Logo of bmcamBioMed Centralsearchsubmit a manuscriptregisterthis articleBMC Complementary and Alternative Medicine
PMC full text:
Copyright/License ►Request permission to reuse

Table 4

Predictors of Knowledge of, Experience with, Expectations about, and Willingness to Try Five Complementary and Alternative Medical (CAM) Therapies

Odds ratios* (95% CI) for the independent predictor variables used in the final models for each CAM therapy
Dependent VariableAcupunctureChiropracticMassageMeditationT'ai chi
High Knowledge of specific therapy (4–5)Tried acupuncture:
43.6 (16.7–113.6)		Tried chiropractic:
12.8 (6.2–26.7)		Tried massage:
7.6 (4.0 – 14.7)		Tried meditation:
11.6 (4.6–29.3)		LOGISTIC NOT VALID**
Bostonian:
4.8 (1.9–12.3)
Previously tried specific therapyNo associationsBostonian:
0.5 (0.3 – 0.8)		65+ yrs:
0.4 (0.2 – 0.7)		Female:
2.5 (1.3 – 4.8)		No associations
Previously tried specific therapy for back painNo associationsNone65+ yrs:
0.3 (0.2 – 0.6)		No associationsLOGISTIC NOT VALID**
High expectations of specific therapy (7 – 10)65+ yrs:
0.4(0.2 – 0.8)		Knowledge:
2.9 (1.6 – 5.2)		65+ yrs:
0.3 (0.2 – 0.5)		No associationsNo associations
Tried acupuncture:
4.3 (2.1 – 9.0)
Very likely to try specific therapy for freeHigh expectations:
15.4 (3.6 – 66.1)		High expectations:
27.4 (9.5 – 79.3)		High expectations:
16.4 (7.4 – 36.5)		High expectations:
3.6 (1.7 – 7.7)		High expectations:
14.3 (5.4 – 38.3)
Tried meditation:
2.4 (1.3 – 4.5)
Very likely to try specific therapy for $10 /visit co-payHigh expectations:
6.8 (3.0 – 15.5)		High expectations:
8.1 (4.2 – 15.7)		High expectations:
6.4 (3.5 – 11.4)		NOT QUERIEDNOT QUERIED
Bostonian:
2.3 (1.4 – 4.0)		Bostonian:
1.8 (1.001 – 3.2)
*These odds ratios describe the significant associations (p < 0.01) for each of the 28 therapy – specific dependent variables. For example, we found that those who had tried acupuncture were 43.6 times more likely to have high knowledge of acupuncture. No other variables were related to high knowledge of acupuncture. ** These logistic regression models did not converge. Categorization for independent variables: Age (<65; 65+) Knowledge of therapy (1–3; 4–5) Gender (M; F) Expectations of therapy (missing through 6; 7+) Geography (Seattle; Boston) Prior Use of therapy (no; yes)