Commissioning a clinical study starts with a practical question: what do you need to learn about your ingredient, and how will that information be used?
You may be preparing for conversations with supplement brands, exploring a new application or building evidence for a specific benefit. Each goal can lead to a different study.
Before choosing a research partner, define the ingredient, intended use, target population and outcome you need to assess. Then compare proposals against that same brief. This makes it easier to understand what you are buying and whether the study can answer the question that matters.
This guide focuses on ingredient research for dietary supplements, with U.S. claims guidance as the regulatory context. The proposal checklists are practical buyer recommendations, not a universal legal standard or a promise of particular study results.
What should an ingredient clinical study help you establish?
A useful study should answer a clearly defined question about a specific ingredient under specific conditions.
“Does this ingredient support wellness?” is too broad to guide a protocol. A more useful question identifies:
- The ingredient and its specifications.
- The dose and formulation being tested.
- The people who will participate.
- The outcome being measured.
- The assessment period.
- The comparison, where appropriate.
For example, a manufacturer investigating sleep would need to decide whether the question concerns perceived sleep quality, time to fall asleep, sleep duration or daytime functioning.
Those are related outcomes, but they are not interchangeable.
FDA’s dietary supplement substantiation guidance recommends evaluating how closely the evidence relates to the particular product and claim, the quality of that evidence and the broader body of research. That makes a focused research question valuable from the start. FDA substantiation guidance
Can ingredient research support a finished supplement?
Ingredient research may be relevant to a finished supplement, but its applicability needs to be assessed.
A brand using your ingredient should be able to understand exactly what was studied and how its product compares. Differences in dose, formulation, delivery format, other ingredients or intended users may affect whether the findings apply.
FDA recommends examining whether the studied ingredient or supplement and its conditions of use are relevant to the marketed product. A shared ingredient name alone does not resolve that question. FDA substantiation guidance
Before commissioning research, prepare an ingredient information sheet covering:
| Information | What to document |
|---|---|
| Ingredient identity | Exact material, source and relevant specifications |
| Standardization | Relevant constituents and how they are measured |
| Study material | Batch, formulation, dose and delivery format |
| Quality documentation | Relevant testing, storage and stability information |
| Intended application | How customers are expected to use the ingredient |
| Existing evidence | Relevant published and unpublished findings |
Ask prospective research partners what additional information they need before designing the study.
Which study design should you choose?
Start with the decision you need to make, then ask the research team to justify the design.
| Research goal | Approach to discuss | Important limitation |
|---|---|---|
| Check whether recruitment and study procedures are workable | Feasibility or pilot study | A feasibility result does not establish a health benefit |
| Explore outcomes and refine a future protocol | Exploratory study | Findings need to be interpreted within the exploratory design |
| Evaluate an effect against a comparator | Randomized controlled study, where appropriate | Conclusions still depend on execution, analysis and relevance |
| Describe patterns of use and outcomes without assigning an intervention | Observational research | Associations do not establish that the ingredient caused the outcome |
These categories can overlap. A pilot can be randomized, and a controlled trial can collect data remotely.
For health benefit advertising claims, the FTC generally expects randomized controlled human testing. It also distinguishes observational associations from evidence of causation. The appropriate evidence depends on the claim and the relevant scientific standards. FTC Health Products Compliance Guidance
Ask the research partner:
What conclusions could this design reasonably support, and what would remain unanswered?
A useful proposal should address both.
How should you choose study outcomes?
Choose measurements because they answer the research question, not simply because a device or platform can collect them.
Ask the team to identify the primary outcome, meaning the main result the study is designed to assess. Secondary and exploratory outcomes should have clear purposes too.
Depending on the protocol, research can combine:
- Wearable measurements to collect relevant physiological or activity data.
- Participant questionnaires to assess experiences, symptoms or functioning.
- Laboratory measurements to evaluate selected biomarkers.
- Other assessments appropriate to the research question.
For each measurement, ask:
- What does it measure?
- Is it suitable for this population and intended use?
- When and how often will it be collected?
- What would a meaningful change look like?
- What are its limitations?
For wearables, also ask about device models, software changes, data access and whether measurements can be compared across devices. FDA’s digital health technology guidance provides a useful reference for evaluating remote measurement tools in medical-product investigations. Its applicability to a particular supplement study should be assessed rather than assumed. FDA digital health technology guidance
For more context, see our guide to wearables in decentralized clinical trials.
What should the proposal say about participants and missing data?
A participant target needs an explanation. For a feasibility pilot, ask how the number supports the feasibility objectives. For an efficacy trial, ask for the assumptions underlying its statistical power or other justified sample-size approach. Do not use a small feasibility pilot’s observed treatment effect as the basis for sizing a later efficacy trial; those estimates can be unstable. Ask the statistician to justify the target difference and other assumptions using appropriate evidence. NCCIH pilot-study guidance
Ask how the proposed sample size relates to the primary outcome, study design and analysis. The proposal should distinguish the number recruited from the number expected to complete the relevant assessments.
Before signing, ask how the team will handle:
- Missed questionnaires.
- Missing wearable recordings.
- Participants who stop taking the study product.
- Dropouts and incomplete follow-up.
- Changes in medications, supplements or relevant behaviors.
- Measurements that fail quality checks.
Missing recordings should not silently become zero values or disappear from the report. Ask how assumptions about missing data will be assessed and whether sensitivity analyses are appropriate. Results based only on participants who complete the study can be biased. The approach must fit the study rather than rely on a universal rule for filling in missing values. Stopping the study product is not necessarily the same as withdrawing from follow-up. The protocol should distinguish those events and explain what follow-up is planned, subject to participant consent. For technical background, see ICH E9(R1) on treatment effects and sensitivity analysis. This pharmaceutical-development guidance is a methodological reference, not a blanket regulatory requirement for supplement studies.
Define the primary outcomes and assessment windows in the protocol before enrollment. Prespecify the analysis methods before unblinding or examining comparative results. Amendments should record what changed, when and why. Exploratory analyses should be clearly distinguished from prespecified confirmatory analyses. You should understand which participants and observations contribute to each result.
Our clinical research methodology and evidence page explains our approach to measurement, missing data and study limitations.
How do you compare research partners?
A contract research organization, or CRO, may manage some or all of the study. The proposal should identify who is responsible for each part.
Use these questions to compare partners:
| Area | Questions to ask |
|---|---|
| Scientific design | Who develops the protocol and justifies the outcomes and sample size? |
| Ethics and participant protection | Who determines applicable review requirements, manages consent and oversees safety? |
| Recruitment | How will eligibility be checked and recruitment progress reported? |
| Study operations | Who handles product distribution, participant questions and adherence tracking? |
| Data management | How are data quality, access, security and corrections managed? |
| Analysis | Who writes the analysis plan and performs the statistical work? |
| Reporting | What reports, datasets and supporting documents are included? |
| Accountability | Who is your main contact, and how are delays or protocol changes handled? |
Request an example report or a redacted deliverables list. That can make differences between proposals easier to see.
What determines the cost of an ingredient clinical study?
A useful budget is tied to a defined protocol and an explicit list of deliverables.
Ask prospective partners to itemize:
- Protocol development and statistical planning.
- Ethics review and study setup.
- Recruitment and participant compensation.
- Study product, comparator and distribution arrangements.
- Devices, questionnaires and laboratory testing.
- Participant support and safety oversight.
- Data management, analysis and reporting.
- Optional manuscript or publication support.
Also ask what triggers additional charges. Examples to clarify include recruitment extensions, repeat testing, extra analyses or protocol changes.
For the timeline, request separate milestones for setup, recruitment, participant follow-up, data review, analysis and reporting.
A study’s participation period is only one part of the project schedule.
What should you receive at the end?
Agree on the deliverables before enrollment begins.
A recommended buyer’s checklist includes:
- Final protocol and documented amendments.
- Statistical analysis plan.
- Participant flow and completion information.
- Results for all planned outcomes, including null or unfavorable findings, with effect estimates and appropriate measures of uncertainty.
- Missing-data and protocol-deviation summaries.
- Protocol-defined adverse-event collection and safety reporting, with monitoring responsibilities, escalation procedures and relevant limitations documented.
- A report explaining methods, findings and limitations.
- Agreed access to de-identified data and a data dictionary, subject to consent and applicable requirements.
- Written terms for ownership, publication and customer-facing use.
Clarify whether manuscript preparation, journal submission fees and responses to peer review are included.
Publication should be treated as a separate process. A research partner can help prepare a manuscript, but should not promise acceptance by an independent journal.
A hypothetical example: planning research on a sleep ingredient
Suppose an ingredient manufacturer wants to investigate perceived sleep quality in adults who report occasional poor sleep.
Before selecting a study package, the manufacturer and research team would work through:
- The study material: the exact ingredient, specifications, dose and formulation.
- The population: eligibility criteria and relevant exclusions.
- The main outcome: an appropriate measure of perceived sleep quality.
- Supporting measurements: whether wearable data or other assessments add useful information.
- The design: the comparator, randomization and blinding arrangements, where appropriate.
- The analysis: sample-size assumptions, assessment periods and missing-data rules.
- The interpretation: what the results could support and where they would not apply.
This is an illustrative planning example, not a recommended protocol or a claim that the ingredient works.
If the study assesses sleep quality, the results should not automatically be presented as evidence of improved recovery, cognition or treatment of insomnia. Those are different claims requiring their own assessment. FTC Health Products Compliance Guidance
Frequently asked questions
Is a pilot study enough?
A feasibility pilot tests whether study procedures are workable, not whether an ingredient is effective. Its size should be justified against feasibility objectives. A confirmatory efficacy trial needs a separate sample-size rationale. Small pilots also cannot reliably establish safety or detect rare harms. See NCCIH’s guidance on pilot studies.
Can an ingredient study be conducted remotely?
Some assessments can be collected remotely. Whether a remote, site-based or hybrid approach is appropriate depends on the protocol, participant needs, measurements and safety arrangements.
Does a clinical study make an ingredient FDA approved?
No. FDA does not approve dietary supplements before they are marketed. Conducting a study does not confer FDA approval on a supplement or establish that an ingredient may lawfully be marketed. Applicable safety, manufacturing and new dietary ingredient notification requirements still need to be assessed. See our guide to FDA approval for wellness products and the FDA’s dietary supplement overview.
Can customers use the results in their marketing?
Potentially, subject to agreed usage rights and a review of whether the evidence supports their specific product and claims. Discuss this before the study so the intended commercial use is clear.
For a closer look at how evidence relates to marketing language, read our guide to structure/function claims and supporting evidence.
Planning your next ingredient study
Bring your ingredient specifications, existing evidence, intended application and key research question to the first conversation.
Ask the research team to explain the proposed design, deliverables and limitations in plain language. You should leave knowing what the study is intended to answer and what decisions it could help you make.
Explore the Reputable Health clinical research platform and our research methodology to prepare for that discussion.