2025 Volume 7 Issue 1 Article ID: 2024-0008
Objective: To clarify the relationship between the implementation of health and productivity management (H&PM) and staff health-related attributes in Japanese hospitals. Method: This study selected 2,000 hospitals from the FY2021 Bed Function Report data and conducted a questionnaire survey from November to December 2023. The questionnaire enquired about the H&PM implementation status, which was the explanatory variable; and health-related attributes, which was the objective variable. The implementation status of the four items and the presence or absence of Excellent H&PM Corporation certification were used to divide the hospitals into three groups: certification, implementation, and non-implementation groups. Logistic regression analysis was conducted with H&PM implementation status and health-related attributes as the variables. Results: Data from 221 hospitals were analyzed. There were 25 hospitals in the certification group, 68 in the implementation group, and 128 in the non-implementation group. Logistic regression used average monthly physician overtime as the outcome, with non-implementation hospitals as the reference. Results showed significant positive associations for the implementation and certification groups. Clear written policies on H&PM promotion and full-time occupational health staff were also significantly associated. However, health issue understanding, plan formulation, and management training were not linked to physician overtime. Other health-related attributes were also unrelated to H&PM implementation status. Conclusion: Hospitals engaging in H&PM may provide an appropriate working environment for physicians.
Medical workers are prone to long working hours owing to a shortage of workers and on-duty shifts1,2,3). Long working hours are associated with increased alcohol intake4), obesity5), and poor sleep6), leading to various health problems. Furthermore, the deterioration of medical workers’ health increases medical errors and influences patient safety7); therefore, it is important to manage the health-related attributes of medical workers and to maintain and promote their health.
The correlation between staff health and organizational productivity has recently attracted increased attention8,9,10). Worsening staff health increases medical costs, affects the survival of an organization, and leads to a decline in productivity11,12); consequently, health management of staff is an important issue for organization managers10).
In Japan, the concept of health and productivity management (H&PM) is becoming widespread13,14). H&PM refers to the “strategic efforts to maintain and promote the health of employees, understanding they are investments to enhance the profitability, etc. from the viewpoint of business management”15).
Additionally, the Excellent H&PM Corporation certification system has been developed to create an environment that can visualize excellent corporations that engage in H&PM and allow them to receive social recognition16). Certification as an Excellent H&PM Corporation requires the dissemination of H&PM policies inside and outside the organization, the appointment of an individual in charge, setting of specific promotion plans (goals), verification of the effectiveness of efforts, and lack of violations of laws and regulations17).
The Study Group on Promoting Health and Productivity Management has identified several indicators for assessing the effects of health investments in H&PM. Among these, output indicators include working hours, smoking rates, and the ratio of exercise habits18).
Studies on H&PM effectiveness in companies show that those certified as Excellent H&PM Corporations enjoy an enhanced image and improved recruitment compared to non-certified ones. This suggests potential benefits, such as lower turnover rates and increased productivity19). Furthermore, organizations that provide health education to managers have a higher proportion of non-smokers19). A previous study related to H&PM in medical institutions found that worse health risk levels have a greater cost of lost productivity and that reducing the health risks of staff improves presenteeism aspects20).
This previous study found that H&PM implementation can improve employee health and productivity, as well as revitalize organizations, in companies and medical institutions. However, few studies have examined the relationship between the H&PM implementation status and the health outcomes or health-related attributes of staff in multi-center medical institutions.
Japan is experiencing a declining birth rate and an aging population, increasing the working generation’s burden. Medical institutions experience issues in maintaining and improving the health of staff, as well as securing and retaining them, owing to the harsh work environments, such as long working hours and night shifts3). Verifying whether H&PM can improve staff health at medical institutions is believed to contribute to improved organizational productivity and the continuous provision of high-quality medical services3) and is of importance from a public health perspective.
Therefore, this study hypothesized that hospitals that implement H&PM will have better health-related attributes for their staff than hospitals that do not.
This ecological study selected hospitals for analysis using a database of hospitals listed in the FY2021 Bed Function Report data21) published by the Ministry of Health, Labour and Welfare of Japan. The Bed Function Report is open data that reports prefectures’ medical functions concerning the beds owned by hospitals, ward equipment, medical staff assigned, medical practices, and other factors22), with 7,019 hospitals reported in the FY2021 Bed Function Report data. This study selected 2,000 hospitals among these and conducted a questionnaire survey regarding the relationship between hospitals and H&PM.
For hospital selection, 403 hospitals were initially excluded for various reasons, including (1) all wards being closed, (2) having ≤1 patient in all wards, (3) having wards with non-zero patient counts despite zero licensed beds, (4) mixing general beds and recuperative care beds in the same ward, and (5) having wards with an average length of stay <1 day (exclusion criteria 1). After excluding these hospitals, the remaining 6,616 hospitals were classified into 99 certified and 6,517 non-certified hospitals. The certified hospitals corresponded to the medical corporations, social welfare corporations, and health insurance associations (“medical corporations, etc.”) that were certified as Excellent H&PM Corporations 2023 from the list of certified corporations23) as of July 2023 published by the Ministry of Economy, Trade and Industry of Japan.
Of the 6,517 non-certified hospitals, 1,049 hospitals that met the exclusion criteria were excluded, resulting in a total of 5,468 hospitals, from which 1,901 were randomly selected. The exclusion criteria were (1) hospitals where the average length of stay could not be calculated; (2) hospitals where any of the variables of the number of beds, nurses, other staff, and patients in the ward was zero; (3) hospitals with <1 registered physician; (4) hospitals with bed occupancy rates exceeding 100% (exclusion criteria 2); and (5) hospitals that could not be linked to addresses in the list of medical institutions by code content as of October 2023, created by each local health and welfare bureau across Japan24,25,26,27,28,29,30,31) (exclusion criteria 3).
After extraction, six hospitals that were not medical corporations and received certification, were revised to certified hospitals. Therefore, 2,000 hospitals were included in the questionnaire distribution, with 105 certified hospitals and 1,895 non-certified hospitals (Figure 1).
A survey was conducted from November to December 2023. The questionnaire asked about H&PM implementation status, health-related attributes, and full-time staff count by sex. Certified hospitals’ H&PM personnel and non-certified hospitals’ health management department personnel were surveyed. Full-time staff referred to directly employed, full-time with no fixed term, and regular staff, excluding contract and part-time workers. No ethics committee approval was needed as the study did not involve human subjects.
Objective variablesThe objective variables, health-related attributes, were assessed using the following six variables: average monthly overtime hours (physician), average monthly overtime hours (nurse), smoking rate, exercise habit percentage, percentage of staff with sufficient rest through sleep, and percentage of staff maintaining appropriate body weight. Respondents were asked in the questionnaire to respond concerning full-time staff. The health outcome items were selected based on items answered in the H&PM survey and previous studies19). The H&PM survey was conducted by the Ministry of Economy, Trade and Industry to understand the H&PM implementation status in corporations and to obtain certification as an Excellent H&PM Corporation32).
The questionnaire included seven items, and respondents were asked to select applicable ones. The response format was made categorical because categorical options were expected to be easier for respondents to answer, thereby increasing the response rate. Hospitals were then categorized based on whether they had many staff with good health-related attributes, serving as a binary outcome variable. The cutoff was set to classify the category closest to 50%. Overtime hours of <15 hours were classified as “low overtime hours.” A smoking rate of <5% was classified as a “low smoking rate.” An exercise habit percentage of ≥20% was classified as “exercise habits present.” The exercise habit percentage was defined as the percentage of staff who exercised twice a week for ≥30 minutes each time. The percentage of staff with sufficient rest through sleep of ≥60% was classified as “good sleep.” The percentage of staff maintaining appropriate body weight of ≥65% was classified as “healthy BMI.” Staff maintaining appropriate body weight was classified as staff with a BMI ranging between 18.5 and <25 kg/m2.
Explanatory variableH&PM implementation status was set as the explanatory variable. The H&PM implementation status was confirmed based on the H&PM survey and previous studies19) and by using the following four items: “Do you have a clearly written policy regarding the promotion of H&PM for your hospital?”, “Does the hospital have full-time professionals (occupational physician, occupational health nurse, nurse) who are involved in promoting H&PM?”, “Do you understand the health issues of staff and have you established a specific H&PM promotion plan?”, and “Do you provide regular training to managers regarding measures to maintain and improve the health of their staff?” If any one of the items were implemented, the hospital was assigned to the implementation group, and if none of the items were implemented, the hospital was assigned to the non-implementation group. Among the implementing hospitals, those that had Excellent H&PM Corporation certification were designated as certified hospitals. Regarding the selection of certified hospitals, the following hospitals from each corporation’s website were extracted from the list of certified corporations as of July 2023 published by the Ministry of Economy, Trade and Industry of Japan23): those with an industry classification of “medical corporation, social medical corporation, health insurance association,” “incorporated association/foundation, chamber of commerce and industry,” and “professional corporation, NPO corporation, public/special corporation, and other corporation.” Furthermore, regarding hospitals where the establishment entity was a “company” in the FY2021 Bed Function Report data21), this study examined the operating company from each hospital’s website and extracted it from the list of certified corporations as of July 202323).
The study used hospital groups (certification, implementation, non-implementation) and the implementation status of four items as explanatory variables. The non-implementation group served as the reference, while the implementation and certification groups were represented as dummy variables. The reference category was set as non-implementation for the implementation status of the four items.
Moderator variablesBed classification, number of full-time staff, and female ratio were used as variables to adjust hospital characteristics. Moderator variables were selected based on previous studies19). Bed classifications were determined from the FY2021 Bed Function Report data. Hospital beds were classified into four categories: highly acute, acute, convalescent, and chronic, with the bed classification that has the largest number of beds defined as the hospital’s bed classification. Subsequently, hospitals were reclassified into two categories, “highly acute or acute care” and “convalescent or chronic care,” with hospitals with highly acute or acute care beds used as a reference, and other hospitals used as a dummy variable. The questionnaire data were used to determine the number of full-time staff and the female ratio.
Characteristics of hospitals responding to the questionnaireTo examine differences between responding and non-responding hospitals, tests were conducted on 19 variables representing hospital characteristics: beds, registered physicians, ward nurses, other staff, total annual ward patients, regional classification, implementing body, being a diagnosis procedure combination (DPC) target status, specific function, community medical support, comprehensive hospitalization system premium, at-home recuperative care support, at-home recuperative care and logistic support, tertiary and secondary emergency facilities, emergency notification, discharge coordination, general beds, and recuperative beds. In the regional classification, the metropolitan type was defined as an area with a population of ≥1 million in the secondary medical area or a population density of ≥2,000 individuals/km2. The regional city type was defined as an area with a population of ≥200,000 in the secondary medical area or a population of 100,000–200,000 and a population density of ≥200 individuals/km2. The depopulated area type was defined as an area that belong to neither the metropolitan nor regional city types33). A DPC target hospital is a hospital that introduced a fixed-rate payment system for inpatient medical expenses. A special function hospital is a hospital that has been certified as providing advanced medical care. A community medical support hospital is a hospital that aims to secure community medical care through support for family physicians. A comprehensive hospitalization system premium assesses if a system has enough personnel and equipment to consistently provide acute care. An at-home recuperative care support hospital offers house calls or home-visit long-term care 24/7. An at-home recuperative care and logistic support hospital is registered in advance to admit patients receiving at-home care in emergencies. A tertiary emergency medical facility treats critically ill patients when primary or secondary emergency response is difficult. A secondary emergency medical facility treats critically ill patients requiring hospitalization or surgery. An emergency notification hospital treats patients more seriously ill than those treated at general acute care hospitals. A discharge coordination department plans discharge destinations for hospitalized patients and provides necessary services. Recuperative beds are used for patients requiring long-term recuperation. This study used the FY2021 Hospital Bed Function Report data21), FY2021 Medical Facility Survey34,35), and Secondary Medical Area Basic Data (Takumi-San) Professional Version_Ver12.0.236) for each variable. These were all open data. Secondary Medical Area Basic Data was created by Wellness Co, Ltd., and was provided free of charge, including medical welfare provision system data and basic statistical information for each secondary medical area across Japan.
The number of general hospitals and general clinics per 100,000 individuals in the secondary medical area to which the hospital belongs was set as variables representing medical resources in the vicinity of the hospital. Furthermore, the proportion of the population aged ≥65 years in the secondary medical care to which the hospital belongs was set as a variable representing patient characteristics in the vicinity of the hospital. The secondary medical area was set as regional units that provide inpatient medical care based on geographical and traffic conditions37). Regarding the regional classification, the metropolitan type was used as a reference, and the other regions were used as the regional classification dummy variables. For the implementing entity, “medical corporation” was used as the reference, and “other than medical corporation” was used as the dummy variable.
Statistical analysisFirst, summary statistics were calculated for the three groups of research hospitals: certified, implementing, and non-implementing. Continuous variables were presented as medians with interquartile ranges (IQRs), while categorical variables were summarized as frequencies and percentages. Comparisons between the three hospital groups were performed using Fisher’s exact test for categorical variables and the Kruskal–Wallis test for continuous variables. However, the implementation status of H&PM was tested between two groups: certified hospitals and implementing hospitals. Additionally, summary statistics comparing respondent and non-respondent hospitals were evaluated, with standardized differences used to assess whether the absolute value exceeded 0.2.
Second, a simple logistic regression analysis (model 1) was conducted to evaluate the association between each outcome indicator and explanatory variables. In a subsequent model (model 2), adjustment variables were introduced. Multiple logistic regression models were further applied to assess the relationship between each outcome and both explanatory and moderator variables. All statistical analyses were conducted using R software (version 4.1.3; R Foundation for Statistical Computing, Vienna, Austria), with a two-sided significance level of 5%. Hospitals that did not respond or were unaware of relevant health-related attributes were excluded from the analysis.
Of the 2,000 hospitals in which the questionnaire was distributed, 221 hospitals (11.1%) provided responses. The breakdown was 25 certified hospitals, 68 implementing hospitals, and 128 non-implementing hospitals. The background of the hospitals in each group is presented in Table 1. Among the health-related attributes, a significant difference (p=0.022) was observed between groups regarding average monthly overtime hours (physicians), with the certification group having the fewest overtime hours. No significant differences were observed in other health-related attributes. Regarding medical institution characteristics, significant differences were observed between groups (p<0.001) concerning the number of full-time staff, with the certification group having the largest number of full-time staff and the implementation and non-implementation group having the fewest number of full-time staff, in descending order (Table 1).
| Certification group (n=25) | Implementation group (n=68) | Non- implementation group (n=128) | p-value | ||||||
|---|---|---|---|---|---|---|---|---|---|
| n (%) Median (Q1, Q3) | n (%) Median (Q1, Q3) | n (%) Median (Q1, Q3) | |||||||
| Medical institution characteristics | |||||||||
| Bed classification | |||||||||
| Highly acute/acute | 9 | 36% | 40 | 59% | 66 | 52% | 0.146 | ||
| Convalescent/chronic | 16 | 64% | 28 | 41% | 62 | 48% | |||
| Number of full-time staff | 397 (192, 596) | 227 (130, 410) | 154 (86, 315) | <0.001 | |||||
| Female ratio | 0.70 (0.68, 0.74) | 0.71 (0.69, 0.78) | 0.73 (0.68, 0.79) | 0.274 | |||||
| H&PM implementation status | |||||||||
| Is it clearly stated in the hospital policy? | |||||||||
| Yes | 23 | 92% | 18 | 27% | 0 | 0% | <0.001* | ||
| No | 2 | 8% | 49 | 73% | 128 | 100% | |||
| Are there full-time occupational health staff? | |||||||||
| Yes | 17 | 68% | 56 | 82% | 0 | 0% | 0.160* | ||
| No | 8 | 32% | 12 | 18% | 128 | 100% | |||
| Are the health issues understood and plans developed? | |||||||||
| Yes | 20 | 80% | 28 | 42% | 0 | 0% | 0.002* | ||
| No | 5 | 20% | 39 | 58% | 128 | 100% | |||
| Is regular management training conducted? | |||||||||
| Yes | 17 | 68% | 25 | 37% | 0 | 0% | 0.010* | ||
| No | 8 | 32% | 42 | 63% | 128 | 100% | |||
| Health-related attributes | |||||||||
| Average monthly overtime hours (physicians) | |||||||||
| Many overtime hours (≥15 hours) | 5 | 24% | 20 | 36% | 53 | 52% | 0.022 | ||
| Few overtime hours (<15 hours) | 16 | 76% | 36 | 64% | 49 | 48% | |||
| Average monthly overtime hours (nurses) | |||||||||
| Many overtime hours (≥15 hours) | 7 | 30% | 15 | 25% | 20 | 19% | 0.360 | ||
| Few overtime hours (<15 hours) | 16 | 70% | 46 | 75% | 88 | 81% | |||
| Smoking rate | |||||||||
| High smoking rate (≥5%) | 15 | 63% | 24 | 62% | 28 | 65% | 0.966 | ||
| Low smoking rate (<5%) | 9 | 38% | 15 | 38% | 15 | 35% | |||
| Exercise habit percentage | |||||||||
| Low exercise habit percentage (<20%) | 13 | 65% | 10 | 67% | 6 | 60% | 1.000 | ||
| High exercise habit percentage (≥20%) | 7 | 35% | 5 | 33% | 4 | 40% | |||
| Percentage of staff with sufficient rest | |||||||||
| Low percentage of staff with sufficient rest (<60%) | 5 | 28% | 7 | 47% | 5 | 50% | 0.418 | ||
| High percentage of staff with sufficient rest (≥60%) | 13 | 72% | 8 | 53% | 5 | 50% | |||
| Percentage of staff maintaining appropriate body weight | |||||||||
| Low percentage of staff maintaining appropriate body weight (<65%) | 8 | 40% | 14 | 47% | 17 | 61% | 0.330 | ||
| High percentage of staff maintaining appropriate body weight (≥65%) | 12 | 60% | 16 | 53% | 11 | 39% | |||
H&PM, health and productivity management; Q1, first quartile; Q3, third quartile.
Fisher’s exact test was used for categorical variables, and the Kruskal-Wallis test was used for continuous variables.
The p-values marked with an asterisk (*) indicate the values that were tested for the significant difference between the Certification group and the implementation group.
Hospitals that did not respond and those with unclear information were excluded.
Confirmation of the summary statistics of the hospitals that responded to the questionnaire and those that did not indicated that no variables had an absolute standardized difference value of >0.2 (eTable 1).
Regression analysisRegarding model 1, a logistic regression analysis was conducted using the average monthly overtime hours (physician) as the objective variable and the hospital group with non-implementation hospitals as the reference variable, with results showing a significant positive association, where the implementation group had an odds ratio (OR) of 1.95 (95% confidence interval [CI], 1.00–3.85) and the certification group had an OR of 3.46 (95% CI, 1.25–11.23) (Table 2). Additionally, a significant positive association was observed for hospitals that “have clearly written policies regarding H&PM promotion within the hospital,” with OR of 3.38 (95% CI, 1.44–8.92). The presence of full-time occupational health staff, understanding health issues and formulating plans, and regular management training were not associated with average monthly overtime hours (physician). Additionally, other health-related attributes were not associated with H&PM implementation status.
| Few overtime hours (physicians) (n=179) | Few overtime hours (nurses) (n=192) | Non-smoking rate (n=106) | Exercise habits present (n=45) | Good sleep (n=43) | Healthy BMI (n=78) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | p | ||
| H&PM implementation status (Reference: Non-implementation group) | |||||||||||||||||||
| Implementation group | 1.95 | 1.00–3.85 | .052 | 0.70 | 0.33–1.51 | 0.351 | 1.17 | 0.47–2.89 | 0.737 | 0.75 | 0.14–4.09 | 0.734 | 1.14 | 0.23–5.84 | 0.870 | 1.77 | 0.63–5.12 | 0.285 | |
| Certification group | 3.46 | 1.25–11.23 | .024 | 0.52 | 0.19–1.50 | 0.205 | 1.12 | 0.39–3.15 | 0.831 | 0.81 | 0.17–4.07 | 0.789 | 2.60 | 0.52–13.82 | 0.245 | 2.32 | 0.73–7.75 | 0.160 | |
| Is it clearly stated in the hospital policy? (Reference: No) | |||||||||||||||||||
| Yes | 3.38 | 1.44–8.92 | .008 | 0.78 | 0.34–1.91 | 0.567 | 1.04 | 0.44–2.45 | 0.921 | 0.63 | 0.18–2.15 | 0.464 | 1.67 | 0.49–5.92 | 0.418 | 1.51 | 0.59–3.90 | 0.390 | |
| Are there full-time occupational health staff? (Reference: No) | |||||||||||||||||||
| Yes | 1.85 | 0.99–3.56 | .058 | 0.65 | 0.32–1.31 | 0.223 | 1.53 | 0.69–3.41 | 0.295 | 0.53 | 0.15–1.83 | 0.312 | 1.12 | 0.32–3.92 | 0.859 | 1.51 | 0.62–3.74 | 0.364 | |
| Are the health issues understood and plans developed? (Reference: No) | |||||||||||||||||||
| Yes | 1.48 | 0.73–3.08 | .289 | 0.80 | 0.37–1.82 | 0.578 | 0.88 | 0.37–2.00 | 0.754 | 0.63 | 0.18–2.15 | 0.464 | 1.31 | 0.38–4.56 | 0.664 | 1.11 | 0.45–2.72 | 0.820 | |
| Is regular management training conducted? (Reference: No) | |||||||||||||||||||
| Yes | 2.03 | 0.95–4.59 | .076 | 1.12 | 0.49–2.81 | 0.800 | 0.90 | 0.37–2.14 | 0.820 | 1.23 | 0.36–4.25 | 0.739 | 1.83 | 0.53–6.74 | 0.345 | 1.59 | 0.62–4.19 | 0.338 | |
BMI, body mass index; CI, confidence interval; H&PM, health and productivity management; OR, odds ratio.
Regarding model 2, which included moderator variables, a significant positive association was found in the hospital groups, with the implementation group having an OR of 2.99 (95% CI, 1.39–6.69) and the certification group having an OR of 3.88 (95% CI, 1.09–17.01) (Table 3). Additionally, a significant positive association was observed for hospitals that “have clearly written policies regarding H&PM promotion within the hospital,” with an OR of 4.84 (95% CI, 1.72–15.85). Furthermore, a significant positive association was observed among hospitals “with full-time professionals involved in promoting H&PM,” with an OR of 2.79 (95% CI, 1.32–6.10). Similar to model 1, no statistically significant differences were found in the relationships between H&PM implementation status and other health-related attributes (Table 3).
| Few overtime hours (physicians) (n=173) | Few overtime hours (nurses) (n=185) | Non-smoking rate (n=101) | Exercise habits present (n=44) | Good sleep (n=41) | Healthy BMI (n=76) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | OR | 95% CI | P | ||
| H&PM implementation status (Reference: Non-implementation group) | |||||||||||||||||||
| Implementation group | 2.99 | 1.39–6.69 | 0.006 | 0.83 | 0.38–1.85 | 0.651 | 1.42 | 0.56–3.63 | 0.460 | 1.11 | 0.17–7.75 | 0.911 | 1.85 | 0.32–11.68 | 0.496 | 1.73 | 0.58–5.33 | 0.330 | |
| Certification group | 3.88 | 1.09–17.01 | 0.049 | 0.69 | 0.22–2.41 | 0.538 | 1.34 | 0.43–4.11 | 0.614 | 0.98 | 0.14–6.97 | 0.986 | 2.49 | 0.40–16.73 | 0.330 | 3.29 | 0.91–12.98 | 0.076 | |
| Is it clearly stated in the hospital policy? (Reference: No) | |||||||||||||||||||
| Yes | 4.84 | 1.72–15.85 | 0.005 | 1.07 | 0.42–2.98 | 0.896 | 1.17 | 0.46–2.96 | 0.735 | 0.60 | 0.12–2.94 | 0.528 | 1.25 | 0.27–5.85 | 0.770 | 2.14 | 0.73–6.74 | 0.176 | |
| Are there full-time occupational health staff? (Reference: No) | |||||||||||||||||||
| Yes | 2.79 | 1.32–6.10 | 0.008 | 0.77 | 0.37–1.64 | 0.492 | 1.71 | 0.75–3.96 | 0.203 | 0.56 | 0.12–2.37 | 0.430 | 1.35 | 0.33–5.43 | 0.673 | 1.89 | 0.73–5.03 | 0.196 | |
| Are the health issues understood and plans developed? (Reference: No) | |||||||||||||||||||
| Yes | 1.55 | 0.68–3.65 | 0.299 | 1.14 | 0.49–2.90 | 0.777 | 1.02 | 0.41–2.47 | 0.969 | 0.62 | 0.13–2.76 | 0.528 | 1.25 | 0.32–5.05 | 0.746 | 1.34 | 0.52–3.58 | 0.549 | |
| Is regular management training conducted? (Reference: No) | |||||||||||||||||||
| Yes | 2.02 | 0.84–5.14 | 0.124 | 1.26 | 0.52–3.42 | 0.624 | 0.85 | 0.33–2.09 | 0.723 | 1.53 | 0.35–7.24 | 0.573 | 2.37 | 0.62–9.98 | 0.216 | 2.05 | 0.75–5.95 | 0.170 | |
BMI, body mass index; CI, confidence interval; H&PM, health and productivity management; OR, odds ratio.
Moderator variables: bed classification (binary value of highly acute or acute, and convalescent or chronic), number of full-time staff, female ratio
This study verified the relationship between H&PM implementation status and staff health-related attributes in hospitals. Among the health-related attributes, average monthly overtime hours of physicians had a significant association with H&PM implementation status, suggesting that hospitals that conduct H&PM may have shorter overtime hours for physicians and may have an appropriate working environment for physicians. However, by clearly stating these in hospital policies, hospitals may have been able to present the reduction of overtime work as an organization-wide issue and promote it. Additionally, the presence of full-time professionals involved in promoting H&PM may have facilitated a more effective response to excessive workloads.
In Japan, from April 2024, an upper limit was applied to physicians’ overtime hours, and hospitals were required to respond. Hospitals that implement H&PM may be implementing appropriate labor management. Meanwhile, no association was found among health outcome items other than physicians’ overtime work. Previous studies found a relationship between managerial education and smoking rates in companies19); however, smoking rates tend to be low in medical institutions38), suggesting that no differences may be found. Additionally, similar to previous studies, the exercise habit percentage, percentage of staff with sufficient rest through sleep, and percentage of staff maintaining appropriate body weight were not associated with H&PM implementation19). Among these, sleep time has a strong association with working hours6), and despite a significant difference being found for the overtime hours of physicians, the fact that no difference was present for the percentage of staff with sufficient rest through sleep was unexpected. This may be due to the recent introduction of work intervals stipulated by Japanese law. Work intervals are designed to provide a certain amount of rest time between the end of a day’s work and the start of the next day’s work to ensure time for daily life and sleep39).
This study has four limitations. First, the low response rate of 11.1% suggests potential bias among responding hospitals. Using the standardized difference, we confirmed that there were no differences in characteristics between the responding and non-responding hospitals. However, this comparison did not extend to the level of interest in H&PM. The response rate was 24.0% for certified hospitals and 10.3% for non-certified hospitals. Among the non-certified hospitals, there may be a tendency for those that are implementing H&PM or have a high level of interest to respond, which could potentially result in an underestimation of the true outcomes. Second, being an ecological study, causal relationships could not be established. It remains unclear whether H&PM implementation reduced physicians’ overtime or if hospitals able to reduce physician overtime implemented H&PM. Future longitudinal studies are needed to clarify causal relationships. Third, being a questionnaire survey, it may not fully reflect objective data. Fourth, adding other variables as adjustment variables may affect the results. For example, there may be differences in health-related attributes based on the age composition of the staff, but such data has not been collected. Additionally, measures against overtime work may differ in their degree of control depending on the type of hospital (eg, private or public), but this has not been examined in the present analysis.
Despite the limited extant studies examining the relationship between H&PM implementation status and employee health-related attributes in multi-center medical institutions, this study statistically showed that H&PM in hospitals was associated with physicians’ overtime work. The results of this study are valuable given that working hours are increasing. Few extant studies verified the effectiveness of H&PM in medical institutions, indicating the necessity of further research.
This study revealed a positive association between H&PM implementation status in hospitals and the percentage of physicians who work shorter overtime hours. Hospitals engaging in H&PM may have an appropriate working environment for physicians.
More than 200 hospitals participated in the survey. We would like to express our sincere gratitude to the personnel who responded.
All authors were involved in study design, data interpretation, and data analysis. HW, SM, and SK prepared the questionnaire. HW compiled the questionnaire and wrote the manuscript. SM, SK, and YN helped prepare the manuscript. All authors revised the manuscript and approved the final manuscript.
The authors declare that there is no conflict of interest.
The data underlying this article will be shared on reasonable request to the corresponding author.
This work was supported by JSPS KAKENHI Grant Number JP22K11940.
