Application of biological variation and sigma metrics to evaluate the performance of HbA_1c in external quality assessment

Figures & data

Figure 1. Biological variation and Sigma Metrics model. Quality criteria are shown in colors for the BV model (A) [minimum (MIN), optimum (OPT), desirable (DES)] and with lines for the SM model (B) (2σ–6σ). Two models were combined as shown in C.

Figure 2. Combined models applied to the performance evaluation for measuring systems. The colors represent five samples: blue, 20210911; red, 20210912; green, 20210913; black, 20210914; yellow, 20210915. The symbols represent specific manufacturers: ellipse, Arkary; rectangle, Bio-Rad; diamond, Medconn; triangle, Primus; down triangle, Roche; cross, Tosoh; star, Mindray; hexagon, Lifotronic.

Table 1. The sigma metrics for measuring systems.

Measuring systems	N	EQA samples
		20210911	20210912	20210913	20210914	20210915
Arkary (Kyoto, Japan)	18	4.50	4.02	5.25	4.44	4.50
Bio-Rad (California, America)	40	2.25	2.42	2.12	2.17	2.30
Lifotronic (Shenzhen, China)	33	4.66	4.19	3.30	2.37	2.65
Medconn (Shanghai, China)	58	3.30	2.21	2.28	3.14	2.55
Mindray (Shenzhen, China)	29	3.92	3.23	3.42	4.96	5.70
Primus (Kansas, America)	36	4.10	3.18	2.95	3.71	2.81
Roche (Mannheim, Germany)	10	3.08	3.68	3.11	4.74	2.95
Tosoh (Tokyo, Japan)	113	4.00	2.90	2.67	4.40	4.65

Figure 3. Combined models applied to the performance evaluation for individual laboratory. The symbols represent specific manufacturers: ellipse, Arkary; rectangle, Bio-Rad; diamond, Medconn; triangle, Primus; down triangle, Roche; cross, Tosoh; star, Mindray; hexagon, Lifotronic.

Table 2. The quality criteria based on biological variation and the number of laboratories that fulfill the criteria.

Performance level	Quality criteria (%)			Laboratories fulfill criteria, n (%)
	TEa	CVa	Biasa	CV	Bias
Minimum	4.7	1.2	2.7	62 (31.6)	174 (88.8)
Desirable	3.1	0.8	1.8	11 (5.6)	124 (63.3)
Optimum	1.6	0.4	0.9	1 (0.5)	51(26.0)