Joint Effect of Sleep Duration and Sleep Quality on Self-Rated Health Among Canadian Adults: Estimating Relative Excess Risk due to Interaction (RERI) from a Nationwide Survey

Shirmin Bintay Kader¹Nahin Shakurun¹Jubayer Mumin^2*Naomi Noor²

• ¹University of Saskatchewan, Saskatoon, Saskatchewan, Canada
• ²Karolinska Institutet (KI), Solna, Sweden

Background: Self-rated health (SRH) is a globally recognized measure of health status. Previous studies have established that inadequate sleep duration and trouble falling asleep combinedly has a greater negative impact on health than either factor alone. This study aims to investigate the excess relative risk due to the interaction between short sleep duration and trouble sleeping on SRH. Method: We used the 2017-18 Canadian Community Health Survey (CCHS) publicly used microdata file. SRH was measured on a 5-point Likert scale from poor to excellent and dichotomized into "Good or Better" and "Fair or Poor." Sleep duration was categorized into "Less than 7 hours" and "More than 7 hours," while trouble sleeping was categorized as "Yes" or "No." A joint variable derived from these created four groups: "no sleep issues," "fewer sleeping hours (<7 hours) only," "trouble sleeping only," and "fewer hours & trouble sleeping." A weighted univariable and multivariable logistic regression with robust variance estimation was conducted to estimate relative risk due to interaction. Results: Among Canadian adults, 40.96% reported less than 7 hours of sleep, and 48.38% reported trouble sleeping. Approximately 11% had Fair or Poor SRH. The odds ratio for Fair or Poor SRH was 1.34 (95% CI: 1.23-1.46) for short sleep duration, 2.38 (95% CI: 2.18-2.61) for troubled sleep, and 2.97 (95% CI: 2.66-3.33) for both conditions. The adjusted RERI was 0.80 (95% CI: 0.40-1.21). Implication: These results imply that shorter duration of sleep and troubled sleeping may increase the negative influence on self-rated health.