Inclusion of Older Adults in Emergency Department Clinical Research: Strategies to Achieve a Critical Goal

Abstract

Medical research across all fields has historically excluded older adults (aged 65 years and older). As older adults have a higher burden of chronic illness, respond differently to treatment, and are more prone to medication side effects, the results of current research may not be applicable to this important population. To address this major research deficiency, the National Institutes of Health (NIH) established the Inclusion Across the Lifespan policy, effective January 2019. We present important considerations and proven strategies for successful inclusion of older adults in emergency care research relating to study design, participant recruitment and retention, and sources of support for investigators.

Including Older Adults: A Critical Goal for Emergency Care Research

There are over 23 million Emergency Department (ED) visits by older adults (aged 65 years and older) annually¹ and this number is expected to grow because of longer lifespans, the aging of baby boomers^2,3 as well as the fact that older adults are increasingly seeking care in the ED rather than other settings.³ In fact, older adults are expected to account for as much as one-third of ED visits by 2030.⁴

Medical research across all fields has historically excluded older adults.^5–8 Of trials published between 1965–2015, 33% had arbitrary upper age limits, and 67% reported on subjects younger than those typically afflicted by the disease under study.⁷ Older adults can also be systematically excluded without an explicit age cutoff by using exclusion criteria, such as multiple comorbid conditions, effectively excluding all or virtually all older adults. (Table 1) For instance, older adults are at greatest risk of severe disease and death due to coronavirus disease (COVID-19), but a recent review found that older adults are likely to be excluded (not based on a hard age cut off) from more than 50% of COVID-19 clinical trials, limiting the ability to evaluate the efficacy, dosage, and adverse effects of these treatments on this population.⁹ Examples of reasons for this type of exclusion included compliance concerns, consent challenges, and technology requirements.⁹ This pervasive exclusion has had profound consequences. Therefore, the results of current research may not be applicable to this important population as older adults have a higher burden of chronic illness, respond differently to treatment, have atypical presentations of disease, functional and/or cognitive impairment and are more prone to medication side effects.^10–13

Table 1.

Examples of exclusion criteria to avoid because they intentionally or unintentionally exclude older adults.

Exclusion Criteria	Potential Effect	Solution
Age	Excludes adults solely on the basis of age.	Ensure that the study population reflects the population affected by the disease. For example, do not limit a study on myocardial infarction to patients less than 65 years of age.
Comorbid conditions	Excludes older adults because they are more likely to have comorbid conditions.	Do not include comorbid conditions as exclusion criteria or, if it is necessary due to an expected effect, only include those specific comorbidities. Also take care to ensure that by excluding that comorbidity, you are not excluding the majority of patients with the condition of interest.
Medications	Excludes older adults because they are more likely to be prescribed many medications	Only exclude based on current medications if there is evidence to support specific medications will impact the outcomes.
Mental status	Excludes older adults as they are more likely to have delirium / dementia.	Apply the strategies outlined to include patients of all ages with acute or chronic changes to their mental status.

To address this major research deficiency, the National Institutes of Health (NIH) established the Inclusion Across the Lifespan policy effective January 2019 which states that “individuals of all ages, including…older adults, must be included in all human subjects research, conducted or supported by the NIH, unless there are scientific reasons not to include them.”¹⁴ Depending on the research question, it may be appropriate to include older adults in addition to younger adults or to focus on older adults exclusively. This inclusion should be based on the scientific basis of the question to ensure the results are meaningful and the older adults are not included solely as “tokens.”¹⁵ Adults with severe or uncontrolled comorbidities may need to be excluded for their own health and safety, but age-based exclusions are rarely justified and other exclusions with disproportionate effects on inclusion of older adults should be limited as much as possible consistent with patient safety and scientific validity.

This will have a profound impact on nearly all emergency medicine research examining topics including resuscitation, cardiology, neurology, toxicology, and many others. Researchers will need to be prepared to respond to these new requirements. The potential impact of this policy on research budgets should be considered early in the planning process. Including older adults may increase the pool of eligible patients, thus decreasing time needed for enrollment. However, their inclusion may increase costs for example due to increased complexity or need for tools for hearing impairment. Researchers can look to both NIH funding and private funding to help assist with these costs.

Although there are barriers to successful recruitment of older adults in the ED,¹⁶ by including older adults, or more importantly discontinuing the exclusion of older adults, clinicians across all subsets of emergency medicine research will be able to make evidence-based treatment decisions that will ensure even our oldest patients experience appropriate, effective treatments. This more inclusive research may also lead to changes in education and policy to align with the most recent evidence for the care of older adults.

Often cited is the concern that older adults will reduce a study’s ability to demonstrate effectiveness of an intervention. Instead, older adults are typically more ill and may increase a study’s ability to demonstrate effectiveness including in preliminary trials. For example, age-adjusted d-dimer expanded the population in which d-dimer can reduce the need for imaging for pulmonary embolism.¹⁷

Previous Success Including Older Adults in Emergency Care Research

Even before this NIH mandate, emergency care research focused on issues specific to older patients was increasing. PubMed indexed 14 publications on “geriatric emergency medicine” in 2000, 48 in 2010 and 575 in 2019. This work and recent successful NIH funding awards^18,19 provide evidence that older adults can be included successfully in emergency care research.

Many organizations have supported geriatric-focused emergency care research including the NIH, the National Institute of Justice, specialty organization foundations including the Emergency Medicine Foundation (EMF) and the Society for Academic Emergency Medicine Foundation (SAEMF), American Geriatrics Society (AGS) and private foundations including the John A. Hartford Foundation and the West Health Institute. The NIH’s National Institute on Aging (NIA) has funding mechanisms dedicated to developing leaders in geriatric care research among sub-specialists including in emergency medicine.¹⁹

In addition to funding individual research studies, NIH is funding the development of interdisciplinary research infrastructure in geriatric emergency care. This includes the Geriatric Emergency Care Applied Research (GEAR) Network (R23 AG058926),¹⁸ which is establishing research priorities; standardizing data approaches and measures for common geriatric emergency care conditions; developing and validating a data bank to support opportunities for investigators to conduct geriatric emergency care research; and facilitating future multi-center studies. This network has already been very successful.^20–22 Additionally, another collaborative NIH research infrastructure grant, GEAR 2.0-Advancing Dementia Care (R61 AG069822), is focused on developing collaborative research to improve emergency care for persons living with Alzheimer’s disease and related dementias.²³

The Geriatric Emergency Department Guidelines have been developed in part based on the results of some of this early geriatric emergency care research but primarily from other settings.²⁴ These guidelines are the basis for the American College of Emergency Physicians’ accreditation program for EDs recognizing on their geriatric-specific care.²⁵ Though the establishment of an accrediting organization represents promising progress, it underscores the need for additional research to identify and support evidence-based guidelines in geriatric emergency care as recommended by national emergency medicine organizations.²⁶

Strategies for Inclusion of Older Adults in Emergency Care Research

While literature exists describing challenges and strategies for including older adults in research in other clinical settings,^5,6,27 we know of none that focus on successful strategies for emergency medicine. Therefore, based on this literature in other settings⁸ and the experiences of Geriatric Emergency Medicine researchers, we present strategies and methods to increase inclusion of older adults in all emergency care research. (Table 2)

Table 2.

Summary of recommendations.

Design and Planning:

• Include older adults in any study examining medical concerns that impact them

• Include geriatric expertise on study team

• Consider patient-centered outcomes

• Facilitate participation of older adults by accommodating vision and hearing impairment

Measuring Relevant Covariates:

• Consider potential confounders (e.g. frailty, comorbidity burden) and measure them

Considering the Importance of Geriatric Syndromes:

• Determine potential effects on your outcomes

• Include older adults with these syndromes to improve generalizability

Cognitive Issues and Capacity to Consent:

• Include patients with both acute and chronic memory problems

• Assess ability to consent and develop a process for enrollment by a legally authorized representative with assent of individual with memory problems

Recruitment / Enrollment:

• Ensure process is appropriate for older adults by reducing survey length and complexity

• Leverage familiar family, friends, and clinicians to assist with enrollment.

Retention:

• Make multiple modalities for follow-up available (in-person, phone, texts, email, and online response)

• Allow flexible follow up windows to accommodate hospitalizations or transitions to other care settings

• Obtain alternate contacts (family, friends, etc.) to use at enrollment to ensure your ability to reach the patient

• Consider study burden and alternative options for outcome assessment if the participant is near the end of life or hospice is initiated

Design and Planning

Successful inclusion of older adults in ED research begins in the study design/planning phase. Consistent with the NIH recommendations, we strongly encourage researchers to include older adults in all studies examining medical concerns that impact them. Including older adults will increase the generalizability and clinical relevance of a study’s findings. Thoughtful, careful initial design/planning is essential before study launch to anticipate and overcome potential barriers.

In addition to study design, planning should include careful consideration of tools required to facilitate older adult participation in research studies. We recommend having tools to help those with vision and hearing impairment; for example, providing study documents printed in larger font, magnifying devices, reading glasses and/or hearing amplification devices to assist with study consent/participation.

As in any study, careful attention should be given to ensure the study team has the right members with complementary skills including someone with geriatrics expertise if this is not a strength of other research team members. In addition, the team should carefully consider the need for appropriate cultural appreciation / sensitivity in your study population. Finally, many studies focus on mortality as the primary or sole negative outcome. For older adults, as this outcome is inevitable for participants near the end-of-life, researchers may want to focus on other patient-centered outcomes. One important example is quality of life²⁸ as it often decreases after an ED visit for acute illness but this may be modifiable.²⁹ Others may include reduction in unpleasant symptoms.

Measuring Relevant Covariates

Rather than excluding older adults from research due to their complexity, investigators should carefully plan rigorous methods that measure and include potential confounders. A major reason that older adults have been excluded from previous studies is concern that the existence of comorbid conditions and multiple medications as well as the physiological changes and functional decline associated with aging may confound results focused on a specific clinical question.^5,6,8,30 For instance, investigators may be concerned that more side effects will be experienced by older adults on multiple medications, and make the therapy appear more hazardous; thus they may want to include polypharmacy as a covariate. This is particularly true in industry-funded trials designed to estimate the efficacy of a new medication or other therapeutic strategy.⁷ Investigators should plan to include factors relevant to older adults in data collection instruments, such as assessing for comorbid conditions, current medications, living situation, cognitive status, social isolation, frailty and functional status among others. The study of older adults with definable clinical states (for example, frailty or comorbidity clusters) may offer important information about risk factors and pathophysiology specific to aging adults.³⁰ Older adults with substantial frailty or a high comorbidity burden are more likely to experience several clinical trial endpoints (e.g. falls, death) and therefore should be measured and reported. Uneven distribution of these factors could be reasons for heterogeneity of treatment effects. Many investigators may want to stratify by these factors as well.

Considering the Importance of Geriatric Syndromes

The relevance of common age-related conditions to a research question should be considered. Geriatric syndromes, such as falls, delirium, and frailty, are outcomes for many risk factors to which older adults are commonly exposed. These syndromes are also associated with significant morbidity and mortality³¹ and affect ED process measures such as disposition and length of stay.³² Depending on the research focus, it may be important to identify these syndromes in research participants. (Table 3) Rigorous approaches to assess for their presence, even in administrative or medical records data, such as insurance claims, have been developed.^33,34 These datasets have been used successfully in emergency medicine research³⁵ and the continued use may help accelerate geriatric inclusiveness in research.

Table 3.

Select geriatric syndromes and possible evidence-based tools for measurement. Investigators should work closely with a geriatrics expert to ensure they utilize the right tool.

Geriatric Syndrome	When to Include	Potential Tools
Falls	When fall events are a planned outcome	Nursing Triage Questions36 4 Stage Balance Test59 Sit-to-Stand Test59
Delirium	To determine the likelihood of delirium during enrollment and follow-up To measure the occurrence of delirium throughout the study To understand how confusion may impact study outcomes	Single Question in Delirium (SQiD)60 Delirium Triage Screen (DTS)61
Frailty	To ensure even distribution frailty in both study arms To provide context during the analysis for why an individual may have had increased vulnerability to adverse health outcomes such as falls, disability, and institutionalization	Identification of Seniors at Risk (ISAR)62 Triage Risk Screening Tool63 The Clinical Frailty Scale40

As an example, a falls history is often screened for as part of the ED nursing triage assessment;³⁶ unfortunately, an ideal ED screening tool for falls history and risk does not currently exist,³⁷ but this is an area of active investigation.²¹ Multiple scoring systems for frailty have been successfully used in the ED setting with good prediction of subsequent complications.³⁸ Notably, there is currently debate about the best frailty tool³⁹ and, similar to falls, selecting tools that are feasible and valid for the ED population is an active area of investigation.^40,41 Table 2 includes examples of tools that may be appropriate, but investigators should reference the most recent literature in consultation with content experts to select the optimal approach for their studies. Other geriatric syndromes exist that we do not fully describe here, and geriatric syndrome measurement is complex and needs further study. Geriatric syndromes often co-exist and impact each other, which further complicates research.

Finally, these geriatric syndromes also make geriatric patients challenging to enroll and thus must be considered so that patients are not systematically excluded. Including patients with geriatric syndromes improves research generalizability.

Cognitive Issues and Capacity to Consent

Older adults may have acutely altered mental status (delirium) or chronic cognitive impairment (dementia) impacting their capacity to consent to participate in research. Researchers may automatically exclude older adults if they think they cannot consent due to a comorbidity, cannot follow instructions necessary for participation, or cannot accurately report outcomes.⁴² This is unfortunate and a missed opportunity, as older adult with cognitive impairment have higher rates of healthcare use⁴³ and therefore could benefit from improvements in medical care across specialties.⁴⁴ Though a larger proportion of older adults have cognitive impairment, younger patients can also suffer from cognitive impairment, and these recommendations are likely to benefit all research patients. There are several tools commonly used to determine capacity to consent for research; Gilbert et al. provides a thorough overview.⁴⁵ The most commonly used tool⁴⁵ is the MacArthur Competence Assessment Tool for Clinical Research (MacCAT-CT).⁴⁶ If the patient lacks capacity, consent and enrollment by a proxy or legally authorized representative (LAR) can and should be utilized.^42,47 The NIH has guidelines for these patients and a hierarchy for whom the LAR should be at an NIH site and notes this could vary by state law or institutional policy.⁴⁸

Institutional review boards (IRBs) that approve research studies, however, often request that the proxy to be available at the time of consent in the ED. Alternatives to consider and discuss with your IRB include electronic consents, telephone consent with or without a paper consent completed later, and/or extended recruitment windows to allow time for the LAR to come to the hospital. Patients at the end of life and on hospice are an important population for inclusion in research but many may also present a challenge for consent. Similar strategies have been successful in this population.

There are many examples of studies that have successfully used LARs for consent. One recent example is the Geriatric Acute and Post-acute Fall Prevention Intervention study.^49,50 If the patient scored less than 4 on the Six-Item Screener,⁵¹ indicating the patient was at high risk for cognitive impairment, an LAR was required for consent. Other investigators prefer a decisional capacity assessment. Seek out experiences at your own institution or discuss with your IRB to determine what would work best at your institution. For longitudinal studies, repeating cognitive assessments may be necessary as cognitive impairment becomes more prevalent over time and patients who did not initially need LAR consent, may later need one.

Recruitment / Enrollment

Recruitment protocols should be piloted with older adults similar to those who will be enrolled to ensure that the process is acceptable to geriatric patients. Additionally, researchers should ensure that instructions, questionnaires, and forms are appropriate for older adults with visual or other sensory impairments and are easily comprehensible.⁵ Proxies, caregivers or other family or friends accompanying an older adult to the ED can assist with communication and support the patient in weighing the merits and disadvantages of participation. Finally, if the patient has a good relationship with the treating physician, engaging the physician in the recruitment process may be helpful.

Older adults often have longer ED visits than younger patients and therefore research staff may have more opportunities to approach them to offer participation in studies. If patients are unavailable initially because they are receiving tests or treatments, continuing to monitor until the patient can be approached should increase enrollment. Ensuring that the members of these programs have excellent skills in working with older adults and enrolling them into research studies will improve recruitment success.

Retention

Older adults have been successfully enrolled in prospective longitudinal studies with follow up rates comparable to younger populations. For example, the related CRASH⁵² and Older Adult CRASH⁵³ studies had similar follow up rates among the younger and older cohorts (91% for younger and 93% for older at 6 weeks and 89% vs. 87% at 6 months). Successful strategies include payment for participation and follow up and engagement of the patient’s healthcare team. Primary care providers and specialists can encourage participation and help locate the patient for follow-up.

As with younger adults, using multiple communication modalities for follow up will increase success. These include mailings, phone calls, texting, emails and online response forms. One previous study in the ED demonstrated that older adults are willing to utilize tablets to directly enter data, but showed poor accuracy.⁵⁴ Other studies have shown older adults can use technology with training including those with dementia.^55,56 Thus, the appropriateness of incorporating technology should be considered within the context of each study.⁵⁷ During recruitment, research staff should collect the names and contact information for alternate contacts such as a spouse, child, or friend. In addition to helping provide data at enrollment, these contacts may help encourage the completion of follow up assessments, provide new contact information should the older adult move or be hospitalized, and provide important outcome information such as unexpected death. For older adults with functional and/or cognitive impairment, a professional or family caregiver may be the appropriate follow-up contact, and successful retention relies on ensuring that this person is reachable and willing to continue to participate. Allowing a flexible follow up window (for example, a longer time period), coordinating with existing healthcare appointments and scheduling longer follow up appointments are techniques that have been successful in previous studies.⁸ Additionally, carefully considering transportation for subjects to participate in the study or offering home visits as part of the research follow up can enhance retention. Finally, there are no known harms of research unique to older adults. All standard human subject precautions should be undertaken with special attention to cognitive impairment.

Strategies - Taking Advantage of Resources and Tools

ED clinical researchers designing and launching studies that include older adults should take advantage of existing resources and tools at every stage of the process. We recommend the following:

• - Building relationships and collaborating with geriatricians and other disciplines and specialties within their institution focused on aging research or those who have experience with clinical research involving older adults.

• - Joining professional organizations such as the Society for Academic Emergency Medicine’s Academy of Geriatric Emergency Medicine and the American College of Emergency Physicians’ Geriatric Emergency Medicine Section to gather best practices and help with research challenges.

• - Collaborating with GEAR (https://gedcollaborative.com/research/gear/) and other NIH-funded networks to benefit from their infrastructure and funding. The aims for many of these networks focus on enhancing health equity for older patients and target new and early stage investigators. The NIH’s NIA, in collaboration with other groups, has created the Recruiting Older Adults into Research (ROAR) project⁵⁸ and the Clinician-Scientist Transdisciplinary Aging Research Coordinating Center (Clin-STAR) (https://clin-star.org/ ) to help encourage and facilitate research.

• - Exploring research-related resources offered by the American Geriatrics Society (AGS) (https://www.americangeriatrics.org/media-center/resource-links) and its partner journal, the Journal of the American Geriatric Society.

• - Utilizing the NIH Toolbox (https://www.healthmeasures.net/explore-measurement-systems/nih-toolbox) to identify screening tools.

• - Following physician scientists and other research organizations on Twitter to stay abreast of new research findings, grant opportunities, and to build a community of health science researchers dedicated to aging research.

Conclusion

Greater inclusion of older adults in emergency care research may be challenging, but will result in more rigorous emergency care research with enormous potential benefits for ED patients. Early attention to avoid the exclusion of older patients should be a priority. The NIH Inclusion Across the Lifespan policy¹⁴ will help ensure that this critical goal is met. Employing strategies to prioritize design and planning, recruitment, and retention will allow geriatric emergency care research to be as relevant and inclusive as possible.