Researchers from the University Medical Center Groningen used the raw data from the SQUASH instrument in 1A Questionnaire 2 to calculate the sum scores for physical activity per activity type and overall for all adult participants who sufficiently filled in the questionnaire (sections: lifestyle & environment and secondary & linked variables).

The SQUASH sum scores can be requested in the Lifelines catalogue. Only the final results are present in the Lifelines catalogue, if you would like to receive the extensive version (including all intermediate variables) we assume you understand the data. The extensive version can be requested by mailing data@lifelines.nl. When the SQUASH sum scores have been used in your research, you can refer to the Chapter 7 of the following thesis:

• Byambasukh, O. (2020). Physical activity and cardiometabolic health: Focus on domain-specific associations of physical activity over the life course. University of Groningen. https://doi.org/10.33612/diss.112903501.
  

The calculation of the SQUASH sum scores according to Wendel-Vos et al. (2004) consisted of three consecutive steps: (1) calculation, (2) conversion, and (3) categorization. Below the steps are briefly described; for a more detailed description we refer to Byambasukh, O. (2020).

• Step 1 - Calculation: In this step the minutes per week of each activity have been calculated using the questionnaire data.
• Step 2 - Conversion: In this step an intensity factor based on the levels of effort has been assigned for each activity.
• Step 3 - Categorization: In this step the intensity categories are determined using the intensity factors from step 2.

After the first SQUASH sum scores were finalized, the following observations made the UMCG researchers who developed the SQUASH sum scores reconsider their results:

1. Total physical activity in minutes/week should not be used as a measure of physical activity because this measure includes light physical activity, and light physical activity entails a high risk of obtaining imprecise estimates of total minutes of light PA and because of potential bias relating to occupational status.
2. Moderate to vigorous physical activity (MVPA) measured in minutes/week is the best variable to rank individuals by level of physical activity.
3. Updated MET values (derived from the 2011 version of compendium) should be applied to calculate the MVPA, but without age correction during data processing. In other words, in the absence of a sound and validated rationale for the inclusion of age correction in data processing, the same categories of intensity should be applied for adults belonging to both younger (18–55 years) and older (>55 years) age groups.
4. An alternative outcome for physical activity that is often used is ‘MET-minutes’. However, if updated MET values derived from the 2011 version of the Ainsworth’s compendium are used, MET values for cycling conducted as a commuting or leisure-time activity should be set at 5.0 to avoid overestimating vigorous PA. Moreover, it is best to use ‘MET-minutes in MVPA’, thus MET-minutes for the moderate to vigorous levels (selecting MET ≥ 4) because of the high risk of obtaining imprecise estimates for light PA minutes.
5. Compared with occupational PA, leisure time and commuting PA demonstrate a much clearer association with health outcomes. Therefore, we recommend to make a substantiated decision to include or exclude occupational PA in the PA measure you want to use, since PA in the various domains seem to represent different kinds of activity with differential health effects.
   

In the end, it was decided to make the following adjustments to the syntax of Wendel-Vos et al. (2004):

• Change the MET value for cycling to 5.0
• Exclusion of age correction

Additionally, it was decided to release the composed variables only. And have