Black, Hispanic, and American Indian people experiencing accelerated aging are understudied in GEM research.

Future GEM researchers should strive for clear and complete reporting of patient sociodemographic details. Our scoping review identified significant inconsistencies in how race, ethnicity, gender, cognitive status, and primary language of participants were reported. As a result, it is difficult to ascertain what patient populations are included in GEM research. Furthermore, because race is a social construct with ambiguous and shifting boundaries that vary geographically, researchers should specify how patient race and ethnicity information was collected. For example, a patient may self-identify differently than a registrar who presumes information about the patient’s race and inputs it into an electronic medical record. There should be careful consideration of reporting based on racial and ethnic categories from the United States census as it can obfuscate inclusion of vulnerable populations. For example, research subjects of Middle Eastern and North African descent would be categorized as “white non-Hispanic,” and accordingly assumed to be in a privileged majority despite facing unique healthcare challenges and needs. For these reasons, United States government agencies, including the Centers for Medicare and Medicaid Services (CMS), have moved towards recommending data collection based on self-reported race and ethnicity.⁶⁹ Of note, the Office of Management and Budget has recently revised guidelines so that 1) race and ethnicity is a combined question that allows respondents to select categores and 2) Middle Eastern or North African has been added as a new minimum reporting category. As these changs are adopted by agencies, researchers should consider adding these options when asking participants how they identify.

JAGS and the Journal of the American Medical Association have published best practices for reporting of race and ethnicity in medical and scientific journals.^7,70 Journals should indeed offer such explicit policies outlining expectations for how race and ethnicity should be reported in original research, ideally citing national guidelines establishing clearly standardized terms. The American Medical Association also offers a guide for inclusive language, which offers clear definitions of social identifiers and preferred terminologies.⁷¹ Because racial and ethnic categories vary from country to country, it may not be possible to create international guidelines about sociodemographic reporting. Complicating matters further, the Enhancing the Quality and Transparency of Research (EQUATOR) Network has explicit criteria on health equity reporting for randomized controlled trials⁷² and systematic reviews,⁷³ but not for other common study designs like observational studies, diagnostic accuracy, prognostic modeling, qualitative research, quality improvement, or economic evaluations.⁷⁴ Such entities and journals should make available educational materials to scholars, scientific editors and reviewers to encourage standardized reporting.

Aspirations must be tempered with the realities of contemporary healthcare, and the resources required to advance GEM research reporting are unequivocally linked to the ultimate success of these efforts. The CMS Framework for Health Equity strives to remedy systemic barriers to equity across race, ethnicity, disability, sexual orientation, gender identity, socioeconomic status, geography, preferred language and other factors that may affect health outcomes.⁷⁵ In addition, the National Institute on Minority Health and Health Disparities (NIMHD) provides a research framework for investigators to consider when designing and reporting research.⁷⁶ Neither of these initiatives provide resources or detail ideas for healthcare systems or investigators to actually begin routinely collecting these patient-level factors. Independent of CMS and NIMHD, some champions of health equity are devising automated approaches to implant elements of potential inequity into electronic medical records without relying upon nurses or physicians. For example, the Housing-Based Socioeconomic Status (HOUSES) Program provides individual socioeconomic status for patients inserted into electronic medical records and is updated annually, but will not illuminate other factors like ethnicity, disabilities, gender identity, or language spoken.⁷⁷

On top of feasibility concerns for healthcare systems to somehow improve the data capture to better understand potential health inequities, policymakers ought to consider the unintended consequences of these efforts in GEM research settings. If opt-in reporting of race, ethnicity, gender identity, sexual orientation, primary language, or disabilities becomes standard, hospital systems and research studies will likely experience missing data, as some patients may prefer not to answer these questions. For example, prior research has shown that people who identify as belonging to minoritized racial and ethnic groups are more likely than Whites to skip questions about their race and ethnicity.⁷⁸ Patients could negatively perceive requests to self-report their sociodemographic characteristics, and in turn decline research participation or even clinical care.

Limitations

The majority of works examined in this scoping review were published prior to 2020. Due to police violence and health disparities that the COVID-19 pandemic magnified,^79,80 2020 was a turning point for collective attention to structural racism both in academic medicine and nationally in the United States. Since then, investigations may have had a greater focus on health equity topics and more meticulous sociodemographic reporting. We did not abstract data related to exclusion of patients based on older age or “age caps” (e.g. exclusion of patients age 85+), which is a known problem in other emergency medicine research. The extent to which capping of age for interventions and treatments delivered in the ED for conditions outside of the 4 specified is unknown and not addressed by this study.

Conclusions

Due to inconsistencies in reporting of sociodemographic characteristics, it is difficult to determine the state of inclusion of older adults with diverse social identities in GEM research, and specifically of those belonging to minoritized racial and ethnic groups. There is a dearth of GEM research focused on health equity topics and marginalized and minoritized populations. Strategies are urgently needed to include in GEM research older adults with cognitive impairment and non-English primary languages.

Supplementary Material

SUP TAB 5

Supplement Table S5 reports the sociodemographic elements of equity and diversity reporting for the delirium studies abstracted.

SUP TAB 6

Supplement Table S6 synthesizes the attributes of the medication safety studies abstracted.

SUP TAB 4

Supplement Table S4 synthesizes the attributes of the delirium studies abstracted.

SUP TAB 7

Supplement Table S7 reports the sociodemographic elements of equity and diversity reporting for the medication safety studies abstracted.

SUP TAB 8

Supplement Table S8 synthesizes the attributes of the elder abuse studies abstracted.

SUP TAB 9

Supplement Table S9 reports the sociodemographic elements of equity and diversity reporting for the elder abuse studies abstracted.

SUP TAB 1

Supplementary Text S1 provides details of the electronic search strategy devised by a medical librarian for this scoping review.

SUP TAB 3

Supplement Table S3 reports the sociodemographic elements of equity and diversity reporting for the fall studies abstracted.

SUP TAB 2

Supplement Table S2 synthesizes the attributes of the fall studies abstracted.