Predictors of use of pain medications for persistent knee pain after primary Total Knee Arthroplasty: a cohort study using an institutional joint registry

Table 5 Multivariable-adjusted^* predictors of use of opioid pain medications after primary total knee arthroplasty

	Multivariable-adjusted^* analysis, 2-year follow-up			Multivariable-adjusted^* analysis, 5-year follow-up
	Odds ratio	(95% CI)	P-value	Odds ratio	(95% CI)	P-value
Male gender (reference female)	0.54	(0.32, 0.92)	0.02^a	0.39	(0.18, 0.84)	0.02^a
Age (reference ≤ 60 yrs)
> 60 to 70 yrs	0.66	(0.36, 1.20)	0.17	0.80	(0.38, 1.70)	0.57
> 70 to 80 yrs	0.27	(0.13, 0.57)	0.001^a	0.34	(0.14, 0.83)	0.02^a
> 80 yrs	0.48	(0.19, 1.22)	0.12	0.39	(0.10, 1.57)	0.19
Body mass index (reference < 25 kg/m²)
25.0 to 29.9	1.00	(0.46, 2.17)	1.00	0.65	(0.24, 1.74)	0.39
30.0 to 34.9	1.10	(0.50, 2.44)	0.81	0.88	(0.31, 2.47)	0.80
35.0 to 39.9	1.25	(0.49, 3.22)	0.64	0.86	(0.29, 2.58)	0.79
≥ 40.0	0.54	(0.19, 1.59)	0.27	0.40	(0.10, 1.64)	0.20
Deyo-Charlson index (5-point change)	1.28	(0.70, 2.33)	0.43	1.03	(0.48, 2.21)	0.93
Depression (reference no depression)	1.05	(0.55, 2.00)	0.88	1.91	(0.83, 4.41)	0.13
Anxiety (reference no anxiety)	3.05	(1.62, 5.72)	0.001^a	3.98	(1.68, 9.43)	0.002^a

*Adjusted for American Society of Anesthesiologists (ASA) score, distance from the medical center, income, operative diagnosis and implant fixation in addition to all the variables in the table; ^aSignificant odds ratios and P-values; CI, confidence interval. Goodness of fit statistic calculated as quasi likelihood ratio with lower value indicating a better fit: 2 years, 824.3; 5 years, 519.1.

ISSN: 1478-6362