Characteristics of Young Adults Admitted to Intensive Care Units at an Academic Health System Free

We obtained a data set from an academic health care system in the midwestern United States with ICUs in an academic hospital and 2 community hospitals. Clinical data were obtained through the health system’s department of enterprise analytics, and extracted data were deidentified using an institutional review board–approved honest broker. The project did not meet the institution’s definition of human subjects research. Data for older patients in the same health system were obtained from a clinical database (Vizient) for a similar time frame for comparison.

Data were obtained for young adults who were admitted to any ICU (medical, surgical, cardiovascular, neuroscience, transplant, and mixed medical-surgical) in the health system from August 2018 through December 2021. We used the National Cancer Institute definition of young adults as those aged 18 to 39 years old.¹⁰  Extracted data included demographic information, age at time of admission, admission and discharge International Classification of Diseases, Tenth Revision (ICD-10), diagnostic codes, discharge disposition, severity of illness (as entered in an electronic billing module [Epic]), length of ICU stay, days of mechanical ventilation, days of dialysis, days of extracorporeal membrane oxygenation (ECMO), code status, advanced care directive on file, palliative care consults, and hospice referrals.

Data were stored as Excel workbooks (Microsoft) and were imported into R, version 4.2.2 (The R Foundation), for analysis. Analyses included descriptive statistics and univariate and multivariate regression models. A type I error level of .05 was considered for hypothesis testing, including the Bonferroni adjustment for multiplicity wherever needed.

The following assumptions were made during analysis:

1. Repeat ICU admissions during the same hospitalization were combined and considered as a single encounter, and repeat ICU admissions during different hospitalizations were considered unique encounters.

2. The diagnostic code used for the patient was the primary ICD-10 code assigned at the time of hospital discharge, collapsed into the 22 ICD-10 diagnostic classification headings.

3. Admissions before March 16, 2020, were considered pre–COVID-19 pandemic, and admissions on or after this date were considered during the COVID-19 pandemic.

From August 2018 through December 2021, there were 3401 unique young adult ICU admissions. The mean age of the group was 29.7 years (SD, 6.0 years), and 58.6% were male. Most of the sample were White (50.7%), non-Hispanic (90.6%), single (81.1%), and with government insurance (63.3%). More than one-half (51.9%) were documented to have extreme illness, indicating highly complex health conditions requiring intensive care, as determined by the hospital’s coding system (3M Health Information Systems). The discharge ICD-10 diagnostic classification heading for nearly one-third (32.9%) of the sample was injury, poisonings, and trauma. Infections and endocrine, circulatory, and digestive disorders were common, each accounting for approximately 9% of classification headings. The mean length of stay in the ICU was 77.2 hours (SD, 82.1 hours), the mean duration of mechanical ventilation was 4.2 days (SD, 18.4 days), and the mean duration of ECMO was 7.3 days (SD, 103.1 days). About 72% of young adults were eventually discharged to home from the hospitalization that included the ICU admission, whereas 6.6% died during the hospitalization. Of the 223 patients who died, 38 (17.0%) received a palliative care consultation and 3 (1.3%) had a hospice referral.

The percentages of Black and Hispanic patients admitted to the ICU were higher among young adults than among older adults. Black and Hispanic young adults comprised 37.8% and 8.7% of the sample, respectively, whereas percentages for Black and Hispanic adults aged 40 to 64 years were 26.9% and 4.7%, respectively. For adults aged 65 years and older, these percentages were 14.9% and 2.0%, respectively. Young adults had a shorter stay in the ICU and lower mortality than older age groups: young adults, 77.2 ICU hours and 6.6% mortality; adults aged 40 to 64 years, 141.6 ICU hours and 10.9% mortality; and adults 65 years and older, 95.5 ICU hours and 13.4% mortality. More young adults were eventually discharged to home (71.8%) compared with the older age groups: 64.1% of adults aged 40 to 64 years and 45.5% of adults 65 years and older were discharged home.

We explored differences in demographic and clinical characteristics between the young adults who were admitted to ICUs before the COVID-19 pandemic and those admitted after the onset of the pandemic. As shown in Table 1, there were significant differences between the groups in race, ICU length of stay, severity of illness, and mortality. Concerning race, the percentage of Black young adults in ICUs increased and that of White young adults decreased from before to during the pandemic. The ICU length of stay, severity of illness, and mortality rate all increased. There were differences in discharge diagnosis ICD-10 diagnostic classification headings, with the most pronounced difference being an increase in the group diagnosed with injury, poisonings, and trauma after the onset of the pandemic. There were no significant differences in age, sex, ethnicity, marital status, or insurance status of young adults between these 2 periods.

In the overall sample, 2267 young adults (66.7%) with an admission to the ICU had no advance directive documented at any time during the hospitalization. The COVID-19 pandemic did not affect the advance directive documentation rate, as the proportion of young adults with no advance directive in the pre–COVID-19 period (1040/1579, or 65.9%) was not significantly different from that in the COVID-19 period (1227/1822, or 67.3%; P = .38). Among the young adults who had a documented advance directive (1134, or 33.3%), 787 (69.4%) had an advance directive documented at the time of hospital admission, 142 (12.5%) had an advance directive documented while in an acute care unit, and 205 (18.1%) had an advance directive documented during their ICU stay.

As shown in Table 2, there were differences in multiple demographic and clinical characteristics of young adults according to where their advance directive was documented. We used multinomial models to compare the likelihood of having an advance directive documented at the time of hospital admission versus while in the acute care unit or while in the ICU (see Table 3). Predictors of having an advance directive documented at the time of hospital admission included age (increasing likelihood by 0.04 per year), male sex (decreasing likelihood by 0.54), and White race (increasing likelihood by 0.79). We did not find significant predictors identified for having an advance directive documented while receiving care in an acute care unit. Predictors for those having an advance directive documented while in the ICU included male sex (decreasing the likelihood by 0.51), single marital status (decreasing likelihood by 0.43), and duration of ICU stay (increasing likelihood by 0.005 per hour) or of ECMO (increasing likelihood by 0.005 per day).

This retrospective study provides a snapshot of young adults admitted to ICUs in one health system throughout an approximately 3.5-year period. Males made up a disproportionate number of young adults admitted to the ICUs, higher than the health system’s catchment area (49.2%) and consistent with other ICUs.¹¹  More than 80% of the young adult sample was single, higher than the overall adult US population, of which approximately 50% to 60% are married or living with a partner.¹²  More than 90% of the sample had health insurance, mostly commercial or government, consistent with improvements noted for young adults after passage of the Affordable Care Act and expansions of its Health Insurance Marketplace, although at the time of this study, this state had not expanded their Medicaid program to include other low-income adults.¹³  Young adults differed from older adults with regard to race and ethnicity, ICU length of stay, discharge status, and mortality rate. The percentages of Black and Hispanic patients admitted to the ICU were higher among young adults than in older adults. The proportion of Black ICU patients in all adult age groups (young [18-39 years], middle-aged [40-64 years], and older [65 years and older]) was disproportionately higher than in the catchment area for the health system (14.4% Black), whereas the proportion of Hispanic ICU patients in every group was disproportionately lower than in the catchment area (11.7%). The shorter stays and lower mortality rates in young adults most likely reflect faster recoveries from acute events than in older adults as well as a different pattern of conditions in the younger population.

Nearly one-third of the sample’s diagnosis at discharge was categorized as injury, poisonings, or trauma, which parallels the 3 leading causes of death for young adults of unintentional injury, suicide, and homicide.⁵  West et al also reported these diagnoses for nearly one-half of young adults in their institution’s ICUs.⁷  These conditions reflect the public health crises in this metropolitan area related to homicides, nonfatal shootings, motor vehicle accidents, self-inflicted injuries, and drug overdoses, all of which increased during this period. Many of these indicators are known to be high for males, Black persons, and those aged 18 to 39 years.^14,15  Endocrine, circulatory, and digestive disorders, and infections made up another one-third of discharge diagnostic categories.

Our analysis suggests differences in the critically ill young adult cohort during the COVID-19 pandemic compared with the prepandemic years. Young adults with COVID-19 had substantial rates of adverse outcomes, including ICU admissions, exacerbated by underlying chronic conditions.¹⁶  The number of Black young adults admitted to ICUs increased during the COVID-19 years. While we do not know the reasons for the increased ICU admissions for Black young adults during the pandemic years, the increase may reflect evidence that Black people experienced higher rates of COVID-19 cases and deaths than White people, especially early in the pandemic.^17,18  This finding may also reflect higher rates of trauma, suicide, and poisonings attributed to psychological stress during COVID-19. Structural vulnerability, defined as negative health outcomes due to the intersection of socioeconomic position and power hierarchies, is an important consideration when interpreting this finding.¹⁹ 

One-third of young adults (n = 1134) in this sample had an advance directive established before or during the hospitalization, similar to the overall rate of 36.7% of adults in the United States. Advance directive completion rates are higher among those who have been ill than among healthy adults.²⁰  These rates are higher than in adults in ICUs in other parts of the world, where advance directive completion in ICUs may be as low as 1%.²¹  In our study, most young adults with a documented advance directive had the directive in place at the time of hospital admission. We speculate that these young adults most likely had an existing chronic condition or prior ICU stay that prompted discussions about preferences for health care decisions.

There is limited research about predictors of advance directive completion among young adults or the timing of when advance directives are completed. Predictors of documented advance directives for young adults in our study included increasing age, being White, being female, having a partner, having a longer ICU stay, and receiving ECMO. This finding is consistent with evidence from other studies of seriously ill adults, which suggests that discussions about goals of care are more common among women, older patients, and those with more comorbidities and less common among patients from underrepresented racial or ethnic groups.²² 

The presence of advance directives is associated with increased use of hospice and palliative care.²³  Among patients who died, fewer young adults had palliative care consultations (17%) and hospice referrals (1.3%) when compared with older adults who died in ICUs at the same institution, among whom palliative consultations averaged 23% to 50% and hospice referrals ranged from 19% to 27%.²⁴  The lower rates in young adults may be due to uncertain prognoses for their conditions, and young age may be perceived as advantageous for survival and recovery. Although palliative care may have been integrated by ICU providers, there may have been missed opportunities to relieve patient and family distress with consultations from palliative care experts.²⁵ 

A major limitation of this inquiry relates to the deidentified nature of the database, which prohibited examining specific information about diagnoses, illness trajectories, experiences, and outcomes for this sample of critically ill young adults. This sample was from a single health system in the midwestern United States and may not translate to ICU settings in other regions where care practices and populations are different. Our analysis was limited to variables available in the data set; we made several assumptions as outlined in the methods. Despite these limitations, this examination of young adults admitted to ICUs provides insight into this distinct subset of ICU patients and how critical care for them may differ from that for older adults in ICUs.

Health care planning during a young adult’s treatment for an acute critical illness or injury may be especially challenging. Because most young adults are single and unlikely to have an existing health care power of attorney (HCPOA) or advance directive, it is critical for providers to identify the person who is legally empowered to make decisions in the event the young adult is unable and to readdress the HCPOA when the young adult regains their decision-making capabilities. It may be that the young adult’s legal decision maker, such as a parent, may not be the person who is closest to the young adult or who best knows the patient’s values and wishes. Facilitating family support for the young adult while in the ICU may also present challenges related to visitation rules that may not accommodate the patient’s own young children or younger siblings.⁶ 

Discharge planning for critically ill young adults can be complex, especially for those recovering from a life-changing injury or illness, who need rehabilitation to regain function or need extensive support from a caregiver. The subset of young adults recovering from an overdose or gun violence have complex postdischarge needs that are often unmet after the young adults return to high-risk communities.^26–28  Coordination of various resources and with a variety of providers requires collaboration directed by a primary care provider or specialist after patients leave the hospital setting and perhaps for years as they age into mature adults. The financial impact of hospitalization and continued treatment after a serious illness or injury can have long-lasting effects on young adults and their families.²⁹ 

Given the limited literature focused specifically on young adults in ICUs, future research is needed to understand many aspects about care for this population and how their conditions and care may differ from those of older and younger patients. Better descriptions are needed for young adults admitted to ICUs with specific acute and chronic conditions, the trajectories and outcomes of their hospitalizations, and their specific care needs, especially their care needs after the hospitalization. Research should focus on the needs and experiences of young adults from their own perspectives as well as disparities in outcomes related to race, gender, or ethnicity. Education is needed to promote “structurally competent” clinicians who can explore the structural factors (social, environmental, and economic) that influence young adults’ experiences with the health care system. Research is necessary to identify effective interventions that address critical health infrastructure at the community level to reduce health disparities.³⁰  Better approaches are needed to promote advance care planning, the establishment of advance directives, and the use of palliative care and hospice services in this population, especially with patients from marginalized groups.

This exploratory study showed that young adults were commonly admitted to ICUs with life-threatening injuries, poisonings, and trauma, with a disproportionate number of single, Black, and male patients. The mortality rate in this population was low, and most had relatively short stays in the ICU and were eventually discharged to home. Despite receiving treatment for a life-threatening condition, most young adults did not have a documented advance directive. Because minimal research has focused on young adults receiving intensive care, their experiences and needs are not well understood. Future studies should examine whether these young patients who survive a critical illness or injury receive effective and age-appropriate care, both in the hospital and after discharge.

This work was performed at the University of Wisconsin-Milwaukee, Medical College of Wisconsin. The authors would like to acknowledge Karen A. Erickson, BS, for her assistance with the data.