Endpoints: The foundation of meaningful evidence in medical device research
We recently ran an internal knowledge-sharing session on endpoints, something we address every day in our post market survey work. We took a step back to define their critical role in protocol and survey design, and how these feed into medical device evaluations and regulations.
In the context of medical device regulations, endpoints are embedded across protocols, Clinical Evaluation Reports (CERs), and post-market activities. Endpoints are predefined, measurable outcomes used to evaluate the safety and clinical performance of a device. They translate objectives into evidence, ensuring that the data collected can answer real clinical and regulatory questions.
There are three key types:
Primary endpoints: the main outcome tied directly to the primary objective and used for decision-making. Often statistically powered, and central to sample size justification. Frequently built around device performance (e.g. procedural or technical success).
Secondary endpoints: support the primary endpoint with additional insights. Often assess safety, usability, or other aspects of device use.
Exploratory endpoints: descriptive, used to explore trends or signals.
Endpoints shape the value of collected data
Endpoints determine whether data is meaningful or merely descriptive. They guide what is collected, how it is analysed, and how conclusions are drawn. In post-market settings, they align survey questions with safety and performance claims, support CER updates, and strengthen PMCF activities.
Without clearly defined endpoints, even large datasets risk failing to demonstrate whether a device is safe, effective, and performing as intended. In this way, endpoints do not simply support evidence; they define its value.
Endpoints don't start in a survey
Strong endpoints are built on evidence, not assumption. They are derived from literature, the State of the Art (SOTA), clinical studies, PMS data, and risk management documentation, and must align with the intended use and claims of the device. They don't start in the survey, they drive its design.
The CER plays a central role in identifying clinically relevant outcomes, which are then translated into measurable endpoints and operationalised through post market surveys and PMCF activities. This creates a continuous link between evidence generation and evaluation.
A critical pitfall: poor questionnaire design can undermine a good endpoint
One of the most common and costly challenges is not defining endpoints, but implementing them effectively. Poor question design can undermine a good endpoint. When survey questions are vague, subjective, or misaligned with endpoints, the resulting data may be abundant but ultimately uninformative.
This disconnect can prevent studies from addressing their core objectives and limit the ability to support benefit–risk conclusions or regulatory submissions. The lesson is simple: data only has value when it directly maps to measurable endpoints.
Strong endpoints, sharper evidence
One of the most common and costly challenges is not defining endpoints, but implementing them effectively. Poor survey and question design can undermine a good endpoint. When survey questions are vague, subjective, or misaligned with endpoints, the resulting data may be abundant but ultimately uninformative.
This disconnect can prevent a survey from addressing its core objectives and limit the ability to support benefit–risk conclusions or regulatory submissions. The lesson is simple: data only has value when it directly maps to measurable endpoints.