How to Write a Research Protocol: A Practical Guide

1. What is a research protocol, and why does it matter?

A research protocol is the documented plan for a study. It describes the scientific question, design, procedures, measurements, analysis and relevant organisational arrangements before researchers begin the work. Its purpose extends beyond satisfying an ethics committee or funder: a well-written protocol helps the research team implement procedures consistently, identify methodological problems and document decisions that may later affect interpretation.

The protocol also allows collaborators, reviewers and other qualified researchers to understand what the investigators intended to do. That transparency becomes particularly important when circumstances require changes after the study has begun.

A protocol is not the same as a research proposal

A research proposal commonly aims to persuade reviewers that a project is worthwhile, credible and feasible. It may form part of a funding application, degree programme or institutional approval process. A protocol has a more operational purpose: it documents the intended research in sufficient detail to support its conduct, oversight and interpretation.

The documents often overlap, and sections of a successful proposal may provide the basis for a protocol. Nevertheless, a persuasive scientific rationale cannot replace a precise description of what researchers will do.

A protocol should be usable, not merely complete

Imagine that another qualified researcher joins the project immediately before recruitment begins. Could they identify who is eligible, which measurements to collect, when to collect them and what to do when a planned assessment cannot be completed?

If important decisions remain unclear, the protocol may not be ready for use. This does not mean documenting every minor action in the main document. Detailed operating procedures and study manuals can accompany the protocol, provided that the essential decisions and procedures are clearly defined and the supporting documents are accessible to those who need them.

2. Identify the study type and applicable requirements

Before writing, establish what kind of study you are documenting and which requirements apply. An observational cohort study, a parallel-group randomised trial and a retrospective analysis of clinical records may share some protocol components, but their procedures, risks, analytical considerations and governance requirements can differ substantially.

Using an appropriate template from the outset can save considerable rewriting later. However, a template is a structure for communicating decisions, not a substitute for making them.

Select appropriate methodological guidance

For randomised trials, the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement provides guidance on the minimum information to include in a protocol. Its updated version, SPIRIT 2025, contains 34 checklist items and a diagram showing the planned schedule of enrolment, interventions and assessments. The guideline aims to improve the completeness and transparency of trial protocols; it does not prescribe how every trial should be designed or conducted (Chan et al. 2025).

An accompanying explanation and elaboration document describes the reasoning behind the checklist items and provides examples of informative protocol reporting (Hróbjartsson et al. 2025). These documents are particularly useful when preparing a randomised-trial protocol because they explain not only what should be reported, but why the information matters.

For observational studies, the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement can help researchers recognise information that should eventually be reported. However, STROBE was developed for reporting completed observational studies. It is not a dedicated observational-study protocol checklist or a guide to designing a methodologically sound study. The appropriate institutional or sponsor template may therefore provide a better starting structure for an observational research protocol.

Understand regulatory guidance in context

The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) develops international guidelines to support the development and regulation of medicines. One of these, ICH E6(R3), is the third revision of its Good Clinical Practice guideline. Good Clinical Practice establishes principles for the ethical conduct of relevant clinical trials, participant protection and the reliability of trial information.

The European Medicines Agency (EMA) publishes the guideline and information about its implementation in the European Union. The EMA lists the ICH E6(R3) Principles and Annex 1 as effective from 23 July 2025. Annex 2, which covers additional considerations for certain clinical-trial designs and data sources, is scheduled to take effect on 15 January 2027 (EMA Good Clinical Practice guidance).

A regulated investigational-medicine trial may require considerably more protocol detail and supporting documentation than a small academic observational study. Researchers should establish which requirements apply to their project rather than assuming that all clinical and biomedical research operates under identical rules.

Use an appropriate template

The United States National Institutes of Health (NIH) provides protocol templates for specified clinical-trial contexts, including certain drug and device trials and behavioural or social sciences research. These contain instructional material and example text, but NIH recommends rather than universally requires their use (NIH protocol templates).

If your institution or sponsor has an approved template, use it unless there is a justified reason to do otherwise. An unfamiliar custom structure may look attractive but create unnecessary problems when reviewers expect particular sections.

3. Write the title, summary and document information

The beginning of the protocol should allow readers to identify the study and understand its main features without searching through the entire document.

Choose an informative title

A useful title identifies the main research question and, where appropriate, the population, study design and intervention or exposure.

Consider the title Exercise and blood pressure in hypertension. It identifies the topic but says little about the planned research. A more informative title for our continuing example would be:

Supervised Exercise Versus Usual Care for Blood Pressure in Adults With Hypertension: Protocol for a Parallel-Group Randomised Trial.

This identifies the comparison, population and design, while making clear that the document concerns a planned study rather than completed findings. The final title must accurately reflect the protocol: do not describe a study as randomised, blinded or controlled unless the methods justify those terms.

Provide a concise protocol summary

Many templates require a structured or unstructured summary covering the rationale, objectives, design, population, intervention or exposure, primary outcome, follow-up and main analytical approach. For a randomised trial, SPIRIT 2025 recommends a structured summary of essential trial information.

The summary should communicate the important features of the plan without reproducing the full methods. Write or revise it after the main protocol sections are sufficiently developed so that it reflects the decisions documented elsewhere.

Record the administrative details

Depending on the applicable requirements, include the investigators’ names and affiliations, sponsor information, funding arrangements, relevant registration identifiers, version number, version date and appropriate contacts.

Document control matters because protocols can change. A collaborator or reviewer must be able to distinguish the version submitted for approval from the currently approved version or a superseded draft.

4. Write the background and scientific rationale

The background explains why the proposed research is needed. Its purpose is not to demonstrate that the authors have read every paper on the topic, but to establish the relevant evidence, identify the unresolved question and explain how the proposed study could contribute.

Build the rationale around the evidence gap

A useful background moves from the clinical or scientific problem to the existing evidence, identifies what remains uncertain and explains why the proposed design is appropriate. For our exercise study, this might involve the clinical importance of blood-pressure control, existing evidence concerning exercise interventions and a particular unresolved question about the population, intervention or comparison.

The rationale must reflect the literature rather than invent a gap to justify the study. Our fictional example does not, by itself, establish that another exercise trial is necessary.

Explain what the proposed study adds

Consider this weak rationale:

Hypertension is an important condition, and exercise is known to be beneficial. Therefore, further research is necessary.

Although the topic is relevant, the statement does not explain why another study would contribute meaningful information. A stronger rationale would specify the uncertainty and explain how the proposed design addresses it.

Replication, evaluation in a different population and more precise estimation can all justify additional research. Novelty is not the only legitimate contribution, but the reason for undertaking the study should be clear.

5. Define the objectives, hypotheses and outcomes

The objectives establish what the study intends to learn, while the outcomes identify what researchers will measure to investigate those objectives. These elements must agree with the study design and intended analysis.

Write objectives that match the research question

For our exercise study, the primary objective might be:

To estimate the effect of assignment to a 12-week supervised exercise programme, compared with usual care, on systolic blood pressure at 12 weeks in adults with hypertension.

This identifies the population, intervention, comparator, outcome and timepoint. It also makes clear that the study concerns the comparison between assigned groups rather than whether one group changes from baseline.

Secondary objectives might address another clinically relevant measure or a specific aspect of intervention implementation. Each should contribute useful information and be realistic within the study’s resources.

Distinguish objectives from hypotheses

An objective describes what researchers aim to establish or estimate. A hypothesis specifies an expected relationship or difference.

For example, the investigators might hypothesise that participants assigned to supervised exercise will have lower mean systolic blood pressure at 12 weeks than participants assigned to usual care. However, not every study requires a directional hypothesis. Descriptive studies and some exploratory projects may be more appropriately framed around estimation or characterisation.

Define outcomes precisely

A statement such as the primary outcome is blood pressure leaves important questions unanswered. Does the study concern systolic or diastolic blood pressure, clinic or ambulatory measurements, and which timepoint?

A more useful definition would be:

The primary outcome is mean clinic systolic blood pressure at the 12-week assessment, obtained using the prespecified standardised measurement procedure.

The protocol must then describe the procedure, including how researchers will obtain and summarise the readings. The same outcome definition should appear consistently in the objectives, measurement schedule, sample-size justification and analysis section.

An inconsistency between these sections may indicate more than an editorial problem. It can reveal that the investigators have not fully resolved what the study is designed to estimate.

6. Describe the study design and setting

The study-design section explains how the research is organised and why its structure is appropriate. The information required depends on whether the study is interventional, observational, prospective, retrospective or uses another design.

State the design clearly

For our example:

This study will use a 2-group parallel randomised design comparing a supervised exercise programme with usual care over 12 weeks.

The statement identifies the broad structure, but a complete protocol also needs to explain how researchers will allocate participants, what each group will receive and when outcomes will be measured.

If the study uses random allocation, specify the randomisation method, allocation concealment procedures and relevant responsibilities. Describe blinding arrangements accurately, including who can and cannot remain unaware of group allocation.

For observational research, explain how the exposure or characteristic of interest will be identified and how participants or records will be selected.

Describe the study setting and schedule

Identify where recruitment, interventions and assessments will occur. Where relevant, include a schedule showing when participants will be screened, enrolled, allocated, assessed and followed up.

SPIRIT 2025 includes a schedule-of-enrolment, interventions and assessments diagram for randomised trials. Such diagrams help readers identify the timing of procedures, particularly when the study involves several visits or outcomes measured at different timepoints.

Identify important dependencies

Repeated measurements from one participant, observations within the same clinic or participants randomised in clusters can create statistical dependence. The protocol should describe these features rather than leaving the statistician to discover them after data collection.

For more detail on choosing a design that matches the research question, see How to Plan a Clinical Research Study: From Research Question to Publication.

7. Specify participants, eligibility and recruitment

This section should explain who can participate, how researchers will identify potential participants and how recruitment will operate.

Define the target population and eligibility criteria

Describe the intended population, inclusion criteria and exclusion criteria. For the exercise study, eligibility could depend on the definition of hypertension, age, medication-related considerations, clinical circumstances and the ability to participate safely.

The exact criteria require justification. Copying another study’s eligibility criteria without considering their relevance can unnecessarily restrict recruitment or exclude participants who belong to the intended population.

Eligibility criteria also affect how broadly the eventual findings may apply. A sample recruited from one outpatient clinic may differ considerably from the wider population of adults with hypertension.

Explain recruitment and consent

Describe how researchers will identify, approach, screen and enrol potential participants. Where informed consent is required, explain when and how it will be obtained, who is responsible and how participants will receive the information needed to decide.

Research involving existing records or identifiable data may require different procedures from an intervention trial. The protocol should reflect the ethical, legal and institutional requirements established for the actual study.

Distinguish the target and recruited populations

The target population is the group to which the research question refers. The recruited population consists of the people who can be identified and enrolled through the study’s practical procedures.

These groups may differ. Describe the recruitment setting honestly rather than implying that findings from a highly selected sample automatically generalise to every member of the target population.

8. Describe interventions, exposures and measurements

This section often determines whether the protocol can be implemented consistently. It should explain what researchers and participants will do, how they will collect measurements and which procedures are essential to answering the research question.

Describe interventions sufficiently for replication

For the exercise trial, stating that participants will complete a supervised exercise programme is inadequate. The protocol may need to specify the exercise mode, session frequency, duration, intensity, progression, supervision, permitted modifications and procedures for monitoring attendance or adherence.

The comparison condition deserves equal attention. Usual care can mean different things across clinical settings and participants, so the protocol should explain what care participants are expected to receive and whether both groups continue their existing treatments.

The Template for Intervention Description and Replication (TIDieR) is a 12-item guide developed to improve descriptions of healthcare interventions. It covers matters such as intervention materials, procedures, delivery, scheduling, tailoring and fidelity (Hoffmann et al. 2014). Although developed as reporting guidance, its questions are useful when planning the information a protocol needs to contain.

Describe exposures in observational research

For observational studies, explain how the exposure or characteristic of interest will be defined and measured.

For example, a study of habitual physical activity might collect exposure data using a questionnaire or wearable device. These methods do not necessarily measure identical constructs, and they may differ in their susceptibility to measurement error.

The protocol should specify the assessment method, observation period and how the recorded information will be converted into variables for analysis.

Make measurement procedures unambiguous

Consider this description:

Blood pressure will be measured before and after the intervention.

It leaves uncertainty about the measurement method, assessment windows and intended summary of the readings.

A more informative description would be:

Clinic systolic blood pressure will be assessed at baseline and 12 weeks using a prespecified standardised procedure. The primary outcome will be the 12-week measurement, summarised according to the study’s defined assessment protocol.

Even this improved wording is not a complete procedure. The protocol or an explicitly referenced measurement manual should specify the device, measurement conditions, number of readings, acceptable assessment windows and handling of unusable measurements.

Not every procedure needs the same level of detail. However, measurements that determine the primary outcome deserve enough specificity to support consistent implementation and interpretation.

9. Explain the sample-size justification

The protocol should explain why the proposed sample size is appropriate for the primary objective and intended analysis.

For a conventional superiority trial, the calculation may involve the target difference, expected outcome variability, statistical power, type I error rate and anticipated missing outcome data. Other studies may use precision-based planning, expected event numbers or design-specific approaches. There is no single calculation suitable for every research protocol.

Report the assumptions and their sources

Consider this statement:

The study will recruit 100 participants to provide sufficient statistical power.

It provides a target without explaining how researchers calculated it or what the target is intended to support.

A stronger justification identifies the primary outcome, intended comparison, target effect or precision, relevant variability assumptions and calculation parameters. It should also explain any adjustment for expected missing outcomes and identify the evidence supporting the assumptions.

Be cautious when using a small pilot study to estimate the expected treatment effect. Such estimates may be highly uncertain and can produce misleading sample-size targets if treated as reliable predictions.

Check consistency with the intended analysis

The sample-size rationale should concern the same primary outcome and scientific comparison as the analysis plan.

If the sample size is calculated to compare groups at 12 weeks but the primary analysis is described as testing changes separately within each group, the protocol contains an important inconsistency. Resolve it before submitting the document for review.

10. Summarise the statistical analysis strategy

A protocol should explain how the planned analysis will answer the research question. This does not necessarily require every technical detail in the main document, but the description must be more informative than a list of statistical tests.

Define the primary comparison

For the exercise study, the primary analysis might estimate the between-group difference in mean systolic blood pressure at 12 weeks using an appropriate model that adjusts for baseline blood pressure.

This would need to be justified in relation to the design, outcome and treatment effect of interest. Importantly, the primary between-group comparison should not be replaced by separate tests of changes within each group.

Explain the main analytical decisions

Describe the primary analysis population and statistical approach, along with how the analysis will account for relevant features such as repeated measurements or clustering. Where appropriate, identify planned covariates, missing-data methods, multiplicity considerations and sensitivity analyses.

The details depend on the study. For example, a prediction-model development study requires a different statistical framework from a randomised intervention trial.

The protocol should also distinguish the primary confirmatory analysis from exploratory analyses. This helps researchers and readers understand which conclusions the study was specifically designed to support.

Explain the relationship with a statistical analysis plan

A statistical analysis plan (SAP) is a document that specifies the intended analyses and related statistical decisions in greater detail.

A separate SAP may be useful for complex studies or required by the sponsor or regulatory framework. Simpler academic studies may be able to document their analytical strategy adequately within the protocol, depending on the applicable requirements.

The protocol and SAP should agree on fundamental features such as the primary objective, outcome, population and comparison. The SAP can elaborate the analysis, but it should not quietly redefine the scientific question.

For guidance on matching methods to the research question and data structure, see How to Choose the Right Statistical Test for Your Study.

11. Address ethics, participant protection and data management

The protocol should describe how the research team will protect participants, manage relevant risks and handle study information.

The World Medical Association’s Declaration of Helsinki, revised in 2024, sets out ethical principles for medical research involving human participants, including research using identifiable human material or data (World Medical Association 2024). The specific requirements for ethical review, consent, privacy and oversight depend on the study and the applicable institutional and legal framework.

Describe risks and participant safeguards

For the exercise study, relevant considerations might include participant screening, intervention-related risks, clinical supervision, stopping procedures and adverse-event management.

An observational study using existing clinical records may raise different concerns, particularly around confidentiality and lawful data use. The nature and extent of the safeguards should reflect the study’s procedures and risks.

Define data-handling responsibilities

Explain how researchers will record, identify, access, store, check and retain data. The protocol may refer to a separate data management plan where the project uses one.

Relevant procedures can include participant identifiers, access restrictions, source-data preservation, documented corrections and safeguards for identifiable information.

The objective is reliable, traceable and appropriately protected data. Researchers should use systems that meet their project’s requirements rather than assuming every study needs the same infrastructure as a regulated investigational-medicine trial.

Address registration and dissemination

Clinical trials subject to prospective registration requirements should be registered in an appropriate public registry at the required time. The World Health Organization (WHO) supports registering clinical trials before recruiting the first participant (WHO trial registration guidance).

Registration requirements depend on the study type and applicable legal, regulatory, funder and journal policies. An observational study does not automatically fall under the same requirements as a prospectively assigned intervention trial, although other preregistration expectations may apply.

The protocol should also describe relevant plans for publishing the findings, communicating results where appropriate and making study information available under applicable ethical and legal conditions.

12. Document oversight, responsibilities and amendments

The protocol should identify responsibility for the study’s main activities and explain how the research team will address important problems or changes.

Depending on the project, these activities may include recruitment, intervention delivery, data quality, participant safety, statistical analysis and reporting. Regulated trials may have additional sponsor, investigator and monitoring requirements.

Describe proportionate oversight

Explain how the research team will identify relevant procedural or data-quality problems, who will review them and what actions may follow.

For interventional research, monitoring arrangements should reflect the risks and applicable requirements. SPIRIT 2025 includes monitoring-related information among its protocol items, but this does not mean every academic study requires identical monitoring procedures.

Maintain a transparent amendment history

Protocols sometimes need to change for legitimate reasons. Recruitment difficulties, newly identified risks or unexpected measurement problems may require revisions to the original plan.

Document what changed, why it changed and when the new version took effect. Where approval or reporting is required, follow the relevant procedure before implementing the change.

The final study report should describe important departures from the original plan rather than silently rewriting the protocol history.

13. Review the protocol for completeness and consistency

Completeness and consistency are related but different. A protocol may contain every required heading yet describe a study whose primary outcome, sample-size justification and statistical analysis do not agree.

The final review should therefore examine both whether the necessary information is present and whether the different sections describe the same intended study.

Check the scientific pathway

For our exercise study, the main protocol components should align as follows:

Each component should support the same scientific question. If the protocol defines clinic blood pressure as the primary outcome but the sample-size section uses assumptions for a different measure, the discrepancy needs resolving.

Check feasibility and implementation

Ask whether another qualified member of the team could understand how recruitment, allocation, measurements and follow-up will operate.

The schedule should be realistic, responsibilities should be clear, and the required equipment and expertise should be available. If the protocol depends on procedures described in supporting documents, check that those documents exist and are consistent with the main plan.

Check the relevant guidance

Finally, review the document against the required institutional template and applicable guidance. For randomised trials, the SPIRIT 2025 checklist and explanation document provide a structured way to assess protocol reporting.

However, completing a checklist does not establish that the study is appropriately designed. That still requires scientific and methodological judgement.

If you would value independent feedback on the document before submission, you can explore Blazon Scientific’s scientific writing, editing and statistical support services. An external review is particularly useful when different contributors have prepared the scientific rationale, protocol methods and analysis sections.

14. Common research protocol mistakes

Several problems recur in protocol drafts, regardless of the study topic or design.

Using vague methodological descriptions

Statements such as standard procedures will be followed or appropriate statistical tests will be used rarely provide enough information to understand the study. Identify the procedures, standards and decisions that matter, and refer to supporting documents where appropriate.

Using inconsistent outcomes and timepoints

The primary outcome should mean the same thing in the summary, objectives, measurement schedule, sample-size justification and analysis section. Differences in wording can sometimes reveal an unresolved methodological decision rather than a simple editing mistake.

Treating templates as substitutes for decisions

A template can identify information that should be provided, but it cannot determine whether the study design is appropriate. Text copied from a previous protocol may contain assumptions or requirements that do not apply to the new research.

Describing the intervention but not the comparator

In a controlled trial, readers need to understand both study conditions. For example, usual care may differ across settings and participants; the protocol should explain it sufficiently to establish the comparison being investigated.

Deferring analysis decisions until after data collection

A statement promising to choose statistical methods once the data are available leaves important decisions unresolved. Some details can be developed in a separate SAP, but the primary analytical strategy should already be coherent with the research question and design.

Neglecting version control

Changing a protocol is not necessarily evidence of poor planning. However, failing to document important amendments can make it difficult to establish which procedures were approved, implemented and later reported.

Research protocol quality checklist

Use this checklist to review the written document before submission or approval.

• The title accurately identifies the study design, population and relevant comparison.
• The protocol includes the required administrative information, version number and date.
• The background establishes a specific evidence gap or justified research need.
• The primary objective is clear and scientifically meaningful.
• The study design is appropriate for the objective.
• The setting, participant population and eligibility criteria are defined.
• Recruitment and applicable consent procedures are explained.
• Interventions, comparators or observational exposures are sufficiently described.
• The primary outcome, measurement procedure and timepoint are unambiguous.
• Secondary outcomes are justified and clearly specified.
• The sample-size justification corresponds to the primary objective and analysis.
• The primary statistical comparison and analytical strategy are defined.
• Relevant missing-data, dependence and sensitivity-analysis considerations are addressed.
• Ethical, safety, privacy and governance arrangements are explained.
• Applicable registration and dissemination requirements are addressed.
• Relevant oversight arrangements and responsibilities are documented.
• The protocol explains how amendments and versions will be managed.
• The summary, objectives, methods, sample-size section and analysis are consistent.
• The document meets the applicable institutional, sponsor and regulatory requirements.

This checklist supports document review. It does not replace an appropriate study-specific template or formal methodological, ethical or regulatory assessment.