1. Evidence-based medicine

For centuries, the medicine has been based on the knowledge and experience of healers. However, as research progressed, individual approaches to treatment were increasingly complemented by scientifically based treatment concepts. In 1990, the term "evidence-based medicine" was coined in the Anglo-American world. This approach found its way into the German-speaking world in the mid-1990s.

1.1. Definition

Evidence-based medicine is the conscientious, explicit and judicious use of the best current external scientific evidence to make decisions about the medical care of individual patients. The practice of evidence-based medicine involves the integration of individual clinical expertise with the best available external evidence from systematic research.

Evidence-based medicine therefore aims to analyse and evaluate scientific knowledge and evidence from different sources according to clear methodological principles that are internationally and nationally accepted. The focus is on the results of clinical research. The result is, for example, high-quality guidelines from which clinically active doctors can extract the best possible "knowledge ingredients" for a clearly justified decision in individual cases.

The German Network for Evidence-based Medicine (www.ebm-netzwerk.de) has been in existence since 1998, and Cochrane Germany was founded in 1999 (www.cochrane.de). Cochrane is considered to be the first and most important global network for solving the problem of the growing flood of information in medicine. It aims to make the transfer of knowledge from clinical research to clinical practice more transparent and easier by systematically searching for, evaluating and presenting the results of trials in an easily accessible way.

The principles of evidence-based medicine serve as the authoritative basis of assessment for the Joint Federal Committee, which decides on the eligibility of statutory health insurance members in Germany. The principles of evidence-based medicine have also become the standard for decision-making in infection prevention. The Commission for Hospital Hygiene and Infection Prevention at the Robert Koch Institute develops recommendations for hospitals, nursing homes and doctors' surgeries on various types of infection, such as post-operative wound infections, pathogens such as C. difficile or other issues such as hand hygiene. In doing so, it fulfils the task assigned to it by Section 23 of the German Infection Protection Act to develop recommendations for the prevention of nosocomial infections in hospitals and other medical facilities based on the latest medical knowledge.

1.2. Approach

In addition to individual clinical experience and the values and wishes of the patient, the current state of clinical research is an essential pillar for the optimal treatment of patients. The process of evidence-based medicine is divided into five steps.

1.2.1. Defining the research question

In this step, the clinical problem is translated into a question to be answered by scientific research. In the context of COVID-19, the questions can be formulated in very different ways, thus influencing the answer. The wearing of masks is an example of this.

"Do masks protect?

This very general question is often asked in the media. It is a yes-no question that is inappropriate from a scientific point of view. This is because a protective effect is usually variable and can be very low or very high. In addition, it does not distinguish which masks are meant, whether the mask protects the wearer or the other person, what the protection should be against (e.g. droplets, specific pathogens or infections) and the environment in which the protective effect is expected. If you are talking only about droplet filtration, you may even be able to attribute a certain level of protection to certain community masks.

"Do surgical masks protect against SARS-CoV-2 viruses?

This question is more nuanced because it defines a type of mask and the protective effect is related to a specific virus. To answer this question, it was possible to analyse all the studies that examined the filtration performance of surgical masks against SARS-CoV-2. However, it remains unclear whether the protective effect relates to the person trying to protect themselves or to a person who is a SARS-CoV-2 carrier trying to protect others in their immediate environment. It is also unclear whether the surgical masks are intended to protect against actual infection or against COVID-19 without knowing the associated symptoms, which are not the same thing.

"Does wearing surgical masks in shops protect against COVID-19?

This question may sound similar to the previous one, but it refers to the disease COVID-19 rather than the virus itself. As a result, the question can only be answered by using studies that have investigated a health benefit from wearing surgical masks (fewer COVID-19 cases). In addition, the setting is clearly defined (shop). Trials in which the health benefit of wearing surgical masks was investigated in a COVID-19 ward in a hospital are therefore irrelevant for this question. This is because the exposure risk in a shop is completely different (few virus carriers, short contact times, little face-to-face contact, little talking) from that in a COVID-19 ward. However, this question still leaves open the question of whether the mask wearer or other people should be protected.

"Will wearing a surgical mask in a shop protect me from COVID-19?

In addition to the previous question, this question also narrows down the person for whom a protective effect is defined.

"To what extent will wearing a surgical mask in a shop protect me from COVID-19?"

Finally, the question may include a quantification of the protective effect to describe the magnitude of the expected health benefit. At this point, the answer can no longer be a simple "yes" or "no". The result can now range from 100% (complete protective effect) to 0% (no protective effect).

1.2.2. Systematic literature search

In this step, the scientific literature is searched using a comprehensible strategy and then systematically analysed. The aim is not only to consider literature that supports a particular hypothesis, but also to obtain an undistorted, unbiased overview of the current state of knowledge on specific issues.

Firstly, search terms (e.g. 'face mask' and 'protection') and databases of scientific literature are defined (e.g. 'PubMed'; the largest publicly available database of scientific medical literature). The definition of the search terms is very important as the results of the search depend heavily on the choice of terms used. You will then receive a list of results. This may include hundreds or thousands of publications, and in science only published results count. The search terms may have been too general and need to be refined. As part of the literature search, you may be able to reduce the size of the publication list by refining your search terms. However, no studies are excluded at this stage on the basis of defined formal or content criteria.

1.2.3. Critical appraisal of evidence

The next step is to define the criteria for including or excluding studies from the analysis. If only studies with SARS-CoV-2 are searched for and some studies with SARS-CoV-1 are displayed in the publication list obtained in the first step, the latter do not fulfil the inclusion criteria and are therefore not included. To make the search comprehensible and reproducible for all other researchers, the total number of hits and the number of excluded studies, including the reasons, are usually reported. This is the only way to ensure that other researchers can perform the same search if they have doubts about the results.

All trials that meet the inclusion criteria are then analysed, ideally by two independent scientists. This is often done using a standardised procedure that all reviewers use to assess the strengths and weaknesses of each study's design. Once the review is complete, the results are summarised and graded according to their strengths and weaknesses, for example using the internationally recognised GRADE criteria.

The evaluation of the study design is particularly important when considering the causality of a risk factor or intervention. Observational studies (cohort studies or case-control studies) are often conducted in which the benefit of an intervention is determined retrospectively for a specific time or in a specific setting, such as a hospital. The same data are collected on a suitable control group for comparison. However, observational studies are only suitable for proving causality of an intervention under certain conditions [1]. Therefore, if fewer SARS-CoV-2 transmissions were measured when masks were worn in one period compared with another, it is not certain that this effect can be explained by the use of masks.

Prospective randomised controlled trials (RCTs) are generally better at establishing the causal relationship between an intervention and its effect.

The quality of the evidence therefore depends mainly on the study design. Systematic reviews, in which as many randomised controlled trials as possible have been analysed (e.g. Cochrane reviews), provide the greatest certainty about a topic.

1.2.4. Evaluating the specific situation

Since medicine is often about the most promising treatment option, the treating physician should then weigh the knowledge gained against the specific clinical situation of his or her patient and assess which treatment or prevention concept is likely to provide the best benefit for this...