The sandwich principle has previously been applied successfully to lectures. In this teaching model, it has been proven that the application is more effective and leads to a significantly better learning outcome compared with classical lectures.  This study intended to investigate whether the application of the sandwich principle to instructional videos leads to improved knowledge gain and how its use is evaluated by students. Therefore, the students’ knowledge was assessed before they watched the instructional video and immediately after the lecture to assess their short-term retention of information. Six months later, a written test to evaluate the long-term recall of knowledge took place. Additionally, the students’ satisfaction with the instructional video and the modification was assessed.
A comparison of the results of the posttest and long-term retention test to the baseline level of knowledge revealed that both groups had significantly better results after watching the instructional video both times. When only the results of the posttest and the long-term retention test were compared, both groups showed a decrease in their mean score. Although this decrease was significant in both groups, the results of the sandwich group were slightly better than those of the control group. This outcome might indicate a beneficial long-term effect for the modified instructional video owing to the activating elements. This long-term effect could be analysed better through with an increased testing scope, a different testing format or a different choice of interval for testing long-term retention. 
In addition, there was no significant difference in the test scores between the sandwich and control groups. The results of the self-assessment likewise show that the students assessed their own knowledge equally good before and after watching the instructional video. Therefore, the objective and subjective results enable us to assume that the instructional video itself must have a good teaching effect. This notion is confirmed by the evaluation of both groups, who verified the advantageous didactic effect of the video itself. The hypothesis that the modification of instructional videos according to the sandwich principle leads to an improved learning outcome could not be proven. Therefore, whether the extra work of developing activating elements and editing the video for the modification is worth the effort must be discussed.
The evaluation showed that the students in the sandwich group highly appreciated the modified video. They found the interruptions useful for repetition of previously learned information. The participants also confirmed that the interruptions improved their concentration and attention to the video. They pointed out that they would like to learn in the future using instructional videos modified according to the sandwich principle. In general, the use of multimedia is highly appreciated by students and can be a powerful supplement and motivator to classical teaching formats. [16–18] A review by Green et al. found improved knowledge, skills performance and learner satisfaction using video-based training resources compared with non-video training groups.  This finding allows us to conclude that, from the students’ point of view, the modified teaching format is effective and indicated.
From the teachers’ point of view, applying the sandwich principle to instructional videos requires a great deal of work. Besides setting a special focus when establishing the learning objectives, the activating elements have to be created and integrated into the video. These preparations are time consuming, especially under the aspect of leaking a beneficial learning outcome. Nevertheless, once created, the modified instructional video is a sustainable teaching method that offers students a flexible, asynchronous study method by being independent of time, place and speed. [20–23] Although video production costs are high in the beginning, after several years of usage, digital learning has been shown to have lower costs due to the reduced need for institutional infrastructure and resources. 
In the present study, the length of the instructional video (45 minutes) can be regarded as a limitation. In general, shorter instructional videos are better for the attention span. Shell et al. found that the optimal length of instructional videos is 5–10 minutes.  According to Bunce et al. the attention span is 20–25 minutes in class, so that in this study the attention span was considered in the sandwich group by placing the activating elements after 15–20 minutes.  The length of the instructional video in this study can also be seen as an advantage because it provides better comparability of the effects of lectures.