Skip to main content
. Author manuscript; available in PMC: 2016 Apr 14.
Published in final edited form as: J Am Geriatr Soc. 2015 Apr 6;63(4):716–721. doi: 10.1111/jgs.13329

Table 3.

Logistic Regression Analysis showing the association between sleep related complaints and persistent fatigue in unadjusted and adjusted models.

Model 1
Odds Ratio (OR)
Confidence Interval (CI)
Model 2
OR (CI)
Model 3
OR (CI)

Difficulty falling asleep
Rarely/Never 1.0 1.0 1.0
Some Nights 1.44 (1.21 – 1.71) 1.36 (1.14 – 1.62) 1.16 (0.94 – 1.36) (P =.163)
Most nights/Every night 1.71 (1.42 – 2.06) 1.61 (1.33 – 1.94) 1.19 (0.93 – 1.52) (P = .164)

Difficulty staying asleep
Rarely/Never 1.0 1.0 1.0
Some Nights 1.35 (1.21 – 1.71) 1.36 (1.15 – 1.61) 1.32 (1.08 – 1.60) (P=.007)
Most nights/Every night 2.00 (1.64 – 2.43) 2.02 (1.66–2.47) 1.40 (1.09 – 1.79) (P =.009)

Use of Sleep Aids
Rarely/Never 1.0 1.0 1.0
Some Nights 1.37 (1.06–1.77) 1.42 (1.10 – 1.24) 1.01(0.75 – 1.30) (P = .917)
Most nights/Every night 2.44 (2.07 – 2.88) 2.36 (1.99 – 2.79) 1.35 (1.08 – 1.67) (P = .009)

Model 1 = Unadjusted model. Model 2: Model adjusted for age, gender, race, education, marital status.

Model 3: Model 2 + body mass index, pain, pain medication use, depression status, physical activity level, disease burden and Short Physical Performance Battery test score.

Other covariates that showed positive association include; Pain that limits activity (OR (CI): 3.82 (3.08 – 4.74));

Depression (OR (CI): 3.35 (2.87–3.91)); Number of chronic diseases 2–3: OR (CI) (1.73(1.35 – 2.24); Number of chronic diseases >3: (OR (CI) 3.16 (2.46 – 4.06)); Low physical activity: (OR (CI) 1.27 (1.06–1.52). Covariates that showed negative association include: Black race (OR (CI): 0.63 (0.49–0.79)), Hispanic race: OR (CI): 0.71 (0.50 – 0.99)), 9–12 years of education: OR (CI): 0.83 (0.71 – 0.99)), not married status: score OR (CI): 0.80 (0.66 – 0.97)) and higher short physical performance battery score OR (CI): 0.89 (0.85 – 0.91)).