Association between weekend catch-up sleep and dyslipidemia among Korean workers

Within competitive sociocultural environments, most Korean workers are likely to shorten their sleep duration during the weekday. Short sleep duration is associated with dyslipidemia; however, studies on the correlation between various sleep patterns and dyslipidemia are still lacking. In hence this study aimed to investigate the association between weekend catch-up sleep (CUS) and dyslipidemia among South Korean workers. Our study used data from the 8th Korea National Health and Nutrition Examination Survey (KNHANES). The analysis covered 4,085 participants, excluding those who were diagnosed with dyslipidemia and not currently participating in economic activities. Weekend CUS was calculated as the absolute difference between self-reported weekday and weekend sleep duration. Dyslipidemia was diagnosed based on the levels of total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides in blood samples collected after 9–12 h of fasting. After adjusting for sociodemographic, economic, health-related, and sleep-related factors, a negative association of weekend CUS with dyslipidemia was observed in male workers (odds ratio: 0.76, 95% confidence interval: 0.61–0.95). Further, workers with total sleep duration of 7–8 h, night workers, and white-collar workers with CUS were at relatively low risk of dyslipidemia compared to the non-CUS group. Less than 2 h of weekend CUS was negatively related to dyslipidemia in Korean workers, especially males. This suggests that sleeping more on weekends for workers who had a lack of sleep during the week can help prevent dyslipidemia.

Participants. The current study used data from the 2019 and 2020 KNHANES, which contains data from 15,469 participants. Participants < 19 years of age (n = 2,730) were excluded from this study. As we aimed to analyze workers, we also excluded individuals not currently participating in economic activities (n = 5,371). In addition, those who were diagnosed with dyslipidemia and current is currently undergoing treatments (n = 1,207) or had missing data (n = 2,076) were excluded. Finally, the study comprised of 4,085 participants (2,206 males and 1,879 females). This study did not require prior consent or approval from an Institutional Review Board because the KNHANES is a secondary dataset and consists of already de-identified data available in the public domain.
Variables. The main variable of interest was weekend CUS calculated using the average weekday and weekend sleep duration from the relevant KNHANES questionnaire. Participants' average weekday and weekend sleep durations were calculated based on their responses to the following questions: On a weekday (or working day), at How many hours do you usually sleep a day? On a weekend (or the day when you do not work, the day before you do not work), How many hours do you usually sleep a day? Weekend CUS was defined as sleep duration in the weekend being longer than that in weekdays 14 Weekend CUS was calculated as the average weekend sleep duration minus the average weekday sleep duration.. Participants were then divided into non-CUS (≤ 0 h) and CUS (0 > h) groups 10 . Additionally, we classified CUS duration into 0 < to 1, 1 < to 2, and > 2 h for subgroup analysis.
The dependent variable was the prevalence of dyslipidemia diagnosed based on the levels of total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides in blood samples collected after 9-12 h of fasting. According to the 2018 Korean Guidelines for the Management of Dyslipidemia, for diagnosis of dyslipidemia, one of the following four criteria was required: (1) total cholesterol ≥ 240 mg/dL, (2) HDL cholesterol ≤ 40 mg/dL, (3) LDL cholesterol ≥ 160 mg/dL, or (4) triglycerides ≥ 200 mg/dL 15 .
The following covariates were included in the analyses: The sociodemographic factors were age (19-29, 30-39, 40-49, 50-59, and ≥ 60 years) and sex (male and female). The socioeconomic factors were education level (middle school or lower, high school, or university or higher), region (metropolitan or rural area), marital status (married or unmarried), occupation (white collar, pink collar, blue collar), and household income (high, middle-high, middle-low, or low). The health-related factors were obstructive sleep apnea calculated by STOP-bang (yes or no), alcohol consumption status (less 1 time per month, 2-4 times per months, over 2 times per week) and smoking status (yes or no). In addition, adjustments were made for average total sleep duration (< 7,7-8,8 <), work pattern (day, night, shift work), physical activity (yes or no), body mass index (underweight, normal, overweight), menopause status (yes or no), hypertension (yes or pre-hypertension or no), and diabetes (yes or prediabetes or no).

Statistical analyses.
Owing to sex differences in physical conditions, all analyses were stratified by sex 16 .
Descriptive analysis using by chi-square test was performed to examine the distribution of the general characteristics of the study population. Multiple logistic regression modelling was used to assess the association between CUS and prevalence of dyslipidemia after adjusting for all covariates. In addition, to find out the association according to the subdivided categories of weekend CUS and dyslipidemia, multiple logistic regression analyses of subgroups were also performed. ORs and 95% CIs were calculated to compare the data of participants with dyslipidemia. Variables were clustered, stratified, and weighted to account for the limited proportion of participants retained in the final analysis 17 . SAS (version 9.4M6; SAS Institute, Cary, NC) was used for all statistical analyses.  Table 2 presents the results from the multiple logistic regression analysis of the association between CUS and dyslipidemia. There was a significant association in males between weekend CUS and dyslipidemia (odds ratio [OR]: 0.76, 95% confidence interval [CI]: 0.61-0.95). However, no such association was found for females.

Results
The results of the subgroup analysis stratified by total sleep duration, work pattern, and occupational categories are shown in Table 3. Male CUS workers who slept for a total average of 7-8 h were less likely to have dyslipidemia compared to non-CUS workers (OR: 0.70, 95% CI: 0.52-0.94). Similarly, male CUS workers with white-collar jobs were at less risk of dyslipidemia compared to non-CUS workers (OR: 0.68, 95% CI: 0.49-0.94). Regardless of sex, night workers with CUS showed a significant association between weekend CUS and dyslipidemia compared to those without CUS (male: OR: 0.38, 95% CI: 0.18-0.83, female: OR: 0.30, 95% CI: 0.13-0.73).

Discussion
In this study, we found that Korean male workers with ≤ 2 h of CUS had a decreased risk of dyslipidemia compared to those without CUS after adjusting for potential covariates. Further, workers with a total sleep duration of 7-8 h, night workers, and white-collar workers with CUS were at relatively low risk of dyslipidemia compared with those without CUS.
Sleep is an important factor in healthcare 18 . Reduced sleep quality or sleep duration could be risk factors for poor physical and psychological health 19,20 . Other studies have shown that excessive sleep has adverse effects on health outcomes 21 . Optimal sleep management is essential for healthcare, but most Koreans, especially those who work, do not get enough sleep 22,23 . Although most people have different lifestyles, Korean workers tend to make up for their lack of sleep on weekdays with weekend sleep 24 . According to studies, to cope with weekly sleep deprivation, weekend CUS is undertaken, which is associated with a lower prevalence of hypertension, obesity, and serum high-sensitivity C-reactive protein levels [25][26][27] . A previous epidemiological study reported that insufficient sleep duration increases the risk of CVD 28 . Likewise, sufficient sleep can reduce the risk of developing CVD 29 . This may explain our finding that supplementing insufficient sleep with weekend CUS is linked to a reduced risk of dyslipidemia.
In our study, workers who had CUS and an optimal sleep duration (7-8 h) on weekdays had a negative relationship with dyslipidemia compared to those who had abnormal sleep durations of < 7 h or > 8 h. A previous study has suggested that abnormal sleep duration during the week is associated with increased mortality in individuals < 65 years old 20 . Similarly, another study showed that those who had appropriate sleep with CUS had a negative correlation with obesity 30 . Therefore, our study suggests the need to keep an optimal sleep duration even if it is supplemented on weekends.
Night work is more strongly associated with dyslipidemia, compared to day or other shift work 31 . Night workers usually receive less sleep than day workers 32 . Sleep deprivation negatively affects metabolism and promotes the development of an atherogenic lipid profile 33 . This may explain our finding that night workers' sleep supplementation on the weekend showed a negative relationship with dyslipidemia compared to day or shift workers. Furthermore, the risk of dyslipidemia is lower when there is ≤ 2 h difference in sleep time between weekdays and weekends. On the other hand, those with > 2 h difference showed a positive relationship, but it was not statistically significant. Obviously, insufficient sleep is associated with negative health effects; however, habitual excessive sleep can also increase the risk of mortality, and if the degree of misalignment is severe, the compensatory effect might disappear 34,35 . Hence, to protect workers from dyslipidemia, we need to identify how to attain enough sleep in general and achieve a balanced sleep duration between weekdays and weekends.
Although the results of this study serve as further evidence in clarifying the negative association between weekend CUS and dyslipidemia, especially among Korean male workers, it has some limitation. First, this study used a cross-sectional data set; thus, we could only determine the association and not investigate the causal relationship between those variables. Therefore, additional research is needed to infer an accurate causality. Second, data regarding sleep time comes from self-report questionnaires; inaccuracies may, thus, occur. As such, the possibility of a difference between actual and reported sleep time cannot be excluded. Third, due to the data limitation, potential risk factors related to sleep and dyslipidemia may existed, such as a diagnosis of insomnia or other medications which affects the levels of lipids not considered in this study.
Despite these limitations, this study has also several strengths. First, dyslipidemia was measured through clinical testing; hence, it was based on more reliable and clear data. Second, since this study was conducted on a nationally representative sample, the results reflect the overall situation in South Korea and could be used to establish health policy.
In conclusion, our results have public health significance because this research provides insight on preventing dyslipidemia, a high-burden disease, by investigating the relationship between weekend CUS and dyslipidemia. Less than 2 h of weekend CUS was negatively related to dyslipidemia, especially among male workers. Workers with 7-8 h of sleep, night workers, and white-collar workers with CUS were at relatively low risk of dyslipidemia