Designing clinical trials for functional neurological disorder requires early recognition that this population differs in important ways from many other neurological cohorts. Symptom variability over time, heterogeneous presentations, and frequent comorbid psychiatric or medical conditions all influence trial design, implementation, and interpretation. A core consideration is aligning the scientific question with a realistic and clinically relevant primary objective, such as reduction in symptom severity, improvement in functional independence, or decreased healthcare utilization, rather than focusing solely on complete symptom resolution. Selecting endpoints that reflect change in real-world functioning is particularly important given the fluctuating nature of the disorder.
Rigorous diagnosis is central to every other design decision. Because misdiagnosis can dilute treatment effects and threaten validity, trials should require confirmation by clinicians with specific expertise in functional neurological disorder, using up-to-date diagnostic criteria and positive clinical signs rather than diagnosis by exclusion alone. Standardized diagnostic pathways and structured documentation help ensure that enrolled participants truly have the condition under investigation, which in turn improves internal validity and the credibility of reported outcomes.
Symptom heterogeneity demands careful definition of the target population. Functional seizures, functional movement disorders, functional weakness, gait impairment, and sensory symptoms each may respond differently to specific interventions. Trials must decide whether to focus on one symptom subtype or include a broader spectrum while pre-specifying stratification or subgroup analyses. Narrower phenotypes simplify measurement and interpretation but may limit generalizability, whereas broader inclusion can increase recruitment but risks masking benefits in specific symptom clusters.
Randomization strategies should anticipate this heterogeneity. Stratified randomization based on key prognostic factors—such as symptom subtype, chronicity, baseline severity, and presence of relevant comorbidities—helps balance groups and reduces confounding. In smaller studies or early-phase work, minimization procedures may be useful to ensure comparable distribution of critical variables across arms. Careful attention to allocation concealment and sequence generation is essential, because knowledge of upcoming assignments can unintentionally influence clinician referrals, patient expectations, or enrollment decisions.
Blinding presents unique challenges in functional neurological disorder trials, especially when evaluating complex psychological or physiotherapy-based interventions. Full double-blind conditions may be impractical, but measures can still be taken to reduce bias. Blinded outcome assessors, standardized scripts for interacting with participants, separation of treating clinicians from evaluators, and pre-specified objective endpoints all help reduce expectancy effects. Where blinding is not possible, rigorous protocolization and detailed reporting of treatment delivery and patient engagement become even more important.
Feasibility and recruitment are recurrent obstacles. Stigma, mistrust of the diagnosis, and previous negative experiences with healthcare systems may lead patients to decline participation. To address this, trial designs need clear, patient-centered communication about the rationale for the study, the nature of the disorder, and realistic expectations about potential benefits. Collaborations with multidisciplinary clinics, involvement of patient advocacy groups, and integration of research pathways into routine clinical care can improve identification of eligible patients and support sustained engagement throughout the trial.
Retention and adherence must be considered from the outset. Many individuals with functional neurological disorder experience fatigue, anxiety, depression, or social and occupational challenges that interfere with regular attendance at sessions and follow-up visits. Flexible scheduling, hybrid in-person and remote assessments, transportation support where feasible, and ongoing patient engagement strategies all help mitigate dropout. Clearly defined minimum adherence thresholds and plans for handling missing data should be specified a priori to avoid biased estimates of treatment effect.
Given the complex psychosocial context of the disorder, potential confounders and co-interventions require close attention. Many participants will be receiving concurrent therapies, such as medication adjustments, psychotherapy, or physical rehabilitation. Trial protocols should delineate which background treatments are allowed, which must be stable prior to enrollment, and how changes during the study will be documented and analyzed. Statistical plans should include sensitivity analyses to assess the impact of these co-interventions on primary and secondary outcomes.
Outcome timing is another key design consideration. Symptom patterns in functional neurological disorder may change rapidly in response to contextual factors, but more durable improvements in function and quality of life often take longer to emerge. Multiple assessment timepoints, including short-term, intermediate, and longer-term follow-up, allow investigators to differentiate immediate response, consolidation of gains, and relapse or deterioration. Trial length must be long enough to capture these trajectories without making participation unduly burdensome.
Sample size calculations should be grounded in realistic assumptions about effect sizes and attrition rates. Because high dropout and variable response are common, conservative assumptions are usually warranted. When existing data are limited, early-phase pilot or feasibility studies can provide estimates of recruitment rates, adherence, and variance in outcomes to inform larger, adequately powered trials. Adaptive designs, including planned interim analyses and possible sample size re-estimation, may be particularly useful in this evolving field, provided they are implemented with strict control of type I error.
Consistency in intervention delivery is crucial, especially for complex non-pharmacological treatments. Trials should specify detailed manuals, training requirements for therapists or clinicians, supervision structures, and fidelity monitoring procedures. Periodic review of session recordings or treatment notes can help ensure adherence to the protocol and identify drift over time. Clear documentation of what constitutes the active components of an intervention allows future replication and facilitates interpretation of which elements are responsible for observed benefits.
Patient and public involvement can significantly strengthen functional neurological disorder trial design. Input from individuals with lived experience can guide selection of outcomes that truly matter to patients, such as return to work or education, improvement in daily independence, or reduction in emergency visits. They can also help refine recruitment materials, consent forms, and communication strategies to increase trust and participation. Incorporating this stakeholder perspective early supports the development of trials that are both scientifically rigorous and more likely to translate into meaningful improvements in clinical practice.
Patient selection and diagnostic standardization
Accurate and consistent patient selection is the foundation of credible clinical trials in functional neurological disorder, because even a modest proportion of misclassified participants can substantially dilute treatment effects. Enrollment should begin with a clearly articulated target population, including age range, symptom subtype, duration of symptoms, and level of functional impairment. Investigators must decide whether to prioritize internal validity with narrowly defined inclusion criteria or increase external validity by allowing a broader, more representative group. In either case, eligibility rules should be explicitly linked to the underlying scientific question and to the mechanisms through which the intervention is expected to exert its effect.
Modern diagnostic approaches to functional neurological disorder emphasize positive clinical signs rather than diagnosis by exclusion. Trial protocols should therefore mandate that participants meet established diagnostic criteria, such as DSM-5 or contemporary consensus guidelines, documented through structured clinical evaluation. Positive signs might include Hoover’s sign for functional weakness, entrainment or distractibility in functional tremor, or internal inconsistency in gait patterns. Requiring such features, and documenting them in a standardized fashion, helps ensure that the trial is studying the intended condition rather than a heterogeneous mix of neurological and non-neurological problems.
Diagnostic standardization benefits from a tiered assessment pathway. A first tier might involve screening by general neurologists or emergency clinicians using brief structured tools to identify probable functional symptoms. A second tier would then involve confirmation by clinicians with specific expertise in functional neurological disorder, ideally within a multidisciplinary setting that includes neurology, psychiatry, and rehabilitation professionals. This tiered approach balances feasibility and recruitment demands with the need for diagnostic rigor, ensuring that only those who truly meet criteria progress to randomization.
Inter-rater reliability is a major concern when diagnosis hinges on clinical judgment. Trials should implement training programs and calibration exercises for all assessors, supported by written manuals that describe diagnostic criteria, required documentation, and example cases. Use of standardized case report forms that prompt clinicians to record specific positive signs and rule out key alternative diagnoses can reduce variation across sites. Periodic reliability checks, such as dual assessments or review of anonymized clinical vignettes, help maintain diagnostic consistency over the course of the study.
Imaging and ancillary investigations should support, but not replace, clinical diagnosis. Brain MRI, EEG, and other tests are typically used to identify structural or epileptic conditions that would constitute exclusion criteria, rather than to diagnose functional neurological disorder per se. Protocols need clear guidance on what investigations are mandatory prior to enrollment, which are recommended but optional, and how to handle incidental findings. Overly extensive testing can slow recruitment and increase participant burden, whereas insufficient testing raises the risk of enrolling individuals with untreated organic disease.
Because functional neurological disorder commonly coexists with other neurological or psychiatric conditions, inclusion and exclusion criteria must carefully specify which comorbidities are permitted. Trials may choose to exclude major neurological conditions that could independently drive symptoms, such as multiple sclerosis or advanced Parkinson’s disease, while allowing stable migraine or peripheral neuropathy. Similarly, psychiatric comorbidities such as depression, anxiety disorders, or post-traumatic stress disorder are highly prevalent and may be central to the clinical picture; automatic exclusion of these patients would yield an unrepresentative sample. Instead, protocols should define thresholds for severity, stability requirements, and whether concurrent psychiatric treatment must be stable for a defined period before enrollment.
Symptom subtype is another key dimension of patient selection. Some studies focus specifically on functional seizures, others on functional movement disorders, functional weakness, or mixed presentations. Grouping different subtypes together can improve recruitment and reflect real-world practice, but it requires careful stratification and pre-planned subgroup analyses to interpret outcomes. For instance, if an intervention is hypothesized to influence motor planning and belief about movement, it may be more applicable to motor symptoms than to functional seizures. Clear documentation of the predominant symptom at baseline, with standardized classification rules, allows meaningful exploration of differential treatment response across subgroups.
Symptom duration and chronicity often influence prognosis and may moderate treatment response. Trials should specify minimum and maximum duration thresholds—for example, excluding ultra-acute presentations where spontaneous remission is likely, or very longstanding cases where reversibility may be limited. Chronicity can be categorized into predefined bands (e.g., 24 months) and used both as an inclusion parameter and a stratification factor. This enables post hoc examination of whether earlier intervention is associated with better outcomes, while preventing imbalances between treatment arms that might bias results.
Standardization of diagnostic communication is as important as standardization of diagnostic criteria. The way clinicians explain the diagnosis to patients affects acceptance, engagement, and ultimately retention in the study. Protocols should include recommended language for conveying the diagnosis in a clear, non-stigmatizing, and non-dualistic way, emphasizing that the symptoms are real, common, and potentially reversible, and that they arise from functional disturbances in brain networks rather than structural damage. Using consistent explanatory models across sites reduces variability in patient expectations, which can otherwise act as a powerful, unmeasured confounder.
Assessment of psychological and contextual factors at baseline provides essential information for both patient selection and interpretation of outcomes. Validated instruments for mood, anxiety, trauma history, dissociation, somatic symptom burden, and illness beliefs should be incorporated into the screening process. Rather than using these measures solely for exclusion, many trials will benefit from treating them as characterization variables and potential moderators of treatment response. For example, high levels of catastrophizing or low treatment expectancy at baseline might predict poorer outcome, informing later personalized treatment strategies and secondary analyses.
Functional neurological disorder frequently involves complex social and occupational circumstances, including disability claims, litigation, or workplace conflicts. Trials must decide how to handle these contextual factors within eligibility criteria. Excluding all individuals involved in compensation processes may yield a less representative sample, but ignoring these issues risks confounding if external incentives influence symptom reporting or participation. A balanced approach involves documenting compensation status, legal proceedings, and employment situation at baseline and including them in pre-specified covariate or subgroup analyses, while only excluding cases where such factors are likely to directly obstruct engagement with the intervention.
Multi-center trials face particular challenges in maintaining uniform diagnostic standards across diverse healthcare systems and cultures. Central adjudication committees can play a critical role by reviewing key diagnostic data—such as structured clinical notes, video recordings of examinations or events suggestive of seizures, and relevant test results—to confirm eligibility. When feasible, remote video review of motor symptoms or events by an expert panel helps harmonize case definitions. This centralized oversight should be clearly described in the protocol, with timelines that are compatible with recruitment needs.
To avoid systemic biases in who is enrolled, recruitment strategies should be designed to reach a broad and inclusive patient population. Functional neurological disorder affects people across age, gender, and cultural groups, yet historically some populations have been underrepresented in clinical trials. Engagement with community clinics, emergency departments, and primary care practices, along with translation of information materials into relevant languages, can broaden access. Screening logs that track who is approached, who declines, and why provide essential data for understanding and addressing barriers to participation.
Diagnostic standardization also extends to the classification of episodes and events over time. In studies of functional seizures, for example, clear criteria are needed to differentiate functional events from epileptic seizures, syncope, or other paroxysmal conditions. Baseline video-EEG documentation of typical events, when available, strengthens diagnostic confidence, and structured seizure diaries can be used to track frequency and semiology. For movement and motor symptoms, video-based rating at baseline and follow-up allows blinded comparison and verification that the same functional phenomenon is being followed throughout the trial.
Baseline severity measures must be consistently applied, because they often form the basis for primary outcomes and for defining clinically meaningful change. Scales such as standardized seizure frequency counts, functional movement disorder rating scales, strength and gait assessments, or global functional status measures should be administered using uniform protocols. Investigators should predefine thresholds for minimal and maximal baseline severity eligible for enrollment, to avoid floor or ceiling effects that could obscure treatment benefits. Documentation of prior treatments and previous exposure to similar interventions is likewise essential, both to reduce contamination and to interpret changes over time.
In pediatric and adolescent populations with functional neurological symptoms, patient selection and diagnostic standardization carry additional complexities. Parental involvement, school context, and developmental considerations must be systematically assessed. Trials should specify age-appropriate diagnostic criteria, including positive signs adapted for younger individuals, and obtain input from pediatric neurologists and child psychiatrists during screening. Because family dynamics and parental beliefs about the disorder strongly influence engagement, standardized approaches to family education and consent are crucial elements of the selection process.
Across all age groups, informed consent procedures should reflect the specific challenges of functional neurological disorder, where patients may have experienced prior invalidating or contradictory explanations. Consent materials need to clearly distinguish participation in research from routine clinical care while acknowledging uncertainties and emphasizing that non-participation will not jeopardize access to treatment. Providing opportunities for questions, allowing time to reflect, and, where appropriate, involving trusted relatives or caregivers can foster informed, voluntary participation and reduce later withdrawal driven by misunderstanding.
Trial protocols must explicitly define how diagnostic uncertainty will be handled. In some cases, initial evaluation may suggest functional neurological disorder but leave residual doubts about alternative diagnoses. Options include exclusion of borderline cases, enrollment contingent on additional investigations, or inclusion with clearly documented diagnostic probability and pre-specified sensitivity analyses excluding uncertain cases. Transparent handling of uncertainty, with clear reporting in trial publications, strengthens confidence in the findings and facilitates comparison across studies that may adopt slightly different thresholds for diagnostic certainty.
Outcome measures and assessment tools
Selecting and operationalizing appropriate outcomes in functional neurological disorder research requires balancing scientific rigor with clinical relevance. Because these conditions affect multiple domains—symptom expression, physical function, emotional well-being, and social participation—no single metric captures the full impact of an intervention. Trial design should therefore begin with a clear hierarchy of primary and secondary outcomes, grounded in a conceptual model of how the intervention is expected to work. The primary outcome should be tightly defined, measurable with acceptable reliability, and directly linked to patient priorities, while secondary measures provide a more comprehensive picture of change and help explain why a treatment did or did not prove effective.
Commonly used primary outcomes fall into three broad categories: symptom-specific measures, functional status indicators, and global improvement ratings. For symptom-specific measures, trials of functional seizures often rely on seizure frequency, time to event remission, or proportion of seizure-free days, typically assessed via prospectively completed diaries. In functional movement disorders, scales that capture tremor frequency, dystonia severity, or gait disturbance—alongside video-based ratings—allow quantification of motor symptoms. Where weakness or gait impairment predominate, standardized strength testing and timed walking or balance tasks provide sensitive indicators of change. Whatever the symptom domain, outcomes need precise operational definitions, such as specifying what constitutes a functional seizure event or what level of assistance counts as independent walking.
Functional status outcomes capture the impact of symptoms on day-to-day life and are often more meaningful to patients and payers than symptom counts alone. Validated instruments such as global disability indices, measures of activities of daily living, and participation scales assessing return to work, school, or caregiving responsibilities are well suited as primary or key secondary outcomes in clinical trials. Because functional neurological disorder can lead to marked disruption in social roles, measuring changes in employment status, absenteeism, or educational participation provides an important complement to purely symptom-focused endpoints. Trials should predefine what constitutes a clinically important improvement in function—for example, progression from being unable to work to part-time employment—so that statistical results can be interpreted in terms of real-world benefit.
Global impression scales offer an integrative perspective that encompasses changes across multiple domains. Clinician-rated tools, such as the Clinical Global Impression–Improvement scale, allow experienced evaluators to synthesize information from history, examination, and collateral reports. Patient-reported global improvement scales capture subjective perceptions of recovery, which may diverge from clinician assessments but are crucial in understanding satisfaction and adherence. Both types of global measures should be anchored to clearly described reference points, such as “compared with the week before starting treatment,” to reduce variability in interpretation over time.
Quality-of-life measures are particularly important, given the complex biopsychosocial impact of these disorders. Generic instruments, such as widely used health-related quality-of-life questionnaires, facilitate comparison with other conditions and support economic evaluations, while disease-specific tools tailored to functional neurological symptoms can provide greater sensitivity to change. When selecting quality-of-life instruments, investigators should consider respondent burden, cultural and language adaptability, and the extent to which domains like role functioning, emotional distress, stigma, and cognitive complaints are represented. Repeated administration at multiple timepoints allows examination of whether symptomatic or functional gains translate into broader improvements in life satisfaction and well-being.
Patient-reported outcome measures occupy a central role in assessing experiences that cannot be reliably observed by clinicians, such as anxiety, mood, fatigue, pain, and illness beliefs. Because comorbid anxiety and depression are common and can drive both symptom expression and help-seeking behavior, trials frequently include standardized scales for mood and anxiety as secondary outcomes. These measures help determine whether an intervention exerts its effect by reducing emotional distress, modifying maladaptive cognitions, or directly altering symptom mechanisms. To ensure interpretability, investigators should predefine hypotheses about how changes in psychological variables relate to the primary endpoint—for example, expecting that decreased catastrophizing will correlate with improved functional independence.
Objective or quasi-objective measures are valuable in a field where symptom self-report can be influenced by expectations, attention, or external incentives. In functional seizures, ambulatory or inpatient video-EEG monitoring may be used in substudies to corroborate event diaries and distinguish functional events from epileptic or other paroxysmal phenomena. For motor symptoms, digital gait analysis, accelerometry, wearable motion sensors, or force-plate assessments can provide granular data on movement patterns, variability, and response to distraction. When feasible, integration of such objective endpoints strengthens the credibility of findings, particularly in unblinded trials, but issues of cost, technical expertise, and participant tolerance must be considered in feasibility assessments during trial planning.
Blinded outcome assessment is a critical safeguard against bias, especially when blinding participants and therapists is not possible. Video recordings of neurological examinations, seizures, or movement tasks can be rated by independent, blinded experts using pre-specified scoring systems. Standardization of camera angles, task instructions, and recording conditions is essential to ensure comparability across time and sites. Blinded raters should be trained using example videos, with inter-rater reliability evaluated before formal data collection begins, and periodic recalibration performed to prevent drift. This approach also facilitates central review, which is particularly useful in multi-center designs where variability in local clinical practice could otherwise distort outcomes.
In trials relying heavily on diaries—such as seizure logs or symptom frequency records—careful attention must be paid to data quality and adherence. Electronic diaries with time-stamping and automated reminders can reduce backfilling and improve completeness compared with paper formats. Clear instructions are needed regarding what constitutes an event, how to handle clusters, and how to record uncertainty. Investigators should predefine rules for handling missing diary days, inconsistent entries, and extreme outliers, as these decisions can significantly influence effect estimates. Regular contact with participants to review diary completion and troubleshoot barriers helps maintain accuracy over follow-up.
Composite outcomes, combining several metrics into a single endpoint, may be considered when no single variable adequately captures meaningful change. For example, a composite might define response as a specified reduction in event frequency plus a minimal improvement in functional status, or as a combination of symptom reduction and return to work. While composites can increase sensitivity to clinically important improvements, they must be constructed and analyzed with caution. Each component should be clinically justified, measured with similar reliability, and reported separately in addition to the composite result to avoid obscuring divergent effects across domains.
Measurement timing strongly affects the interpretation of results. Functional neurological symptoms may fluctuate from day to day or even hour to hour, while deeper changes in coping strategies and neural circuitry may take months to consolidate. Trials should therefore use multiple assessment points, aligning them with hypothesized mechanisms of action. Early timepoints, such as immediately following a psychoeducational session or an intensive therapy block, can capture rapid, context-dependent shifts, whereas intermediate and long-term follow-up evaluates durability, relapse, and late-emerging benefits. Statistical analysis plans should specify how repeated measures will be modeled, including approaches to handle non-linear trajectories and differential timing of peak improvement among participants.
Handling missing data is particularly important in this population, given the relatively high risk of dropout and inconsistent attendance. Pre-specified strategies—such as mixed-effects models using all available data, multiple imputation under clearly stated assumptions, or sensitivity analyses comparing different missing-data approaches—should be detailed in the protocol. Simple methods such as last observation carried forward are generally discouraged because they can introduce substantial bias, especially when missingness is related to symptom change. Documenting reasons for missing assessments, and distinguishing between missingness due to logistical issues versus worsening symptoms or dissatisfaction, allows more nuanced interpretation of outcomes.
Psychometric properties of the selected tools should be scrutinized, particularly when adapting instruments from other neurological or psychiatric populations. Reliability, validity, responsiveness to change, and minimal clinically important difference values are essential considerations. Where evidence is limited for use in functional neurological disorder, pilot work or embedded validation substudies can assess whether a given scale behaves as expected. For example, investigators can explore correlations between scale scores and independent markers of severity, test-retest stability over short intervals without treatment changes, and responsiveness following known effective interventions in small cohorts.
Cultural and linguistic adaptation of outcome measures is vital in multi-national or multi-ethnic clinical trials. Direct translation without appropriate cultural validation can distort meaning, especially for constructs such as somatic symptom burden, stigma, or explanatory beliefs about illness. Forward and backward translation, cognitive interviewing with patients from target populations, and psychometric testing in each language group help ensure that measures retain their intended properties. Where cross-cultural differences in norms are anticipated—such as varying thresholds for reporting psychological distress—analysis plans should consider stratified or adjusted approaches rather than assuming equivalence across all sites.
Beyond symptom and function, healthcare utilization and economic outcomes provide important information for stakeholders and policymakers. Functional neurological disorder is associated with frequent emergency visits, hospitalizations, diagnostic testing, and specialist consultations. Trials may therefore track emergency department attendances, inpatient days, imaging and other investigations, and overall healthcare costs as secondary outcomes. These data, linked with quality-of-life measures, enable cost-effectiveness and cost-utility analyses that compare the value of novel interventions with usual care. Clear definitions of each utilization endpoint, along with standardized methods for extracting data from medical records or administrative databases, are necessary to avoid misclassification.
Process measures and intermediate mechanisms should not be overlooked in outcome planning, as they can illuminate why an intervention succeeds or fails. For psychological and rehabilitative treatments, process variables might include treatment expectancy, therapeutic alliance, engagement with homework or self-management tasks, and changes in illness beliefs or avoidance behaviors. Physiological process measures could encompass heart rate variability, startle responses, or neuroimaging markers, depending on the hypothesized pathways. Integrating such measures into trial design enables mediation analyses that test whether improvements in key processes account for changes in clinical outcomes, guiding refinement of future interventions.
Safety and adverse event reporting require particular nuance in functional neurological disorder trials, where increases in symptom intensity or frequency may represent transient destabilization on the way to improvement, non-specific fluctuation, or genuine harm. Protocols should specify how adverse events related to symptom exacerbation are classified, graded, and attributed to the intervention. Standardized questions about new or worsening symptoms, suicidality, self-harm, or significant functional decline should be incorporated into follow-up visits. Importantly, reporting formats must differentiate between serious and non-serious events and between anticipated, mechanism-consistent reactions and unexpected problems, to avoid over- or underestimating risk.
Integration of patient and caregiver perspectives into the selection and refinement of outcome measures strengthens the relevance and credibility of clinical trials. Structured consultations, focus groups, or advisory panels can identify outcomes that standard scales may overlook, such as confidence in self-management, perceived validation by clinicians, or family burden. Where feasible, investigators can co-develop brief supplementary questionnaires that address these domains, pilot them for clarity and acceptability, and incorporate them as exploratory endpoints. This participatory approach not only improves alignment between research and lived experience but may also enhance recruitment and retention by signaling that patient priorities are central to the study.
Intervention strategies and control conditions
Intervention strategies in functional neurological disorder research span pharmacological, psychological, rehabilitative, and hybrid models, each posing distinct design challenges. Trials must begin with a clearly articulated treatment rationale grounded in current theories about functional symptoms, such as altered predictive processing, abnormal attention to bodily sensations, or disrupted motor control and agency. This theoretical anchor guides selection of treatment components, dosing or session frequency, and the mechanisms expected to change. For example, interventions targeting maladaptive expectations might prioritize psychoeducation, cognitive restructuring, and exposure to feared movements or situations, whereas those focusing on motor retraining may emphasize graded practice of normal movement patterns and sensory-motor integration.
Psychoeducation is almost universally considered a core therapeutic element and often serves as the first intervention step. In clinical trials, its content, timing, and delivery format must be standardized to avoid unmeasured variability. Protocols should define key messages about the nature of functional neurological disorder, the role of brain network dysfunction, and the potential for reversibility, along with specific language to avoid stigmatizing or dualistic explanations. Education may be delivered in individual or group settings, by neurologists, psychologists, or therapists; whichever model is chosen, session length, use of written or digital materials, and opportunities for patient questions should be specified. Importantly, when psychoeducation is embedded within a broader therapy, investigators must decide whether it is part of the active intervention or also provided in control conditions to maintain ethical standards.
Psychological treatments, particularly cognitive-behavioral approaches and related modalities, are prominent intervention strategies. Manuals should clearly describe session structure, therapeutic techniques, and homework tasks, including adaptations tailored to symptom subtype (e.g., seizure-focused CBT versus approaches for functional movement disorders or weakness). Treatment “dose” encompasses the number of sessions, their duration, spacing, and any booster sessions, all of which must be predefined and justified in relation to expected mechanisms of change. To ensure fidelity, therapists require training, supervision, and adherence checklists or rating scales; a sample of sessions may be recorded and independently evaluated. Randomization plans should anticipate therapist effects, for example by having therapists provide both active and control interventions, or by stratifying by therapist in smaller studies.
Physiotherapy, occupational therapy, and other rehabilitation-based interventions play a central role in many functional motor presentations. Effective programs typically combine education, movement retraining, graded exposure, and strategies to shift attention away from symptoms toward functional goals. In clinical trials, rehabilitation protocols need detailed specification of exercises, progression criteria, use of assistive devices, and integration of dual-task or distraction techniques. Intensity and duration of therapy blocks are critical parameters, as brief intensive programs may have different effects from weekly outpatient sessions. To minimize contamination, therapists in the intervention arm should be trained in the theoretical model underpinning the protocol, while those providing usual care or control therapies should avoid incorporating key active elements unless intentionally designed as a comparative effectiveness study.
Multidisciplinary interventions integrating neurology, psychology, physiotherapy, occupational therapy, and sometimes social work or speech therapy may be especially promising but difficult to standardize. Trial designs must map out the roles of each discipline, the sequence or parallel nature of inputs, and mechanisms for team communication. Care pathways might include joint assessment sessions, shared goal-setting meetings with the patient, and coordinated discharge planning. Because these complex packages are resource-intensive, feasibility and recruitment planning should consider clinic capacity, staff availability, and the ability of participants to attend frequent appointments. When evaluating such multi-component programs, it is helpful to collect process measures describing which components were delivered, in what combination and intensity, to aid later interpretation of outcomes.
Pharmacological interventions, though less established, are sometimes evaluated to target comorbid conditions or hypothesized neurobiological mechanisms. Trials of medications must specify inclusion and exclusion criteria around concurrent psychotropic or anti-seizure drugs, washout periods if indicated, and procedures for dose titration and monitoring. Standard drug-trial features such as double-blind randomization, matching placebos, and rigorous adverse event surveillance are often more feasible for medications than for psychotherapy or physiotherapy. However, investigators must carefully distinguish between treating functional symptoms directly and treating associated mood, anxiety, or sleep problems; outcome measures and hypotheses should reflect the intended target of the pharmacological agent.
Digital and remotely delivered interventions, such as internet-based CBT modules, telehealth physiotherapy, or app-supported self-management programs, are increasingly relevant. These approaches can enhance access and scalability, but require careful attention to technology literacy, data security, and engagement strategies. Protocols should detail how participants are onboarded, how adherence is monitored (e.g., log-ins, completed modules, or wearable use), and what kind of therapist support is available (e.g., brief check-ins versus fully guided therapy). Remote designs may improve recruitment across wider geographic regions and can support long-term follow-up, but they also introduce attrition risks if participants lose interest or face technical barriers; pilot work can help refine approaches before large-scale trials.
The choice of control condition is particularly challenging in functional neurological disorder trials because of powerful expectancy, contextual, and therapeutic relationship effects. Simple “no treatment” or waitlist controls may maximize contrast but risk ethical concerns and differential dropout, especially when symptoms are severe or disabling. More commonly, investigators use enhanced usual care, supportive therapy, or attention control conditions that balance contact time and perceived support without incorporating the specific active ingredients of the experimental intervention. For example, a control for CBT might provide non-directive supportive counseling with the same number of sessions and similar session length, while deliberately avoiding structured cognitive restructuring or systematic exposure tasks.
Usual-care controls reflect real-world practice and can be valuable for pragmatic trials, but usual care is often poorly defined and variable across sites. To maintain interpretability, protocols should describe what usual care typically entails (e.g., follow-up with neurology, access to general physiotherapy, pharmacotherapy for comorbidities) and set minimal standards, such as ensuring that all participants receive a clear diagnostic explanation. Documentation of concomitant interventions, including frequency and type of healthcare contacts, is essential for understanding what patients actually receive in both arms. In cluster-randomized or stepped-wedge designs comparing implementation of specialized services versus usual pathways, careful baseline characterization of pre-existing practices helps contextualize outcomes.
Sham or placebo controls can be appropriate for certain interventions, such as neuromodulation (e.g., transcranial magnetic stimulation or transcranial direct current stimulation), biofeedback, or some physiotherapy techniques. Constructing a credible sham requires that the control condition mimic procedural aspects (e.g., device noise, scalp sensations, clinic setup) while lacking the hypothesized active component. Trials should predefine and test credibility of sham conditions—often via short questionnaires assessing participants’ beliefs about whether they received the real or control treatment—because differential beliefs between arms can influence outcomes. Blinding of participants and, when possible, treating personnel is critical; separate staff may be required to set up equipment and to conduct outcome assessments to preserve masking.
When using active comparator designs, such as comparing specialized FND-focused physiotherapy to generic neurological rehabilitation, outcomes must be chosen to detect both superiority and potential non-inferiority or equivalence, depending on the research question. Active comparators may be ethically preferable and more informative for health systems deciding among options, but they often reduce detectable effect sizes and require larger sample sizes. Treatment protocols for the comparator should be as detailed and standardized as for the experimental arm, including training and fidelity monitoring, to avoid unbalanced “intervention quality” that could bias interpretation.
In psychotherapy and multidisciplinary interventions, therapist allegiance and differential enthusiasm can inadvertently function as an uncontrolled “active ingredient.” Explicit strategies to manage this include training clinicians in both the experimental and control approaches (where appropriate), encouraging equipoise, and providing structured supervision that emphasizes adherence to the assigned manual rather than personal preference. Trial designs may use cross-over of therapists across study arms or allocate therapists to a single modality but ensure comparable levels of training and support. Measuring therapist factors—such as years of experience, attitudes toward FND, and perceived credibility of the interventions—allows exploratory analyses of their influence on outcomes.
Adherence and engagement with interventions require systematic attention, as partial exposure to therapy can dilute apparent treatment effects. Trial protocols should define adherence thresholds (e.g., completion of at least 70% of planned sessions or modules) and prespecify how low-adherence participants will be handled in primary analyses (typically intention-to-treat, with per-protocol analyses as secondary). Strategies to support engagement may include flexible session scheduling, reminder systems, brief motivational interviewing at baseline, or involvement of family members in goal setting. For interventions that require daily practice, such as home exercises, seizure self-monitoring, or cognitive exercises, adherence logs and, where feasible, objective monitoring (e.g., app usage data or device-recorded activity) can help distinguish lack of effect from insufficient exposure.
For some patients, abrupt introduction of intensive therapy may be counterproductive if they have not yet accepted the diagnosis or remain ambivalent about change. Stepwise intervention strategies, beginning with diagnostic explanation and brief stabilization or preparatory sessions before more demanding therapies, may therefore be advantageous. Clinical trials can compare stepped-care models—where patients begin with low-intensity interventions and escalate only if response is inadequate—against immediate provision of high-intensity treatment. Such designs require clear decision rules for stepping up care, based on standardized early response criteria, and careful tracking of the proportion of participants who ultimately receive each level of intervention.
Cross-over designs, in which participants receive both experimental and control treatments in different periods, are less common in functional neurological disorder because of concerns about carryover effects and durable learning or behavior change. However, they may be appropriate in early-phase feasibility studies or when testing brief, reversible procedures such as neuromodulation or specific behavioral experiments. If cross-over is used, adequate washout intervals and careful selection of outcomes that can return to baseline—or explicitly modeling carryover in the analysis—are essential. Investigators must also consider ethical implications if evidence suggests that one treatment is substantially more effective than the other.
Implementation considerations intersect with intervention and control design, particularly in multi-center trials. Variability in local expertise, service structures, and resource availability can lead to inconsistent delivery of both active and control conditions. Strategies to mitigate this include centralized training workshops, ongoing supervision via teleconferencing, and centralized review of selected session recordings or case summaries. Fidelity monitoring should be embedded in routine trial operations rather than treated as an optional add-on, and corrective feedback mechanisms should be in place to address identified deviations promptly.
Intervention safety monitoring is integral to trial design, especially given the potential for symptom exacerbation or emotional distress during therapy. Procedures should outline how worsening symptoms during treatment are assessed and managed, including criteria for temporary treatment pauses, crisis interventions, or referral to additional services. In exposure-based or trauma-informed psychological treatments, for example, protocols need clear guidance on managing acute distress, self-harm risk, or dissociative episodes. A data safety monitoring plan, proportionate to trial risk and size, helps ensure that adverse signals are recognized early and that appropriate modifications—such as adjustments in inclusion criteria or additional support measures—are considered systematically.
Decisions about intervention intensity and control conditions must be integrated with the broader trial design, including randomization schemes, sample size calculations, and selection of primary and secondary outcomes. High-intensity, resource-demanding interventions may necessitate smaller sample sizes and focus on proof-of-concept mechanisms, whereas lower-intensity, scalable approaches might be evaluated in larger pragmatic clinical trials emphasizing real-world applicability and cost-effectiveness. Aligning the complexity of the intervention and control arms with available resources, recruitment capacity, and the maturity of the evidence base is essential to produce interpretable results that can meaningfully inform clinical practice and health policy.
Ethical and regulatory challenges in fnd trials
Ethical and regulatory oversight of clinical trials for functional neurological disorder hinges on recognizing the dual vulnerability of this population: disability from symptoms and a history of stigmatization or diagnostic invalidation. Protocols must therefore go beyond generic protections and directly address how communication, consent, risk management, and post-trial responsibilities will be handled. Research ethics committees and regulators increasingly expect investigators to demonstrate an understanding of these specific vulnerabilities and to show how the study design mitigates them while still permitting robust, hypothesis-driven research.
Informed consent presents one of the most prominent ethical challenges. Many individuals with functional neurological disorder have previously received conflicting explanations or been told that their symptoms are “psychological,” “not real,” or “all in the mind.” These experiences can erode trust and complicate discussions about study participation. Consent processes should allocate sufficient time and use clear, non-judgmental language that aligns with contemporary explanatory models, emphasizing that symptoms are genuine, can be understood in terms of brain function, and are potentially reversible. Investigators must explicitly separate research aims from routine clinical care, while acknowledging that participation may or may not provide direct benefit, to avoid therapeutic misconception.
Capacity to consent is usually preserved but may be influenced by comorbid depression, anxiety, dissociation, cognitive complaints, or high distress levels. Protocols should set out procedures to assess understanding and voluntariness using teach-back methods, especially for more complex interventions. For individuals whose symptoms include episodic altered awareness, such as functional seizures, consent should be obtained during clear periods and documented carefully, with options for re-confirmation later in the study. In pediatric or adolescent populations, additional safeguards include age-appropriate explanations, assent procedures, and careful attention to the balance between parental wishes and the young person’s autonomy.
The ethical obligation to avoid harm is complicated by the possibility that intensive assessment or treatment can transiently worsen symptoms, even when ultimately beneficial. For example, exposure-based psychological therapies, detailed revisiting of triggering events, or challenging longstanding patterns of avoidance can provoke short-term spikes in anxiety or symptom severity. Similarly, repeated neurological examinations or monitoring procedures may heighten attention to bodily sensations. Study protocols must therefore include clear descriptions of anticipated transient reactions, how they will be explained to participants, and what supportive measures will be available during periods of exacerbation. These elements should be transparent in patient information sheets so that participants can make genuinely informed decisions about risk.
Monitoring safety requires particular nuance in interpreting symptom fluctuations, which are common in this disorder and not always attributable to the intervention. Adverse event frameworks should distinguish between expected variability, mechanism-consistent transient worsening, and signals of serious or unanticipated harm. Standardized questions at each visit, use of structured severity grading systems, and pre-defined thresholds for urgent review or treatment modification support consistent handling across sites. Data Safety Monitoring Boards or independent safety officers should be briefed about the typical course of functional neurological disorder so they can interpret patterns of adverse events appropriately, rather than assuming that any symptom increase indicates unacceptable risk.
Another ethical consideration is the choice of control conditions, particularly when individuals experience significant disability. Purely observational or waitlist controls may be methodologically attractive but problematic if they involve withholding any meaningful intervention from participants with marked impairment or high healthcare utilization. Many ethics committees will require that all participants at least receive a clear diagnostic explanation and access to usual clinical care, even when the experimental arm offers more intensive or specialized treatment. In stepped-wedge or delayed-start designs, investigators can balance the ethical imperative to eventually offer beneficial interventions with the need for a comparative phase, but must justify the length of delay and provide explicit criteria for early crossover if serious deterioration occurs.
Transparency about uncertainty and equipoise is essential when describing the rationale for randomization. Given the paucity of definitive treatments, many trials legitimately operate in conditions of clinical uncertainty, yet participants may understandably perceive that specialized therapies are superior to usual care. Investigators must explain that the purpose of randomization is to determine which approaches truly help, and that current evidence does not yet justify routine access to the experimental intervention outside a controlled setting. Overstating potential benefits of new treatments, especially those framed as cutting-edge or brain-based, risks coercive influence and undermines ethical recruitment.
The diagnostic label itself can carry ethical and regulatory implications. Some patients fear that a diagnosis of functional neurological disorder will jeopardize access to services, disability benefits, or legal recognition of their difficulties. Others worry that participating in a study might “lock in” a contested diagnosis. Consent materials should frankly address how the diagnosis will be recorded, who will have access to that information, and how it may interact with existing records. Investigators should avoid language that implies symptoms are fabricated or voluntary, and they should support participants in communicating the diagnosis to other clinicians or stakeholders if needed.
Privacy and confidentiality raise additional complexities when trials involve video-EEG monitoring, movement recordings, or detailed personal narratives, including trauma histories. Video clips of seizures or abnormal movements, while invaluable for blinded rating and central adjudication, are highly identifiable. Regulatory frameworks typically require that such data be encrypted, stored on secure servers, and shared only under strict data-sharing agreements. Participant information should spell out who may view recordings (e.g., local clinicians, external blinded raters), how long they will be retained, and whether they may be used in future research or teaching, with separate consent options where appropriate.
For interventions that explicitly explore trauma, adverse life events, or interpersonal dynamics, the risk of psychological distress and re-traumatization must be carefully mitigated. Ethics committees expect protocols to include detailed safety plans: criteria for pausing therapy, escalation pathways for acute suicidality or self-harm risk, and close liaison with local mental health services. Therapists should receive training not only in the specific treatment model but also in managing emergencies and in culturally sensitive approaches to discussing trauma and stigma. Regular supervision sessions can help therapists recognize when treatment is straying beyond the scope of the research protocol or their own competencies.
Stigma, both internalized and from healthcare professionals, can influence recruitment and participation. Some individuals may interpret an invitation to join a functional neurological disorder trial as confirmation that clinicians believe their symptoms are “not real,” while others may feel invalidated when approached for primarily psychological or behavioral interventions. To minimize harm, recruitment materials should emphasize that the disorder reflects functional changes in brain networks and that the study aims to improve understanding and care, not to question the legitimacy of symptoms. Training for all staff involved in recruitment and assessment should include explicit discussion of stigma, language to avoid, and strategies for validating patient experiences.
Equitable access to research participation is a core ethical requirement but can be difficult to achieve in this field. Specialized centers conducting functional neurological disorder trials are often located in urban academic hospitals, raising concerns about geographic and socioeconomic barriers. Moreover, cultural beliefs about illness and mental health may shape how functional symptoms are understood and whether research is considered acceptable. Investigators should work to reduce these disparities through flexible scheduling, partial reimbursement of travel costs where permissible, options for remote assessments, and translation of materials into relevant languages. Screening logs and monitoring of demographic characteristics over time help identify systematic exclusion of particular groups and support corrective strategies.
Inclusion and exclusion criteria raise delicate questions about who is “too complex” or “too high-risk” for participation. Excluding individuals with comorbid severe mental illness, substance use disorders, or active litigation may simplify safety management and outcome interpretation, but risks producing unrepresentative samples and limiting generalizability. Ethics committees may challenge criteria that systematically exclude those with the greatest need. Protocols should therefore justify each exclusion criterion in terms of safety, data integrity, or feasibility and consider alternative approaches, such as enhanced monitoring or stratified analyses, rather than blanket exclusion wherever possible.
Regulatory classification of interventions can create obstacles that are particularly salient in functional neurological disorder trials. Psychological and rehabilitative treatments do not always fit neatly into existing regulatory categories, leading to variable expectations across jurisdictions regarding trial registration, monitoring, and reporting requirements. For example, complex multidisciplinary programs might be viewed as service development in one system and as investigational interventions in another. Investigators must familiarize themselves with the applicable frameworks and engage early with regulators to clarify whether elements such as neuromodulation devices, digital therapeutics, or data-collection apps trigger additional obligations, including device registration or specific safety reporting standards.
Data protection regulations influence not only storage of personally identifiable information but also cross-border collaboration and secondary use of datasets. Multi-center or international trials must specify how data will be transferred, anonymized or pseudonymized, and accessed by external collaborators. Consent forms should distinguish between use of data for the specific trial and potential future research on related questions, including genetic or biomarker analyses if specimens are collected. In disorders where patients may feel particularly scrutinized or judged, explicit assurances about the limits of data sharing, along with options to participate in the trial without agreeing to broad secondary use, can bolster trust.
Another ethical and regulatory concern involves the handling of incidental findings from investigations performed as part of the study, such as brain imaging, EEG, or laboratory tests. Protocols should detail how clinically significant abnormalities will be communicated to participants and their treating clinicians, who will be responsible for follow-up, and how this information will be recorded. Investigators must guard against “diagnostic drift,” where a study aimed at functional symptoms inadvertently becomes a broad screening program without the resources or mandate to manage all discoveries. Clear thresholds for what constitutes an actionable incidental finding help maintain focus while respecting participants’ right to know about clinically relevant information.
Financial and non-financial incentives require careful calibration. Compensation for time, travel, and inconvenience is often necessary to enable participation, especially in lengthy or intensive protocols, but payments that substantially exceed local norms may unduly influence economically vulnerable individuals. Ethics committees typically expect justification for incentive levels and structures, including whether bonuses are tied to complete follow-up or simply to attendance at each visit. In functional neurological disorder, where fluctuations in symptoms or life circumstances can make sustained engagement difficult, punitive structures that withhold compensation for missed visits may be particularly problematic and should generally be avoided.
Post-trial obligations raise difficult questions in a context where effective treatments are scarce and often available only within specialized research settings. When an intervention shows clear benefit during the trial, investigators should plan how continued access will be managed once the formal study ends, especially for participants randomized to control conditions who did not receive the active treatment. This may involve offering the experimental therapy after study completion, integrating successful components into routine services, or providing structured referrals. Regulatory and ethical guidance increasingly emphasizes that such plans be outlined in the protocol and communicated during consent, to avoid perceptions that patients are being used solely as research subjects without regard for long-term care.
Publication and reporting ethics are central to maintaining confidence in a still-emerging evidence base. Selective reporting of favorable outcomes, non-publication of negative trials, or inadequate description of diagnostic standards and control conditions can distort perceptions of efficacy and safety. Investigators should prospectively register trials, including detailed specification of primary and key secondary outcomes, and commit to publishing results regardless of direction. Transparent reporting of recruitment numbers, reasons for ineligibility or refusal, attrition rates, and protocol deviations is particularly important in this field, where feasibility and recruitment challenges are common and can themselves inform future trial design.
Ensuring the independence of safety monitoring and analysis is another key safeguard. Where strong allegiances exist to specific therapeutic models—common in psychotherapy and rehabilitation research—there is a risk that adverse events or null results may be downplayed. Data Safety Monitoring Boards and, where possible, independent statisticians should have clear authority and access to unblinded data when necessary to protect participants. Trial steering committees should include clinicians with diverse perspectives, methodologists, and, ideally, patient representatives who can raise concerns about burden, acceptability, and emerging ethical issues.
Patient and public involvement is not merely a methodological strength but an ethical asset. Engaging individuals with lived experience of functional neurological disorder in the planning and governance of trials can help identify language that feels invalidating, procedures that are overly burdensome, or outcomes that fail to capture what matters most. Involvement might include participation in protocol development workshops, membership on oversight committees, or consultation on recruitment materials and consent forms. Integrating this input into formal ethical and regulatory submissions demonstrates respect for participants as partners rather than passive subjects and can improve both the ethical robustness and practical success of functional neurological disorder research.
