Workplace Social Capital and Well-Being in Medical Residents

Abstract

The authors analyzed survey data from 276 medical residents at 32 hospitals in Japan to examine associations among workplace social capital, subjective well-being, and work engagement. Scores on both outcomes increased across social capital quartiles, even after adjusting for confounders. The authors recommend interventions to strengthen trust and reciprocity within training environments. Because of the cross-sectional design, low response rate, and reliance on self-reported data, future research must establish causal relationships.

Background

Medical residents face heavy workloads, time pressure, and psychological and physical demands from supervisors and patients. Social isolation, irregular sleep, and limited exercise also threaten their well-being. Previous studies found burnout in approximately half of all residents. The coronavirus disease (COVID-19) pandemic further worsened physicians’ emotional and mental health. Burnout and reduced well-being can increase medical errors and compromise both care quality and patient safety.

Work engagement describes a positive, fulfilling psychological state characterized by dedication, vigor, and absorption. Past research links this state to fewer mental health issues, reduced absenteeism, greater life satisfaction, and improved work performance. Although studies of residents remain scarce, higher work engagement correlates with better care quality, greater training satisfaction, and fewer positive screens for depression.

Workplace social capital (WSC) encompasses the resources available through professional social networks. A Danish study excluded healthcare workers, while a Swedish study used a general measure that was not specific to the workplace. A Japanese study included only 7 physicians among 440 participants and did not record their years of training. The authors examined whether workplace social capital was associated with subjective well-being and work engagement among medical residents.

Methods

The research team identified clinical training hospitals through Regional Bureau of Health and Welfare websites across Japan and invited training directors at 78 institutions to participate. 32 hospitals agreed to participate. The multicenter cross-sectional survey, conducted in July–August 2022, ran concurrently with a validation study of a workplace social capital scale for Japanese medical residents.

Training directors distributed online SurveyMonkey invitations to their residents. The survey instructions emphasized that participation was voluntary and that declining carried no penalty. The researchers sent three reminders to initial nonrespondents.

Table 1. Characteristics of participating hospitals

Characteristic	N (%)
Hospital type
Community hospital	30 (94)
University hospital	2 (6)
Hospital size
≤ 500 beds	19 (59)
501–800 beds	8 (25)
801–1000 beds	3 (9)
≥ 1001 beds	2 (6)
Hospital location
Hokkaido and Tohoku	5 (16)
Kanto	6 (19)
Chubu	6 (19)
Kinki	4 (13)
Chugoku and Shikoku	5 (16)
Kyushu	6 (19)

The authors assessed the primary outcome using the Subjective Well-Being Scale (SWBS), a 12-item instrument validated in Japanese. Each item is scored from 1 to 4, yielding a total score of 12 to 48. They evaluated the secondary outcome with the Japanese version of the Utrecht Work Engagement Scale (UWES). This instrument comprises 9 equally weighted items scored from 0 to 6, which are summed to yield a total score from 0 to 54. Higher total scores reflect greater well-being and work engagement, respectively.

To measure the explanatory variable, the team used the Japanese medical resident version of the Workplace Social Capital Scale (JMR-WSC). They adapted this 8-item instrument from an original Finnish scale and validated it for residents. Respondents rate items from 1 (“strongly disagree”) to 5 (“strongly agree”), and the scores are averaged.

The researchers reported continuous variables as means and standard deviations and categorical variables as frequencies and percentages. Because the intraclass correlation coefficients (ICC) for the SWBS and UWES were below 0.1, the researchers used conventional multivariable linear regression. They adjusted the models for gender, postgraduate year, hospital type and size, and specialty. To address potential violations of the linearity assumption, they divided WSC scores into quartiles. Given the small number of missing values, the team conducted a complete-case analysis. The two-sided significance threshold was P < 0.05.

The authors determined that a minimum sample size of 240 participants was required, assuming at least 20 participants for each of the 12 independent variables. They performed all statistical analyses using SPSS Statistics 29.0 (IBM Japan, Tokyo, Japan). Participants provided consent via a checkbox on the opening screen. 10 respondents received ¥5,000 gift vouchers through a lottery. The University of Tokyo institutional review board approved the study (approval number 2022062NI).

Results

Of the 1,290 residents invited, 289 (22.4%) completed the survey. After excluding 13 responses with missing data, the final analysis included 276 participants. The median hospital response rate was 21.9%, with an interquartile range of 10.4% to 35.9%.

Table 2. Participant and measurement characteristics (N = 276)

	Value
Gender, n (%)
Women	99 (35.9)
Men	176 (63.8)
Nonbinary	1 (0.4)
Postgraduate year (PGY), n (%)
PGY 1–2 (junior residents)	177 (64.1)
PGY 3–6 (senior residents)	99 (35.9)
Hospital type, n (%)
Community hospital	164 (59.4)
University hospital	112 (40.6)
Hospital size, n (%)
≤ 500 beds	87 (31.5)
501–800 beds	56 (20.3)
801–1000 beds	17 (6.2)
≥ 1001 beds	116 (42.0)
Specialty, n (%)
Medical specialtiesᵃ	160 (58.0)
Surgical specialtiesᵇ	73 (26.4)
Other specialtiesᶜ	43 (15.6)
WSC scale, mean (standard deviation)	4.05 (0.74)
SWBS, mean (standard deviation)	34.39 (5.16)
UWES, mean (standard deviation)	30.07 (11.22)

ᵃ Internal medicine, pediatrics, and general practice. ᵇ General surgery, neurosurgery, obstetrics and gynecology, ophthalmology, orthopedic surgery, otolaryngology, plastic surgery, and urology. ᶜ Anesthesiology, dermatology, emergency medicine, psychiatry, and radiology. PGY stands for postgraduate year.

Well-being and work engagement increased across successive WSC quartiles, even after adjustment for confounders. The adjusted mean differences between the highest and lowest quartiles were 6.55 points for SWBS and 15.12 points for UWES.

Table 3. Associations of residents’ WSC scores with SWBS and UWES scores (N = 276)

	Unadjusted mean difference	95% CI	Adjustedᵃ mean difference	95% CI
SWBSᵇ
WSC Q1ᵈ	Ref	Ref	Ref	Ref
WSC Q2ᵉ	4.07	2.44–5.69*	3.83	2.16–5.49*
WSC Q3ᶠ	4.45	2.88–6.03*	4.54	2.96–6.13*
WSC Q4ᵍ	6.64	5.06–8.22*	6.55	4.96–8.15*
UWESᶜ
WSC Q1ᵈ	Ref	Ref	Ref	Ref
WSC Q2ᵉ	8.46	4.93–11.99*	8.63	5.03–12.23*
WSC Q3ᶠ	8.48	5.06–11.90*	8.94	5.51–12.37*
WSC Q4ᵍ	14.66	11.22–18.10*	15.12	11.66–18.57*

CI denotes confidence interval, Q denotes quartile, and Ref denotes the reference group. ᵃ Adjusted for postgraduate year, gender, hospital type, hospital size, and specialty. ᵇ Score range 12–48. ᶜ Score range 0–54. WSC intervals, as reported in the original article, are ᵈ 0.000–3.750, ᵉ 3.750–4.125, ᶠ 4.125–4.625, and ᵍ 4.625–5.000. *p < 0.01.

Discussion

The authors identified three potential pathways linking social capital to the outcomes. First, social capital can improve communication, cooperation, and coordination with colleagues and supervisors, thereby enhancing work performance and self-efficacy. Second, informal social control—in which peers and supervisors monitor residents’ professional attitudes—can promote a fair and positive workplace atmosphere. Third, supportive networks provide knowledge, skills, and mentoring.

The authors recommend interventions that build reciprocity and trust among members and across hierarchical levels. Potential activities include improving communication, strengthening social cohesion and cooperation, and involving residents in decision-making. They identify medical escape rooms—an immersive, simulation-based learning format—as one approach. Specific methods for building WSC remain under development and require further empirical support.

The study documented both associations specifically among medical residents and used an internationally recognized WSC instrument. The participating hospitals varied in region, size, and type. The team judged the study’s external validity to be high, while acknowledging uncertainty about the representativeness of the participating hospitals.

Given the cross-sectional design, the findings are best interpreted as associations rather than causal relationships. The authors suggest longitudinal follow-up studies. They identify the low response rate and the brief survey window as limitations. If residents with lower WSC, well-being, or work engagement avoided the survey, the analysis might underestimate the true associations. Measuring both the explanatory variable and the outcomes through self-report introduces common method bias, which could overestimate the relationships. Unmeasured confounding also remains a possibility. The participating hospitals might share a pre-existing interest in occupational health.

Conclusions

The authors conclude that workplace social capital correlates positively with subjective well-being and work engagement among Japanese medical residents. They infer that interventions to expand social capital may improve both outcomes.

Reading notes

Because all three variables were measured through self-report at the same point in time, well-being and work engagement may be reflected in assessments of networks. The JMR-WSC, SWBS, and UWES measured assessments of networks, well-being, and work engagement. Knowledge sharing and mentoring remain hypotheses about the processes underlying the observed association. The proposed explanation that networks provide knowledge, skills, and mentoring opportunities can also be considered from the perspective of workplace learning in HRD. It raises a further question about how residents’ opportunities to learn from colleagues and attendings, as well as the learning experiences themselves, relate to well-being and work engagement.

References

Fujikawa, H., Aoki, T., & Eto, M. (2024). Associations between workplace social capital, well-being, and work engagement in medical residents: A multicenter cross-sectional study. BMC Medical Education, 24(1), Article 1063. https://doi.org/10.1186/s12909-024-06055-0