https://orcid.org/0009-0008-4214-456X
Figures
Abstract
Background
Family caregivers of children with autism spectrum disorder (ASD) often experience substantial and multidimensional caregiver burden, which may adversely affect their physical and psychological well-being, family functioning, and the child’s rehabilitation. Although caregiver-focused psychosocial interventions have shown promise, structured and theory-informed programs specifically designed to address caregiver burden in this population remain limited.
Methods
A multi-stage intervention development design was used, guided by an integrated framework of the ABC-X family stress model and stress and coping theory. First, best-available evidence was synthesized to identify relevant intervention strategies. Second, findings from prior quantitative and qualitative phases of a mixed-methods research program were integrated to define intervention targets and draft content. Third, the draft program was refined through a two-round asynchronous modified Delphi consultation to improve content relevance, clarity, and implementation planning.
Results
The finalized program comprised three core modules (stressors, cognitive appraisal, and coping), seven themes, and 22 components. It incorporated psychoeducation, mindfulness-based approaches, gratitude-based expressive writing, acceptance and commitment-informed strategies, self-compassion training, and problem-solving skills. Expert response and authority were high; however, Kendall’s W values were modest across the two rounds (0.23 and 0.25). These findings suggest that the Delphi process contributed to systematic refinement and expert appraisal of the program and reflected experts’ perceptions of the relevance, clarity, and appropriateness of its content; practical feasibility remains to be tested directly with caregivers.
Conclusions
This study developed a theory-driven and evidence-informed psychosocial intervention framework for caregiver burden among family caregivers of children with autism spectrum disorder. The program provides a structured basis for ongoing caregiver-informed feasibility testing and subsequent effectiveness evaluation; its acceptability, delivery fidelity, and later effectiveness require further empirical validation.
Citation: Tian X, Qian W, Zhang J, Li J, Chen B, Li X, et al. (2026) An evidence-informed psychosocial intervention program for caregiver burden among family caregivers of children with autism spectrum disorder: a multi-stage development and Delphi study. PLoS One 21(8): e0354823. https://doi.org/10.1371/journal.pone.0354823
Editor: Yilin Jiang, UTM Skudai: Universiti Teknologi Malaysia, MALAYSIA
Received: April 13, 2026; Accepted: July 13, 2026; Published: August 12, 2026
Copyright: © 2026 Tian et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The anonymized data underlying the Delphi consultation findings of the present intervention-development study are available in Zenodo at https://doi.org/10.5281/zenodo.20730110. The deposited files include the anonymized Round 1 and Round 2 Delphi rating datasets, a data dictionary, SPSS syntax, and summary analysis outputs. Expert identifiers, institutional information, contact details, and potentially identifiable free-text comments were removed or summarized to protect participant confidentiality. The earlier quantitative and qualitative phases informed the developmental rationale only and were not re-analysed in the present manuscript. The evidence synthesis is registered with the Shanghai Evidence-Based Nursing Center (ES20245306).
Funding: The study was supported by the Shenzhen Longgang District Science and Technology Innovation Special Fund—Medical and Health Technology Project (Grant No. LGWJ2023-098). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Autism spectrum disorder (ASD) is widely recognized as a major public health concern because of its long-term impact on affected individuals, families, and health and social care systems [1,2]. The World Health Organization estimated that approximately 1 in 127 people globally had autism in 2021, while emphasizing substantial variation across studies and settings [3]. Epidemiological data from multiple regions also indicate marked increases in identified prevalence over recent decades, resulting in a growing population of children who require sustained care and support across developmental stages [4–6]. In China, limited diagnostic resources, regional disparities in service accessibility, and under-identification may further contribute to an underestimation of the true prevalence of ASD, while shifting a substantial and often unrecognized caregiving burden onto families [6].
ASD is a lifelong neurodevelopmental condition characterized by persistent difficulties in social communication and restricted, repetitive patterns of behavior [7,8]. Because children with ASD often require long-term and coordinated support, family caregivers frequently assume primary responsibility for managing daily care, coordinating therapeutic services, and making ongoing care-related decisions [9]. These responsibilities place family caregivers at the center of care delivery and expose them to sustained psychosocial strain.
Family caregivers of children with ASD consistently report higher levels of caregiver burden than caregivers of children with other chronic conditions or developmental disabilities [10–12]. In this population, caregiver burden is commonly understood as a multidimensional construct involving physical exhaustion, emotional distress, social restriction, time dependency, and limitations in personal development [9]. Elevated caregiver burden has been associated with anxiety, depressive symptoms, and poorer quality of life, which may in turn undermine caregivers’ capacity to sustain long-term caregiving and to support children’s rehabilitation and developmental needs [13–15].
Despite increasing recognition of caregiver burden as a critical issue in ASD care, current interventions remain limited in their capacity to address caregivers’ complex and interrelated needs in a coherent manner. Recent reviews have shown that interventions for families of children with ASD are often fragmented and tend to focus on isolated psychological outcomes or short-term symptom relief, rather than addressing the broader psychosocial determinants of caregiver burden through a structured program [16,17]. This fragmentation may limit their integration into routine services and reduce their sustainability and scalability.
Existing interventions for families of children with ASD are commonly categorized as caregiver-mediated or caregiver-focused. Caregiver-mediated interventions primarily seek to improve child outcomes by training caregivers to deliver therapeutic strategies [16]. Although these approaches may benefit child functioning, they may also intensify caregiving demands and increase caregiver stress and burden [14,17]. In contrast, caregiver-focused interventions aim to support caregivers’ own psychological well-being, coping capacity, and adjustment, and appear more directly relevant to reducing caregiver burden [16,17]. However, substantial heterogeneity remains in intervention content, delivery formats, and outcome measures, highlighting the need for a more integrated and theory-informed program.
To guide intervention development, the present study integrated the ABC-X family stress model and Lazarus and Folkman’s stress and coping theory [18,19]. Within this framework, caregiver burden can be understood as the product of interactions among family caregiving stressors, caregivers’ appraisal of those stressors and available resources, and their coping responses. This scope differs from the job demand-control-support model, which was developed to explain strain arising from paid work demands, decision latitude, and workplace support [20]. Although occupational models are valuable for employed populations, the integrated family-stress and coping framework was selected because it more directly represents unpaid family caregiving, illness-related meaning-making, family resources, stigma, and coping processes. It therefore provides a closer conceptual fit for a caregiver-focused intervention targeting multiple interrelated domains rather than isolated symptoms (Fig 1).
Accordingly, this study aimed to develop and refine an evidence-informed psychosocial intervention program for family caregivers of children with ASD. Using a multi-stage intervention development approach, the study integrated findings from a best evidence synthesis, prior quantitative and qualitative phases of a mixed-methods research program, and Delphi expert consultation to construct a theoretically grounded program for subsequent feasibility and effectiveness testing.
Methods
Study design
This study employed a multi-stage, evidence-informed design to develop a psychosocial intervention program targeting caregiver burden among family caregivers of children with ASD. The development process comprised three sequential phases: (1) best evidence synthesis; (2) program conceptualization through integration of theoretical frameworks and empirical findings from prior quantitative and qualitative phases of a mixed-methods research program; and (3) a two-round asynchronous modified Delphi consultation to refine the program content and implementation plan. Reporting of the consensus component was guided by the ACCORD (ACcurate COnsensus Reporting Document) guideline [21].
Phase 1. Best evidence synthesis
Formulation of the research question.
A best evidence synthesis was conducted to inform the development of the intervention program. The review question was formulated using the PIPOST framework, which specified the following elements: population (family caregivers of children with ASD); intervention (strategies targeting caregiver burden and its related intervention targets); professionals (clinical healthcare providers); outcomes (alleviation of caregiver burden); settings (hospital or home); and types of evidence (clinical practice guidelines, evidence summaries, systematic reviews, and expert consensus documents). The best evidence synthesis was registered with the Shanghai Evidence-Based Nursing Center (Registration No. ES20245306).
Literature search.
Guided by the 6S evidence hierarchy, literature was searched from the top down across evidence-based resources, guideline databases, evidence summary databases, systematic review databases, and comprehensive bibliographic databases. The search period covered database inception to September 15, 2024. Search terms were developed in both English and Chinese. Boolean operators were used, and both subject headings and free-text terms were applied. Detailed search strategies are presented in S1 File.
Study selection.
Eligibility criteria were developed according to the PIPOST framework. The inclusion criteria were as follows: (1) family caregivers of children with ASD aged ≥18 years; (2) interventions targeting caregiver burden or related intervention targets; (3) outcomes including at least caregiver burden or parenting/parental stress; (4) hospital or home settings; (5) evidence types including clinical practice guidelines, evidence summaries, systematic reviews, and expert consensus documents; and (6) publications in English or Chinese. The exclusion criteria were as follows: (1) full text unavailable; (2) interventions or outcomes not extractable; and (3) duplicate publications. Retrieved records were imported into EndNote for deduplication using both software-assisted and manual methods. Two reviewers independently screened titles and abstracts in two consecutive rounds, followed by full-text screening of retained records. Disagreements were resolved through discussion within the research team.
Quality appraisal.
Two reviewers with relevant clinical experience and formal training in evidence-based nursing independently appraised the methodological quality of the included literature. Disagreements were resolved through discussion within the research team. Appraisal tools were selected according to the type of evidence. For evidence summaries, methodological quality was assessed by tracing the original references. The Appraisal of Guidelines for Research and Evaluation II (AGREE II) instrument was used for clinical practice guidelines. The Joanna Briggs Institute (JBI) critical appraisal tools (2016 version) were used for systematic reviews and expert consensus documents.
Evidence extraction, synthesis, and grading.
Two reviewers independently extracted, synthesized, and graded the evidence. Disagreements were resolved through discussion within the research team. For evidence with consistent content, concise evidence statements were extracted. For complementary evidence, findings were integrated according to logical consistency. For conflicting evidence, priority was given to evidence-based sources, higher-quality evidence, and more recent evidence. According to the research design types of the original studies underpinning the included evidence, evidence was graded from Level 1 to Level 5 using the JBI 2014 pre-grading system, with Level 1 representing the highest level and Level 5 the lowest.
Phase 2. Program conceptualization and draft formulation
The preliminary intervention program was formulated through an iterative process integrating the theoretical framework, empirical inputs from the prior quantitative and qualitative phases, and recommendations from the best evidence synthesis. No primary data from the earlier phases were re-analysed in the present study. Their relevance was assessed according to population and setting correspondence, temporal proximity, conceptual alignment with the intervention framework, and convergence with the evidence synthesis. Methodological confidence was judged from the design-specific safeguards used in each phase, including standardized multicentre procedures and validated measures in the quantitative study and maximum-variation sampling, reflexivity, team-based analysis, participant feedback, and an audit trail in the qualitative study. Because these phases were components of the same research program rather than external evidence studies selected for the review, they were not assigned separate external quality-appraisal scores.
Theoretical foundation of the intervention program.
A theoretical framework for understanding caregiver burden among family caregivers of children with ASD was established by integrating the ABC-X family stress model with stress and coping theory. This framework informed three core modules: stressors, cognitive appraisal, and coping. Cognitive appraisal was operationalized as two related evaluative processes: appraisal of stressors and appraisal of available coping resources. Appraisal of stressors included illness cognition and affiliate stigma, whereas appraisal of available coping resources included psychological capital and perceived social support. Both processes were placed within the Cognitive Appraisal module because they represent evaluative judgments that precede and shape coping responses. Coping referred to the behavioral and cognitive strategies adopted in response to caregiving demands. Accordingly, psychological capital, perceived social support, illness cognition, affiliate stigma, and coping strategies were identified as key intervention targets (Fig 1).
Contribution of quantitative findings to intervention development.
The prior quantitative phase was a multicentre cross-sectional survey conducted from January to April 2024 in four rehabilitation centres in China. Using convenience sampling, 320 family caregivers completed the survey and 307 valid responses were analysed. Validated measures, standardized site procedures, confirmatory factor analysis, multiple regression, structural equation modelling, and bias-corrected bootstrap analyses were used to examine caregiver burden and its psychosocial pathways. This phase is available as a preprint and has not been treated as peer-reviewed evidence in the present study [22]. Psychological capital, perceived social support, illness cognition, affiliate stigma, and coping tendency showed direct and/or indirect associations with caregiver burden and were therefore selected as intervention targets. Caregivers of an only child with ASD, female caregivers, unemployed caregivers, and caregivers of children with greater ASD severity were identified as priority populations for later intervention targeting and recruitment.
Contribution of qualitative findings to intervention development.
The prior qualitative phase used descriptive phenomenology and purposive maximum-variation sampling. Sixteen family caregivers from the same four rehabilitation centres participated in face-to-face semi-structured interviews between May and August 2024. Data were analysed using Colaizzi’s method, with reflexive journaling, team-based analysis, participant feedback, and an audit trail used to support trustworthiness; reporting followed COREQ. Caregivers described the causes and influencing factors of burden, different appraisal and coping patterns, expectations regarding illness prognosis, and preferences concerning intervention content, format, timing, and personnel. These findings informed the draft program and represented indirect caregiver contribution to its development. However, caregivers did not directly review, co-design, or validate the finalized program after Delphi refinement. The integrated program outline based on the theoretical framework and findings from the quantitative and qualitative phases is presented in Table 1.
Evidence-informed intervention strategies.
The best evidence synthesis provided specific intervention strategies and content to support program development. The evidence recommended the use of two or more core intervention components to address caregiver burden [23,24]. Based on the synthesized evidence, mindfulness-based interventions were aligned with psychological capital [23,25]; gratitude-based interventions combined with expressive writing were aligned with perceived social support [24]; acceptance and commitment therapy was aligned with illness cognition [16,26]; self-compassion practices were aligned with affiliate stigma [27]; and problem-solving skills training was aligned with coping strategies [16]. These themes were therefore matched with corresponding intervention strategies in the preliminary program outline, and the detailed intervention content is presented in Table 2. In addition, evidence supporting the provision of accurate and reliable information about ASD led to the inclusion of “basic knowledge of ASD” as an additional intervention theme [25]. The evidence synthesis also informed decisions regarding intervention personnel, delivery format, and intervention setting [16,28].
Integration and structuring of the intervention program.
Integration of the theoretical framework, quantitative and qualitative findings, and evidence-informed intervention strategies resulted in a preliminary program outline comprising three core modules, eight intervention themes, and 26 specific components. Based on this outline, a preliminary draft of the intervention program was developed and subsequently submitted for refinement through Delphi expert consultation.
Phase 3. Delphi expert consultation
Expert selection.
A purposive sampling strategy was employed to recruit experts from relevant disciplines, including autism treatment, psychological therapy, psychiatric nursing, clinical nursing, and nursing education. The inclusion criteria were as follows: (1) a bachelor’s degree or above; (2) an associate senior professional title or above, or an equivalent senior academic or clinical rank; (3) at least 10 years of professional or research experience in the relevant field; and (4) willingness to participate in the study. A total of 16 experts were included in the Delphi consultation. This panel size was consistent with methodological recommendations that Delphi studies generally include 15–30 experts [29].
Delphi procedure.
After obtaining the experts’ agreement to participate, a two-round asynchronous modified Delphi consultation was conducted. Experts were recruited between 01/01/2025 and 28/02/2025. Structured questionnaires were distributed electronically by email or WeChat. The first round was administered between 01/01/2025 and 31/01/2025, and the second between 15/02/2025 and 28/02/2025. After round 1, quantitative ratings and anonymized written comments were summarized. The research team revised the program through documented discussion and used the revised program and questionnaire in round 2. The process ended after two rounds because no further structural changes were indicated, all retained items met the predefined item-level criteria, and an additional round was not expected to resolve the remaining multidisciplinary variation [30,31].
Data analysis.
Quantitative data were analyzed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Delphi indicators included expert response rate, expert authority coefficient (Cr), mean item importance score, coefficient of variation (CV), and Kendall’s coefficient of concordance (W). The authority coefficient was calculated as the mean of the judgment basis coefficient (Ca) and familiarity coefficient (Cs). Items were retained, modified, merged, or deleted using predefined thresholds for importance and variation together with written comments. Divergent recommendations were reconciled by the research team through comparison with the theoretical framework, prior empirical findings, best evidence synthesis, clinical feasibility, and internal coherence. Decisions and reasons were documented between rounds; no item was retained or removed solely on the basis of Kendall’s W. Detailed thresholds and procedures are provided in S2 File.
Ethical considerations.
This study was approved by the Ethics Committee of Shenzhen Hospital, Beijing University of Chinese Medicine (Approval No. KY-2024–368). Before participation, all experts were fully informed about the study aims and procedures, and written informed consent was obtained from all experts prior to participation. All expert information was kept confidential and was not disclosed publicly without authorization.
Use of artificial intelligence tools.
During manuscript revision, OpenAI ChatGPT was used to support English-language editing, consistency checking, and revision-related wording refinement. No AI tool was used to generate or analyse primary data, conduct statistical analyses, create figures or tables, select references, or make scientific interpretations. All AI-assisted text was critically reviewed and revised by the authors, who take full responsibility for the final content of the manuscript.
Results
Phase 1. Best evidence synthesis
Evidence identification and selection.
The literature search yielded 1,103 records. After removal of 380 duplicates, 723 records remained for screening. Of these, 704 were excluded based on title and abstract screening, and six additional records were excluded after full-text review. Ultimately, one guideline and 12 systematic reviews were included in the final synthesis (Fig 2) [16,17,23–28,32–36].
Characteristics of the included evidence.
All included documents were published in English and comprised one guideline and 12 systematic reviews. Their main characteristics are presented in S3 File.
Methodological quality of the included evidence.
The included guideline [36] was rated as high quality using AGREE II. The standardized domain scores were 97.0% for Scope and Purpose, 88.9% for Stakeholder Involvement, 94.4% for Rigor of Development, 87.5% for Clarity of Presentation, 75.0% for Applicability, and 100% for Editorial Independence. Inter-rater agreement was 0.747. Although the guideline did not report explicit recommendation grades, it was still considered suitable for informing practice.
All 12 systematic reviews [16,17,23–28,32–35] met the JBI methodological appraisal criteria and were retained for synthesis. Detailed appraisal results are provided in S3 File.
Evidence synthesis and thematic categorization.
A total of 64 evidence statements were extracted from the 13 included documents. Through inductive synthesis, these statements were grouped into 12 thematic domains, including caregiver rights, macro-level policy support, multidisciplinary team establishment, intervention formats and settings, and specific intervention approaches. After integration and refinement, 31 best evidence statements were generated and are presented in S3 File.
Overview of key evidence themes.
The synthesized evidence highlighted several domains relevant to intervention development. Evidence on caregiver rights emphasized the importance of ensuring caregivers’ access to ASD-related information, support resources, and psychological services, as highlighted in the NICE guideline [36]. Evidence at the policy level underscored the importance of incorporating both children with ASD and their family caregivers into formal mental health support systems to promote family well-being [25]. Evidence on service organization supported multidisciplinary collaboration among healthcare, psychological, and educational professionals to provide comprehensive support for caregivers [16,28].
Regarding intervention delivery, the evidence suggested that group-based formats and implementation in rehabilitation settings may improve feasibility and caregiver engagement [16,28]. Caregiver-focused interventions, which directly address caregivers’ psychological and emotional needs, showed more consistent benefits for caregiver well-being than caregiver-mediated, child-centered approaches [23,28]. Evidence on specific intervention approaches supported the use of mindfulness-based interventions, Acceptance and Commitment Therapy, psychoeducation, problem-solving skills training, expressive writing, and multicomponent programs. Collectively, these approaches were associated with reduced caregiver burden and improved psychosocial outcomes, providing an evidence-informed basis for the preliminary development of the psychosocial intervention program in the present study [16,23–25].
Phase 2. Preliminary intervention program
The preliminary psychosocial intervention program comprised three core modules, eight intervention themes, and 26 intervention components. The draft program was developed by integrating theory-derived domains with context-specific targets identified from the prior quantitative and qualitative phases and by mapping these targets to evidence-informed intervention strategies. Where multiple sources converged, priority was given to strategies that were both theoretically aligned and practically deliverable within a structured caregiver-focused program.
Phase 3. Delphi expert consultation
Characteristics of Delphi experts.
A total of 16 experts participated in the Delphi consultation. Their mean age was 47.50 ± 4.89 years, and their mean length of professional experience was 23.88 ± 6.11 years. Of the 16 experts, three held bachelor’s degrees, seven held master’s degrees, and six held doctoral degrees. All experts held senior professional titles. Their areas of expertise included autism intervention, psychological therapy, psychiatric nursing, clinical nursing, and nursing education. Detailed characteristics of the experts are presented in Table 3.
Expert engagement and authority.
Two rounds of Delphi consultation were completed. In each round, 16 questionnaires were distributed by WeChat or email, and all were returned with valid responses, yielding a response rate of 100%.
Expert authority was assessed using the authority coefficient (Cr). The judgment basis coefficient (Ca) was 0.90 and the familiarity coefficient (Cs) was 0.76, resulting in a Cr of 0.83, which indicated a high level of expert authority. Detailed results on expert engagement and authority are shown in Table 4.
First-round Delphi results and program revisions.
In the first round, all items met the predefined item-level retention thresholds, with mean importance scores of at least 3.50 and CVs of no more than 0.25. Kendall’s coefficient of concordance was 0.23 (χ² = 132.04, P < 0.001), indicating statistically significant but modest agreement across the multidisciplinary panel. These results supported continued refinement of the item set but were not interpreted as strong consensus.
At the first level, all three core modules including stressors, cognitive appraisal, and coping met the predefined retention criteria and were retained. At the second level, all eight intervention themes were retained. At the third level, all 26 intervention components were also retained based on the quantitative ratings. Detailed first-round ratings are presented in Table 5.
In addition to the quantitative ratings, experts provided qualitative suggestions to improve the clarity, feasibility, and implementation of the program. Regarding intervention delivery, experts suggested combining online and face-to-face formats, providing recorded sessions for caregivers unable to attend in person, limiting group size to facilitate interaction, and adjusting the frequency of check-in activities to accommodate caregivers’ time constraints. Regarding program structure, experts recommended reviewing key content from the previous session at the beginning of each session, specifying the intervention provider for each component, and distinguishing between whole-group and small-group discussion formats. Regarding evaluation and outcome assessment, experts suggested aligning intervention objectives with corresponding evaluation indicators, clarifying evaluation criteria, and specifying assessment time points. In addition, several experts recommended clarifying whether post-intervention support would be provided through the online communication group and developing contingency plans for unexpected situations or participant non-adherence during implementation. All qualitative comments were reviewed and discussed by the research team and informed subsequent program refinement.
Based on the first-round results and research team discussions, the preliminary intervention program was revised to improve conceptual clarity, feasibility, and internal coherence while preserving all core intervention targets. Revisions mainly involved merging conceptually overlapping items, refining terminology and session focus, and streamlining components with substantial content overlap. For example, content related to ASD etiology, clinical manifestations, and treatment was integrated into one component on basic ASD-related knowledge within the stressor module. Introductory content related to mindfulness, Acceptance and Commitment Therapy, and self-compassion was reframed as intervention introduction content rather than theoretical overview content. In addition, several components were merged or removed when expert feedback indicated redundancy or limited added value within the available session time. For example, mindfulness meditation was integrated with mindfulness breathing exercises, and advanced acceptance and cognitive defusion content was merged to streamline program delivery. These revisions were made on the basis of feasibility considerations and expert recommendations rather than statistical exclusion, and no core intervention target was removed. After revision, the program comprised three core modules, seven intervention themes, and 22 intervention components.
Second-round Delphi results and program finalization.
In the second round, all retained items continued to meet the predefined item-level criteria, with mean importance scores ≥3.50 and CVs ≤ 0.25. Kendall’s W increased only marginally from 0.23 to 0.25 and remained modest. The statistically significant coefficient therefore indicated non-random concordance, but not strong agreement among experts.
All three core modules, seven themes, and 22 components met the predefined item-level criteria and were retained after review of their ratings and written comments. Retention reflects satisfaction of the prespecified decision rules, not strong unanimity across the panel (Table 6).
In addition to quantitative ratings, experts provided qualitative feedback during the second round of consultation, primarily focusing on further refinement of intervention delivery, clarity, and risk management. Suggestions related to intervention format included replacing the term “self-introduction” with “ice-breaking activity,” clearly distinguishing between whole-group discussion and small-group discussion, adjusting session duration, and specifying acceptable formats for homework submission (e.g., video, photographs, written text, or audio recordings). Experts also recommended refining the names of intervention modules and themes and providing more detailed descriptions of video-based materials used during the intervention to enhance clarity and consistency. With respect to evaluation and safety considerations, experts suggested elaborating on intervention evaluation methods and developing an emergency response plan to address potential situations of marked emotional distress among participants. All qualitative suggestions were reviewed and discussed by the research team and informed further refinement of the intervention program.
Based on the second-round ratings, written feedback, and subsequent research-team discussion, the overall structure was retained while content-level refinements were made. Intervention objectives were sharpened; selected exercises and homework activities were specified; and supporting materials, including basic ASD information, structured writing templates, and interview or dialogue guides, were added to support standardized delivery. After two rounds, the finalized program comprised three core modules, seven themes, and 22 components. A concise summary of the finalized program is provided in Table 7 to enhance transparency and replicability, whereas detailed session objectives, activities, and supporting materials are presented in S4 File.
Implementation planning incorporated several safeguards to reduce burden and support fidelity during subsequent testing. The eight weekly sessions were retained because the content spans distinct informational, appraisal, and coping targets and because the evidence synthesis supported multicomponent programs delivered over approximately 5–8 weeks. Planned burden-reduction measures include hybrid attendance, recorded review materials, small groups, scheduling alongside the child’s rehabilitation where possible, brief breaks, flexible homework formats, and follow-up after missed sessions. Planned fidelity procedures include a standardized manual, facilitator training and rehearsal, session checklists, attendance and homework records, documentation of adaptations and deviations, and regular supervision. Facilitators will monitor fatigue and distress during each session; activities will be paused or shortened when needed, and marked distress will trigger immediate support and referral by a qualified mental-health professional. These procedures are implementation plans, not evidence that the program is already feasible.
Discussion
Family caregivers of children with ASD frequently experience substantial and multidimensional caregiver burden, which may adversely affect caregivers’ well-being, family functioning, and the child’s rehabilitation and developmental support [9,10]. Although caregiver burden has been widely recognized, existing interventions for this population have often focused on isolated psychological outcomes, such as stress, anxiety, or depressive symptoms, or have been embedded within caregiver-mediated programs primarily targeting child outcomes [16,17,23]. Against this background, the present study aimed to develop a structured psychosocial intervention program specifically designed to target caregiver burden among family caregivers of children with ASD.
A major strength of the present study lies in its theory-driven and evidence-informed development process. The integrated ABC-X family stress model and stress and coping theory organized the intervention around family caregiving stressors, appraisal, resources, and coping. The prior quantitative phase identified potentially modifiable psychosocial targets through a multicentre survey and pathway analyses, while the qualitative phase provided caregiver-derived accounts of lived burden, prognosis expectations, appraisal and coping patterns, and preferences for intervention delivery. These empirical inputs were considered most relevant where they converged with the theoretical framework and best evidence synthesis. Nevertheless, the cross-sectional design of the quantitative phase limits causal inference, and the qualitative findings reflect a small, service-engaged sample; the earlier phases therefore informed target selection rather than independently proving that the selected components will be effective.
The final program addresses the heterogeneity and fragmentation described in previous reviews by organizing multiple caregiver-focused strategies within one explicit framework [23,24,26]. It integrates psychoeducation, mindfulness-based approaches, gratitude-based expressive writing, Acceptance and Commitment Therapy-informed strategies, self-compassion training, and problem-solving skills. Reviews suggest that caregiver-focused and multicomponent approaches may improve psychosocial outcomes, but heterogeneity in content, delivery, and outcomes precludes assuming that the present combination is superior before direct testing [23,24]. The program should therefore be viewed as a structured synthesis of plausible components rather than an established effective package.
The systematic evidence synthesis, transparent item-level decision criteria, high expert participation, and multidisciplinary review enabled systematic development, expert appraisal, and refinement of the program. However, Kendall’s W remained modest in both rounds. The ratings reflect experts’ perceptions of the relevance, clarity, and appropriateness of the program content; they do not constitute evidence of formal content validity, strong consensus, practical feasibility, or effectiveness.
The program provides a structured framework that nurses and other professionals may use in later feasibility testing of caregiver-focused support. Its planned hybrid delivery, small-group activities, standardized materials, and multidisciplinary facilitation were selected to reduce access barriers and support consistent implementation. Nevertheless, eight sessions of 90–120 minutes may be demanding for caregivers who already face substantial time and emotional constraints. Flexible attendance, breaks, recorded materials, follow-up after missed sessions, fatigue monitoring, and distress-management procedures are intended to mitigate this burden. Facilitator training, manuals, session checklists, supervision, and documentation of adaptations are planned to support fidelity. Whether these measures are acceptable and sufficient must be assessed directly with caregivers using recruitment, attendance, retention, completion, adverse-event, fidelity, and qualitative acceptability indicators.
This study has several limitations. First, the program has not yet completed empirical feasibility or effectiveness evaluation. Pilot testing has begun and 12 family caregivers had enrolled at the time of manuscript preparation, but no preliminary process, feasibility, or outcome data are included in the present report. Second, caregivers contributed indirectly through the prior qualitative phase but did not directly co-design or validate the final post-Delphi program. Focus groups or cognitive interviews with caregivers are therefore needed to assess acceptability, cultural fit, burden, emotional safety, and practical feasibility before a definitive trial. Third, the consensus exercise was a two-round asynchronous modified Delphi rather than a synchronous deliberative process. Delphi designs vary and no universal number of rounds is required [37], but the format limited real-time clarification and may have contributed to the modest W values. Fourth, the panel was small and recruited within one national context, and the evidence synthesis included one guideline and 12 systematic reviews with heterogeneous underlying interventions. Finally, the eight-session multicomponent program may create participation burden and implementation complexity; the adequacy of the planned flexibility and fidelity safeguards remains to be established.
Conclusion
Guided by an integrated theoretical framework and informed by prior quantitative and qualitative findings, best evidence synthesis, and a modified Delphi consultation, this study developed a structured psychosocial intervention framework for caregiver burden among family caregivers of children with ASD. The final program comprised three core modules, seven themes, and 22 components. The Delphi process enabled systematic refinement and expert appraisal of the program but did not establish formal content validity, feasibility, or effectiveness. The framework is ready for direct caregiver-informed feasibility testing, including assessment of acceptability, delivery fidelity, participant burden, cultural fit, and emotional safety, before subsequent effectiveness evaluation.
Supporting information
S2 File. Basis for calculating expert authority.
https://doi.org/10.1371/journal.pone.0354823.s002
(DOCX)
S4 File. Finalized psychosocial intervention program for caregiver burden among family caregivers of children with ASD.
https://doi.org/10.1371/journal.pone.0354823.s004
(DOCX)
Acknowledgments
We thank the experts who participated in the Delphi consultation for their detailed ratings and written recommendations, which helped refine the program content and implementation plan. We also acknowledge the family caregivers whose participation in the prior qualitative phase informed the intervention targets and delivery considerations.
References
- 1. Zeidan J, Fombonne E, Scorah J, Ibrahim A, Durkin MS, Saxena S, et al. Global prevalence of autism: a systematic review update. Autism Res. 2022;15(5):778–90. pmid:35238171
- 2. Zhou H, Xu X, Yan W, Zou X, Wu L, Luo X, et al. Prevalence of autism spectrum disorder in China: a nationwide multi-center population-based study among children aged 6 to 12 years. Neurosci Bull. 2020;36(9):961–71. pmid:32607739
- 3.
World Health Organization. Autism. Geneva: World Health Organization; 2025. https://www.who.int/news-room/fact-sheets/detail/autism-spectrum-disorders
- 4. Maenner MJ, Shaw KA, Bakian AV, Bilder DA, Durkin MS, Esler A, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years — autism and developmental disabilities monitoring network, 11 sites, United States, 2020. MMWR Surveill Summ. 2023;72(2):1–14.
- 5. Delobel-Ayoub M, Saemundsen E, Gissler M, Ego A, Moilanen I, Ebeling H, et al. Prevalence of autism spectrum disorder in 7-9-year-old children in denmark, finland, france and iceland: a population-based registries approach within the ASDEU project. J Autism Dev Disord. 2020;50(3):949–59. pmid:31813107
- 6. Luo Z, Qi Q, Wang K, Zhou J, Chen S, Wang L. Current status of and challenges posed by autism spectrum disorders in China: prevalence, legal issues, and public awareness. Biosci Trends. 2022;16(2):167–71. pmid:35283396
- 7.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed., text rev. ed. Washington, DC: American Psychiatric Association; 2022.
- 8. Lord C, Elsabbagh M, Baird G, Veenstra-Vanderweele J. Autism spectrum disorder. Lancet. 2018;392(10146):508–20. pmid:30078460
- 9. Dückert S, Gewohn P, König H, Schöttle D, Konnopka A, Rahlff P, et al. Multidimensional burden on family caregivers of adults with autism spectrum disorder: a scoping review. Rev J Autism Dev Disord. 2023;12(4):723–40.
- 10. Picardi A, Gigantesco A, Tarolla E, Stoppioni V, Cerbo R, Cremonte M, et al. Parental burden and its correlates in families of children with autism spectrum disorder: a multicentre study with two comparison groups. Clin Pract Epidemiol Ment Health. 2018;14:143–76. pmid:30158998
- 11. Purpura G, Tagliabue L, Petri S, Cerroni F, Mazzarini A, Nacinovich R. Caregivers’ burden of school-aged children with neurodevelopmental disorders: implications for family-centred care. Brain Sci. 2021;11(7):875. pmid:34208983
- 12. Lin Y-N, Iao L-S, Lee Y-H, Wu C-C. Parenting stress and child behavior problems in young children with autism spectrum disorder: transactional relations across time. J Autism Dev Disord. 2021;51(7):2381–91. pmid:32965625
- 13. Baker JK, Fenning RM, Preston AE, Chan N, McGregor HA, Neece CL. Parental distress and parenting behavior in families of preschool children with and without ASD: spillover and buffering. J Autism Dev Disord. 2024;54(12):4661–73. pmid:37957427
- 14. Brunt S, Nevill R, Mazurek MO. Associations among autism symptom domains and facets of caregiver strain. J Autism Dev Disord. 2024;54(4):1507–16. pmid:36637596
- 15. Lovell B. Predicting potentially harmful psychological and physical behaviours by parental caregivers toward children with autism spectrum disorder. J Child Fam Stud. 2018;27(8):2422–9.
- 16. Merriman K, Burke T, O’Reilly G. A systematic review of the effectiveness and efficacy of clinician-led psychological interventions for parents of children with ASD. Research in Autism Spectrum Disorders. 2020;76:101584.
- 17. MacKenzie KT, Eack SM. Interventions to improve outcomes for parents of children with autism spectrum disorder: a meta-analysis. J Autism Dev Disord. 2022;52(7):2859–83. pmid:34189683
- 18.
Weber JG. Individual and family stress and crisis: history and definition of stress theory. Thousand Oaks, CA: Sage Publications; 2010.
- 19.
Lazarus RS, Folkman S. Stress, appraisal, and coping. New York: Springer Publishing Company; 1984.
- 20. Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q. 1979;24(2):285–308.
- 21. Gattrell WT, Logullo P, van Zuuren EJ, Price A, Hughes EL, Blazey P. ACCORD (ACcurate COnsensus Reporting Document): a reporting guideline for consensus methods in biomedicine developed via a modified Delphi. PLoS Med. 2024;21(1):e1004326.
- 22. Tian X, Qian W, Chen B, Zhang J, Li J, Li X, et al. Psychosocial pathways to caregiver burden among families of children with autism spectrum disorder: A multicenter cross-sectional study in China. Research Square. 2026.
- 23. Li SN, Chien WT, Lam SKK, Chen ZY, Ma X. Effectiveness of parent-focused interventions for improving the mental health of parents and their children with autism spectrum disorder: a systematic review and meta-analysis. Res Autism Spectrum Disorders. 2024;114:102389.
- 24. Yu Y, McGrew JH, Boloor J. Effects of caregiver-focused programs on psychosocial outcomes in caregivers of individuals with ASD: a meta-analysis. J Autism Dev Disord. 2019;49(12):4761–79. pmid:31451967
- 25. Catalano D, Holloway L, Mpofu E. Mental health interventions for parent carers of children with autistic spectrum disorder: practice guidelines from a critical interpretive synthesis (CIS) systematic review. Int J Environ Res Public Health. 2018;15(2):341. pmid:29443933
- 26. Juvin J, Sadeg S, Julien-Sweerts S, Zebdi R. A systematic review: acceptance and commitment therapy for the parents of children and adolescents with autism spectrum disorder. J Autism Dev Disord. 2022;52(1):124–41. pmid:33635420
- 27. Kulasinghe K, Whittingham K, Mitchell AE, Boyd RN. Psychological interventions targeting mental health and the mother-child relationship in autism: systematic review and meta-analysis. Dev Med Child Neurol. 2023;65(3):329–45. pmid:36208472
- 28. Ağırkan M, Koç M, Haskan Avcı Ö. How effective are group-based psychoeducation programs for parents of children with ASD in Turkey? A systematic review and meta-analysis. Res Dev Disabil. 2023;139:104554.
- 29. Hsu CC, Sandford BA. The Delphi technique: making sense of consensus. Pract Assess Res Evaluation. 2007;12(10):1–8.
- 30. Hasson F, Keeney S, McKenna H. Research guidelines for the Delphi survey technique. J Adv Nurs. 2000;32(4):1008–15. pmid:11095242
- 31. Skulmoski GJ, Hartman FT, Krahn J. The Delphi method for graduate research. J Inf Technol Educ Res. 2007;6:1–21.
- 32. Agarwal R, Wuyke G, Sharma U, Burke SL, Howard M, Li T, et al. Stress and anxiety among parents of transition-aged children with autism spectrum disorder: a systematic review of interventions and scales. Rev J Autism Dev Disord. 2022;:1–23. pmid:36065251
- 33. Hartley M, Due C, Dorstyn D. Barriers and facilitators to engaging individuals and families with autism spectrum disorder in mindfulness and acceptance-based therapies: a meta-synthesis. Disabil Rehabil. 2022;44(17):4590–601. pmid:34033733
- 34. Rutherford M, Singh-Roy A, Rush R, McCartney D, O’Hare A, Forsyth K. Parent focused interventions for older children or adults with ASD and parent wellbeing outcomes: a systematic review with meta-analysis. Res Autism Spect Disorders. 2019;68:101450.
- 35. Yesilkaya M, Magallón-Neri E. Parental stress related to caring for a child with autism spectrum disorder and the benefit of mindfulness-based interventions for parental stress: a systematic review. Sage Open. 2024;14(2).
- 36.
National Institute for Health and Care Excellence. Autism spectrum disorder in under 19s: recognition, referral and diagnosis. London: NICE; 2011. https://www.nice.org.uk/guidance/cg128
- 37. Shang Z. Use of Delphi in health sciences research: a narrative review. Medicine (Baltimore). 2023;102(7):e32829. pmid:36800594