180 Chapter 8 Model BIC LL Npar Entropy Patients per class Edinburgh cohort 1 2 3 4 5 6 7 1-class 14578·4 -7200·3 29 1 458 – – – – – – 2-class 14167·1 -6902·8 59 0·8 307151– – – – – 3-class 13968·5 -6711·6 89 0·8 173 152133– – – – 4-class 13901·7 -6586·3 119 0·8 162 128 127 41 – – – 5-class 13829·1 -6458·1 149 0·9 147 132 103 39 37 – – 6-class 13869·1 -6383·2 149 0·9 130 123 83 42 42 38 – 7-class 13980·5 -6350·0 209 0·9 113 103 65 57 42 41 37 ARREST cohort 1 2 3 4 5 6 7 1-class 19279·6 -10157·3 26 1 758 – – – – – – 2-class 18636·4 -9741·5 53 0·8 576182– – – – – 3-class 18360·2 -9515·8 80 0·8 410 232116– – – – 4-class 18284·9 -9374·1 107 0·8 284 197 162 115 – – – 5-class 18269·8 -9259·3 134 0·8 276 139 121 115 107 – – 6-class 19409·7 -9171·1 161 0·8 171 153 119 116 106 93 – 7-class 19462·4 -9107·9 188 0·8 172 131 117 112 88 78 60 SAFO cohort 1 2 3 4 5 6 7 1-class 2065·0 -976·1 21 1 214 – – – – – – 2-class 1963·3 -866·3 43 0·8 12589– – – – – 3-class 1945·9 -798·6 65 0·9 1256722– – – – 4-class 1981·6 -757·4 87 0·8 79 714420– – – 5-class 2035·2 -725·1 109 0·9 59 54453620– – 6-class 2121·1 -709·1 131 0·9 58 53 383418 13– 7-class 2174·1 -676·5 153 0·9 51 40 363723 207 Table 2: Model fit statistics. Model fit statistics for latent class models from one to seven classes in the Edinburgh, ARREST, and SAFO cohorts. BIC: Bayesian Information Criteria, defined in Methods. LL: Loglikelihood, measures the fit of the model to the data. Npar: number of parameters, measure of model complexity. Entropy is a measure for class separation: it ranges from zero to one and values of ≥0·8 indicate good separation of the different classes.
RkJQdWJsaXNoZXIy MTk4NDMw