Classifying Patients with Chronic Pelvic Pain into Levels of Biopsychosocial Dysfunction Using Latent Class Modeling of Patient Reported Outcome Measures

Table 1

Statistical fit criterion for multiple latent class models based on four PROMIS-based superdomains evaluating patients with chronic pelvic pain. The number of classes tested is shown in the first column, and statistical fit parameters for these are shown in the subsequent columns.

Classes	Parameters	Log-likelihood	BIC^1	aBIC^2	Entropy
1	8		13940	13914	N/A
2	13		13509	13468	0.74
3	18		13376	13319	0.76
4	23		13310	13237	0.82
5	28		13303	13214	0.76
6	33		13295	13190	0.74

Bayesian Information Criterion. ²Sample size-adjusted BIC.