Impairments in arousal and awareness after severe brain injury are ubiquitous to disorders of consciousness (DoC) which include coma, unresponsive wakefulness syndrome (UWS)/vegetative state (VS), and the minimally conscious state (MCS). MCS is a clinically heterogeneous category; as it is further stratified into MCS plus (+) and MCS minus (−). MCS+ is applied when observed behavioral responses demonstrate some level of preserved language functioning, as evidenced by the ability to follow commands, discriminate objects, or attempts to communicate (Bruno et al., 2011b; Giacino et al., 2022). Subtle behavioral differences that distinguish between these conditions are not easily detected on informal non-standardized bedside evaluation. For example, the re-emergence of spontaneous eye opening without evidence of purposive behavior is considered the hallmark of transition from coma to UWS/VS, yet even patients in MCS+ may demonstrate poor sustained arousal and highly inconsistent purposive behaviors. Conversely, reflexive vocalizations, eye, and limb movements are all commonly seen in UWS/VS, and may be misinterpreted as purposive responses to stimuli. Expected variability and ambiguity of behavioral responses further complicates the clinical phenotype and limits diagnostic certainty at the individual level. Consequently, informal bedside assessment and team consensus carries a 40% misdiagnosis rate (Schnakers et al., 2009; Wang et al., 2020). The potential consequences of misdiagnosis are great, as one’s level of consciousness plays a central role in driving important care decisions such as withdrawal of life sustaining treatments and access to specialty post-acute and rehabilitation services.

This paper will present a structured approach to evidence-based assessment of DoC to apply clinically to improve diagnostic accuracy across the continuum of care. An overview of practice guidelines and program recommendations will be provided, which include the use of standardized neurobehavioral rating scales to reduce diagnostic error. Common confounds germane to DoC will be discussed along with strategies to help address and mitigate their impact on behavior responsiveness and optimize diagnostic certainty. The clinical strategies to neurobehavioral assessment highlighted in this paper were included based on published evidence including the American Academy of Neurology (AAN) DoC Practice Guideline Recommendations (Giacino et al., 2018) and European Academy of Neurology Guidelines for Diagnosis of Coma and DoC (Kondziella et al., 2020) in conjunction with various evidence-informed recommendations such as the American Congress of Rehabilitation Medicine’s DoC Minimal Competency Recommendations (Giacino et al., 2020a). Specifically, published Guidelines and Recommendations underwent intensive expert investigation, systematic review, data analysis, application of the Grading Recommendation Assessment, Development and Evaluation (GRADE) process. Additionally, the AAN DoC Guideline development applied the AAN Clinical Practice Guideline Process Manual to direct the methods of creating the 2018 Practice Recommendations.

Published American and European DoC practice guidelines and American minimal competency recommendations for rehabilitation programs support the use of valid and reliable standardized neurobehavioral rating scales as the “gold standard” for assessment of persons with DoC (Giacino et al., 2018, 2020a; Kondziella et al., 2020). Their superior diagnostic accuracy as compared to team-based consensus has been supported through the past published evidence (Schnakers et al., 2009; Wang et al., 2020). Patient performance on these scales can assist in identifying level of consciousness within the DoC spectrum, facilitate detection of diagnostic confounds and guide development of strategies aimed at accessing latent cognition to maximize rehabilitation potential and functional outcomes. Moreover, serial assessments can be used to identify trends in the rate and trajectory of recovery that can help inform prognosis and degree of long term disability (Giacino et al., 2018, 2020a). Tables 1, 2 present a complete list of practice guidelines and program recommendations related to diagnostic assessment.

There are several evidence-supported standardized behavior scales that can be employed in clinical practice, at all levels of care, to aid diagnosis, prognosis and family counseling for DoC. Irrespective of specific scale used, assessment of persons with DoC typically evaluates behavioral responsiveness in the common domains of sensory process and function including: auditory, visual, motor, oral motor, communication and arousal (Kalmar and Giacino, 2005; Pape et al., 2009, 2014; Morrissey et al., 2018). Often responses are graded based on a hierarchy of behaviors that demonstrate neurological functioning at either a brainstem, subcortical or cortical level (Giacino et al., 2022). Seel et al. (2010) conducted a review of available behavioral DoC assessment scales and provided recommendations for use based on the psychometric qualities (validity and reliability) of each scale and other criteria. The Coma Recovery Scale-Revised (CRS-R) was the only tool recommended for clinical use with minor reservations secondary to its strong reliability, validity, standardized administration and scoring procedures, interpretative scoring guidelines, and ease of accessibility for clinicians. Five additional scales were recommended for practice with moderate reservations including the SMART, WNSSP, SSAM, WHIM, and DOCS. One scale, the CNC, was recommended, however, with major reservations. Four scales were specifically not recommended for bedside assessment of DOC due to poor validity, reliability or a lack of standardization. These included the RLS85, LOEW, and CLOCS (Seel et al., 2010). See Table 3.

Since the review Seel et al. (2010), the CRS-R has undergone further extensive investigation. Bodien et al. (2016) performed sensitivity and specificity analyses using CRS-R derived diagnoses to determine that a total cut-off score of eight or higher reliably distinguishes between patients in UWS/VS and MCS in 93% of cases (Bodien et al., 2016). Collective evidence evaluating the utility of the CRS-R, compared to other behavior rating scales, diagnostic modalities, and neurophysiological studies, demonstrates the superiority of the CRS-R as a sensitive and reliable tool to accurately identify and discriminate among the levels of DoC (Lechinger et al., 2013; Annen et al., 2019; Formisano et al., 2019a; da Conceição Teixeira et al., 2021). Additional evidence focusing on the utility of the CRS-R identifies the benefit of serial use of the CRS-R to improve accuracy of identifying behavioral presentation of DoC (Wannez et al., 2017; Yang et al., 2021). Further evaluation and investigation of the CRS-R has produced development of a CRS-R index to improve total score interpretation and translation of the CRS-R into multiple languages for international use (Lombardi et al., 2007; Tamashiro et al., 2014; Binder et al., 2018; Annen et al., 2019; Zhang et al., 2019). Moreover, research also supports the use of the CRS-R to help inform the trajectory of DoC recovery and prognosis at the individual level (Bodien et al., 2016, 2022; Giacino et al., 2018, 2020a).

Standardized behavior rating scales such as the CRS-R are rarely utilized in the intensive care unit (ICU) for diagnostic assessment of Doc after severe brain injury (Chaturvedi et al., 2021). Time demands imposed by these tools, along with use of sedation, paralytics, mechanical ventilation and movement restricting equipment all serve as practical barriers to the implementation of standardized assessment of consciousness in DoC patients (Chaturvedi et al., 2021). Consequently, physicians in the neurological ICU routinely perform non-standardized bedside evaluations to determine level of consciousness.

The Glasgow Coma Scale (GCS) is the most widely known and utilized tool for assessing brain injury severity and level of coma in ICU/acute care settings due to its feasibility and time efficient implementation required at this level of care (Formisano et al., 2019b; Helbok et al., 2022). However, the GCS is an observational scale and lacks sensitivity to distinguish among different levels of consciousness, and to identify salient features of MCS (−/+) in particular (Bodien et al., 2021). Bodien et al. (2021) compared GCS score combinations to CRS-R scores and found great variability and diagnostic error rates when the GCS is used to identify consciousness. Specifically, they found that GCS total scores did not differentiate among DoC subtypes and that when GCS scoring criteria are used, many persons in MCS were erroneously classified as being “comatose.” The Full Outline of UnResponsiveness (FOUR) is an additional neurological assessment implemented in the ICU that is recommended by the European (EU) DoC guidelines for assessment of level of consciousness in the ICU (Kondziella et al., 2020). The EU recommends the use of the FOUR over the GCS in light of its convenience of serial use by clinicians and nurses. Additionally, the FOUR is more sensitive in capturing certain MCS and locked-in syndrome behaviors involving eye movement which decreases the risk of misdiagnosis (Bruno et al., 2011a; Kondziella et al., 2020; Bodien et al., 2021). Although the FOUR is a recommended assessment for this patient population, there are currently efforts underway to develop and validate an abbreviated version of the CRS-R and other standardized rating scales adapted for DoC patients in the ICU (Aubinet et al., 2021; Bodien et al., 2021; Sanz et al., 2021).

Notably, even standardized behavioral rating scales are limited in their ability to differentiate a subset of ICU patients at risk for being misidentified as having a DoC due to the presence of related clinical features such as complete motor paralysis or language impairment (Kondziella et al., 2020). Recent research has found approximately 15–20% of persons classified as having a DoC in the ICU actually have cognitive motor dissociation (CMD), a condition of covert consciousness characterized by the retained capacity for volitional thought in the absence of overt behavioral manifestations or motoric output (Schiff et al., 2005; Owen et al., 2007; Owen and Coleman, 2008). CMD can only be detected with the use of advanced technologies such as functional MRI and electroencephalograph (EEG). These modalities have demonstrated the ability to identify cases of higher-order cortex motor dissociation by eliciting accurate responses to language and music based tasks in persons behaviorally presenting as UWS/VS (Edlow et al., 2017; Claassen et al., 2019; Kondziella et al., 2020; Thibaut et al., 2020). Active and passive paradigms in using fMRI and EEG have demonstrated utility in identifying CMD in behaviorally unresponsive patients. However, it has been found that passive paradigms have a greater likelihood of capturing preserved consciousness (Kondziella et al., 2016; Aubinet et al., 2022). The scientific understanding of CMD is evolving, but current evidence suggests it is likely a distinct phenomenon separate from the DoC spectrum (Kondziella and Stevens, 2022), more akin to a functionally locked-in syndrome. Evidence suggests those who are identified as CMD while in the ICU have an improved functional recovery as compared to those unresponsive patients who demonstrate no evidence of consciousness with advanced neuroimaging (Edlow et al., 2017, 2021). This is a critical issue, given detection of consciousness, or lack thereof, can have significant impact on surrogate decisions regarding withdrawal of care while in ICU (Giacino et al., 2018; Graham et al., 2018; Naci and Owen, 2022; Pruvost-Robieux et al., 2022).

Beyond CMD, persons with DoC present with a wide range of overlying complications and comorbidities that can exacerbate the complexity of the clinical picture (Majerus et al., 2009; Schnakers et al., 2015; Bodien et al., 2022). US practice guidelines recommend that prior to making a final determination regarding level of consciousness, efforts be made to identify and treat confounding conditions that impede accurate diagnosis and directly impact the ability to actively participate and interact with others (Giacino et al., 2018). Similarly, minimal competency recommendations (Giacino et al., 2020b) state that rehabilitation programs should have a protocol in place to detect and treat confounds that can mask evidence of conscious awareness and lead to misdiagnosis. For purposes of the present paper, the authors conceptualize these confounds within three primary categories: medical/neurological issues, overlying functional (motor/sensory/cognitive) impairments, and adverse environmental influences on behavior responsiveness (see Table 3). Some confounds may be present in the acute/ICU setting, whereas others may not develop or become apparent until the post-acute setting.

Serial monitoring over time, using recommended neurobehavioral assessment tools, can facilitate early detection of behaviors that may indicate emerging awareness and thus guide development of individualized rehabilitation strategies. Collated results from repeat assessments can help identify trends in recovery that can inform long-term prognosis for persons with DoC. A compendium of evidence supports the prognostic utility of CRS-R scores and the trajectory of those scores over time to predict recovery of consciousness and functional outcome (Pignat et al., 2016; Portaccio et al., 2018; Annen et al., 2019; Lucca et al., 2019; Hamilton et al., 2020; Boltzmann et al., 2021). Ultimately, when results are to be used to help inform prognosis, serial CRS-R scores must be considered along with other significant factors such as patient age, premorbid conditions, injury comorbidities and severity, frequency of complications and effective acute management (Estraneo et al., 2018; Steppacher et al., 2020; Kowalski et al., 2021; Nekrasova et al., 2021; Siegert et al., 2022). Figure 2 presents a recommended structured approach to applying serial assessment to outcome monitoring.

Ongoing tracking of scores over time provides objective data that can be used to help guide family education and counseling efforts regarding clinical care decisions and long term care planning. When communicating diagnosis and prognosis with family caregivers, rely on use of simple language that is easy to understand, and provide periodic updates (Giacino et al., 2020a). Counseling should include education about their loved one’s behavioral assessment results, information about the assessment tools used and how obtained results relate to expectations for recovery. Presenting a graph of scores on the CRS-R and other measures throughout the course of care is a useful tool to visually demonstrate a person’s recovery trajectory and areas of progress (or lack thereof). This approach to counseling is aimed at helping family caregivers understand their loved one’s condition and care needs so they can develop realistic expectations for recovery and collaboratively establish an appropriate short- and long-term plan of care (Giacino et al., 2020a).

Assessment of Doc is rapidly evolving. As previously mentioned, there are efforts underway to develop and validate consciousness screens and short-form versions of existing scales to facilitate expedient, accurate assessment in critical care settings. Additionally, a valid and reliable DoC assessment in young children is needed. Slomine et al. (2019) have developed the Coma Recovery Scale for Pediatrics (CRS-P) to evaluate DoC in children 12 months and older. The CRS-P is undergoing continued investigation related to strength of psychometric properties and utility of use in the pediatric DoC population (Slomine et al., 2019). Ongoing exploration into ways to expand the use of neuroimaging and electrophysiological technologies to aid detection of consciousness and to identify CMD early post injury is a high priority to better inform medical decision-making (e.g., withdrawal of care) and overall care planning. Brain Computer Interface (BCI) is an additional modality that has been studies extensively as an assessment tool to identify consciousness or CMD through “cerebral communication” (Farisco et al., 2014; Ortner et al., 2017; Wang et al., 2017). BCI is evolving through research and ideally will become a clinical tool feasible for utilization at the bedside. Evaluating the comparative sensitivity, specificity, cost, and overall ease of implementation among these technologies will help direct future efforts to make these tools more accessible. There is a significant need to develop a prognostic algorithm where neurobehavioral assessment results, in combination with evidence-based biomarkers (e.g., neuroimaging, electrophysiological studies, etc.) can be applied to promote diagnostic accuracy and enhance the precision of prognostic estimates to support informed care decisions (Hammond et al., 2021; Mainali et al., 2022; Olson et al., 2022). Finally, operationalizing an interdisciplinary education, training, and mentorship methods can help ensure reliability and validity of assessment results and enhance clinical application of results to guide quality DoC care.

Standardized neurobehavioral assessment is a primary feature of quality DoC care essential to ensuring diagnostic accuracy, appropriate rehabilitation planning, and outcome monitoring. Given the high prevalence of medical, neurological, functional and environmental confounds in persons with DoC, it is imperative to have tools that facilitate accurate bedside assessment of consciousness. Evidence-based neurobehavioral rating scales are widely available and accessible tools for bedside use across the continuum of care. Serial and multimodal assessment can improve diagnostic certainty, identify trends in recovery over time, and guide prognostic counseling with families. As technology continues to advance through future funding and research, the application of multimodal assessment tools will likely continue to evolve and play an increasingly important role in supporting DoC assessment and overall care planning for this population.

BM and AS equally contributed to the development and completion of manuscript content, including review of the literature, manuscript prose, development of tables and figures, and the editing process. They completed the final review of the manuscript together over various virtual meetings and achieved consensus on final manuscript and submission. Both authors contributed to the article and approved the submitted version.