Lower reference limits of quantitative cord glucose-6-phosphate dehydrogenase estimated from healthy term neonates according to the clinical and laboratory standards institute guidelines: a cross sectional retrospective study

Table 2 Characteristics of G6PD values and lower reference limits (2.5th percentiles) in absence of the outliers

	Males (n = 199)	Females (n =176)	Combined (n = 375)
Mean (SD)^a	14.1 (1.7)	14.1 (1.4)	14.1 (1.6)
Median (interquartile range)^a	14.1 (13.0-15.2)	14.1 (13.2-15.1)	14.1 (13.1-15.2)
Mode^a	14.7	15.1	13.3
z-score of skewness^b	-0.20/0.17 = -1.18	-0.30/0.18 = -1.67	-0.24/0.13 = -1.85
z-score of kurtosis^b	-0.13/0.34 = -0.38	-0.11/0.37 = -0.28	-0.06/0.25 = 0.24
Skewness coefficient (P-value)	.24	.11	.06
Kurtosis coefficient (P-value)	.87	.89	.92
Anderson-Darling test (P-value)	>.999	>.999	.12
Shapiro-Wilk test (P-value)	0.64	0.18	0.14
Parametric LRL (95% CI)^a	10.8 (10.3–11.3)	11.0 (10.5–11.6)	11.0 (10.5–11.3) ^c
Bootstrap LRL (95% CI)^a	10.7 (10.1–11.2)	10.6 (10.2–11.6)	10.7 (10.3–11.1)
Non-parametric LRL (95% CI)^a	10.7 (10.0–11.2)	10.6 (9.6–11.6)	10.7 (10.2–11.1)
Harris-Boyd method combination test
Larger SD/Smaller SD		1.2
Normal deviate z test		Zero
Critical z value of 5		6.2

^aG6PD activity expressed as Units/gram Hemoglobin.
^bz-score = Skewness (or Kurtosis)/SD of Skewness (or Kurtosis).
^cThe suggested lower reference limit (LRL) and its 95% confidence interval (CI).

ISSN: 1471-2431