106 Reporting 02a58157

Reporting Guidelines for Scientific Studies

Overview

Reporting guidelines are evidence-based recommendations for what information should be included when reporting specific types of research studies. They provide checklists and flow diagrams to ensure complete, accurate, and transparent reporting, which is essential for readers to assess study validity and for other researchers to replicate the work.

The EQUATOR Network (Enhancing the QUAlity and Transparency Of health Research) maintains a comprehensive library of reporting guidelines. Using appropriate reporting guidelines improves manuscript quality and increases the likelihood of publication acceptance.

Why Use Reporting Guidelines?

Benefits

For authors:

• Ensures nothing important is forgotten
• Increases acceptance rates
• Improves manuscript organization
• Reduces reviewer requests for additional information

For readers and reviewers:

• Enables critical appraisal of study validity
• Facilitates systematic reviews and meta-analyses
• Improves understanding of what was actually done

For science:

• Enhances reproducibility
• Reduces research waste
• Improves transparency
• Enables better evidence synthesis

When to Use

• During study design: Many guidelines include protocol versions (e.g., SPIRIT for trial protocols)
• During manuscript drafting: Use checklist to ensure all items are covered
• Before submission: Verify adherence and often submit checklist with manuscript
• Many journals require: Reporting guideline checklists as part of submission

Major Reporting Guidelines by Study Type

CONSORT - Randomized Controlled Trials

Full name: Consolidated Standards of Reporting Trials

When to use: Any randomized controlled trial (RCT), including pilot and feasibility trials

Latest version: CONSORT 2010 (updated statement)

Key components:

• Checklist: 25 items covering title, abstract, introduction, methods, results, discussion
• Flow diagram: Participant flow through enrollment, allocation, follow-up, and analysis

Main checklist items:

1. Title identifies study as randomized trial
2. Structured abstract
3. Scientific background and rationale
4. Specific objectives and hypotheses
5. Trial design description (parallel, crossover, factorial, etc.)
6. Eligibility criteria for participants
7. Settings and locations of data collection
8. Interventions described in sufficient detail for replication
9. Primary and secondary outcomes defined
10. Sample size determination and power calculation
11. Randomization sequence generation
12. Allocation concealment mechanism
13. Blinding implementation
14. Statistical methods
15. Participant flow with reasons for dropouts
16. Recruitment dates and follow-up dates
17. Baseline characteristics table
18. Analysis results for each outcome
19. Harms and adverse events
20. Trial limitations
21. Generalizability
22. Interpretation consistent with results
23. Trial registration number
24. Full protocol access
25. Funding sources

Extensions for specific designs:

• CONSORT for cluster randomized trials
• CONSORT for non-inferiority and equivalence trials
• CONSORT for pragmatic trials
• CONSORT for crossover trials
• CONSORT for N-of-1 trials
• CONSORT for stepped wedge designs

Where to access: http://www.consort-statement.org/

STROBE - Observational Studies

Full name: Strengthening the Reporting of Observational Studies in Epidemiology

When to use: Cohort studies, case-control studies, and cross-sectional studies

Latest version: STROBE 2007 (widely adopted standard)

Key study designs covered:

• Cohort: Follow exposed and unexposed groups forward in time
• Case-control: Compare exposure history between cases and controls
• Cross-sectional: Measure exposure and outcome simultaneously

Main checklist items (22 items):

1. Title and abstract indicate study design
2. Background and rationale
3. Objectives
4. Study design with rationale
5. Setting, locations, and dates
6. Eligibility criteria and selection methods
7. Variables clearly defined (outcomes, exposures, confounders)
8. Data sources and measurement methods
9. Bias management strategies
10. Study size justification
11. Handling of quantitative variables
12. Statistical methods including confounding and interactions
13. Sensitivity analyses
14. Participant flow with reasons for non-participation
15. Descriptive data including follow-up time
16. Outcome data
17. Main results with unadjusted and adjusted estimates
18. Other analyses (subgroups, sensitivity analyses)
19. Key results summary
20. Limitations with potential bias discussion
21. Interpretation and generalizability
22. Funding sources and role

Extensions:

• STROBE-ME (Molecular Epidemiology)
• RECORD (Routinely collected health data)
• STROBE-RDS (Respondent-driven sampling)

Where to access: https://www.strobe-statement.org/

PRISMA - Systematic Reviews and Meta-Analyses

Full name: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

When to use: Systematic reviews with or without meta-analysis

Latest version: PRISMA 2020 (significant update)

Key components:

• Checklist: 27 items covering all sections
• Flow diagram: Study selection process

Main sections:

1. Title: Identify as systematic review/meta-analysis
2. Abstract: Structured summary
3. Introduction: Rationale and objectives
4. Methods:
   • Eligibility criteria
   • Information sources (databases, dates)
   • Search strategy (full strategy for at least one database)
   • Selection process
   • Data collection process
   • Data items extracted
   • Risk of bias assessment
   • Effect measures
   • Synthesis methods
   • Reporting bias assessment
   • Certainty assessment (e.g., GRADE)
5. Results:
   • Study selection flow diagram
   • Study characteristics
   • Risk of bias assessment results
   • Synthesis results (meta-analysis if applicable)
   • Reporting biases
   • Certainty of evidence
6. Discussion:
   • Limitations
   • Interpretation
   • Implications

Extensions:

• PRISMA for Abstracts
• PRISMA for Protocols (PRISMA-P)
• PRISMA for Network Meta-Analyses
• PRISMA for Scoping Reviews (PRISMA-ScR)
• PRISMA for Individual Patient Data
• PRISMA for Diagnostic Test Accuracy
• PRISMA for Equity-focused reviews

Where to access: http://www.prisma-statement.org/

SPIRIT - Study Protocols for Clinical Trials

Full name: Standard Protocol Items: Recommendations for Interventional Trials

When to use: Protocols for randomized trials and other planned intervention studies

Latest version: SPIRIT 2013

Purpose: Ensure trial protocols contain complete descriptions before trial begins

Main checklist items (33 items):

• Administrative information (title, trial registration, funding)
• Introduction (background, rationale, objectives)
• Methods: Trial design
  • Study setting
  • Eligibility criteria
  • Interventions in detail
  • Outcomes (primary and secondary)
  • Participant timeline
  • Sample size calculation
  • Recruitment strategy
  • Allocation and randomization
  • Blinding
  • Data collection methods
  • Data management
  • Statistical methods
  • Monitoring (data monitoring committee)
  • Harms reporting
  • Auditing
• Ethics and dissemination
  • Ethics approval
  • Consent procedures
  • Confidentiality
  • Dissemination plans

Where to access: https://www.spirit-statement.org/

STARD - Diagnostic Accuracy Studies

Full name: Standards for Reporting of Diagnostic Accuracy Studies

When to use: Studies evaluating diagnostic test accuracy

Latest version: STARD 2015

Main checklist items (30 items):

1. Study design identification
2. Background information and objectives
3. Study design description
4. Participant selection criteria and recruitment
5. Data collection methods
6. Index test description and execution
7. Reference standard description
8. Rationale for choosing reference standard
9. Test result definition and cutoffs
10. Flow of participants with timing
11. Baseline demographic and clinical characteristics
12. Cross-tabulation of index test results by reference standard
13. Estimates of diagnostic accuracy with confidence intervals
14. Handling of indeterminate results
15. Adverse events from testing

Flow diagram: Shows participant flow and test results

Where to access: https://www.equator-network.org/reporting-guidelines/stard/

TRIPOD - Prediction Model Studies

Full name: Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis

When to use: Studies developing, validating, or updating prediction models

Latest version: TRIPOD 2015

Types of studies:

• Model development only
• Model development with validation
• External validation of existing model
• Model update

Main checklist items (22 items):

1. Title identifies study as prediction model study
2. Abstract summarizes key elements
3. Background and objectives
4. Data source and participants
5. Outcome definition
6. Predictors (candidate and selected)
7. Sample size justification
8. Missing data handling
9. Model building procedure
10. Model specification (equation or algorithm)
11. Model performance measures
12. Risk groups if used
13. Participant flow diagram
14. Model development results
15. Model performance
16. Model updating if applicable

Where to access: https://www.tripod-statement.org/

ARRIVE - Animal Research

Full name: Animal Research: Reporting of In Vivo Experiments

When to use: All in vivo animal studies

Latest version: ARRIVE 2.0 (2020 update)

Two sets of items:

ARRIVE Essential 10 (minimum requirements):

1. Study design
2. Sample size calculation
3. Inclusion and exclusion criteria
4. Randomization
5. Blinding
6. Outcome measures
7. Statistical methods
8. Experimental animals (species, strain, sex, age)
9. Experimental procedures
10. Results and interpretation

ARRIVE Recommended Set (additional items for full reporting):

• Abstract, background, objectives
• Ethics statement
• Housing and husbandry
• Animal care and monitoring
• Interpretation and generalizability
• Protocol registration
• Data access

Where to access: https://arriveguidelines.org/

CARE - Case Reports

Full name: CAse REport Guidelines

When to use: Case reports and case series

Latest version: CARE 2013

Main checklist items (13 items):

1. Title with "case report"
2. Abstract summarizing case
3. Introduction with case background
4. Patient information (demographics, primary concern)
5. Clinical findings
6. Timeline of events
7. Diagnostic assessment
8. Therapeutic intervention
9. Follow-up and outcomes
10. Discussion with strengths and limitations
11. Patient perspective
12. Informed consent

Where to access: https://www.care-statement.org/

SQUIRE - Quality Improvement Studies

Full name: Standards for QUality Improvement Reporting Excellence

When to use: Healthcare quality improvement reports

Latest version: SQUIRE 2.0 (2015)

Main sections (18 items):

1. Title and abstract
2. Introduction (problem description, available knowledge, rationale, objectives)
3. Methods (context, intervention, study design, measures, analysis, ethical review)
4. Results (intervention, outcomes)
5. Discussion (summary, interpretation, limitations, conclusions)
6. Other information (funding)

Where to access: http://www.squire-statement.org/

CHEERS - Economic Evaluations

Full name: Consolidated Health Economic Evaluation Reporting Standards

When to use: Health economic evaluations

Latest version: CHEERS 2022 (major update from 2013)

Main checklist items (28 items):

1. Title identification as economic evaluation
2. Abstract
3. Background and objectives
4. Target population and subgroups
5. Setting and location
6. Study perspective
7. Comparators
8. Time horizon
9. Discount rate
10. Selection of outcomes
11. Measurement of effectiveness
12. Measurement and valuation of costs
13. Currency and price adjustments
14. Choice of model
15. Assumptions
16. Analytical methods

Where to access: https://www.equator-network.org/reporting-guidelines/cheers/

SRQR - Qualitative Research

Full name: Standards for Reporting Qualitative Research

When to use: Qualitative and mixed methods research

Latest version: SRQR 2014

Main sections:

• Title and abstract
• Introduction (problem formulation, purpose)
• Methods (qualitative approach, researcher characteristics, context, sampling strategy, ethical issues, data collection, data analysis, trustworthiness)
• Results (synthesis and interpretation, links to empirical data)
• Discussion (limitations, implications)

Alternative: COREQ (Consolidated criteria for reporting qualitative research) for interviews and focus groups

Where to access: https://www.equator-network.org/reporting-guidelines/srqr/

How to Use Reporting Guidelines

During Study Planning

1. Identify relevant guideline based on study design
2. Review checklist items that require planning (e.g., randomization, blinding)
3. Design study to ensure all required elements will be captured
4. Consider protocol guidelines (e.g., SPIRIT for trials)

During Manuscript Drafting

1. Download checklist from guideline website
2. Work through each item systematically
3. Note where each item is addressed in manuscript (page/line numbers)
4. Revise manuscript to include missing items
5. Use flow diagrams as appropriate

Before Submission

1. Complete formal checklist with page numbers
2. Review all items are adequately addressed
3. Include checklist with submission if journal requires
4. Note guideline adherence in cover letter or methods

Example Checklist Entry

Item 7: Eligibility criteria for participants, and the settings and locations where the data were collected
Page 6, lines 112-125: "Participants were community-dwelling adults aged 60-85 years with mild cognitive impairment (MCI) as defined by Petersen criteria. Exclusion criteria included dementia diagnosis, major psychiatric disorders, or unstable medical conditions. Recruitment occurred from three memory clinics in Boston, MA, between January 2022 and December 2023."

Finding the Right Guideline

EQUATOR Network Search

Website: https://www.equator-network.org/

How to use:

1. Select your study design from the wizard
2. Browse by health research category
3. Search for specific keywords
4. Filter by guideline status (development stage)

By Study Design

Multiple Guidelines

Some studies may require multiple guidelines:

Example 1: Pilot RCT with qualitative component

• CONSORT for quantitative arm
• SRQR for qualitative component

Example 2: Systematic review of diagnostic tests

• PRISMA for review methods
• STARD considerations for included studies

Extensions and Adaptations

Many reporting guidelines have extensions for specific contexts:

CONSORT Extensions (examples)

• CONSORT for Abstracts: Structured abstracts for RCT reports
• CONSORT for Harms: Reporting adverse events
• CONSORT-EHEALTH: eHealth interventions
• CONSORT-SPI: Social and psychological interventions

PRISMA Extensions (examples)

• PRISMA-P: Protocols for systematic reviews
• PRISMA for Abstracts: Conference abstracts
• PRISMA-NMA: Network meta-analyses
• PRISMA-IPD: Individual patient data reviews
• PRISMA-S: Search strategies
• PRISMA-DTA: Diagnostic test accuracy reviews

STROBE Extensions (examples)

• STROBE-ME: Molecular epidemiology
• RECORD: Routinely collected health data

Creating Flow Diagrams

CONSORT Flow Diagram

Four stages:

1. Enrollment: Assessed for eligibility
2. Allocation: Randomly assigned to groups
3. Follow-up: Received intervention, lost to follow-up
4. Analysis: Included in analysis

Example:

Assessed for eligibility (n=250)
    ↓
Excluded (n=50)
  • Did not meet criteria (n=30)
  • Declined to participate (n=15)
  • Other reasons (n=5)
    ↓
Randomized (n=200)
    ├─────────────────┬─────────────────┐
    ↓                 ↓                 ↓
Allocated to       Allocated to      Allocated to
Intervention A     Intervention B     Control
(n=67)            (n=66)            (n=67)
    ↓                 ↓                 ↓
Lost to follow-up  Lost to follow-up  Lost to follow-up
(n=3)             (n=5)             (n=2)
    ↓                 ↓                 ↓
Analyzed          Analyzed          Analyzed
(n=64)            (n=61)            (n=65)

PRISMA Flow Diagram

Stages:

1. Identification: Records from databases and registers
2. Screening: Records screened, excluded
3. Included: Studies included in review and synthesis

New features in PRISMA 2020:

• Separate tracking for database and register searches
• Tracking of duplicate removal
• Clear distinction between reports and studies

Common Mistakes and How to Avoid Them

Mistake 1: Not Using Guidelines at All

Impact: Missing critical information, lower chance of acceptance

Solution: Identify and use appropriate guideline from study planning stage

Mistake 2: Using Guidelines Only After Manuscript is Complete

Impact: May realize key data were not collected or documented

Solution: Review guidelines during study design and data collection

Mistake 3: Incomplete Checklist Completion

Impact: Missed items remain unreported

Solution: Systematically address every single checklist item

Mistake 4: Using Outdated Guidelines

Impact: Missing recent improvements in reporting standards

Solution: Always check for latest version on official guideline website

Mistake 5: Using Wrong Guideline for Study Design

Impact: Important design-specific elements not reported

Solution: Carefully match study design to appropriate guideline

Mistake 6: Not Submitting Checklist When Required

Impact: Editorial desk rejection or delays

Solution: Check journal submission guidelines and include checklist

Mistake 7: Generic Reporting Without Specificity

Impact: Insufficient detail for replication or appraisal

Solution: Provide specific, detailed information for each item

Journal Requirements

Many Journals Now Require:

1. Statement of adherence to reporting guidelines in Methods
2. Completed checklist uploaded as supplementary file
3. Page/line numbers on checklist indicating where items are addressed
4. Flow diagrams as figures in manuscript

Example Methods Statement:

"This study is reported in accordance with the Strengthening the Reporting of
Observational Studies in Epidemiology (STROBE) statement. A completed STROBE
checklist is provided as Supplementary File 1."

Journals with Strong Requirements:

• PLOS journals (require checklists for specific designs)
• BMJ (requires CONSORT, PRISMA, and others)
• The Lancet (requires adherence statements)
• JAMA and JAMA Network journals (require checklists)
• Nature portfolio journals (encourage guidelines)

Resources

Official Guideline Websites

• EQUATOR Network: https://www.equator-network.org/
• CONSORT: http://www.consort-statement.org/
• STROBE: https://www.strobe-statement.org/
• PRISMA: http://www.prisma-statement.org/
• SPIRIT: https://www.spirit-statement.org/
• ARRIVE: https://arriveguidelines.org/
• CARE: https://www.care-statement.org/

Training Materials

• EQUATOR Network provides webinars and training resources
• Many guidelines have explanatory papers published in medical journals
• Universities often provide workshops on reporting guidelines

Software Tools

• Some reference managers can insert reporting guideline citations
• Covidence, RevMan for systematic review reporting
• PRISMA flow diagram generator: http://prisma.thetacollaborative.ca/

Checklist: Using Reporting Guidelines

Before starting your study:

• Identified appropriate reporting guideline(s)
• Reviewed checklist items requiring prospective planning
• Designed study to capture all required elements
• Registered protocol if applicable

During manuscript drafting:

• Downloaded latest version of guideline checklist
• Systematically addressed each checklist item
• Created required flow diagram
• Noted where each item is addressed (page/line)

Before submission:

• Completed formal checklist with page numbers
• Verified all items adequately addressed
• Included adherence statement in Methods
• Prepared checklist as supplementary file if required
• Checked journal-specific requirements
• Mentioned guideline adherence in cover letter

Venue-Specific Reporting Requirements

Reporting Standards by Venue Type

ML Conference Reporting Standards

NeurIPS/ICML/ICLR reproducibility requirements:

• Datasets: Names, versions, access methods, preprocessing
• Code: Availability statement; GitHub common
• Hyperparameters: All settings reported (learning rate, batch size, etc.)
• Seeds: Random seeds for reproducibility
• Computational resources: GPUs used, training time
• Statistical significance: Error bars, confidence intervals, multiple runs
• Broader Impact statement (NeurIPS): Societal implications

What to include (typically in appendix):

• Complete hyperparameter settings
• Training details and convergence criteria
• Hardware specifications
• Software versions (PyTorch 2.0, etc.)
• Dataset splits and any preprocessing
• Evaluation metrics and protocols

Enforcement and Evaluation

What gets checked:

• Medical journals: Checklist uploaded; adherence statement in Methods; systematic completeness
• PLOS/BMC: Mandatory checklists for certain designs; reproducibility emphasized
• High-impact: Methods sufficiency for replication (checklist often not required)
• ML conferences: Reproducibility checklist (NeurIPS); code availability increasingly expected

Common issues leading to rejection:

• Missing required checklists (medical journals)
• Insufficient methods detail for reproduction
• Missing key information (randomization, blinding, power calculation)
• No data/code availability statement when required

Methods statement examples:

Journal (STROBE):

This study followed STROBE reporting guidelines. Checklist provided in Supplement 1.

ML conference (reproducibility):

Code available at github.com/user/project. All hyperparameters in Appendix A.
Training used 4×A100 GPUs (~20 hours). Seeds: {42, 123, 456}.

Pre-Submission Reporting Checklist

For clinical trials (medical journals):

• CONSORT checklist complete with page numbers
• Trial registration number in abstract and methods
• CONSORT flow diagram included
• Statistical analysis plan described
• Adherence statement in Methods

For observational studies (medical/epidemiology):

• STROBE checklist complete
• Study design clearly stated
• Statistical methods detailed
• Confounders addressed
• Adherence statement in Methods

For systematic reviews:

• PRISMA checklist complete
• PRISMA flow diagram included
• Protocol registered (PROSPERO)
• Search strategy documented
• Risk of bias assessment included

For ML conference papers:

• All datasets named with versions
• Code availability stated (GitHub link if available)
• Hyperparameters listed (appendix acceptable)
• Random seeds reported
• Computational resources specified
• Error bars/confidence intervals shown
• Broader Impact statement (if required)