Mapping of disease-specific Oxford Knee Score onto EQ-5D-5L utility index in knee osteoarthritis

Table 4 Measures of performance (accuracy) of the best models in the five domains on the estimation and external validation sample

Domain	Model type	Preparation of predictors	Tuned hyperparameters	Accuracy
				Estimation sample			External validation sample
				Baseline¹	Crude²	CV³	Crude²
				(95% CI)	(95% CI)	(SD)	(95% CI)
MO	Penalized regression	Pre-processed	alpha = 1	26.5% (22.6,.30.9)	0.658 (0.612, 0.701)	0.656	0.687
			criteria = aic			(0.037)	(0.593, 0.770)
			link = cauchit
SC	Random forest	RFE	nsets = 150	59.4% (54.7, 63.9)	0.840 (0.803, 0.872)	0.724	0.669
			ntreeperdiv = 150			(0.039)	(0.575, 0.754)
			ntreefinal = 600
UA	Random forest	All predictors	nsets = 50	25.4% (21.5, 29.7)	0.882 (0.848, 0.91)	0.604	0.687
			ntreeperdiv = 100			(0.044)	(0.593, 0.770)
			ntreefinal = 200
PD	Cumulative probability model	Pre-processed	parallel = TRUE	30.7% (26.5, 35.2)	0.686 (0.642, 0.729)	0.671	0.678
PD	Cumulative probability model	Pre-processed	link = cauchit	30.7% (26.5, 35.2)	0.686 (0.642, 0.729)	(0.039)	(0.584, 0.762)
AD	CART	RFE	cp = 0.00645	30.3% (26.1, 34.7)	0.452 (0.405, 0.499)	0.435	0.357
			split = abs			(0.038)	(0.269, 0.451)
			prune = mc

¹Baseline accuracy in each domain is the proportion of the most common level
²The crude accuracy is the proportion of all correct predictions
³The cross-validated accuracy rate was the average of the 5 × 3-folds accuracy rates conducted on the estimation sample. MO is mobility, SC self-care, UA usual activities, PD pain/discomfort, and AD anxiety/depression. Pre-processed predictors extracted using PCA; alpha (α), the term of penalty; criteria, criteria used for selecting the optimum magnitude of penalty; Link, link function used to transform cumulative probability into an unbounded scale; RFE, recursive feature elimination; nsets, number of score sets tried before the approximation of the optimal score set; ntreeperdiv, number of trees in the smaller forests; ntreefinal, number of trees in the final OF constructed using the optimized score set; parallel, parallel curves or not; CP, complexity parameter; Split, criteria for splitting; Prune, criteria for pruning

ISSN: 1749-799X