The research protocol
- Concept
The written plan of a study: what is asked, how, with whom, when and with what resources. It is approved and registered before the start.
What it is
The protocol is the document that describes, before the first datum is collected, everything the study will do: the question, the design, the population, the variables, the analysis, the ethical aspects, the timeline and the budget. It applies the scientific method (observation, question, hypothesis, test and conclusion) to a concrete plan. The ethics committee reviews it; clinical trials are publicly registered before the first participant is enrolled, and later changes are documented.
Key facts
- Research process: idea → literature search → question → protocol → ethics approval and registration → pilot → data collection → analysis → publication.
- Recruitment time = participants needed / (sites × enrolments per site per month).
Worked example
- Data: a 2-site trial needs 300 participants; each site enrols 10 a month and each person is followed for 3 months.
- Combined rate: 2 × 10 = 20 a month. Months of recruitment: 300 / 20 = 15.
- Before recruitment: 3 months for the ethics committee, trial registration and staff training.
- Recruitment ends in month 3 + 15 = 18; the last person enrolled finishes follow-up 3 months later, in month 21.
- Cleaning, analysis and report: 3 months. Total length: 3 + 15 + 3 + 3 = 24 months.
- If each site enrols only 7 a month: 300 / 14 = 21.4, rounded up to 22 months; total 3 + 22 + 3 + 3 = 31 months.
- Reading: a recruitment rate 30 % lower lengthens the study by 7 months; the protocol timeline must rest on real figures from each site.
Protocol sections and the question each one answers
| Section | What it answers |
|---|---|
| Background and rationale | What is known, and why is this study needed? |
| Question, objectives and hypothesis | What exactly do we want to know? |
| Design | Which type of study answers it best? |
| Population, criteria and sample | With whom, and how many? |
| Variables and measurements | What is measured, and how? |
| Analysis plan | How will the data answer the question? |
| Ethical aspects | How are participants protected? |
| Timeline and budget | When, and with what resources? |
Markers
- SPIRIT
- ICTRP
- REC
In practice
- In Colombia the protocol goes to the institution's research ethics committee, under Resolución 8430 de 1993 (see Ethics).
- Journals require clinical trials to have been registered before they started in order to consider them for publication.
- Publishing or registering the protocol lets readers compare what was planned with what was reported.
Common pitfalls
- Writing the protocol after the data have been collected.
- Timelines based on optimistic recruitment rates that nobody checked.
- Changing the primary outcome without an amendment or a reason.
Test yourself
Think of the answer, then tap the question to see it.
When must a clinical trial be registered?
Before the first participant is enrolled, in a registry recognised by the WHO ICTRP platform.
Which guideline describes the items of a clinical trial protocol?
SPIRIT.
240 participants are needed and 3 sites enrol 5 a month each. How many months of recruitment?
3 × 5 = 15 a month; 240 / 15 = 16 months.
What is done if the protocol must change during the study?
A dated, justified amendment, approved by the ethics committee and recorded in the registry.
Elsewhere in DurgiPedia
- Research ethics committees
- Resolution 8430 of 1993: health research in Colombia and its risk categories
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: SPIRIT 2025 Statement: Defining standard protocol items for clinical trials (EQUATOR Network, 2025); International Clinical Trials Registry Platform (ICTRP) (World Health Organization, 2026); E6(R3) Good Clinical Practice (GCP) (U.S. Food and Drug Administration, 2025).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.