Study design in clinical and other clinical investigations of medical devices

The study design plays a fundamental role in clinical research and medical studies, which also include clinical investigations with medical devices as a special form, as errors in the design phase cannot be reversed later or can only be reversed at great expense.

This applies equally to all clinical-scientific projects.

It all starts with the study design!

The study design includes

• research question
• Study type
• Study population
• Observation unit
• Measurement method and
• calculation of case numbers.

The study design is determined at the beginning of a scientifically sound project, in the so-called conception phase, and determines whether the project can succeed at all. This also applies in particular to clinical investigations of medical devices or digital health or care applications (DiGA and DiPA). The regulatory framework is important for the choice of study design. This means whether it is a feasibility study, an authorisation study, a purely scientific study or a study after placing on the market.

Essential elements of the study design

As part of the study design concept, the guiding question and the reason for the clinical study are defined, e.g. with what effort which question is to be answered, which findings are to be gained in which persons under which circumstances. The research question then logically determines the type of study with a possible comparison to be tested against. This question must then be operationalised, i.e. made measurable and assessable.

A distinction can be made between primary and secondary data. Primary data are obtained from the studies themselves. Secondary data, on the other hand, is data that can be obtained, for example, from the literature review as part of a clinical evaluation. A clinical study is usually only conducted if the research of secondary data is not sufficient to answer a new question. A rough distinction can be made between observational, comparative, non-interventional and interventional clinical studies. The regulatory framework often determines which type of study is to be selected.

When selecting the study population, care must be taken to ensure that a group of people is included in the study that later corresponds to the actual group of people for whom the medical device or diagnostic is intended, the so-called target group. In addition to the underlying disease, concomitant diseases, gender and age, this also includes the institution (practice, hospital, general population, several different institutions) from which the study population is selected. This is important to ensure the transferability of the results and thus the external validity of the study.

In the context described here, the units of observation are individual people or a sub-collective or, for example, different skin areas in a clinical study and, in the case of systematic reviews, the individual studies. The choice of the observation unit is decisive for the interpretation of the study.

In order to answer the research question, the appropriate measurement method must be selected, consisting of measurement instruments and measurement methodology. In clinical investigations, it is important to consider the regulatory framework, feasibility, suitability and quality of the measurement instruments. These factors determine how precisely a study is interpreted. In addition, it is important to pay close attention to the number, time points, the time sequence, the reference to any influencing variables and the like in the measurements. The value of a measurement method also depends on the scale level, very simply whether the variables are nominal, ordinal or metric. The statistical method to be selected will then be derived from this later.

It is of course necessary to include a sufficient number of people in a clinical study. For ethical and economic reasons, the number of cases should not be too large. Therefore, a case number estimate is made with knowledge of an expected effect in order to answer the main or, if applicable, secondary questions.

It is essential that a clinical investigation is both internally and externally valid. Internal validity shows how reliable a result is in relation to the actual research question. It improves when the study is well planned, clear criteria for participation are defined and external influences are minimised. External validity means that the results of a study are applicable to the entire target group, not just the study participants. Both are scrutinised by authorities and later users.

Errors in study design cannot be corrected during and after the trial. It is therefore very important to design the appropriate study or clinical investigation very carefully and with sufficient time in the company of experienced specialists to ensure a meaningful, significant and feasible study.

At MEDIACC, we offer comprehensive support in developing a suitable study design. Feel free tocontact us for an initial non-binding consultation!