Table 3.

Variables Associated with 12-Month SF-36 PCS Scores (n=196)

Variable	Bivariable Analysis (unstandardized regression coefficients (95% CI))	p-value	R-square	Multivariable Analysis (unstandardized regression coefficients (95% CI))	p-value
Gender		0.58	<0.01		0.39
- male	reference category			reference category
- female	−1.07 (−4.76 to 2.62)			−1.33 (−4.37 to 1.70)
Age (for each 10-year increment)	−0.78 (−1.80 to 0.23)	0.13	0.01	−1.40 (−2.25 to −0.56)	<0.01
Smoking Status		<0.01	0.06		<0.01
- not currently smoking	reference category
- current smoker	−5.96 (−9.32 to −2.60)			−4.28 (−7.09 to −1.48)
Fracture type		<0.01	0.08		<0.01
- closed	reference category			reference category
- open	−6.77 (−9.93 to −3.61)			−5.00 (−7.65 to −2.35)
Multi-Trauma		0.08	0.02		0.14
- no multi-trauma	reference category			reference category
- multi-trauma	−2.89 (−6.09 to 0.31)			−1.96 (−4.54 to 0.62)
SPOC Score (for each 14-point increment)	−2.92 (−3.58 to −2.25)	<0.01	0.28	−2.67 (−3.31 to −2.03)	<0.01

95% CI = 95% confidence interval

Legend: The unstandardized regression coefficients represent the expected differences in SF-36 PCS scores between the subgroup and the reference category. For example, in our multivariable analysis, for every increase in 10-years of age patient’s will score 1.4 less on the SF-36 PCS (95% CI = −2.25 to −0.56) at 12-months. The R-square value represents the amount of variation explained in 12-month SF-36 PCS scores by each independent variable (e.g. smoking status explained 6% of variation in 12-month SF-36 PCS scores).