科研 Skill

RCT伦理与试验注册

rct-ethics-registration

Covers RCT IRB approval, informed consent, trial registration, and pre-analysis plans for research transparency. Use when preparing ethics approval or a PAP for an RCT.

发布者 Codar安全审核通过v0.1.0更新于 2026-11-18

SKILL.md

<skill_content>

Ethics and transparency are prerequisites, not afterthoughts.

<mandatory_requirements>

</mandatory_requirements>

<thinking_process> When preparing ethics and registration:

  1. Draft research protocol with theory of change and procedures
  2. Conduct power analysis to justify sample size
  3. Prepare IRB protocol (purpose, procedures, risks, benefits, consent)
  4. Submit to IRB 2-3 months before planned start
  5. Register trial immediately after IRB approval
  6. File pre-analysis plan before data collection begins
  7. Implement data security protocols
  8. Monitor for adverse events throughout study
  9. Submit amendments if protocol changes
  10. Report final results to registry and participants </thinking_process>

<implementation_pattern>

<code_template>

@app.cell
def ethics_compliance_checklist():
    # Comprehensive ethics and registration checklist
    # Use before launching any RCT to ensure compliance

    checklist = {
        "Pre-Study (Before Participant Contact)": [
            "[ ] IRB protocol drafted",
            "[ ] Informed consent forms prepared",
            "[ ] Data security plan documented",
            "[ ] IRB approval obtained",
            "[ ] Trial registered (AEA/ClinicalTrials.gov)",
            "[ ] Pre-analysis plan filed",
            "[ ] Research team trained on ethics",
        ],
        "During Study": [
            "[ ] Informed consent obtained for all participants",
            "[ ] Adverse events monitored and reported",
            "[ ] Data encrypted and securely stored",
            "[ ] IRB amendments submitted for protocol changes",
            "[ ] Annual IRB renewal obtained",
            "[ ] Trial registry updated with progress",
        ],
        "Post-Study": [
            "[ ] Final IRB report submitted",
            "[ ] Trial registry updated with results",
            "[ ] Data de-identified before sharing",
            "[ ] Results shared with community",
            "[ ] Consent forms archived per retention policy",
            "[ ] PII destroyed per protocol timeline",
        ],
    }

    print("ETHICS & REGISTRATION COMPLIANCE CHECKLIST")
    print("=" * 60)

    for phase, items in checklist.items():
        print(f"\n{phase}")
        print("-" * 40)
        for item in items:
            print(f"  {item}")

    print("\nCritical deadlines:")
    print("- IRB submission: 2-3 months before start")
    print("- Trial registration: Before data collection")
    print("- PAP filing: Before accessing outcome data")
    print("- Annual renewal: Before approval expiration")

    return checklist,

</code_template>

</implementation_pattern>

consent_text = """
INFORMED CONSENT FOR RESEARCH PARTICIPATION

Study Title: [Title]
Principal Investigator: [Name, contact]

1. PURPOSE
You are invited to participate in research studying [purpose in plain language].
We are enrolling approximately [N] participants.

2. PROCEDURES
If you agree, you will:
- [Specific procedures]
- [Time commitment]
- [Number of visits/contacts]

3. RISKS
Possible risks include:
- [List all risks, even if minimal]
- [Be specific and complete]
These risks are [minimal / greater than minimal].

4. BENEFITS
Possible benefits to you:
- [Direct benefits, if any]
- [Or state clearly if no direct benefit]

Possible benefits to others:
- [Broader societal benefit]

5. COMPENSATION
You will receive [specific compensation amount and timing].
[If no compensation, state clearly]

6. CONFIDENTIALITY
Your information will be kept confidential:
- Data will be de-identified (names removed)
- Stored on encrypted, password-protected computers
- Only research team will access identifiable data
- Identifiers destroyed after [time period]

7. VOLUNTARY PARTICIPATION
Participation is completely voluntary. You may:
- Refuse to participate
- Skip any questions
- Withdraw at any time without penalty

Your decision will not affect [services, relationships, etc.].

8. CONTACT INFORMATION
Questions about study: [PI contact]
Questions about rights: [IRB contact]

CONSENT
[ ] I have read this form
[ ] My questions have been answered
[ ] I voluntarily agree to participate

_________________________  _________
Participant Signature      Date

_________________________  _________
Researcher Signature       Date
"""

print(consent_text)

print("\nKey requirements:")
print("- 8th grade reading level")
print("- All 8 basic elements included")
print("- Voluntary nature emphasized")
print("- Specific risks and benefits")
print("- Contact information provided")

return consent_text,
</code>
<lesson>
Informed consent requires: (1) purpose, (2) procedures, (3) risks, (4) benefits, (5) alternatives, (6) confidentiality, (7) voluntary nature, (8) contacts. Written at 8th grade level. Emphasize voluntary participation and right to withdraw.
</lesson>
</example>

</examples>

<common_mistakes>

<mistake severity="critical">
  <what>Starting participant contact before IRB approval</what>
  <consequence>Research misconduct, legal liability, inability to use data, institutional sanctions</consequence>
  <prevention>IRB approval is prerequisite. Never recruit, consent, or collect data before approval</prevention>
</mistake>

<mistake severity="critical">
  <what>Not registering trial or filing pre-analysis plan</what>
  <consequence>Journal rejection, results perceived as data-mined, loss of scientific credibility</consequence>
  <prevention>Register trial immediately after IRB approval, file PAP before accessing outcome data</prevention>
</mistake>

<mistake severity="high">
  <what>Inadequate informed consent (missing elements, coercive, not voluntary)</what>
  <consequence>IRB violations, invalid consent, ethical breach, unusable data</consequence>
  <prevention>Include all 8 required elements, ensure voluntary participation, use plain language</prevention>
</mistake>

<mistake severity="high">
  <what>Weak data security (unencrypted data, no de-identification, broad access)</what>
  <consequence>Data breaches, participant harm, IRB sanctions, legal penalties</consequence>
  <prevention>Encrypt all data, separate PII from analysis files, limit access, destroy PII per protocol</prevention>
</mistake>

<mistake severity="medium">
  <what>Not reporting adverse events or protocol deviations</what>
  <consequence>IRB sanctions, continued participant harm, research misconduct findings</consequence>
  <prevention>Monitor for adverse events, report serious events within 24 hours, document all deviations</prevention>
</mistake>

</common_mistakes>

<interpretation_guide>

<irb_review_types>
- **Exempt**: Minimal risk, no intervention, secondary data (still requires IRB determination)
- **Expedited**: Minimal risk, specific categories (surveys, interviews, some RCTs)
- **Full board**: Greater than minimal risk, vulnerable populations, deception

Timeline:
- Exempt/Expedited: 2-4 weeks
- Full board: 4-8 weeks (meets monthly typically)
- Always add 2-3 months buffer to timeline
</irb_review_types>

<vulnerable_populations>
Extra protections required for:
- Children (parental consent + child assent)
- Prisoners (special IRB composition, no coercion)
- Pregnant women (consider fetal risk)
- Cognitively impaired (capacity assessment, LAR if needed)
- Economically disadvantaged (ensure compensation not coercive)

For vulnerable populations, demonstrate why inclusion is necessary and how risks are minimized.
</vulnerable_populations>

<registration_platforms>
- Economics/social science: AEA RCT Registry (socialscienceregistry.org)
- Health/medical: ClinicalTrials.gov
- General: OSF Registries (osf.io)
- Politics: EGAP Registry (egap.org)

Minimum info: title, PIs, outcomes, sample size, timeline, hypotheses
</registration_platforms>

</interpretation_guide>

<references>
<paper>Belmont Report (1979). Ethical Principles and Guidelines for the Protection of Human Subjects of Research. National Commission.</paper>
<paper>45 CFR 46 - Protection of Human Subjects (Common Rule). U.S. Department of Health and Human Services.</paper>
<paper>CONSORT Statement. Consolidated Standards of Reporting Trials. http://www.consort-statement.org/</paper>
<paper>Casey, K., Glennerster, R., & Miguel, E. (2012). Reshaping institutions: Evidence on aid impacts using a preanalysis plan. Quarterly Journal of Economics, 127(4), 1755-1812.</paper>
<paper>Christensen, G., & Miguel, E. (2018). Transparency, reproducibility, and the credibility of economics research. Journal of Economic Literature, 56(3), 920-980.</paper>
</references>

</skill_content>

示例提示词

帮我准备这项随机实验的IRB申请材料和预分析计划。

所属插件

社会学、政治学与城市研究面向社会科学研究:问卷与访谈设计、质性编码、文本即数据、社会网络、ABM、比较政治与选举数据、RCT 与 DID、城市与区域分析。

安装该插件会一并获得同领域的 58 个 Skill / MCP。

同插件的其他科研 Skill

常见问题

RCT伦理与试验注册能做什么?

Covers RCT IRB approval, informed consent, trial registration, and pre-analysis plans for research transparency. Use when preparing ethics approval or a PAP for an RCT。

如何安装RCT伦理与试验注册?

在科搭 Codar 左侧点击「插件」,在「组件」页签搜索「RCT伦理与试验注册」并安装;在 Claude Code 或 Codex 中可以发送本页的安装提示词,或下载 Skill 包后解压到 ~/.claude/skills 或 ~/.codex/skills。

RCT伦理与试验注册需要哪些权限?

该科研 Skill没有声明额外权限;执行具体操作时仍由 Agent 按自身规则请求确认。

RCT伦理与试验注册经过官方审核吗?

是。SpeedSkills 收录的每个版本都会记录内容摘要与权限声明,当前版本已通过安全审核。