Once the initial question has been set, then the process of protocol development can begin. The protocol turns the research question into a study. The protocol should be practical and achieves the balance of ensuring that an accurate answer can be obtaining in a safe, ethical and practicable way in the chosen clinical setting.

A protocol is a study plan, designed to describe the objectives, background, methodology, organisation, participants, procedures and assessment tools of the clinical trial. It is the recipe for the clinical research study and ensures that all researchers perform the trial in the same way. A well designed and clearly written protocol yields good data. The protocol describes the top-level requirements. The standard operating procedures (SOP) then describe precisely how these steps and tasks will be conducted, which is particularly important in multicentre studies. Every site should work from the same version of exactly the same protocol, but the standard operating procedures then set out how each step is conducted in each centre (ideally this is the same in every site, but some things could differ based on logistical differences). For example, the protocol might say ‘2ml sample collected and analysed in the laboratory’, and the SOP would then describe precisely how this happens at each site to ensure consistency and uniform data.

The process of developing a protocol is a collaborative one, with the involvement of multiple contributors. To achieve a well-designed protocol, input might be required from a number of sub-specialists, including physicians, nurses, data managers, regulatory experts and statisticians.

Protocol development is a cyclical process of review and modification by the contributors. A single central person should have responsibility for the coordination of the process, circulating the drafts for review and collating the comments. This person should also have responsibility for integrating contributions, maintaining a master file, and controlling file access during the draft process. A process for deciding on accepting or rejecting comments or suggestions should be established. It is also advisable to document the outcome of discussions and agreed decisions, in case the issue needs to be revisited at a later date. A method of allocating version numbers and version dates to protocols during the drafting process is also advisable.

The overall aim of the protocol, which is a helpful concept to check along the way, is ‘is the protocol answering the question that is set’? One of the hardest, yet most important steps is deciding on the primary objective. This should be the single key question the study is asking. The next key step is agreeing how that primary objective will be measured. This is your primary endpoint. For example the study might be asking ‘Is drug A better than drug B in treating Leishmaniasis?’ So what is the primary objective? Clearance of parasite? Reduction in symptoms, discharge from hospital? Quality of life? This step often requires much discussion within the field. Then once this is agreed, you need to think about how you are going to measure this primary outcome. If you settled on ‘resolution of symptoms’ what exactly are you measuring and can it be done consistently?

A protocol for a trial of an Investigational New Drug (IND) must follow the format outlined in Item 6 “Protocols” Section of the ICH Guideline for GCP . Although not a requirement for all trials, it forms an internationally accepted basis for preparing a study protocol and is a helpful starting point. Template protocols are available here (https://globalhealthtrials.tghn.org/resources/downloadable-tools-and-templates/) and there is also a helpful protocol development tool (PRACTIHC tool http://www.practihc.org/index.htm). At the very least the ICH-GCP outline provides a useful checklist for the items that should be considered for inclusion in a study protocol. If the study is observational, or the intervention is not an investigational product, then many of the items can simply be missed. However you might wish to add further headings, depending on the nature of your study. Examples of helpful sections are ‘lay summary’ at the beginning, or a section on ‘community engagement’.

The essential issues that should be included in a study protocol for an interventional study are presented below and are based on the ICH GCP guidelines. For non-interventional studies, some sections will not be applicable (e.g. treatment of subjects) and can be omitted or modified where appropriate.

Title page

• The title should be informative and indicate the nature of the study and design.
• A concise but informative statement of the purpose of the trial.
• The names, roles and contact details of physicians and/or researcher(s) responsible for conducting the study as well as the sponsor details where appropriate.
• The protocol effective date and version number.

Background information

• A detailed explanation of the research question, i.e. the hypothesis to be tested.
• A justification for the trial including results of any relevant literature search in the area of study
• A description of the study population, the indication, the disease prevalence and any current treatments available.
• A description of the treatment under investigation and any previous studies or data available
• Study objective and purpose
• Objectives should be precise and measurable by endpoints
• There should be one main or primary objective of the study that reflects the aim of conducting the study. The primary objective should dictate the design and methods of the study.
• If several objectives exist, then these should be divided into primary and secondary objectives
• Study design
• Experimental design (e.g., parallel, crossover, single group).
• Control method (e.g., placebo, active comparator, low dose, historical, or none [uncontrolled]).
• Level and method of blinding (e.g., open-label, single-blind, double-blind, matching placebos, or "double-dummies").
• Treatment assignment method (e.g., randomisation, stratification, or both).
• A definition of each study period and stage as necessary, including sequence and duration.
• Total duration of the study including duration of the active treatment period(s), screening or run-in period(s), and any follow-up or washout period(s).
• Dose regimen(s) in each study period and stage (if applicable), including frequency of administration.
• Method and criteria for individualising dosing (e.g., subject weight or plasma concentrations).
• Rules for dose changes, including flexible dosing; modifications in dosage and frequency of dose changes or adjustments; dose reductions, interruptions or tapering and permanent discontinuation for reasons of efficacy or safety, re-challenges, and any circumstances for resuming investigational product.
• Definition of subject completion. This is usually completion of all phases of the trial including any follow-up.
• Justification for the study design including the choice of control and duration of treatment, dose of investigational product and comparator/control group(s). Discussion of suitability of design to achieve the objectives and any known or potential problems of the design.
• Selection and withdrawal of subjects
• As relevant, the number of subjects to be enrolled, randomised and evaluable based on stated assumptions of attrition/withdrawal, or both. The number of subjects (sample size) should be targeted to the primary objective of the study.
• The population from where subjects will be drawn and, where applicable, how they will be approached and recruited.
• A list of inclusion and exclusion criteria that subjects must meet to be eligible for enrolment in the study. These may include sex, age range, disease duration and severity etc.
• Predetermined reasons for withdrawal of subjects or withdrawal of investigational product. Information on the duration of required follow up period, type and timing of data to be collected for withdrawn subjects and whether withdrawn subjects are to be replaced.

Treatment of subjects
For non-interventional methodology studies, this section can be omitted. For studies involving medicinal products include the following information as applicable:

• Description of study drug: chemical name; trade name; relevant pharmaceutical properties (e.g., type of salt, free base, or free acid); dosage form; unit dose strength.
• Description of dose regimen(s) in each study period and stage, including frequency (e.g., BID), route of administration, and duration of treatment (total and by study period)
• Method for assigning subjects to treatment and blinding of the study drug (patients, assessors, physicians). Discussion of any bias-reducing procedures when randomisation is not used.
• Dosing instructions (e.g., before/after meals, time of day, time before study visits, missed doses).
• Method for individualising dosing, including criteria used (e.g., subject weight or plasma concentrations).
• Rules for dose changes, including flexible dosing; modifications in dosage and frequency of dose changes or adjustments; dose reductions, interruptions or tapering and permanent discontinuation; and any circumstances for resuming investigational product, if applicable.
• Details of who will supply the study drug, the packaging and labelling of the product
• Description of any preparation of study drug to be performed at the site along with the handling and storage requirements.
• Description of procedures for study drug dispensing records, accountability, and disposal procedures during the trial. 
• Description of assessment of subject compliance with study drug administration and any protocol-required compliance criteria.
• Any medication and non-protocol defined treatments that can be administered during the study. Medications permitted during the study and any prohibited during the study or at certain times during the study.
• Arrangements for continuation of treatment for subjects after the end of the study.

Assessment of efficacy

• Primary efficacy and any secondary efficacy endpoint(s)
• Discussion of relevance of the measurement(s) if needed.
• Description of the methods and timing for assessing efficacy data. Any methods/training to ensure consistency across centres should be included.

Assessment of safety

• Identification of specific variables for each type of safety assessment
• How to perform, collect and record each assessment
• Methods used to obtain and/or interpret the assessment
• Criteria for any rating scales used to classify laboratory or other safety assessments.
• Guidelines used for management of relevant laboratory or other safety assessment abnormalities
• Techniques used to standardise or compare laboratory results or other safety assessments
• Identify any non-investigator party responsible for evaluation of laboratory or other safety assessments and describe any procedures used, including centralized reading/measurement
• Define an adverse event (AE), a serious AE, drug-related AEs and severity of AEs. Define any disease-related outcomes that do not need to be reported as (S)AEs.
• Define the time period for detecting AEs and SAEs for each phase of the study, how they will be detected and recorded, the type and duration of follow up.

Statistics
This section of a protocol is likely to be written by a statistician. The key issues that need to be included in this section are:

• The hypothesis being tested
• Sample size determination detailing methods of determining sample size, power calculations and clinical justifications
• Variables to be used to assess outcomes and how these data will be reported
• Detailed plans for the statistical analyses of the primary and secondary endpoints
• Plans for handling missing data and drop-outs
• Data handling
• Study stopping rules

Direct access to source data/documents
The protocol should state the investigator will permit trial-related monitoring, audits, ethics committee review, and regulatory inspections (if appropriate) by providing direct access to source data and documents.

Quality control and quality assurance
Details of the quality assurance and quality control systems to be used during the study to assure the quality of the data, for example:

• Investigator training sessions
• Instruction manuals
• Monitoring procedures
• Auditing procedures

Ethics
The protocol should state that the study will be conducted according to the Declaration of Helsinki; that the ethical aspects of the study will be reviewed by an appropriately constituted ethics committee, and that each subject will give their informed consent to participate in the study. If there are any local guidelines for the ethical conduct of a clinical trial then these should be provided. The protocol should also include the rights of subjects to withdraw from the study at any point.

Data handling and record keeping

• Description of procedures for data collection and recording
• Description of coding and processing of data, including by whom and software used
• Validation and data cleaning

Financial and insurance matters

• Details of the finance and insurance for the trial may be included in the protocol or in a separate agreement between the sponsor and the investigator.

Publication policy

• The publication policy should be included in the protocol if not addressed in a separate agreement.

Supplements/Appendices
Example of documents to be included (where relevant):

• Subject information sheets,
• Consent forms
• Data collection forms
• Summary of product
• Questionnaires

Reference information:1 http://www.ich.org/fileadmin/PublicWebSite/ICHProducts/Guidelines/Efficacy/E6R1/Step4/E6R1Guideline.pdf

The SPIRIT group have also developed a useful checklist for protocol writing. http://www.spirit-statement.org/