Rehabilitation and Social Determinants of Health in Critical Illness Recovery Literature: A Systematic Review Authors: Lindsey E. Fresenko, Charlotte Rutherfurd, Lauren E. Robinson, Cayla M. Robinson, Ashley A. Montgomery-Yates, Rachel Hogg-Graham, Peter E. Morris, Tammy L. Eaton, Joanne M. McPeake, Kirby P. Mayer Published: 2024 ABSTRACT OBJECTIVES:Patients who survive critical illness navigate arduous and disparate recovery pathways that include referrals and participation in community-based rehabilitation services. Examining rehabilitation pathways during recovery is crucial to understanding the relationship on patient-centered outcomes. Furthermore, an understanding of social determinants of health (SDOH) in relation to outcomes and rehabilitation use will help ensure equitable access for future care. Therefore, there is a need to define and understand patient care pathways, specifically rehabilitation after discharge, through a SDOH lens after surviving a critical illness to improve long-term outcomes.DATA SOURCES:MEDLINE, PubMed, Web of Science Core Collection (Clarivate), the CINAHL, and the Physiotherapy Evidence Database.STUDY SELECTION AND DATA EXTRACTION:A systematic review of the literature was completed examining literature from inception to March 2024. Articles were included if post-hospital rehabilitation utilization was reported in adult patients who survived critical illness. Discharge disposition was examined as a proxy for rehabilitation pathways. Patients were grouped by patient diagnosis for grouped analysis and reporting of data. Two independent researchers reviewed manuscripts for inclusion and data were extracted by one reviewer using Covidence. Both reviewers used the Newcastle-Ottawa Scale to assess risk of bias.DATA SYNTHESIS:Of 72 articles included, only four articles reported detailed rehabilitation utilization. The majority of the studies included were cohort studies (91.7%) with most articles using a retrospective design (56.9%). The most common patient population was acute respiratory diagnoses (51.4%). Most patients were discharged directly home from the hospital (75.4%). Race/ethnicity was the most frequently reported SDOH (43.1%) followed by insurance status (13.9%) and education (13.9%).CONCLUSIONS:The small number of articles describing rehabilitative utilization allows for limited understanding of rehabilitation pathways following critical illness. The reporting of detailed rehabilitation utilization and SDOH are limited in the literature but may play a vital role in the recovery and outcomes of survivors of critical illness. KEY POINTS Purpose: The systematic review aims to quantify critical illness survivors’ rehabilitation utilization following discharge and to identify the reporting of SDOH in the literature. Findings: Only four articles reported detailed rehabilitation utilization. Discharge disposition was the only data available and used to describe pathways. Variability existed within SDOH reporting with the most commonly reported SODH being race/ethnicity and insurance status. Meaning: With limited reporting of detailed rehabilitation utilization and SDOH in critical illness literature, defining pathways and connecting to patient outcomes are difficult to achieve. After surviving critical illness, patients transition through facilities, healthcare providers, and rehabilitative settings in their recovery. Many survivors of critical illness develop or continue to experience physical, cognitive, and mental health concerns, commonly referred to as post-intensive care syndrome (PICS) (1, 2). Consequently, ICU survivors require ongoing care to monitor and address these healthcare needs. The amount of care received or required after hospital discharge for optimal patient outcomes is unknown. In addition to medical, rehabilitation, and mental health needs, there are social needs that may be preexisting to critical illness or exacerbated by new deficits upon returning home. These social needs are individual-level factors directly impacting patients. Social determinants of health (SDOH) are the societal and environmental level classification, which are defined as the environments in which we work, play, and live that impact quality of life and health outcomes (3). SDOH are estimated to impact as much as 50% of the county-level variation in health outcomes, in contrast to clinical care, which is thought to impact around 20% of health outcomes (4). Many patients experience a change in occupation or loss of job, which in turn impacts healthcare coverage and income following critical illness (5). The loss of earnings after discharge can be substantial, ranging from 30,000 to 40,000 dollars over a 5-year period (5). Previous research has demonstrated that SDOH and a patient’s social needs have a relationship with utilization of post-hospital healthcare services and patient outcomes in the general population (6). Patients often receive rehabilitative services such as physical therapy, occupational therapy, and speech language pathology during their hospitalization; however, some patients require facility admissions and ongoing rehabilitative services at home after hospital discharge. The outcomes associated with rehabilitation for survivors of critical illness and evidence continues to expand to demonstrate value and critical need (7). Several studies demonstrate the relationship between inpatient rehabilitation services and positive patient outcomes (8). After hospital discharge, receiving rehabilitation and at higher frequency, is associated with a lower risk of 10-year mortality in survivors of critical sepsis (9). Experts in critical illness recovery and rehabilitation have defined the importance and areas of focus for critical care survivors’ rehabilitation (10). However, the dosage of rehabilitation and the impact on long-term outcomes after hospital discharge is unclear. In order to plan for future interventional research in this area, it is crucial to understand care pathways and rehabilitation use in the post-hospital discharge period. Furthermore, a thorough understanding of the SDOH in relation to outcomes and rehabilitation use will help ensure the design of equitable access for future care. Therefore, there is a need to define and understand patient care pathways, specifically rehabilitation after discharge, through a SDOH lens after surviving a critical illness to improve long-term outcomes. Our systematic review aims to quantify critical care survivors’ rehabilitation utilization reported in the literature following discharge. Our secondary aim is to identify the reporting of SDOH in the literature. METHODS The purpose of our search was to identify recovery pathways for survivors of critical illness after hospital discharge through rehabilitation utilization reported in the literature. Two medical librarians assisted with the development of our search strategy and performed the final search. This systematic review was registered through PROSPERO (CRD42022292907) in December 2021 before the first search being completed (11). Our systematic review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (12). Search Strategy Two librarians collaborated to create an initial MEDLINE PubMed systematic search in February 2022. The two librarians consulted with the researchers to form a relevant keyword list and to workshop the PubMed search. After the PubMed search was finalized, the librarians translated the PubMed search into three additional databases: Web of Science Core Collection (Clarivate), the CINAHL via Elton B. Stephens Company Host, and the Physiotherapy Evidence Database (Neuroscience Research Australia and the University of Sydney). Full database coverages can be found in the Supplemental Materials (http://links.lww.com/CCX/B434) of the article (11). The PubMed search was translated into the three additional databases using the Systematic Review (SR)-Accelerator’s Polyglot (13). All four searches were reported by adhering to the PRISMA Searching Extension (14). The references were exported to EndNote on February 24, 2022. An update of the search was conducted on May 30, 2023, and March 12, 2024, and results were uploaded in Covidence. No limits were included in the searches. The full search strategies for each database can be found in the Supplemental Materials (http://links.lww.com/CCX/B434) and have been deposited into SearchRxiv. After exportation to EndNote, the citations were uploaded to Covidence. Covidence is a systematic review software used to help aid with the screening processes (15). Covidence also provides a deduplication service to eliminate all duplicates of articles in the library. The librarians ensured that the deduplication process was successful. Upon deduplication, the researchers began conducting their initial title and abstract screening. The librarians added the full-text of articles to Covidence following selection for full-text screening. After full-text screening was complete and the articles were selected for final inclusion, the librarians used the SpiderCite tool from SR-Accelerator to retrieve articles for citation chaining (16). The articles that cited the articles chosen for inclusion as well as the references of the chosen articles were reviewed. The final number of articles considered at each stage are included in the PRISMA flowchart (Fig. 1). Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart. *Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). **If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. PEDro = Physiotherapy Evidence Database. Article Selection We included articles that were written in English language and included adult patients (≥ 18 yr old) surviving an admission to the ICU. We defined mixed populations as articles containing both critically ill patients admitted to an ICU and patients hospitalized outside of the ICU. Because we wanted to focus on critically ill patients only, articles were only included if the author could provide ICU specific patient data after being contacted. If ICU specific patient data were not obtained, the article was excluded. We only included articles that reported discharge disposition data as a proxy if rehabilitation utilization was not reported (Table 1). TABLE 1.Inclusion and Exclusion CriteriaInclusionExclusionn ≥ 6 subjectsCase report/case series < 6 subjectsAdults (> 18 yr old)Pediatric subjects onlyICU admission or critically ill patients (any diagnosis)Acutely hospitalized patientsFollow-up after hospital dischargeMixed populationsaSystematic reviewsQualitativeNonpeer reviewed articles (conference abstracts, perspective papers, dissertations, and magazines)aMixed populations—authors were contacted for ICU patients only, articles were excluded if authors could not provide data or did not respond. Inclusion and Exclusion Criteria Mixed populations—authors were contacted for ICU patients only, articles were excluded if authors could not provide data or did not respond. We excluded all nonpeer reviewed content in addition to case series and case reports with less than six subjects, abstracts without full-text, mixed populations, and qualitative methodology (Table 1). With our initial goal of quantifying post-acute rehabilitation, we excluded studies regarding admission to neurologic ICUs due to the established discharge disposition and recovery pathways for these patients (17–20). For abstract and title review, four reviewers independently screened all studies using Covidence (Melbourne, VIC, Australia) (15). Two reviewers independently completed full-text review to identify eligible studies for data extraction. Any discrepancies were resolved by consensus. One of the authors completed all the data extraction with a second review to verify correct data and extract additional variables. Risk of Bias Risk of bias was independently assessed by two independent appraisers using the Newcastle-Ottawa Scale for cohort studies. The score consists of domains to assess quality and risk of bias of the articles. The following areas were assessed: the selection of study groups, the comparability of the groups, and the ascertainment of either the exposure or outcome of interest (21). Statistical Analysis To capture rehabilitation used after discharge, discharge disposition was used for statistical analysis as a surrogate for rehabilitation utilization, which is the closest proxy to describe rehabilitation utilization and pathway. Descriptive statistics were analyzed to describe study characteristics such as study type, methodology, patient population represented, and location. Patients were grouped according to diagnosis to describe the frequency of discharge location reported. Pooled estimates were created for the percent of patients who were discharged to each location reported in the article with a 95% CI (Supplemental Fig. 1, http://links.lww.com/CCX/B434). In this review, we also sought to understand if there was a relationship with the patient population and discharge disposition. To test the hypothesis that patient population had a relationship with discharge disposition, we used a Kruskal-Wallis test. We grouped patients by admitting diagnosis and analyzed each discharge disposition. All statistics were completed on IBM SPSS (Version 28, Armonk, NY). For our secondary aim, descriptive statistics were completed to describe the frequency of SDOH reporting in the studies included (Table S1, http://links.lww.com/CCX/B434). A heat map was used to describe the frequency of reporting by patient population (Fig. 2). Articles that included at least one SDOH were included in the analysis and heatmap. GraphPad Prism 9.4.1 (GraphPad Software, Boston, MA) was used to create all the figures. Figure identifying social determinants of health (SDOH). Figure outlining social determinants of health reported in each article included by patient diagnosis represented by each article. Articles were included in the figure if at least one SDOH were reported. Each color represents a different diagnosis group. RESULTS Our initial search identified 12,736 articles with 5,003 duplicates removed from February 2022 to inception. We screened 7733 for inclusion through title and abstract with 49 articles included in final data extraction and data analysis from original search. An updated search to include articles from March 2022 to May 2023 identified 1211 articles for title and abstract screening with an additional 11 articles included in the final data extraction. The most recent search from May 2023 to March 2024 identified 866 articles for title and abstract screening with eight articles included. Four additional articles from hand searching were included for a total of 72 articles (Fig. 1). There were 857,846 patients represented in these studies and 765,474 survivors of critical illness. Of the 72 studies included, 66 (91.7%) were cohort studies. There was a mix of retrospective (n = 41, 56.9%) and prospective (n = 31, 43.1%) studies. Over half of the studies were conducted in the United States (n = 38) followed by Europe (n = 17) and Canada (n = 7) (Table S2, http://links.lww.com/CCX/B434). From the risk of bias using the Newcastle-Ottawa Scale, no studies had a high risk of bias identified. A summary of the studies included is presented in Table 2. TABLE 2.Study CharacteristicsStudy Characteristicsn = 72, n (%)Study type Cohort study66 (91.7) Randomized control trial4 (5.6) Case series2 (2.8)Location of study United States38 (52.8) Europe17 (23.6) Canada7 (9.7) Australia6 (8.3) South America3 (4.2) Asia1 (1.4)Methodology Retrospective41 (56.9) Prospective31 (43.1)Patient population Acute respiratorya37 (51.4) Chronic respiratory/mixed9 (12.5) Older adults10 (13.9) Sepsis7 (9.7) Trauma6 (8.3) Otherb7 (9.7)Discharge disposition In-hospital mortality41 (56.9) Long-term acute care hospital33 (45.8) Skilled nursing facility30 (41.7) Inpatient rehabilitation facility51 (70.8) Home with home health18 (25.0) Home with no services59 (81.9)aAcute respiratory = patients with COVID-19, acute respiratory distress syndrome, acute respiratory failure, and post-lung transplant.bOther = cardiac, oncology, and postoperative infection.Discharge disposition represents how many articles report specific discharge disposition out of all the articles. Study Characteristics Acute respiratory = patients with COVID-19, acute respiratory distress syndrome, acute respiratory failure, and post-lung transplant. Other = cardiac, oncology, and postoperative infection. Discharge disposition represents how many articles report specific discharge disposition out of all the articles. All 72 studies were included in all of the descriptive and associative statistical analysis (9, 22–92) (Table S3, http://links.lww.com/CCX/B434). Only four articles out of 64 explicitly reported rehabilitation utilization (43, 60). These four studies were comprised of patients with acute respiratory distress syndrome (n = 839), acute lung injury (n = 205), and older adult (n = 145) populations, where two studies relate to one patient cohort. The data reported included a median skilled nursing facility (SNF) length of stay (LOS) of 60 days (interquartile range [IQR], 16–104 d) and inpatient rehabilitation facility (IRF) LOS of 21 days (IQR, 9–30 d) up to 2 years after discharge for survivors of acute lung injury (60). A second article reported a median IRF stay of 20 days (IQR, 10–30 d) and a SNF stay of 50 days (IQR, 23–134 d) for survivors of acute respiratory distress syndrome (59). The LOS reported in the third study, which focused on older adults discharged to an IRF, was a median of 3 weeks with a range of 1–24 weeks (43). The fourth study reported a median LOS for COVID-19 patients discharged to an IRF of 119 days with an IQR of 11–31 days (84). The number of patients with a discharge disposition reported was 353,075. For discharge disposition, 66,131 patients (18.7%) were discharged to a facility, including SNF, IRF, and long-term acute care hospital (LTACH) and 286,944 (81.3%) discharged home (Table 3). When analyzing the number of studies that reported discharge disposition, the most common was home with no services (81.9%) or inpatient rehabilitation (70.8%). Likewise, home with home health services (25%) was the least commonly reported discharge disposition. The average percent of patients discharged to each location per the included articles are as follows: 43.5% home with no services, 30.1% inpatient rehabilitation, 16.2% SNF, 12.5% LTACH, and 8.7% home with home health. TABLE 3.Discharge Disposition by Patient DiagnosisDischarge Disposition, n = 353,075Long-Term Acute Care Hospital, 1,753 (0.50%)Skilled Nursing Facility, 18,635 (5.28%)Inpatient Rehabilitation Facility, 45,743 (12.96%)Home With Home Health, 20,649 (5.85%)Home With NS, 266,295 (75.42%)Diagnosis, n (%) Acute respiratorya1,125 (64.18)2,100 (11.27)39,048 (85.36)1,905 (9.23)73,255 (27.51) Chronic respiratory212 (12.09)16,353 (87.75)6,002 (13.12)14 (0.07)104,494 (39.24) Geriatric28 (1.60)89 (0.48)36 (0.17)2,170 (10.51)580 (0.22) Sepsis179 (10.21)7 (0.04)68 (0.15)839 (4.06)6,371 (2.39) Trauma205 (11.69)8 (0.04)549 (1.20)0 (0)3,416 (1.28) Otherb4 (0.23)78 (0.42)40 (0.09)15,721 (76.13)78,179 (29.36)Age, mean (sd)59.6 (9.5)60.9 (5.9)57.4 (11.7)59.3 (7.0)60.3 (10.6)Sex, males (%)57.0653.8358.8357.0660.77NS = no services.aAcute respiratory = patients with COVID-19, acute respiratory distress syndrome, acute respiratory failure, and post-lung transplant.bOther = cardiac, oncology, and postoperative infection. Discharge Disposition by Patient Diagnosis NS = no services. Acute respiratory = patients with COVID-19, acute respiratory distress syndrome, acute respiratory failure, and post-lung transplant. Other = cardiac, oncology, and postoperative infection. Grouped by Patient Population In contrast to the number of studies reporting discharge disposition, patient level data are presented with grouped analysis. Most of the reported discharge data were represented in the acute respiratory diagnosis group, which accounted for the majority of studies represented in this review (51.4%). The majority of patients represented were discharged home with no services (85.7%; Table 3). LTACH was the least common discharge disposition representing only 0.32% of the patients (Table 3). There was no significant relationship between population diagnosis and discharge disposition. Additionally, there was no difference between patient age and sex between discharge locations (Table 3). Social Determinants of Health Our secondary aim was to describe the frequency of SDOH reported in the critical care literature represented in this study. We used the framework set out by the Centers for Disease Control and Prevention (CDC) to classify the SDOH into the following categories: Individual-level factors, economic stability, education access and quality, healthcare access and quality, neighborhood and built environment, and social and community context (3). Using this framework, we categorized the SDOH in Figure 2. The majority of the SDOH reported were in the individual level (race) and healthcare (insurance) categories. The most commonly reported SDOH was race (n = 31) followed by insurance (n = 10) and education level (n = 10). Tobacco use (n = 5) and residential area (urban vs. rural) (n = 4) were the least reported SDOH. None of the articles in our results reported mode of transportation. Additionally, no articles demonstrated a relationship across rehabilitation utilization and SDOH. A summary of the results is presented in Table S3 (http://links.lww.com/CCX/B434) and Figure 2. DISCUSSION The aim of this review was to quantify the amount of rehabilitation services provided after discharge for critical care survivors. There was heterogeneity in the reporting of the rehabilitation course after discharge. From our search, four articles reported detailed post-acute rehabilitation services utilization. Due to the small number of articles reporting our primary outcome, discharge disposition was used as a surrogate measure to represent different potential rehabilitation pathways, that is, discharging to a rehabilitation center or discharging home with home health rehabilitative services. The lack of reporting of post-hospital rehabilitation data precludes our ability to synthesize the potential disparate recovery pathways faced by survivors of critical illnesses. Furthermore, we are unable to examine how SDOH influence participate in rehabilitation. In relation to our secondary aim, we found the reporting of SDOH was sporadic. The CDC framework for grouping SDOH may elude to what SDOH are easily accessible and which are currently emphasized or collected for survivors of critical illness (3). All of the subjects in our articles were survivors of critical illness; therefore, insurance status should be accessible. However, the variability in reporting of insurance data may be related to the health system model of each country the patients were hospitalized in. The majority of articles were based in the United States where insurance is not universal and highly variable (93) (Table S2, http://links.lww.com/CCX/B434). Based on the nature of critical illness and the large number of patients who were admitted as a result of acute respiratory failure in the studies included, the lack of data on tobacco use was surprising. Residential location was also under reported therefore utilizing zip code or established indexes to identify areas of social need or vulnerability was not possible with included literature. Recent literature has demonstrated the crucial impact of social support in this patient population; therefore, the reporting of marital status was expected. Because our population is critically ill and hospitalized, transportation may not be a focus during hospitalization. However, we anticipated that a few studies would address transportation due to the need to have safe and reliable transport to attend outpatient follow-up. Transportation to rehabilitation as well as other medical services may be critical for patients living in rural regions. Previous research suggests that patients living greater than 45 miles away from the main clinic are less likely to follow-up (94). Thus, transportation is a key determinant that likely influence the rehabilitation pathway. A recent randomized controlled trial of a “mobile” PICS clinic reduces the need for transportation bringing services directly to the patient. The study, however, failed to demonstrate a significant benefit on quality of life (95). In contrast, a nurse navigator primarily providing telecommunications was beneficial for reducing mortality and readmission in first 30 days after hospital discharge for patients surviving sepsis (96). A previous systematic review reported that post-hospital rehabilitation is related to quality of life, but it remains unknown why some patients participate, which may be related to healthcare system (United States vs. other countries), providers (referrals to rehabilitation), or patient (financial barriers) (97). Research on follow-up (medical or rehabilitation) for ICU survivors is limited, and more work is necessary to understand the relationship between SDOH and rehabilitation delivery. We also must highlight that studies conducted in countries outside the United States will have differing SDOH important for their region. Thus, it is expected in this field that SDOH have some variability. In general, however, understanding which SDOH are important in recovery and how they impact patient outcomes can only happen with increased assessment and reporting. After a patient is discharged from the hospital, healthcare providers recommend follow-up care and referral to other healthcare services aligned with patient need. Expert consensus highly recommends physical rehabilitation after discharge, specifically delivered in the community (i.e., home health or outpatient) for ICU survivors (10, 98, 99). However, there are several challenges with the referral process on the day of hospital discharge, which does not always equate to participation once patients get home. One study suggests that when discharge recommendations made by physical therapist are not followed that patients are 2.9× more likely to be readmitted (100). Again, emphasizing referral does not always equate to participation. Clinically, follow-up care may not always be accessible due to financial status, insurance type, and loss of healthcare insurance coverage. Proximity to services, which can exacerbate health conditions, is also an important consideration (6). Additionally, post-ICU clinics were developed to address the specific impairments faced by patients after critical illness to mitigate any gaps in care. In relation to rehabilitation services, referrals for additional rehabilitation were provided to 30% of patients who attended indicating a need (101). Thus, more work is necessary to determine who may benefit from rehabilitation, and how we address the SDOH to ensure our patients in need get the services. The expertise of clinicians and patient education are important; however, patients may be more concerned about their unmet social needs and quality of life than visiting additional healthcare providers (102). The unmet social needs may serve as a barrier to any further medical intervention leading to poorer health outcomes (103). The results of this review demonstrate that data on social needs are not systematically collected or accounted for as a factor for patients’ overall health and recovery. Recent literature also reflects this trend of SDOH not always being screened and rarely addressed (104). A hospital admission is a pivotal point in a person’s life to address unmet social needs and prevent further decline or readmission. Recent studies have included SDOH in their predictive models, which highlights the importance of social needs on negative health outcomes (105–107). Therefore, personalized care is necessary to address not only the clinical components of care but recognize the impact unmet social needs may have on health outcomes and recovery. Standardized care will not be the standard of care for every patient. To improve patient outcomes, a holistic approach to patient care is required with assessment of everything that can contribute to their health. Each patient’s social needs and SDOH impact their care, hospital course, recovery, and outcomes. The inability to work or loss of employment contribute to a change in income and insurance after surviving a critical illness (108). Within healthcare systems, there is a realization that SDOH impact recovery and patient outcomes (109). More clinicians and researchers are advocating for screening and addressing SDOH during critical illness (104). Many tools to screen SDOH exist and can be used in healthcare settings (110). However, barriers exist including time, fear of addressing sensitive topics, and lack of confidence of providers (111). Therefore, the application in the critical care population is limited. Future research should examine optimal methods for assessing the SDOH in the context critical care recovery setting. With increased assessment and evidence, intervention and implementation studies using existing social and community services to impact recovery can be completed to highlight the direct impact on long-term recovery. Our review has several strengths. The searched conducted is reproducible and was executed by two medical librarians. All articles included had a low risk of bias and transparent methodology. The authors who completed screening were all clinicians and the two authors who completed full-text review are both rehabilitative professionals. Our search was also broad as we did not focus on one population within the ICU, which allows representation of different diagnoses and generalization. Finally, the data extracted from the articles were completed by one author to be consistent and maintain rigor. Our review does have notable limitations. We expected to find literature that reported rehabilitation utilization. However, the data and ability to follow the trajectory of patients were challenging. Discharge disposition was used as a surrogate to rehabilitation utilization, which was only present in four articles. Discharge locations provide a glimpse of rehabilitative services received after discharge but not the frequency, duration, or movement through additional rehabilitative settings. Patients who were discharged home may also represent a population that used outpatient or home health services depending on how the data was collected and reported in each article. Discharging to home may also signify limited access to rehabilitative services due to patient’s geographic location or medical coverage. Furthermore, our search may have missed literature pertinent to our analysis to provide a more thorough description due to the aims of our original search strategy. Many studies were close to meeting the inclusion criteria or included as reference material but were unable to be included in the data analysis because the articles did not include discharge disposition. Finally, we only included articles that were completed and reported in English. INTERPRETATION Rehabilitation services received by ICU survivors and their SDOH have a role in recovery, but the impact of these factors continues to remain unclear. Clinicians should assess patient’s SDOH and be mindful of patient resources for receiving rehabilitation after discharge when making discharge recommendations for post-acute rehabilitation services. Our review highlights the need for more data and evidence to understand rehabilitation use by ICU survivors and how SDOH might influence access and outcomes. ACKNOWLEDGMENTS We thank two clinicians who assisted with review articles, Nicholas D. Fresenko, PT, DPT, and Madison S. Tackett, DO. We appreciate and acknowledge their time and commitment to this study. Supplementary Material