Table 1. Example of cumulative exposure time calculation.

Year	Time (T)	Residential characteristics: walkability category 1WC(I) and geographic classification 2GC(Z)	3CET to RA	CET to MW in MPC	CET to HW in MPC	CET to LW in LPC	4PCET to RA	PCET to MW in MPC	PCET to HW in MPC	PCET to LW in LPC
1994	0	MW in MPC	0	0	0	0	0	0	0	0
1996	2	MW in MPC	0	2	0	0	0	2/2	0	0
1998	4	RA**	0	4	0	0	0	4/4	0	0
2000	6	RA**	2	4	0	0	2/6	4/6	0	0
2002	8	HW in MPC	4	4	0	0	4/8	4/8	0	0
2004	10	HW in MPC	4	4	2	0	4/10	4/10	2/10	0
2006	12	LW in LPC	4	4	4	0	4/12	4/12	4/12	0
2008	14	LW in LPC	4	4	4	2	4/14	4/14	4/14	2/14
2010	16	LW in LPC	4	4	4	4	4/16	4/16	4/16	4/16

¹ WC(I): Walkability category- I ranging from 1 to 3 represented (low (LW), medium (MW) and high walkability (HW))

² GC(Z): Geographic classification—Z ranging from 1 to 4 represented (rural areas (RA), small (SPC), medium (MPC) and large population centers (LPC)).

³CET: Cumulative exposure time spent in neighbourhoods with different characteristics.

⁴ PCET: Proportion of cumulative exposure time spent in neighbourhoods with different characteristics.

The proportion of cumulative exposure time (PCET) of respondent X to category I Walkability (WC) and category Z geographic classification (GC) at survey year T = (Total years having resided in WC[I]

* GC[Z])/T, with I ranging from 1 to 3 representing the 3 Walk score^® categories (low, medium and high walkability), Z ranging from 1 to 4 representing (rural areas, small, medium, and large population centers) and T ranging from 2 to 16 (in multiples of 2, i.e. 8 follow-ups). For each respondent, the sum of all PCETs values across all walk score categories in different geographic classifications is equal to the number of follow-up years from baseline, and for each follow-up, the sum of PCETs is equal to one.

** Note: rural areas were always in the low walkability category.