Develop a cross-specialty long COVID consensus guideline process with Delphi rounds,…
- User
Develop a cross-specialty long COVID consensus guideline process with Delphi rounds, living literature review, and dissemination strategy.
- Assistant
Cross-Specialty Long COVID Consensus Guideline Development Process
Overview and Objectives
This process outlines a rigorous, transparent, and adaptable framework for developing living consensus guidelines on long COVID (post-acute sequelae of SARS-CoV-2 infection). It integrates a multidisciplinary Delphi method for expert consensus, a living systematic literature review for evidence integration, and a multi-channel dissemination strategy to ensure rapid uptake.
Key Objectives:
- Achieve ≥80% consensus on key recommendations across specialties (e.g., pulmonology, cardiology, neurology, immunology, rheumatology, psychology, primary care, rehabilitation).
- Produce actionable, evidence-based guidelines on diagnosis, risk stratification, management, rehabilitation, and research priorities.
- Enable continuous updates as new evidence emerges.
- Promote equity by including global experts from diverse regions and underrepresented groups.
Guideline Scope: Modular sections covering symptoms (fatigue, dyspnea, cognitive impairment, dysautonomia), diagnostics, therapeutics, multidisciplinary care pathways, and patient-centered outcomes.
Timeline: Initial guidelines in 6-9 months; quarterly updates thereafter.
Governance: Overseen by a Steering Committee (10-15 members: methodologists, clinicians, patients, policymakers).
Step 1: Preparation and Panel Assembly (Months 1-2)
- Steering Committee Formation: Recruit via professional societies (e.g., ATS, ESC, AAN), patient advocacy groups (e.g., Long COVID SOS), and WHO/CDC networks. Include 2-3 patient representatives.
- Expert Panel Recruitment: 50-100 experts (20-30 per specialty cluster: respiratory/cardio; neuro/psych; immune/rheum; rehab/primary care). Criteria: ≥5 years clinical/research experience in COVID/long COVID; geographic diversity; no major conflicts of interest (managed via disclosure).
- Use snowball sampling and open calls via platforms like ResearchGate and specialty journals.
- Tools Setup:
- Secure online platform (e.g., REDCap or DelphiManager) for anonymous surveys.
- Living review infrastructure: Automated PubMed/Embase/Cochrane alerts + AI screening (e.g., DistillerSR or ASReview).
Step 2: Living Systematic Literature Review (Ongoing, Starting Month 1)
- Methodology: Follow PRISMA-Living guidelines.
- Search Strategy: Weekly automated searches (MeSH: "Post-Acute COVID-19 Syndrome" OR "Long COVID"; filters: humans, English, post-2020). Include grey literature (preprints, guidelines).
- Screening: Dual human-AI review (title/abstract → full-text). Risk-of-bias assessment (ROBINS-I for non-randomized studies; AMSTAR-2 for reviews).
- Synthesis: GRADE-rated evidence summaries in living format (e.g., interactive Evidence-to-Decision tables via GRADEpro). Focus on PICO questions (e.g., "In long COVID patients with fatigue, does graded exercise therapy improve outcomes?").
- Team: 5-7 reviewers (librarian, epidemiologist, 2 clinicians, AI specialist).
- Outputs: Quarterly evidence briefs shared pre-Delphi rounds; version-controlled repository (e.g., GitHub or Zenodo).
Step 3: Delphi Consensus Process (Months 2-5, Iterative)
- Structure: 3-4 anonymous electronic rounds + virtual consensus meeting. Pre-defined consensus threshold: ≥80% agreement (Likert scale: 1-9; 7-9 = agree).
- Round 1 (Open-Ended):
- Distribute 20-30 draft statements (developed by Steering Committee + initial lit review).
- Experts suggest additions/revisions.
- Rounds 2-3 (Structured):
- Present anonymized group statistics (median, IQR, % agreement).
- Revised statements + rationale/justification fields.
- Stable items removed; <50% agreement items discussed.
- Round 4 (If Needed): Final voting with forced choice (agree/disagree).
- Consensus Meeting (Month 5): Virtual (Zoom), 2-4 hours. Discuss <80% items live; vote via Slido. Record rationales.
- Patient Input: Parallel patient Delphi (n=20-30) for lived experience statements.
- Monitoring: Response rate target >80%; reminders via email/SMS.
Example Statement: "We recommend multidisciplinary assessment for long COVID patients with ≥3 organ systems involved (GRADE: Moderate evidence; Consensus: 85%)."
Step 4: Guideline Drafting and Approval (Months 5-6)
- Integration: Steering Committee synthesizes Delphi outputs + living review into draft using AGREE II-compliant template.
- Sections: Executive summary, methods, recommendations (with GRADE/strength), flowcharts, research gaps.
- Visual aids: Algorithms, infographics (e.g., for primary care triage).
- External Review: 20-30 independent reviewers (AGREE tool); 2-week feedback window.
- Final Approval: ≥90% panel sign-off; minority opinions appended.
Step 5: Dissemination and Implementation Strategy (Months 6+ Ongoing)
-
Core Principles: Open-access, multilingual (English + top 5 languages), patient-friendly versions.
-
Channels and Timeline:
Channel Tactics Timeline Metrics Publications Lead paper in Lancet or BMJ; modular sections in specialty journals (e.g., Chest, Neurology). Preprint on medRxiv. Month 6; updates quarterly Citations (>500/yr); Altmetric score >1,000 Digital Hub Dedicated website (e.g., LongCOVIDGuidelines.org) with living PDF, app (iOS/Android), interactive tools. Launch Month 6 100K visits/yr; app downloads >50K Professional Networks Endorsement by societies (ATS, ERS); webinars; integration into UpToDate/Epocrates. Month 7+ 80% society adoption Conferences Plenary sessions (e.g., ATS 2025); posters/booths. Ongoing 5K attendees reached Social/Outreach Twitter/X threads (#LongCOVIDGuidelines); TikTok/YouTube explainers; patient org partnerships. Launch Month 6 1M impressions; 10K shares Policy/Implementation Briefings to WHO/CDC/NICE; EHR integration (e.g., FHIR codes); audits in pilot clinics. Month 8+ Policy citations; 20 clinics adopting Global Equity Translations via Google/DeepL + experts; low-resource adaptations (e.g., WHO essential package). Month 9+ Usage in 50+ countries -
Evaluation: Annual audits (uptake surveys, guideline adherence studies). Feedback loop via website form.
Step 6: Living Update Mechanism (Ongoing, Quarterly)
- Triggers: New high-impact evidence (e.g., RCTs), Delphi re-vote if >10% shift.
- Mini-Delphi: Subset panel (20-30) for rapid updates (<4 weeks).
- Versioning: Semantic versioning (e.g., v1.2); changelog with GRADE re-assessments.
- Sustainability: Funding via grants (NIH, Wellcome); volunteer core team.
Resources and Risks
- Budget Estimate: 50K, personnel 150K).
- Risk Mitigation: COI management (annual declarations); bias audits; inclusivity quotas.
- Success Metrics: Consensus rate ≥80%; guideline cited in >50% new long COVID papers; improved patient outcomes in audits.
This process ensures evidence-driven, expert-consensus, adaptable guidelines that evolve with the science. For implementation, contact the Steering Committee lead.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.