. 2016 Apr 15;31(6):984–991. doi: 10.1093/ndt/gfw062

Table 4.

Nocturnal Trial: comparison of changes for frequent (six times per week) versus conventional (three times per week) in-center hemodialysis

Variable	Trt	Observed data^a (mean ± SD)			Adjusted means and treatment effects^b (± SE or 95% CI)
		Baseline	Month 4	Month 12	Month 4		Month 12
		Baseline	Month 4	Month 12	Change from baseline	Treatment comparison (6× versus 3×)	Change from baseline	Treatment comparison (6× versus 3×)
SPI II	3×	32.0 ± 18.4	33.3 ± 19.1	33.0 ± 23.1	1.7 ± 2.4	−2.9 (−9.3–3.4)	+1.2 ± 2.8	−4.5 (−12.2–3.2)
SPI II	6×	33.8 ± 17.4	30.9 ± 16.3	29.8 ± 17.7	−1.3 ± 2.3	−2.9 (−9.3–3.4)	−3.3 ± 2.8	−4.5 (−12.2–3.2)
Hours of sleep	3×	6.37 ± 1.45	6.72 ± 1.83	6.24 ± 1.55	0.25 ± 0.22	−0.13 (−0.75–0.49)	−0.16 ± 0.19	0.43 (−0.09–0.96)
Hours of sleep	6×	6.51 ± 1.43	6.59 ± 1.70	6.80 ± 1.71	0.12 ± 0.23	−0.13 (−0.75–0.49)	+0.27 ± 0.19	0.43 (−0.09–0.96)

SE, standard error; Trt, treatment frequency.

^aStatistics apply to subjects with measures at all three time points.

^bValues adjusted for baseline score.