Sleep Quality and Its Correlates among Adolescents of Western Nepal: A Population-Based Study

Table 3

Multivariate analysis of factors associated with sleep quality.

Characteristics	COR (95% CI)^#	AOR (95% CI)^##
Age (≤17 years)	1.59 (1.12-2.27)	1.55 (1.03-2.33)
Ethnicity (Brahmin/Chhetri)	1.44 (1.01-2.07)	1.27 (0.80-2.02)
Religion (Hindu)	2.11 (1.19-3.79)	1.46 (0.72-2.95)
Residence (urban)	1.64 (1.08-2.51)	1.53 (0.92-2.52)
Tobacco user (no)	2.38 (1.19-4.90)	0.98 (0.40-2.39)
Drinking status of father (yes)	1.72 (1.20-2.48)	1.18 (0.77-1.81)
Reasons of selecting current study (own decision)	1.54 (1.64-3.98)	1.77 (1.05-3.00)
Satisfied with academic performance (yes)	1.87 (1.22-2.86)	1.44 (0.86-2.40)
Family members scold you (no)	5.01 (2.39-11.51)	2.13 (0.90-5.38)
Relationship with friends (good)
Average	1.15 (0.77-1.72)	1.01 (0.63-1.63)
Poor	3.20 (1.92-5.40)	2.12 (1.13-4.00)
Relationship with teachers (good)
Average	1.37 (0.89-2.10)	1.07 (0.64-1.78)
Poor	2.90 (1.80-4.71)	1.24 (0.66-2.29)
Run away from home (no)	4.40 (2.06-8.42)	3.15 (1.44-7.21)
Hours spent on internet(≤ 2)	2.10 (1.47-3.10)	1.07 (0.66-1.72)
Hours spend on internet before going to bed(≤ 1)	3.27 (2.24-4.81)	2.67 (1.61-4.48)

(): reference category. Significant at <0.000, 0.01, and 0.05, respectively; #COR calculated from the bivariate logistic regression; ^##AOR calculated from the multivariate logistic regression. Note: if, in the bivariate analysis, the variable was insignificant, it was not included in the multivariate analysis. Fitting null model for pseudo- (pseudo- was calculated by using the “Pscl” package). IIH: . IIhNull: the . G2: minus two times the difference in the . McFadden: . ML: . CU: . The multicollinearity was assessed by using VIF (variance infection factor) by using “car” package; all the VIF were less than 5. So, there is no multicollinearity in this model.