Table 4 Univariable regression analysis of ΔBMD after 1 year of follow-up, in tertiles with high, low and no BMD loss, in the spine and the hip (dependent variables) and disease and demographic variables (independent variables)

RR (95% CI)	High BMD loss	Low BMD loss	No BMD loss
Age, years†	1.02 (1.01 to 1.03)	1.00 (0.99 to 1.01)	1
Male*	1.64 (1.08 to 2.53)	1.55 (1.03 to 2.46)	1
Postmenopausal women*	1.44 (1.22 to 1.57)	1.18 (0.90 to 1.40)	1
Current smokers†	1.13 (0.78 to 1.50)	1.39 (1.01 to 1.75)	1
BMI, kg/m²*	0.99 (0.96 to 1.02)	1.01 (0.97 to 1.04)	1
SHS baseline†	1.03 (1.01 to 1.05)	0.99 (0.98 to 1.02)	1
SHS progression >SDC*	1.99 (1.13 to 3.84)	2.41 (1.19 to 4.40)	1
DAS baseline*	1.01 (0.86 to 1.17)	1.02 (0.87 to 1.18)	1
ΔDAS after 6 months*	1.16 (0.92 to 1.47)	1.06 (0.84 to 1.34)	1
HAQ baseline*	1.17 (0.97 to 1.36)	1.07 (0.87 to 1.27)	1
ΔHAQ after 6 months*	1.03 (0.69 to 1.54)	1.31 (0.86 to 1.98)	1
Positive RF*	1.03 (0.82 to 1.20)	0.96 (0.75 to 1.15)	1
Bisphosphonate users*	0.34 (0.13 to 0.77)	0.44 (0.20 to 0.93)	1
Vitamin D and/or calcium supplement*	0.66 (0.36 to 1.15)	0.71 (0.39 to 1.20)	1

BMD, bone mineral density; BMI, body mass index; RF, rheumatoid factor status; DAS, disease activity score; HAQ, health assessment questionnaire; SHS, Sharp–van der Heijde score; SDC, smallest detectable change.
The tertile with no BMD loss is used as reference population.
*Spine L2–4. †Total hip.