Table 2 

Respondent Health Need and Association with Alternative Medicine Use (Versus Use of Neither Alternative Medicine nor Conventional Care 

Model 1^{1} 
Model 2^{2} 
Model 3^{3} 
Model 4^{4} 



UOR^{5} 
CI^{6} 
AOR^{7} 
CI^{6} 
AOR^{7} 
CI^{6} 
AOR^{7} 
CI^{6} 
AOR^{7} 
CI^{6} 



One or More Health Needs^{9} 



Yes 
1.84 
1.592.12 
1.84 
1.592.12 
1.98 
1.702.32 
1.83 
1.562.15 
1.75 
1.492.06 


No (ref) 
1.00 
1.00 
1.00 
1.00 
1.00 



Wald chisquare (d.f.; pvalue) 
68.95 (1; p < .001) 
91.51 (4; p < .001) 
234.36 (18; p < .001) 
305.55 (23; p < .001) 
432.10 (30; p < .001) 



Improvement chisquare^{8 }(d.f.; pvalue) 
 
22.56 (3; p < .001) 
142.85 (14; p < .001) 
71.19 (5; p < .001) 
126.55 (7; p < .001) 



^{1 }The dichotomous, composite health need measure was entered into a model controlling for the external environment (region of residence) measures significantly associated (p < .05) with the dependent variable in Table 1. ^{2 }The dichotomous, composite health need measure was entered into a model controlling for the external environment and predisposing measures (sex, age, race and ethnicity, education, class of worker, and born in the U.S.) significantly associated (p < .05) with the dependent variable in Table 1. ^{3 }The dichotomous, composite health need measure was entered into a model controlling for the external environment, predisposing measures, enabling measures (poverty status and health insurance coverage), and barriers to conventional care (delayed care due to cost and/or noncost barriers) significantly associated (p < .05) with the dependent variable in Table 1. ^{4 }The dichotomous, composite health need measure was entered into a model controlling for the external environment, predisposing measures, enabling measures, barriers to conventional care, and personal health practice measures (leisuretime physical activity, alcohol drinking status, smoking status) significantly associated (p < .05) with the dependent variable in Table 1. ^{5 }UOR=unadjusted or crude odds ratio. ^{6 }CI= 95% confidence interval. ^{7 }AOR=adjusted odds ratio. ^{8 }Analogous to the Fchange statistic in ordinary least squares regression, the improvement chisquare is a test statistic used to determine if the variables entered in each step improve the fit of the model. ^{9 }This measure is defined as any one of: poor or fair health; health is worse off than it was 12 months ago; one or more serious chronic or acute conditions (see definition in Table 1); a functional limitation (see definition in Table 1); and/or back problems. 

Nahin et al. BMC Health Services Research 2010 10:220 doi:10.1186/1472696310220 