Table 3.

Logistic regression to identify factors that influence the probability that the ICER is greater than the weighted median for ratios in the north-east quadrant for cost-per-QALY studies

	Model 1: disease area			Model 2: full
Variable	OR	SE	P value	OR	SE	P value
Disease area=oncology	3.03***	1.12	0.003	3.22***	1.27	0.003
Sponsor=industry	–	–	–	0.66	0.38	0.466
Author=industry	–	–	–	0.43	0.39	0.352
Study quality (subjective score)	–	–	–	1.27*	0.17	0.069
Constant	0.70*	0.15	0.086	0.21**	0.15	0.026
Observations	208			208

Variables were coded as follows: disease area oncology=1, disease area not oncology=0; study sponsor included industry=1, sponsor did not include industry=0; author list included industry=1, author list did not include industry=0; study quality=1–7 interval scale (half-points were possible).

***p<0.01; **p<0.05; *p<0.10.

ICER, incremental cost-effectiveness ratio; OR, odds ratio; QALY, quality-adjusted life-year; SE, standard error.