Protocol for evaluating the effects of the intelligent multimodal symptom assessment and response system in patients with pemphigus vulgaris: a randomised controlled trial Authors: Yi Chen, Ling Wang, Juan Gao, Runxi Xiao, Shuyu Yan, Xi Pan, Yuhua Yu, Changrong Yuan, Yongyong Yuan Published: 2026 ABSTRACT AbstractIntroductionPemphigus vulgaris is a rare autoimmune blistering disease characterised by recurrent mucocutaneous erosions, high symptom burden and unpredictable relapse. Current management relies mainly on pharmacological therapy and hospital-based follow-up, with limited real-time monitoring and individualised support in home-based disease management. To address these challenges, this trial aims to evaluate the effectiveness of an Intelligent Multimodal Symptom Assessment and Response System, integrating patient-reported outcomes, wearable physiological data and image-based lesion assessments, to improve symptom management and quality of life in pemphigus vulgaris patients. Primary objective is to evaluate the effectiveness of the Intelligent Multimodal Symptom Assessment and Response System in improving symptom alerting and symptom management outcomes in patients with pemphigus vulgaris. Secondary objectives are to enhance patients’ self-management ability in symptom monitoring and control, to improve treatment adherence throughout the follow-up period, to promote health-related quality of life among pemphigus vulgaris patients, to assess the usability and acceptability of the system from the patients’ perspective.Methods and analysisThis is a multicentre, parallel-group, randomised controlled trial. 160 participants will be randomly assigned to either the intervention group (receiving the Intelligent Multimodal Symptom Assessment and Response System) or the control group (receiving standard care). Eligible participants will be adults aged 18 years or older with a confirmed diagnosis of pemphigus vulgaris and active skin or mucosal lesions, who are able to use digital devices and provide written informed consent. Individuals with severe comorbidities, concurrent participation in other clinical trials or cognitive impairments that may interfere with study adherence will be excluded. The intervention will be delivered via a digital platform, integrating electronic patient-reported outcomes, wearable physiological data and lesion images over a 12-week follow-up period.Ethics and disseminationEthical approval was obtained from the Institutional Review Board of the School of Nursing, Fudan University (IRB 2025-07-13), the Medical Ethics Committee of the Institute of Dermatology, Chinese Academy of Medical Sciences (2025-KY-034) and the Ethics Committee of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine (2025-452). Approvals were granted on 10 September, 18 July and 23 July 2025, respectively. This protocol is based on V1.0, 13 July 2025 of the protocol. The results of this study will be disseminated through peer-reviewed publications and academic conferences.Trial registration numberChiCTR2500109711.Keywords: Health informatics, China, Adult dermatology, Disease Management, Protocols & guidelines STRENGTHS AND LIMITATIONS OF THIS STUDY This is a multicentre randomised controlled trial to evaluate the efficacy and feasibility of an intelligent multimodal symptom assessment and response system in patients with pemphigus vulgaris. This is a multicentre randomised controlled trial to evaluate the efficacy and feasibility of an intelligent multimodal symptom assessment and response system in patients with pemphigus vulgaris. Stratified randomisation with computer-generated sequences and allocation concealment was applied, reducing selection bias and improving group comparability. Stratified randomisation with computer-generated sequences and allocation concealment was applied, reducing selection bias and improving group comparability. The use of multimodal data sources (patient-reported outcomes and objective physiological indicators) strengthens the robustness and comprehensiveness of outcome assessment. The use of multimodal data sources (patient-reported outcomes and objective physiological indicators) strengthens the robustness and comprehensiveness of outcome assessment. The intelligent algorithm supports individualised symptom alerts and response recommendations based on multimodal data fusion, promoting precision symptom management. The intelligent algorithm supports individualised symptom alerts and response recommendations based on multimodal data fusion, promoting precision symptom management. However, the requirement for digital literacy and internet access may lead to selection bias, potentially limiting the generalisability of the findings. However, the requirement for digital literacy and internet access may lead to selection bias, potentially limiting the generalisability of the findings. Introduction Pemphigus vulgaris (PV) is a rare autoimmune blistering disorder that primarily affects the skin and mucous membranes.1 The peak onset of PV typically occurs between 50 and 65 years, and it is characterised by flaccid blisters and painful mucosal erosions or ulcers.2 3 Current treatment goals aim to reduce new blister formation, promote lesion healing and maintain remission with the lowest dose of systemic therapy. Standard therapeutic modalities include systemic corticosteroids, immunosuppressants, biologics, intravenous immunoglobulin and specialised wound care.4,6 Although immunotherapy has significantly reduced PV-related mortality from 60% to 90% to approximately 10%,7 the relapse rate remains as high as 81%, with a median disease-free survival of about 14 months,8 and long-term use of corticosteroids and immunosuppressants effectively controls disease activity but increases the risk of infection, metabolic disturbances, osteoporosis and psychiatric complications.9 10 PV presents management challenges beyond pharmacological treatments, due to fluctuating symptoms, unpredictable relapses and psychosocial distress.11 12 Existing guidelines offer clinical recommendations for medication but fall short in providing comprehensive strategies for symptom management.4 6 Symptom monitoring, patient education, regular follow-up, pain control, wound care, diet management, infection prevention, medication management and psychosocial support are core components of symptom management for optimising long-term outcomes in PV.11,15 Currently, clinical follow-up for PV primarily relies on face-to-face visits and passive symptom reporting, with limited capacity for real-time, multimodal monitoring. This traditional care model hinders timely recognition of symptom deterioration and delays early intervention, resulting in gaps in continuous, individualised management—particularly in outpatient and home settings. Digital health technologies, such as patient-generated health data (PGHD), which are health-related data created, recorded or gathered by patients (or caregivers) outside clinical settings to support ongoing monitoring and care decisions.16 17 PGHD provide new opportunities for precision symptom management, by enabling real-time, multimodal monitoring and feedback, digital systems can strengthen home-based care and support continuity post-discharge. To address these gaps, our research team is developing the Intelligent Multimodal Symptom Assessment and Response System (IMSARS), a platform designed to deliver precise symptom-threshold alerts, strengthen treatment adherence and improve quality of life through high-sensitivity multimodal data collection and individualised response algorithms. This multicentre, parallel-group, randomised controlled trial is designed to evaluate the clinical efficacy and feasibility of the system, with the aim of generating robust evidence to support intelligent, data-driven nursing interventions for rare autoimmune skin diseases. Rationale Despite advances in immunotherapy, long-term management of pemphigus vulgaris remains constrained by fragmented follow-up and the lack of continuous, multimodal symptom monitoring in home-based settings. To address this gap, we developed the IMSARS, grounded in precision nursing and digital intelligence, to enable real-time symptom tracking and individualised, assessment-triggered responses during post-discharge and remote care. To isolate the incremental benefit attributable to the intelligent multimodal response algorithm, the control group will use the same WeChat-based platform with assessment-driven response functions disabled, thereby controlling for digital contact, reporting burden and platform exposure. This trial will therefore evaluate whether IMSARS improves symptom management and treatment adherence, and ultimately enhances quality of life, beyond routine follow-up delivered through an identical user interface. Objectives Primary objective To evaluate the effectiveness of the IMSARS in improving symptom alerting and symptom management outcomes among patients with PV. Secondary objectives To enhance patients’ self-management ability in symptom monitoring and control. To enhance patients’ self-management ability in symptom monitoring and control. To improve treatment adherence throughout the follow-up period. To improve treatment adherence throughout the follow-up period. To promote health-related quality of life among PV patients. To promote health-related quality of life among PV patients. To assess the usability and acceptability of the system from the patients’ perspective. To assess the usability and acceptability of the system from the patients’ perspective. Methods and analysis Study design Design This study is a multicentre, randomised, parallel-group, single-blind controlled trial designed to evaluate the efficacy and feasibility of the IMSARS in patients with PV, with exploratory analysis to further investigate potential outcomes and insights. The study is planned to be conducted between 1 June 2026 and 1 March 2027. Figure 1 presents the summary graphic illustration. The protocol follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2025 and the SPIRIT-Patient-Reported Outcome (PRO) Extension for patient-reported outcomes,16 17 and the trial results will be reported in accordance with the Consolidated Standards of Reporting Trials (CONSORT 2025) statement.18 Summary graphic illustration. ePRO, electronic patient-reported outcome; HR, heart rate; HRV, heart rate variability; IMSARS, Intelligent Multimodal Symptom Assessment and Response System; PROMIS, Patient-Reported Outcomes Measurement Information System; RCT, randomised controlled trial;QoL, quality of life. Blinding This is a single-blind trial. Participants will be blinded to group allocation because both groups use the same digital platform with similar user-facing interfaces; however, the response functions are enabled only in the intervention group. Study personnel responsible for intervention delivery and system management will be aware of allocation due to operational requirements. Data analysts will be blinded to allocation during the primary analysis where feasible by using coded group labels. Unblinding Unblinding of participants is not routinely planned. If unblinding is deemed necessary for participant safety (eg, serious adverse events requiring knowledge of intervention features), it will be authorised by the principal investigator and documented, including the reason, date and personnel involved. Eligibility criteria of participants Eligibility criteria Adults diagnosed with pemphigus vulgaris in China. Recruitment will occur among outpatients at dermatology departments of the participating tertiary hospitals. The study will be conducted at two tertiary dermatology centres in China: Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, and Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. Inclusion criteria: (1) Diagnosis of PV confirmed by clinical manifestations and at least one of the following: histopathology, direct immunofluorescence or ELISA positivity for anti-desmoglein 1/3 antibodies. (2) Age ≥18 years. (3) Presence of active skin and/or mucosal lesions. (4) Ability to use a smartphone or electronic device and willingness to comply with digital follow-up. (5) Provision of written informed consent. Exclusion criteria: (1) Concomitant severe dermatologic or systemic disorders that may interfere with symptom assessment. (2) Participation in another clinical trial within the past 3 months. (3) Severe psychiatric illness or cognitive impairment preventing adherence to the protocol. Withdrawal criteria: (1) Participants may be withdrawn from the study under any of the following circumstances: failure to complete self-reported data for seven consecutive days. (2) Occurrence of a serious protocol violation that compromises data integrity or participant safety. (3) Voluntary withdrawal on the participant’s request. All withdrawal cases and reasons will be documented in the study record, and data collected prior to withdrawal will be retained for intention-to-treat analyses unless the participant explicitly requests data removal. Sample size The primary endpoint is the mean number of symptom-threshold events—defined as any target symptom (pain, pruritus, fatigue, sleep disturbance, anxiety, depression or physical function) reaching a T-score ≥60 and/or persistent increase in new blister counts without improvement for more than 1 week. Assuming a minimal clinically important difference of 0.5 SD, a two-sided α=0.05 and power=80% (β=0.2), 64 participants per group are required. Considering a 20% attrition rate, 80 participants per group (total n=160) will be enrolled. The dropout assumption is based on prior experimental studies reporting ≈17% withdrawal and ≤3% loss to follow-up. Wearable correlation substudy: correlation analyses will be conducted to examine the associations between wearable-derived physiological and behavioural indicators, including heart rate (HR), heart rate variability (HRV), step count, and sleep metrics, and symptom severity or disease activity measures. Assuming a moderate correlation effect size (r=0.45–0.50), with α=0.05 and power=0.80, approximately 35–40 participants completing paired measurements are required. Participant recruitment Potentially eligible patients will be identified by qualified dermatologists from the Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, and Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, both of which are tertiary teaching hospitals collaborating with the research team. The clinicians will invite eligible patients to participate and provide them with detailed study information and the electronic or paper-based informed consent form through the WeChat-based recruitment platform or in person at the clinic. Before enrolment, study coordinators will contact each patient by telephone to confirm eligibility, explain the study procedures and answer any questions. Patients who agree to participate will provide electronic informed consent or paper version according to their preference and the recruitment setting. Reasons for ineligibility or refusal will be recorded by the coordinators and categorised as ‘patient refusal’ or ‘ineligibility’. To ensure sufficient recruitment and achieve the planned sample size, participants will receive educational materials on symptom self-management after enrolment. After providing informed consent, all participants will be informed that a study coordinator will contact them shortly for initial training. During this training, participants will receive on-site instruction from clinical nurses on the use of the IMSARS at the time of enrolment. This will be followed by online reinforcement training via WeChat, including illustrated step-by-step guidance and screenshots to facilitate system operation. Participants will also be informed that they can contact study coordinators at any time through WeChat or mobile phone for technical or procedural assistance. Each participant will be registered in the system and automatically assigned a unique study identification code by the backend server. All research data will be stored and analysed using coded identifiers to ensure anonymity. Patients who sign informed consent but fail to complete baseline data entry will be considered screening failures and excluded from the study population. Participants who complete baseline assessments but fail to report weekly PROs will remain enrolled, and the research team will attempt to re-engage them through automated reminders and direct coordinator support. All enrolment and training records will be securely stored in the encrypted study database. Assignment Eligible participants will be randomly assigned in a 1:1 ratio to either the intervention group, which will use the multimodal assessment and response system for real-time symptom monitoring and personalised management, or the control group, which will receive routine follow-up and standard nursing care. Randomisation will be stratified by study centre and baseline disease severity (mild, moderate or severe), based on predefined clinical criteria. Within each stratum, a permuted block randomisation approach with varying block sizes will be employed to ensure allocation balance and minimise predictability. The randomisation sequence will be generated automatically through the WeChat-based application backend using a Python randomisation module. This system will maintain allocation concealment, preventing both participants and data collectors from accessing the assignment sequence. Follow-up assessments will occur at baseline, week 2, week 4, week 8 and week 12 post-enrolment. The study design flowchart is shown in figure 2. Study flow chart. ePRO, electronic patient-reported outcome. Intervention Digital platform intervention IMSARS components. IMSARS comprises a mobile front-end and a secure web-based backend that integrates a classification engine and a knowledge-graph rule base (figure 3). Core functions include: (i) electronic patient-reported outcomes (ePROs) capture; (ii) lesion image upload and standardised skin/mucosal assessment; (iii) integration of PGHD from wearables (eg, HR, HRV, step count and sleep metrics), wearable devices are used as an exploratory and supplementary data collection component in a subset of participants. Wearable-derived physiological and behavioural indicators will be used for exploratory correlation analyses with symptom severity and disease activity measures but will not be directly incorporated into the automated decision-making framework or recommendation generation in the current intervention model; (iv) automated alerts with tailored self-management recommendations (eg, medication guidance and patient education); and (v) audit logging for monitoring engagement and adherence. Workflow for IMSARS in PV. Lay summary. IMSARS helps patients and clinicians spot symptom worsening earlier by combining symptom self-reports, lesion photos and wearable data to deliver timely, risk-matched self-care guidance and escalation prompts between clinic visits. ePRO, electronic patient-reported outcome; IMSARS, Intelligent Multimodal Symptom Assessment and Response System; PDAI, Pemphigus Disease Area Index; PV, pemphigus vulgaris. Clinically oriented workflow description. IMSARS implements a multimodal assessment–response workflow (figure 3). Patients submit ePROs and relevant contextual information (eg, medication use and disease severity), upload lesion images and when available, contribute wearable-derived physiological/behavioural data. These inputs are processed using predefined rules to stratify symptom risk and assign participants to three response zones: Zone 1 (high symptom risk), Zone 2 (moderate symptom risk) and Zone 3 (low symptom risk), which determine notification frequency and intensity. Symptom early-warning rules and decision framework (figure 4). To operationalise symptom monitoring and response generation, an exploratory symptom early-warning and decision framework was developed based on Patient-Reported Outcomes Measurement Information System (PROMIS) T-score interpretation standards, disease activity indicators, expert consultation and preliminary longitudinal observations from the pilot cohort. The framework integrated multidimensional patient-reported and disease-related data, including PROMIS symptom T-scores, quality-of-life measures and disease activity indicators (eg, Pemphigus Disease Area Index (PDAI) score and number of new blisters).19 PROMIS symptom thresholds served as the primary trigger for symptom warning activation, whereas quality-of-life impairment and disease activity indicators were incorporated as supportive contextual factors for risk stratification and individualised decision-making. Any target symptom domain (pain, pruritus, fatigue, sleep disturbance, anxiety, depression or impaired physical function) reaching a T-score ≥60 and/or persistence of newly developed blisters without reduction for more than 1 week was considered indicative of a clinically relevant symptom event. PROMIS symptom T-scores were categorised into three levels. T-scores <60 were considered within the normal-to-mild symptom range. A T-score ≥60 was defined as a moderate symptom warning threshold requiring closer monitoring or supportive intervention, whereas a T-score ≥70 was considered a severe or high-risk threshold requiring intensified attention and potential clinical management. In addition to cross-sectional severity thresholds, longitudinal symptom dynamics were incorporated into the decision framework. An increase of ≥5 points from baseline or from the previous assessment was regarded as clinically meaningful symptom worsening. Persistent abnormalities were defined as either: (1) two consecutive assessments with T-scores ≥60; or (2) continuous symptom escalation across sequential assessments. Primarily based on symptom severity, with quality-of-life impairment and disease activity indicators serving as supportive contextual factors, the system automatically stratified patients into three response zones (Zone 1–3), corresponding to red, gold and blue warning levels for healthcare providers. Zone-specific decision rules then generated tailored recommendations, including (i) urgency stratification with recommended care-seeking actions (eg, urgent consultation vs home observation); (ii) adaptive educational content; (iii) symptom-specific self-management guidance; and (iv) adjusted push frequency and follow-up intensity. Generally, Zone 1 corresponded to severe or progressive symptom deterioration (eg, T-score ≥70 or persistent symptom escalation), Zone 2 reflected moderate symptom burden requiring adaptive monitoring (eg, T-score ≥60), whereas Zone 3 represented relatively stable symptom status requiring routine follow-up. For example, a patient presenting with persistent blister progression for more than 1 week and a pain T-score increasing from 58 to 71 across sequential assessments would be automatically classified into Zone 1 (high-risk), triggering clinician alerts, intensified follow-up and recommendations for urgent medical evaluation. Symptom early-warning rules and decision framework. PDAI, Pemphigus Disease Area Index; PROMIS, Patient-Reported Outcomes Measurement Information System;QoL, quality of life. Brief practical interpretation of real-world benefits. In practical terms, IMSARS helps patients and clinicians ‘stay ahead’ of deterioration by combining what the patient feels and reports—symptoms, what the skin/mucosa looks like—lesion photos, and what the body is doing—wearable metrics. Instead of waiting for the next clinic visit, the system can identify symptom worsening earlier, deliver clear, risk-matched self-management guidance and—when needed—prompt timely escalation to clinical review. By adjusting message frequency and intensity across low-risk, moderate-risk and high-risk response zones, the platform aims to provide more support when risk is high while avoiding unnecessary burden when risk is low, thereby strengthening continuity of care after discharge and supporting adherence and day-to-day self-management. Interventions and treatment protocol Intervention group. Participants will receive the full IMSARS functionality described above (Section 3.3.1), including automated risk stratification, alerts, tailored recommendations and audit-logged engagement metrics. Control group. Participants will use the same mobile application; however, the assessment-driven response functions will be disabled, and no automated feedback, personalised recommendations or stratified responses will be provided following symptom or data assessment. Concomitant care. All participants will continue to receive usual medical care and pharmacological treatment as determined by their treating dermatologists. Concomitant medications, dose adjustments and treatment changes during follow-up are permitted and will be recorded, including the indication and timing. Use of additional non-study digital symptom-management tools will be discouraged where feasible and documented if used. Criteria for discontinuing or modifying the intervention Participants may discontinue use of the digital platform at any time. The research team may suspend system-based recommendations for an individual participant if system use is judged to increase psychological burden, if repeated technical failures prevent reliable monitoring or if continued participation is deemed unsafe by the treating clinician. Any discontinuation or modification will be documented, and outcome assessments will be sought as feasible to support intention-to-treat analyses. Participant burden and contingency plans To support sustained self-reporting and standardised outcome assessment, ePROs will be collected at prespecified timepoints (baseline and weeks 2, 4, 8 and 12). Each scheduled ePRO assessment is expected to take approximately 8–10 min. In addition to the scheduled assessments, participants will be encouraged to submit unscheduled ePRO reports at any time when symptoms change, to facilitate timely identification of deterioration between visits. Lesion photographs will be participant-initiated and symptom-driven. Participants will upload skin and/or mucosal images whenever they perceive meaningful changes in lesions or related symptoms; each photo submission is expected to take approximately 1–2 min. Wearable monitoring is optional and will be implemented in a subset of approximately 35–40 participants due to device availability constraints. Wearable participation is not required and will not affect eligibility or retention; primary and secondary outcomes do not rely on wearable data. Wearable-derived metrics will be analysed as exploratory outcomes in the wearable subset. Technical support will be available via telephone and WeChat during business days. To ensure continuity of intervention delivery and safety-relevant data capture, predefined contingency procedures will be applied for technical failures. If app access or data upload is interrupted (eg, network instability, device malfunction or synchronisation failure), participants will be instructed to contact the study team for remote troubleshooting. When in-app reporting remains unavailable, the study team will collect essential safety-relevant symptom information via telephone/secure messaging, document the event as a protocol deviation and continue follow-up per protocol. Missed scheduled ePRO assessments may be completed within the allowed window when feasible; otherwise, the assessment will be recorded as missing and handled according to the prespecified missing-data approach. Data collection, management, and monitoring Data sources and collection. Data will be collected via the platform and synchronised devices, including (i) baseline characteristics and clinical history recorded at enrolment and we will assess disease activity and vesicle severity in patients with pemphigus by using the internationally recognised Pemphigus Disease Area Index,19 which includes standardised assessments of lesion extent, blister activity and disease severity, (ii) scheduled patient-reported outcomes captured through the PROMIS-57 profile,20 (iii) image-based lesion documentation, (iv) wearable-derived physiological/behavioural metrics where available, and (v) system logs for adherence and engagement. Data capture and quality control. Electronic laboratory notebook was not used. Study data were captured via the study platform and exported to Excel for analysis. In the intervention group, symptom reports, lesion images and wearable-derived physiological/behavioural metrics (eg, HR, HRV, step count and sleep parameters) will be uploaded automatically or entered in-app according to the system workflow. In the control group, participants will complete scheduled ePRO assessments at prespecified follow-up timepoints. The backend will automatically timestamp entries and maintain audit trails. Electronic forms will include range and logic checks to minimise missing or implausible values. Intervention fidelity Fidelity will be ensured through standardised training of all study personnel, centralised oversight of intervention delivery and automated backend logging within the system to verify adherence to the predefined intervention protocol. Fidelity will be systematically assessed by examining system utilisation metrics, including frequency and completeness of symptom reporting, engagement with recommended content and overall patient interaction patterns recorded by the platform. Personnel and assessment time points Data will be collected by designated study personnel. The detailed timeline for intervention delivery and outcome assessments is presented in figure 5 (Timeline of Interventions and Assessments). Assessments will be performed at baseline and at weeks 2, 4, 8 and 12 (±3 days) after enrolment. Continuous data capture (wearable metrics and platform use logs) will occur throughout the 12-week follow-up period for participants with compatible devices. Timeline intervention and assessment notes. Assessments will be performed at baseline and at weeks 2, 4, 8 and 12 (±3 days) after enrolment. PRO, patient-reported outcome; PROMIS, Patient-Reported Outcomes Measurement Information System;CRA = Clinical Research Assistant. Trial monitoring and auditing Due to the low-risk nature of the digital intervention and the lack of investigational drugs/devices, no independent Data Monitoring Committee will be established. The study steering team will oversee recruitment, protocol adherence, data completeness, and safety using backend audit logs and periodic reviews. Any protocol deviations will be recorded and addressed promptly. No interim analysis is planned. The trial may be paused or terminated if safety concerns arise or as instructed by the ethics committee. Outcomes and measurements Outcomes are organised into five domains—clinical symptom control (primary), adherence and patient-reported outcomes (secondary), longer-term disease status and quality of life, healthcare utilisation, and system usability/safety. The assessment schedule, corresponding instruments and assessment windows are summarised in figure 5. Primary outcome. The primary endpoint is the number of symptom-threshold events within 12 weeks after discharge. A symptom-threshold event defined as the occurrence of any target symptom (pain, pruritus, fatigue or sleep disturbance) reaching a T-score ≥60, and/or worsening disease activity as assessed by PDAI, including progression in vesicle number or lesion severity meeting predefined clinical thresholds. PDAI-related assessments will be conducted by trained clinicians/researchers based on patient-uploaded lesion photographs and corresponding clinical information collected during follow-up. The outcome will be aggregated as the cumulative number of symptom-threshold events per participant during follow-up. Secondary outcomes. Secondary outcomes include composite scores of psychological and physical function interference (anxiety, depression, fatigue, sleep disturbance and physical functioning) and disease-specific quality of life. The analysis metrics will include the weekly completion rate of ePROs and composite scores for psychological and physical function interference. The method of aggregation will be the proportion of participants completing ePROs and the aggregation of composite scores. Self-reported symptom data collection will be conducted using the Chinese version of the PROMIS-57 profile.20 21 We selected the PROMIS-57 because of its strong psychometric properties, broad validation across chronic disease populations and its ability to comprehensively assess multidimensional symptom burden across physical, psychological and social domains.22 In addition, its standardised T-score metric facilitates longitudinal symptom monitoring, threshold-based risk stratification and integration into digital symptom management and early-warning frameworks. The PROMIS Profile-57 assesses symptoms experienced over the previous 7 days, making it particularly suitable for repeated longitudinal assessments and dynamic symptom monitoring during follow-up. Its multidomain structure and standardised scoring system also support individualised response generation within digital health interventions.23 24 The dermatology-specific quality-of-life instrument used in this study was the Chinese version of the Autoimmune Bullous Disease Quality of Life questionnaire, authorised for use by Professor Dédée Murrell25 and Professor Baoqi Yang.26 Assessments will be conducted at protocol-defined timepoints. Additional indicators. The indicators include ePRO reporting adherence, medication adherence, overall patient assessment, Physician Holistic Assessment Visual Analog Rating Scale (Dermatology), Physician Subjective Assessment of Disease Severity Score, Family Dermatology Life Quality Index and unplanned readmission rate. ePRO reporting adherence will be calculated as the weekly completion rate, defined as the proportion of participants who complete all expected ePRO assessments at each scheduled time point. Medication adherence will primarily refer to adherence to prescribed pharmacological treatment regimens. Healthcare resource utilisation. Healthcare resource indicators include the number of outpatient visits, emergency visits, unplanned readmissions, medication changes, and nurse or specialist telephone follow-ups, summarised as counts throughout follow-up. System performance and usability. System-related outcomes include user acceptance measured by the System Usability Scale and patient satisfaction assessed through a Likert-type questionnaire. The method of aggregation will involve calculating the mean scores. Assessments will be conducted at the end of the trial. Safety indicators. Safety outcomes include the incidence of adverse events (AE), both medication-related and system-related (eg, data collection failure, increased psychological burden or other unexpected issues during system use). Serious adverse events (SAE) will be reported to the ethics committee in accordance with institutional requirements within the required timeframe, and participants will receive appropriate clinical management. All AEs/SAEs will be documented with onset date, severity, relatedness, action taken and outcome. The analysis metrics will include the frequency and severity of AEs/SAEs. The method of aggregation will involve counting the occurrences. Data will be collected throughout the trial. Statistical considerations All statistical analyses will follow a prespecified analysis plan finalised prior to database lock. Primary analyses will use the intention-to-treat principle, with sensitivity analyses in the per-protocol population. Baseline characteristics, outcome variables and feasibility metrics will be summarised using descriptive statistics. Continuous variables will be expressed as mean±SD for normally distributed data or median (IQR) for non-normal data; categorical variables will be presented as frequencies and percentages. All tests will be two-sided with α=0.05. To limit inflation of type I error across multiple secondary outcomes, the Benjamini–Hochberg false discovery rate procedure will be applied to secondary endpoints. Analyses will be performed in International Business Machines (IBM) SPSS Statistics for Windows, Version 27.0 (IBM Corp., Armonk, NY, USA). International Business Machines Corp., Armonk, NY, USA). Primary outcome. Because event counts are expected to be overdispersed, the primary analysis will use negative binomial regression, adjusting for baseline symptom burden, disease duration, PV disease severity and study centre. If follow-up time varies, an offset for log (person-time) will be included. Results will be reported as adjusted rate ratios with 95% CIs. Subgroup analysis. Prespecified subgroup analyses by PV disease severity will be conducted by adding a treatment-by-severity interaction term to the primary model; severity-stratified estimates with 95% CIs will be presented where relevant. Secondary outcomes. Model selection will be based on outcome scale and repeated-measures structure. Continuous variables will be analysed using linear or generalised linear mixed-effects models for repeated continuous outcomes (such as PROMIS domain T-scores and disease-specific quality-of-life measures), accounting for within-participant correlation over follow-up and allowing inclusion of participants with incomplete follow-up under a missing-at-random assumption. Binary outcomes will be analysed using logistic regression. Count outcomes (eg, healthcare utilisation) will be analysed using Poisson regression when equidispersion is reasonable; otherwise, negative binomial models will be used. Effect estimates will be reported as adjusted β coefficients for continuous outcomes, ORs for binary outcomes and rate ratios for count outcomes, all with corresponding 95% CIs. Model fit and assumptions will be evaluated using residual diagnostics, checks for overdispersion in count models and collinearity diagnostics. Handling of missing data. Missing data will be handled according to the recommendations outlined in the SPIRIT-PRO Extension.27 Patterns and mechanisms of missingness will be examined usingLittle’s Missing Completely at Random test. When data are deemed to be missing at random, multiple imputation will be employed using a fully conditional specification method, and 20 imputed datasets will be generated for analysis. Sensitivity analyses, including complete-case and worst-case scenarios, will be performed to evaluate the robustness of the findings. For PRO measures, partial missingness such as skipped items will be managed following the scoring guidelines of each instrument, including the PROMIS-Profile 57 manual. The proportion of missing ePRO assessments will be reported at each time point, and differences between groups will be tested to identify any systematic bias. Patient and public involvement Patients with pemphigus vulgaris and their caregivers contributed to the early stages of the study by providing input on symptom management priorities and the usability of the digital platform, which informed refinement of the research questions and patient-reported outcome measures. Participants will remain actively involved throughout the trial through self-reporting of symptoms and outcomes, and study findings will be shared with them in plain language via the digital platform on completion. Ethics and dissemination Ethics approval The protocol was approved by the Institutional Review Board of the School of Nursing, Fudan University on 10 September 2025 (Approval No. IRB 2025-07-13). Furthermore, ethical approval was also obtained from the Medical Ethics Committee of the Institute of Dermatology, Chinese Academy of Medical Sciences, under acceptance number 2025-KY-034, approved on 18 July 2025, with an approved amendment (EC-AF-38.10) on 24 October 2025. Additional ethical approval was granted by the Ethics Committee of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine (Reference No. 2025-452), approved on 23 July 2025. Any amendments to the protocol will be submitted for further review and approval by the Institutional Review Board, and the updated version will be registered in the clinical trial registry. The results of this study will be disseminated through peer-reviewed publications and presentations at academic conferences. The study will be conducted in accordance with the Declaration of Helsinki, all procedures will be conducted in accordance with the principles of Good Clinical Practice and the General Data Protection Regulation. Written informed consent will be obtained from all participants prior to enrolment (see supplemental appendix 1). All study data will be securely stored, de-identified and accessible only to authorised research personnel. Any study-related adverse events will be reported to the ethics committee, and appropriate medical care will be provided as required. Data sharing statement De-identified individual participant data and the statistical code may be made available upon reasonable request after publication of the primary trial results, subject to approval by the steering team and applicable ethics and institutional policies. Requests should specify the research question and analysis plan, and data will be shared under a data use agreement to protect participant confidentiality. No identifiable data will be shared. Trial sponsor and funding This is an investigator-initiated study with no commercial sponsor. Institutional resources for platform maintenance, staff time and data management are provided by the participating institutions. The research team is responsible for the study design, conduct, data analysis, manuscript preparation and publication decisions. Additional consent No biological specimens will be collected for this study. Therefore, additional consent for specimen storage or ancillary genetic testing is not applicable. Ancillary care and compensation Participants will continue to receive standard clinical care throughout the study. If any study-related harm occurs, appropriate medical care will be provided in accordance with institutional policies. The study team will follow local regulations and institutional requirements regarding compensation for research-related injury. There is no planned financial incentive for participation beyond study-related educational materials and technical support. Discussion This study represents an important exploration of integrating an ePRO system into the long-term management of pemphigus vulgaris, a rare autoimmune blistering skin disease. Against the backdrop of widespread use of WeChat mini-programmes among patients in China, the development of a multimodal symptom assessment and stratified response system that integrates ePROs, wearable physiological data and wound images demonstrates strong feasibility and, to some extent, enhances patient engagement as well as the continuity and completeness of symptom data. Using symptom threshold events as the primary outcome enables a more direct reflection of symptom changes and intervention responsiveness, aligning well with a symptom-control–centred disease management objective. Nevertheless, several limitations should be acknowledged. First, the two-centre, single-blind study design may affect the generalisability and internal validity of the findings. Future studies could expand the number of participating centres and adopt multicentre designs to improve sample representativeness; additionally, incorporating more objective outcome measures or refining blinding strategies may help mitigate potential bias. Second, the study relies to some extent on participants’ basic digital literacy, which may pose participation barriers for patients who are unable or unfamiliar with using smartphones, mini-programmes or uploading wound images, thereby introducing selection bias. Although WeChat mini-programmes are highly accessible and widely used in China, future research and real-world implementation should explore alternative data-collection approaches or provide additional human support for individuals with limited digital skills to enhance inclusiveness and equity. Moreover, the single-blind design may still introduce observer and implementation bias. Although participants in both groups used the same digital platform following computer-based randomisation and were unaware of their group allocation due to differences in system functionality, researchers were aware of group assignments and intervention details. This awareness may have influenced intervention delivery and outcome assessment to some extent. While the single-blind design is operationally feasible and reduces participant-level expectation bias, future studies could further minimise bias by standardising intervention procedures, emphasising objective endpoints or incorporating third-party outcome assessment. Finally, although stratified randomisation was used to balance patients with different disease severity levels between the intervention and control groups, substantial heterogeneity remains in disease severity, disease course and individual responses, which may influence intervention effectiveness. Accordingly, subgroup analyses will be conducted to examine differential effects of the system across severity levels and disease stages. Future research may also explore the applicability of this multimodal symptom management system to other pemphigus subtypes or related autoimmune skin diseases to establish its broader clinical value. In summary, this study represents an important step towards integrating intelligent digital health technologies into the long-term management of pemphigus vulgaris. By combining electronic patient-reported outcomes, wearable physiological data and image-based lesion assessment within a multimodal symptom assessment and response system, this trial addresses key gaps in real-time symptom monitoring and home-based disease management for patients with PV. The system’s design, aligned with commonly used digital platforms (WeChat) in China, supports patient engagement and continuous symptom tracking, which are essential for managing a chronic, relapsing condition. Although limitations related to study design and digital literacy must be acknowledged, the findings of this trial are expected to provide clinically relevant evidence on the feasibility and effectiveness of multimodal, ePRO-driven symptom management strategies. If proven effective, this approach may offer a scalable and patient-centred model for symptom monitoring and supportive care in PV and potentially other chronic wound-related or autoimmune skin conditions, thereby contributing to improved symptom control and quality of life. Protocol access and trial status The protocol can be accessed at https://www.chictr.org.cn/showproj.html?proj=286980. The randomised controlled trial is currently in the active recruitment phase. Supplementary material Acknowledgements We would like to thank all those who participated in this study. Footnotes Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2026-116731). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details. Ethics approval: Ethical approval was obtained from the Institutional Review Board of the School of Nursing, Fudan University (IRB 2025–-07-13), the Medical Ethics Committee of the Institute of Dermatology, Chinese Academy of Medical Sciences (2025-KY-034), and the Ethics Committee of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine (2025–-452). Approvals were granted on 10 September, 18 July, and 23 July 2025, respectively. References Supplementary Materials