Relationship between long working hours and periodontitis among the Korean workers

We aimed to investigate the association between long working hours and periodontitis, and whether such an association constitutes an exposure-response relationship. Data for this study were collected from the Korea National Health and Nutrition Examination Surveys conducted from 2007 to 2014; 17,533 workers (9,483 of men and 8,050 of women) were included. The odds ratios (OR) and 95% confidence intervals (CI) for the analysis of periodontitis defined as positive of Community Periodontal Index in relation to working hours were calculated using multiple logistic regression models with various stratifications. Compared to participants who worked ≤40 hours per week, the prevalence ratio (95% CI) for the periodontitis was 1.19 (1.14–1.24) and full adjusted OR (95% CI) was 1.09 (1.02–1.18) in participants who worked over 40 hours per week. The OR (95% CI) for periodontitis were 1.09 (0.99–1.19) in working group of 40< and ≥52 and 1.10 (1.01–1.20) in working group of >52 hours per week with a significant trend (p = 0.0233) even after adjusting for age, socioeconomic status, healthy behaviour, chronic disease, and dental care status. Long working hours are associated with periodontitis among Korean workers in an exposure-response manner.

Continued worldwide ranges from 15% to 46%, and is 33% in South Korea [17][18][19][20] . The rate of severe periodontal disease is 8.9% in the United States 17 . Furthermore, according to the Global Burden of Disease 2015, prevalence of periodontal diseases was over 25% higher than 2005, disability-adjusted life year estimates of severe periodontitis was over 75% higher than 1990 21,22 . Considering the high prevalence of periodontitis and its detrimental effect on the general population 23 , it is considered a public health problem 24 . Especially, working population could be vulnerable for periodontal disorders considering age distribution of severe periodontitis and lack of time to medical examination. The prevalence of most of adults' oral disorders peaked after 70 years old, while that of severe periodontal disease peaked nearly 2 decades earlier 21 . A previous study indicated that workers with LWH have increased risks of forgoing healthcare needs because of the lack of time 25 . Hence, periodontitis may be aggravated by the lack of regular dental examinations 26 . Therefore, investigating an association between periodontal disease and an undesirable work environment would be a major area of interest within the field of both occupational and dental health. Furthermore, linking LWH and strenuous shift work to periodontitis would reveal another factor contributing to the deterioration of health. While a recent study showed that undesirable work environments may adversely impact periodontal health 27 , too little is known about the relationship between undesirable work environments and periodontitis, particularly in terms of LWH.
The purpose of this investigation was to explore the relationship between LWH and periodontitis.

Results
The baseline characteristics of participants with and without periodontitis are shown in Table 1. Of the 17,533 participants in this research, 5,201 had periodontitis (29.7%); the prevalence of periodontitis was significantly higher in males (36.1%) than in females (22.1%). A significant association was also found between the prevalence of periodontitis and age; with the oldest age group (51-65 years) having the highest prevalence (45.1%) and the youngest (19-30 years) having the lowest (5.3%). The prevalence of periodontitis was higher among individuals with lower educational and household income levels. Current smokers had a significantly higher prevalence of periodontitis (P < 0.0001). Those who drank alcohol heavily (34.5%) and those with abnormal sleep durations (31.1%) also showed higher rates of periodontitis, although with no statistical significance. The prevalence of periodontitis was lower among those who engaged in vigorous physical activity 3 days and more (27.3%) compared to those who engaged in less than 3 days or no activity per week. Central obesity, diabetes, hypertension, elevated triglycerides, and decreased high density lipoprotein cholesterol were associated with significantly higher rates of periodontitis. Daily oral care or preventive annual dental examinations were associated with significantly lower periodontitis rates. Green-collar, shift, and LWH workers showed significantly higher rates of periodontitis than other worker categories. The specific results of CPI according to oral sextant anatomy are presented in Supplementary Table 1.
The prevalence ratio (PR) and 95% confidence interval (CI) for periodontitis by working hours and CPI score are presented in Table 2. The prevalence of periodontitis was significantly associated with LWH (PR: 1.19, 95% CI: 1.14-1.24). Moreover, there were significant relationship between working hours and periodontitis, depending on increased working hours. The highest PR between periodontitis and LWH was founded in workers who working over 52 hours per week with CPI 4 group (PR: 1,46, 95% CI: 1.30-1.65).
The relationship between LWH and periodontitis according to our multiple logistic regression models are shown in Table 3. Among all participants, the LWH group was more likely to experience periodontitis after adjusting for all confounding factors. Both men and women in the LWH group were more likely to have periodontitis when stratified individually.
The correlations between periodontitis and the numbers of weekly work hours are presented in showed higher periodontitis rates. There was a significantly increased risk of periodontitis according to LWH in those who did not undergo preventive annual dental health examinations (P for trend, 0.0028). However, there was no significant association between working hours and periodontitis among participants who did not care their oral health daily. The association between periodontitis and LWH after stratification according to smoking and drinking status were shown in Table 4 too. There were increased, but statistically non-significant association between periodontitis and working hours. Finally, when stratified by working schedule, a significant exposure-response relationship between periodontitis and working hours was only observed among subjects with fixed working schedules.

Discussion
We found that LWH in workers are associated with a significantly increased prevalence of periodontitis. Furthermore, this effect exhibited an exposure-response relationship even after subjects were adjusting for, and stratified according to, a number of covariates.  Table 3. Odds Ratio (OR) and 95% confidence interval (CI) for periodontitis according to long working hours with subgroup analysis. All models were adjusted for age, sex, socioeconomic status (educational level and household income level), occupational classification, job position, health behaviour (alcohol drinking, smoking, physical activity level, and daily sleep duration), chronic disorders (metabolic syndrome), and dental care (daily dental care and annually dental examination) except a stratified variable, respectively. This is the first study to link the effect of excessive working hours, to periodontitis. However, our results are consistent with those of a previous study that found a relationship between shift work and periodontitis 27 . Nowadays, many individuals are exposed to the shift work and/or LWH; such substandard working conditions can interfere with workers' recreation and rest periods, which are fundamental to one's wellbeing 5 . Our results support the notion that engaging in LWH or abnormal schedules is related with decreased heath condition, with periodontitis being one of the signs.
There are several possible explanations for our results. Extended work can activate the hypothalamicpituitary-adrenal (HPA) stress-response system, which is a possible precursor for periodontal disease as activation of the HPA axis stimulates the adrenal cortex to raise the level of glucocorticoids 28 . Work stress also stimulates the sympathetic nervous system to secrete catecholamines (norepinephrine and epinephrine) in the adrenal medulla. Stress-related hormones have been widely implicated in periodontal disease; previous studies revealed that patients with periodontal disease have above-normal concentrations of total urinary metanephrine (a metabolite of epinephrine) and salivary cortisol [29][30][31] . The stress-response system may also attenuate immunity, leading to periodontal tissue breakdown 32 .
Due to lack of time to recovery stress from work, the immune system might bring dysfunction by being not attentive enough so that increased chance to infectious agents (viruses and bacteria) enter the whole body and cause infectious disease 33 . Periodontitis is an inflammatory chronic disease of gingival tissue caused by bacterial infection 34 . Thus, periodontitis among works who worked long hours could be aggravated by increased vulnerability from infectious agents due to not enough time to recovery of work stress. Further researches are required to establish the association between LWH and periodontitis as an infectious disease.
Work-related stress can also associated with periodontal disease, as it can induce habits such as smoking, excessive alcohol consumption, and neglect of oral hygiene. Tobacco use is a widely-accepted risk and aggravating factor for periodontal disease 18,32,35 . It is well known that cigarette smoke exposure is a strong risk factor for periodontitis, as it reduces elastase and neutrophil levels while elevating T-cells in the oral cavity 36 Table 4. Odds Ratio (OR) and 95% confidence interval (CI) for periodontitis by long working hours category with subgroup analysis. All models were adjusted for age, sex, socioeconomic status (educational level and household income level), occupational classification, job position, health behaviour (alcohol drinking, smoking, physical activity level, and daily sleep duration), chronic disorders (metabolic syndrome), and dental care (daily dental care and annually dental examination) except a stratified variable, respectively. on effects of alcohol consumption, the linkage of periodontitis is debated 38,39 . Current research also indicated closed but non-statistical significant relationship between LWH and periodontitis according to drinking or smoking status. Further studies are needed to determine the association between alcohol consumption or smoking and periodontitis focused on working population. We demonstrated that a lack of regular dental examinations is significantly linked to the association between periodontitis and LWH. A previous study examined the relationship between LWH and accessibility to hospital facilities 25 . Many workers with LWH could not meet their healthcare needs because of the lack of time. As periodontitis is one important early indicator of poor health 9 , it may be considered an identifier of workers-at-risk. Thus, there is a need for enhancing accessibility to regular medical or dental check-ups in the workplace.
Inadequate sleep is a potential risk factor for increased inflammatory and pro-inflammatory factors that could aggravate periodontal diseases 40 . However, previous research failed to show a significant correlation between short or long sleep durations and periodontitis 41 .
In current study, subjects who had vigorous physical activity at least 3days and more per a week showed decreased risk for periodontitis. Previous cross-sectional study using 12,110 general population also indicated that "5 and more episodes of moderate or 3 and more episodes of vigorous-intensity physical activity per week" related to lower periodontitis 42 . Furthermore, another cross-sectional study from Jordan with 340 participants indicated that low (38%) and moderated physical activity (28,7%) showed higher prevalence of periodontitis than high physical activity (13.1) which was measured by International Physical Activity Questionnaire (IPAQ, http://www. Ipaq.ki.se/ ipaq.htm) 43 . Moderate physical activity from Jordan study same as our study's 3 times and more vigorous physical activity level. In summary, the periodontitis might be protectable by at least 3 days and more vigorous physical activity. Nevertheless, without comprehensive assessment for the physical activity such as the IPAQ.
A previous study demonstrated a significant association between shift work and periodontitis 27 ; our study showed a significant relationship between LWH and periodontitis only among fixed-scheduled workers, but not shift workers. Furthermore, there were no synergic or additive effects between work schedules and working hours on periodontitis (data not shown). To understand workers' health statuses as related to the workplace, it may be helpful to use alternative analysis methods including stratification by undesirable working conditions such as LWH and shift work.
While this is the first study to investigate the relationship between LWH and periodontal diseases in the working population with a statistically strong power (17,517 participants), there are some limitations to our study. Our research was based in South Korea, which has the second-longest average working hours in OECD countries 44 ; therefore, our results may not be generalizable. However, our findings are still highly relevant to understanding workers' health because working more than 40 hours per week is clearly defined as LWH in many countries. Unfortunately, data from the KNHANES did not include advanced clinical approaches such as imaging and bone loss evaluation for diagnosing periodontitis. Periodontitis positivity in our study was only determined by using the CPI. Furthermore, dental examination for periodontitis was based on 10 index teeth. Previous investigators warned about the possibility of underdiagnosing periodontitis by not examining the mouth fully 45,46 . Thus, results from current investigation of periodontal status linked to working hours were limited only positive of CPI as a screening test for periodontitis which was insufficient to reflect individuals' periodontal status perfectly. Further investigations are required to estimate the association between working conditions and periodontitis using full-mouth examination.
Also, KNHANES had two information (daily oral care and annually dental examination) for oral health behaviors. Periodontal health status might be closely related with tooth brushing, flossing, other supplementary devices or specific oral conditions. Due to nature of data, we could demonstrate oral health behaviors status insufficiently.
Our findings may be somewhat limited by the cross-sectional nature of the KNHANES. study design. It difficult to determine a cause-and-effect relationship between LWH and periodontitis; thus, longitudinal investigations are required for this purpose.
In conclusion, our large, nationally representative study demonstrated a significant association between periodontitis and workers engaged in LWH. The current findings extend our knowledge of the adverse impacts of LWH to include maladies such as periodontitis. Future detailed investigations with longitudinal studies on the link between LWH and periodontal diseases will provide better insight into the effects of undesirable working condition on workers' health. All KNHANSE subjects understood about anonymity to protect themselves and potential risks and benefits with written informed consents before they participated survey.

Study participants.
LWH and occupational characteristics. To assess the total number of weekly working hours, we relied on the following question in the self-reported questionnaire: "How many hours do you work at your job per week on average, including overtime but excluding meal times?" LWH are defined as exceeding the hours of full-time work as designated by the law. In Korea, weekly working hours are limited to 40 as specified by the Standard Labour Act. This Act allowed up to 12 hours of overtime with the agreement of all parties 25,48,49 . For our purposes, LWH was defined as working for >40 hours per week. To investigate whether an exposure-response association existed between LWH and periodontitis, working hours were categorized as ≤40, 40< and ≤52, and >52 (severe LWH) per week.
Occupational classifications were categorized into 4 groups based on the 10 major categories of the International Standard Classifications of Occupations: white collar workers (managers, professionals, technicians, and associate professionals), pink collar workers (clerical support, service, and sales workers), green collar workers (skilled agricultural, forestry, and fishery workers), and blue collar workers (crafts and related trades, plant and machine operators and assemblers, and elementary occupations) 50 . Employment status were divided into three categories; paid workers, self-employed, and another. Another category was included non-paid familial workers. Working schedules were binomially classified according to self-reported questionnaire responses. Those who usually worked during the daytime (06:00-18:00), evening hours (14:00-24:00), or night-time (21:00-08:00) were classified as having fixed schedules, while those who worked according to any other schedule (24 hours shifts, split shifts, or irregular shifts) were classified as shift schedule workers.
Periodontitis and dental care. Dentists examined participants' oral health and periodontal conditions using the Community Periodontal Index (CPI; modified based on the Community Periodontal Index of Treatment Needs or CPITN) created by the World Health Organization 51 . A CPI probe has a 0.5-mm ball tip and a black band 3.5-5.5 mm long that measures the depth of periodontal pockets. Ten index teeth (#11, #16, #17, #26, #27, #31, #36, #37, #46, and #47) were assigned to a sextant (upper right: #17-14, upper anterior: #13-23, upper left: #24-27, lower right: #47-44, lower anterior: #43-33, and lower left: #34-37) and examined respectively. If index teeth were extracted or could not be evaluated, the remaining teeth were checked at their sextants and the highest CPI was scored. The CPI was categorized as code 0 (healthy status), 1 (bleeding on probing, no gingival pockets), 2 (calculus present, no gingival pockets, black band of probe fully visible), 3 (periodontal pockets 4-5 mm), and 4 (periodontal pockets ≥6 mm). In this study, a code of 3 or 4 (periodontal pockets ≥4 mm) for at least 1 sextant was defined as positive for periodontitis.
Oral health can be affected by individuals' oral care habits, and the KNHANES separately questions subjects about daily dental care and annual dental examinations. Daily dental care included questions such as whether teeth were brushed the day before; participants were also asked whether they had a regular dental check-up (not related to disease) in the past year.
Other covariates. We also investigated socioeconomic and lifestyle factors related to both LWH and periodontitis, as people with lower socioeconomic status (i.e., educational and income levels) or behaviours adverse to health (smoking, heavy alcohol drinking, and lower physical activity level) may have an increased risk of periodontitis 52,53 . Furthermore, LWH are also linked to socioeconomic status and health behaviour 54,55 . Additional covariates were age (19−30, 31−40, 41−50, and 51−65 years), education level (below middle school, high school, or college/university), household income quartile, smoking (never, former, or current), alcohol drinking (never, moderate, or heavy), weekly vigorous physical activity day (none, less than 3 days, and 3 days and more), abnormal daily sleep duration (<6 or 8< vs. 6 ~ 8 hours), and metabolic syndrome.
Metabolic syndrome was diagnosed according to previous study based on National Cholesterol Education Program Adult Treatment Panel III criteria and the Korean Society for the Study of Obesity 50 . The presence of more than three of following five abnormalities were defined as a metabolic syndrome: (i) central obesity (waist circumference >90 cm in males or >80 cm in females); (ii) hypertension (blood pressure ≥130/85 mm Hg or antihypertensive drug treatment); (iii) hyperglycemia (fasting glucose level of serum ≥ 100 mg/dl or use of diabetes medication); (iv) high triglyceride (TG) levels (TG ≥ 150 mg/dl or specific treatment for elevated TG); or (v) low high density lipoprotein cholesterol (HDL-C) levels (HDL-C < 40 mg/dl in males and <50 mg/dl in females or the use of treatment for low HDL-C). Household income was compared with the standard income level of Korean citizens, calculated using standardized classification by sex, residence, and age groups. Alcohol consumption was defined differently according to sex; heavy drinking was defined as at least 7 and 5 glasses of alcohol consumed more than twice per week by men and women, respectively.
We used physical activity status focused on vigorous physical activity. Vigorous physical activity level was defined as 20 minutes or more per a day of exertion that produces increased respiration, it were divided into three categories according to vigorous physical activity days per a week following previous studies 42, 43 . Statistical analysis. We analyzed data using the SAS 9.4 software (SAS Institute Inc., Cary, NC). First, the demographics of the study population and the prevalence of periodontitis were calculated. Chi-square tests were used to compare differences in baseline characteristics according to periodontitis. The PR and 95% CI were calculated with working hours and CPI stratification. The OR and 95% CI of the relationship between LWH and periodontitis were estimated using a logistic regression model. The model was adjusted for adjusted for age, sex, socioeconomic status (educational level and household income level), occupational classification, job position, health behaviour (alcohol drinking, smoking, physical activity level, and daily sleep duration), chronic disorders (metabolic syndrome), and dental care (daily dental care and annually dental examination) except a stratified variable (sex, abnormal sleep duration, annual dental examination, daily oral care, vigorous physical activity, smoking status, alcohol drinking, and working schedule), respectively. Next, we used weekly work hours as a categorical variable (≤40, 41-52, or >52 hours) to assess the exposure-response association with periodontitis using fully adjusted logistic models with subgroup analysis according to sex, abnormal sleep duration, annual dental examination, daily oral care, vigorous physical activity, smoking status, alcohol drinking, and working schedule.