Table 3

Logistic regressions of random effect for MHT treatment in menopause outpatient visits
	Outpatient visits (% prescribing MHT)		Model 1 – no interaction	Model 2 – interaction of WHI and woman education	Model 3 – interaction of WHI and hospital level

	n	%^a	Odds Ratio (95% CI)	Odds Ratio (95% CI)	Odds Ratio (95% CI)

WHI report announcement
After	1157	72.95	0.36 (0.25–0.52)
Before (reference group)	1392	82.97

Medical care institution
Metropolitan hospital	627	82.93	14.87 (7.28–30.37)	11.98 (5.82–24.68)	7.02 (3.23–15.27)
Local community hospital	492	81.10	17.00 (7.92–36.49)	12.09 (5.81–25.14)	12.13 (4.84–30.41)
Physician clinic	955	83.87	26.42 (13.34–52.32)	16.52 (8.43–32.39)	18.22 (8.12–40.89)
Ref: Academic medical center	475	58.74

WHI on Medical care institution^b
WHI effect on metropolitan hospital					6.25 (2.08–18.72)
WHI effect on local community hospital					1.69 (0.52–5.49)
WHI effect on physician clinic					2.13 (0.76–5.98)
WHI effect on academic medical center					0.15 (0.34–0.63)

Physician specialty
Obstetrician or gynecologist	1295	85.10	5.34 (3.45–8.26)	6.26 (3.92–10.00)	5.56 (3.60–8.58)
Ref: other specialty	1254	71.53

Demographic factors
Age			0.73 (0.41–1.32)	0.72 (0.37–1.40)	0.66 (0.36–1.20)
55–65	719	75.52
65+	348	83.62	2.33 (0.81–6.69)	2.43 (0.86–6.92)	4.38 (1.41–13.54)
Ref: 45–55⁺	1482	78.61
Education
Senior high school	373	75.87	0.53 (0.21–1.35)	0.33 (0.09–1.21)	2.46 (1.02–5.93)
Junior high school	446	78.03	0.54 (0.19–1.55)	0.38 (0.11–1.31)	3.11 (1.19–8.10)
Elementary school or lower	1180	80.76	0.95 (0.41–2.19)	0.62 (0.22–1.80)	4.50 (1.83–11.04)
No	322	79.19	1.18 (0.43–3.20)	1.10 (0.31–3.84)	4.90 (1.75–13.72)
Ref: Undergraduate or graduate	228	70.18

WHI effect on woman education^c
WHI effect on women with senior high school				1.64 (0.43–6.22)
WHI effect on women with junior high school				1.10 (0.29–4.16)
WHI effect on women with elementary school or lower				1.28 (0.40–4.08)
WHI effect on women with no education				0.85 (0.20–3.53)
WHI effect on women with undergraduate or graduate				0.30 (0.11–0.83)

Marital status
Married	2141	77.11	1.03 (0.51–2.10)	0.64 (0.34–1.22)	1.18 (0.62–2.23)
Ref: Unmarried	408	85.29

Economic factors
Employment status
Employed	841	79.31	0.94 (0.45–1.94)	0.37 (0.17–0.76)*	0.63 (0.30–1.30)
Ref: Unemployed	1708	77.99
Personal income (NT $)
<4,999	293	78.16	0.73 (0.31–1.72)	0.66 (0.28–1.59)	0.76 (0.31–1.89)
5,000~19,999	562	79.89	0.72 (0.30–1.71)	0.36 (0.28–1.59)	0.74 (0.32–1.73)
20,000~39,999	431	79.35	1.20 (0.57–2.55)	0.39 (0.18–0.85)	1.14 (0.53–2.45)
>= 40,000	259	71.43	0.24 (0.09–0.59)	0.33 (0.13–0.85)	0.86 (0.31–2.39)
Ref: None	993	79.15

Health Status
Illness in the past six months
Yes	2188	78.11	1.08 (0.49–2.39)	2.18 (1.07–4.47)	0.81 (0.38–1.73)
Ref: No	361	80.33
Osteoporosis
Yes	741	80.43	1.63 (0.91–2.89)	1.48 (0.84–2.59)	1.37 (0.80–2.33)
Ref: No	1808	77.60

%^a: Percentages in the second column of table 3 were the proportion of outpatient visits with a prescription of MHT in each category. For example, among the 1,157 outpatient visits after the publication of WHI report, there're 72.95% (844 visits) with a prescription of MHT. WHI and Medical care institution^b: A linear function of the coefficient was estimated and tested in Model 3 with the interaction terms. In order to simplify interpretation, we showed the effect of publication of the WHI on the four types of hospital instead of the interaction terms. WHI effect and woman education^c: A linear function of the coefficient was estimated and tested in Model 2 with the interaction terms. In order to simplify interpretation, we showed the effect of publication of the WHI on the five categories of level of women education instead of the interaction terms.
Huang et al. BMC Public Health 2007 7:56 doi:10.1186/1471-2458-7-56