Telecommuting-related health outcomes during the COVID-19 pandemic in South Korea: a national population-based cross-sectional study | BMC Public Health

This study investigated the association between health-related outcomes and telecommuting by comparing comprehensive issues, including physical health, mental health, absenteeism, and presenteeism between telecommuters and daily commuters in South Korea. Six hundred fifty workers met the telecommuting definition; among them, 338 were males (312 females). Compared with daily commuters, telecommuters showed a higher prevalence of anxiety, insomnia, fatigue, musculoskeletal pain, headache/eye strain, and presenteeism. In addition, there were gender differences in the health status of telecommuters. For both depression and insomnia, a significant association between telecommuting was observed only in females. Overall, female workers seemed to be more susceptible to mental illness due to telecommuting.

Following the selection criteria of our study, 4.21% (n = 650) of all the participants (n = 15,451) were classified as telecommuters, which is a two-fold increase from the 5th KWCS (Table S1). Considering that the previous survey did not encompass the questionnaires on ICT device utilization, comparison between the two data should be performed with caution. The proportion of telecommuters in the working population of South Korea during the pandemic has been explored in prior studies using other secondary data. The 23rd Korean Labor and Income Panel Study identified the status of flexible work arrangements in March 2020 and showed 2.27% of paid workers worked from home [26]. Similarly, the South Korean Economically Active Population Survey held in August 2020 revealed an upsurge in telecommuters from 0.47% in 2019 to 2.49% in 2020 among paid workers [18].

Our findings are consistent with those of previous studies in which telecommuters showed worse health indicators than commuters. A cross-sectional study using data from the 2010, 2012, and 2013 American Time Use Survey assessed the subjective well-being of 3,962 wage workers and showed that telework was associated with higher psychological stress than office work, with coefficients of 0.298 (p-value of < 0.01) from fixed-effect regression models [27]. Another cross-sectional study investigated the effect of teleworking on physical discomfort among university faculty and staff who were forced to transition into teleworking during the COVID-19 pandemic. A total of 131 university members (86%) complained of new or worsening physical discomfort after telecommuting, while 7% reported improvements in their existing problems [28]. One multilevel regression from 25,465 workers in the 6th EWCS data showed that teleworking several times a week or daily resulted in a higher probability (11%) of experiencing presenteeism at least once per year with never teleworking as a reference [29]. However, the results of other studies differ from those of the present study. Henke et al. (2016) analyzed self-reported data from 3,703 financial workers. The results showed that telecommuters (≥ 73 h/month) had a lower overall health risk score than non-telecommuters (coefficient, -1.233; p-value < 0.05) [24]. A retrospective case-control study using screening data from 1,978 South Korean workers during the COVID-19 pandemic suggested that certain workplace interventions, including telecommuting, led to a significant decrease in depression and anxiety [30]. The former study differs from our research by analyzing only a specific occupation group, while the latter integrated paid leave and telecommuting into a single intervention. Such differences in study design would have led to opposite results.

To understand the link between telecommuting and its detrimental influence on health, differences in the working environment between remote and physical offices should be considered. A major concern of telecommuting is the blurring of boundaries between work and private life. Telecommuters are more likely to engage in less structured, longer working hours and have non-regular work schedules [7, 31, 32]. The negative impact of long working hours, such as chronic fatigue and physical/mental health problems, is well established [33]. These intense and extended hours at remote workstations have been associated with increased physical discomfort, musculoskeletal pain, burnout, and eyestrain [34]. A high prevalence of insomnia was found among employees with non-regular working patterns [35]. Telecommuters are also at greater risk of developing poor dietary habits and scarce exercise [36, 37]. Such unhealthy lifestyles can, in turn, cause workers’ fatigue and negatively affect their mental well-being [36, 38]. Another proposed risk factor for telecommuters is inappropriate ergonomic environment. Frequent use of non-office equipment, including chairs without armrests or laptops with no external monitor, exacerbates neck, shoulder, and lower back pain among telecommuters [39, 40]. Additionally, during the transition to remote work, telecommuters may lose resources at work, such as coworkers’ support, resulting in social isolation [41]. Many studies have suggested that social isolation and lack of support from coworkers lead to health impairments and are predictive of depressive disorders and burnout [42, 43]. The role of social support in moderating psychosocial stress responses is well known [44]. Regardless of widespread social distancing due to COVID-19, social isolation among telecommuters has consistently been observed [13, 45]. Vander Elst et al. (2017) demonstrated that the extent of telecommuting is negatively related to well-being because of the lack of social support from colleagues [46]. More importantly, during the global COVID-19 pandemic, the number of companies and governments implementing social distancing has surged, requiring employees to telecommute. Given that workers who engaged in involuntary teleworking reported greater emotional exhaustion than their voluntary counterparts, this sudden and unplanned shift to remote work may exacerbate the negative impact of telecommuting on their mental and physical health statuses [47]. The higher risk of presenteeism among telecommuters is partly attributable to their attitudes toward working from home. Notably, telecommuters may perceive their condition as a privilege to work comfortably at home and have the motivation to work incessantly in order not to miss the opportunity even if their health is compromised [48].

Interestingly, we found that female workers were more likely to have telecommuting-related health problems than male workers, indicating a gender difference in the effect of telecommuting on health-related outcomes. Similar findings were also reported in two recent studies [15, 49]. Giménez-Nadal et al. (2020) analyzed data from the American Time Use Survey and observed gender differences in the well-being of teleworkers, showing that male teleworkers had significantly lower levels of subjective stress, pain, and tiredness than commuters, while corresponding results were not found among female workers [15]. Graham et al. (2021) studied Australian telecommuters during the pandemic and found that females had higher levels of musculoskeletal discomfort/pain and psychosocial stress (OR = 2.06; 95% CI:1.38–3.08) than males [49]. The observed gender difference may be partially explained by the fact that teleworking has different connotations for males and females due to traditional gender roles.

In traditional households, females are responsible for domestic roles such as house chores and parenting, especially in East Asian countries [48, 50]. Excessive household chores are known to decrease work productivity, cause telecommuters to disengage from work, and increase job stress and sleep disturbances [13]. Although telecommuters spend less time commuting and gain more free time, females tend to invest extra time in household chores, unlike males, who are able to reinforce work-life balance with more leisure time [6, 51]. This may lead to a higher workload for female teleworkers, and the demands of work and family are combined, which become more harmful to their health. In addition, although it is unclear whether females are more vulnerable to social isolation caused by telecommuting, female workers reported feeling more loneliness and anxiety than males during the COVID-19 pandemic. [51, 52] A combination of disconnected relationships from co-workers and restrictions on social activity may pose a threat to the mental health of female telecommuters.

To the best of our knowledge, this is the first study to describe multifaceted, increased health problems among telecommuters in relation to daily commuters during the COVID-19 pandemic. The strength of our study is based on a large-scale, representative sample of South Korean workers. By applying the definition of telecommuting as inclusion/exclusion criteria into the participant’s selection process, employees who work from home were screened out. Our study also included overall health issues, specifically mental/physical health conditions and absenteeism/presenteeism, and we considered the gender aspects of telecommuting-related health problems. However, this study has several limitations. Most importantly, this was a cross-sectional study with its inherent inability to certify a causal relationship between exposure and outcome. In other words, whether telecommuting itself causes physical or mental illness or whether unhealthy workers are more inclined to work from home is unclear. Second, there were a few defects in the setting of independent variables. Although telecommuters’ workplace covers all non-central office spaces where both employers and employees agree, the current survey only inquired about “home” as a telecommuting place. The classification criteria for distinguishing telecommuters from commuters were a consequence of the author’s discretion, but were not referred from previous studies. Moreover, as the questionnaire item regarding telecommuting status followed qualitative criteria, we could not evaluate the quantitative intensity of telecommuting. Third, due to a lack of survey items, other possible confounders, such as lifestyle information (i.e. drinking and smoking), personal attitudes toward work, residential environment, or past medical history were not measured. Finally, rather than documented medical insurance claims or clinical records, outcome variables were measured by responses to questionnaires; therefore, a recall bias cannot be ruled out.