Before developing and writing the protocol, you should have reviewed the research and determined the research question.
The International Conference on Harmonization defines the protocol as a document that describes the objective(s), design, methodology, statistical considerations and organization of a trial. The protocol usually also gives the background and rationale for the trial, but these could be provided in other protocol referenced documents. The protocol lays out uniform standards for the way each participant is handled - this allow researchers to combine data and reach valid conclusions. It also allows for replication if others were to conduct the study themselves.
This page provides detail on the structure, procedures, and organization for the conduct of research. It provides information on the types of things that should be considered when developing a protocol.
Duke Approach and Tools
A protocol details the structure, procedures, and organization for the conduct of research. A protocol can be part of a grant made available via an agreement or contract, or created by the investigator who will conduct the research. For investigator-initiated research, the investigator would be responsible for writing the protocol and the CRC and other key staff members (such as a statistician) may be asked to contribute to the development of the protocol.
Required protocol elements and structure can vary on the phase of the trial, the type or design of the study, and the funding agency. However, many elements are common. The elements that should be covered by all protocols, regardless of design include:
- Background and Rationale
This section of the protocol includes an evidence-based background and justification for why you want to do the study. It establishes the significance of the topic to be researched and provides the conceptual framework for addressing the hypothesis.
Introduce the topic of interest. Indicate the specific population of interest and the literature that has been reviewed (including citations). Provide a brief and focused description of the key information that you’ve learned about this topic from your literature review. Include a statement placing the study in the context of the development or proposed use of the test article, if applicable.
Highlight the gap or area where there is missing information in the literature that your study will address. Indicate why this research is important and how this research is different from existing research or literature.
- Study Design
· Type of Study - For specific information about study design, see the Study Design and Statistical Considerations web-based on-demand training. The FDA provides a guidance document about Study Design.
· Participants - This section describes the plan for recruiting subjects to the study. Describe the study inclusion and exclusion criteria. Indicate the expected number of subjects to be enrolled and age range of the subjects. Indicate where you will access the population and how you will find the number of patients needed. Outline what is required of the subject to participate in the study, steps for withdrawing from the study, and circumstances under which the investigator can withdraw a participant from the study.
· Sampling Frame - This section provides details of planned data analyses and statistical considerations. Describe how the participant sample is determined. Describe the proposed statistical analysis, how missing data will be handled such as what will happen when a data field is left blank.
· Risks and Benefits - Risk is the probability of harm or injury (physical, psychological, social, or economic) occurring as a result of participation in a research study. Provide detailed information about the potential risks and benefits associated with this study and ways or approaches used to minimize any risk. When writing this section of the protocol, include a description of ethical considerations, issues, or concerns relating to the study. Identify the risks associated with the research for this study, as distinguished from the risks of therapies the subjects would receive even if not participating in the study. Identify the probable benefits to be derived from participation in this study. If the protocol includes vulnerable populations in the study, indicate any additional protections that are built in to the protocol to protect this population. Describe in detail how potential subjects will be provided with an accurate and fair description (during consent) of the risks or discomforts and the anticipated benefits. Include the informed consent form as an attachment, if applicable.
· Duration of the study
Treatment or Intervention
For prospective treatment and intervention studies, you must thoroughly describe the investigational product or intervention itself. Indicate the preparation required. Describe the procedures used to assign or randomize patients to a treatment and the process for blinding or masking the treatment. Describe who will administer the intervention, how it will be administered and where. Will the staff require certain credentials to administer the intervention? Describe the certifications required and training that will be provided to the staff. Describe how the inventory and supply are managed. Describe how the product will be stored. Are there any test results that will require termination of treatment? If so, describe the procedure. Describe any allowed concomitant medications. Does the subject perform any part of the intervention? If so, include a description of your plan to monitor the subject’s compliance.
- Study Aims
Describe the general intention of the research. Indicate your specific primary research question. State the research objective, aims, and hypothesis of your study. List clearly defined objectives that provide measurable endpoints. Some objectives are more important than others. Distinguish between them. Identify the objectives that are primary, secondary, or exploratory. Include the hypotheses to be tested.
- Study Procedures
The study procedures section can be lengthy. This section of the protocol maps out the tests and procedures that will occur in the study along with time points. Describe screening and recruitment, consent, laboratory evaluations, assays or specimens, Case Report Forms (CRFs) or surveys, standard-of-care evaluations or procedures, and subject payment or compensation. Include a schedule, timeline, study matrix, protocol table, flowchart, or study calendar describing the procedures, tests, and other activities that happens at each visit and when the subject receives treatment or assessment. This item is usually housed in the Appendix and referenced here in the study procedures section.
- Adverse Events and Reporting
In studies with human subjects, describe the frequency and intensity of any anticipated events. Describe how the study team should collect information about adverse events, serious adverse events or unanticipated adverse events. Describe how the event should be recorded, and how and when it should be reported to the sponsor, the FDA, and the IRB. Enter the forms to use for reporting in the Appendix. All investigators are required to have a way to address any subject complaints or issues. Indicate your plan for Principal Investigator availability and addressing these issues here.
Data Management and Analysis
This section of the protocol describes the plan for dealing with the data including the sample, capture, collection, quality assurance, integrity, analysis, and storage of the data. Provide specific details about what types of data will be captured and how the data is captured. Describe any electronic data capture procedures and the platform for data entry. Describe your sources of clinical data. Describe any data monitoring, quality assurance or quality control systems to be used. If such systems do not apply to your study, indicate that as well. Describe how you will address data confidentiality, any de-identification of data, storage, and security measures in detail. Describe the planned data analysis. Include a description of the what data points are collected and what data will be analyzed for each of the study objectives. Describe plans for any interim or subgroup analysis including what will be analyzed and the timing. Describe how any missing data will be handled. Indicate any data criteria for terminating the study. Include specific plans for storing the data, whether it is physical or electronic. Include a description of how the storage will ensure that the data are secure, and are maintained for the appropriate length of time. In instances when a key to identifiers is needed, include a description of how that key is maintained. Your Data Manager can assist in writing this section of the protocol and creating the Research Data Security Plan.
Tips for Specific Types of Studies
Describe the type of randomization to be employed, with further information about whether stratified randomization is needed. The stratification variable should be well-defined. Describe how the control group is derived. For example, if the researcher is enrolling patients on an investigational product during pregnancy and she wants to stratify on weeks of gestation at enrollment, the researcher should clearly define how weeks gestation. Describe how the subject is allocated to a specific treatment group or arm. Is the assignment of intervention versus control a 1:1 ratio? How are the arm assignments generated and by whom? Is special software used? Will this need to be concealed and if so, how will that happen? How will the randomization procedure be implemented and by whom? Are there contingencies if this may need to happen after hours?
Describe the blinding procedures including who is blinded to what. For example, is it just the participant who is blind to their assignment? Or investigators or others on the study team? Describe the criteria and procedures used to break the blind.
Provide a complete description of the investigational product. If drugs will be administered, describe the dose, frequency, route, and duration. Include the investigator’s brochure in the appendix. If the dose will be escalated over time, indicate how. If there is any specific toxicity levels that trigger a modification to the dose, indicate what and how the dose will be modified. Describe acceptable and not acceptable concomitant treatments.
For case control studies, look for an objective definition for the disease or outcome of interest that is used to define the cases. Typically, controls should be sources from a similar population as the active cases, the only difference being that they do not have the outcome of interest. Understand how this will be derived and make sure that obtaining that information is feasible. Matching is sometimes used to attempt to account for confounding. Outline the procedure for matching.
Blinding is frequently used in case control studies to minimize bias. Ensure that the protocol fully describes what is being blinded and the procedure for blinding. Other methods may be used to reduce bias and these should be described.
Completely describe your process for selecting the data to use in the study. List the criteria for considering studies for review. What are the study-level inclusion and exclusion criteria? Should only certain designs be selected? Describe the search strategies. List the databases to be used, key terms, and decision trees. For meta-analysis studies, describe how the different data from different studies will be extracted, converted to a unit of analysis, compared, and analyzed. If the abstraction is not clear-cut, how will resolutions be determined? How will you determine the unit of analysis? How will missing data be handled? The studies will likely be quite different from one another with regards to how data were collected, and how they were analyzed. How will this be handled?
In general, a quality improvement project does not need to be submitted to the IRB. Check the Quality Improvement Activities in Health Care Versus Research Policy for specific details of what is a quality improvement project and an accompanying checklist.
Often the Study Procedures section of a quality improvement protocol does not provide the level of detail required. List all actions of the study as the subject will experience it. For example: The subject will go to station 1 and complete a pre-assessment survey, then the subject will go to station 2 to view a video for the intervention, finally the subject will go to station 3 to complete a post-survey. All procedures will be completed in one subject visit.
Retrospective Chart Review
Ensure that you indicate how you are going to find the target subject population. List the specific ICD code, CPT code, or other criteria that you will use for the search. Indicate the specific databases or repositories that you will investigate. Specifically indicate calendar dates for each step of the process. List the specific fields of data collected and the specific repository or source for the field of data. Indicate if you are collecting data for deceased subjects. Include a list of sources such as databases that may have been used, key terms, and decision trees. Describe how charts or studies were selected for inclusion in the review or exclusion. If you are accessing images, consider the format of the images and how you will get it. If the images are in dicom file format, you may need to have the files copied to a disk. Be sure to indicate if you will destroy the disk after accessing the file.
If you intend to contact the subjects, you can provide the medical record number to the Decision Support Repository (DSR) group who will then provide the contact information for the subject. To follow up with the subjects by phone, be sure to include the telephone script for the phone discussion. The script ensures that the team are consistent in gathering the data which allows it to be analyzable.
DOCR provides the Research Summary Template to assist with survey protocols. Consider the amount of time required to develop the survey, validate the survey, make adjustments, pilot the survey (usually with your peers), and make adjustments again before you can use the survey in your study. Often, researchers don’t consider the duration of this task in the timeline required to complete their study. Therefore, they might write a protocol that is unrealistic. If you are going to design the study questions yourself, consider starting the study question development six months to a year before starting the study itself. It is unrealistic to think that a new survey tool can be developed, used in a trial, and published in one year.
One option is to use existing validated survey tools that gather the information that you need to examine. Your literature research should guide you toward tools that are validated and used regularly in the field. Consider selecting and using only certain questions from existing validated instruments. You may not need to use all of the existing survey and you don’t want the study subject to answer questions that aren’t needed. Using questions from existing validated tools saves time. If the survey will be conducted by phone, include a copy of a telephone script. Because a survey study doesn’t need a full informed consent, implied consent language is used in the survey itself.
Include specific details about how the data is collected and stored. If the data is stored on paper, the paper must be in a double locked location such as a locked cabinet in a locked office. Indicate who enters the data into a system to be analyzed and how that happens.
Duke Policies and Guidance
Contraceptive Use (IRB)
Emergency Use (IRB)
IRB Use of Consultants (IRB)
Pregnancy Testing (IRB)
Duke Tissue Exemptions List (DUSOM, Duke Pathology) - This site provides guidance about which specimens may be excluded from routine submission to surgical pathology.
You can find additional policies and guidance on the Duke University School of Medicine Institutional Review Board webpage.
Some pages and documents are located behind the Duke firewall and require a username and password to access. If you are a Duke employee or affiliate and are unable to access the information, please contact DOCR.Help@dm.duke.edu for assistance.
Forms and Templates
Quality Improvement Project Instructions and Template (IRB) - The DUHS IRB provides a template for the summary section of the protocol for a quality improvement project.
Training and Job Aids
Developing and Writing the Protocol - Web-based training available on demand.
Ask others to review your protocol. Consider requesting a protocol review from the Clinical Research Unit, your mentor, a subject-matter expert in the field, or the DOCR Outreach Team.
The Office of Corporate Research Collaboration - The OCRC drafts, reviews, and negotiates agreements for commercially supported research, both clinical and non-clinical, in the School of Medicine and related departments, and approves these agreements for institutional signature.
Consultation for Protocol Development for Studies Involving Mobile Technology - Are you planning a study involving mobile technology, such as smartphone apps or texting? DOCR study planning services, subsidized by the Duke CTSA and School of Medicine, are free to investigators and study teams and can help you navigate the nuances of protocol development that are particular to mobile health studies. If you are interested in setting up a planning meeting or have questions about this service, please email firstname.lastname@example.org.
Regulations and Guidance
Good Clinical Practice - (US FDA)
Drug Study Designs (FDA, 1998)