Association of teriparatide adherence and persistence with clinical and economic outcomes in Medicare Part D recipients: a retrospective cohort study

Table 7 Logistic regression of adherence and fracture risk* ^†

Variable	Odds ratio	95% confidence interval	P value
PDC (≥ 80% vs. <50%)	0.81	(0.515, 1.231)	0.48
PDC (≥ 50% ≤80% vs. < 50%)	0.91	(0.479, 1.741)	0.94
Age	1.00	(0.976, 1.016)	0.66
Gender (Female vs. Male)	1.37	(0.796, 2.368)	0.25
Race (Black vs. White)	1.41	(0.558, 3.538)	0.48
Race (Hispanic vs. White)	0.42	(0.048, 3.599)	0.29
Race (Other vs. White)	1.80	(0.689, 4.703)	0.22
Fracture in 6 Month Period before Initiating Teriparatide	2.93	(2.084, 4.122)	<.0001
Charlson Comorbidity	0.96	(0.861, 1.07)	0.46
Geographic Region (West vs. Northeast)	1.38	(0.347, 5.513)	0.99
Geographic Region (Southeast vs. Northeast)	2.07	(0.566, 7.531)	0.05
Geographic Region (Midwest vs. Northeast)	1.29	(0.341, 4.889)	0.76
Bisphosphonate Use	1.04	(0.738, 1.469)	0.82
Calcitonin Use	0.62	(0.283, 1.378)	0.24
Selective Estrogen Receptor Modulator Use	1.16	(0.551, 2.433)	0.70
Estrogen Use	0.61	(0.197, 1.891)	0.39

* Results of multivariate logistic regression analyses, with ‘Fracture in 12 Month Post-Period’ as the outcome variable.
^† Among patients with a fracture.