Clinical Trial Design: Best Practices for 2026

When utilising human participants and their health history, the plan for how that data is collected, used, and protected must be rigorously defined, defensible, and built on informed consent from the outset. That is why a well-structured, strategic clinical trial design is crucial. It should clearly define the health questions being addressed, establish a clear consent and data governance framework, and protect patient data in line with relevant regulation (such as GDPR in the EU or HIPAA in the US), all of which is essential for securing approval from ethics committees and regulatory agencies.

This ultimate guide to the best practices for designing a clinical trial defines the core elements of any clinical trial design and offers insider advice for how to efficiently work with patient data in a compliant and resourceful manner.

In short

• What is it? Clinical trial design is the structured framework of protocols, methodologies, and data collection plans used to evaluate the safety and efficacy of clinical interventions.
• Core elements in 2026: A shift toward flexible, patient-centric, and data-driven methodologies, including adaptive trial designs, pragmatic clinical trials, and the integration of AI, driven in part by evolving regulatory expectations such as the EU MDR revision and the growing role of real-world evidence and registries in post-market evidence generation.
• How it is executed: Replaces traditional, rigid protocols with agile workflows managed via modern clinical trial design software, allowing sponsors to maintain direct control over study structures and variables.
• Importance of software solution: Modern no-code software eliminates CRO programming bottlenecks, reduces study setup timelines from months to days, and ensures real-time data validation and compliance.
• Who is Ledidi? Ledidi is a Norwegian software company founded by academic surgeons and software engineers. We provide a unified, secure cloud platform, comprising Ledidi Core, Ledidi Trials, and Ledidi Registries, that integrates data capture, real-time analytics, and global collaboration. The platform is built on a compliance-first foundation, including ISO 27001, HIPAA, and GDPR alignment.

What is a clinical trial design?

A clinical trial design is a strategic, structured plan for how to conduct a clinical trial. The design acts as the blueprint for effectively conducting the trial while also ensuring complete compliance throughout the process. It consists of specific protocols, methodologies, and data collection plans that answer research questions, ensure patient safety, and define the study parameters.

Common types of clinical trial designs

You can group most clinical trial designs into two broad categories. The first category is interventional designs (also known as experimental), and the other category is observational designs. The former covers studies that evaluate a specific medical, surgical, or behavioural intervention, while in the latter, researchers observe participants and record outcomes without active intervention.

Interventional trial designs and observational trial designs can be broken down into several types of design. The most common types to choose in a clinical trial are as follows:

Randomised Controlled Trials (RCT)

• What? This is known as the gold standard of clinical trial designs. It involves the random assignment of participants to either the experimental group that receives the treatment or the control group that is treated with a placebo or existing, standard treatment.
• Type of design: Interventional/experimental design.
• Relevant for these studies: Drug efficacy and safety studies, vaccine trials, surgical intervention studies, and any study where establishing a causal relationship between an intervention and an outcome is the primary objective.

Crossover Trials

• What? Each participant receives a sequence of different treatments over a set period.
• Type of design: Interventional/experimental design.
• Relevant for these studies: Chronic condition and pain management studies. Best suited to stable conditions where carryover effects between treatment periods can be adequately managed through washout intervals.

Factorial Trials

• What? Participants are randomly assigned to one of several groups, each receiving a different combination of treatments. This design allows researchers to assess the individual and combined effects of multiple interventions within a single trial population.
• Type of design: Interventional/experimental design.
• Relevant for these studies: Particularly useful when researchers need to assess whether two or more treatments interact with each other, or when running separate trials for each intervention would be prohibitively costly.

Single-Arm (First-in-Human) Studies

• What? All participants receive the same intervention, with no concurrent control group. Outcomes are typically compared against historical data, a pre-defined performance goal, or the patient's own baseline.
• Type of design: Interventional/experimental design.
• Relevant for these studies: Early feasibility and first-in-human medtech studies, Phase I Pharma studies, orphan or rare disease indications, and situations where randomisation is not ethical or practical due to small eligible populations.

Cohort Studies

• What? Researchers follow a defined group of participants over a set period of time to observe whether any risk factors or exposures correlate with health outcomes.
• Type of design: Observational design.
• Relevant for these studies: Epidemiological research, long-term disease surveillance, and occupational health studies.

Innovative clinical trial design: Moving beyond traditional models

For pharmaceuticals, the randomised controlled trial (RCT) has long been considered the gold standard due to how precisely it can be structured. Conditions are strictly controlled to ensure internal validity, including the establishment of cause-and-effect relationships, typically through comparison against a placebo. For this reason, these trials are often referred to as explanatory trials, designed to isolate the exact effect of a treatment under controlled conditions.

Medical devices rarely follow this same model. Blinding is often impractical, since an operator generally knows which device or technique is being used, and a placebo equivalent, a sham surgical procedure, raises ethical concerns that a placebo pill does not. As a result, device trials more commonly use a non-inferiority design against standard of care, or a single-arm study measured against a pre-defined performance goal. In the US, this often supports a 510(k) substantial equivalence submission or a PMA; in the EU, it supports a clinical investigation relying on equivalence to an existing device under MDR.

Moving from explanatory trials to pragmatic trials

Pragmatic clinical trials take a different approach. Rather than optimising for internal validity, they test effectiveness under real-world conditions. Eligibility criteria are deliberately broader, comparators reflect standard-of-care rather than placebo, and the outcomes selected are those that matter to patients and clinicians. External validity takes precedence.

Regulators are responding to this shift. The MHRA has developed its own real-world evidence framework, recognising that data generated through pragmatic designs and electronic health records can support regulatory submissions where traditional RCT evidence is impractical or insufficient. For pan-European submissions, the EMA's real-world evidence framework applies in parallel.

Pragmatic designs do, however, place greater demands on data infrastructure. Multi-site studies with heterogeneous patient populations and real-world workflows require software that can accommodate variation in data collection while maintaining the standardisation necessary for regulatory compliance. This makes the choice of clinical trial design software more consequential, not less. The ongoing Post-Market Clinical Follow-up (PMCF) required under MDR is, in effect, pragmatic real-world evidence generation, collecting data on how a device performs in routine clinical use rather than under trial conditions.

How to design a clinical trial?

The core elements of any clinical trial serve as the guiding principles for carrying out the design. When a sponsor is designing a clinical trial, they focus on four different core elements (PICO).

Population

The sponsor starts by defining the disease group being targeted for the study. The population should be clearly and specifically defined.

Intervention

The treatment, device, or procedure being tested is defined and applied to the study population.

Comparator

The intervention is measured against a comparator. This is commonly a placebo or standard treatment for pharmaceutical studies, while device studies more often use standard of care, an existing predicate device, or a pre-defined performance goal, since placebo comparators are frequently impractical or unethical for surgical or implantable devices.

Outcome

The objective, measurable results of the clinical trial are being evaluated.

The rise of adaptive trial design

Traditional clinical trial designs are fixed at the point of protocol finalisation. Sample sizes, dosing arms, and interim analysis schedules are agreed before the first participant is enrolled, and the design does not change regardless of what the accumulating data shows.

Adaptive trial design addresses this directly. By allowing pre-specified modifications to the trial based on interim data, such as adjusting sample sizes, dropping ineffective treatment or device arms, or changing allocation ratios, adaptive designs can identify effective treatments more quickly and with fewer participants than traditional fixed designs.

Common types of adaptations

Adaptations can take many forms depending on the goals of the study:

• Sample Size Re-estimation: If interim data suggests the initial sample size is too small to detect a meaningful effect, researchers can increase the number of enrolled participants midway.
• Dose or Parameter-Finding: In early-stage pharmaceutical studies, doses can be escalated or de-escalated based on how the first groups of patients respond to the drug. Device studies use an equivalent approach for adjustable parameters, such as energy delivery settings, catheter or implant sizing, or stimulation intensity.
• Dropping Arms: If multiple treatments or device configurations are being tested and one is performing poorly or showing severe toxicity or adverse events, it can be dropped early to shift patients to more promising options.
• Population Enrichment: The trial focus can be narrowed to the specific demographic, genetic, or anatomical subgroup that is responding best to the treatment or device.
• Seamless Stage Combinations: In pharma, two distinct phases of clinical research can be combined into a single, continuous trial. The FDA's adaptive design guidance for devices describes an equivalent approach, a seamless feasibility-to-pivotal design, where a study can transition directly from an early feasibility study into a pivotal study without stopping to launch a separate trial.

Benefits vs. challenges

Adaptive designs are increasingly favoured by regulators, including the FDA in the US and the EMA in the EU, because they maximise efficiency and safety. By halting unsuccessful trials early or moving successful treatments or devices to market faster, sponsors and researchers save time and resources.

However, the approach introduces severe operational and statistical complexities. If not planned carefully, mid-study alterations can introduce statistical bias and inflate the risk of false-positive results. Rigorous simulations and stringent statistical oversight are required to maintain trial integrity.

The role of AI in clinical trials: How it is streamlining clinical trial designs

Artificial intelligence (AI) plays a practical role in modern day clinical trial design. There are many use cases where AI ensures efficiency and improves the workflow for researchers and sponsors. They can use it for patient recruitment modelling, protocol feasibility analysis, and anomaly detection in real-time data streams.

Clinical trial startup is often delayed by the time it takes to design electronic case report forms (eCRFs), configure study workflows, and verify that data collection aligns with the study protocol. Ledidi's AI Form Builder streamlines this process by helping research teams create high-quality, protocol-aligned forms faster, reducing manual effort while maintaining compliance and data quality.

Faster study build

Instead of building forms field by field, the AI Form Builder helps transform protocol requirements into structured data collection forms. This enables teams to:

• Build eCRFs significantly faster than traditional manual configuration.
• Reduce repetitive setup work through AI-assisted form generation.
• Make updates quickly as protocols evolve during study planning.
• Accelerate the path from protocol approval to first patient enrolment.

Protocol-aware form design

One of the biggest challenges in study setup is ensuring that every required assessment, visit, and data point defined in the protocol is captured correctly. The AI Form Builder assists by checking forms against protocol requirements, helping teams identify potential gaps before a study begins. Key benefits include:

• Support for mapping forms to protocol-defined visits and assessments.
• Identification of missing or inconsistent data fields.
• Improved consistency between the study protocol and eCRFs.
• Reduced risk of protocol deviations caused by incomplete study configuration.

Clinical trial protocols serve as the blueprint for how studies are conducted and what data must be collected to ensure participant safety and scientifically reliable results. These capabilities help maintain data integrity and traceability while enabling rapid study configuration.

The result is a faster, more efficient study build process that allows sponsors, CROs, and research sites to focus less on configuring software and more on advancing clinical research.

Example: In silico clinical trials

In silico clinical trials see researchers use validated computer simulations and mathematical models to test drugs or medical devices before or during human trials. This reduces dependency on physical testing and accelerates development timelines.

Get a demo of Ledidi's AI form builder

Choosing the right clinical trial design software

Executing modern trial designs, whether they be adaptive, pragmatic, or multi-site, places specific demands on data infrastructure. The software must support complex branching visit structures, real-time data validation, and role-based access across multiple sites.

Legacy EDC systems were not designed for this level of operational flexibility. They typically require CRO programmers to configure the database and eCRFs from a written specification, a process that can take months. This introduces costs at every protocol amendment, and removes the sponsor from direct control of their own study design.

Ledidi Trials is built to address this directly. As a no-code, low-code platform designed for research teams rather than programmers, it allows sponsors and CROs to design, configure, and publish their own eCRFs and visit structures with minimal programming dependencies.

More importantly, Ledidi enables you to move beyond a single study view and manage your entire evidence programme in one place. It offers a single platform for all your clinical data needs that reflects that long-term strategic view.

Explore Ledidi Trials

A real-world example

In collaboration with Ledidi, the Cambridge-based robotic surgical device manufacturer, CMR Surgical, used our software solution, Ledidi Core, to ensure a data-driven clinical practice that would improve the quality of care and patient outcomes. This enabled them to evaluate the Versius robotic system in paediatric urology across three NHS clinical sites (Southampton Children's Hospital, Royal Manchester Children's Hospital, and Evelina London Children's Hospital).

By taking advantage of the Ledidi no-code designer, which allows for building forms that mirror their actual clinical workflows, they managed to save significant time and cost without compromising compliance.

Did you know?

Ledidi Trials has officially launched as a dedicated, specialised module for GxP-regulated trials. This serves as the primary software for regulatory clinical trials, designed for medtech, biotech, and Clinical Research Organisations (CROs).

Ledidi Trials was also independently audited and confirmed audit-ready in January 2026 by Thomas Pauly CISA / DHC, ensuring compliance with key regulatory expectations, including FDA 21 CFR Part 11, ICH E6 (R3) GCP (Good Clinical Practice), and ISPE GAMP 5.

Explore Ledidi Trials

Best practices for implementing a modern clinical trial design

Once the principles of the clinical trial design have been established, the theoretical foundation of the trial must be carried out in practice. Once sponsors execute the designs, success relies on whether the design is implemented correctly. The data integrity of the patient information and compliance with any relevant regulations or standards is critical.

When a sponsor is implementing the trial design in practice, they should abide by the ALCOA+ (Attributable, Legible, Contemporaneous, Original, Accurate, Complete, Consistent, Enduring, and Available) Framework for data integrity. This framework is the foundation for compliant and secure data handling.

Attributable

It is imperative that there is a clearly defined structure for the collection of data. Records identify the specific individual who performed the action, recorded the data, or made a change. This is typically tied to an electronic or physical signature and timestamp.

Legible and contemporaneous

The data collection should be accurate and executed in real-time at the point of care. All recorded data must be readable, understandable, and permanent for the entire duration of the required retention period.

Original

One should only collect and include relevant, primary data. The data must be the first recorded version (or a formally certified true copy).

Accurate and consistent

The design should operate based on predefined validation rules like range checks and logical consistency checks to catch errors at the point of entry.

Complete and enduring

All data should be securely stored in a cloud-based software solution. The software must include an immutable audit trail, allowing for data validation at any point.

Read more about the ALCOA+ Framework in our ultimate guide to eCRFs in clinical trials

Clinical trial phases

Each industry has different standards and procedures for carrying out a clinical trial. For that reason, the phases of each clinical trial differ based on what kind of study is being carried out. We explain the phases of pharmaceutical and biotech trials, medical device trials, and SaMD and IVD trials below.

Pharmaceutical and biotech clinical trial phases

• Phase 1 focuses on safety and finding the best dosage using a small group of participants.
• Phase 2 tests the treatment on a larger group to evaluate the effectiveness of the drug. Side effects are also identified here.
• Phase 3 confirms efficacy and monitors adverse reactions by comparing the new treatment against the standard of care in a large, diverse population.
• Phase 4 is the post-market surveillance of the drug to identify and track long-term effects and real-world safety after approval.

Medical device clinical trial phases

Feasibility, FIH, and EFS (Pilot Phase)

This initial stage gathers early safety and operational data in a small group of human participants. It includes Early Feasibility Studies (EFS) and First-in-Human (FIH) trials to test basic concepts and allow for modifications. In the US, this runs under an Investigational Device Exemption (IDE), while the EU requires an early clinical investigation under the Medical Device Regulation (MDR).

Pivotal evaluation (Pre-Market Phase)

This is a statistically powered, larger-scale trial designed to definitively confirm safety and effectiveness in a diverse patient population. Because blinding can be difficult with devices, performance is typically measured against the current standard of care or pre-defined performance goals rather than a placebo. The data generated serves as the primary evidence base supporting regulatory submissions, such as a US 510(k) or PMA, or an EU CE Mark.

Post-Market Clinical Follow-up and Surveillance (Post-Approval Phase)

Once approved, the device enters long-term monitoring to evaluate safety and clinical performance during real-world, routine use. This stage is crucial for detecting rare or delayed side effects that pre-market studies cannot capture. Under EU MDR rules, a Post-Market Clinical Follow-up (PMCF) plan is a strict mandate that must be actively maintained and legally justified.

SaMD and In Vitro Diagnostic (IVD) clinical trial phases

Scientific validity and analytical performance (Technical Phase)

This initial stage establishes the science and digital/technical precision of the tool before it interacts with real-world patient populations. It proves that the software algorithm or biological biomarker genuinely correlates with the target clinical condition, and verifies analytical performance using curated data sets, reference materials, or bench testing to ensure the algorithm or assay processes its inputs reliably.

Clinical validation and performance (Pre-Market Phase)

This stage evaluates the finalised digital tool or assay in its intended-use population to confirm it accurately identifies, predicts, or diagnoses the target condition in real patients. Performance is tested against a validated reference standard or a cleared comparator method. The data generated during this phase serves as the primary evidence base required for regulatory submissions, such as an FDA clearance/approval or an EU CE Mark.

Post-Market Performance Follow-up (Post-Approval Phase)

Once launched, the product enters continuous, proactive monitoring to evaluate safety, diagnostic accuracy, and software stability during routine, real-world use. This stage is critical for detecting algorithmic bias, software bugs, or rare clinical issues across diverse, unselected patient populations. Under EU frameworks, an active Post-Market Clinical/Performance Follow-up (PMCF/PMPF) plan is a strict mandate that must be legally maintained.

Read more about the phases of clinical trials

In summary: What characterises a successful clinical trial design

• Simple: Conceptually simple and operationally feasible.
• Relevant: Addresses clinically relevant questions.
• Ethical: Designed to protect participant safety and comply with GCP standards.
• Feasible: Achievable within available resources, sites, and recruitment timelines.
• Measurable: Built around clearly defined, objective primary and secondary endpoints.
• Reproducible: Standardised enough that results can be independently verified or replicated.

Watch: Realise your clinical trial design with us

The clinical trial design plays a vital role in the continuous clinical data lifecycle. With Ledidi, a research group can run a clinical registry on Ledidi Registries and a manufacturer-sponsored clinical trial on Ledidi Trials within the same secure environment. This can be done without migrating data or re-training staff.

Ledidi Trials demo video

Watch more videos of Ledidi in practice

How Ledidi can benefit your clinical trial design

In the medical device and biotechnology sectors, timelines dictate market success. Delays in clinical trial setup cost sponsors months of lost time, directly pushing back first-patient enrolment and increasing capital burn rates. Translating a written protocol into an Electronic Data Capture (EDC) platform has historically been the primary operational bottleneck.

The Ledidi Trials platform changes this dynamic by giving sponsors complete operational control over their clinical trial design. Operating directly within an intuitive, drag-and-drop browser workspace, research teams can independently create, configure, and publish protocol-aligned electronic Case Report Forms (eCRFs) without any external vendor dependency. This agile approach eliminates traditional IT infrastructure delays and removes hidden setup fees.

Our AI study builder meaningfully compresses study setup timelines by translating complex protocol parameters and written clinical investigation plans directly into structured study forms and data points, which reduces a process that traditionally takes several weeks or months down to minutes or hours.

Data governance and regulatory compliance are built into the backend architecture rather than layered on afterward. Automated system audit trails, version control, role-based access restrictions, and compliant electronic signatures ensure the infrastructure meets FDA 21 CFR Part 11 requirements in the US and Annex 11 requirements in the EU, alongside ICH GCP data integrity standards, from day one.

About Ledidi

Ledidi is a Norwegian software company founded in 2016 by two academic surgeons and three software engineers with extensive expertise in system architecture and cloud computing.

Frequently asked questions

What is the difference between an adaptive trial design and a traditional trial design?

A traditional trial features a fixed, unalterable protocol set from the beginning. Conversely, an adaptive trial allows for pre-planned modifications (such as altering dosage levels, dropping treatment arms, or adjusting sample sizes) while the trial is ongoing, based on real-time interim data analysis.

How does clinical trial design software improve compliance?

Clinical trial software supports participant and site compliance by automating protocol workflows, enabling eConsent, and maintaining audit trails and electronic signatures in line with regulatory requirements such as FDA 21 CFR Part 11. It tracks participant adherence to treatment regimens in real time, minimises data entry errors through built-in validation checks, and helps sites maintain alignment with Good Clinical Practice (GCP) guidelines throughout the study.

What are pragmatic clinical trials?

Pragmatic clinical trials (PCTs) are designed to evaluate the effectiveness of an intervention in routine, real-world clinical practice. Unlike strictly controlled traditional randomised controlled trials (RCTs) run in idealised conditions, PCTs feature broader eligibility criteria, operate in everyday healthcare settings, and focus on outcomes applicable to the general patient population.

What is the difference between MHRA and HRA in clinical trials?

In the UK, the Medicines and Healthcare products Regulatory Agency (MHRA) is the statutory body responsible for ensuring that medical products and trial protocols are safe, effective, and compliant with statutory regulations. The Health Research Authority (HRA), on the other hand, safeguards the ethical standards, rights, and well-being of trial participants. While the MHRA assesses scientific safety, the HRA handles ethical governance and reviews.

How does Ledidi's AI study builder speed up study design?

Ledidi's AI Form Builder turns protocol requirements directly into structured eCRFs and study workflows, cutting a process that traditionally takes weeks or months down to minutes or hours while checking for gaps against the protocol as it builds. Human oversight stays central: the AI runs in foreground mode, so a person must initiate every interaction and review and approve each AI-generated change before it takes effect, with all actions logged in the project's audit trail. AI features are also opt-in only, requiring activation at both the organisation and project level, so no study uses AI unless a sponsor has actively enabled it. Full details are in Ledidi's Data Processing Addendum.

Sources

• Fergusson D et al. Five good reasons to be disappointed with randomized trials. PLOS Medicine / PMC.
• Zwarenstein M et al. Pragmatic vs explanatory trials: the Pragmascope tool to help measure differences in protocols of mental health randomized controlled trials. PMC.
• Dalrymple K V Real World Evidence Versus Randomised Controlled Trials: Is the Future of Nutritional Sciences Research in Electronic Health Records? PMC.