Table 1.

Challenges and potential solutions for DoC research—design considerations

Considerations	Challenges	Potential solutions
Source population	Reliance on behavioural classification for patients with coma or DoC Composite reference population (multi-step management and care) Multiple aetiologies Multiple entry points after injury Acute stage: missing chronic follow-up Chronic stage: missing data on acute covariates	Disease continuum approach Consider multiple recruitment scenarios with flexible study entry points (reduce selection bias) Ensure homogeneous access to long-term follow-up (reduce selection bias) Retrieve acute stage covariates in studies on chronic DoC Endotype approach to DoC (e.g. ACCESS framework) to allow longitudinal investigations based on mechanistically defined subgroups Multi-modal evaluation (clinical examination, neurophysiology, neuroimaging) at subsequent steps Adopt a two-stage strategy for enrolment including caregiver evaluations Large-scale population studies including initially healthy persons
Case definition	Lack of a widely agreed, generalizable, operative definition for coma Unknown timing of the transition from coma to chronic DoC	Use of a single scale for diagnosis, capturing the defining features of DoC (e.g. CRS-R, SECONDs)
Case verification	Neurological deficits confounding clinical diagnosis (e.g. aphasia, critical illness neuropathy/myopathy) Extra-neurological impairments confounding clinical diagnosis (e.g. sepsis, medications) Cognitive motor dissociation Lack of verification of initial diagnosis	Standardized, systematic evaluation with clearly described examination protocols for clinical, neuroimaging and neurophysiology assessment Verification at successive timepoints through multi-modal tools
Selection of outcomes (dependent variables)	GOS-E, CPC and mRS provide inadequate representations of residual cognitive impairments, do not capture subtle differences in treatment effectiveness and have floor and ceiling effects Lack of precise definitions for VS/UWS or MCS in the above-mentioned scales Medical complications caused by DoC impact scores on general functioning scales	Use of dedicated scales for DoC assessment (e.g. CRS-R, SECONDs) to establish case definitions and outcome measures concurrently Define transitions between different levels of consciousness, particularly timing, trajectories and associations with medical and biological factors Development of telephone-based versions of dedicated scales to improve follow-up retention Concurrent use of traditional disability scales Integration of PROMs for patients with residual communication abilities Integration of tools for caregivers’ burden of care
Selection of covariates	Choice of relevant covariates for DoCs is not straightforward (distinction from confounders) Endpoints in previous research (e.g. in traumatic brain injury) not explicitly tailored to DoCs Small monocentric datasets for subacute and chronic DoC, inconsistently accounting for acute-phase covariates	Build conceptual frameworks to classify covariates based on different interactions with DoC constructs (e.g. pre-event, baseline and secondary modifiers) Avoid mixing processes taking place at different times Define endotype-based covariates, related to mechanisms Develop CDE for DoC to foster a standardized collection of covariates

CRS-R, Coma Recovery Scale-Revised.