Lifestyle factors associated with presenteeism among city government office workers: a cross-sectional study

Abstract

Objectives: Presenteeism is a critical issue in occupational health. This study aimed to examine the association between presenteeism and subjective sleep quality, smoking status, and alcohol consumption.

Methods: Anonymous data of 777 workers in a Japanese city were retrospectively obtained from City Government Office A. They included variables like absolute presenteeism scores (measured using the Japanese version of the World Health Organization Health and Work Performance Questionnaire short form), gender, age, family status, subjective sleep quality, smoking status, and alcohol consumption. A binary logistic regression analysis was performed with gender, age, family status, subjective sleep quality, smoking status, and alcohol consumption as the independent variables, and absolute presenteeism scores equal to or below 40 as the dependent variable. A gender-stratified binary logistic regression analysis was also performed.

Results: The logistic regression analysis results revealed that absolute presenteeism was positively associated with poor subjective sleep quality among all respondents (odds ratio [OR], 1.70; 95% CI, 1.18-2.44) and men (OR, 1.85; 95% CI, 1.12-3.05) and with current drinkers among women (OR, 3.49; 95% CI, 1.36-8.92); it was negatively associated with age among those who were ≥50 years old (OR, 0.50; 95% CI, 0.27-0.93) and with current drinkers among men (OR, 0.43; 95% CI, 0.20-0.92).

Conclusions: The factors associated with presenteeism differed between men and women office workers, suggesting that gender differences need to be considered when working toward improving workers’ productivity.

Key points: • What is already known on this topic.

A number of health problems are associated with productivity loss. Several recent studies have reported that productivity loss due to presenteeism exceeds that due to absenteeism, increasing the importance of studies on presenteeism. Several studies suggested that subjective sleep quality, smoking, alcohol consumption, and alcohol-related problems were associated with absenteeism and presenteeism. However, few studies have been conducted on city government office workers, and gender differences in the relationship between lifestyle factors and presenteeism have been rarely reported.

• What this study adds.

This study found that poor subjective sleep quality was associated with presenteeism among city government office workers, which is consistent with previous studies of other occupational workers. The association was particularly pronounced among men. Furthermore, this study found that women and men who consume alcohol are at higher and lower risk for presenteeism, respectively. This study contributes to the existing literature by identifying gender differences in the relationship between alcohol consumption and presenteeism.

• How this study might affect research, practice, or policy.

This study suggests that considering gender differences is crucial in conducting research on presenteeism. Gender differences may exist in the relationship between lifestyle and presenteeism among Japanese city government office workers, and gender-sensitive strategies should be implemented to increase productivity.

1. Introduction

A number of health problems are associated with lost productivity,¹ which can be measured by the costs associated with absenteeism or presenteeism.² Whereas absenteeism is generally defined as “not showing up for scheduled work,” presenteeism has 2 major aspects: “sickness presenteeism” and “impaired work functions.”^1,3 “Sickness presenteeism” refers to why workers engage in work when their health status gives them a legitimate reason to take days off.¹ “Impaired work functions” refers to impaired work performance related to health problems.¹ This study focuses on “impaired work functions.” As several recent studies have reported that productivity loss from presenteeism exceeds that from absenteeism,^1,4 the importance of research on presenteeism is growing. It is assumed that the occurrence of presenteeism in administrative agencies, such as City Hall, not only decreases individual and workplace productivity but also increases coworker burden. As a result, the workplace becomes more prone to increased overtime work and mental health problems. Therefore, it is crucial for industrial health activities to identify the factors related to presenteeism and promote improvements.

Previous studies suggest that insomnia is significantly associated with absenteeism and presenteeism.^4,5 Additionally, some studies showed that insomnia-related presenteeism, rather than absenteeism, increases costs.^6,7 Smoking might also be associated with absenteeism and presenteeism.^8,9 It has been estimated that the average annual per smoker excesses for absenteeism and presenteeism costs were $516.56 and $461.92, respectively.¹⁰ According to various studies, including systematic reviews, alcohol consumption and related problems might be associated with absenteeism and presenteeism,^11,12 and have been referred to as “alcohol-related absenteeism” and “alcohol-related presenteeism,” respectively.^11,13 As with insomnia and smoking, alcohol-related presenteeism, rather than absenteeism, is a greater cause of lost productivity.¹⁴

The aim of this study was to examine the association between presenteeism and subjective sleep quality, smoking status, and alcohol consumption. Additionally, because gender differences have been found in disease predispositions to insomnia, smoking, and drinking behaviors or attitudes,^15-18 this study also explored the gender differences in the association between those lifestyle factors and presenteeism.

2. Methods

2.1. Study design and survey respondents

This was a cross-sectional study using existing data. In 2015, as part of its industrial health activities, City Government Office A in Tottori Prefecture, conducted a self-administered, anonymous questionnaire survey related to presenteeism with 893 of its workers. The anonymous data of 777 workers (response rate: 85.3%), which were retrospectively obtained from City Government Office A, contained the following variables: absolute presenteeism score, gender, age, family status, subjective sleep quality, smoking status, and alcohol consumption. However, in this study, data from only 713 of the 777 (91.8%) respondents were used, after excluding data of those who did not indicate their gender, age, and absolute presenteeism score. Among the excluded 64 respondents, 15 did not provide gender or age information and 49 did not indicate their absolute presenteeism score. The overall sample size of 713 was sufficient to detect a weak effect size (d = 0.2). However, the sample size was not large enough to detect the weak effect size by gender. Therefore, the analysis by gender was conducted as a subanalysis.

2.2. Presenteeism

Presenteeism was assessed using the validated Japanese version of the World Health Organization Health and Work Performance Questionnaire (WHO-HPQ) short form.^19-22 The WHO-HPQ is an instrument designed to measure respondents’ productivity using a self-reported number of absences and work performance. The present study assessed the respondents’ absolute presenteeism. Absolute presenteeism is based on one’s own self-rated work performance. The WHO-HPQ contains the following questions: “Using the 0–10 scale, how would you rate your overall job performance on the days you worked during the past four weeks?” The absolute presenteeism score was calculated by multiplying the question’s rating by 10. Therefore, the absolute presenteeism score ranged from 0 (total lack of performance during respondents’ time on the job) to 100 (no lack of performance during their time on the job). Based on previous studies, absolute presenteeism was defined as absolute presenteeism scores that were ≤40.²³ This cut-off value was set by a Japanese study.²³

2.3. Demographic variables

Multiple-choice responses were required for all questions pertaining to demographic variables, which included gender, age, and family status. Although the question on age had 6 options: 10-19, 20-29, 30-39, 40-49, 50-59, and ≥60 years, age was classified into only 4 categories: <30, 30-39, 40-49, and ≥50 years, because among this study’s respondents no women were found in the 10-19 age category and only a few workers (6 men and 1 woman) were found in the ≥60 age category. The question on family status was answered by the respondents using the following options: whether they were living only with their spouse (ie, as a married couple), as a married couple living with their child or children, as a married couple living with their parent(s), living in a 3-generation family, living as singles, living as singles with their parent(s), living as singles with their child or children, and other family structures.

2.4. Subjective sleep quality

Subjective sleep quality was assessed with the question: “During the last 30 days, how would you rate your overall sleep quality?” and the responses were obtained using the following 4 options: “very good,” “fairly good,” “fairly poor,” and “very poor.” The “very good” and “fairly good” responses were then categorized as “good,” and the “very poor” and “fairly poor” as “poor.”

2.5. Smoking status and alcohol consumption

The question “What is your current smoking status?” was used to elicit responses from among 5 options: never smoked; smoked once, but not since then; smoked quite a bit, but not currently, at least for the past year; smoked occasionally; and smoked daily. As this variable pertained mainly to smoking “frequency,” respondents with the following 2 responses “never smoked” and “smoked once, but not since then” were categorized as “nonsmokers,” whereas those who answered: “smoked quite a bit, but not currently, at least for the past year,” “smoked occasionally,” and “smoked daily” were categorized as “ex-smokers,” “occasional smokers,” and “current smokers,” respectively.

In response to the question “What is your drinking status?” respondents had to choose 1 of the following options: “never drank alcoholic beverages,” “drank once, but not since then,” “used to drink daily, but not for the past year or more,” “drink occasionally,” and “drink on a daily basis.” As this variable was mainly about drinking “frequency” (similar to smoking status), the respondents who answered “never drank alcoholic beverages” and “drank once, but not since then” were categorized as “nondrinkers,” whereas those who answered “used to drink daily, but not for the past year or more,” “drink occasionally,” and “drink on a daily basis” were categorized as “ex-drinkers,” “occasional drinkers,” and “current drinkers,” respectively.

2.6. Statistical analysis

First, descriptive statistics were used to analyze the respondents’ characteristics by gender and, thereafter, a chi-squared test was used to analyze gender differences in the association between absolute presenteeism and each variable. In addition, a binary logistic regression analysis was conducted with the binary outcome of the absolute presenteeism score as the dependent variable and gender, age, family status, subjective sleep quality, smoking status, and alcohol consumption as the independent variables. Associations were examined between absolute presenteeism scores (≤40) and each variable. Nonresponses and those with missing variables about family status, subjective sleep quality, smoking status, and alcohol consumption were classified as missing values and included in the analysis. A gender-stratified binary logistic regression analysis was also performed and has been presented as a subanalysis. In the subanalysis, gender was excluded from the independent variables. All analyses were performed using SPSS Statistics 28.0 for Macintosh (IBM Corp., Armonk, NY, USA).

3. Results

The characteristics of the study’s respondents are shown in Table 1. Of the 713 respondents, 412 (57.8%) were men and 301 (42.2%) were women. Most respondents were either in their 40s and 50s or older. Poor subjective sleep quality was reported by 314 (44.0%) respondents, including 190 (46.1%) men and 124 (41.2%) women. Current smokers were extremely few among female respondents, compared with 105 among their male counterparts. Smokers (occasional as well as current) comprised 117 (28.4%) men and 4 (1.3%) women. Current drinkers were 182 (25.5%), namely 145 (35.2%) men and 37 (12.3%) women. The number and percentage of respondents with presenteeism for each variable are also presented in Table 1. For example, among all respondents aged less than 30 years, 32 (33.0%) respondents reported absolute presenteeism. Respondents aged ≥50 years had a lower absolute presenteeism percentage compared with those of other ages among all respondents. Respondents with poor subjective sleep quality had a larger proportion of absolute presenteeism than those with good subjective sleep quality. Women aged <40 years tended to have more absolute presenteeism than those >40 years. Among men, those with poor subjective sleep quality had a larger proportion of absolute presenteeism than those with good subjective sleep quality.

Table 1.

Numbers and percentages of respondents with presenteeism by age, family status, lifestyle factors, and gender.

	Overall				Men				Women
	n	Number of respondents with WHO-HPQ ≤40	Percentage of respondents with WHO-HPQ ≤40	P values a	n	Number of respondents with WHO-HPQ ≤40	Percentage of respondents with WHO-HPQ ≤40	P values *	n	Number of respondents with WHO-HPQ ≤40	Percentage of respondents with WHO-HPQ ≤40	P values a
Overall	713	174	24.4		412	102	24.8		301	72	23.9
Age, y				.025				.193				.008
<30	97	32	33.0		47	14	29.8		50	18	36.0
30-39	148	44	29.7		72	21	29.2		76	23	30.3
40-49	235	51	21.7		148	40	27.0		87	11	12.6
≥50	233	47	20.2		145	27	18.6		88	20	22.7
Family status				.215				.100				.448
A married couple only	74	16	21.6		46	10	21.7		28	6	21.4
You as a married couple and your child or children	264	49	18.6		168	32	19.0		96	17	17.7
You as a married couple and your parent(s)	24	6	25.0		15	3	20.0		9	3	33.3
Three-generation family	160	48	30.0		92	30	32.6		68	18	26.5
Single	59	18	30.5		30	8	26.7		29	10	34.5
You and your parent(s)	83	24	28.9		43	16	37.2		40	8	20.0
You and your child or children	20	5	25.0		6	2	33.3		14	3	21.4
Other family structures	27	7	25.9		12	1	8.3		15	6	40.0
Missing values	2	1	50.0		0	0	0.0		2	1	50.0
Subjective sleep quality				.009				.012				.453
Good	376	76	20.2		211	40	19.0		165	36	21.8
Poor	314	94	29.9		190	60	31.6		124	34	27.4
Missing values	23	4	17.4		11	2	18.2		12	2	16.7
Smoking status				.816				.677				.411
Nonsmokers	544	130	23.9		249	60	24.1		295	70	23.7
Ex-smokers	47	10	21.3		46	10	21.7		1	0	0.0
Occasional smokers	15	4	26.7		12	2	16.7		3	2	66.7
Current smokers	106	30	28.3		105	30	28.6		1	0	0.0
Missing values	1	0	0.0		0	0	0.0		1	0	0.0
Alcohol consumption				.818				.059				.034
Nondrinkers	132	34	25.8		48	19	39.6		84	15	17.9
Ex-drinkers	31	9	29.0		21	5	23.8		10	4	40.0
Occasional drinkers	366	89	24.3		197	51	25.9		169	38	22.5
Current drinkers	182	41	22.5		145	27	18.6		37	14	37.8
Missing values	2	1	50.0		1	0	0.0		1	1	100.0

^aDifferences in percentage of respondents with presenteeism among each variable by gender were tested using the chi-squared test.

Abbreviation: WHO-HPQ, World Health Organization Health and Work Performance Questionnaire.

The results of the logistic regression analysis are shown in Table 2. Absolute presenteeism was positively associated with poor subjective sleep quality among all respondents (odds ratio [OR], 1.70; 95% CI, 1.18-2.44) and men (OR, 1.85; 95% CI, 1.12-3.05) and with current drinkers among women (OR, 3.49; 95% CI, 1.36-8.92); it was negatively associated with age among those who were ≥50 years old (OR, 0.50; 95% CI, 0.27-0.93), with current drinkers among men (OR, 0.43; 95% CI, 0.20-0.92), and with women aged 40-49 years (OR, 0.24; 95% CI, 0.09-0.66). Among male workers, no difference in presenteeism was observed between nonsmokers and ex-smokers.

Table 2.

Association between lifestyle factors and presenteeism (n = 713).^a

Covariates entered	Absolute presenteeism score <40
	OR (95% CI)
	Overall b  n = 713	Men c  n = 412	Women c  n = 301
Gender (Ref. = Men)
Women	0.95 (0.63-1.44)
Age, y (Ref. = <30)
30–39	1.00 (0.55-1.84)	1.14 (0.46-2.83)	0.87 (0.37-2.05)
40–49	0.59 (0.32-1.07)	0.97 (0.42-2.27)	0.24 (0.09-0.66)
≥50	0.50 (0.27-0.93)	0.59 (0.24-1.48)	0.53 (0.22-1.30)
Family status (Ref. = A married couple only)
You as a married couple and your child or children	0.90 (0.47-1.74)	0.86 (0.37-1.99)	0.93 (0.30-2.86)
You as a married couple and your parent(s)	1.44 (0.47-4.39)	1.04 (0.23-4.69)	3.01 (0.48-18.73)
Three-generation family	1.92 (0.98-3.76)	2.09 (0.88-5.01)	1.87 (0.61-5.76)
Single	1.27 (0.56-2.88)	1.10 (0.35-3.40)	1.85 (0.52-6.61)
You and your parent(s)	1.28 (0.60-2.71)	1.56 (0.59-4.14)	0.88 (0.25-3.12)
You and your child or children	1.18 (0.36-3.89)	1.89 (0.26-13.86)	1.06 (0.21-5.41)
Other family structures	1.36 (0.48-3.89)	0.32 (0.04-2.99)	2.24 (0.54-9.35)
Poor subjective sleep quality (Ref. = Good)	1.70 (1.18-2.44)	1.85 (1.12-3.05)	1.28 (0.70-2.33)
Smoking status (Ref. = Nonsmokers)
Ex-smokers	1.00 (0.46-2.17)	1.03 (0.46-2.30)	—
Occasional smokers	1.31 (0.39-4.40)	0.98 (0.20-4.86)	3.47 (0.27-44.37)
Current smokers	1.39 (0.27-2.36)	1.51 (0.87-2.62)	—
Alcohol consumption (Ref. = Nondrinkers)
Ex-drinkers	1.36 (0.55-3.34)	0.54 (0.16-1.87)	3.24 (0.72-14.56)
Occasional drinkers	0.97 (0.60-1.57)	0.58 (0.29-1.16)	1.17 (0.55-2.47)
Current drinkers	0.99 (0.56-1.75)	0.43 (0.20-0.92)	3.49 (1.36-8.92)

^aThere was no difference in the direction of association between the analyses including and not including missing values for each independent variable. The results of the analysis including missing values are presented.

^bAdjusted for gender, age, family status, subjective sleep quality, smoking status, and alcohol consumption.

^cAdjusted for age, family status, subjective sleep quality, smoking status, and alcohol consumption.

Abbreviations: n, number of respondents included in the analysis on absolute presenteeism; OR, odds ratio; Ref., reference category coded 0. Dash indicates no available value.

4. Discussion

This study examined the association between presenteeism and factors such as age, subjective sleep quality, smoking, and alcohol consumption among Japanese government office workers. The results indicated that absolute presenteeism was positively associated with poor subjective sleep quality among all respondents and men, and with current drinkers among women, whereas absolute presenteeism was negatively associated with age for all respondents and with current drinking for men.

4.1. Demographic variables

Regarding sociodemographic characteristics and presenteeism, presenteeism was negatively associated with age among those aged ≥50 years. This result corresponds with some previous studies that have reported that presenteeism is negatively associated with increasing age.^3,8,24,25 However, in this study, such an association was found among women, but not men. Some Japanese women quit full-time employment after marriage or childbirth. In general, Japanese women’s participation rate in the labor force decreases in their 30s and increases in their 40s; however, such a trend is not observed among Japanese men.²⁶ In fact, this study’s respondents only comprised regular workers. The number of workers in their 40s and over 50 years of age was lower for women than men, indicating a decrease in the number of regular female workers in their 40s. One possible reason for presenteeism being negatively associated with female workers in their 40s was that given women who are regular workers have relatively good support at home for their work, even after their 40s, they may have relatively high work performance. Another reason was menstruation, which might cause presenteeism in younger rather than older female workers. As older female workers are more likely to experience menopause, they are less likely to have menstrual symptoms. A previous Japanese study reported that presenteeism was associated with increasing severity of menstrual symptoms and self-reported heavy bleeding, given that 74.0% of women workers reported having menstrual symptoms.²⁷ Conversely, older workers were those who could have continued working, possibly resulting in the low proportion of older workers with presenteeism.

4.2. Subjective sleep quality

This study found that poor subjective sleep quality was associated with presenteeism among all respondents. This finding is in line with those of previous studies.^24,25,28 According to these results, including those of our study, poor subjective sleep quality is probably associated with presenteeism, and this association was clearer among men in this study. When it comes to presenteeism, men may be more vulnerable to poor sleep quality than women. In fact, such gender differences in the effects of poor sleep quality were also found in other studies. Previous research suggested an inverse association: men with poor sleep quality were more likely to have low physical performance, whereas women with poor sleep quality were less likely to have low physical performance.²⁹ Another study has shown that poor sleep quality is associated with overweight/obesity in men, but not in women,³⁰ and obesity is known to be a risk factor for presenteeism.³¹ Furthermore, it was reported that men had more sympathetic nervous system activity caused by poor sleep quality than women.³²

Gender differences have also been reported with respect to subjective sleep quality ratings. Morris et al³³ reported that men were more likely to equate subjective sleep quality with sleep duration and efficiency, whereas women may regard it as a sleep disturbance or daytime dysfunction. van den Berg et al³⁴ stated that although sleep, when actigraphically measured, was better in women than in men, women reported shorter and poorer sleep than men. These gender differences may contribute to gender differences in the association between presenteeism and subjective sleep quality.

Finally, the association between presenteeism and poor subjective sleep quality could have been influenced by confounders, such as stress, depression, and working hours, which may have been the cause of gender differences in this association. Regarding reverse causality, subjective sleep quality might be impaired by presenteeism.

4.3. Smoking status

This study’s result, that being a current smoker was not associated with presenteeism among all respondents, and also among men and women separately, differs from those of some previous studies.^8,9 However, other previous studies did not find a positive association between smoking and presenteeism.^35,36 The explanation of Robroek et al⁸ is useful to understand these differences in association; they suggested they were caused by differences between the focus of presenteeism: degree or presence. If the focus was on the degree of presenteeism, a positive association with smoking would be found, but if the focus was on its presence, as in this study, no positive association with smoking would be found.

As the proportion of women smokers (1.3%) in this study, was much lower than that in the general population (7.6%),³⁷ the association between presenteeism and smoking status could not be assessed. Point estimates of frequency show a higher percentage of presenteeism among women who were occasional smokers than nonsmokers. Therefore, a positive association between smoking and presenteeism might have been found if a larger number of smokers had been examined. Although the percentage of men smokers in this study was 28.4%, almost the same as the percentage of Japanese male smokers generally, the male respondents showed no significant association between smoking status and presenteeism. A previous study explained that the association between smoking and presenteeism may be due to withdrawal symptoms ,¹⁰ suggesting that the disadvantage is less likely to appear in this study’s setting.

Conversely, the current study found no differences between nonsmokers’ and ex-smokers’ association with presenteeism. This result is supported by a previous Japanese study that reported no differences on presenteeism between persons who had never smoked and former smokers.⁹

4.4. Alcohol consumption

In this study, no associations between alcohol consumption and presenteeism among all respondents were identified. Absolute presenteeism was positively associated with alcohol consumption in current drinkers among women, whereas it was negatively associated with alcohol consumption in current drinkers among men. As a result, an overall association between presenteeism and alcohol consumption was not detected.

Women are more vulnerable to the negative effects of alcohol and develop alcohol-related health problems at shorter durations and through lower consumption compared with men.³⁸ Therefore, female current drinkers may be more likely to be associated with presenteeism. Furthermore, as women perceive greater social sanctions for drinking,³⁸ to hide their drinking habits they may be more likely to come to work despite having a “hangover” after a night of drinking. Additionally, women might have the tendency to underreport their drinking or presenteeism. Furthermore, it is possible that women with presenteeism consume alcohol to relieve their negative feelings.

By contrast, the results were inconsistent for men respondents with alcohol-related presenteeism. A previous study indicated that alcohol-related presenteeism was more associated with men than women,¹³ which may be because risky drinking is more common in men than women and binge drinking is associated with alcohol-related presenteeism.^13,39,40 However, the men in this study showed the opposite result of alcohol-related presenteeism. Additionally, heavier drinkers may be absent from work due to ill health caused by alcohol.

4.5. Limitations

This study has some limitations. First, its data were based entirely on self-reports, making it susceptible to recall and social desirability biases. For presenteeism measured using the WHO-HPQ, these biases were considered to have a small effect because calibration studies have shown strong concordance with the WHO-HPQ, archival records, and ratings by supervisors and peers.^19,20 Second, because of self-reporting, there were some missing values. Given that no differences in the direction of association of presenteeism between the analyses were found, the impact of missing values is suggested to be small. We compared respondents with deficient absolute presenteeism scores with respondents without deficient absolute presenteeism scores in terms of family structure, sleep quality, smoking status, and alcohol consumption to assess the impact of missing data. No differences were found, but the proportion of deficient lifestyle factors was greater among respondents with deficient absolute presenteeism scores. By contrast, the absolute presenteeism scores among respondents with deficient absolute presenteeism scores cannot be discussed. Third, generalizability is limited, as respondents were office workers in a local city government office in Japan. Hence, for the findings in this study to be directly applicable to other occupations, there is a need for further analyses. However, at least for Japanese local government office workers, the findings might be generalizable because the style of working is similar among local government office workers. Fourth, as this study did not assess several factors that might have been associated with presenteeism, those factors could have influenced the associations that were identified in this study. For example, as mentioned above, stress, depression, and working hours might confound the association between presenteeism and poor subjective sleep quality. Moreover, as this study’s respondents’ data were obtained from City Government Office A, the factors that could be included in this study and the sample size were limited. Finally, being a cross-sectional study, beyond a point it was unable to demonstrate causality. The risk of reverse causality must be considered when interpreting the results. Therefore, it is necessary to follow up with a longitudinal study to prove causality.

4.6. Conclusion

This study examined the association between presenteeism and factors such as age, subjective sleep quality, smoking, and alcohol consumption among Japanese government office workers. The results indicated that absolute presenteeism was positively associated with poor subjective sleep quality among all respondents and men, and with current drinkers among women, whereas absolute presenteeism was negatively associated with age for all respondents and with current drinking for men. The direction of the association between lifestyle factors and presenteeism may differ by gender among Japanese city government office workers; therefore, different methods, by gender, are needed for improving productivity related to subjective sleep quality or alcohol consumption. This study may be useful for devising measures to improve labor productivity and prevent mental health problems in administrative workplaces, based on gender differences. However, as various instruments were used to measure presenteeism and lifestyle factors across studies, it is difficult to compare the results of the various studies, including the current study. A gold standard of instruments is needed to establish the association between presenteeism and lifestyle factors. Moreover, future research would have the advantage of adopting longitudinal methods to elucidate the causality of presenteeism by gender.

Author contributions

A.K. and Y.O. conceptualized the study. Y.O. formulated its methodology and supervised its statistical analysis through SPSS. T.O. and H.K. validated the results. T.O. and Y.K. took care of the formal analysis. A.K. investigated and curated the data. T.O. was a major contributor to writing the manuscript and preparing its original draft. All the authors read and approved the final manuscript.

Funding

This study was funded by the Ministry of Health, Labour and Welfare of Japan as part of a project on addiction. However, the funding body did not have any role in the study’s design; its collection, analysis, and interpretation of data; or in writing the manuscript.

Conflicts of interest

The authors declare no conflict of interest for this article.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.