The Impact of Productivity Loss From Presenteeism and Absenteeism on Mental Health in Japan

Hidden costs of mental health issues, such as presenteeism and absenteeism, are significantly high. Therefore, preventive measures, screening, and early intervention for individuals with mental health problems are not only essential for their well-being but also hold great significance for society as a whole. Actively implementing these measures is crucial.

Abstract

Objective

Mental health issues among employees cause significant productivity losses through presenteeism and absenteeism. This study aimed to quantify productivity losses caused by employees with mental health issues in Japan.

Methods

Participants were recruited to match the Japanese population distribution by gender, age, and region. Mental health status and productivity loss were assessed using self-administered questionnaires. The results were extrapolated to estimate nationwide impact calculated using probabilistic sensitivity analysis.

Results

We analyzed 27,507 individuals. Productivity loss due to mental health–related presenteeism was estimated at $46.73 billion, and absenteeism at $1.85 billion, equivalent to 1.1% of Japan’s GDP and over seven times the medical costs for mental disorders. Women in their 20s reported more mental health issues than men.

Conclusions

These results highlight the urgent need for businesses and governments to enhance workplace mental health measures.

CME Learning Objectives

• After completing this enduring educational activity, the learner will be better able to:

• Quantify the national economic impact of productivity loss due to presenteeism and absenteeism associated withmental health issues in Japan, utilizing large-scale survey data and probabilistic sensitivity analysis.

• Identify demographic trends, including higher mental health–related productivity losses among young women, to inform targeted workplace interventions.

• Evaluate the necessity of workplace mental health initiatives by comparing productivity losses to national economic indicators, emphasizing the need for preventive strategies at both corporate and policy levels.

Mental health issues among employees have become a significant concern, particularly in developed countries. According to the World Health Organization, in 2019, 970 million people (one in eight of the global population) were reported to be living with mental disorders.¹ In the United States, the widespread prevalence of burnout has been highlighted,² and similarly in Germany, modern society has been described as a “fatigue society.”³ In Japan, according to a survey by the Ministry of Health, Labour and Welfare, approximately 80% of workers reported experiencing significant anxiety, concerns, or stress related to their work.⁴ To prevent mental health issues among workers, a stress-check program was introduced in 2015 under the Industrial Safety and Health Act. The prevention of mental health problems has been recognized as a critical issue and various efforts are being made to address it.⁵

Corporate measures to prevent mental health issues include not only implementing stress checks but also conducting training sessions, improving work environments, promoting work-life balance, and establishing consultation services.⁶ However, the implementation of mental health measures has not progressed sufficiently because of the associated costs and need for specialized knowledge. In practice, only 65% of workplaces in Japan have implemented mental health measures, with smaller enterprises exhibiting lower implementation rates.⁴ Although numerous studies have demonstrated the cost effectiveness of workplace mental health interventions,^7,8 their implementation remains insufficient.

The limited progress in implementing mental health measures within corporations and government institutions may stem from inadequate assessments of the impact of mental health issues. When employees take leave of absence, resign, or incur medical expenses through healthcare consultations, these effects are rendered visible. However, presenteeism and absenteeism are often referred to as “hidden costs,” making their impact difficult to quantify. Notably, presenteeism has been reported to generate losses several times greater than medical expenses, representing a substantial and often overlooked financial burden.^9–12 Hemp highlights that hidden costs such as presenteeism and absenteeism often exceed medical expenses, with presenteeism costing 2.6 times more and absenteeism about one-fourth the cost of medical and pharmaceutical expenses.¹³

Several studies have estimated the economic losses associated with mental health issues. Evans-Lacko and Knapp estimated the absenteeism and presenteeism costs of depression across eight countries. The prevalence of depression was 3.33% in Japan, 9.66% in the US, and 8.28% in Canada, with presenteeism costs at $8.2 billion, $84.6 billion, and $6.81 billion, respectively.¹⁴ Many individuals with mental health issues such as depressive symptoms do not seek medical care.^15–17 Therefore, estimating presenteeism costs based solely on those diagnosed with depression may lead to significant underestimation.¹⁸ Therefore, it is essential to estimate productivity losses among individuals with mental health issues, regardless of whether they have been formally diagnosed by a physician. This study thus aimed to estimate presenteeism and absenteeism losses due to mental health issues in Japan.

METHODS

Data Resource

In this study, the following data were used based on a large-scale Internet survey conducted in 2022. This survey targeted workers aged 20 years and older, stratified by gender, age group, and region. It received 27,693 valid responses. An explanatory document was presented, and a checkbox was provided to obtain confirmation of consent. Consent was obtained from all respondents. After excluding 186 respondents aged 75 years and older, data from 27,507 individuals were analyzed. Of these, 15,091 were men (54.9%) and 12,416 were women (45.1%), with a mean age of 45 years. These data are part of a prospective cohort study on occupational health and safety in Japan.¹⁹ The research protocol followed the Checklist for Reporting Results of Internet E-Surveys to ensure survey quality.²⁰ Sampling was stratified by gender, age, and region, targeting workers aged 20 years or older and yielded proportions for employment status and industry that closely matched government-led representative data.

Additional data by gender and age group were sourced from the latest government statistics. Population data were obtained from the Population Estimates of the Ministry of Health, Labour and Welfare (2023).²¹ Labor force participation rates by gender and age group were sourced from the Basic Survey on the Labor Force of the Ministry of Internal Affairs and Communications (2023).²² Average daily wages by gender and age group were drawn from the Basic Survey on Wage Structure of the Ministry of Health, Labour and Welfare.²³

Identification of Individuals With Mental Health Issues

In the survey, participants were asked to select the health issues that most impacted their work from a list of options, including health impairments caused by allergies, pain-related issues, mental health issues such as depressive symptoms (eg, low mood), anxiety, sleep disturbances (eg, difficulty falling asleep), general fatigue or exhaustion, eye-related issues such as vision problems, eye strain, dry eyes, glaucoma, and other health issues, or an option indicating that none of the above significantly affected their work. Individuals who selected mental health issues or sleep disturbances were identified as experiencing mental health challenges. Sleep disturbances were included in this category because they are known to have a high comorbidity rate with mental health disorders.^24,25 In addition, several studies have shown that sleep disturbances can predict future depressive symptoms. For instance, poor sleep quality and sleep disturbances in adolescence were linked to increased depression scores over time.^26,27 Similarly, poor sleep quality has been associated with higher odds of developing depressive symptoms in older adults.^28,29

Presenteeism

For presenteeism, participants who selected “mental health issues” or “sleep-related issues” for the question whether they experienced “Health problems or issues over the past month” were included in the analysis. Presenteeism was calculated based on the responses to the quantity and quality method, which has been widely validated and used in previous studies.^9,11,30,31 Specifically, participants were asked to rate—on an 11-point scale from 0 to 10—the extent to which their “work quantity” was affected during periods of symptoms compared to when they were symptom-free (normal conditions). Similarly, they were asked to rate their “work quality” on the same 11-point scale. Work quality was explained with examples such as the frequency of errors and the ability to demonstrate creativity, focusing on the quality of work outcomes. Participants were then asked how many days in the past 30 days they had experienced such symptoms. Finally, annual presenteeism loss days were estimated using the following formula:

$$\mathit{\text{Annual Presenteeism Loss days}}=\mathit{\text{Quantity}}\times \mathit{\text{Quality}}\times \mathit{\text{Days with Symptoms}}\mathit{per}\mathit{\text{month}}\times 12$$

This formula accounts for the impact of symptom days on work performance over 1 year.

Absenteeism

For absenteeism, participants were asked to report the number of days they had been absent from work due to illness over the past 12 months. However, the specific illnesses causing the absences were not identified. Therefore, the analysis was restricted to participants who reported “mental health issues (eg, depressive symptoms such as low mood or feelings of anxiety)” or “sleep-related issues (eg, inability to sleep despite trying).”

Participants reported their days of absence using the following five categories: “none (0 days),” “1–less than 5 days,” “5–less than 10 days,” “10–less than 15 days,” and “15 days or more.” To enable the calculations, the midpoints of each category were defined as follows: 0, 2.5, 7.5, 12.5, and 15, respectively.

Estimated Nationwide Productivity Loss Due to Presenteeism and Absenteeism

The nationwide productivity loss due to presenteeism and absenteeism resulting in mental health issues were estimated. The formula for calculating presenteeism loss was as follows:

$$\mathit{PL}=\sum _{i=1}^n\left(N_i\times L_i\times Y_i\times S_i\times P_i\times \left(H_i\times 12\right)\right)$$

where i represents each subgroup stratified by gender and age, and PL represents presenteeism loss. N is the population, L is the labor force participation rate, Y is the prevalence of individuals with symptoms, S is the average daily income, and P is the presenteeism rate. Additionally, H represents the number of symptomatic days per month, which was extended to 12 months to calculate the total annual loss.

The formula for calculating absenteeism loss is as follows:

$$\mathit{AL}=\sum _{i=1}^n\left(N_i\times L_i\times Y_i\times S_i\times A_i\right)$$

where i represents each subgroup stratified by gender and age, and AL represents the losses due to absenteeism. N is the population, L is the labor force participation rate, Y is the prevalence of individuals with symptoms, S is the average daily income, and A is absenteeism (annual number of days absent).

Probabilistic Sensitivity Analysis

Probabilistic sensitivity analysis (PSA) was conducted to account for the uncertainty in the estimation based on the collected data. The distributions used in the analysis were determined based on previous studies and the characteristics of the data.³² Specifically, the population data, labor participation rates, and wages were assumed to follow a deterministic distribution, the prevalence of individuals with symptoms followed a beta distribution, and presenteeism and absenteeism followed normal distributions. Monte Carlo simulations were performed with 10,000 iterations to calculate 95% confidence intervals (CIs). All analyses were conducted using R (ver. 4.2.3; R Foundation for Statistical Computing, Vienna, Austria) software.

In this study, PSA was initially performed separately for each gender and age group. However, this resulted in high variability owing to the small sample size within each subgroup. To reduce this variability and ensure consistency, the total value was recalculated by aggregating all variables across subgroups, and PSA was performed at the aggregate level. Additionally, the total mean value was fixed to align with the sum of the subgroup results, thereby maintaining internal consistency within the analysis. This study was approved by the ethics committee of the University of Occupational and Environmental Health, Japan (R3-076).

Strengthening the Reporting of Observational Studies in Epidemiology Guidelines

The reporting of this study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines (Supplementary Digital Content, Supplementary Data 2, http://links.lww.com/JOM/B943).

RESULTS

The number of participants with mental health issues was 778 men and 765 women, totaling 1543 individuals. The prevalence of individuals with mental health issues (those who reported “mental health issues or sleep-related issues”) by gender and age group is shown in Figure 1. The prevalence of individuals with symptoms was the highest among women aged 25–29 years (9.5%) and among men aged 30–34 years (8.1%). Women in their 20s had an approximately 1.7 times higher prevalence than men in the same age group.

FIGURE 1.

Prevalence of mental health–related symptoms by gender and age group.

Figure 2 shows the annual number of presenteeism days calculated by gender and age group. Among women, the highest number was observed in the 30–34 year age group (equivalent to 132.4 days per year), whereas among men, it was highest in the 40–44 year age group (145.2 days per year). Although the prevalence of mental health issues peaked among the participants in their 20s and 30s and subsequently declined, presenteeism remained high until their early 50s.

FIGURE 2.

Annual presenteeism loss (days) by gender and age group.

Figure 3 shows the annual absenteeism loss (days) by gender and age group. Among women, the highest loss was observed in the 20–24 year age group (5.4 days per year), whereas among men, it was the highest in the 40–44 year age group (6.3 days per year).

FIGURE 3.

Annual absenteeism loss (days) by gender and age group.

By combining the findings so far with gender- and age-specific data on population, labor force participation rates, and average daily income (see Appendix A., http://links.lww.com/JOM/B944), presenteeism and absenteeism losses were calculated (Table 1). For presenteeism loss, men aged 45–49 years had the highest value ($6,214,728,000), while for absenteeism loss, men aged 40–44 years had the highest value ($233,855,000). Overall, the total presenteeism loss was $46,738,761,000, and the total absenteeism loss was $1,855,609,000.

TABLE 1.

Annual Monetary Value of Productivity Loss From Presenteeism and Absenteeism Due to Mental Health Issues in Japan

	Age Group (Years)	Presenteeism Loss (USD)		Absenteeism Loss (USD)
Gender		Estimate	95% CI	Estimate	95% CI
Male	20–24	647,661	415,253–922,893	18,578	5,361–34,776
	25–29	2,017,587	1,663,792–2,411,437	67,389	47,334–90,289
	30–34	3,670,279	2,977,672–4,440,752	161,614	117,083–212,425
	35–39	3,748,890	3,130,561–4,414,076	132,875	97,288–172,066
	40–44	5,370,401	4,438,593–6,390,915	233,855	176,215–299,561
	45–49	6,214,728	5,240,860–7,269,739	211,043	157,656–270,897
	50–54	3,849,622	3,050,470–4,729,167	168,208	116,738–227,213
	55–59	2,308,657	1,759,556–2,916,602	134,031	87,233–189,891
	60–64	637,926	449,525–856,020	45,596	24,735–71,094
	65–74	443,095	269,717–660,758	15,235	5,290–28,355
Female	20–24	1,657,319	1,209,770–2,167,709	73,136	43,671–109,317
	25–29	2,603,888	2,242,779–2,995,379	94,287	72,425–117,671
	30–34	2,352,645	1,901,300–2,837,587	92,855	66,374–122,040
	35–39	2,405,093	1,950,762–2,897,800	80,767	55,206–110,485
	40–44	2,211,208	1,786,924–2,672,649	84,422	58,439–114,905
	45–49	2,242,182	1,830,750–2,694,058	72,399	49,513–98,297
	50–54	2,132,055	1,697,134–2,614,474	83,647	56,805–114,648
	55–59	1,567,250	1,201,765–1,986,261	54,491	32,649–80,149
	60–64	487,848	335,175–672,667	19,845	7,895–34,406
	65–74	170,427	102,956–254,804	11,336	3,722–21,762
Total		46,738,761	44,809,092–49,350,694	1,855,609	1,705,207–1,996,782

Note: All values in 1,000 USD units.

All values are converted using an exchange rate of 1 USD = 157 JPY.

DISCUSSION

In Japan, the productivity loss due to mental health–related presenteeism was estimated at $46.73 billion (95% CI: $44.80 billion–$49.30 billion), while absenteeism accounted for $1.85 billion (95% CI: $1.70 billion–$1.99 billion). The total productivity loss due to mental health issues corresponds to 1.11% of Japan’s GDP ($4.38 trillion). The medical expenses for mental and behavioral disorders among individuals under 65 years of age amount to approximately $7 billion,³³ whereas the productivity loss due to presenteeism was approximately seven times higher. Previous estimates of productivity loss in Japan due to mental health issues include $8.2 billion for general mental health disorders,¹⁴ $31.7 billion for mental illnesses,⁹ $21.5 billion for sleep-related problems,⁹ and approximately $20.3 billion for depression and anxiety disorders.³⁴ Because of its larger sample size and distribution, which closely mirror Japan’s population structure, this estimation offers superior representativeness, rendering its findings highly generalizable. This study highlighted a greater productivity loss than previous estimates, further underscoring the importance of mental health measures by businesses and governments. The differing results of this study compared to previous research can be attributed to the inclusion of individuals with mental health symptoms regardless of whether they had a formal diagnosis; larger sample size; greater diversity in terms of sex, age, and occupation; and the potential influence of conducting the survey after the COVID-19 pandemic. The shift in work practices triggered by the COVID-19 pandemic, including the rise in remote work and decline in social interactions, has been reported to contribute to increased isolation and stress, leading to a surge in mental health issues.^35,36 The estimation results of this study are believed to more accurately reflect the actual situation in Japan.

The present findings revealed that the prevalence of presenteeism was 25.2 times greater than that of absenteeism, which is a substantial difference. According to Evans-Lacko and Knapp, the ratio of presenteeism to absenteeism associated with depression across eight countries typically ranges from 5 to 10, making the disparity observed in this study significantly larger.¹⁴ A potential reason for this discrepancy is that the participants were identified based on the presence of symptoms, irrespective of whether they had been formally diagnosed with depression. Additionally, the Japanese workplace culture makes it challenging to take time off work, which may result in employees continuing to work despite experiencing symptoms, thereby contributing to higher levels of presenteeism.³⁷ Engaging in rest and self-care during periods of mental health challenge can help maintain a stable mental state without compromising on productivity. Promoting and integrating self-care practices into Japanese workplace culture is essential to foster employee well-being and sustain performance.

Overall, productivity loss tended to be higher among men than among women. However, no gender differences were observed in the prevalence of mental health–related symptoms, presenteeism, or absenteeism. In fact, the prevalence of mental health–related symptoms was higher among the female participants in their 20s than among their male counterparts. This suggests that the gender gap in productivity loss is influenced by lower labor participation rates and lower wages among women.³⁸ If women’s labor force participation rate and average daily income were equal to those of men, the estimated productivity loss due to presenteeism would be $59.95 billion, and $2.34 billion due to absenteeism. In fact, women’s labor force participation in Japan has been increasing annually.³⁹ As Japan strives to achieve a society with equality between men and women, implementing measures that address productivity losses caused by mental health issues regardless of gender is essential.

Deady et al proposed an updated framework for mentally healthy workplaces, focusing on three pillars: protect, promote, and respond.⁶ It emphasizes preventing harm, enhancing wellbeing, and supporting recovery through multilevel strategies tailored to organizational needs. This evidence-based model aims to create safer, more supportive, and thriving work environments. In Japan, while “protect” measures such as stress checks and antiharassment laws were recently established,^5,40 “promote” and “respond” efforts remain underdeveloped, requiring greater focus on well-being and recovery support. Factors such as a desire to avoid burdening colleagues, strong occupational responsibility, inability to refuse tasks, low income, ignoring symptoms, or difficulty staying at home, combined with workplace challenges such as insufficient staffing, restrictive leave policies, weak leadership, and job insecurity, contribute to individuals continuing to work despite illness.³⁷ To prevent productivity losses caused by mental health issues, it is essential to enhance workplace systems, ensure the availability of staff substitutes, provide stable employment, implement effective management, promote health initiatives, and facilitate treatment through medical institutions.³⁷

Japan’s Ministry of Economy, Trade and Industry promotes Health and Productivity Management and certifies outstanding companies through its recognition program for exemplary employee health initiatives.⁴¹ This is in line with international discussions on the shareholder economic value of health-related productivity loss, as emphasized by Loeppke and Hymel (2023), who highlighted a case study in a publicly traded US company and discussed the economic returns of health and productivity management initiatives, including those in Japan.⁴² It is reported that every $1 invested in expanding treatment for depression and anxiety yields a $4 return through improved health and productivity.⁴³ As demonstrated by the current estimates, the impact of productivity loss due to mental health issues is profound. It is crucial to enhance diverse services tailored to individuals with various mental health challenges, while ensuring that these services are scientifically evaluated for their effectiveness. For example, Nakao et al conducted a 2-year cohort study on depression levels and suicide attempts in relation to employee assistance programs, highlighting the importance of such scientific evaluations.⁴⁴

Despite our robust and rigorous analyses, this study has several limitations. First, the estimation of productivity loss was based on subjective evaluations from questionnaires, leaving the actual monetary loss unclear. Although the quantity and quality method used in this study is a validated measure widely employed globally to assess productivity loss, further research is needed to verify the extent of actual losses. Second, the estimation of absenteeism was based on the absences reported by individuals with mental health issues; however, it was not explicitly determined whether these absences were directly caused by mental health problems. This may have led to an overestimation of absenteeism. Conversely, it is possible that individuals who did not report mental health issues also experienced absences owing to such problems, which were not accounted for in this estimation, potentially resulting in an underestimation. Additionally, because the survey targeted individuals who were currently employed, those on long-term leave at the time of data collection were excluded. As a result, absenteeism may have been somewhat underestimated. Future studies should aim to obtain more precise measurements of absenteeism attributable to mental health issues.

CONCLUSIONS

This study reveals a substantial economic impact of mental health issues on productivity loss in Japan, with presenteeism accounting for $46.73 billion and absenteeism for $1.85 billion annually, corresponding to 1.11% of the nation’s GDP. These findings underscore the critical need for the prioritization of mental health measures by businesses and governments. Productivity loss due to presenteeism is particularly significant, being approximately seven times higher than mental health–related medical expenses, emphasizing the need for proactive workplace interventions. To mitigate these economic and social impacts, targeted mental health strategies including early intervention, workplace support systems, and comprehensive policy measures are essential to improve employee well-being and reduce productivity losses on a national scale.