writing-irb-submissions

# Writing IRB Submissions

## Why This Skill Exists

Every human-subjects research study in the United States requires IRB review under the Common Rule (45 CFR 46) and, for FDA-regulated research, 21 CFR Parts 56 and 50. Incomplete or non-compliant submissions cause weeks of delays, stalling enrollment timelines and burning sponsor confidence. This skill encodes the complete IRB submission package assembly workflow so that initial submissions, amendments, and continuing reviews meet institutional and federal requirements on the first pass.

---

## Checkpoint A — Intake and Scoping

Before drafting any submission documents, confirm:

### Required Intake Questions
1. What type of IRB review is needed (initial, amendment, continuing review, reportable event, closure)?
2. Which IRB has jurisdiction (institutional IRB, central/commercial IRB such as WCG, Advarra)?
3. Is the study FDA-regulated (IND, IDE) or non-regulated?
4. Does the study qualify for exempt or expedited review per 45 CFR 46.104 or 46.110 categories?
5. Does the study involve vulnerable populations (children, prisoners, pregnant women, cognitively impaired)?
6. Is this a single-site or multi-site study? If multi-site, is a single-IRB arrangement required per the 2018 Common Rule revision?
7. What is the funding source (federal, industry, foundation)?
8. Are there HIPAA considerations requiring a separate authorization or waiver?
9. Is a Data Use Agreement (DUA) or Material Transfer Agreement (MTA) involved?
10. What is the anticipated date of first enrollment?

### Required Source Documents
- Final or near-final protocol (version-controlled)
- Investigator's Brochure (for IND/IDE studies)
- Draft informed consent form(s)
- HIPAA authorization or waiver request
- Recruitment materials (flyers, scripts, advertisements)
- Investigator CVs and medical licenses
- Conflict-of-interest disclosures (financial and non-financial)
- Data management and safety monitoring plan
- Sponsor-investigator agreements (if applicable)
- Certificate of Confidentiality (if applicable)

---

## Step 1 — Determine Review Category

Classify the study into the correct regulatory category:

- **Exempt** (45 CFR 46.104): Educational research, surveys/interviews with non-identifiable data, benign behavioral interventions, secondary research on existing data. Note: exempt does not mean no review — the IRB or designee must confirm exemption.
- **Expedited** (45 CFR 46.110): Minimal-risk research fitting one of the nine expedited categories (e.g., collection of small blood volumes, non-invasive data collection, research on existing specimens). Document which specific category applies.
- **Full Board**: Greater-than-minimal-risk research, or any study involving vulnerable populations where expedited review is insufficient. Requires convened IRB quorum with at least one non-scientist member.

Flag studies that the PI believes are exempt but that may actually require expedited or full review due to identifiability, sensitive topics, or deception.

---

## Step 2 — Draft Protocol Summary for IRB

Create a lay-language protocol summary (distinct from the full protocol) that addresses IRB reviewers' key concerns:

1. **Study purpose**: One-paragraph plain-language description of the research question
2. **Background and rationale**: Why this research is needed; gap in current knowledge
3. **Study design**: Type (RCT, observational, qualitative), arms, blinding, duration
4. **Study population**: Target N, key eligibility criteria, recruitment strategy
5. **Study procedures**: What participants will experience at each visit — described in participant-facing language
6. **Risks**: Enumerate known risks by likelihood and severity; distinguish drug-related, procedure-related, and privacy risks
7. **Benefits**: Direct benefits to participants (if any) and benefits to society; do not overstate therapeutic benefit for early-phase studies
8. **Risk-benefit analysis**: Explicit statement that risks are minimized and reasonable in relation to anticipated benefits and the importance of the knowledge to be gained (45 CFR 46.111(a)(2))
9. **Alternatives**: What alternatives to participation exist (standard treatment, other studies, no treatment)
10. **Data confidentiality protections**: De-identification methods, data storage security, access controls, Certificate of Confidentiality status

---

## Step 3 — Prepare Informed Consent Document

Draft the consent form per 45 CFR 46.116 required elements and 21 CFR 50.25 (if FDA-regulated):

### Eight Required Elements (45 CFR 46.116(b))
1. Statement that the study involves research; explanation of purposes, expected duration, procedures
2. Description of reasonably foreseeable risks or discomforts
3. Description of benefits to subject or others
4. Disclosure of appropriate alternative procedures or treatments
5. Description of confidentiality protections
6. For more-than-minimal-risk research: explanation of compensation and treatment availability for injury
7. Contact information for questions about research, rights, and research-related injury
8. Statement that participation is voluntary; refusal involves no penalty

### Additional Elements When Appropriate (45 CFR 46.116(c))
- Unforeseeable risks (e.g., reproductive risks, unknown long-term effects)
- Circumstances for investigator termination of participation
- Additional costs to participant
- Consequences of withdrawal and procedures for orderly termination
- Notification of significant new findings during the study
- Approximate number of participants

### Special Consent Provisions
- **Children** (45 CFR 46 Subpart D): Assent form (age-appropriate language, typically 7–12 and 13–17 versions) plus parental/guardian permission
- **Non-English speakers**: Consent translation with back-translation verification; short-form consent process per 21 CFR 50.27
- **Cognitively impaired**: Legally authorized representative (LAR) consent procedures
- **Emergency research** (21 CFR 50.24): Exception from informed consent with community consultation documentation

---

## Step 4 — Compile HIPAA Documentation

If the study involves protected health information (PHI):

- **HIPAA Authorization**: Standalone or embedded in consent; must specify PHI elements, who will access, purpose, expiration
- **Waiver of Authorization** (45 CFR 164.512(i)): If seeking waiver, document that (1) use involves no more than minimal risk to privacy, (2) research could not practicably be conducted without waiver, (3) adequate plan to protect identifiers and destroy them at earliest opportunity
- **Limited Data Set with Data Use Agreement**: Alternative to full waiver for research using dates and geographic data

---

## Step 5 — Prepare Supplementary Materials

Assemble all ancillary documents:

1. **Recruitment materials**: Advertisements, social media posts, screening scripts — all must be IRB-approved before use
2. **Questionnaires and survey instruments**: Include validated instruments with permission documentation
3. **Investigator qualifications**: Current CVs (NIH biosketch or sponsor format), medical licenses, GCP training certificates
4. **Financial disclosure**: Per 42 CFR Part 50 (PHS-funded) or 21 CFR 54 (FDA-regulated) — significant financial interests >$5,000
5. **Sponsor agreements**: Clinical trial agreements, budgets affecting conflict assessment
6. **Data and Safety Monitoring Plan (DSMP)**: Required for all NIH-funded clinical trials; describe monitoring entity, frequency, stopping rules

---

## Step 6 — Complete IRB Application Form

Fill in the institution-specific or central-IRB application addressing:

- Study title, PI information, co-investigators, study team roles
- Funding source and grant/contract number
- IND/IDE number (if applicable)
- Study sites and single-IRB documentation
- Participant population characteristics
- Recruitment methods and locations
- Consent process description (who obtains, where, waiting period)
- Data management and storage plan
- Conflict-of-interest attestations
- CITI training completion dates for all study personnel

---

## Checkpoint B — Pre-Submission Review

Before submitting the package to the IRB, verify:

1. [ ] All eight required consent elements are present and in lay language (8th-grade reading level target)
2. [ ] Risk-benefit analysis is explicit and balanced — not promotional
3. [ ] HIPAA documentation matches the data elements actually collected
4. [ ] Recruitment materials do not overstate benefits or use coercive language
5. [ ] All study personnel have current GCP/CITI training
6. [ ] Financial disclosures are complete and consistent with institutional COI records
7. [ ] Protocol version matches the consent form version
8. [ ] If amendment: tracked-changes version of consent and protocol are included with summary of changes
9. [ ] If continuing review: enrollment numbers, adverse events, and protocol deviations are summarized
10. [ ] Application form is complete — no blank required fields

---

## Quality Audit

- [ ] Consent form reading level has been assessed (Flesch-Kincaid or SMOG formula)
- [ ] Regulatory citations (45 CFR 46, 21 CFR 50/56, HIPAA) are correctly applied
- [ ] Vulnerable-population protections match the actual study population
- [ ] Consent form version number and date are in the footer of every page
- [ ] Contact information includes three required contacts (PI, IRB, institutional representative)
- [ ] All referenced attachments are actually included in the submission package
- [ ] No exculpatory language exists in the consent form (prohibited by 45 CFR 46.116)
- [ ] Sponsor-required language does not conflict with federal requirements
- [ ] All [VERIFY] flags have been resolved or escalated

---

## Guidelines

1. Never include exculpatory language in consent forms — statements that waive participant rights or release investigators from liability are prohibited under 45 CFR 46.116
2. Consent forms must be understandable to a lay audience; avoid medical jargon or define it in parentheses
3. Risk descriptions must be specific and quantified where possible (e.g., "nausea occurs in approximately 15% of patients" not "nausea may occur")
4. Do not promise therapeutic benefit in Phase I studies — use language like "this study is not designed to treat your condition"
5. For multi-site studies under single-IRB arrangements, identify the reviewing IRB and document reliance agreements
6. IRB submissions are living documents — amendments must be submitted before implementing any change (except to eliminate immediate hazards per 21 CFR 56.108(a)(4))
7. Recruitment material review by the IRB is required before any advertising begins
8. When consent processes involve electronic consent (eConsent), ensure 21 CFR Part 11 compliance for electronic signatures
9. Mark any uncertain risk characterization with [VERIFY] for PI review
10. This skill produces submission drafts — final approval requires PI signature and IRB determination