Atherton 1982.

Methods	Design: Controlled before‐after study Duration of follow‐up: Up to 12 months
Participants	Country: United States Setting: Urban workplaces Intervention group: 594 employees from an unknown number of workplaces (total of 29 agencies between intervention and control groups, but not reported separately) Control group: 154 employees from an unknown number of workplaces Age: working age. Gender not reported
Interventions	Adoption of a compressed work week, with participants working normal weekly hours over fewer days (either four‐day weeks or nine‐day fortnights)
Outcomes	Weekly household vehicle miles travelled Rideshare and transit use Vehicle emissions outcomes apparently estimated based on vehicle miles travelled
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	High risk	Not randomised
Allocation concealment?	High risk	Not randomised
Blinding? All outcomes	High risk	Blinding not feasible. Could have influenced outcomes
Incomplete outcome data addressed? All outcomes	High risk	2309 employees surveyed at baseline (combined intervention and control groups) but primary outcome measure (weekly household vehicle miles travelled) reported for only 748 (32%). 2464 employees surveyed at 12 month follow‐up (combined intervention and control groups) but primary outcome measure reported for only 533 (22%). Results for agencies in the central business district were excluded due to 'an insufficient number of observations for this group'. Thus, a substantial amount of incomplete outcome data. Participant numbers were not always reported separately for intervention and control groups, making it difficult to fully assess incomplete outcome data.
Free of selective reporting?	Unclear risk	Not clear that all pre‐specified outcomes were included
Free of other bias?	Unclear risk	Numbers of workplaces in intervention and control groups were not reported. Participant numbers were not always reported separately for intervention and control groups, making it difficult to fully assess risk of bias.
Adequate matching of intervention / control groups?	Unclear risk	Travel mode at baseline similar for intervention and control groups. Demographic characteristics for each group were measured but data not presented. The authors reported that 'participation rates among employees that have different socioeconomic characteristics can vary considerably', suggesting socio‐economic differences between intervention and control groups, but did not present data to support this