Table 2.

Studies showing negative results between OA and fracture risk

Name of study	Participant numbers	Type of study	Had OA	OA diagnosis	Comparator	BMD measurement site	Result
Arden et al. (1996) [19]	937 (92 fractures)	Case-control study	319 with OA	OA was classified radiologically using standard x-rays of the pelvis, thoracolumbar spine, hands and weight-bearing knees. Radiographs were scored to the KL method	618 no OA	Lumbar spine and femoral neck	Despite having increased BMD of 5.3%, subjects with hip OA had a significantly increased risk of fracture compared to controls
Arden et al. (2006) [20]	6641	Randomised control trial	422 with OA and 277 with prevalent OA (clinically diagnosed)	The knee pain and OA questionnaire. They were also asked if they had ever received a clinician diagnosis of knee OA: “Has a doctor ever told you that you have OA of the knee?”	No knee OA 5774 (clinically diagnosed)	BMD not recorded	Patients with a clinical diagnosis of knee OA and with knee pain have an increased risk of non-vertebral and hip fracture
Bergink et al. (2003) [21]	4239	Cohort study	1466 fracture group contains 320 OA cases	ROA was assessed by means of the KL grading system in 5 grades (from 0 to 4)	2773 non-fracture group contained 675 OA cases	Lumbar spine and femoral neck	Although people with ROA had a higher BMD, their incident fracture risk was increased as compared with those without ROA
Chan et al. (2014) [22]	3864	Population-based prospective study	1077 with OA 325 fractures	The presence of OA was ascertained at baseline by self-reported diagnosis	1787 no OA 745 fractures	Lumbar spine and femoral neck	Overall, 29% of women and 26% of men had reported a diagnosis of OA. Fracture risk was significantly higher in women with OA than those without OA
Jones et al. (1995) [23]	1821	Longitudinal population-based study	462 with OA	Medication use and self-reported arthritis were assessed by a structured personal interview	1359 no OA	Lumbar spine and femoral neck	Individuals with self-reported OA, despite higher BMD, are not protected against non-vertebral osteoporotic fracture
Lee et al. (2014) [24]	1829	Cross-sectional study	34.20%	Radiographic knee OA was defined as KL grade ≥ 2	65.80%	Lumbar spine and femoral neck	In both sexes, the prevalence of vertebral fractures increased with age and was higher in the knee OA group than in the control group (in men, 13.2% in the OA group and 7.9% in the control group; in women, 27.7% in the OA group and 14.7% in the control group)