Table 3.

Summary of the analyses for the dependent variable Continuity of care (PREM), multilevel linear regression

Potential case-mix variable	Fixed effect significant (y/n)	Slope effect GP level important* (y/n)	Slope effect country level important (y/n)	Case-mix control (y/n)
Self-reported general health	Yes Worse self-reported health → better experience continuity	No	Yes	No Importantly more variation at country level for people with very good self-reported health
Longstanding disease	Yes Longstanding disease → better experienced continuity	No	No	Yes
Patient’s age	Yes Older than 40 → better experienced comtimuity	No	No	Yes
Patient’s sex	Yes Women → better experienced continuity	No	No	Yes
Education	Yes Higher education → worse experienced continuity	No	No	Yes
Income	No	No	No	No
Migrant status	Yes Migrant status → worse experienced continuity	No	No	Yes
Place of living	Yes Outside big cities and suburbs → better experienced continuity	No	No	Yes

^*Important means that the difference in variance between categories is more that 0.25*variance in the model with fixed effect and random intercept