Spss v20

Clinical characteristics are expressed as the means ± standard deviations (SDs) for continuous variables and as percentage values for discrete variables. The distribution of the variables was analyzed via the Kolmogorov‒Smirnov test. The clinical characteristics of patients were compared via Student’s t test for continuous variables and Fisher’s exact test for discrete variables. Differential RNA expression analysis between conditions was performed via the DESeq2 method (version 3.4, available at http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html)^{19 (link)}. We considered those RNAs with a p value corrected by FDR (P adj) ≤ 0.05 as differentially expressed to avoid the identification of false positives across the differential expression data^{20 (link)}. Gene predictions were estimated using the Cufflinks method^{21 (link)}, and the expression levels were calculated via HTSeq software (version 0.5.4p323, available at https://pypi.python.org/pypi/HTSeq)^{22 (link)}. This method eliminates the multimapped reads, and only the unique reads are considered for gene expression estimation. The edgeR method (version 3.2.4, available at https://bioconductor.org/packages/3.20/bioc/html/edgeR.html) was applied for differential expression analysis between conditions^{23 (link)}. This method relies on different normalization processes based on in-depth global samples, CG compositions and lengths of genes. In the differential expression process, this method relies on a Poisson model to estimate the variance of the RNA-seq data for differential expression. Significant mean differences in molecule levels were analyzed via Student’s t test for variables with a normal distribution and the Mann–Whitney U test for variables with a nonnormal distribution. Finally, Pearson’s correlation coefficients were calculated to determine the relationships among variables with a normal distribution. p < 0.05 was considered statistically significant. All statistical analyses were performed via SPSS (v.20.0) software (IBM SPSS Inc., USA), R (version R-4.3.1) or GraphPad Prism 8.

We used κ (kappa)20
21 to measure inter-rater reliability for
individual Jadad and NOS questions. We interpreted κ values as follows:
>0.80 was very good, 0.61–0.80 was good, 0.41–0.60 was moderate,
0.21–0.40 was fair and <0.21 was poor.22
For test–retest reliability, each rater re-assessed half of the articles to
which they had been assigned during the inter-rater reliability phase. The
re-assessments took place 2 months after the inter-rater reliability
phase13 (link) to minimise the possibility that
recall of the first assessments would influence the second assessments.
We employed the intraclass correlation coefficient-model 2,1 or ICC(2,1)23 (link) to measure inter-rater and test–retest
reliability for the Jadad and NOS total scores. We computed separate ICC(2,1) values
for consistency (systematic differences between raters are considered irrelevant) and
absolute agreement (systematic differences between raters are considered
relevant).24 ICC(2,1) values were
interpreted as follows: >0.75 was excellent, 0.40–0.75 was fair to good
and <0.40 was poor.25
We calculated two sets of ICC(2,1)s for the Jadad Scale. The first set pertained to
the six-item Jadad Scale,19 (link) and the second set
pertained to the original three-item Jadad Scale.6 (link)
SAS V.9.2 (The SAS Institute) was used to calculate κ; SPSS V.20 (IBM Corp.)
was used to calculate ICC(2,1). The level of significance was
α=0.05.

Since all nine participants had completed nine study nights, the PSG and SWA collected 81 data points for each sleep variable except for AW2, which collected 79 data points due to technical issues. All data obtained from the PSG, AW2 and SWA were aligned to the timing of PSG lights-out and lights-on. A linear mixed model was applied to analyze differences in sleep variables between the devices at each temperature condition, with temperature conditions and sleep devices (PSG, AW2 and SWA) as fixed factors and participants as random factors. Further post-hoc pairwise comparisons were performed using the Fisher’s Least Significant Difference procedure. The Bland–Altman (B–A) plots (MedCalc software, Belgium) displayed the mean bias (the average of the differences between two methods) and 95% limits of agreement (the mean bias plus or minus 1.96 times its SD) [13] (link). Additionally, B–A plot with multiple measurements per subject with the true value varies model was also performed due to the repeated measure design. Since PSG is considered the gold standard measurement for sleep, plots of the differences between AW2/ SWA and PSG against PSG rather than the mean of the two methods were displayed in this manuscript [14] (link). The linear regression was used to evaluate the associations between sleep parameters (SOL, WASO, TST and SE) collected from AW2/SWA and PSG (SPSS v20, Chicago, IL). Wake and sleep epoch (one epoch, 30 s) agreements were analyzed for AW2/SWA against PSG using the kappa statistic, which determines the amount of agreement that can be expected by chance [15] . The kappa statistic ranges from 1 which demonstrates perfect agreement, to 0 which demonstrates agreement based on chance alone, and to −1 which demonstrates complete disagreement [1] (link). As the SWA is limited to estimating sleep and wake in 1-min epochs, each 1 min output was divided to provide an equivalent measure in two 30 s epochs as reported previously [10] (link). For sleep and wake epoch analysis, data were coded as 0=wake and 1=sleep. Overall agreement rates (percentage agreement), sensitivity, specificity and kappa statistic were calculated using SPSS v20. Sensitivity is a measure of the ability of the AW2 or SWA to detect sleep when the PSG has also scored sleep, and calculated as the number of true sleep epoch/(number of true sleep+number of false wake epoch). Specificity is a measure of the ability of the AW2 or SWA to detect wake when the PSG indicated the same, and calculated as the number of true wake epoch/(number of true wake+number of false sleep epoch) [2] (link).