The effect of adherence to statin therapy on cardiovascular mortality: quantification of unmeasured bias using falsification end-points

Table 3 Effect estimates of adherence to statin therapy on cardiovascular mortality by calendar period

Calendar years	Small model		Large model		Difference in HR	# events
Calendar years	HR	95 % CI	HR	95 % CI	Difference in HR	# events
1994–2002	0.85	0.55 to 1.34	0.79	0.44 to 1.40	0.06	158
2003–2006	0.58	0.45 to 0.76	0.46	0.35 to 0.59	0.12	642
2007–2012	0.77	0.65 to 0.92	0.54	0.45 to 0.65	0.23	765
Overall (1994–2002)	0.70	0.61 to 0.81	0.53	0.46 to 0.61	0.17	1565
Falsification outcome (1994–2002)	0.93	0.97 to 1.09	0.68	0.58 to 0.80	0.25	1179

ISSN: 1471-2458