Before and during a voluntary 6-week fasting intervention, parents and their children were encouraged to address their screen media use habits and time spent using them (see Figures 1 and 2). For this purpose, various alternatives and offers for joint family activities in the areas of nature, play, sport, culture and nutrition were suggested. In a previous article 16, we presented the MediaFasten pilot study (www.MedienFasten.org), an open, prospective, multicenter, pseudonymized survey conducted in spring 2019. In a collaboration with the German professional association of pediatricians and adolescent doctors (Berufsverband der Kinderärzte und Jugendaerzte) an announcement was made and practices were able to contact the team if they were interested in taking part. A convenience sampling methods was used. Six practices showed their interest and contact was made where the study purpose and design were introduced to the practice team. In six pediatric and adolescent medicine practices in the state of North Rhine-Westphalia, Germany, 250 families with children from all social classes were to be recruited to actively participate in the 6-week fasting period parallel to Christian Lent. This began on March 6, 2019, and ended on April 16, 2019. The intervention was primarily designed to raise awareness among families about changing usage behavior.

The study was carried out as a cooperation project with the professional association of pediatricians and adolescent doctors (BVKJ). The families received a media fasting set from the pediatrician's office to take home. The set contained a cover letter with pediatric recommendations for parents on careful use of screen media (including the quality of the content), a media fasting calendar with 44 surprise doors to hang on the wall, 5 before and after questionnaires on the family's use of digital media (2 for parents, 3 for children), a post-questionnaire to assess the media fasting intervention, as well as postage-free addressed return envelopes for sending the completed questionnaires.

The media fasting calendar provided daily suggestions for games, fun, nutrition, exercise and relationships for 44 days.

For participation there were no exclusion criteria other than a certain level of reading comprehension skills when answering the questionnaires; therefore, the minimum age for children to participate was set at 6 years.

For data collection, consent forms had to be signed by an adult or legal guardian of the children. A pre- and post-survey took place. In this publication, the focus is on the intentionality before the six-week Lent and the pre-intervention motivation of the role model motif. The study was reviewed and approved by the Ethics Committee of the University of Witten/Herdecke (application no. 194/2018). For a detailed description of the methodology used, see Schwartz et al. (2019).

In addition to socio-demographic information, the questionnaire for parents contained eight items on the intentions of media fasting, which had an acceptable internal consistency with Cronbach's alpha = .71 and formed a common factor (“I want to help my child(ren) through media fasting be a role model.”; “I think it would be good for me to use less digital media.”; “I think it would be good for my child(ren) to use less digital media.”; “I feel annoyed/stressed by the constant availability of digital media.”; “I hope that the media fast will give me more introspection and silence.”; “I hope that my child(ren) will have more time for themselves again.”; “I hope that as a family we become more attentive and calm.”; “There is something cathartic about the idea of media fasting.”). The items were rated on a 4-point Likert scale (0 – no, not at all; 1 – somewhat no; 2 – somewhat yes; 3 – yes, definitely).

The questionnaire for children and adolescents contained eight similar items, four of which concern intentions (“I want to be a role model for my parents through media fasting.”; “I want to be a role model for my siblings through media fasting.”; “I hope that we can do more together as a family when everyone fasts on media."; "I hope that through the media fast I will have more time for friends."); while the content of the other four items was about screen use (“I think my parents use too much digital media.”; “My parents say that I am too busy with digital media.”; “I feel stressed by the amount of digital media.”; “I find that I spend too much time on digital media.”) are informative items that were not taken into account in the scale structure. The items were rated on a 4-point Likert scale (0 – no, not at all; 1 – somewhat no; 2 – somewhat yes; 3 – yes, definitely).

The analysis was open-ended and descriptive; group comparisons (gender, age groups) were carried out for usage behavior, as well as correlation analyses (Spearman's rho) and stepwise regression analyses on the role model function. The significance level of this exploratory study was set at p < 0.05. For the correlation analyses, an r > 0.5 was considered a strong relationship, an r between 0.3 and 0.5 as moderate, r between 0.2 and 0.3 as weak and r < 0.2 as marginal and insignificant. All statistical analyses were carried out using SPSS (version 27.0).

In the parent group (n = 365), mothers (66%) were more frequently represented than fathers (34%) (Table 1a). The parents had an average of 2 children. The group of parents aged <40 was the strongest at 50%, followed by those aged 40-50 (42%). The parent group was slightly dominated by people with higher education (51% with a university degree and 21% with a vocational school diploma). Media-free days (based on general/average media consumption during everyday life) per week were hardly present (0.3 ± 0.9 days): 88% had no media-free days, 9% 1-2 days, 3% used more than 2 no-media days per week. There were no significant differences between men and women regarding media use (not shown).

In the group of children and young people (n = 407), 20% were 0-5 years old (but were not taken into account for the statements on media use and intentions), 34% were 6-9 years old, 29% were 10-13 years old and 17% 14-18 years old. The group of toddlers/kindergarten children was not taken into account for the following analyses so 280 school children (56% boys and 44% girls) with an average age of 10.5 ± 3.1 years were examined. Media-free days per week were hardly present (0.8 ± 1.5 days) (Table 1b): 72% had no media-free days, 16% had 1-2 days, and 12% did not use media more than 2 days per week. There were no significant differences between boys and girls (not shown). In the group of primary school children, the number of media-free days was relatively high (1.2 ± 1.8 days), medium in the 10 to 13-year-olds (0.6 ± 1.3 days), and in the 14 to 18-year-olds low (0.1 ± 0.4 days); these differences were statistically significant (F = 13.4; p < 0.0001).

The parents' intention to be a role model for their children in media fasting (89% agreement; r = 1) correlated moderately, in particular, with the idea that media fasting has “something cleansing about it” (76% agreement; r = .41) and with the hope of “becoming more attentive and calmer as a family” (76% agreement; r = .41) as well as with the feeling that it would be good for their children to use less digital screen media ( 77% agreement; r = .36), and with the wish that they would “have more time for themselves again” (61% agreement; r = .36). Weakly associated was the desire for more “inwardness and silence” (77% agreement; r = .28) and marginally the feeling of being “annoyed/stressed by the constant availability of digital media” (59% agreement; r = .17) as it does not make sense (Table 2a).

Among the children, there was a strong connection between the intention to be a role model for their parents (50% agreement) and to be a role model for their siblings (55% agreement; r = .73) (Table 2b). The intention to be a role model for parents was only marginally associated with the assessment that they use too much digital screen media (48% agreement; r = .16), but significantly more strongly (moderately) associated with the desire to do more together as a family (70% agreement; r = .40) and to have more time for friends again (47% agreement; r = .31). However, their own assessment of being too busy with digital screen media (36% agreement; r = .10) or being stressed by them (15% agreement; r = .02) did not show a significant connection.

Since parents' and children's role model intentions were significantly associated with different variables, stepwise regression analyses were carried out to estimate the predictive significance of these influences. An evaluation of the predictors of the parents' role model intention as a dependent variable showed (Table 3a) that this intention can be predicted most strongly by the hope of becoming “more attentive and calm as a family” (23% explanation of variance), as well as by the notion that there is “something cathartic about media fasting” (additional 7% explanation of variance) and by noting that it would be good for their children to use less digital screen media (another 3% explanation of variance). The age of the parents only had a small predictive value (1% additional explanation of variance). These four variables could explain 35% of the variance. The hope that media fasting would enable them to gain more inwardness and silence and that their children would have more time for themselves was not significant in the regression model.

A similar evaluation of the predictors for the role model intention of children (older than 5 years) and adolescents showed (Table 3b) that their intention to be a role model to their parents was significantly predicted by their intention to be a role model to their siblings “To be a role model” (53% variance explanation). The age of the child explains another 3% of the variance, and the intention to “do more together as a family” only explains an additional 1%. Hoping to “have more time for friends” had no significant influence on this model. The three variables mentioned can explain 57% of the variance.

Although the families were being recruited from medical practices to which different social classes have access, the intervention primarily reached families of slightly higher education levels than the average population. This is therefore a self-selection bias that is difficult to control since the focus here was on the voluntary nature of participation.

A response behavior in the sense of social desirability cannot be ruled out, since all information is self-reported by the respondents.

Furthermore, most of the younger children's questionnaires were completed by their parents and are therefore of limited validity with regard to children's digital screen media use. Here too, the younger children in particular may have answered in the way their parents might have wanted to hear. Already in the first study, it became clear that parents rated their children's digital screen media use higher than their children's statements about its use.

Accelerated by the coronavirus pandemic, digital screen media are taking on an unprecedented importance in everyday life. Adults and children alike are finding more ways than ever to pass the time with digital information, communication, films, and games. The need to participate in classes online has also led to an increase in media ownership or at least availability in childhood, which is a mixed blessing due to the risk of excessive and/or harmful use. In this context, it will be interesting to investigate to what extent this development has an influence on the willingness and intentions of media fasting interventions and whether significant changes in usage behavior can be identified permanently or whether the boost in media usage due to the pandemic was just a short-term effect slowly fading away. Despite all the advantages we are currently attributing to digital screen media, we must not forget that there are also dark sides. Here it is important to strengthen the role models of parents and siblings in terms of setting an example for healthy use of media because, as the results of this intervention showed, a many participants were interested in being a good role model for other family members, and strive to ensure a healthier use of media.

Intermittent media fasting periods can make important short-term and long-term contributions to the health of society as a whole, given the rapid increase in digital usage times for both children and adults. To date, the supportive intentions within families to reduce the screen time of individual family members from within the community have not been addressed sufficiently. Based on the results of the present study, consideration could be given to making targeted use of appeals to the role model function of family members as a relevant resource in this and other health-related issues.