. 2005 Sep 8;5:31. doi: 10.1186/1472-6947-5-31

Table 3.

Psychosocial factors and other variables predicting intention to prescribe HT

	Quebec and France	Quebec		France

	All doctors N = 1472	General Practitioners N = 333	Gynaecologists N = 129	General Practitioners N = 419	Gynaecologists N = 577
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)

TPB Constructs
RB	2.08**** (1.61 – 2.69)	-	-	4.60**** (2.17 – 9.73)	1.47* (1.03 – 2.11)
MN	2.38**** (1.70 – 3.34)	3.42** (1.52 – 7.66)	84.58** (3.64 – 4.00)	-	1.75* (1.11 – 2.77)
PBC	-	-	-	1.65* (1.07 – 2.55)	-
Power of control beliefs (POC's)	4.92**** (3.51 – 6.90)	21.92**** (6.86 – 70.05)	15.86** (1.97 – 128.23)	5.80**** (3.05 – 11.03)	6.20**** (3.63 – 10.59)
Other Variables
Cytology (PAP smear)	-	11.05* (1.01 – 123.08)	-	-	-
Hormonal dosage as menopausal criteria	-	4.36* (1.15 – 16.50)	-	-	-
Cessation of menses as menopausal criteria	-	-	-	4.44*** (1.45 – 13.59)	-
Lipid profile	-	-	-	-	2.64* (1.23 – 2.68)

	Likelihood Ration X²= 1580; degrees of freedom = 3; p < 0.0001	Likelihood Ration X²= 384; degrees of freedom = 4; p < 0.0001	Likelihood Ration X²= 153; degrees of freedom = 2; p < 0.0001	Likelihood Ration X²= 477; degrees of freedom = 4; p < 0.0001	Likelihood Ration X²= 584; degrees of freedom = 4; p < 0.0001

Note: Due to incomplete information in some sections of the questionnaire, 996 questionnaires from France (419 general practitioners and 577 gynaecologists) and 462 questionnaires from Quebec (333 general practitioners and 129 gynaecologists) were used in the regression analysis. The differences in the total numbers of subjects in the regressions are due to missing information for some of the explanatory variables in the subgroup models. The outcome variable in the regressions is a dichotomous variable representing high vs. low intention to prescribe.

OR: Odds ratio; 95% CI: 95% Confidence Interval

* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001