Table 3.

Multinomial regression sensitivity analysis of hip data after n = 33 vignettes scored as “disagreement” by the expert voting panel have been removed. “Rarely appropriate” is the reference category.

Classification	AAOS Prognostic/Predictor Variable	Rating for Each Prognostic/Predictor Variable	B	OR
Appropriate	Intercept		−987.51
	Function-limiting pain	Pain at rest or night	270.20	2.21 × 10¹¹⁷
		Short distances	212.73	2.43 × 10⁹²
		Moderate to long distances	0^a
	Range-of-motion limitation	Severe	85.84	1.89 × 10³⁷
		Moderate	52.12	4.28 × 10²²
		Minimal	0^a
	Risk of negative outcome	No modifiable risk factors	154.53	1.30 × 10⁶⁷
	Risk of negative outcome	Modifiable risk factors	0^a
	Hip radiographic evaluation	Severe OA^b	665.48	1.03 × 10²⁸⁹
		Moderate OA	267.66	1.75 × 10¹¹⁶
		Minimal OA with FAI	2.74	15.43
		Minimal OA with AD	−52.08	2.42⁻²³
		Minimal OA	0^a
	Age	Elderly	543.03	6.83 × 10²³⁵
		Middle-aged	328.15	3.27 × 10¹⁴²
		Young	0^a
May be appropriate	Intercept		−82.29
	Function-limiting pain	Pain at rest or night	24.89	6.46 × 10¹⁰
		Short distances	28.47	2.31 × 10¹²
		Moderate to long distances	0^a
	Range-of-motion limitation	Severe	25.63	1.35 × 10¹¹
		Moderate	21.20	1.60 × 10⁹
		Minimal	0^a
	Risk of negative outcome	No modifiable risk factors	1.96	7.09
	Risk of negative outcome	Modifiable risk factors	0^a
	Hip radiographic evaluation	Severe OA	81.31	2.05 × 10³⁵
		Moderate OA	81.31	2.05 × 10³⁵
		Minimal OA with FAI	3.67	39.24
		Minimal OA with AD	8.71	6079.75
		Minimal OA	0^a
	Age	Elderly	53.87	2.49 × 10²³
		Middle-aged	22.75	7.59 × 10⁹
		Young	0^a

↵a Parameter fixed to zero because it is redundant.
↵b Computing the exponent of the coefficient for severe OA in the “appropriate” relative to “rarely appropriate” classification exceeds computational limits, producing a value of infinity. Therefore, a lower value than the coefficient was used to compute the reported OR. OA: osteoarthritis; AAOS: American Academy of Orthopaedic Surgeons; FAI: femoroacetabular impingement; AD: acetabular dysplasia.