**5. Conclusions**

We propose VO2net (VO2gross <sup>−</sup> VO2stand, mL·kg−1·min<sup>−</sup>1), instead of MET (VO2gross/VO2rest), to be a criterion measure for accelerometer calibration in children, adolescents and adults. This conclusion is based on the finding that these age-groups showed similar relationship between Speedeq and VO2net, and consequently VO2net may be considered a measure of absolute PA intensity equivalent by age. The same similarity between the age-groups was not found for the relationship between Speedeq and MET. New VO2net-based accelerometer cut-points were developed and when they were applied to free-living data, the difference in PA between children, adolescents and adults was increased compared to applying MET-based accelerometer cut-points. The implications of these findings are that PA can now be directly compared between age-groups in epidemiological research to investigate changes in PA behaviors from childhood into adulthood and associated factors related to risk for cardiometabolic disease. We contribute with new accelerometer cut-points for absolute PA intensity for the hip and thigh placement in children, adolescents and adults (Table 1), to be applicable to accelerometer output generated from raw triaxial acceleration data [17,18]. The algorithms for processing the raw acceleration data are available by request to the authors. Still, this is a small and explorative study and further research is needed to evaluate the robustness of the VO2net calibration method proposed.

**Author Contributions:** Conceptualization, D.A. and J.F.; methodology, D.A. and J.F.; software, J.F.; formal analysis, D.A. and J.F.; visualization, J.F.; investigation, D.A., J.F., C.B., Ö.E., E.E.-B., L.L., M.H. and M.B.; resources, Ö.E., L.L. and M.B.; data curation, D.A. and J.F.; writing—original draft preparation, D.A. and J.F.; writing—review and editing, C.B., Ö.E., E.E.-B., L.L., M.H. and M.B.; funding acquisition, Ö.E., L.L. and M.B.

**Funding:** The calibration study was supported by grants from ALF, Sahlgrenska University Hospital, Gothenburg, Sweden. In the free-living study, the data in children was collected as part of the I.Family study with financial support from the European Commission within the Seventh Framework Programme no. 266044. Data collection was also part of the EpiLife-Teens research program supported by the Swedish Research Council, the Swedish Research Council for Health, Working Life and Welfare, and the National Research Council for Sustainable Development (Formas). The adult data in the LIV-2013 study was financially supported by the Swedish School of Sport and Health Sciences funding, the National Board of Health and Welfare, ICA Sverige AB, Monark Exercise AB, and Apoteket Hjärtat.

**Conflicts of Interest:** The authors declare no conflict of interest.
