by Sumudu Narayana
A clinical protocol is the framework of a trial, outlining how the trial will be conducted and providing all the necessary guidelines for the study team.
Writing a research protocol for a clinical trial is an intricate and time-consuming process. It can also be a little daunting at first, especially coming from a more general biomedical science background with limited understanding of the clinical world, as I did.
We all make mistakes, but some can be more taxing that others. Keeping this in mind, here are a few things I wish I’d known before I wrote my first clinical protocol.
1. Have a defined research question and primary outcome
A well-defined research question and primary outcome are essential for writing a clinical protocol. Your research question is the very purpose of the study. This can range from gaining scientific knowledge on a therapeutic area in need of innovation or conducting a crucial study testing a new drug.
Research questions should contain neutral wording and define the study’s purpose and scope. A poorly formed question will immediately get you off on the wrong foot.
For instance:
BAD: To assess the effect of rifaximin on liver disease.
Why? The question does not specify the effects of rifaximin that are being assessed in the study. Similarly, liver disease encompasses a broad range of conditions and complications, and this research question does not clarify what is being examined in the study.
BETTER: To assess the efficacy of rifaximin in reducing portal hypertension in a small cohort of patients with compensated cirrhosis.
The measure by which your research question is answered is called the primary outcome. It is important to have one defined primary outcome, and it has a limited number of related secondary outcomes. This is because in a clinical setting it is easy to collect a great deal of data that might ultimately be completely unnecessary for addressing the study question.
Having clear definitions for the study focus helps in avoiding the “do-it-all” scenario and will help you narrow down the study methods. It also helps avoid generating imprecise results.
Always remember: the quality of data you produce is a reflection of the thought and effort put into designing the study.
Tip: When designing your research study cross-check your topic, and related topics, on clinical trial registries to ensure that your research question isn’t a currently registered study or been studied already.
2. Conduct statistical planning
Statistical planning was something I didn’t know I had to consider when I wrote my first protocol. But I soon found out how essential clinical statistics were in writing up the protocol.
It’s key to determine how many patients are needed in your study, and this is the first stage of statistical planning.
Power calculations and sample size estimations are used to calculate the number of participants required to answer the research question. These calculations help generate an appropriate sample size that can then be used to make inferences about the whole population.
It’s also important to define your statistical methods and the plan for data analysis at this stage. This is because sometimes the type of statistical analysis you plan to use at the end of the study can affect the sample size needed for the study. For example, a study using a non-parametric test (e.g., Mann–Whitney U) will need more patients than one using a parametric test (e.g., Student’s t-test).
I recommend working with a trained biostatistician during the early stages of protocol development to ensure that the study is designed appropriately. This will also help you brush up on your clinical statistics, and even learn some new techniques as I have.
You will also need access to some statistical software. IBM SPSS is commonly used as it is easy to use and allows for a wide-range of statistical analyses such as descriptive analysis, univariate and multivariate analysis and predictive modeling. However, a number of alternatives exist, including MATLAB, STATA, SAS, and the free and open-source package R.
Tip: You’ll need some basic statistical knowledge in the following areas to work in clinical trials: descriptive statistics, estimation and hypothesis testing, randomization and survival analysis. There are a number of free online courses that you can do to learn, or refresh, your statistical skill set.
3. Collaborate with your research team and study support staff
Establishing positive and productive relationships with other members of the research team and key support staff will yield many rewards. Your research team and support staff can include anyone from doctors, nurses, pharmacists (if dispensing drugs), physiotherapists, and laboratory services staff (blood processing).
As a researcher with a basic science background, I found it daunting delving in to the clinical aspects of patient care and study procedures. Having a good working relationship with all the different members of the research team allowed me to ask questions (I had many for my first protocol). For example, I had to ask the doctors and nurses in our team about a complication of liver disease (my research topic) and the related procedures to understand the topic of my clinical project.
Don’t be afraid to ask questions, it’s better to ask a lot of questions and learn than it is to write something wrong in your protocol because you didn’t understand a technique or clinical terminology. Equally important is to take on board your research team’s comments in designing the study protocol. These relationships are also essential when designing the study as it allows for clear discussions about the roles and expectations of each team member during the study.
Tip: I learned early that all members of a research team are busy in their primary roles. Always make appointments with team members, prepare a list of specific questions you have and be mindful of their time. This will help create a good foundation for your working relationship.
4. Eligibility and study methods
The most important part of the research protocol is outlining the study processes and logistics of how the study will be completed.
The protocol should cover topics that range from the identification of participants to recruitment and data collection and management.
The CONSORT 2010 guidelines are a useful tool for study design. These are an internationally recognized “minimum set of recommendations for reporting randomized trials”. The CONSORT 2010 Checklist provides a detailed 25-item list of things that need to be considered in study design and reporting.
The SPIRIT 2013 statement is another valuable source. This was an initiative by leading experts to “improve the completeness and quality of trial protocols”. The statement provides comprehensive guidelines in trial design, trial registration and ethical principles.
Eligibility criteria for a study are crucial and should be kept simple to maximize recruitment into the study. Inclusion criteria are intended to define the study population. Exclusion criteria further refine the study population and should include relevant contraindications.
As a basic scientist, I found this section the most difficult to write and relied on my relationships with other members of the research team to define criterion. It’s also a good idea to include a list of study assessments and a schedule of when these assessments will occur in the study methods. Ensure that this is clear and easy to understand.
I’ve found using a table or flow diagram to describe the study assessments and schedule effective.
Try, where possible, specifying when an assessment will occur and whether it is study-specific or part of standard care. This will reduce the number of questions and comments you get from the ethical review board, while also making it easy for every member of the team to know how the study will proceed.
Tip: Get your study protocol reviewed by someone outside of your study team but would have a basic understanding of your research topic. Informal peer-review is valuable as it will help identify potential flaws in study design or additional study methods that can be added. Also look for other free peer-review systems that might be available in your institute or region. For example, in Australia, we have the NHMRC National Scientific Committee that provides advice on the scientific merit and integrity of a research protocol.
5. International and national regulations
In 1964, the World Medical Association developed the Declaration of Helsinki as a statement of ethical principles for medical research involving human subjects. These guidelines form the basis of the current international standard, International Conference on Harmonization – Good Clinical Practice (ICH-GCP).
ICH-GCP is “an international ethical and scientific quality standard for designing, conducting, recording and reporting trials that involve the participation of human subjects.” Ethical review boards and other regulatory committees execute the 13 main principles of the ICH-GCP. You can find these guidelines published online.
It is the responsibility of the principal investigator and protocol developer to ensure that their study adheres to these guidelines, as well as any additional local or national legislation.
In Australia, I’ve needed to get clearances from the Therapeutic Goods Act for drug trials and confirm that protocols adhere to both state and federal legislation.
So, when you are designing a protocol, ask yourself some of the following questions:
- Does the protocol include testing a new/unapproved drug or device?
- Is the protocol testing an approved drug or device outside of its intended use? Or outside of its intended patient population?
If any of these are applicable, work alongside the research office to ensure you follow the appropriate regulatory standards and ask for advice if you are unsure on any aspect. Liaising with your institution’s ethics officer will definitely be beneficial for this section of the protocol.
Safety reporting in a trial is one of the main requirements of the ICH-GCP.
As part of the clinical protocol, it is important to state what is, and is not, a reportable adverse event. Additionally, it should also clarify who is responsible for reporting adverse events and to whom they should be reported. An adverse event is an unexpected or unintended symptom or disease that may or may not be related to a medical treatment or procedure.
There are standard guidelines (for example, in Australia, we have the government’s Australian Clinical Guidelines) and templates such as the Common Terminology Criteria for Adverse Events (CTCAE) that you can refer to.
Depending on the type of study you are conducting, and the participants involved, it can be advisable to tailor the adverse event wording to the research study to avoid unnecessary over-reporting.
Tip: One thing that has helped me with safety reporting is creating a table of potential adverse events for the trial I am designing. Also useful is creating an adverse events log for each participant enrolled into the study, as this lets me document and keep track of any events that occur during the study. This log should include details such as adverse event, onset date, CTCAE grade, and relationship to study drug/procedure.
6. Clinical trial registration
Trial registration is a step that comes after protocol development and ethical review but is essential prior to enrollment of any patients into your clinical trial.
There a number of reasons why a trial must be registered on a public clinical trial registry. These include prevention of selective data reporting and publication, preventing the unnecessary duplication of research and to help the public, and potential participants, be aware of all ongoing research studies.
There are a number of public clinical trial registries, with the two most commonly referenced registries being the WHO International Clinical Trials Registry Platform (ICTRP) and ClinicalTrials.gov. These registries are not-for-profit, are accessible to the public free-of-charge and have mechanisms in place to ensure the validity of the data registered.
Some countries have trial registries that are specific for their region. For example, in Australia, we have the Australian New Zealand Clinical Trials Registry (ANZCTR), which is partly funded by the Australian Government. All clinical trials are required to be registered on ANZCTR upon ethics approval, with the ANZCTR contributing data to the ICTRP.
Trial registry of a clinical protocol is now a major requirement by journals prior to publication. You will need to provide evidence that the trial was registered prior to the enrollment of the first patient into your study. The best way to do this is by providing your trial registration number in your manuscript. The registration number will also be published, usually in the abstract or body of the manuscript.
The International Committee of Medicine Journal Editors (ICMJE) provides a comprehensive summary of the requirements of trial registration and data sharing prior to publication of your research.
Tip: When planning your research protocol, spend some time looking through trial registries to get an idea on what sort of information you need to register. This will help you to design your study accordingly. Also, if you have an idea of which journal you’d would like to submit your research to, look through its requirements for trial registration. For example, you might want to check if the journal has a preference on which trial registry research protocols should be registered.
As even this relatively short, and by no means exhaustive, list shows, there are many things to consider when writing a clinical research protocol.
A good starting point is looking through the EQUATOR Network. This is an umbrella organization that provides links and information guidelines for all aspects and styles of clinical research. You will find links to some of the guidelines (CONSORT, SPIRIT) and information sources referenced in this article, and many more.
Get to know your local research office personnel, as they are the best information source to get a comprehensive overview on what you need. Also, work closely with your research team and check for any ethics board guidelines and checklists to ensure you have all you need to write a great first research protocol.
Here’s a handy chart to show you the common flow of the process.
Sumudu Narayana currently works as a clinical trials coordinator in Adelaide, South Australia. She has a PhD in Biological Sciences from the University of Adelaide and transitioned into clinical research to help bring research from the laboratory to patients.