Table 4.

Simulation parameters and estimands for different scenarios in simulation study 1. For all scenarios, we set α = 0, β_trt = 3, β_ep = 1, β_M = 1, ${\mathrm{\sigma }}_{\mathrm{\mu }}^2=5$ and ${\mathrm{\sigma }}_{\mathrm{\epsilon }}^2=5$

	Parameters				Estimand values
Scenario	βTRTxEP	βTRTxM	γ	δ	${\mathrm{\beta }}_E^{\mathit{AB}}$	${\mathrm{\beta }}_P^{\mathit{AB}}$	${\mathrm{\beta }}_E^{\mathit{PB}}$	${\mathrm{\beta }}_P^{\mathit{PB}}$
Scenario 1: Constant treatment effect	0	0	0	0	3	3	3	3
Scenario 2: Treatment effect varies across episode	1.5	0	0	0	3.5	3.38	3.5	3.38
Scenario 3: Treatment effect varies across patients with different values of Mi	0	3	0	0	5	4.5	5	4.5
Scenario 4: Treatment effect carries forward	0	0	1	0	3	3	3.33	3.25
Scenario 5: Treatment becomes less effective on re-use	0	0	0	-3	2.5	2.63	2	2.25
Scenario 6: Treatment effect varies across episodes, across patients with different values of Mi, carries forward, and becomes less effective on re-use	1,5	3	1	-3	5	4.5	4.83	4.38