Table 5.

CMS 50-DL SpO₂ bias and A_RMS statistics for differences from SpO₂ at Hct = 40%

O2Hb range	HCT = 30		HCT = 20		HCT = 10
	Bias (%)	ARMS (%)	Bias (%)	ARMS (%)	Bias (%)	ARMS (%)
60–69.9	− 0.18	0.41	2.52	2.52	4.93	4.94
70–79.9	0.69	0.70	2.10	2.10	3.50	3.53
80–89.9	0.79	0.79	1.72	1.72	1.92	1.98
90–100	0.10	0.33	1.37	1.37	0.18	0.56
70–100	0.52	0.64	1.73	1.76	1.86	2.35

For each device, the curve fit at hematocrit 40% was used as a “reference” performance level and was subtracted from the curve fits at hematocrits of 30%, 20%, and 10%. The bias (%) and A_RMS (%) were calculated over the 70–100% O₂Hb range and at each 10% OHb range down to 60%. A_RMS is used to characterize pulse oximeter accuracy and A_RMS g₂reater than 3% used as a threshold for inaccuracy