Table 2.

Associations between number of switches and patient outcomes post-period

Outcome	1 switch n = 4653		2 switches n = 2693		≥ 3 switches n = 2579
	Odds ratio	95% CI	Odds ratio	95% CI	Odds ratio	95% CI
TSH out of range	1.08	1.00–1.17	1.29	1.17–1.42	1.23	1.08–1.40
Composite endpoint	1.21	1.08–1.35	1.18	1.03–1.36	1.52	1.24–1.86
Chronic kidney disease	1.40	1.06–1.84	1.34	0.94–1.90	1.74	1.15–2.66
Depression	1.14	1.02–1.27	1.08	0.94–1.24	1.29	1.08–1.53
Fatigue	1.10	1.02–1.20	1.20	1.09–1.33	1.30	1.15–1.48
Heart failure	0.87	0.67–1.13	0.73	0.51–1.05	1.55	1.06–2.28
Hyperlipidemia	1.03	0.95–1.13	1.00	0.89–1.12	1.19	1.2–1.38
Hypertension	1.13	1.03–1.24	1.05	0.93–1.18	1.22	1.05–1.43
Obesity	1.37	1.21–1.55	1.31	1.12–1.53	1.39	1.14–1.69

Odds ratios represent likelihood of being diagnosed with outcome of interest compared to continuous users (N = 9925) of Synthroid^®. Results from multivariable logistic regression models of clinical outcomes associations with number of switches including controls for patients’ demographics (age, sex, region of residence, and insurance type), index dose and copayment of levothyroxine prescription, pre-period Charlson Comorbidity Index score, pre-period visit to endocrinologist, and lagged dependent variable

CI confidence interval, TSH thyroid-stimulating hormone