Study protocol for a prospective, cluster randomised control, hybrid type 1 effectiveness-implementation trial to evaluate the Reduction Of Seizures through Education and Support: the ROSES study Authors: Hamada H Altalib, Poojith Nuthalapati, Sana F Ali, Margaret Gopaul, Chinenye M Okafor, Bridget Gaglio, Abigail Herbst, Sohail Dewain, Saifeldin Hossameldin, Fabio Speranza, Priyanka Deep, Kelly Fischbein, Lawrence J Hirsch, Mary Jo Pugh, Heidi Munger Clary Published: 2026 ABSTRACT AbstractIntroductionEpilepsy affects 1% of the US population, imposing a significant burden through seizures, distress and medication side effects. Many adverse outcomes can be reduced with patient education and an acute seizure action plan. Electronic health record clinical templates can support this care but are often seen as burdensome. This study will develop a clinical template embedding an acute seizure action plan into routine neurology visits and examine its effect on care quality and healthcare use, as well as barriers and facilitators to adoption.Methods and analysisThis study is a single-site, prospective, quality improvement, hybrid type 1 effectiveness-implementation study, sequential mixed-method design. Providers are randomised to either the acute seizure action plan plus general health education, or general health education alone control group, with 10 patients per provider enrolled over 12–15 months and followed for 1 year. Phase 1 will collect baseline interviews and pre-enrolment seizure outcomes. Phase 2 will implement the intervention with quantitative data along with provider interviews at 6 months. Phase 3 will include 12-month interviews to assess real-world use of the acute seizure action plan. The primary outcome is change in epilepsy quality measure scores before and after implementation between groups. Secondary outcomes include changes in individual quality measures, seizure metrics and healthcare utilisation. Implementation processes will be assessed by the Unified Theory of Acceptance and Use of Technology and Theoretical Domains Framework.Ethics and disseminationThe study was approved by the Yale Institutional Review Board (Protocol ID: 2000032913). Findings will be disseminated through peer-reviewed publications and conference presentations.Keywords: EPILEPSY, EVIDENCE-BASED NEUROLOGY, HEALTH POLICY & PRACTICE, QUALITY OF LIFE, ANTICONVULSANTS WHAT IS ALREADY KNOWN ON THIS TOPIC WHAT THIS STUDY ADDS This quality improvement, hybrid type 1 effectiveness-implementation study will develop and evaluate a point-of-care electronic health record clinical template that embeds an acute seizure action plan into routine visits to standardise counselling, documentation and safety planning. This approach pairs workflow changes with the clinical template with an acute seizure action plan to improve education and counselling, document epilepsy quality measures, ensure appropriate use of rescue medications and reduce emergency visits for breakthrough seizures. This quality improvement, hybrid type 1 effectiveness-implementation study will develop and evaluate a point-of-care electronic health record clinical template that embeds an acute seizure action plan into routine visits to standardise counselling, documentation and safety planning. This approach pairs workflow changes with the clinical template with an acute seizure action plan to improve education and counselling, document epilepsy quality measures, ensure appropriate use of rescue medications and reduce emergency visits for breakthrough seizures. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY This study offers a replicable way to monitor and improve specialty care quality, using epilepsy as an example, and may show that integrating an acute seizure action plan into a clinical template improves care and safety. Expected benefits include better clinical care, improved patient quality of life, fewer injuries, faster recognition of seizure clusters, lower rates of status epilepticus, improved access to and use of rescue medications and reduced overall costs for people with epilepsy. This study offers a replicable way to monitor and improve specialty care quality, using epilepsy as an example, and may show that integrating an acute seizure action plan into a clinical template improves care and safety. Expected benefits include better clinical care, improved patient quality of life, fewer injuries, faster recognition of seizure clusters, lower rates of status epilepticus, improved access to and use of rescue medications and reduced overall costs for people with epilepsy. Introduction Epilepsy is a chronic disease that affects 1% of the US population1 and millions of individuals worldwide.2 People with epilepsy (PWE) face numerous challenges, continually underscoring the importance of finding ways to improve standards of epilepsy care and meaningfully improve quality of life (QoL).3 Approximately 30% of PWE have uncontrolled or poorly controlled seizures, despite treatment with antiseizure medication,4 placing them at risk for premature mortality5; injury from falls, drowning, and automobile accidents6; underemployment7; social stigma8; psychological distress9 and adverse drug effects.10 Furthermore, PWE may suffer from seizure clusters, many seizures in a short period of time, and are associated with early mortality and frequent emergency department visits.11 Professional societies, such as the American Academy of Neurology (AAN) and the National Association of Epilepsy Centers (NAEC), have established epilepsy quality measures to standardise and improve care.12,16 The AAN, NAEC outpatient team and the Quality Indicator for Epilepsy Treatment (QUIET) team recommend systematic screening for behavioural health conditions, counselling on safety (eg, driving, swimming), review of medication side effects and interactions, discussions of pregnancy-related risks and development of an acute seizure action plan that delineates steps to take during a seizure, strategies to prevent clusters and indications for urgent evaluation.14 16 Despite clear recommendations, implementation into routine practice is limited. Unfortunately, <20% of clinical encounters document counselling and behavioural screening by neurologists,17 and few PWE have a documented acute seizure action plan.18,20 Tracking the use of these measures is impeded by variability in documentation habits of healthcare professionals, which makes longitudinal tracking of the quality metrics challenging. Healthcare providers often record care in free-form electronic health record (EHR) notes, which obscures key data needed for longitudinal assessment. Several measures have been retired due to the difficulty of extracting reliable information from unstructured documentation.14 17 18 EHR-based tools, including templates and registries, can embed prompts, standardise data capture and support clinical decision-making.21,23 Previous implementation of EHR-based templates by the NorthShore University Health System, the Epilepsy Learning Healthcare System and the Veterans Health Administration (VHA) Epilepsy Centers of Excellence demonstrated feasibility and potential benefit.22 24 25 NorthShore University Health System created a structured clinical documentation support tool that captures extensive data, ranging from imaging and labs to behavioural health screening and integrates it into patient visits. Although this model supports comprehensive data collection, barriers such as added visit time, need for additional staff and workflow disruption have limited adoption and implementation.22 26 Similarly, the Epilepsy Learning Healthcare System is a patient-centred network of academic hospitals, patient and family partners, and nonprofit organisations that seeks to improve care through data collection and iterative interventions.24 27 Its approach captured key seizure data (frequency, type, date of last seizure) via case report forms, but many quality AAN (eg, medication adherence, behavioural health screening) relied on patient-reported outcomes or manual entry by additional clinical staff, creating workflow and efficiency challenges.21 23 In contrast, the VHA integrated its epilepsy registry directly into the EHR to enable real-time data collection and decision support.25 Nevertheless, clinician behaviour and workflow disruptions have limited adoption, and acute seizure action plans have not been integrated into these tools.28 This quality improvement, hybrid type 1 effectiveness-implementation study will develop and implement a point-of-care EHR template that operationalises epilepsy quality measures and generates an acute seizure action plan during the clinical encounter. We will evaluate clinical outcomes, QoL and healthcare resource utilisation. We hypothesise that this approach will improve quality of care, decrease injuries, expedite recognition and management of seizure clusters, enhance access to and use of rescue medications, and ultimately decrease the burden and cost of care for PWE. Preparatory stakeholder interviews and tool development We consulted leaders from the Northshore University Health System, the Epilepsy Learning Healthcare System and the VHA Epilepsy Centers of Excellence to inform the design of a clinical template that incorporated an acute seizure action plan. Across systems, clinicians cited scepticism, resistance to workflow changes, and added workload as persistent barriers. In response, we prioritised a clinical template with an acute seizure action plan that fits into healthcare providers’ routine visits, requires minimal training and aligns with existing documentation practices. The design emphasises usability, fewer clicks, streamlined data collection using toggles, dropdown menus and clearly organised sections to minimise cognitive load while capturing structured data. The template captures seizure details (frequency, type, date of last seizure), epilepsy history (risk factors, duration, triggers, severity), social history, diagnostic imaging and patient education and counselling within a clinician-focused design. Pilot testing indicates completion in approximately 6 min for new patients and less for follow-ups, addressing concerns about visit length, staffing and workflow integration. Epilepsy Quality Measures to Improve Performance Score We developed a clinical template that automatically populates a database to simplify quality measurement by capturing key elements of epilepsy care, including seizure frequency and type, cause of seizures, treatment response and tolerance and counselling delivered during visits. Building on existing quality measures from professional societies (eg, AAN, NAEC, QUIET), including those no longer in use (retired) due to feasibility constraints, we defined a structured scoring system, the Epilepsy Quality Measures to Improve Performance (EQUIP) score, to enable routine monitoring within standard workflows (table 1). Table 1EQUIP scoreMeasure categoryMeasureEndorsing body/yearCommentsPrimary measuresSeizure-specific assessment for all seizure types at all visitsSeizure frequencyAAN 202112For all seizure types: Key measure of efficacy: if ‘zero’ then seizure free, then data of last seizure can confirm the length of seizure freedom. We did not include ‘seizure freedom’ as recommended in 2022 AAN EQMSeizure type*AAN 202112For all distinct semiologies: Plays a crucial role in patient awareness, treatment decisions and monitoring for treatment efficacySeizure aetiologyAAN 201414Patient visits with seizure type and epilepsy aetiology or syndrome documented, or where testing was ordered to determine epilepsy aetiology, seizure type or epilepsy syndromeTime since last seizureAAN 202112Supports understanding of seizure control over timeSevere seizures†Not endorsed explicitlySeizure severity discussed in AAN 2021,12 but no guidelines recommended evaluating seizure clusters or status epilepticusAAN 2021:The work group discussed seizure severity but decided not to include it as a quality measureTreatmentMedication adherenceNAEC,16 QUIET15Indicator of treatment complianceCounsel on antiseizure medication side effects and/or interactionsAAN 2014,14 NAEC,16 QUIET15Informs shared decision-making and monitoringComorbidityBehavioural health screening/referralAAN 2017,14 NAEC,16 QUIET15Recognises psychiatric comorbidity and ensures access to careCounselling and educationDriving/safety/injury prevention (eg, SUDEP, swimming, driving)QUIET15Patient safety educationLifestyle/trigger counselling (eg, missed meds, alcohol, sleep)QUIET15Avoid seizure recurrence through lifestyle changesAdd’l measure for womenWomen of reproductive age: counsellingAAN 2017,14 NAEC,16 QUIET15Contraception/family planning (documentation of patient being pre/post-menstrual, surgically sterile or absent reproductive organs)Secondary measuresMagnetic resonance imaging (MRI) / electroencephalography (EEG) orderedNAEC,16 QUIET15Several EQM studies show that neurologists order MRIs and EEGs >90% of the time. However, in resource-poor regions, it may be important to document as a barrier to serviceEducated or referred patients to one or more Managing Epilepsy Wellness programmesNot proposed in prior guidelinesEmerging evidence demonstrates effectiveness in improving seizure outcomes, quality of life and independenceASAPNot proposed in prior guidelinesPractical tool for managing breakthrough seizures and emergency planningIn drug-resistant cases, referral to comprehensive epilepsy centre or discussion of epilepsy surgical optionsAAN 2017,14 NAEC,16 QUIET15Ensures access to advanced care pathways for refractory epilepsyEQUIP score calculation: each primary measure receives one score when documented by the provider. EQUIP score=(completed measures/total applicable measures)×100. Total applicable measures: 10 for men and non-reproductive age women; 11 for women of reproductive age (includes reproductive health counselling). Score range: 0%–100%.*Seizure types are categorised in accordance with the 2017 International League Against Epilepsy operational classification of seizure types.36†Severe seizures such as status epilepticus and seizure clusters are at higher risk for morbidity and mortality. More aggressive treatment, including appropriate counselling and referral for advanced care, such as neurosurgery, is warranted.AAN, American Academy of Neurology; ASAP, acute seizure action plan; EQM, epilepsy quality measures; EQUIP, Epilepsy Quality Measures to Improve Performance; NAEC, National Association of Epilepsy Centers; QUIET, Quality Indicator for Epilepsy Treatment; SUDEP, sudden unexpected death in epilepsy. EQUIP score EQUIP score calculation: each primary measure receives one score when documented by the provider. EQUIP score=(completed measures/total applicable measures)×100. Total applicable measures: 10 for men and non-reproductive age women; 11 for women of reproductive age (includes reproductive health counselling). Score range: 0%–100%. Seizure types are categorised in accordance with the 2017 International League Against Epilepsy operational classification of seizure types.36 Severe seizures such as status epilepticus and seizure clusters are at higher risk for morbidity and mortality. More aggressive treatment, including appropriate counselling and referral for advanced care, such as neurosurgery, is warranted. AAN, American Academy of Neurology; ASAP, acute seizure action plan; EQM, epilepsy quality measures; EQUIP, Epilepsy Quality Measures to Improve Performance; NAEC, National Association of Epilepsy Centers; QUIET, Quality Indicator for Epilepsy Treatment; SUDEP, sudden unexpected death in epilepsy. The EQUIP score includes seizure-specific assessments (eg, frequency, type, aetiology, date of last seizure), occurrence of severe events (eg, seizure clusters, status epilepticus), treatment adherence and counselling on medication side effects. It also includes quality indicators related to reproductive health counselling for women of childbearing age, behavioural health screening and referrals and patient education on safety and lifestyle-related seizure triggers. A detailed summary of the EQUIP set and their alignment with existing quality measurement sets is provided in table 1. Each applicable primary measure is scored as completed (1 point) or not completed (0 points) and the percentage score is calculated as (completed measures/total applicable measures) × 100, with 10 primary measures for men and women not of reproductive age and an additional reproductive health counselling measure for women of reproductive age. Secondary measures include documentation of neurodiagnostic evaluations (electroencephalogram and MRI), referrals for surgery in drug-resistant epilepsy, acute seizure action plans and referrals to epilepsy self-management and wellness programmes. This evaluation framework is designed to be feasible, scalable and clinically actionable for improving epilepsy care delivery. Study objectives Primary objective To determine whether the clinical template with an acute seizure action plan improves the overall EQUIP score among neurologists. Secondary and other objectives To evaluate the performance of the clinical template with acute seizure action plan at the individual quality measure level, including seizure assessment measures (frequency, type and date of last seizure), education and counselling measures (on topics such as medication side effects, interactions, treatment adherence, comorbidities and safety). To evaluate the performance of the clinical template with acute seizure action plan at the individual quality measure level, including seizure assessment measures (frequency, type and date of last seizure), education and counselling measures (on topics such as medication side effects, interactions, treatment adherence, comorbidities and safety). To evaluate the impact of the clinical template with an acute seizure action plan on the number of seizure-related emergency department visits and the number of inpatient neurology admissions over a 12-month period. To evaluate the impact of the clinical template with an acute seizure action plan on the number of seizure-related emergency department visits and the number of inpatient neurology admissions over a 12-month period. To evaluate barriers and facilitators to implementing the clinical template with an acute seizure action plan in the real world, evaluate the impact of the clinical template with an acute seizure action plan on QoL of PWE compared with standard of care, and evaluate the impact of the clinical template with an acute seizure action plan on the number of monthly seizure cluster visits compared with standard of care. To evaluate barriers and facilitators to implementing the clinical template with an acute seizure action plan in the real world, evaluate the impact of the clinical template with an acute seizure action plan on QoL of PWE compared with standard of care, and evaluate the impact of the clinical template with an acute seizure action plan on the number of monthly seizure cluster visits compared with standard of care. To describe and compare rescue medication use differences on clinical and healthcare resource utilisation outcomes compared with standard of care. To describe and compare rescue medication use differences on clinical and healthcare resource utilisation outcomes compared with standard of care. To describe baseline demographics and clinical characteristics of patients with and without refractory epilepsy and seizure clusters by randomised group (clinical template with an acute seizure action plan plus general health education vs general health education alone). To describe baseline demographics and clinical characteristics of patients with and without refractory epilepsy and seizure clusters by randomised group (clinical template with an acute seizure action plan plus general health education vs general health education alone). Methods and analysis Overall design This study is a single-site, prospective, quality improvement, hybrid type 1 effectiveness-implementation study,29 using a sequential mixed-methods design. Data collection Primary outcome The primary outcome is the change in EQUIP scores from baseline to 12 months post enrolment. Secondary and other outcomes Secondary outcomes will be measured at the patient level and include individual quality measures outlined in the EQUIP scoring system, as well as clinical seizure-related outcomes, treatment patterns and healthcare resource utilisation. These outcomes will be assessed at baseline and compared with the follow-up time point at 12 months post-enrolment: Change in individual EQUIP quality performance measures from baseline to 6 and 12 months, including seizure assessment measures (frequency, type and date of last seizure), education and counselling measures (on medication side effects, interactions, treatment adherence and comorbidities). Change in individual EQUIP quality performance measures from baseline to 6 and 12 months, including seizure assessment measures (frequency, type and date of last seizure), education and counselling measures (on medication side effects, interactions, treatment adherence and comorbidities). Change in number of seizures reported from baseline to 6 and 12 months post-enrolment. Change in number of seizures reported from baseline to 6 and 12 months post-enrolment. Frequency and proportion of refractory epilepsy (yes/no) at 6 and 12 months post-enrolment, Frequency and proportion of refractory epilepsy (yes/no) at 6 and 12 months post-enrolment, Frequency and proportion of seizure type at 6 and 12 months post-enrolment. Frequency and proportion of seizure type at 6 and 12 months post-enrolment. Change in seizure cluster frequency from baseline to 6 and 12 months post-enrolment. Change in seizure cluster frequency from baseline to 6 and 12 months post-enrolment. Change in seizure duration (minutes) from baseline to 6 and 12 months post-enrolment. Change in seizure duration (minutes) from baseline to 6 and 12 months post-enrolment. Change in number of antiseizure medications prescribed from baseline to 6 and 12 months post-enrolment. Change in number of antiseizure medications prescribed from baseline to 6 and 12 months post-enrolment. Change in postindex rescue medication prescription and utilisation from baseline to 6 and 12 months. Change in postindex rescue medication prescription and utilisation from baseline to 6 and 12 months. Change in healthcare resource utilisation from baseline to 6 and 12 months post-enrolment (inpatient, emergency department, outpatient, urgent care, referrals for other care, number of antiseizure medications prescribed in their medical records). Change in healthcare resource utilisation from baseline to 6 and 12 months post-enrolment (inpatient, emergency department, outpatient, urgent care, referrals for other care, number of antiseizure medications prescribed in their medical records). Theoretical frameworks Implementing new practices in clinical care requires changes in healthcare providers’ behaviour, which is facilitated by understanding the determinants of current and goal behaviour. To better understand these determinants and thus support implementation strategy development, this study will be guided by the Unified Theory of Acceptance and Use of Technology Model and the Theoretical Domains Framework. The Unified Theory of Acceptance and Use of Technology Model is one of the most frequently used models of information technology adoption. It proposes that behavioural intention to use a technology is influenced by several factors, specifically, performance expectancy, effort expectancy, social influence and facilitating conditions. Sex, age, experience and voluntariness of use are identified moderators.30 31 The Theoretical Domains Framework is an integrated framework consisting of 14 domains and was initially developed for implementation research to identify influences on healthcare professional behaviour related to implementation of evidence-based practices.32 These frameworks will guide the development of the interview and focus group guides, and the results will be mapped back to them (online supplemental material). Procedure This study will be conducted in patients with diagnosed epilepsy receiving care within the Yale New Haven Health System. Ten healthcare providers will be randomised to either general health education alone (control group) or general health education plus the clinical template with an acute seizure action plan (intervention group), and all patients enrolled in the study will receive the routine clinical care allocated to them. Figure 1 summarises the study design and study procedures. ROSES study design: Gantt chart. Observational period 1 is the index date. Observational period 2 is the closest visit to 6 months post index date ±30 days. Observational period 3 is the closest visit to 12 months post index date ±30 days. Clinical template with an ASAP, acute seizure action plan; EQUIP, Epilepsy Quality Measures to Improve Performance; GHE, General health education; QOLIE-10, Quality of Life in Epilepsy Inventory-10; ROSES, Reduction Of Seizures through Education and Support. Phase 1: preimplementation period Qualitative measures One-on-one provider interviews: In-depth, semistructured interviews will be conducted with all healthcare providers at baseline to assess current epilepsy care practices and identify barriers and facilitators to implementing a clinical template with an acute seizure action plan. Healthcare providers randomised to the clinical template with an acute seizure action plan plus general health education arm will also provide feedback to refine the clinical template with an acute seizure action plan. One-on-one provider interviews: In-depth, semistructured interviews will be conducted with all healthcare providers at baseline to assess current epilepsy care practices and identify barriers and facilitators to implementing a clinical template with an acute seizure action plan. Healthcare providers randomised to the clinical template with an acute seizure action plan plus general health education arm will also provide feedback to refine the clinical template with an acute seizure action plan. Patient focus groups (up to 10 participants): Focus groups with PWE will explore their experiences with epilepsy care, their knowledge and comfort in managing epilepsy, and the tools and resources they currently use. Patient focus groups (up to 10 participants): Focus groups with PWE will explore their experiences with epilepsy care, their knowledge and comfort in managing epilepsy, and the tools and resources they currently use. Caregiver focus groups (up to 10 participants): Similar focus groups will be conducted with caregivers to understand their experiences and perspectives on epilepsy care management. Caregiver focus groups (up to 10 participants): Similar focus groups will be conducted with caregivers to understand their experiences and perspectives on epilepsy care management. Quantitative measures Collection of patient seizure outcomes: Seizure measures will be collected from medical charts, covering up to 3 years of baseline data to evaluate the effectiveness of the intervention. Collection of patient seizure outcomes: Seizure measures will be collected from medical charts, covering up to 3 years of baseline data to evaluate the effectiveness of the intervention. Phase 2: implementation period Quantitative measures Clinic visits: Patients will attend clinic visits where they will receive either the acute seizure action plan plus general health education or general education alone based on provider randomisation. Clinic visits: Patients will attend clinic visits where they will receive either the acute seizure action plan plus general health education or general education alone based on provider randomisation. Monthly survey of seizure outcome measures: Seizure frequency, severity, and occurrence of seizure clusters will be recorded monthly from study patients. Monthly survey of seizure outcome measures: Seizure frequency, severity, and occurrence of seizure clusters will be recorded monthly from study patients. QoL scale: The Quality of Life in Epilepsy Inventory (QOLIE-10) will be administered every 4 months to evaluate patients’ QoL. QoL scale: The Quality of Life in Epilepsy Inventory (QOLIE-10) will be administered every 4 months to evaluate patients’ QoL. Epilepsy quality measures (EQUIP): Epilepsy quality measures will be collected from clinic notes within 7 days after study visits, every 4 months. Epilepsy quality measures (EQUIP): Epilepsy quality measures will be collected from clinic notes within 7 days after study visits, every 4 months. Qualitative measures One-on-one provider interviews: In-depth, semistructured interviews will be conducted with all healthcare providers at 6 months postimplementation. Healthcare providers in the acute seizure action plan plus general health education arm will discuss their experiences, while those in the control (general health education alone) arm will identify anticipated barriers and facilitators to the use of the clinical template with an acute seizure action plan. One-on-one provider interviews: In-depth, semistructured interviews will be conducted with all healthcare providers at 6 months postimplementation. Healthcare providers in the acute seizure action plan plus general health education arm will discuss their experiences, while those in the control (general health education alone) arm will identify anticipated barriers and facilitators to the use of the clinical template with an acute seizure action plan. The implementation period will evaluate the comparative effectiveness of the acute seizure action plan plus general health education intervention against the general health education alone control arm. The general patient education will be delivered to the patients on the nature of epilepsy, emphasising the importance of medication adherence, diet and exercise. Quantitative data collection will include baseline QoL, seizure frequency (including seizure clusters), severe seizures, 1-year health service utilisation, and basic demographic and clinical characteristics. Seizure outcome measures will be collected monthly, while QOLIE-10 will be collected within 7 days before the study visits. EQUIP will be collected within 7 days after the study visits during the respective observational periods. Each participant is expected to have follow-up visits during the 12-month study period, with evaluations of seizure outcome measures and QOLIE-10 outcomes at baseline, 6 months and 12 months post-enrolment. Qualitative data from interviews with healthcare providers in the acute seizure action plan plus general health education arm at 6 months postimplementation will focus on their experiences with the intervention and barriers and facilitators, while interviews with healthcare providers in the general health education alone arm will focus on anticipated barriers and facilitators of using a clinical template with an acute seizure action plan. Phase 3: postimplementation period Qualitative measures One-on-one provider interviews: At 12 months postimplementation, in-depth, semistructured interviews will be conducted with all healthcare providers to understand their experiences with the clinical template with an acute seizure action plan. One-on-one provider interviews: At 12 months postimplementation, in-depth, semistructured interviews will be conducted with all healthcare providers to understand their experiences with the clinical template with an acute seizure action plan. Patient focus groups (up to 10 participants): Focus groups with PWE will be held at 12 months to evaluate the utility of the acute seizure action plan, including barriers and facilitators to its real-world use, as well as insights into health choices and behaviours. Patient focus groups (up to 10 participants): Focus groups with PWE will be held at 12 months to evaluate the utility of the acute seizure action plan, including barriers and facilitators to its real-world use, as well as insights into health choices and behaviours. Caregiver focus groups (up to 10 participants): Similar focus groups will be conducted with caregivers at 12 months to understand their perspectives on the utility of the acute seizure action plan, barriers and facilitators to its use, and the factors influencing health choices and behaviours. Caregiver focus groups (up to 10 participants): Similar focus groups will be conducted with caregivers at 12 months to understand their perspectives on the utility of the acute seizure action plan, barriers and facilitators to its use, and the factors influencing health choices and behaviours. The postimplementation period aims to gather qualitative data on the experiences of healthcare providers, patients, and caregivers with the acute seizure action plan intervention, focusing on its practicality, effectiveness and the factors driving health-related decisions. General health education alone providers will receive the acute seizure action plan intervention at 6 months. To minimise contamination risk, EHR tool access will be restricted from control group providers until their 6-month transition time point. Study population and selection criteria Providers (cluster level) This study will include 10 healthcare providers in the Yale Epilepsy Division, each responsible for providing care for a minimum of 10 patients enrolling in the study. The total patient enrolment is expected to be a minimum of 100 PWE. Healthcare providers will be considered clusters and will be randomised to the acute seizure action plan plus general health education arm or the general health education alone control arm. Inclusion criteria for providers Providers with the Yale Comprehensive Epilepsy Center who provide outpatient clinical care for PWE. Providers with the Yale Comprehensive Epilepsy Center who provide outpatient clinical care for PWE. Exclusion criteria for providers Providers who do not use the Yale EPIC EHR. Providers who do not use the Yale EPIC EHR. Providers who only dictate their notes and are not willing to type any information or complete a template into the note. Providers who only dictate their notes and are not willing to type any information or complete a template into the note. People with epilepsy Adult patients receiving epilepsy care within the Yale New Haven Health System will have the option to enrol in the study. On enrolment of the first patient, the healthcare provider will be randomised 1:1 to either general health education alone (control) or acute seizure action plan plus general health education (treatment). Inclusion criteria for PWE People ≥18 years of age with an International Classification of Diseases, 10th Revision diagnosis of epilepsy seen in Yale Comprehensive Epilepsy Center. People ≥18 years of age with an International Classification of Diseases, 10th Revision diagnosis of epilepsy seen in Yale Comprehensive Epilepsy Center. Focal-onset or generalised epilepsy. Focal-onset or generalised epilepsy. Exclusion criteria for PWE People with active psychogenic seizures or people whose primary diagnosis is psychogenic seizures or functional neurological disorders. People with active psychogenic seizures or people whose primary diagnosis is psychogenic seizures or functional neurological disorders. Severe cognitive impairment that precludes them from reporting seizure history. Severe cognitive impairment that precludes them from reporting seizure history. Active suicidality. Active suicidality. Progressive neurological disorder that is associated with terminal illness such as malignant brain tumour or Alzheimer’s disease. Progressive neurological disorder that is associated with terminal illness such as malignant brain tumour or Alzheimer’s disease. Currently have and are utilising an acute seizure action plan. Currently have and are utilising an acute seizure action plan. Caregivers Caregivers of adult patients receiving care for epilepsy within the Yale New Haven Health System will enrol in the study. Inclusion criteria for caregivers Provides formal or informal care for PWE. Provides formal or informal care for PWE. Exclusion criteria for caregivers In an active legal dispute with the PWE. In an active legal dispute with the PWE. Active substance abuse, psychotic, or actively suicidal. Active substance abuse, psychotic, or actively suicidal. Data collection Patients will be followed as per current clinical practices for their condition, except for the addition of monthly follow-up phone calls, which also include QOLIE-10 assessment. No additional diagnostic or monitoring procedures (aside from the randomisation of clinical template with an acute seizure action plan plus general health education vs general health education alone) will be applied. Table 2 illustrates the assessment schedule. The choice of medical treatment is made independently by the treating physician in the regular course of practice and is not influenced by the study protocol. Table 2Schedule of activities for quantitative dataAssessmentsOP1*OP2†OP3‡Unplanned termination OPDemographic dataXVerification of selection criteriaXWithdrawal criteriaXXXSeizure, medical/procedures historyXGeneral health educationXRecording of prescribed treatment and proceduresXXXXRecording of concomitant medicationXXXXEQUIPXXXQOLIE-10XXXX*OP 1 is the enrolment date.†OP 2 is the closest visit to 6 months post enrolment date ±30 days.‡OP 3 is the closest visit to 12 months post enrolment date ±30 days.EQUIP, Epilepsy Quality Measures to Improve Performance; OP, observational point; QOLIE-10, Quality of Life in Epilepsy Inventory-10. Schedule of activities for quantitative data OP 1 is the enrolment date. OP 2 is the closest visit to 6 months post enrolment date ±30 days. OP 3 is the closest visit to 12 months post enrolment date ±30 days. EQUIP, Epilepsy Quality Measures to Improve Performance; OP, observational point; QOLIE-10, Quality of Life in Epilepsy Inventory-10. Data sources Data for this study will be derived from multiple sources, including patient medical records, patient-reported outcomes, healthcare provider interviews and patient and caregiver focus groups. Both primary and secondary data will be collected to examine baseline characteristics and outcomes. Data linkage is possible due to a single site and a single system that has inter-operability between the data sets. Study size This cluster-randomised trial assigns healthcare providers to either an acute seizure action plan plus general health education or general health education alone control arm (1:1 ratio). We aim to recruit ≥10 providers (5 per arm) and 10 patients per healthcare provider. This sample size provides 80% power to detect a two-point improvement on the EQUIP scale (SD=4.572). We defined this two-point change—a 20% increase on the 10-point scale—as our threshold for clinical significance.33 This sample size calculation assumes an estimated intracluster correlation is 0.01 with a two-sided test at the 0.05 significance level using the two means completely randomised statistic. Data analysis Quantitative analysis (survey) Data will be collected electronically (through REDCap). Quantitative data will be analysed using the principle of intention-to-treat. This means patients will be analysed according to the initial healthcare provider’s randomisation (acute seizure action plan plus general health education or general health education alone control arm), with secondary analyses of per-protocol results. Patient characteristics will be described in the overall population and the treatment and control arms over the baseline period. Categorical variables will be detailed by counts and proportions. Continuous variables will be described by mean, SD, median, and interquartile range (first and third quartiles). For the QoL measures, a multilevel mixed model will be used that includes the healthcare provider as a random cluster effect and the treatment group as a fixed effect. The cluster will be included as a random effect to account for the dependency between patients treated by the same healthcare provider. Analysis of the primary outcome measure To account for the dependency between patients treated by the same healthcare provider, a multilevel mixed model will be used. The primary analysis will compare the change from baseline in EQUIP measures between patients in the clinical template with an acute seizure action plan plus general health education arm and those in the general health education alone control arm. This means a random effects least-squares mean regression model will be applied with the healthcare provider as the cluster level random effect and the treatment group as the fixed effect. Additional variables to potentially adjust for include age at enrolment, sex, number of years with epilepsy, seizure severity, race and ethnicity. Precision will be assessed using two-sided 95% CIs. Analysis of the secondary outcome measures For secondary outcomes, this study will evaluate the number of seizure-related emergency department visits and the number of inpatient neurology admissions over a 12-month time period, treating these as separate outcomes. Change in QoL will be evaluated from baseline between patients in the clinical template with an acute seizure action plan plus general health education arm and those in the general health education alone control arm. To analyse these outcomes, a random effects least squares mean regression model will be applied with the healthcare provider as the random cluster effect and the treatment group as fixed effect. In addition, we will examine the number of monthly seizure clusters over a 6-month and 12-month time period using a zero-inflated Poisson mixed model with healthcare provider as the cluster level random effect and treatment group as the fixed effect. This method will be applied to account for the anticipated high level of patients with zero hospital visits. If overdispersion is detected by dividing the residual deviance by the degrees of freedom, then a zero-inflated negative binomial mixed model is recommended. Precision for the rate ratios will be assessed using two-sided 95% CIs. Similar modelling approaches will be applied to the clinical (average monthly seizures) and healthcare resource utilisation (neurology visit, inpatient visit, emergency department visit) outcomes for the objective of describing and comparing rescue medication use (yes/no) differences on clinical and healthcare resource utilisation outcomes. Depending on the type of measurement, the hierarchical model will always include the healthcare provider as a random cluster effect, along with treatment and rescue medication use as a fixed effect. Descriptive analyses of baseline demographics and clinical characteristics will be displayed and stratified by refractory epilepsy (yes/no), seizure clusters (yes/no) (online supplemental table S1), and treatment arm. Outcome definition and measurement for quantitative variables are illustrated in online supplemental table S1. Sensitivity analysis A sensitivity analysis on treated patients will be performed. This will help to identify the proportion of patients that may have been misallocated and analyse them according to the treatment (clinical template with an acute seizure action plan plus general health education and those with general health education alone) that they received, instead of the intent to treat, which analyses patients according to the initial healthcare provider randomisation (clinical template with an acute seizure action plan plus general health education or general health education alone) that they are allocated to. Analysis of safety variables A separate safety analysis will not be conducted. Safety measures such as status epilepticus, adverse medication events, seizure clusters and healthcare resource utilisation will be included during the primary analysis. Any safety results found will be described. Qualitative analysis (interviews and focus groups) All interview and focus group audio recordings will be uploaded within 24 hours to a secure study-specific file-sharing site and deleted from audio recorders. The recordings will be transcribed verbatim, and the transcribed data will be reviewed by qualitative scientific staff to ensure that all personally identifiable information and protected health information are removed and to correct obvious transcription errors. The narrative data from the interview and focus group transcripts will be entered into ATLAS.ti qualitative data analysis software (V.22.0) by a study team member.34 ATLAS.ti is a specialised software designed for the qualitative analysis of textual, graphical, audio and video data. The coding structure used for thematic analysis will be created from a combination of inductive and deductive coding and will take the three participant groups (healthcare providers, patients and caregivers) into account. Data collection and analyses will be conducted iteratively. Limitations of the research methods As a single-site study, conducted at a tertiary epilepsy centre with relatively few healthcare providers, generalisability may be constrained. Lack of masking and potential provider contamination given they are all in the same practice setting are additional limitations, along with the risk of differential loss to follow-up. However, patients in the control arm continue to receive standard clinical care, mitigating ethical concerns. Additionally, no minimum seizure frequency threshold was specified as an inclusion criterion, as the Acute Seizure Action Plan is intended to benefit patients across the full spectrum of epilepsy presentations (including newly diagnosed, refractory and treatment-responsive cases), regardless of seizure burden.35 Baseline seizure frequency and epilepsy subtype will be included as covariates in all analyses to account for this heterogeneity. Second, seizure types are classified using the 2017 International League Against Epilepsy (ILAE) operational classification,36 which remains the standard embedded in current quality frameworks (AAN, NAEC, QUIET); retaining the 2017 framework ensures comparability with the existing literature. The 2025 ILAE classification37 was published after the clinical template had been developed, deployed in the EHR. We intend to add the equivalent new terms to the old ones in the database and in future reporting. In light of these limitations, this study evaluates whether a clinical template with an acute seizure action plan may improve epilepsy quality performance and recommends the use of consolidated epilepsy quality performance measures to evaluate clinical and QoL outcomes. Ethics and dissemination Patient consent was waived because this study is evaluating an implementation strategy to improve care quality. However, patients were informed about the study, and their participation was entirely voluntary. Patients were also informed by the study team that their participation choices would not affect their clinical care or relationship with providers. We intend to disseminate and publish the results of this study whether they are positive, negative or inconclusive. The results of the study will be published in peer-reviewed journals and presented at national conferences. Patient and public involvement Patients were not involved in the initial protocol development, formulation of the research question, selection of outcome measures or the design of study procedures and recruitment strategies. Patients and caregivers will not be directly involved in recruitment or conduct of study. However, patient and caregiver perspectives will be sought during the prospective phase of the trial through focus groups as part of the qualitative component of the study to support implementation and guide future research. Supplementary material The funder did not influence the results/outcomes of the study despite author affiliations with the funder. Footnotes Funding: This project has received funding from UCB (grant/award number: not applicable). UCB was involved in the design of the study and interpretation of the data. UCB was responsible for the quantitative data analysis. The authors, some of whom are UCB employees, were responsible for data interpretation, revising the manuscript for intellectual content and approving the manuscript for submission. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study was approved by a Yale University Institutional Review Board (ID: 2000032913). References Associated Data Supplementary Materials