As current prevalence rates show, around 40% of the population in Germany over the age of 16 can be considered chronically ill (Bundesministerium Für Gesundheit et al., 2021). This is reflected in high rates of hypertension, diabetes, chronic obstructive pulmonary disease, cardiovascular disease, chronic back pain and obesity (Rathmann et al., 2018; Hoebel et al., 2019; Güthlin et al., 2020). In addition to these physical diseases, prevalence rates also include mental disorders, the most common of which are anxiety, affective disorders and addictive disorders (Jacobi et al., 2014). The impact of chronic diseases is associated with a high burden of disease, reduced quality of life for patients and high health care costs (Statistisches Bundesamt, 2017).

Triggers and perpetuators of chronic disease include unhealthy lifestyles and poor health choices, which often lead to a complex treatment process and complicate doctor-patient interactions (Frost et al., 2018). Unhealthy behaviors are often difficult to change or abandon due to long-standing habits and short-term positive effects. Individuals therefore experience strong ambivalence caused by the simultaneous occurrence of conflicting desires, thoughts or feelings for or against behavior change. Behavior change is crucial in the recovery process of various mental and physical illnesses, as a healthy lifestyle is associated with a significantly reduced risk of chronic disease (Ford et al., 2009). Thus, it is possible to prevent or positively influence the development of a chronic disease by changing behavior toward a healthier lifestyle, e.g., through more exercise and a healthier diet (Hu and Willett, 2002; Vainio et al., 2002) or abstaining from nicotine (Mannino and Buist, 2007). Furthermore, internal and external motivation can be considered key variables for behavior change in patients (Ryan and Deci, 2017).

Healthcare workers and physicians should therefore be involved in improving patients’ motivation to change, as professional support can facilitate active participation in managing their own health and making decisions that promote self-care (Barbosa et al., 2021). But how can physicians support their patients in changing a certain behavior—especially when, due to its positive short-term consequences and its habitual character, the behavior in question is hard to give up? The ambivalence which arises from these conditions often prevent patients from making a decision.

Addressing this issue, Miller and Rollnick (1991), Miller and Rollnick (2004), and Miller and Rollnick (2012) developed motivational interviewing (MI). This approach provides well-defined negotiation techniques as well as a specific attitude, called the motivational interviewing spirit, that help build a reliable physician-patient relationship as well as evoke and enhance patients’ intrinsic motivation for behavior change (e.g., utilization of an indicated medical treatment). Notably, MI aims to foster patients’ intrinsic (vs. extrinsic) motivation. Therefore, performing MI does not imply convincing the patient to do what the physician thinks is best. Especially in the presence of ambivalence, confronting patients or persuading them toward a change in behavior tends to provoke arguments against that very change. Such a confrontation highlights the advantages of the status quo and the disadvantages of the behavior change, making actual behavior change less probable.

In contrast, practicing MI means supporting the patient in identifying behavior changes that are desirable from their perspective as well as in finding out which motives or aspects they perceive to be helpful in pursuing the specific change. According to Miller and Rollnick (2012), the interviewer should take into account the following ethical values: nonmaleficence, beneficence, autonomy and justice. These values provide the ethical framework for MI. This framework implies, for example, that the interviewer does not support the patient in pursuing harmful behaviors, but that the interviewer accepts if the patient decides against a healthy behavior change. Thus, MI can augment the traditional role of a physician as an expert—someone who informs the patient of the diagnosis and recommends treatment options—and can open up a communication style that focuses on the patient’s perspective, particularly when a patient shows ambivalence regarding a certain behavior change.

MI was initially developed for the treatment of addictive disorders. Since then, the field of application has gradually been extended to other mental and physical diseases. In a Cochrane Review (Smedslund et al., 2011), MI was shown to significantly reduce the amount of alcohol consumption. In comparison to other interventions investigated in that study, MI was equally or more effective. Positive effects of MI have also been shown in the treatment of eating disorders (Treasure et al., 1999; Hettema et al., 2005; Cassin et al., 2008; Geller and Dunn, 2011; Westra et al., 2011) as well as overweight and obesity (Van Dorsten, 2007; Armstrong et al., 2011; Teufel et al., 2011).

In a meta-analysis (Lundahl et al., 2013), MI was found to be associated with significant effects regarding behavior change in several health-related behavior categories. Among others, MI enhanced the participation of pain patients in workshops imparting pain management strategies (Habib et al., 2005), reduced risk behavior in adolescents who previously had been treated due to an injury (Johnston et al., 2002), increased health-promoting behaviors in adolescents with HIV (Naar-King et al., 2008), improved the attitude of patients with type 2 diabetes toward health-promoting behavior changes (Rubak et al., 2009) and reduced regeneration of caries mediated by mothers’ preventive behaviors (Weinstein et al., 2006).

These findings demonstrate the usefulness of MI for widespread use in the health sector and the associated need for knowledge transfer about MI.

Training in basic communication skills is already an integral part of medical training at various faculties (Smith et al., 2007). In addition, the use of simulated patients for in-depth practical training is widely accepted and perceived to be of great value for students (Kaplonyi et al., 2017), as is the use of peer role play as a simulation-based training method (Gelis et al., 2020). Both the German Master Plan 2020 for medical education and the national catalog of competency-based learning objectives for medicine reinforce the importance of doctor-patient communication skills, including MI (Wissing, 2018).

Research has demonstrated the success of teaching MI techniques in the medical field, both in the training of medical students and in the training of fully qualified doctors (Frost et al., 2018; Kaltman and Tankersley, 2020). A review by Kaltman and Tankersley (2020) found that participation in MI courses led to increased knowledge and improved practical skills in medical students. A recent study by Jacobs et al. (2021) showed that MI training in pre-medical students has a positive effect on knowledge and specific skills. Another review by Dunhill et al. (2014) showed that training in MI is particularly effective when it is “intensive,” i.e., when it involves several sessions or is integrated into a longitudinal curriculum, and when interactive exercises are an integral part of the training.

Several studies were able to show that MI can be successfully taught through training and workshops (e.g., Evangeli et al., 2009; Britt and Blampied, 2010; Walters et al., 2010) for an overview see Schwalbe et al. (2014) and that training MI improves knowledge and practical skills (Madson et al., 2009). In terms of didactic modality, MI is best learned in communication workshops with feedback and targeted coaching (Miller et al., 2004) while focusing on the underlying assumptions and spirit of MI rather than on specific techniques (Miller and Moyers, 2006). Embodying the spirit is associated with an increased likelihood of internalizing other relevant MI values and behaviors such as acceptance, respect for autonomy, empathy and warmth (Moyers et al., 2005). The reported prevalence and high use of MI in the health care system highlights the need for MI to be taught at an early stage of medical education.

Blended learning describes the combination of face-to-face teaching with online materials and courses (Delialioglu and Yildirim, 2007) and has been shown to have better outcomes in terms of knowledge acquisition compared to traditional learning and online-only learning in health education (Morton et al., 2016; Westerlaken et al., 2019; Vallée et al., 2020).

The benefits of both formats, such as the efficiency and flexibility of online materials and the interaction with peers and tutors afforded by in-person learning, can be combined in one approach. Meta-analytic results suggest that the use of blended learning in medical education and training is significantly superior to traditional teaching in terms of growth in theoretical knowledge (Vallée et al., 2020) and satisfaction with the curriculum (Li et al., 2019).

In line with the need for more high-quality research on MI in education (Frost et al., 2018), our study integrates the factors identified in the literature as helpful and practical, such as workshops with feedback and targeted coaching, spirit in the early stages of medical education. We complement the literature by implementing our curriculum in a blended learning format, both to optimize time management and to ensure that it can be used independently of the face-to-face seminar.

To the authors’ knowledge, the only published German study on MI training for medical students in a blended learning format is the pilot study of the present study (Keifenheim et al., 2019). In the pilot study, significant improvements in subjective and objective knowledge as well as (subjective) practical skills were achieved after the first of three parts of the MI curriculum. The present study investigates the success of MI training in a blended learning format (i.e., a combination of lecture, simulation patient videos, face-to-face practical sessions and role-play scenarios) for medical students in their 6th to 9th semester onwards as part of their mandatory medical courses.

The research questions were:

Q1: Does participation in an MI curriculum offered in a blended learning format lead to an increase in students’ subjectively rated theoretical knowledge of MI?

Q2: Does participation in an MI curriculum offered in a blended learning format lead to an increase in students’ subjectively assessed practical skills in MI?

Q3: Does participation in an MI curriculum offered in a blended learning format lead to an increase in students’ objectively assessed theoretical knowledge of MI?

This evaluation was conducted at a German medical university faculty among medical students between the 6th and 9th semesters, prior to the COVID-19 pandemic. Using a longitudinal pre-post design, students were compared with assignment to the intervention or control group (non-randomized) and the two time points of measurement (T0/T2) as independent variables. Only data from subjects who met the following criteria were used for analysis: (i) completed questionnaires at T0 (6th-semester medical students) and T2 (9th-semester medical students) and (ii) participated in at least two of the three videos and two of the three practical components.

The MI curriculum consisted of three parts (MI 1–3). While the first part focused on MI rationale, MI spirit and the physician-patient relationship, the second and third parts concentrated on specific MI techniques. MI spirit was briefly summarized at the beginning of MI 2 and MI 3 and constituted the basis of all role-play interactions when training specific MI techniques. In order to show and train students on different uses, each part of the curriculum highlighted a specific field of application (eating disorders, addictive disorders and health-promoting behavior).

The materials were prepared by a psychotherapist and resident psychiatrist who has acquired theoretical and practical knowledge through literature studies and several MI workshops, as well as several years of experience in the application of MI in addictive disorders, health-promoting behaviors and eating disorders. MI trainers were psychotherapists and physicians with at least 1 year of professional experience. They studied MI literature, curriculum videos and instructions for the practical sessions. In addition, they received a practical training of about 3 h with the psychotherapist or physician. Throughout the course of the curriculum, trainers were supervised by the psychotherapist or physician. Table 1 summarizes the content of the curriculum.

The link to the online questionnaire was sent to 6th-semester medical students of the respective medical faculty (control group in the winter semester, intervention group in the summer semester) (time T0). The intervention group then took part in a compulsory first course of the newly introduced curriculum on MI (MI 1).

The course (3 units, 135 min) on MI was integrated into an existing seminar (“Psychosomatic Medicine and Psychotherapy”) as one of six sessions. In advance, the students were asked to watch online teaching and demonstration videos (2 units, 90 min). In the so-called “teaching videos,” theoretical knowledge was conveyed to the students in the form of a recorded lecture. For the “therapy demonstration videos,” example conversations (with simulation patients) were filmed in which a doctor applies MI. In the 8th and 9th semesters, the students in the intervention group took part in further seminars on MI (MI 2 and MI 3). Here, too, a 1.5-h session on MI was integrated into an existing course (in the 8th-semester seminar “Psychiatry” and in the 9th-semester weekly internship “Psychiatry”). In addition, the students were asked to watch teaching and demonstration videos on an online platform (approximately 30 min).

The control group studied in the 6th semester according to the original timetable, without the opportunity to participate in the MI sessions. At the end of the 9th semester, all participants in the control group who had completed the questionnaire at time T0 received an email with a link to a very similar, slightly adapted questionnaire (time T2). Figure 1 summarizes the study procedure in detail.

The Questionnaire on Motivational Interviewing (see Figure 1) consisted of a “knowledge test” section, an “assessment of subjective theoretical knowledge and practical skills” section and a section on “satisfaction with the curriculum.”

SPSS for Windows (version 22.0) was used for quantitative analysis. Data were analyzed visually and by Kolmogorov–Smirnov test under the assumption of normal distribution. However, if a non-normal distribution was suspected, non-parametric Mann–Whitney U test was also used. Homogeneity of variance was tested using Levene’s test. Descriptive statistics of participants between the two groups were tested for differences in means and distributions using t-tests and chi-squared tests. Analyses of variance were calculated using T0 and T2 as within-subject factors and group (control vs. intervention) as a between-subject factor to examine the impact of the MI curriculum. The Mann–Whitney U test was used to test for homogeneity of variance. Partial eta squared (η²) was used as the effect size for ANOVA and Pearson’s correlation coefficient r for t-tests. Correlations were calculated using Pearson’s correlation coefficient r, Spearman’s correlation coefficient r_s or Kendall’s correlation coefficient τ depending on the scale level. At the nominal scale level, possible correlation was tested using Fisher’s exact chi-squared test and the non-parametric Mann–Whitney U test. p-values below 0.05 were considered significant.

For qualitative analysis, feedback in the free-response format was categorized according to qualitative content analysis (Mayring and Fenzl, 2010). Missing answers in the multiple-choice knowledge test were defined as “wrong,” similar to the procedure in a written exam. In the case of missing questionnaire data, the item, but not the participant, was excluded from the analysis of variance. If the questionnaire was canceled, the participant was excluded from the data analysis. Students who attended fewer than two parts of the seminar or watched fewer than two information videos were also excluded.

In the total sample, interest in learning MI was, on average, high to very high (M = 2.92; SD = 1.00; 0 [none] to 4 [very high]). Prior to the start of the study, 83.7% of the subjects had participated in one or two general communication training sessions offered as part of their medical training. With regard to MI in particular, 98% of participants reported that they had received no training or input in this area prior to the start of their studies. In line with this, students rated their theoretical knowledge (M = 0.24; SD = 0.48; 0 [none] to 4 [very high]) and practical skills in MI (M = 0.37; SD = 0.64; 0 [none] to 4 [very high]) as rather low.

There were also some limitations to this study. One limitation relates to the measurement tools used. As the questionnaires were not validated, but developed in-house, psychometric data on validity has limited interpretability. In addition, no objective assessment of practical skills was made. The authors are aware of the limited sample size in the control group, and future studies should pay attention to the selection of appropriate measurement instruments and assessment forms in general and also for raters such as the MITI or the MI-SCOPE (Hannöver et al., 2012; Kitzmann et al., 2019).