7 Practical Ways a Literature Review De-Risks an SaMD Pivotal Clinical Study

A strong literature review shapes SaMD study design, from comparators to sample size. Learn how it reduces risk in pivotal clinical studies.

A pivotal clinical study is often the single most expensive scientific activity in a SaMD's lifecycle. Most failures are not due to poor device performance, but because the study was not designed to demonstrate performance effectively.

One of the most underutilised tools for avoiding that outcome is a well-conducted and systematic literature review. Done properly, a literature review directly informs an SaMD’s study design decisions, which regulators will likely question.

Here are seven ways it does that:

1. It identifies the state of the art - which is usually your comparator

The state of the art refers to the current best available clinical practice for a given condition or clinical need. Understanding it tells you what your device needs to be measured against in your study - at minimum from a safety perspective, and often from a performance perspective too.

A robust literature review addresses key questions: What is the current standard of care? What clinical gaps exist? What level of performance has the current standard of care demonstrated? Without this, there is no objective basis for defining what "good enough" looks like for your device, and no defensible rationale for the comparator you propose in your study design.

2. It identifies similar devices, their study design, and findings 

Before designing a study from scratch, you need to understand what is already out there and how well similar devices perform. If cleared or approved predicates exist (for FDA 510(k)) or sufficiently similar devices have been evaluated under EU MDR, the clinical data surrounding those devices is directly useful. It tells you what study designs regulators have previously accepted, what comparator arms have been used, what populations were enrolled, and what endpoints were considered sufficient.

It also establishes scientific validity, a requirement under IMDRF SaMD clinical evaluation guidance, by demonstrating that the clinical association underpinning your device is scientifically established.

3. It informs sample size calculations

Sample size calculations are a common point of query from FDA reviewers and UK/EU notified bodies. A justification that appears optimistic or poorly supported will be challenged. More importantly, the fundamental risk is an underpowered study. A study that is too small can run to completion and still fail to detect a real effect, not because the effect isn't there, but because the study was never large enough to find it. You will have spent time, budget, and patient resources and still have no actionable result. Published studies on similar devices or clinical populations provide the baseline performance estimates and effect sizes that make a sample size calculation credible. 

4. It identifies performance metrics

A literature review helps you identify how similar devices with comparable intended uses have assessed their performance. This includes the specific metrics used to evaluate accuracy, safety, and any associated clinical claims or benefits. Aligning your evaluation approach with these established metrics allows for more meaningful comparisons between your device and those already on the market. It also ensures that your assessment framework is grounded in accepted practices, which strengthens the credibility of your results.

5. It helps you set acceptance criteria

The review also provides a benchmark for what level of performance is considered acceptable based on existing devices and clinical standards. This enables you to define clear acceptance criteria for your own product. For example, to demonstrate safety, you would typically need to show that your device performs at least as well as current clinical practice or human performance. Depending on your intended claim, you may need to demonstrate equivalent or superior performance to humans. If comparable products already exist, you should aim to meet or exceed their reported outcomes, and clearly justify why your product offers a clinical advantage.

6. It identifies the best ground truthing practices

Some SaMD pivotal studies will need a ground truth established by expert clinicians against which the device's output is compared. There are important factors to consider when establishing this ground truth, including the competence of the clinicians, and the risk of variability and bias when assigning ground truth labels.  A poorly defined ground truth runs the risk of affecting the outcomes of your study, and may also be contested by regulators. 

The literature can help to inform what ground truthing processes have been used in comparable studies, and whether those standards are considered acceptable by regulators. This is particularly valuable for newer SaMD categories where no universal gold standard has been established. In those cases, the literature is your main tool for identifying what reviewers will consider a credible ground-truthing process.

7. It identifies known risks, limitations, and safety signals

A literature review is not only used to demonstrate performance; it is also critical for identifying and understanding risks. Your clinical strategy should reflect awareness of harms, limitations, and safety signals that have already been observed in similar devices. Regulators expect this to be clearly captured in both your risk management file in line with ISO 14971and your clinical investigation plan.

In addition to published studies, a regulatory-focused review requires systematic searching of adverse event databases, safety notices, and global post-market surveillance data. This is a key distinction between a regulatory literature review and an academic one. For a more detailed comparison between these two types of literature reviews, see our “SaMD Literature Reviews Made Simple” blog. Evidence from similar devices may highlight adverse events, usability issues, or contraindications that are directly relevant to your product. Incorporating these insights early allows you to proactively design your study to monitor, mitigate, and appropriately report these risks.

The bottom line

The cost of getting study design wrong can render study outcomes unusable for a regulatory application and require a complete restart. A literature review, conducted systematically and before the protocol is finalised, is one of the most practical tools available for avoiding that outcome. Done well, it gives you a defensible, appropriately powered study design that is aligned with regulatory expectations from the outset.

If your company is navigating complex scientific or regulatory pathways for a digital health product, Hardian Health can help. Get in touch to discover how our expertise can support your journey from concept to clearance.