Section-by-section guide for writing CIHR Project Grant applications, covering RCTs, AI/technology, observational studies, with scoring rubrics and templates. Use when drafting, reviewing, or scoring a CIHR grant application.
Installation
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Section-by-section guide for writing CIHR Project Grant applications, covering RCTs, AI/technology, observational studies, with scoring rubrics and templates. Use when drafting, reviewing, or scoring a CIHR grant application.
CIHR Grant Writing Skill
Overview
A section-by-section guide for writing CIHR Project Grant applications, modeled on the structure of the PANTHEON SLIM (Randomized Controlled Trial) grant and cross-referenced with the CardioAgent (AI/Technology) grant. This skill is generalizable across clinical trials, observational studies, AI/technology development, and other health research study designs.
Each section includes: Requirements (what must be present), Expectations (what reviewers look for), Scoring Rubric (weighted scoring criteria), and Templates (fill-in-the-blank scaffolds).
For CIHR-specific language tips, study type adaptations, and common reviewer critiques, see the appendices.
SECTION 1: The Need for a Trial / Study
Weight: 25/100
This section must build the case that your study addresses a critical, unmet need. It follows a hierarchical argument structure.
1.1 What is the problem to be addressed?
Weight: 5/25
Requirements
State the clinical/scientific problem in 2-3 sentences maximum
Define the specific patient population or target group
Identify what is currently unknown or what practice gap exists
State why current evidence is insufficient (e.g., "no RCT has addressed...", "current AI tools lack...")
Use underline/bold formatting to emphasize the key population and the key gap
Expectations
The problem statement should be immediately understandable to a non-specialist reviewer
Frame as a patient-centered or health-system problem, not just a scientific curiosity
Quantify the problem where possible (mortality rates, error rates, prevalence)
Explicitly state the knowledge gap that this study will fill
Template
The problem to be addressed is that [specific clinical/scientific gap] in [target population] is unknown/unresolved. [Current approach or standard of care] represents [limitation], but [it has only been studied in X / no evidence supports Y / current tools cannot Z]. [Quantify the consequence of not addressing this gap].
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Clarity of problem
2/5
Immediately clear, specific, and compelling
Understandable but somewhat vague
Confusing or overly broad
Evidence of gap
2/5
Systematic review or guideline gap cited; no existing RCT/study identified
Literature cited but gap not well-defined
No evidence that gap exists
Patient/population focus
1/5
Specific population clearly defined with prevalence data
Population mentioned but not well-characterized
No clear target population
1.2 What are the principal research questions to be addressed?
Weight: 5/25
Requirements
State the overarching objective in one sentence (use the study acronym if applicable)
State the central hypothesis clearly
Define the primary objective with a feasibility/success threshold (for pilot) or primary endpoint (for phase III)
List secondary objectives (numbered, 2-4 items)
List exploratory objectives (numbered, 2-5 items)
For each objective level (primary/secondary/exploratory), state whether results will be evaluated by sex and gender subgroups
For pilot/feasibility studies: primary objective MUST be operational (e.g., recruitment rate), not clinical
Expectations
Objectives must follow a clear hierarchy: Primary > Secondary > Exploratory
The primary objective must be answerable with the proposed design and sample size
For pilot studies: clinical outcomes are exploratory only (state this explicitly)
Sex and gender analysis commitment should be present at every objective level
Use CIHR-specific language: "patient-oriented," "sex and gender-based analysis (SGBA+)"
Template
Overarching objective: The overarching objective of the "[STUDY ACRONYM]" [study type] is to [determine/evaluate/develop] [intervention/technology] [in what population] [to achieve what].
Central hypothesis: We hypothesize that [intervention/approach] will [expected effect] compared with [comparator] in [population].
Primary objective: To [specific measurable objective]. Success threshold: [define].
Secondary objectives:
i. To [objective 1];
ii. To [objective 2].
Exploratory objectives:
i. To [objective 1];
ii. To [objective 2];
iii. To [objective 3].
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Hypothesis clarity
2/5
Testable, specific, directional hypothesis with clear comparator
Hypothesis present but vague
No hypothesis or untestable statement
Objective hierarchy
2/5
Clear primary with feasible threshold; logical secondary and exploratory
Objectives present but hierarchy unclear
Objectives conflated or missing levels
SGBA+ integration
1/5
Sex/gender analysis specified at each objective level
Mentioned but not integrated into objectives
Absent
1.3 Why is a trial/study needed now?
Weight: 10/25
This is the most critical subsection of the entire "Need" section. It builds the scientific rationale through a logical chain of evidence. Use numbered subsections (1.3.1, 1.3.2, etc.).
Requirements
1.3.1 Burden of disease/problem: Canadian and global epidemiological data with citations
1.3.2 Current standard of care/approach: What is currently done and why it works (partially)
1.3.3 Limitations of current approach: Why the standard of care is insufficient (quantify risks, error rates, costs)
1.3.4 Emerging evidence for the proposed approach: What new evidence suggests a better strategy exists
1.3.5 Evidence gap in the target population: Demonstrate that the proposed approach has NOT been studied in your specific population (cite systematic review or guideline gap)
1.3.6 Specific technical/clinical gap: Address any unique safety or feasibility concerns
1.3.7 Sex and gender representation gap: Demonstrate that prior studies lacked adequate sex/gender representation and analysis
Use bold/underline for the key concluding statement of the evidence gap
Include at least one citation from your own team's prior work in this field
Expectations
Build the argument like a legal brief: each subsection leads logically to the next
Use Canadian data prominently (CIHR is a Canadian funder) or primarily
Demonstrate clinical equipoise with real-world data (e.g., practice variation showing disagreement)
The final statement should be a bold, underlined, definitive assertion of the knowledge gap
Reference your own prior work to show you are the right team to address this gap
For AI/technology studies: address current limitations of existing AI approaches and why yours is different
Template for Key Concluding Statement
There is thus an important knowledge gap to support the use of [proposed approach] in [target population], despite the fact that this population represents [quantify the size/importance of the population].
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Burden of disease (Canadian context)
2/10
Canadian-specific epidemiology with recent data and economic impact
Some Canadian data but incomplete
No Canadian data or only global
Logical argument chain
3/10
Each subsection flows naturally to the next; unavoidable conclusion
Generally logical but some jumps
Disjointed; conclusion not supported
Evidence of gap
3/10
Systematic review or guideline development process cited showing no existing evidence
Literature review shows gap but not systematic
Assertion without evidence
Own prior work cited
1/10
Multiple relevant team publications integrated into rationale
One team publication mentioned
No team publications cited
Sex/gender gap identified
1/10
Specific data on under-representation with quantification
General statement about sex/gender gaps
Not mentioned
1.4 How will the results of this trial/study be used?
Weight: 3/25
Requirements
For pilot studies: state that results will inform the design of a phase III confirmatory study
For confirmatory studies: state that results will influence clinical guidelines
Describe the dissemination plan: conferences, journals, social media, public engagement
State whether the study could transition to the next phase seamlessly (adaptive design)
Explain why CIHR funding is required (e.g., no industry interest due to generic drugs / public health focus)
Name the specific guideline body or clinical practice that would be impacted
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Impact pathway
2/3
Clear pipeline from pilot to phase III to guidelines with named bodies
Comprehensive, evidence-based risk assessment with mitigation
Risks mentioned with some mitigation
Risks not addressed or dismissed
Safety monitoring
1/2
DSMB, adverse event protocol, stopping rules defined
Some monitoring described
No safety monitoring plan
SECTION 2: The Proposed Trial / Study Design
Weight: 40/100
This section covers all methodological details. Each subsection maps to a specific CIHR review criterion.
2.1 What is the proposed trial/study design?
Weight: 5/40
Requirements
State the study design using standard terminology (e.g., "pilot, multi-center, double-blinded, pragmatic, patient-centered RCT")
State whether it is a pilot/feasibility study or a confirmatory study
If pragmatic: reference PRECIS-2 tool and justify pragmatic elements
Describe patient engagement strategy: patient partners on steering committee, co-development of protocol
For AI/technology: describe the validation framework (retrospective + prospective phases)
Include a study flowchart figure
Template
The proposed [study] is a [phase], [number of sites]-center, [blinding], [pragmatic/explanatory], and [patient-centered] [study type] designed to [primary purpose]. [Population] will be eligible. [Brief description of randomization/allocation]. The [pragmatic/explanatory] design ensures that [justification].
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Design appropriateness
2/5
Design perfectly matches research question; standard terminology used
Appropriate design but not fully justified
Design-question mismatch
Patient engagement
2/5
Named patient partner on steering committee; co-developed protocol; EDI principles
Patient input mentioned but not structured
No patient engagement
Study flowchart
1/5
Clear, comprehensive flowchart with all study phases and timelines
Flowchart present but incomplete
No flowchart
2.2 What are the planned trial interventions?
Weight: 4/40
Requirements
Describe the experimental intervention with dose/frequency/route/duration
Describe the control intervention with the same level of detail
Justify the choice of comparator with evidence and guideline references
Address regulatory requirements (e.g., Health Canada Clinical Trial Application)
Describe management of participants on prior therapies (switching protocols)
State that all other treatments follow standard of care (pragmatic principle)
Describe post-study care plan
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Intervention clarity
2/4
Fully specified with dose, route, frequency, duration for both arms
Mostly specified but some gaps
Vague intervention description
Comparator justification
1/4
Evidence-based with guideline reference; addresses evidence gaps in comparator choice
Some justification
No justification for comparator
Regulatory and practical
1/4
Regulatory pathway identified; switching protocols; post-study care
Some practical issues addressed
Regulatory/practical issues ignored
2.3 Allocation to trial groups
Weight: 2/40
Requirements
State allocation ratio (e.g., 1:1)
State stratification variables with justification (cite validation of stratification tool)
State block sizes
Name the randomization platform/system
For non-RCT designs: describe sampling or allocation strategy
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Randomization rigor
2/2
Validated stratification tool cited; appropriate block sizes; named platform
Randomization described but not fully detailed
No randomization details
2.4 Methods for protecting against sources of bias
Weight: 2/40
Requirements
Describe blinding strategy (who is blinded: participants, investigators, outcome assessors)
Describe allocation concealment method
For AI studies: describe blinding of human evaluators to AI outputs
Address potential sources of bias specific to your design
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Bias protection
2/2
Double-blind with allocation concealment; all bias sources addressed
Single-blind or partial concealment
Open-label without justification
2.5 Inclusion/exclusion criteria
Weight: 3/40
Requirements
List inclusion criteria as bullet points
List exclusion criteria as bullet points
For pragmatic trials: explicitly state that few exclusion criteria are used to maximize generalizability
Justify any exclusion criterion that removes a specific subpopulation
Describe screening log data collection: baseline characteristics, sex, gender, race, ethnicity, reasons for exclusion
Address EDI: describe strategies to include underrepresented populations (e.g., Indigenous communities)
For heterogeneous populations: acknowledge heterogeneity and state how subgroups will be characterized
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Criteria appropriateness
1/3
Pragmatic criteria maximizing generalizability; each exclusion justified
Reasonable criteria but some unjustified exclusions
Overly restrictive or unjustified
EDI integration
1/3
Named strategies for diverse recruitment; screening log captures demographics
EDI mentioned but no concrete strategy
No EDI consideration
Population characterization
1/3
Heterogeneity acknowledged; subgroup plan described
Some acknowledgment
Assumed homogeneity
2.6-2.7 Treatment duration and follow-up
Weight: 2/40
Requirements
State treatment duration with start and end points
State follow-up visit schedule with specific time points
State visit modalities (in-person, telephone, video)
State total study duration (recruitment + follow-up)
State recruitment period duration
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Timeline completeness
2/2
All durations specified; flexible visit modalities; realistic timeline
Mostly specified
Incomplete or unrealistic
2.8 Primary and secondary outcome measures
Weight: 6/40
Requirements
Primary outcome: State with exact definition, measurement method, and success threshold
Secondary outcomes: List each with definition (numbered, 2-5 items)
Exploratory outcomes: List each with definition (numbered, 3-6 items)
For pilot studies: primary outcome MUST be feasibility-related (recruitment rate, adherence, etc.)
For clinical trials: use standardized endpoint definitions (e.g., Academic Research Consortium, BARC bleeding)
For AI studies: define accuracy metrics (AUROC, F1, sensitivity, specificity) and reference standards
Include patient-oriented outcomes
For novel endpoints: describe the development methodology (e.g., discrete-choice experiment)
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Primary outcome definition
2/6
Precisely defined with validated measurement method and threshold
Defined but threshold unclear
Vague or inappropriate primary outcome
Outcome hierarchy
2/6
Clear primary/secondary/exploratory with appropriate scope at each level
Hierarchy present but some misclassification
No hierarchy or outcomes conflated
Standardized definitions
1/6
All endpoints use published consensus definitions with citations
Most endpoints standardized
Custom definitions without justification
Patient-oriented outcomes
1/6
Named patient-oriented outcomes with development methodology
Patient outcomes mentioned
No patient-oriented outcomes
2.9 How will outcomes be measured at follow-up?
Weight: 2/40
Requirements
State the data source for each outcome category (screening logs, medical charts, self-report, imaging)
State whether endpoints will be adjudicated (and by whom) or not adjudicated (justify for pilot)
Describe adverse event and serious adverse event monitoring
For AI studies: describe the reference standard / ground truth generation process (e.g., central reader model)
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Measurement rigor
2/2
Each outcome has specified data source; adjudication plan clear; AE monitoring
Mostly specified
Measurement methods unclear
2.10 Sample size justification
Weight: 4/40
Requirements
For pilot/feasibility studies: state explicitly that formal sample size calculation is not applicable; provide expected enrollment range based on recruitment assumptions
For confirmatory studies: provide full power calculation with alpha, beta, effect size, and assumptions
State the basis for effect size assumptions (prior studies, pilot data, clinical significance)
Address multiple comparisons if applicable
For AI studies: justify the number of cases for training/validation/testing; describe stratified sampling for rare conditions
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Sample size appropriateness
2/4
Fully justified with transparent assumptions; sensitivity analyses
Calculation present but assumptions not fully justified
No calculation or unrealistic assumptions
Effect size basis
2/4
Based on own pilot data or meta-analysis of comparable studies
Based on literature but indirect evidence
Arbitrary or unjustified effect size
2.11-2.14 Practical considerations
Weight: 4/40
Covers: health service research issues, recruitment, compliance, and loss to follow-up.
Requirements
Recruitment: State expected rate per site per month with evidence; describe recruitment process; state total expected enrollment
Compliance: Describe adherence monitoring strategy; cite expected adherence/discontinuation rates from prior trials; describe patient-partner involvement in adherence strategies
Loss to follow-up: State expected rate with evidence; describe retention strategies; commit to identifying barriers
For all three: state that sex/gender disparities will be monitored and mitigated
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Recruitment feasibility
2/4
Evidence-based rate; named sites with capacity; prior recruitment experience
Rate stated but evidence weak
No evidence of feasibility
Adherence/retention plan
2/4
Proactive monitoring with iterative strategies; patient partner involvement; sex/gender analysis
Basic plan
No plan
2.15 Number of centers
Weight: 1/40
Requirements
List all centers with site PI name and role
Include geographic diversity (for Canadian multi-center: multiple provinces)
Reference letters of support from each site
For international sites: describe regulatory coordination plan
2.16-2.18 Analysis plan
Weight: 5/40
Requirements
Type of analyses: For pilot: explicitly state "most analyses will be descriptive"; for confirmatory: state primary statistical test
Describe analysis for each outcome level (primary, secondary, exploratory)
State whether intention-to-treat or per-protocol analysis (or both)
Frequency of analyses: State whether interim analyses are planned; describe DSMB access to data
Subgroup analyses: List pre-specified subgroups (sex, gender, site, risk score); state whether interaction testing will be performed
For AI studies: describe performance metrics, calibration, and fairness analyses across demographic subgroups
Scoring Rubric
Criterion
Weight
5 (Excellent)
3 (Good)
1 (Weak)
Statistical rigor
3/5
Appropriate methods for each outcome; multiple comparison handling; ITT specified
Generally appropriate but some gaps
Inappropriate methods or no plan
Subgroup/SGBA+
2/5
Pre-specified sex/gender subgroups at all objective levels; fairness metrics for AI
Some subgroup analysis planned
No subgroup analysis
2.19 Prior pilot work / Preliminary data
Weight: 2/40
Requirements
Describe all preliminary studies that inform this proposal (surveys, observational studies, pilot data)
For each: state IRB status, funding source, and how results will inform the current study
Demonstrate that these studies were conducted independently (no overlapping funds)
For AI studies: present preliminary performance data in table format (model comparison)
State how the totality of the preliminary program will inform the proposed study
