The study aimed to investigate how children and adolescents in Germany differ in reaching the WHO Guideline of at least 60 min MVPA per day depending on whether PA was self-reported or measured by accelerometer. As expected, the PAQ values were higher than those measured by the ACC but still, both values are alarmingly low.
The low overall adherence to the PA Guideline as obvious from the PAQ can also be found in the results of a recently published pooled data study by the WHO, according to which less than 15% of school-going adolescents aged 11–17 did meet the Guidelines . The lower PA Guideline adherence as measured by the ACC is consistent with the findings of [7, 29, 30] who also reported less light to moderate PA determined by device-based measurements compared to questionnaires. The difference between self-reported and device-based measured PA might be because the PAQ only asks for physically active time, which is subjective and depends on the physical fitness of the participant, as was stated by other studies before [4,5,6]. Additional, well-known factors that influence the validity of self-reports are recall bias and social desirability .
The differences between the results of both methods are significantly smaller for younger children than in the older age groups. Differences between age and gender groups are found in both methods. The older the subjects are, the lower is the proportion of those who meet the WHO Guideline on each day, with girls meeting the guideline less frequently than boys in all age groups. A closer look at the differences between genders in the youngest age group reveals that boys almost match their answers given in the questionnaire with a difference of 0.2 days. The previously mentioned overestimation of PA by the questionnaires [4,5,6] cannot be found here. The significant difference between the youngest age group and the older ones may be caused by the fact that in this group an external observer (usually a parent) fills out the questionnaire together with the child and may therefore be better able to assess the activity . Found that parent-reported MVPA corresponds to one to two-thirds of the child’s activity measured by the accelerometer. This could be a clue why the gap between the methods is smaller. A more plausible explanation is that the activity patterns of children are more spontaneous, impulsive, and of shorter duration [1, 2]. These short activities, when measured in total, often result in a small amount of light or moderate activity and are poorly captured by questionnaires . Accelerometers, by contrast, register these short and spontaneous movements which are sometimes overlooked when filling out the questionnaire. This does not lead to an overestimation in the questionnaire for small children, but an underestimation due to the short, spontaneous movements not recorded. Since accelerometers measure these movements, this would explain the reduced difference between the two methods.
This is also reflected by the increasing difference between the methods used in the older age groups. The older the participants are, the more structured is their everyday life and the less spontaneous movements are registered by the accelerometer. The familiar overestimation by the questionnaire reoccurs. The older the participants were, the less often they reached the WHO recommendation of 60 min MVPA per day. The lower adherence in older age groups is confirmed by findings of [34, 35] in other European countries and might be due to longer times spent at school or work.
In comparison to boys, girls reach the 60 min MVPA on fewer days in all age groups, but the difference decreases to about half of the initial value with increasing age. The difference in gender and the lower adherence for girls is consistent with the worldwide gender gap of physical activity reported in [28, 35, 36]. Mielke and colleagues  found a similar prevalence of inactivity in women and men in a study based on worldwide data of the WHO. Normally, girls and boys might be expected to be equally active until puberty, with the gap starting to open up at this point in time. However, different interests may probably be the reason for this earlier gap - girls tend to be sociable and do esthetic sports, while boys tend to romp, scuffle, and do run-intensive sports. Further examination of the data on PA intensity and sport disciplines in MoMo could give a more detailed answer as to where this difference comes from and where interventions could be useful to close this gap.
Strength and limitations
The present study is limited to its observational nature and we do not intend to infer causality from paralleled trends or significant correlations. The main goal of MoMo is to track and report PA and fitness of children and adolescents in a nationwide sample, and significant effort was put into collecting representative data from 167 sample points all over the country.
A major strength is the large number of participants and recording of physical activity of each participant by PAQ and ACC. However, this also leads to the restriction that the PAQ assessed PA of an average week, whereas the ACC measured PA during one specific week. An additional comprehensive and elaborate diary was avoided during the week by wearing an ACC. Study participants carried the ACC following the completion of the already very time-consuming fitness test and surveys on activity and health.
Even though the wearing times were very long on average, some participants told us that wearing had been prohibited in some sports competitions like soccer. Wearing electronic devices was forbidden to prevent trainers from having an unfair advantage in knowledge. Even if documented by the non-wear time protocol , the unrecorded activities could not be taken into account retrospectively. The manual input of the data from the handwritten non-wear protocol is very time-consuming. Besides, the information in the protocols is very inconsistent and manual input would distort the acceleration data. This missing data could be another link to the difference between PAQ and ACC results in this study. A wearing time of 24 h and a consistent ambulatory assessment for the non-wear time could solve this problem in future studies.
A check of the WHO guidelines is easy to implement with an accelerometer at first glance. However, evaluation results in a multitude of possibilities for implementation. When examining the average time spent with physical activity each week (as now recommended by the new WHO Guidelines of 2020 ), days with activity times longer than 60 min would compensate for those with less activity . Still, the daily stimulus is very important in children . This study determined whether the subject was active for at least 60 min or not on each day individually. To look at the exact times spent with MVPA on every single day will result in fewer days of at least 60 min MVPA when both evaluation methods are compared . The main reason, however, since the study was already planned and started in 2014, the questions in the questionnaire still referred to the 2010 WHO Guidelines . Only now the recommendations on youth activity have changed from a recommendation of at least 60 min per day to a recommendation of an average of 60 min per day . This adaptation will require changes in survey questions and sampling methods for future monitoring. However, changing the question wording is unlikely to address the need for PA monitoring among children who, especially at young ages, are unable to answer a complex question about average behavior over the past few days, weeks, or months. In the future, this may require the use of proxy reports from multiple respondents, including parents and teachers, though both may also miss observing large portions of the day . The alternative of asking daily duration for an entire week may be more accurate but increases survey response time. Therefore, measuring daily PA remains a strength of the portable devices for now, and adapting the questionnaires to the new WHO guideline remains a real challenge.
However, we have looked at the accelerometer data with the background of the new WHO guideline. It should be noted, however, that these results cannot be compared with the results of the question used in this study about the number of days on which the subjects have MVPA for more than 60 min. However, if one compares the number of subjects meeting the old versus the new guideline based on the accelerometer data, we see that the percentage increases from 3% to a full 34% of the study participants. This means that 31% of the participants who do not reach 60 min MVPA on all days still have days in the week on which they do so much physical activity that these outweigh the remaining days under the new guideline compared to the old one. This drastically reduces the proportion of children and adolescents who are too inactive, which is also likely to cause some political controversy in the future.
By using an epoch length of one second in MoMo, short activities can be recorded with the accelerometers. This could be another reason why younger children have a more consistent PA outcome with both methods. These short activities are less frequent for older children, which is associated with the fact that an increasing number of older children only practice organized sports. According to MoMo data from previous waves, organized PA in extracurricular activities and sports clubs increased by 8 %, while unorganized PA decreased by 7 % .
Apart from PA, many other parameters were collected in MoMo. This results in a multitude of evaluation options, and further examination of the data in MoMo (such as PA intensity, sports disciplines, socioeconomic status, migration status, etc.) will give a more detailed answer as to the reasons behind the differences regarding age and gender.