Article Text
PDF
Abstract
Objectives Overwork-related cerebrovascular and cardiovascular diseases (CCVDs) constitute a major occupational and public health issue worldwide. However, to our knowledge, few studies have reported the underlying pathophysiological mechanisms. We aimed to determine whether patients with extreme workload have a greater risk of developing hypertensive intracerebral haemorrhage (ICH) located in the deep brain areas than patients without extreme workload. We also determined the association between the number of hours of overtime work and the risk of developing hypertensive ICH.
Design Unmatched case–control study.
Setting Database of patients claiming compensation for overwork-related CCVDs in Japan.
Participants A total of 1215 patients who claimed overwork-related ICH in Japan, of whom 621 had their compensation claim approved (patients with extreme workload) and 622 did not.
Primary and secondary outcome measures Logistic regression analysis was performed to calculate the risk of developing hypertensive ICH in patients with extreme workload compared with those without extreme workload, adjusted for confounders such as age, occupation, smoking status, alcohol consumption and medical history. We also calculated the risk of developing hypertensive ICH in compensated patients by average monthly overtime working hours.
Results Patients with extreme workloads had a significantly higher odds ratio (OR) for developing hypertensive ICH (1.44, 95% CI: 1.10 to 1.88) than those without extreme workloads. ORs for developing hypertensive ICH according to overtime working hours showed a dose−response relationship; an overtime of 100 hours/month was associated with a significantly higher OR (1.31, 95% CI: 0.89 to 1.91; 1.41, 95% CI: 0.95 to 2.11; and 1.50, 95% CI: 1.01 to 2.22 for 60–79.9, 80–99.9 and≥100 hours/month, respectively) than that for workloads of less than 60 hours/month.
Conclusions Regarding Japanese workers, overtime work could be associated with the risk of developing hypertensive ICH, and hypertension may play an important role in overwork-related ICH.
- stroke
- case-control studies
- occupational & industrial medicine
- hypertension
- neurosurgery
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The Research Center for Overwork-Related Disorders collected information on all Japanese patients with claims for overwork-related cerebrovascular and cardiovascular diseases (CCVDs) between 2010 and 2017.
For data collection, the Labor Standards Inspection Office objectively evaluated the diagnosis and pathology based on the records maintained by the physician in charge and investigated the working hours and work conditions at the company.
One major strength of this study is that we collected data on all work-related CCVD claims in Japan and objectively gathered information on work hours and disease diagnoses.
The main limitation relates to the fact that the study participants developed cerebral haemorrhage and claimed compensation for overwork-related CCVDs, and this prevented us from determining the absolute incident risk of developing hypertensive intracerebral haemorrhage by extreme workload.
Introduction
Overwork-related cerebrovascular and cardiovascular diseases (CCVDs) constitute a major occupational and public health issue worldwide, particularly in East Asian countries, including Japan.1 2 The number of claims and compensations provided for overwork-related CCVDs remains high (approximately 800 claims are made per year, of which 200–300 are approved for compensation).2 The Research Center for Overwork-Related Disorders (RECORDS), established at the National Institute of Occupational Safety and Health, Japan, in 2014, is responsible for conducting research aimed at understanding the conditions leading to overwork-related deaths and disorders. This centre retrospectively collected the information of all Japanese patients with claims for overwork-related CCVDs between 2010 and 2017. Then, the centre created a database of the information collected and has been analysing this data.2 The centre also updates the database each fiscal year for all compensated patients as soon as possible, followed by creating a database for uncompensated patients.
Previous studies have demonstrated a relationship between overwork and CCVDs.3–6 However, the detailed pathology, such as the site of intracerebral haemorrhage (ICH) or subtype of cerebral infarction, which is related to overtime work, is still unknown.7 We focused on ICH, which is the most common overwork-related CCVD in Japan.2 Although hypertension is a known risk factor for both lobar and non-lobar ICH, it has a greater role in non-lobar (especially deep) ICH.8–11 Determining the site of ICH in patients with extreme workloads may help to confirm the pathophysiological mechanism underlying overwork-associated ICH.
We hypothesised that patients with extreme workloads would have a greater risk of developing deep ICH than patients without extreme workloads. Moreover, we assessed the associative risk of developing deep ICH with the number of hours of overtime work. Thus, we conducted an unmatched case–control study using a database of patients with claims for overwork-related CCVDs and tested this hypothesis.
Methods
Database
The RECORDS has created an anonymous database of 3988 patients with claims for overwork-related CCVDs in Japan, including 2027 patients from 2010 to 2017 whose claims were approved (compensated patients) and 1961 patients from 2010 to 2015 whose claims were rejected (uncompensated patients). It includes data about the patients’ diagnoses, radiological findings (eg, CT and MRI), medical history (eg, hypertension, diabetes and hyperlipidaemia), age at ICH onset, number of hours of overtime work in 1–6 months before onset, occupation (based on the Japan Standard Occupational Classification), smoking status (current smoking, past smoking, never smoking, unknown (missing data)) and alcohol consumption (g/day).
The criteria of compensation for extreme workload in Japan
From 2001 to 2021, workers with overwork-related CCVDs in Japan received compensation if they met the following criteria12: (1) they developed CCVDs, such as ischaemic cerebral infarction, ICH, subarachnoid haemorrhage, myocardial infarction, angina, cardiac arrest, aortic dissection or hypertensive encephalopathy and (2) they had an excessive workload. Excessive workload was divided into three categories: (a) abnormal event, defined as an event that induced extreme tension, excitement, fear or wonder or an accidental or unpredictable incident that urgently places a physical burden on the worker or causes a rapid and significant change in the work environment; (b) short-term excessive heavy work, defined as a clearly excessive workload that imposes a very significant physical or mental burden (compared with that of normal work) taken on within approximately 1 week prior to the appearance of a disease; and (c) long-term excessive heavy work, defined as excessive heavy work performed for a long period. A strong correlation between work and the appearance of a disease can be assumed if the worker has worked more than 100 hours of overtime in the month prior to the appearance of the disease or more than 80 hours per month of overtime in 2–6 months prior to the appearance of the disease. Factors other than working hours contributing to an excessive work burden should also be adequately considered. These include an irregular work pattern, frequent business trips, shifts, late-night work, work with long compulsory attendance hours, the working environment (in terms of temperature, noise level and time differences with the worker’s home country) and mentally stressful work. When a worker claims compensation, the Labor Standards Inspection Office objectively evaluates the diagnosis and pathology based on the records maintained by the physician in charge and investigates the working hours and work conditions at the company to check whether the compensation criteria are met (criteria (1) and (2): a, b or c).
Participants
To avoid the effect of selection bias associated with claiming compensation, we compared the site-specific risk of ICH in compensated patients (exposed with extreme workloads) with that in uncompensated patients (exposed without extreme workloads) as opposed to comparing to patients in the general population. We analysed 1306 patients with ICH from the database and excluded 44 compensated patients with abnormal events (n=18) and short-term overload categories of excessive work (n=26) because we focused on the relationship between long-term overwork and ICH. We also excluded recurrent ICH patients (with a medical history of ICH; n=15) and patients with missing data of medical history (n=32) of ICH because recurrent ICH may have a different pathophysiological mechanism of onset from that of the initial ICH. Finally, 1215 (621 compensated and 594 uncompensated) patients were analysed (figure 1).
Flowchart of the participant selection process. CCVDs, cerebrovascular and cardiovascular diseases; ICH, intracerebral haemorrhage.
RECORDS does not have information regarding patient race, and as we were unable to directly handle information that includes patient names, we requested RECORDS staff members who can handle that information to check the percentage of Japanese individuals among the patients with claims for overwork-related CCVDs. According to their estimates based on patient names, the majority of them (at least 95%) were believed to be Japanese.
who had passed away) were informed of the study goals and allowed to opt-out if they did not want their information to be used for research purposes via the official website of the Ministry of Health, Labour and Welfare or the National Institute of Occupational Safety and Health, Japan, although none of them opted out of participation. The study protocol, including opt-out, was approved by the ethics committee of the National Institute of Occupational Safety and Health in Japan (approval number: 2021N26).
Participant and public involvement
No participants were involved in deciding the investigation or outcome measures for the present study, nor were they asked to advise on the interpretation or the writing up of this study.
Analysis
All statistical analyses were performed with SPSS software for Windows (V.13.0). We categorised ICH according to the site into the following categories: lobar, deep (putamen, thalamus and basal ganglia), infratentorial (brain stem and cerebellum), unknown (no available information regarding the site of ICH) and other (49 patients: 16 malformations, 5 tumours and 3 intraventricular haemorrhages in compensated patients; 23 malformations, 1 tumour and 1 cerebral venous sinus thrombosis in uncompensated patients).8 9 We defined the patients exposed to extreme workload as compensated patients for overwork-related CCVDs and those who were not exposed as uncompensated patients. We categorised the mean overtime working hours in the 6 months before ICH onset into four categories: <60, 60–79.9, 80–99.9 and ≥100 hours/month1 and ‘incomplete data’ (incomplete data on the number of hours of overtime work in the 6 months before ICH onset). We also categorised alcohol consumption into five categories: non-drinker, <19, 20–39, ≥40 g/day and unknown (missing data).5 13 The year of onset range was 2003–2017, because some patients developed ICH before the claim year. Thus, we divided the year of onset into two categories—before 2010 and after 2010—based on the median year of the range, that is, 2010.
We compared the characteristics of patients according to the site of ICH using the χ2 test and residual analysis. Multivariate logistic regression models were computed to estimate the site-specific risk of ICH in patients with extreme overload.
We used four models. In model 1, the dependent variable was the occurrence of ICH (yes or no) at each site (deep, lobar, infratentorial, unknown and other), and the independent variables were the exposure to extreme overload and covariates such as the age of onset, sex, occupation, medical history, alcohol consumption, smoking status and year of onset category. In model 2, we added medical history of hypertension, diabetes and hyperlipidaemia as additional covariates to model 1. In model 3, the site of ICH was the dependent variable, and the above-mentioned covariates and overtime hours (<60, 60–79.9, 80–99.9, ≥100 hours/month in the 6 months before onset and incomplete data) were the independent variables. In model 4, we added medical history of hypertension, diabetes and hyperlipidaemia as additional covariates to model 3.
For sensitivity analysis, we repeated the same analysis (models 1–4) after excluding participants with an unknown site of ICH or excluding participants with claims for 2016–2017.
Results
Characteristics of patients with claims according to ICH location
The most frequent stroke type was deep ICH in all age groups above 30 years. Older patients (≥60 years) tended to develop deep ICH rather than ‘other’. Deep ICH patients had a significantly lower rate of ‘current smoker’, whereas infratentorial ICH patients had a significantly higher rate of current smoker. Regarding medical history, deep ICH patients tended to have a medical history of hypertension, whereas patients in the ‘other’ category tended to have less hypertension. Moreover, deep ICH patients were significantly exposed to extreme overload (compensated). In contrast, lobar ICH patients were significantly less exposed to extreme overload and long overtime working hours (table 1).
Characteristics of patients with claim by the location of ICH
Younger patients (<30 or 30–39 years old) were more likely to be in the ‘other’ category than other age groups. There was no significant difference in alcohol consumption among the ICH location groups.
Details regarding patient occupation are provided in the online supplemental file 1.
Supplemental material
Deep ICH risk associated with exposure to extreme overload or with the number of overtime working hours
Compensated patients with extreme overload showed a significantly high OR for deep ICH (adjusted OR 1.45, 95% CI: 1.11 to 1.89 in model 1; OR 1.44, 95% CI: 1.10 to 1.88 in model 2) compared with uncompensated patients, although the crude OR of compensated patients was marginally not statistically significant: 1.26 (0.99 to 1.60). Moreover, the crude OR of long-term overtime work was not statistically significant (OR: 1.15, 95% CI: 0.80 to 1.63; OR: 1.25, 95% CI: 0.86 to 1.81; and OR: 1.33, 95% CI: 0.93 to 1.91 for 60–79.9, 80–99.9 and≥100 hours/month, respectively). After adjustment for covariates in model 3, long-term overtime work showed a dose−response relationship with deep ICH compared with less than 60 hours/month (OR: 1.26, 95% CI: 0.87 to 1.83; OR: 1.44, 95% CI: 0.97 to 2.16; and OR: 1.49, 95% CI: 1.01 to 2.21 for 60–79.9, 80–99.9 and ≥100 hours/month, respectively) (table 2). Young age (<30 years and 30–39 years compared with 40–49 years), ‘current smoker’ (compared with non-smokers) and having a medical history of hyperlipidaemia were associated with a low risk of developing deep ICH. Patients receiving treatment for hypertension had a 1.36 times increased risk of developing deep ICH. There were no statistically significant differences in risk status by alcohol consumption and patients’ occupation after adjusted covariates. The ORs for deep ICH risk of compensated patients and long-term overtime workers were similar irrespective of the adjustment for medical history (model 4, table 2).
Relationship between deep ICH and extreme overload or overtime work hours
The sensitivity analysis (excluding the participants whose site of ICH was unknown or the participants with claims for 2016–2017) showed almost the same results.
Site-specific risk of ICH according to overtime work
The OR for lobar haemorrhage was significantly lower in patients with an extreme overload (OR: 0.60, 95% CI: 0.37 to 0.98 in model 2) than in patients without an extreme overload. The ORs for the infratentorial and ‘other’ categories were not significant (OR: 0.85, 95% CI: 0.62 to 1.18 and OR: 0.88, 95% CI: 0.46 to 1.71 in model 2, respectively).
Figure 2 shows the site-specific risks of developing ICH in patients according to overtime work (calculated in model 4). As mentioned above, long-term overtime work had a dose−response relationship with deep ICH. In the lobar ICH category, workers with 80–99.9 hours/month of overtime work had a significantly low OR, although there was a very small number of participants (n=3) in this category. There was no significant relationship with long-term overtime in the infratentorial or ‘other’ categories.
Site-specific risks of ICH according to overtime work. ORs for extreme overload or the number of hours of overtime work in the 6 months before onset of site-specific ICH versus the other site of ICH, after adjustment for covariates (sex, age, onset year, smoking status, alcohol consumption, medical history of hypertension, diabetes and hyperlipidaemia). Deep (in the putamen, thalamus or basal ganglia), infratentorial (in the brain stem or cerebellum), other (tumour, malformation, intraventricular haemorrhage or cerebral venous sinus thrombosis). ICH, intracerebral haemorrhage
Discussion
This study aimed to test the hypothesis that compensated patients with an extreme workload would have a higher risk of developing hypertensive ICH (deep ICH) than uncompensated patients (without extreme overload). We found that compensated patients had a 1.45 times higher risk of developing deep ICH than uncompensated patients. In addition, long-term overtime work has a dose−response relationship with developing deep ICH and that patients engaged in ≥100 hours/month of overtime work had a significant risk of developing deep ICH. To the best of our knowledge, this is the first study to show the relationship between extreme workload and site-specific ICH risk. The strength of our study was that we used data regarding working hours and CCVD diagnosis collected objectively from the employers and the physicians in charge, respectively, by the Labor Standards Inspection Office, unlike previous studies that used self-reported data on working hours.5 14 15
ICH has two subtypes according to pathology. Lobar ICH is associated with cerebral amyloid angiopathy, which is characterised by the deposition of congophilic material, preferentially in the vessels of the cortex and leptomeninges.16 The other subtype of ICH is hypertensive ICH, which is related to degenerative changes in the small penetrating arteries of the deep brain areas.17 A meta-analysis showed that the risk ratio of hypertension was higher for deep ICH than for lobar ICH or other types of ICH.8 11 Hypertension is thought to cause arteriolosclerosis with lipohyalinosis, microatheromas and microaneurysms in the small and deep perforating intracranial arteries. Additionally, a study showed that long working hours were associated with elevated systolic blood pressure (BP).18 Elevated BP as a potential biological mechanism could be the most important risk factor for overwork-related CCVDs.4 18 19 Our result demonstrates that an extreme workload is significantly related to hypertensive ICH and that there seems to be a dose−response relationship between long-term overtime work and hypertensive ICH, as indicated by previous studies.6 20 21 Therefore, our results could support the hypothesis that overtime work causes ICH through hypertension.
Unexpectedly, there was no statistically significant relationship between workload and infratentorial haemorrhage, although infratentorial haemorrhage is thought to have a similar hypertensive aetiology to deep haemorrhage.17 A potential explanation for this discrepancy may be related to the calculation method used in logistic regression analysis. We calculated the OR for infratentorial haemorrhage, compared with non-infratentorial haemorrhage (ie, in the deep, lobar and ‘other’ categories). We think that the high proportion of deep haemorrhage in the reference population (non-infratentorial) may contribute to the absence of a significant association between infratentorial haemorrhage and overtime work. Then, we tried to calculate the OR for infratentorial haemorrhage in model 4 by excluding the deep haemorrhage category patients (data shown in online supplemental file 1). The results showed that overtime work for 80–99.9 hours/month was associated with significantly high ORs for developing infratentorial ICH compared with those for less than 60 hours/month (OR: 0.96, 95% CI: 0.51 to 1.80; OR: 3.06, 95% CI: 1.47 to 6.36; and OR: 1.12, 95% CI: 0.57 to 2.20 for 60–79.9, 80–99.9 and ≥100 hours/month, respectively). Infratentorial haemorrhage was also related to long-term overtime work, although the pattern of the relationship differed from that for deep haemorrhage.
The observation of cerebral microbleeds (CMBs) on magnetic resonance images is generally regarded as indicative of small vessel diseases of the brain, which play an important role in stroke.22 A previous study reported that systolic BP variability was an independent risk factor for deep and infratentorial CMB progression, whereas diastolic BP variability was independently associated with CMB development in deep regions.23 Another study reported that the impacts of hypertension on deep ICH and infratentorial haemorrhage differed according to ethnicity.9 Thus, the impact of hypertension on infratentorial haemorrhage may differ from that on deep haemorrhage.
Notably, the ORs of the effect of extreme workload on hypertensive ICH (deep ICH) were similar with and without adjustment for hypertension treatment, although hypertension treatment was significantly associated with hypertensive ICH. In other words, extreme workload was an independent risk factor for hypertensive ICH. The potential explanations for the significant risk of developing deep ICH associated with extreme workload are as follows: first, individuals performing extreme workload may develop elevated BP over a short period without it being detected during health check-ups. Second, extreme workload could lead to masked hypertension19 24 (normal office BP but high out-of-office BP) or a morning surge in BP. These BP abnormalities have been reported to predict cerebrovascular events, increase arterial damage and lead to sustained hypertension.25–27 Third, individuals performing extreme workloads could have no time to receive treatment even if their BP abnormality was detected. We cannot test these hypotheses without a detailed analysis of the participants’ BP, such as ambulatory BP or BP in the last health check-up, and the medications taken by them. Further clinical or experimental studies are needed for clarifying the mechanism underlying the relationship between extreme workload and hypertensive ICH.
A previous meta-analysis showed that diabetes had a significant association with an increased risk of non-lobar compared with lobar ICH, except in Asian participants.11 Additionally, a Japanese study reported that diabetes was not associated with any significant difference in risk of ICH at any site.28 Our result that diabetes was not significantly correlated with deep ICH is consistent with these previous findings. The current study also showed that hyperlipidaemia was associated with a significantly lower risk for deep ICH compared with the ‘other’ category. The above meta-analysis also demonstrated that hypercholesterolaemia was not significantly associated with ICH at any site.11 On the other hand, a case–control study reported that hypercholesterolaemia was less frequent in non-lobar ICH cases.29 The Rotterdam Scan Study found that triglyceride levels had a strong inverse association with the risk of ICH and that patients with the highest quartile of triglyceride levels had the lowest rate of deep or infratentorial CMB, whereas no such relationship was seen with lobar CMB.30 Thus, hyperlipidaemia, which includes both hypercholesterolaemia and hypertriglyceridaemia, might play a role in the current study results, although we were unable to determine whether participants had hypercholesterolaemia or hypertriglyceridaemia.
A study in Japan showed that patients with brainstem haemorrhage had the highest rate of current smoking (52%), whereas the rates of current smoking were around 25% and 30% in patients with deep (putamen or thalamus) and lobar haemorrhage, respectively.28 The current results are consistent with the previous findings: the highest rate of current smoking was 56.2% in participants with infratentorial ICH, and the smoking rate was lower for those with deep ICH (39.1%). Thus, the OR for deep ICH in those who were current smokers seems to be significantly low.
Our study had some limitations. First, we only included patients with claims for compensation. This prevented us from determining the absolute incident risk of deep ICH by extreme workload. However, we believe that the OR of site-specific ICH can provide meaningful information on the mechanism of the relationship between ICH and extreme workload. Second, we could only use the clinical information about patients which was collected for determining compensations. Thus, we could not gather comprehensive information about the ICH, such as its size and the extent to which the patient’s hypertension, diabetes and hyperlipidaemia were under control. Additionally, we had no available data regarding the use of anticoagulants or antiplatelets, which could have influenced ICH development.31 However, the prevalence rate of individuals with a previous medical history of cerebral infarction, myocardial infarction or angina who were likely to use anticoagulants or antiplatelets was approximately 1%. Therefore, we considered that using anticoagulants or antiplatelets could have had a relatively small impact on the current results. Third, we only analysed data from 2010 to 2015 for uncompensated patients, although data from 2010 to 2017 were analysed for compensated patients. This is because the database of compensated patients was created first. However, the working style or criteria for compensation for overwork-related CCVDs did not change in this period and our sensitive analysis showed almost the same results. Therefore, we prioritised increasing the sample size over uniformity in the data collection years.
In conclusion, ICH patients with extreme workloads have a higher risk of hypertensive ICH than ICH patients without extreme workloads. Our findings indicate that hypertension could play an important role in Japanese overwork-related ICH. Therefore, occupational health professionals and employers should pay attention to the management of overtime workers’ BP. Further research, such as cohort studies focusing on the site-specific ICH risk associated with long working hours and case–control studies investigating the relationship between deep ICH and long working hours with detailed clinical data, such as variations in ambulatory BP or sleep disturbance, are needed to reveal the biological mechanisms underlying the relationship between overwork, hypertension and ICH.
We excluded patients with missing data regarding medical history of hypertension, diabetes and hyperlipidaemia (2.8%) from this analysis.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. The ethics committee of the National Institute of Occupational Safety and Health in Japan (approval number: 2021N26). In this study, we used only anonymised medical data and did not involve active invasion or intervention. Participants (and their family members/relatives for dead cases) were informed of the study goals and allowed to opt-out if they did not want their information to be used for research purposes via the official website of the Ministry of Health, Labour and Welfare or National Institute of Occupational Safety and Health, Japan. None of the workers opted out of participation.
Acknowledgments
We thank the staff members of the Research Center for Overwork-Related Disorders, National Institute of Occupational Safety and Health, Japan, for their assistance with data collection.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @rollpanna9999
Contributors MT and TY designed and directed the project. YM analysed the data and wrote the manuscript in consultation with TY and MT. All authors reviewed the study report,made comments or suggestions on the manuscript drafts and approved the final version. MT is the guarantor.
Funding This work was supported by Industrial Disease Clinical Research Grants from the Ministry of Health, Labor and Welfare, Government of Japan (180902-01, 211001-01).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.