Table 2. Multivariable conditional logistic regression analysis for hip fracture^†.

Parameter	OR‡	95% CI	P
Smoker (Yes vs. no)	1.07	1.02-1.22	0.04
History of falls during the last year	1.67	1.13-2.58	0.02
Cardiovascular disease (Yes vs. no)	1.15	1.03-1.35	0.03
Serum calcium	1.43	1.21-1.65	0.01
Serum ALP	1.03	1.01-1.06	0.009
Serum Hs-CRP	1.11	1.04-1.22	0.006
Serum TMAO	1.16	1.07-1.25	<0.001

^†Multivariable model included all of the following variables: smoker, history of falls during the last year, cardiovascular disease, serum levels of calcium, ALP, hS-CRP and TMAO which had been confirmed in the univariable logistic regression analysis

^‡Note that the odds ratio corresponds to a unit increase in the explanatory variable.