Disorders of Consciousness in Acute Care: Application of the CRS-R and Respiratory, Swallowing, and Discharge Outcomes at a Level 1 Trauma Hospital Authors: Sara Penrod, Zoe Samson, Stephanie C Chan Published: 2026 ABSTRACT AbstractObjectiveTo exemplify the use of the Coma Recovery Scale-Revised (CRS-R) during critical acute care and to characterize associated functional and rehabilitative trajectories of patients with disorders of consciousness (DoC). Although most research emphasizes postacute assessment and recovery, early rehabilitation in the initial days and weeks after injury remains unexplored.DesignRetrospective, observational cohort studySettingLevel 1 Trauma and Certified Stroke Center acute care hospitalParticipantsA total of 68 adults admitted with a severe brain injury with an initial Glasgow Coma Scale score of ≤8.Main Outcome MeasuresOutcomes were assessed across 3 domains: (1) consciousness, via the CRS-R; (2) respiratory function, including cough categorization, tracheostomy status, and decannulation readiness trials (speaking valve or capping); and (3) swallowing/eating, measured by the Functional Oral Intake Scale (FOIS) and engagement in evaluations for oral diet advancement.ResultsInitial CRS-R assessments showed 2.9% (n=2) in coma, 52.9% (n=36) in unresponsive wakefulness syndrome (UWS), 27.9% (n=19) in minimally conscious state minus (MCS−), 13.2% (n=9) in MCS plus (MCS+), and 2.9% (n=2) emerged. By discharge, no patients remained in coma; 2.9% (n=2) were in UWS, 17.7% (n=12) MCS−, 13.2% (n=9) MCS+, and 41.2% (n=28) emerged. Among those emerging, 53.6% did so within 30 days. Tracheostomy was required in 73.5% (n=50); most participated in capping (72%, n=36/50) and speaking valve trials (88%, n=44/50) while cannulated, and 68% (n=34/50) were decannulated. Instrumental swallowing evaluations were completed in 42.6% (n=29), with 39.7% (n=27) discharged on oral diets. Overall, under half (39.7%, n=27) were discharged to specialized or rehabilitative settings.ConclusionPatients with DoC make demonstrable improvements in consciousness, respiratory function, and swallowing during acute care, challenging prevailing assumptions that rehabilitation potential begins later in the recovery trajectory. These findings underscore the importance of initiating structured assessments and interventions early to optimize outcomes and inform discharge planning. Future research should focus on the development and evaluation of standardized screening protocols for identifying DoC in acute care, as well as examining national screening, assessment, and treatment practices.KEYWORDS: Brain injuries, Decannulation, Disposition, Disorders of consciousness, Dysphagia, Rehabilitation, Tracheostomy Disorders of consciousness (DoC) after severe acquired brain injury span a continuum from coma through emergence.1 Despite their clinical significance, standardized DoC assessments are infrequently used in intensive and acute care settings,2 limiting diagnostic accuracy, prognostication, and monitoring of recovery. Treatment approaches also lack consensus among intensive care unit (ICU) and rehabilitation providers, despite evidence supporting early and intensive rehabilitation in non-DoC populations.3,4 Existing literature predominantly focuses on postacute rehabilitation settings,4, 5, 6 or broader brain injury cohorts,3,7,8 leaving a limiting understanding of recovery trajectories during initial hospitalization. This study examines recovery during acute care hospitalization among patients with DoC across 3 domains: consciousness, breathing, and eating. We emphasize early use of the Coma Recovery Scale-Revised (CRS-R) for screening and longitudinal assessment before inpatient rehabilitation. Outcomes include levels of consciousness, tracheostomy status and progression, instrumental swallowing assessments, and discharge disposition. We propose that systematic CRS-R administration during acute hospitalization is feasible and may improve diagnostic accuracy, inform clinical decision-making, and support disposition planning. Consciousness Outcomes for hospitalized patients with DoC are highly variable and influenced by clinical, social, and system-level factors. Limited availability of postacute resources, including closures of skilled nursing facilities (SNFs), has contributed to prolonged hospitalizations.9 Underdiagnosis of DoC in acute care may result in undertreatment and inappropriate withdrawal of care in patients with recovery potential.10 Early, structured assessment and rehabilitation are therefore critical. Breathing Patients with DoC frequently require tracheostomy, which is associated with increased costs and longer hospital stays.11,12 Although common, tracheostomy placement carries risks such as stenosis and tracheomalacia.13 Although timely decannulation is recommended,14 patients with DoC are often excluded from standard pathways.15 Speaking valves (SV) and swallowing trials remain inconsistently implemented due to historical wakefulness requirements16 and practice variability. Emerging evidence supports earlier use of SV and incorporation of physiologic markers such as cough characteristics to guide decannulation readiness.17 Eating Swallowing rehabilitation is inconsistently used in DoC despite evidence of preserved reflexive swallowing even in low-awareness states.18 Dysphagia and aspiration risk are common,19 and instrumental assessments, such as a modified barium swallow study (MBS) or flexible endoscopic evaluation of swallowing, are recommended before oral intake.20, 21, 22 However, eligibility criteria often require sustained arousal and oral acceptance, limiting access. Early swallowing intervention may support recovery and signal emerging consciousness.19 Study Aim This study aimed to characterize assessment practices and rehabilitation progress among hospitalized patients with DoC during early acute care, with a focus on trajectories in consciousness, breathing, and eating. Methods Design, setting, and participants This retrospective cohort study was conducted at a level 1 trauma center and approved by the hospital’s Institutional Review Board (IRB 2089066-1). Adults (≥18y) diagnosed with coma, unresponsive wakefulness syndrome (UWS), or minimally conscious state (MCS) after traumatic brain injury, stroke, or hypoxia/anoxia were included. All patients were admitted to the neurocritical or trauma/surgical ICUs with an initial Glasgow Coma Scale score of ≤8, indicating a severe impairment of consciousness. This analysis includes the first 68 patients enrolled between October 2019 and April 2025. Patients were identified through physiatry consultation and assessed by a trained speech-language pathologist specializing in DoC. All CRS-R evaluations were conducted off sedation. Patients with pre-existing DoC or prior brain injury were excluded. Measures Demographics included age (18-40, 41-60, and >60y to reflect phases of early, middle, and older adult development), sex assigned at birth was reported as a male/female binary, length of stay, and etiology (trauma, stroke, and hypoxia/anoxia). Comorbidities were characterized as none versus 1-2, including neurologic, pulmonary, cardiac, or psychiatric, existing conditions. Level of consciousness was assessed using the CRS-R, the gold standard evaluation tool.23 Assessments were performed once patients were medically stable and off sedation, typically in a sitting position whenever possible; those unable to tolerate upright positioning were assessed semi-upright or supine. Due to clinical variability and the presence of comorbidities in this population,24 CRS-R assessments were conducted once or twice weekly, depending on the patient’s medical condition, consistent with international assessment patterns in DoC.25 CRS-R findings were categorized as coma (lowest), UWS, lower-level MCS (MCS−) without language preservation, higher-level MCS with language preservation (MCS+), and emergence from MCS (EMCS) (highest). The MCS+/MCS− distinction was operationalized by Bruno et al26 and refined by Thibaut et al,24 carrying prognostic and rehabilitative relevance in this population. Time to emergence was defined as days from hospital admission to the first EMCS classification. Of note, patients received physical therapy, occupational therapy, and speech-language pathology services at variable frequencies, in addition to physiatry follow-up and coordinated interdisciplinary care involving nurses, registered dietitians, social workers, and other disciplines. Variables describing participants’ respiratory characteristics and hospital progress included the presence of coughing at admission, categorized as voluntary, reflexive, neither, or both. The presence or absence of moderate to copious secretions at admission was noted. A binary variable indicated whether the patient had a tracheostomy. For tracheostomized patients, additional variables captured whether a SV (Passy Muir) trial was performed, if trials were tolerated, and if any adverse events occurred while the tracheostomy was in situ. Other variables noted whether tracheostomy capping trials were attempted, whether capping was tolerated, and whether adverse events occurred. Tolerance was defined by the absence of the following during occlusion trials: respiratory distress, respiratory arrest, or death. Adverse events were defined as any of the after while the tracheostomy was in situ: the need for tracheostomy upsize, reintubation, transfer back to ICU, respiratory arrest, or death. A separate variable noted whether the patient was ultimately decannulated during hospitalization. Time to decannulation and CRS-R scores at decannulation were recorded. Functional eating ability was evaluated using the validated Functional Oral Intake Scale (FOIS),27 a 7-point scale where higher scores reflect greater oral intake. FOIS scores were documented at both hospital admission and before discharge. An additional variable captured whether patients underwent MBS fluoroscopic swallowing assessment. Categorical data on CRS-R findings and tracheostomy status (categorized as no tracheostomy, cuff inflated, cuff deflated, SV, capped, or decannulated) concurrent with the initial MBS provided information on consciousness level and respiratory support at the time of the study. Another variable indicated whether a trial of oral (per os [PO], oral intake by mouth) intake was performed during hospitalization, along with CRS-R score and tracheostomy status at the time of the initial PO trial. Finally, diet orders at discharge were documented, specifying the presence and type of solids and/or liquids. Discharge location was characterized as specialized programs for DoC or traumatic/acquired brain injury, skilled nursing or rehabilitation facilities, home, hospice, or death. Analysis Descriptive statistics were used. CRS-R trajectories were cross-tabulated, and relationships between age, etiology, and outcomes were explored using SPSS version 30.0.0.0.a Results A total of 68 patients with severe acquired brain injury and an initial Glasgow Coma Scale score ≤8 were included. Mean age was 46 years (SD, 18.9), with 45.6% (n=31) aged 18-40 years, 26.5% (n=18) aged 41-60 years, and 27.9% (n=19) >60 years. The cohort was predominantly male (63.2%, n=43). Average hospital length of stay was 75.6 days (median 53, SD, 80.2) with an average ICU stay of 25.6 days (median 21, SD, 18.9) and average stay on acute care floors of 50 days (median 29.5, SD, 78.4). Etiologies included stroke (42.6%, n=29), traumatic (36.8%, n=25), and hypoxic/anoxic injury (20.6%, n=14). Most patients (64.7%, n=44) had no significant premorbid neurologic, pulmonary, cardiac, or psychiatric diagnoses (table 1). Table 1Characteristics of patients with DoC, n=68Table 1 dummy alt textVariableEffective nPercentageMean (Median; SD)Age (y, continuous)6846.0 (44; 18.9)Age (y, categorical) 18-403145.6 41-601826.5 >601927.9Sex assigned at birth Female2536.8 Male4363.2Length of hospital stay (d)6875.6 (53; 80.2)Length of ICU stay (d)6825.6 (21; 18.9)Length of acute unit stay (d)6850 (29.5; 78.4)Pathogenesis Hypoxia/anoxia1420.6 Stroke2942.6 Trauma2536.8Premorbid conditions* 04464.7 1-22435.3Initial CRS-R score686.4 (5; 4.2)Final CRS-R score6813.5 (14.5; 7.6)Change in CRS-R score687.1 (5; 6.5)Initial CRS-R FindingsComa22.9Unresponsive wakefulness3652.9Minimally conscious (−)1927.9Minimally conscious (+)913.2Emerged from MCS22.9Final CRS-R findingsComa00Unresponsive wakefulness1927.9Minimally conscious (−)1217.6Minimally conscious (+)913.2Emerged from MCS2841.2Time to emerge (d, continuous)2841 (30; 34.9)Time to emerge (d, categorical)28≤30 d1553.6%31-60 d1035.7%61-90 d13.6%>90 d27.1%DispositionDoC program1014.7TBI/ABI program1116.2Inpatient rehab68.8Skilled nursing913.2Home913.2Hospice1116.2Deceased1217.6Abbreviations: ABI, acquired brain injury; TBI, traumatic brain injury.⁎Neurologic, pulmonary, cardiac, psychiatric. Characteristics of patients with DoC, n=68 Abbreviations: ABI, acquired brain injury; TBI, traumatic brain injury. Neurologic, pulmonary, cardiac, psychiatric. Initial CRS-R assessments were completed after a physiatry consultation, and patients were off sedation. A total of 2.9% (n=2) of patients were categorized as being in coma, 52.9% (n=36) in UWS, 27.9% (n=19) in MCS−, 13.2% (n=9) in MCS+, and 2.9% (n=2) had EMCS. The median CRS-R score at initial evaluation was 6.4 (range, 1-19). By the time of discharge, no patients remained in coma; 2.9% (n=2) were in UWS, 17.7% (n=12) in MCS−, 13.2% (n=9) in MCS+, and 41.2% (n=28) in EMCS. Among the 28 patients who eventually reached EMCS status, about half (53.6%, n=15) emerged within the first 30 days of hospital admission, 35.7% (n=10) between 31 and 60 days, 3.6% (n=1) between 61 and 90 days, and 7.1% (n=2) after 90 days (table 1). Cross-tabulations of initial and final CRS-R scores demonstrated improvement in consciousness across the cohort. Both patients who were in coma at admission progressed to UWS. Of the 36 patients initially in UWS, 25% (n=9) emerged by discharge. Among those admitted in MCS+, 77.8% (7 of 9) reached EMCS. Emergence was most frequent in the youngest age group (18-40y), though improvements were observed across all age categories (fig 1). By discharge, approximately half of the patients with stroke (44.8%, 13 of 29), trauma (52%, 13 of 25), and hypoxia/anoxia (50%, 7 of 14) showed emergence or progression to higher levels of consciousness (fig 2). Changes in consciousness across age groups measured by CRS-R findings. Initial and final CRS-R findings refer to the first and last CRS-R assessments obtained during hospitalization. Changes consciousness by neurologic pathogenesis measured by CRS-R findings. Initial and final CRS-R findings refer to the first and last CRS-R assessments obtained during hospitalization. Tracheostomy was performed in 73.5% of patients (n=50) (fig 3). Of those, 88% (n=44) completed SV trials, 72% (n=36) capping trials, and 68% (n=34) were decannulated. Eleven patients (16.2%) underwent instrumental swallowing assessments (MBS) while tracheostomized; among these, 44.8% (n=5) were later decannulated. At the time of decannulation (n=34), CRS-R scores spanned the full range of recovery, where 11.8% (n=4) were in UWS, 17.7% (n=6) in MCS−, 29.4% (n=10) in MCS+, and 41.2% (n=14) had achieved EMCS. Time to decannulation varied, where about a quarter (23.5%, n=8) of patients were decannulated within the first 30 days of their admission, 52.9% (n=18) between 31 and 60 days into their admission, and the remaining quarter (23.6%, n=8) were decannulated past the 60-day mark (fig 3). Respiratory characteristics of patients with DoC flowchart. CRS-R at decannulation indicates the CRS-R score most closely preceding a patient’s decannulation. CRS-R assessments were conducted once or twice weekly, depending on the patient’s medical condition. Injury to decannulation was calculated as the date of decannulation subtracted from the hospital admission date. Tolerance was defined by the absence of the following during occlusion trials: respiratory distress, respiratory arrest, or death. Adverse events were defined as any of the following while the tracheostomy was in situ: the need for tracheostomy upsize, reintubation, transfer back to ICU, respiratory arrest, or death. All patients were nil per os, no oral intake by mouth (NPO) on admission. At discharge, 47.1% (n=32) remained at FOIS level 1, receiving exclusively nonoral nutrition (fig 4). The remainder showed progression across the FOIS scale: 16.2% (n=11) reached level 2 (oral intake only in structured settings), 13.2% (n=9) level 3 (consistent oral intake with supplemental feeding), and 1.4%-14.7% reached levels 4-7, indicating varying degrees of functional oral intake. Among the 41 patients still NPO at discharge, most were in UWS (46.3%, n=19) or MCS− (26.8%, n=11), although a subset had reached MCS+ (14.6%, n=6) or EMCS (12.2%, n=5). Among the 41 patients discharged NPO, 14 (34.15%) had received a PO trial, and 27 (65.85%) did not have a PO trial. Among the 41 patients discharged NPO, 6 (14.63%) had participated in MBS, and 35 (85.37%) did not. One was assessed via flexible endoscopic evaluation of swallowing whereas in UWS, PO trials were not offered because of poor secretion management and diminished responsiveness to stimulation. Swallowing and eating characteristics of patients with DoC. 1. No oral intake. 2. Tube dependent with minimal/inconsistent oral intake. 3. Tube supplements with consistent oral intake. 4. Total oral intake of a single consistency. 5. Total oral intake of multiple consistencies requiring special preparation. 6. Total oral intake with no special preparation, but must avoid specific foods or liquid items. 7. Total oral intake with no restrictions. Discharge locations varied: 14.7% (n=10) were transferred to specialized DoC programs (of which 8 were trauma patients and 2 were stroke patients), 16.2% (n=11) to skilled rehabilitation centers with brain injury programs, 8.8% (n=6) to inpatient rehabilitation facilities, 13.2% (n=9) to SNFs, 13.2% (n=9) to home with caregivers, and 16.2% (n=11) to inpatient hospice. Care was withdrawn, and patients died in-hospital in 17.6% (n=12) of cases (table 1). Discussion To our knowledge, this is the first study to examine the use of the CRS-R in conjunction with rehabilitative characteristics of hospitalized patients with DoC early in acute care. As access to postacute services—including SNFs, long-term acute care hospitals, and inpatient rehabilitation programs—declines,9 the need for accurate recognition and management of patients with DoC in the acute setting becomes increasingly critical. Barriers to postacute placement may prolong hospitalization and delay access to specialized care, underscoring the importance of optimizing early assessment and care pathways. Despite high medical complexity, all patients in our cohort underwent CRS-R assessments, and many engaged in early, foundational rehabilitation activities, including SV and capping trials, decannulation, standardized swallowing assessments, and therapeutic oral intake. These findings parallel postacute reports and demonstrate the feasibility of initiating structured assessment and intervention during acute hospitalization. Routine CRS-R use during acute care represents a key contribution of this study and offers a scalable approach for broader implementation. All patients demonstrated some improvement in consciousness, with no patients discharged or dying in a comatose state. More than half progressed to a higher level of consciousness (MCS or emerged), which has been associated with more favorable long-term functional outcomes.28,29 Recovery varied by etiology, with higher emergence rates in traumatic and stroke-related DoC (52%, 13 of 25, and 44.8%, 13 of 29, respectively) than hypoxia/anoxic injury (14.3%, 2 of 14), although etiology alone should not guide prognostication.30 Future studies should evaluate patterns of CRS-R utilization, rehabilitation frequency, and patient outcomes across trauma centers to support longitudinal assessment and inform best practices in the care of individuals with DoC. Age-related differences were also observed. Consistent with previous studies, patients in the young adult age group (18-40y) were more likely to be discharged at higher levels of consciousness,31 despite being more frequently presenting with lower levels of consciousness at admission. This finding suggests that younger adults may have greater recovery potential.32 Respiratory progression was common. Nearly three-quarters of the cohort underwent tracheostomy, and the majority tolerated SV and capping trials with minimal complications, regardless of consciousness level. Although predictive analyses were limited by sample size, these findings suggest that respiratory progression—including decannulation—can be safely initiated during acute hospitalization. Individual clinical factors, such as secretion burden and cough reflex, appeared to play a more influential role than level of consciousness alone.20,33 Swallowing function similarly improved. All patients were NPO upon admission and had a baseline FOIS score of 1, consistent with prior findings in prefeeding dysphagia among patients with DoC.21 By discharge, more than half of the cohort showed improvement in swallowing status, and 23 patients (85.2%) who achieved a full PO diet had also emerged from MCS. Importantly, patients across varying levels of consciousness—including those in lower states—participated in structured dysphagia management, including clinical swallowing evaluations, bolus trials, and sensory stimulation.27,34,35 Notably, 2 UWS patients underwent PO trials in the absence of reflexive lip pursing and jaw clamping to oral stimulation, a commonly observed oromotor response in this population. Thirteen additional patients were discharged with therapeutic PO intake recommendations. Collectively, these findings support the integration of regular swallowing evaluations in the care of hospitalized patients with DoC, even at lower levels of consciousness. Timing of swallowing trials was guided by multiple factors, including medical stability, emergence of reflexive oral responses, and secretion control. No patients in unresponsive wakefulness underwent MBS due to insufficient prerequisite feeding skills and oral-stage function.21 These observations further support the hypothesis that oral-phase swallowing may be, at least in part, cognitively mediated.36 Within our cohort, 18 patients underwent MBS before DOC emergence, and 11 were evaluated while tracheostomized. Together, these findings reinforce the importance of considering instrumental swallowing assessments in acute care patients with DoC, including those with open tracheostomies or undergoing SV or capping trials.36 Since the completion of this study, Herr et al37 have introduced the Swallowing Assessment in Disorders of Consciousness, a structured tool designed to quantify swallowing behaviors in patients with DoC. Unlike traditional bedside swallow evaluations, Swallowing Assessment in DoC provides a structured, quantitative instrument tailored to individuals who are often unable to communicate or actively participate in assessments. The present findings complement this emerging work and may inform clinical decision-making for speech-language pathologists evaluating swallowing function in hospitalized patients with DoC. As this study reports on cohort-level outcomes, specific intervention strategies and treatment frequency were beyond the scope of analysis and remain important targets for future investigation. Discharge destinations varied widely, reflecting both clinical and socioeconomic factors—including insurance coverage, financial resources, and geographic access—that may influence postacute care. These disparities highlight the importance of prioritizing structured rehabilitation assessments during the acute care period and highlight the need to improve awareness, utilization, and longitudinal application of tools such as the CRS-R. Study limitations As a single-center study, findings may not generalize to the broader population of hospitalized patients with DoC, and case identification may have missed eligible patients. The small sample size further limits the detection of subtle patterns. This ongoing study will benefit from cohort expansion, and multicenter designs are needed to improve generalizability. We did not assess the efficacy or safety of specific swallowing or tracheostomy interventions, nor did we analyze individual treatment plans or dysphagia evaluations; findings are therefore presented at the cohort level. Prolonged acute care length of stay—partly due to barriers to postacute placement—may have influenced observed recovery and decannulation rates. Although the CRS-R is considered the gold standard for DoC classification,23 misclassifications may occur due to testing variability and confounders, such as severe motor impairment, spasticity,24 pain, agitation,38 akinetic mutism,39 iatrogenic effects, or caregiver exclusion.40 This study focuses on the acute phase; longer-term outcomes were not examined. Complications, such as hydrocephalus, subdural hematomas, infections, dysautonomia or sympathetic storming,41 lesion location,42 and use of neurostimulants, such as amantadine,43,44 were not tracked. Additionally, objective measures of covert consciousness,45 such as electrophysiology or functional magnetic resonance imaging (functional magnetic resonance imaging), were not used. Future studies should incorporate multicenter designs, larger samples, longitudinal follow-up, and standardized tracking of clinical variables to better define recovery trajectories and optimize care. Conclusion In this cohort of hospitalized, medically complex patients with DoC, measurable progress during the acute care phase underscores that recovery begins early and can be reliably detected through structured assessment. Patients demonstrated meaningful advancement across domains of consciousness, respiratory function, and swallowing, challenging the traditional notion that rehabilitation gains are limited to later stages of care. These findings support a paradigm shift in which rehabilitation is initiated based on physiologic readiness, rather than delayed until overt emergence. Early, structured assessment and intervention in acute care, therefore, represent a critical opportunity to shape recovery trajectories, inform prognostication, and optimize care planning. Acute hospitalization should be recognized not as a passive interval but as a decisive phase in recovery, where timely rehabilitation can meaningfully influence outcomes for patients with DoC. Supplier a. SPSS version 30.0.0.0; IBM. Disclosures The investigators have no financial or nonfinancial disclosures to make in relation to this project. Acknowledgments Data statements The data supporting study findings are not publicly available due to restrictions related to patient confidentiality. Deidentified data may be available upon request with permission of the MaineHealth IRB. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used NotebookLM to create a graphical abstract, using only our study findings as a data source to generate the image. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. References