Sas enterprise version 6

We used a chi-square test to define significant differences in mortality rates among patients grouped according to underlying diseases. We used linear and logistic regression analyses to define the risk factors for the clinical outcomes. The data were analyzed using SAS Enterprise version 6.1 (SAS Institute Inc., Cary, NC, USA). A P<0.05 was used to indicate statistical significance.

All statistical analyses were conducted with SAS Enterprise version 6.1. (SAS Institute Inc. USA). Single imputation was used to impute missing values using chained equations [29 (link)] by invoking Proc MI statement in SAS. Categorical variables were described using percentages and frequencies while continuous variables were described using mean and standard deviations. Statistical difference in background characteristics of study individuals were assessed using t-tests for continuous and Fischer’s exact test and Chi-square tests for categorical variables. Analysis of covariance (ANCOVA) was used to assess association of chronotype with quality of life domains and depression inventory. A similar method was also used to assess association between the quality of life domains and depression. ANCOVA assumptions of same slope across the groups were assessed by including the grouping variable along with covariates and the interaction term. Results did not indicate that there were interactions among covariates and the groups under study. Therefore, models without interaction were run for analysis. Age and sex-adjusted correlations were used to assess an association between quality of life domain and depression with objectively measured AHI, cardio-metabolic risk factors and nutrient intake. All analyses were adjusted for age and sex.