Keywords
delirium - disparities - social determinants of health - racial/ethnic - socioeconomic
status - contextual factors
Delirium and Marginalized Populations: Frameworks
Delirium and Marginalized Populations: Frameworks
Delirium is a neuropsychiatric syndrome that represents global brain dysfunction and
is frequently observed in hospitalized patients, with an incidence range of 11 to
47% in the general hospital wards and up to 87% in critically ill patients.[1]
[2]
[3]
[4] It is characterized by acute onset of deficits in attention, awareness, and cognition
that fluctuate in severity over time[1] and portends both poor short- and long-term outcomes. In the short term, this includes
longer length of hospital stay and increased rate of hospital-acquired complications;
in the long term, this includes increased risk of dementia and earlier mortality.
A growing body of research has underscored the importance of nonmedical factors such
as education, housing, and income/job security on the risk of morbidity and mortality
across a vast spectrum of diseases, including delirium.[4]
[5]
[6]
[7]
[8]
[9] The social–ecological and social determinant of health frameworks can clarify how
these nonmedical factors impact delirium—its diagnosis, treatment, and outcomes. The
social–ecological model recognizes that an individual's health is impacted not only
by individual-level personal behaviors and genetics, but also by a complex interplay
of institutional, community, and policy-level factors ([Fig. 1]).[10]
[11]
[12] For example, at the organizational level of hospital institutions, environmental
and physical factors such as visitor restrictions and disparate use of physical restraints
among different racial or ethnic groups can influence the development and trajectory
of delirium.[13]
Fig. 1 The social–ecological model grounds the individual in a broader environment that
involves an interplay of individual, interpersonal, institutional, community, and
policy factors that can inform both a framework for identifying the extent of disparities
and solutions to address them.
Social determinants of health are the conditions in which people are born, grow, live,
work, and age. These nonmedical factors include education and socioeconomic status
(SES) that have the ability to impact health outcomes.[14]
[15] Especially pertinent to delirium is the concept of the “cognitive reserve model,”
defined as the brain's ability to deal with an insult through preexisting cognitive
capacity or compensation through alternative processes.[16]
[17] A patient's “cognitive reserve” may be impacted by social determinants of health
such as level of education and occupational role.[5]
[17] Social determinants of health can also influence a patient's social engagement (i.e.,
volunteering and participating in group activities) and social capital (i.e., resources
derived from social networks such as the availability of caregivers), which have been
shown to reduce the incidence of delirium.[5]
[18]
[19]
[20] For example, these resources are not uniformly available across racial/ethnic groups
who may have less flexibility and accommodations in their work environments, creating
caregiving challenges.[21]
This article uses these two frameworks to review the existing literature on disparities
in delirium, building on seven previously proposed domains relevant to delirium in
older adults: demographic factors, social and economic factors, physical and mental
well-being, social capital, activities and social engagement, built and social environment,
and life course factors.[5] We then review the extant literature and risk factors across the continuum of care:
prehospitalization, perihospitalization involving emergency care, hospitalization
including inpatient and/or intensive-care-unit setting, as well as posthospitalization
transition to a nursing home or community ([Fig. 2]). The current literature is nascent, although there are studies that have identified
a higher incidence of delirium among non-Hispanic Black and Hispanic/Latinx patients.[22] In instances where the literature is less established, we draw from literature in
other areas of medicine, within and outside of neurology, to discuss how disparities
may manifest. Given that much of the relevant literature focuses on racial and/or
ethnic disparities, we do so as well, although we recognize that “race” actually serves
as a proxy for racism. We also recognize that disparities can occur across other dimensions
of an individual's identity, that is, gender/sexual orientation,[23] immigration status,[24] and ability/disability (vision and hearing deficits[5]), which can intersect and exert a compounding negative influence for those with
multiple marginalized identities.
Fig. 2 The role of social determinants of health in delirium across the continuum of care.
Prehospital Risk Factors for Delirium
Prehospital Risk Factors for Delirium
The Cognitive Reserve Model and Delirium Risk
A patient's cognitive reserve—the brain's ability to remodel after a medical insult—is
impacted by social determinants of health across a patient's life course (i.e., including
childhood and early adulthood experiences, and factors like education and SES).[25] Patients with lower cognitive reserve have increased brain vulnerability and are
subsequently more likely to develop delirium later in life.[17] Conversely, high-reserve individuals are better able to cope with disruptions to
brain networks and less likely to develop delirium.[26]
Studies in the United States have shown that early life experiences (life course factors[5]) such as educational attainment,[27] neighborhood socioeconomic disadvantage,[28] parental educational attainment,[29] and paternal occupation[30] influence rates of delirium, possibly via lowering cognitive reserve capacity. Notably,
it is not simply a matter of “years of education” but also educational quality, with
one study showing attending higher-quality schools is associated with improved later-life
cognitive performance among older adults.[31] Life experience is shaped by SES. SES can be conceptualized as education, income,
wealth and assets, employment, or a combination of these at individual or neighborhood
levels. SES also broadly influences a patient's cognitive reserve and, as a result,
the risk of delirium.[32]
There are multiple proposed mechanisms that may explain the association between low
SES and the risk of delirium. One mechanism is that higher cognitive reserve, found
in higher SES populations, protects against the deliriogenic effects of inflammation.
For example, in a study of geriatric adults without dementia, higher vocabulary knowledge
(a measure of cognitive reserve) was generally protective and modified the association
between C-reactive protein and postoperative delirium.[33] Another proposed mechanism is that higher cognitive reserve is associated with the
ability to compensate in the face of neuropathological burden or medical insult through
the recruitment of novel brain networks.[26] Finally, lower SES populations, often those who are racial and/or ethnic minorities,
are also at higher risk for developing cognitive decline and dementia, which in turn
puts them at higher risk of delirium. Patients with dementia need fewer precipitating
factors to increase their likelihood of developing delirium.[34]
In support of this theory, studies have shown that the risk for dementia and cognitive
decline occurs at higher rates in Black and Hispanic populations,[35] as well as in patients with low SES in terms of occupation, education, and employment
status.[36] Other studies have also identified immigrant communities, across racial and/or ethnic
groups, as being at higher risk of cognitive decline owing to unique barriers such
as heightened social isolation exacerbated by the immigration experience, low acculturation
to their host country, and increased trauma exposure.[37] Immigrants often have lower social capital and emotional health as a result of social
isolation.[5]
[38] Therefore, adverse social risk factors increase the risk of cognitive decline among
marginalized communities, including racial and/or ethnic minority populations, immigrant
patients, and patients with lower SES, which increases the risk of developing delirium.
Since SES and other life course factors are racially patterned due to historic and
ongoing societal discrimination, residential segregation, and racism, there exist
multiple potential interventions at the policy level (i.e., further investment in
disadvantaged schools and neighborhoods) that could mitigate these risks later in
life ([Table 1]).
Table 1
Examples of social determinants of health-related interventions for addressing delirium
risk and care
|
Social determinant of health category
|
Potential interventions
|
|
Education
|
- Early childhood education
- Investment in schools in disadvantaged communities
|
|
Employment
|
- Enhanced paid sick leave and family leave policies to support caregivers of patients
- Enhanced family caregiver reimbursement policies
|
|
Social and community context
|
- Investment in the community, working to improve access to infrastructure such as
grocery stores and public transport
- Improved urban planning, i.e., increased street lighting or signage, increased green
spaces to promote physical activity, and reduce risk of falls
|
|
Healthcare access and quality
|
- Expansion of health care insurance eligibility, i.e., Medicaid expansion, universal
health care, coverage for immigrant populations
|
The Role of Emergency Management Services in Delirium Risk and Care
More proximal to an individual's presentation to the hospital encounter is care management
by Emergency Management Services (EMS) personnel—emergency medical technicians and
paramedics. There are limited data examining the practices and outcomes of EMS personnel
in delirium. However, some of the documented disparities in care at this stage may
put marginalized patients on a steeper trajectory to the development of delirium once
hospitalized. For example, studies have shown that EMS providers inadequately treat
racial and/or ethnic minority patients' pain, which is a strong risk factor for developing
delirium.[5]
[39]
[40] In a cohort study of 4.7 million patient encounters from the National Emergency
Medical Services Information System from 2019 to 2021, racial and/or ethnic minority
populations—Black, Native American, Alaskan Native, Asian, and Hispanic—were less
likely to have a recorded pain score after an acute traumatic injury.[41] Even when they had a recorded pain score, Black patients were half as likely to
receive pain medications as compared with White patients.[41]
[42]
[43] Similarly, in a 2015 to 2017 study of data from the Oregon Emergency Medical Services
Information System, patients with limited English proficiency (LEP) were less likely
to have a recorded pain score, most likely due to the lack of professional interpreter
use, as only 5.5% EMS encounters used a professional interpreter.[44] While EMS providers commonly note their inability to communicate with a patient
due to a language barrier, they do not take steps to overcome this barrier.[44] Other studies have found racial and/or ethnic disparities in EMS recognition of
symptoms of acute myocardial infarction or stroke, which could delay care, increase
the length of hospitalization, and potentially increase the risk of delirium.[45]
[46]
EMS personnel have identified the treatment of agitation in the setting of hyperactive
delirium as particularly challenging. One qualitative study involving focus groups
with EMS providers found that although they felt comfortable recognizing delirium,
they desired additional screening tools and guidance on the management of agitation
due to delirium during ambulance rides.[47] EMS personnel frequently treat agitation using both chemical (ketamine, benzodiazepines,
antipsychotics) and physical restraints.[48] There is a lack of data analyzing social factors and disparities involved in the
use of physical and chemical restraints by EMS, although racial and/or ethnic disparities
have been identified in restraint use in emergency departments (EDs), as discussed
further below[49]; presumably the same mechanisms of racism at the interpersonal, institutional, and
structural levels exert influence on EMS providers similar to ED providers even if
that has yet to be documented. Furthermore, it is important to highlight the use of
“excited delirium” by law enforcement to justify violence against Black and Hispanic
patients outside of the hospital setting. The term “excited delirium” is not recognized
as a valid diagnosis and is not to be confused with agitation in the context of delirium.[50]
Delirium Risk Factors in the Emergency Department
Delirium Risk Factors in the Emergency Department
Patients often spend a prolonged period of time in the ED from triage to either admission
or discharge. The risk of delirium during this time is impacted by social determinants
of health.
Triage Factors
Racial and/or ethnic minority patients wait longer before being seen by a provider
in the ED. This is significant for the development and trajectory of delirium as studies
have shown that total ED length of stay, including waiting room time, treatment time,
and boarding, is significantly associated with the development of delirium in older
patients.[51]
[52] Time to triage has become one of the most important determining factors for a patient's
waiting time in the ED, and based on the analysis of 2013 to 2017 data from the National
Hospital Ambulatory Care Survey, Black, Hispanic, and Asian patients were more likely
to wait for “extreme” wait times (>139 minutes) compared with White patients, even
when adjusted for age, sex, comorbidity status, health system indicators, and geographic
region.[53] This can potentially be attributed to interpersonal racism leading to these patients
being skipped over despite the same acuity triage level.[54] Black patients have also been shown to be triaged to a lower acuity than White patients,
although this is seen most frequently with subjective diagnoses like “shortness of
breath” and less with observable diagnoses like “altered mental status.”[55]
[56] Given the documented disparities in wait times and the association between long
wait times and delirium, it is highly plausible that increased delirium in marginalized
patients is in part driven by their treatment during triage, although future studies
will be needed to confirm the association.
Emergency Department Factors
Even once triaged, patients suffer from disparities in ED experiences based on insurance
and SES. For example, patients on Medicaid are placed in ED hallway beds at higher
rates than those with private insurance, which can increase the risk of delirium.[51]
[57] In one study of two academic EDs from 2013 to 2016, patients with Medicaid or self-pay
had 22% greater odds of being placed in a hallway bed after adjusting for various
ED factors and patient demographics.[58] This trend was similarly found in a separate study looking at an academic and a
community ED from 2017 to 2020.[59] As insurance status is likely unknown at the point of triage, these data serve as
a proxy for broader socioeconomic factors; for example, individuals are potentially
placed in hallways based on provider biases from appearances alone.
Another potential delirium risk factor is the use of both physical and chemical restraints
in the ED. Both are highly influenced by a patient's race, SES, and medical history.
Physical restraints include mittens or limb restraints, while chemical restraint refers
to the use of psychoactive and sedating medication. Although there are no published
data on the influence of physical restraints in the ED in the development of delirium,
studies in both the ICU[60] and patients with stroke[61]
[62] show that physical restraints increase the risk of delirium. Chemical restraints
are often used alongside physical restraints, though few studies examine the use of
chemical restraints alone. Most medications used as chemical restraints (intramuscular
benzodiazepines or antipsychotics) are known to be delirium inducing. A patient is
more likely to be physically restrained if they are Black,[49]
[63]
[64]
[65]
[66] have Medicaid,[63]
[66] are unhoused,[63]
[66] or have a diagnosis of bipolar or psychotic disorder.[49]
[63]
[64] In a systematic review/meta-analysis involving 10 articles, Black patients had a
31% higher relative risk of being physically restrained than White patients in the
ED, even when adjusting for various sociodemographic and medical indicators.[49] One study of all ED visits at a regional healthcare network in New England found
that Black, non-Hispanic patients had significantly higher adjusted odds of receiving
chemical restraints compared with White, non-Hispanic patients.[67]
Taken together, there are a host of potentially addressable delirium risk factors
in the ED that are influenced by race and/or ethnicity and SES.
Hospital Course of Patients with Delirium
Hospital Course of Patients with Delirium
Diagnosis of Delirium
Delirium can be challenging to diagnose due to its multifactorial nature, variable
manifestations (including three subtypes of delirium—“hyperactive,” “hypoactive,”
and “mixed”), and fluctuating time course. There are multilevel factors that could
contribute to disparities in the diagnosis of delirium.
At the individual level, screening for delirium also necessitates the investigation
of possible underlying organic disease processes that may precipitate this altered
state, including existing cognitive impairment, sepsis, metabolic disturbances, and
respiratory failure.[68]
[69] Multiple studies have highlighted disparities across several social risk factors
in the timely diagnosis and treatment of these predisposing illnesses. For example,
Black patients are more likely to experience delays in the initiation of sepsis treatment
compared with White patients, and hospitals that treat predominantly Black patients
are less likely to provide timely antibiotic therapy.[70] An Argentinian-based study identified longer delays in sepsis treatment for patients
admitted to public hospitals compared with private hospitals; these patients were
also significantly more likely to be from lower SES and educational backgrounds.[71] Racial and/or ethnic biases in the interpretation of pulse oximetry accuracy have
been associated with greater occult hypoxemia in Black and Hispanic patients with
COVID-19, contributing to unrecognized or delayed recognition of eligibility for therapies.[72] Black patients admitted to the ICU for heart failure were less likely than their
White counterparts to be seen by a cardiology specialist, which was associated with
higher mortality.[73] These findings strongly suggest that racial and/or ethnic and socioeconomic disparities
likely exist in the timely and appropriate diagnosis and treatment of underlying disease
processes that precipitate delirium.
Disparities in delirium incidence and diagnosis likely also exist between LEP patients
compared with English-proficient patients, with one study citing a twofold increase
in risk for LEP patients.[74] A systematic review of 26 internationally published articles on clinical psychiatric
care for LEP patients concluded that language discordance and the use of untrained
interpreters can cause errors in assessing mental status, delusions, and disorganized
thinking, findings that have been replicated across clinical settings.[75]
[76]
[77] Despite this, professional interpreters continue to be underutilized.[78] Furthermore, immigrants from an LEP background may present with different predominant
delirium symptoms than nonimmigrant patients. For example, one large retrospective
study in Australia found that immigrants were more likely to present with agitation/aggression,
potentially related to frustration with a language barrier.[79] Nursing staff in one qualitative study reported lower confidence in administering
the Confusion Assessment Method (CAM) screening test for delirium in LEP patients,
citing uncertainty about adapting the tool to languages other than English, difficulty
assessing baseline mental status, and technical difficulties with virtual interpreters.[80]
[81] However, one single-center study that made a concerted effort to include patients
with LEP demonstrated no statistical difference in the incidence of delirium based
on LEP status. Interpretation is limited because the sample size was small and delirium
was assessed only at one time point, which may have led to bias.[82] Overall, these data suggest possible underdiagnosis or delay in the diagnosis of
delirium among LEP patients which requires further dedicated attention and study.
At the hospital level, there may be differences in delirium diagnosis due to patient
volume, academic versus community setting, and proportion of patients from racial
and/or ethnic minority backgrounds. For example, one retrospective study in the United
States showed that patients in small, nonteaching hospitals were diagnosed with delirium
less frequently compared with those at large, urban teaching centers.[83] Based on national survey data, 75% of hospitals in the United States are nonteaching
hospitals, and they account for 50% of admissions, indicating that a vast number of
patients are inherently at risk of underdiagnosis of delirium due to lack of proximity
to teaching hospitals in urban areas.[84]
Further hindering the diagnosis of delirium in marginalized populations is that delirium
can be time-consuming to diagnose. Several focus group studies have identified time
and heavy workload among nursing staff as diagnostic barriers.[85]
[86] However, nurses serving marginalized patients may be the ones least equipped to
diagnose the condition. For example, there is evidence that institutions that serve
predominantly Black patients tend to employ less trained nurses, and that those nurses
work fewer hours than nurses employed by institutions primarily serving White patients.[87]
[88] Additionally, a study on nursing staffing ratios for post–cardiac arrest patients—who
are predisposed to delirium—found that each additional patient per nurse lowered the
odds of survival for Black patients more than White patients.[88]
[89] Together, these findings suggest that systemic factors like nursing hiring and staffing
practices may contribute to racial and/or ethnic disparities in delirium diagnosis
and management.
Management of Delirium
The management of delirium is highly heterogeneous across institutions and healthcare
providers given its poorly understood pathophysiology and multifactorial nature.[90] It includes both prevention and treatment, which can involve various social factors
that could result in disparate outcomes among patients.
In terms of prevention, there is consensus that multicomponent nonpharmacologic interventions
such as the “Hospital Elder Life Program” and “ABCDEF bundle” are highly effective
in preventing delirium.[91]
[92] These approaches encompass frequent reorientation and cognitive stimulation, pain
control, sleep optimization, regular sedation weaning trials, and early mobilization.
As discussed earlier, underutilization of professional language interpreters may limit
the use of reorientation techniques with patients to prevent delirium, introducing
disparities in the prevention for LEP patients.[78] Although there is a lack of research on racial and/or ethnic and other disparities
in these interventions pertaining to delirium outcomes, there is evidence of racial
and/or ethnic disparities in these interventions more generally. Multiple studies
have demonstrated that patients identifying as racial and/or ethnic minorities receive
less acute pain therapies than their White counterparts, leading to suboptimal pain
control and consequently likely increasing the risk of delirium incidence.[93]
[94] Early mobilization as part of delirium prevention typically invokes consultation
of physical and occupational therapy services (PT and OT), and studies have found
that racial and/or ethnic minorities and LEP patients received fewer PT and OT consults.[95] This suggests potential mechanisms for likely racial, ethnic, and linguistic disparities
in the prevention of delirium.
Once delirium is recognized and diagnosed, patients with delirium are often managed
with physical and/or chemical restraints due to perceived concerns about their risk
of self-harm. Although concerns about self-harm can be valid, the use of physical
restraints has been widely shown to be a precipitating factor for delirium, and recent
clinical trials have shown that antipsychotic use is ineffective in the prevention
and treatment of ICU delirium.[61]
[96]
[97] Despite these concerns, there is evidence to suggest a lack of standardized selection
criteria, inconsistent documentation in the chart, and insufficient reevaluation for
discontinuation of physical restraints.[60]
[90]
[98] This invites bias and consequently racial, ethnic, and socioeconomic disparities
in the treatment and prevention of delirium.
In fact, just as prior studies have shown disparities in the use of restraints in
ED settings across multiple dimensions of an individual's identity—by race,[49] but also modified by age and gender[65]
[67]
[99]
[100]—at least one study in the inpatient setting found that patients with LEP had significantly
increased risk of receiving both physical restraints and antipsychotic medications.[80] While these studies were not conducted with a focus on delirium, it is reasonable
to suspect that the use of physical and chemical restraints is similar for patients
with hyperactive delirium that manifests as agitation, across hospital settings.
Hospital Discharge Destination
Hospital Discharge Destination
Patients can be discharged to a variety of settings, including homes, rehabilitation
centers, or nursing homes. Social determinants of health and risk factors can impact
discharge destination.
Nursing home placement is common after hospitalization. The 2013 National Inpatient
Sample showed that 22.3% of patients were discharged from the hospital to postacute
care settings with Skilled Nursing Facilities being the most common institutional
setting.[101] Based on the National Post-Acute and Long-term Care Study, there are 1.3 million
nursing home residents in the United States, and this number is likely to rise as
the population ages.[102] Like other healthcare services, nursing home placement is often determined based
on insurance status which frequently manifests as segregated facilities with unequal
resources. For example, homes with ≥50% of Black patients had more Medicaid-funded
residents and had lower ratios of registered nurses (RNs) and medical aide (MA) hours
per resident.[87] In a study analyzing the 2014 Medicare admissions to nursing homes, Black and Hispanic
patients with Alzheimer's dementia were less likely to go to nursing homes with special
dementia care units and more likely to go to for-profit institutions.[103] These for-profit institutions yielded poorer outcomes for their patients.[104]
[105] This unequal care leads to medical complications. For example, the prevalence of
pressure ulcers among high-risk nursing home residents is higher in Black residents
compared with White residents, which increases a resident's risk of delirium.[106] While there are no studies that specifically look into the disparities of delirium
diagnosis and treatment in marginalized patients in nursing homes specifically, it
is known that delirium is underdiagnosed in nursing homes and long-term care facilities,
possibly due to lack of proper training or lack of documentation.[107]
[108] It is reasonable to hypothesize that less resourced facilities do a less adequate
job in the diagnosis and management of delirium, and that marginalized populations
are likely disproportionally effected. Studies will be needed to investigate the extent
to which this occurs.
As in other healthcare settings, racial and ethnic minority patients are put in physical
restraints at a higher rate than White patients in nursing homes, which increases
their risk of delirium. Chemical and physical restraints continue to be used in nursing
homes to varying degrees, with an estimated 9% using physical restraints and 26 to
27% using chemical restraints in the United States in 2003.[109] The use of physical restraints, especially chair restraints, in nursing homes is
a significant risk factor for delirium.[110] In an analysis of the National Nursing Home Survey 2004, Black nursing home residents
were more likely to be physically restrained (bed rails, side rails, and trunk restraints)
compared with White nursing home residents.[111]
Just as with nursing homes, racial and/or ethnic disparities have been identified
in rehabilitation settings.[112]
Patients from neighborhoods with lower SES are less likely to be discharged home.
While this may be attributed to a desire to facilitate safe discharges, a home discharge
is less deliriogenic than other institution-based discharge destinations.[113] However, even when patients are discharged home, differential neighborhood physical
disorder (presence of trash, vandalism, observed or perceived crime) and social cohesion
(connectedness and solidarity within groups in a community) can affect recovery. There
is limited research on these posthospitalization factors with respect to delirium
risk and recovery, as well as long-term cognitive trajectory, representing an area
of future study in better understanding and ultimately mitigating delirium disparities.
Recommendations for Future Research and Action
Recommendations for Future Research and Action
The role of social determinants of health in the development, diagnosis, treatment,
and outcome of delirium is understudied, particularly in the prehospital and posthospital
settings. However, existing literature suggests there are multiple potential areas
for intervention to prevent and treat delirium based on social determinants of health
and the social–ecological frameworks, like increased investment in schools to improve
later life cognitive function, as well as more proximal factors to patient care, like
institutional policies and individual provider training to mitigate disparate restraint
use.[31]
[49] We outline some potential interventions for various social determinants of health
categories in [Table 1]. Using the social–ecological model, we can also consider multilevel interventions.
At the interpersonal level, research and associated actions should be inclusive of
family caregivers who are impacted by delirium and can play a facilitative role in
recovery.[114]
[115]
[116] At the institutional level, other actions could include increasing collection and
monitoring of disparity metrics as part of broader quality improvement efforts (i.e.,
around restraint use, and triage decisions), improving access to and training around
the use of interpreters with patients with delirium, and increasing workforce diversity
to facilitate race and language-concordant care. At the policy level, other actions
could include increasing resource allocation to nursing homes serving minoritized
populations or enhancing reimbursement of interpreter services including for EMS.
Lack of action, whether that is increased research to expand the limited research
on delirium disparities or targeted multilevel interventions, would simply mean that
modifiable risk factors are left unaddressed and an opportunity for improving patient
care is missed.
