Reporting clinical study results is a critical step in medical research publication. Even when a study has a strong design and produces valuable findings, poor reporting can make its results difficult to interpret, evaluate, or apply in clinical practice. Researchers must present their findings clearly, accurately, and objectively so that readers can understand what the study discovered and how confidently those findings can be interpreted.

The Results section is particularly important because it provides the evidence on which the paper's conclusions are based. It should communicate the study findings without exaggeration, selective reporting, or unsupported interpretation. From describing participant recruitment and baseline characteristics to presenting outcome measures, confidence intervals, statistical tests, and adverse events, every element should contribute to a transparent account of the research.

For medical students, clinical investigators, postgraduate scholars, and healthcare professionals, learning how to report clinical study results is essential for producing credible, publication-ready manuscripts. This guide explains the structure, statistical principles, tables, figures, reporting guidelines, and common mistakes to consider when preparing a medical research article in 2026. For related guidance on manuscript preparation, see How to Write the Materials and Methods Section of a Medical Research Paper.

What Is the Results Section in a Medical Research Article?

The Results section presents the findings obtained from the study methods described earlier in the manuscript. It explains what the researchers observed after collecting and analyzing the data.

Depending on the study design, the Results section may include:

  • The number of participants screened, enrolled, and analyzed
  • Baseline demographic and clinical characteristics
  • Primary and secondary outcome findings
  • Comparisons between intervention and control groups
  • Statistical estimates and confidence intervals
  • Safety outcomes and adverse events
  • Missing data, participant withdrawals, and protocol deviations
  • Findings from prespecified subgroup or sensitivity analyses

The purpose is to provide a factual account of the evidence, allowing readers to assess the findings independently.

Results vs. Discussion: Understanding the Difference

One of the most common challenges in medical manuscript writing is distinguishing between results and interpretation.

The Results section answers: What did the study find?

The Discussion section answers: What do these findings mean, and how do they relate to existing evidence?

For example:

Results: "The intervention group had a lower mean systolic blood pressure than the control group at 12 weeks, with a between-group difference of −6.2 mmHg (95% confidence interval [CI], −9.1 to −3.3)."

Discussion: "The observed reduction suggests that the intervention may improve blood pressure control in the studied population, although the clinical implications should be considered alongside the study's limitations and other available evidence."

The numerical statement belongs in the Results section. The explanation of its meaning belongs primarily in the Discussion section.

1. Begin With Participant Recruitment and Study Flow

The first part of the Results section should explain how participants progressed through the study. This is particularly important for clinical trials and observational studies involving participant screening, enrollment, follow-up, or exclusions.

Readers should be able to determine how many individuals were considered for participation, how many met the eligibility criteria, and how many contributed data to the final analysis.

Depending on the study design, report:

  • The number of individuals assessed for eligibility
  • The number excluded and the reasons for exclusion
  • The number enrolled or randomized
  • The number assigned to each study group
  • The number who received the intervention, where applicable
  • The number who completed follow-up
  • The number included in each analysis
  • The reasons for withdrawal, loss to follow-up, or exclusion from analysis

For randomized controlled trials, a participant flow diagram can communicate this information more effectively than a lengthy paragraph. The diagram should align with the manuscript text and any supplementary material.

Example:

"Of the 240 individuals assessed for eligibility, 180 met the inclusion criteria and were enrolled. Participants were allocated to the intervention group (n = 90) or control group (n = 90). During follow-up, 84 participants in the intervention group and 86 in the control group completed the study."

This is an illustrative example, not actual research data. In a real manuscript, explain the reasons for missing follow-up and distinguish participants who completed the study from those included in each outcome analysis.

A transparent account of participant flow helps readers assess potential attrition bias and understand whether the final analysis reflects the population originally recruited.

Medical Journal IMJ Health Call for Papers

2. Present Baseline Characteristics Clearly

Baseline characteristics describe the participants before the intervention or during the initial assessment period. These details help readers understand the study population and assess the comparability of groups where relevant.

Depending on the research question, baseline characteristics may include:

  • Age and sex
  • Relevant medical history
  • Disease duration and severity
  • Existing comorbidities
  • Medication use
  • Laboratory measurements
  • Lifestyle factors
  • Other prognostic variables relevant to the primary outcome

For continuous variables, report the mean and standard deviation when appropriate, or the median and interquartile range when the distribution or study context supports that summary. For categorical variables, provide counts and percentages.

Example:

"At baseline, the mean age of participants was 54.2 years (standard deviation, 10.6), and 52% were female. The median duration of diagnosed hypertension was 6 years (interquartile range, 3–10 years)."

Use actual values from the study dataset and select summaries appropriate to the data.

Should Baseline Differences Be Tested Statistically?

In randomized trials, significance tests comparing baseline characteristics are generally discouraged because randomization, rather than a nonsignificant p-value, is the basis for expecting groups to be comparable. Descriptive summaries are usually more informative.

For observational studies, baseline differences may reveal important differences between exposure groups. Researchers should describe these differences and explain how relevant confounding variables were handled in the analysis.

Avoid suggesting that groups are equivalent simply because a baseline comparison did not reach statistical significance.

3. Report Primary Outcomes First

The primary outcome is the principal endpoint used to address the study's main research question. It should receive clear and prominent reporting in the Results section.

Before reporting the result, ensure that the reader can identify:

  • The outcome that was measured
  • The time point at which it was assessed
  • The number of participants contributing data
  • The result in each group, where applicable
  • The estimated difference or association between groups
  • The relevant measure of uncertainty
  • The statistical analysis used, where necessary for interpretation

For clinical trials, report the between-group effect rather than relying solely on separate within-group changes. For observational studies, report the association estimate relevant to the research question, such as a risk ratio, odds ratio, hazard ratio, or regression coefficient.

Illustrative example:

"At 12 weeks, the mean reduction in systolic blood pressure was 11.4 mmHg in the intervention group and 5.2 mmHg in the control group. The estimated between-group difference in change was −6.2 mmHg (95% CI, −9.1 to −3.3)."

The example communicates the findings and their uncertainty. The sign and interpretation of an estimate must be consistent with the order in which groups are compared.

Why the Primary Outcome Should Come First

Presenting the primary outcome early helps readers understand the central finding before considering secondary endpoints or exploratory analyses. It also reduces the risk of giving disproportionate attention to statistically significant secondary findings when the primary outcome does not support the original hypothesis.

The outcome and its time point should correspond to the study protocol or analysis plan. If either changed, report the change transparently and explain it where appropriate.

4. Report Secondary Outcomes and Additional Findings

Secondary outcomes provide supplementary information about the effects, associations, or clinical characteristics under investigation.

Examples include:

  • Changes in laboratory values
  • Symptom severity
  • Functional status
  • Quality of life
  • Hospital admissions
  • Treatment adherence
  • Disease progression
  • Healthcare utilization

Present secondary outcomes in a logical order, preferably following the study objectives or the sequence established in the protocol.

For each outcome, identify the measure, time point, analysis population, and relevant estimate. Where appropriate, provide confidence intervals and p-values.

Distinguish prespecified secondary outcomes from exploratory analyses conducted after examining the data. This distinction matters because exploratory findings may generate hypotheses but generally require confirmation in further research.

Avoid reporting only favorable secondary findings while omitting outcomes that did not show the expected effect.

5. Understand and Report Statistical Significance Correctly

Statistical significance is often included in medical research articles, but it must be interpreted carefully. A p-value does not measure the size of an effect, the probability that a hypothesis is true, or the clinical importance of a finding.

Report p-values accurately

Where appropriate, report exact p-values rather than simply stating that a result was significant or nonsignificant.

For example:

  • "The between-group difference was statistically significant (p = 0.018)."
  • "The analysis did not provide strong evidence of a between-group difference (p = 0.21)."

If the p-value is extremely small, follow the journal's reporting convention, such as reporting p less than a specified threshold rather than displaying p = 0.000.

A nonsignificant result does not prove that two treatments are equivalent or that an intervention has no effect. It may reflect a small effect, limited precision, inadequate sample size, or other factors.

Report effect sizes and confidence intervals

Whenever appropriate, report the magnitude of the observed effect and its confidence interval. These help readers understand the estimated size and precision of the finding.

Depending on the study, relevant measures may include:

Difference in continuous outcomes: Mean difference.

Difference in change from baseline: Between-group difference in change.

Binary clinical outcome: Risk difference, risk ratio, or odds ratio.

Time-to-event outcome: Hazard ratio.

Diagnostic performance: Sensitivity, specificity, and likelihood ratios.

Association between variables: Correlation or regression coefficient.

Prognostic model performance: Discrimination and calibration measures.

Select measures appropriate to the study design and analysis plan.

Statistical significance is not clinical significance

A small difference can be statistically significant in a large study but may not be meaningful to patients. Conversely, a potentially important difference may have a wide confidence interval in a small study.

When relevant, compare the observed effect with a clinically meaningful threshold established for the outcome. Avoid calling a treatment clinically effective solely because the p-value falls below a conventional significance threshold.

6. Report Safety Outcomes and Adverse Events

Safety reporting is particularly important in clinical trials, drug studies, device evaluations, and investigations involving medical procedures.

Describe the adverse events and other safety outcomes that were collected according to the protocol. Depending on the study, this may include:

  • The number and proportion of participants experiencing at least one adverse event
  • The number and types of adverse events
  • Serious adverse events
  • Events considered related to the intervention, where assessed
  • Treatment discontinuations due to adverse events
  • Deaths and other prespecified safety endpoints
  • Relevant laboratory abnormalities

Clearly identify the denominator used for percentages. For example, a percentage may be calculated among all randomized participants, those who received treatment, or those who completed a particular assessment. These denominators are not interchangeable.

If participants can experience multiple events, distinguish the number of participants with events from the total number of events.

Example:

"At least one adverse event was reported in 12 participants in the intervention group and 9 participants in the control group. Two participants in the intervention group discontinued treatment because of adverse events."

This is an illustrative reporting format. Actual manuscripts should include relevant denominators, event definitions, and the required level of detail.

Do not describe an intervention as safe merely because no serious adverse events were observed. The sample size, duration of follow-up, event ascertainment, and statistical uncertainty affect how confidently safety can be assessed.

7. Address Missing Data, Withdrawals, and Protocol Deviations

Missing data can influence the reliability and interpretation of clinical study findings. Researchers should report the extent of missing information and explain how it was handled.

Relevant information may include:

  • The number of participants with missing outcome data
  • The timing and reasons for missing assessments, where known
  • The number lost to follow-up
  • Withdrawals and their documented reasons
  • The analysis population used for each outcome
  • The methods used to address missing data
  • Sensitivity analyses evaluating the impact of key assumptions, where appropriate

Avoid automatically treating missing data as evidence of treatment failure or assuming that participants lost to follow-up had outcomes similar to those who remained in the study.

The appropriate approach depends on the study design, the amount and pattern of missingness, and the assumptions underlying the analysis. Researchers should explain whether they used complete-case analysis, multiple imputation, weighting, or another method and provide sufficient information to assess its suitability.

For randomized trials, report the analysis population and explain whether the analysis followed the intention-to-treat principle, a modified intention-to-treat definition, or another prespecified approach. Do not use these terms interchangeably.

Protocol deviations and changes to the analysis plan should also be reported transparently when relevant. Readers need to distinguish planned analyses from decisions made after the data were examined.

8. Use Tables and Figures to Present Results Effectively

Tables and figures make complex results easier to interpret, especially when a study includes several outcomes, time points, or comparison groups.

However, visual presentation should support the written Results section rather than duplicate every numerical detail.

When to use tables

Tables are useful for presenting:

  • Baseline demographic and clinical characteristics
  • Primary and secondary outcome estimates
  • Comparisons between study groups
  • Adverse events
  • Regression models and adjusted estimates
  • Detailed laboratory or diagnostic findings

A well-designed table should have a clear title, understandable row and column headings, appropriate units, and explanatory footnotes. Define abbreviations and indicate the meaning of statistical summaries.

Example table structure:

Outcome: Mean change in systolic blood pressure. Intervention group: −11.4 mmHg. Control group: −5.2 mmHg. Between-group estimate: −6.2 mmHg.

95% confidence interval for difference: −9.1 to −3.3.

Assessment time point: 12 weeks.

The figures are illustrative only. A real manuscript should follow the target journal's table conventions and report the appropriate sample sizes, estimates, and uncertainty measures.

When to use figures

Figures may be useful for:

  • Participant flow diagrams
  • Changes in outcomes over time
  • Kaplan–Meier survival curves
  • Forest plots
  • Diagnostic performance curves
  • Relevant laboratory or imaging findings

Choose a visual format that matches the data. For continuous outcomes, show variability or uncertainty where appropriate. For time-to-event analyses, include numbers at risk when recommended or necessary for interpretation.

Avoid manipulating axes in ways that exaggerate small differences, using unexplained symbols, or presenting decorative graphics that do not improve understanding.

Avoid duplicating information

The text should highlight the main findings, while tables and figures provide supporting detail. If a table already presents all the baseline characteristics, there is usually no need to repeat every value in the paragraph.

Always ensure that the values in the narrative, tables, figures, and supplementary files agree.

9. Follow the Appropriate Clinical Research Reporting Guideline

Reporting guidelines help authors provide the information needed to evaluate a study. The appropriate guideline depends on the design and purpose of the research.

CONSORT: Randomized controlled trials.

STROBE: Observational studies, including cohort, case-control, and cross-sectional studies.

PRISMA: Systematic reviews and meta-analyses.

STARD: Diagnostic accuracy studies.

CARE: Case reports.

ARRIVE: Animal research.

TRIPOD: Prediction model studies.

SQUIRE: Quality improvement studies in healthcare.

SPIRIT: Clinical trial protocols.

These guidelines are not interchangeable. Some are intended for completed studies, while others focus on protocols or specialized research designs. Use the current version relevant to the manuscript and check the journal's author instructions.

For example, CONSORT provides guidance for transparent reporting of randomized trials, including participant flow, outcome reporting, and analysis populations. STROBE supports reporting of observational studies, while STARD addresses the evaluation of diagnostic accuracy.

A checklist can help identify missing information, but it cannot compensate for a poorly designed study or inaccurate data.

10. Maintain Transparency When Reporting Unexpected or Negative Results

Clinical research does not always produce the outcomes researchers anticipate. An intervention may show no statistically significant difference, a secondary outcome may contradict the hypothesis, or an unexpected safety concern may emerge.

These findings should be reported objectively.

How to report negative findings

Instead of writing "The intervention had no effect," consider reporting the estimated difference, confidence interval, and p-value, then describing the result as not demonstrating a statistically significant difference if that is supported by the analysis.

The confidence interval may reveal whether the data are compatible with both a meaningful benefit and a meaningful harm. That information is often more useful than a nonsignificant p-value alone.

Avoid selective reporting

Selective reporting occurs when researchers emphasize favorable outcomes or analyses while omitting unfavorable or inconclusive findings.

To support transparent reporting:

  • Report the prespecified primary outcome regardless of statistical significance
  • Include relevant secondary outcomes
  • Distinguish prespecified analyses from exploratory analyses
  • Disclose material deviations from the study protocol or analysis plan
  • Report important safety findings
  • Avoid presenting subgroup findings as definitive when the study was not designed to support them

Unexpected findings can be scientifically valuable, but their limitations and uncertainty should be acknowledged.

11. Report Subgroup and Sensitivity Analyses Carefully

Subgroup analyses investigate whether an effect differs across categories, such as age groups, disease severity, or baseline risk. Sensitivity analyses assess whether findings change under different analytical assumptions or methodological choices.

When reporting these analyses, specify:

  • Which subgroups or alternative methods were examined
  • Whether the analyses were prespecified
  • The number of participants or observations included
  • The estimates and confidence intervals
  • Any relevant interaction tests
  • The rationale for the analysis

A statistically significant result in one subgroup and a nonsignificant result in another does not automatically establish that the treatment effect differs between those subgroups. A formal interaction analysis is generally needed to assess evidence of a difference.

Similarly, numerous exploratory analyses increase the likelihood of chance findings. Such results should be described cautiously and, where appropriate, presented as hypothesis-generating.

12. Ensure Consistency Between the Results, Methods, and Abstract

The Results section should align with the study protocol, statistical analysis plan, Methods section, abstract, tables, and figures.

Before submission, verify that:

  • Every reported outcome is described in the Methods section
  • The primary outcome matches the stated research objective
  • Sample sizes and analysis denominators are consistent
  • Statistical estimates and confidence intervals match the underlying analysis
  • Units and decimal places are consistent
  • The abstract accurately reflects the main findings
  • Any changes to the planned analyses are disclosed
  • Conclusions are not stronger than the results justify

The abstract deserves particular attention because it is often the first part of a paper readers encounter. It should communicate the most important findings without replacing estimates with vague statements such as "the results were significant."

A final numerical audit can help prevent transcription errors. Researchers should verify manuscript values against the final statistical output or validated analysis dataset rather than relying on manually copied figures.

13. Common Mistakes to Avoid When Reporting Clinical Study Results

Several reporting errors can reduce the clarity and credibility of a medical research article.

Reporting only p-values

A p-value alone does not communicate the size or precision of an effect. Include effect estimates and confidence intervals where appropriate.

Confusing within-group and between-group differences

A significant change in one group and a nonsignificant change in another does not prove that the groups differ. Report the appropriate between-group comparison or interaction estimate.

Omitting denominators

Percentages are difficult to interpret without knowing the number of participants or observations on which they are based. Provide denominators where relevant.

Overstating nonsignificant findings

A nonsignificant result does not establish equivalence, absence of an effect, or lack of clinical benefit. Interpret the estimate and confidence interval in context.

Ignoring missing data

Failing to describe withdrawals or missing outcomes can conceal potential sources of bias. Report the extent of missingness and the methods used to address it.

Presenting exploratory findings as confirmatory

Post hoc analyses may be useful, but they should be identified as exploratory and interpreted cautiously.

Repeating all table values in the text

Excessive repetition makes the Results section unnecessarily long. Use the narrative to highlight key findings and let tables or figures provide the detailed data.

Reporting conclusions in place of findings

Statements such as "the treatment was highly effective" are insufficient without supporting numerical results. Present the evidence first and reserve broader interpretation for the Discussion.

14. A Practical Template for Writing the Results Section

The following structure can help organize a clinical research manuscript.

1. Participant Flow – Report screening, enrollment, allocation, follow-up, exclusions, and analysis populations.

2. Baseline Characteristics – Summarize the participants' demographic and relevant clinical characteristics.

3. Primary Outcome – Present the primary outcome at the prespecified time point, including the effect estimate and uncertainty.

4. Secondary Outcomes – Report additional prespecified outcomes in a logical order.

5. Safety Outcomes – Describe adverse events, serious adverse events, and relevant discontinuations when applicable.

6. Missing Data and Additional Analyses – Explain relevant missingness, sensitivity analyses, subgroup analyses, and protocol deviations.

This structure should be adapted to the research design. A diagnostic study, case report, observational study, or systematic review may require a different sequence.

15. Final Checklist Before Submitting a Medical Research Article

Use the following checklist to assess whether the Results section is complete and publication-ready.

  • Participant flow and analysis populations are clearly reported
  • Baseline characteristics are summarized appropriately
  • The primary outcome is presented prominently
  • Secondary outcomes are reported transparently
  • Effect estimates and confidence intervals are included where appropriate
  • Statistical tests and p-values are reported accurately
  • Clinical significance is not confused with statistical significance
  • Safety outcomes are reported where applicable
  • Missing data and withdrawals are addressed
  • Subgroup and exploratory analyses are clearly identified
  • Tables and figures are accurate, readable, and consistent with the text
  • The manuscript follows the relevant reporting guideline
  • The abstract matches the main findings
  • No results have been fabricated, selectively omitted, or exaggerated

Reporting clinical study results effectively requires more than presenting numerical findings. Researchers must organize their results logically, describe participant flow, report primary and secondary outcomes accurately, communicate statistical uncertainty, and document relevant safety findings and missing data.

A strong Results section separates factual reporting from interpretation, uses tables and figures appropriately, and follows the reporting guideline suited to the study design. Most importantly, it presents favorable, unfavorable, and inconclusive findings with equal scientific honesty.

By checking the Results section against the study protocol, statistical output, and journal requirements, researchers can improve the clarity, reproducibility, and credibility of their medical research articles. Transparent reporting helps clinicians, reviewers, researchers, and policymakers assess the evidence and understand its relevance to healthcare decisions. For those considering doctoral research in this area, Top 10 Pharmaceutical Research Topics for PhD offers guidance on selecting impactful research directions.

Continue Exploring Our Blog