Assessing the agreement of biomarker data in the presence of left-censoring

BMC Nephrology

Table 2 Simulation results based on 1000 data sets with sample size of 100 -- Per cent censoring (40%, 25%)

Per cent censoring	True ρ	True ρ _c	Method	Mean ${\hat{ρ}}_{c}$	Relative bias (%)	Empirical SD	Mean SE	The percentage of 95% confidence intervals that include true value of CCC
(40%, 25%)	0.25	0.238	1. Delete the pair	0.1240	−47.90	0.1279	0.1230	84.9
			2. Replace by LOD	0.1929	−18.95	0.1319	0.0843	86.8
			3. Replace by 0.5 × LOD	0.1856	−22.02	0.1092	0.0845	85.9
			4. Replace by c × LOD	0.1850	−22.27	0.1180	0.0844	85.7
			5. ML	0.2340	−1.68	0.0984	0.0971	93.9
	0.50	0.476	1. Delete the pair	0.2754	−42.14	0.1190	0.1108	60.2
			2. Replace by LOD	0.3901	−18.05	0.1219	0.0727	80.1
			3. Replace by 0.5 × LOD	0.3661	−23.09	0.1488	0.0736	73.7
			4. Replace by c × LOD	0.3664	−23.03	0.1603	0.0735	74.0
			5. ML	0.4692	−1.43	0.0847	0.0811	93.8
	0.75	0.714	1. Delete the pair	0.4966	−30.45	0.0978	0.0851	30.7
			2. Replace by LOD	0.6268	−12.21	0.1054	0.0500	68.7
			3. Replace by 0.5 × LOD	0.6055	−15.20	0.1389	0.0512	64.7
			4. Replace by c × LOD	0.6083	−14.80	0.1306	0.0513	65.0
			5. ML	0.7066	−1.04	0.0554	0.0528	94.8

Abbreviations: LLD Lower Limit of Detection, ML Maximum Likelihood, C = Random number from uniform distribution (0, 1), CI, confidence interval, Relative bias = $\frac{(Mean {\hat{ρ}}_{c} - True ρ_{c})}{True ρ_{c}} \times 100 %$ , CCC⁵= $ρ_{c} = \frac{2 COV (X, Y)}{VAR (X) + VAR (Y) + {(E (X) - E (Y))}^{2}}$ .

ISSN: 1471-2369