Utilizing distributional analytics and electronic records to assess timeliness of inpatient blood glucose monitoring in non-critical care wards

Table 2 Mean, standard deviation and mixture probability estimates from the mixture models among wards with high power

	Components with SD < 1				Components with SD ≥1
	First	Second	Third	Fourth	Others
Mean (converted to hours according to 24-h clock)	7.23 (6.29, 7.52)	11.38 (11.11, 11.61)	17.18 (17.02, 17.37)	21.94 (21.57, 22.26)	11.22 (4.38, 21.56)
Standard deviation (hours)	0.28 (0.13, 0.75)	0.34 (0.19, 0.44)	0.25 (0.13, 0.38)	0.26 (0.21, 0.93)	5.94 (1.58, 6.55)
Mixture probabilities	0.22 (0.18, 0.29)	0.18 (0.16, 0.2)	0.22 (0.19, 0.23)	0.2 (0.15, 0.23)	0.12 (0.06, 0.22)

Median (minimum, maximum) were reported for mean and standard deviation (SD). For mixture probabilities, we first took the average estimates within each ward and reported the median (minimum, maximum)

ISSN: 1471-2288