Page 27 - qp_17

a) Examine the SCATTER and BIAS plot done in step (6) for constant
scatter.
b) Examine for the dramatic and significant differences (approximately 3:1 or
greater) between standard deviations at the upper and lower ends of the
range.

11. Assessment of Acceptability
a) Calculate difference (bias) between the methods
• Calculate BIAS (differences between 2 results) at the medical decision
limits or at lower or upper reference ranges using linear regression
formula obtained in step (8) above.
Y= bX + a
• bias at medical decision limit (Xmd)
• Replace X value with Xmd value to get Y value
• PercentageBias = Y- Xmdx 100%
Xmd
If BIAScalculated at medical decision limit < Bias at the desirable or minimum
specification (BIAS as shown in biological variation table)  then the
method is acceptable

b) Calculate Total Error
• Calculate Total Error between 2 method using formula:
• TEcalculated= bias + 1.65 CV
• (CV of new method) @ 90% CI.
• TE CALCULATED< Total Allowable Error in Biological Variation (minimal
@ desirable?)ACCEPTABLE