Structure the Report Around the Reviewer's Questions
A population PK report is easiest to review when it provides a visible path from the question being addressed to the evidence supporting the answer. Reviewers need to locate the analysis objective, understand what data and methods were used, evaluate the results, and see how those results support the interpretation and conclusion. A report can contain technically sound analyses yet be difficult to assess if these pieces are scattered, inconsistently labeled, or separated by unclear references.
There is no single section order that is mandatory for every submission. The report should follow applicable regional requirements and the submission's established document structure. Within that framework, organize material so that essential reasoning is in the main text and detailed supporting material is findable without interrupting the narrative. The earlier material on regulatory questions, analysis plans, data cutoffs, model development decisions, uncertainty, diagnostics, sensitivity analyses, limitations, and exposure simulations provides the analysis content to report; the task here is to arrange that content for review.
A useful organizing principle is to keep the main report in the same conceptual order as the analysis argument: why the analysis was done; what data and methods were used; what it found; how those findings should be interpreted; and what conclusion is supported. This does not mean every technical detail belongs in the main text. It means that important decisions and conclusions should not require a reviewer to infer the logic from disconnected appendices.
A Practical Section Architecture
A report can adapt the following architecture to its purpose and the required submission format. Short studies may combine sections; complex analyses may need subsections. Keep section numbering and labels stable so that references remain precise.
| Report component | Review question it should answer | Typical content |
|---|---|---|
| Synopsis or executive summary | What was asked, what was done, and what is the main finding? | Objective, analysis scope, key result, qualification, and conclusion |
| Objectives and context | Why was this analysis needed? | Regulatory or clinical question, intended use, and relevant context |
| Data and methods | What evidence and procedures produced the results? | Analysis population, data cutoff, data handling, planned methods, and relevant implementation details |
| Results | What did the analysis show? | Model development outcome, parameter estimates and uncertainty, covariate findings, diagnostics, sensitivity results, and simulations as relevant |
| Interpretation and limitations | How much weight can be placed on the findings? | Meaning in context, uncertainty, limitations, and applicability |
| Conclusion | What answer is supported? | A concise response to each objective, qualified to match the evidence |
| References and appendices | Where can supporting details be checked? | Source documents, detailed tables and figures, technical output, and other supplementary material |
The synopsis is a map, not a substitute for the report. It should state the principal finding and any important qualification, but it should also point the reader toward the relevant detailed results. Methods should be sufficiently clear to understand how the reported estimates were obtained, while detailed implementation information can be provided in an appendix when it is too extensive for the main narrative. The results should present evidence before drawing a broad interpretation. The conclusion should answer the stated objectives rather than introduce a new claim.
Give each objective a concise identifier or heading. Define the scope of the answer, including the population or setting it concerns.
Describe the data and analysis decisions needed to understand the evidence, and identify where fuller details are provided.
Place key numerical findings and diagnostic evidence in clearly titled, numbered tables or figures, with units and relevant context.
Explain what the evidence supports, acknowledge material limitations, and give a conclusion that answers the original question without exceeding the evidence.
Make the Evidence Easy to Find and Evaluate
A table or figure should do a defined job. Use a table when readers need to compare exact values, groups, estimates, or uncertainty measures. Use a figure when a pattern, distribution, relationship, or diagnostic feature is easier to see than to describe. Avoid reproducing the same full set of values in prose and a table. Instead, the prose should direct attention to the important pattern and cite the exhibit containing the detail.
Number tables, figures, and appendices consistently, and give each a descriptive title. A title such as “Population parameter estimates” is more useful than “Table 4”; a table title should make clear what population or analysis it represents when that distinction matters. Define abbreviations and symbols in a note, specify units, and identify the analysis set or data cutoff when necessary. If a result is central to a conclusion, it should be visible in the main report or be directly and unambiguously referenced there.
Appendices are valuable for material that supports verification but would make the main argument cumbersome: extended parameter listings, supplementary diagnostic plots, detailed data descriptions, or implementation records. They should not become an unindexed repository. The main text should explain why a supporting item matters and give a precise reference, such as “Appendix B, Figure B.3,” rather than “see appendix.” If an appendix contains several related items, provide a contents list or a clear internal organization.
Worked Example: Linking an Objective to Its Evidence
Consider a hypothetical analysis intended to assess whether body weight is associated with apparent clearance in adults receiving a study medicine. The report has one objective: “Characterize the association between body weight and apparent clearance in the analysis population.” The hypothetical analysis finds an estimated 12% higher apparent clearance at 90 kg than at 60 kg, with a 95% confidence interval for the relative change from 2% to 23%.
First, give the objective a stable label, such as Objective 1. In the methods, state the relevant analysis population and the prespecified approach used to assess the association. The result belongs under a matching subsection, “Objective 1 results,” and should identify the estimated contrast, its interval, and the supporting table or figure. For example, Table 6 could report the covariate estimate and uncertainty, while Figure 8 could show the relevant diagnostic or model-based relationship if that display supports the interpretation.
The interpretation should say that the fitted analysis estimated higher apparent clearance at the higher body weight, while acknowledging the interval and the limits of the analysis. It should not say that weight caused clearance to increase. The conclusion should answer the objective at the same level: “In this analysis, body weight was associated with apparent clearance; the estimated contrast and uncertainty are reported in Table 6.” Thus, a reader can trace Objective 1 to a method, a result, and a qualified conclusion without searching for an unstated link.
The example's numerical result is hypothetical. Its purpose is to show the reporting chain: the objective determines what method and evidence must be identifiable, and the evidence constrains the conclusion.
Worked Reporting Decisions
Worked Example: Choosing Main-Text and Appendix Material
Suppose a hypothetical report contains 24 parameter estimates, a concise summary of 4 parameters central to the primary objective, and 18 diagnostic plots. A reviewer needs to see the key estimates and assess whether the diagnostics support the model, but placing every value and plot in the main text would obscure the main findings.
Put the 4 objective-relevant estimates in a main-text table with units and uncertainty. Put the complete set of 24 estimates in an appendix table, and organize the 18 plots into a numbered appendix series with descriptive captions. In the results text, state the principal diagnostic conclusion and cite the relevant plots by number; do not claim that diagnostics were satisfactory without indicating which evidence supports that statement.
This division retains essential evidence in the main report while making complete supporting material available. It is not a rule that exactly 4 estimates belong in a main-text table. The decision depends on the objectives and the reviewer's need to assess the result. If one of the other 20 estimates is material to interpretation, it should be surfaced in the main text rather than hidden solely because it is not among the initial 4.
Worked Example: Auditing Cross-References Before Finalization
A hypothetical report has five conclusions. A traceability check records whether each conclusion has an identified objective, a methods section, a results exhibit or clearly stated result, and a limitations discussion where relevant. Four conclusions have all required links; one conclusion about a simulated exposure comparison has no cited results exhibit.
Define coverage for this audit as the number of conclusions with all required links divided by the total number of conclusions. Here:
The 80% is a document-check result, not evidence that the analysis is 80% valid. The uncovered conclusion identifies a specific repair: add the relevant result or exhibit reference, or revise the conclusion if the evidence does not support it. After the repair, check all five links again. This simple audit is useful because it tests whether a report's argument can be followed; it does not replace scientific review of the analysis.
Deriving a Traceable Report
The reason to organize by objective can be stated as a simple completeness argument. Let an objective be supported only when the report makes its method, evidence, and interpretation identifiable. If any link is missing, the reader cannot evaluate the full argument for that objective from the report. Therefore, for each objective, the report needs a path from the question to the conclusion:
This is not a statistical theorem or a score of analytical quality. It is a document-design justification: a conclusion that cannot be traced to results is difficult to verify, and results that cannot be connected to an objective may not answer the question of interest. A traceability matrix makes the links explicit. For each objective, list the methods subsection, principal result or exhibit, interpretation, and conclusion location. Missing cells reveal a report-structure problem to resolve before submission.
Diagnostics, Failure Modes, and Limitations
Reviewability can fail even when every expected section is present. Common warning signs include:
- Objectives are broad or absent. Results then appear as a catalogue rather than evidence addressing a defined question.
- Methods are detached from results. A reader cannot tell which population, data cutoff, or analysis decision produced a displayed estimate.
- Exhibits are difficult to interpret independently. Missing units, undefined abbreviations, unclear group labels, or an unspecified analysis set force readers to guess.
- Interpretation outruns evidence. A qualified estimate is described as definitive, or an association is expressed as a causal effect.
- Appendices are hard to navigate. Vague references and inconsistent numbering make supporting evidence functionally unavailable.
- The conclusion repeats the synopsis without answering the objectives. A concise conclusion should state what each objective's evidence supports and any important qualification.
A table of contents, stable numbering, descriptive captions, and a final cross-reference audit can detect navigation and traceability problems. These checks do not establish that the data or analysis are correct. A well-organized report can still contain flawed methods, and a technically sound result can still be interpreted too broadly. Report structure improves the visibility of evidence; it does not substitute for evaluating that evidence.
There are trade-offs. Concision can improve navigation, but excessive compression may hide decisions or uncertainty needed for evaluation. Completeness can support verification, but reproducing every technical detail in the main narrative can bury the findings. A suitable balance keeps the objective, essential methods, principal results, interpretation, limitations, and conclusion in the main report, while placing supplementary detail in clearly indexed appendices. Adapt that balance to the analysis complexity and the applicable submission requirements.
Check Your Understanding
Answer each question, then compare with the solutions below.
- A report gives an important conclusion but cites no result or exhibit. What is the structural problem, and what should be checked?
- A table contains estimates but no units or description of the analysis population. Why does this impair review, and what information should be added?
- A report has 6 objectives, and the traceability audit finds complete objective-to-conclusion links for 5. What is the coverage, and what does that number mean?
- When should detailed diagnostic plots be placed in an appendix rather than the main text?
Solutions. (1) The conclusion is not traceable to evidence. Identify the objective and results that support it; add a precise result or exhibit reference, or revise the conclusion if support is absent. (2) Readers cannot interpret the numerical scale or determine which analysis produced the estimates. Add units, define labels and abbreviations, and state the analysis population in the title, heading, or a clear table note. (3) Coverage is \(5/6 \times 100\%=83.3\%\). It describes completion of this document-link audit, not the validity or accuracy of the analysis; the remaining objective needs review. (4) Put supplementary plots in an indexed appendix when they support verification but would interrupt the main narrative. Keep key diagnostic evidence or a clear summary in the main report when it is central to evaluating a principal result, and provide precise references to the additional plots.