Table 2. Associations between patient satisfaction with healthcare and adherence to antidepressant (AD) medications.

AD users	Adherence to ADs			Discontinuation of ADs
	Prevalence of adherence by patient satisfaction level, %		OR (95% CI)*	Prevalence of discontinuation by patient satisfaction level, %		OR (95% CI)*
	Satisfied	Unsatisfied		Satisfied	Unsatisfied
All users, n = 4990	44.8	40.3	1.26 (1.08, 1.47)	36.0	41.5	0.83 (0.71, 0.96)
New users, n = 1940	26.2	21.1	1.41 (1.06, 1.88)	n/a	n/a	n/a

Abbreviations: CI, confidence interval

Prevalence, odds ratios and 95% CIs are weighted by MEPS’s sampling weights.

Referent group is “Unsatisfied”

^* ORs are adjusted for: 1) Individual-level factors: Age; Gender; Race/ethnicity; Geographic region of residence; Education; Speake English at home; Marital status; Employment status; Year of MEPS survey; Poverty status; Ever delay, forego or make change in prescription medicine because of cost; Ever delay, forego or make change in treatment because of cost; Type of antidepressant used; severity of depression symptoms (PHQ2≥3); Depression; Bipolar disorder; Anxiety; Psychotherapy; Poor physical health; Poor mental health; Have cognitive limitations; Hypertension; Coronary heart disease; Angina; Myocardial infarction; Other heart diseases; Stroke; Diabetes; Arthritis; Asthma; Chronic bronchitis; Cancer. 2) Provider-level factors: Provider specialty; Provider-patient gender concordance; Provider-patient race/ethnicity concordance. 3) Healthcare system factors: Uninsured; Have usual source of payment; Payment source; Out-of-pocket payments; Counts of office visits.