Healthcare Workers in Japan Felt Less Happy Even After COVID Emergency Ended

When Japan declared a state of emergency in the spring of 2020, healthcare workers stood on the front line of an unprecedented public health crisis. New research now suggests that the psychological toll of that period did not simply evaporate once the emergency measures were lifted. In fact, according to a study published in Discover Social Science and Health, the gap in self-reported happiness between healthcare workers and other workers was largest after the state of emergency ended, a finding that carries important implications for how health systems support their staff during the recovery phases of a pandemic.

The study, conducted by economist Eiji Yamamura of Seinan Gakuin University, Fumio Ohtake of Osaka University, and Yoshiro Tsutsui of Kobe International University, draws on a uniquely rich source of evidence: a series of five internet surveys administered to a large panel of Japanese residents between March 13 and June 12, 2020. This window captures the entire arc of Japan’s first COVID-19 state of emergency, from its announcement through its lifting, allowing the researchers to observe how subjective well-being evolved in near real time as the epidemiological and policy landscape shifted.

The analytical sample comprised 10,901 person-wave observations for the panel models and 10,689 observations for the cross-sectional estimation techniques. Respondents rated their own happiness on an eleven-point scale, ranging from zero, indicating the least happy, to ten, indicating the most happy. This self-reported measure, widely used in economics and psychology research, captures an individual’s overall cognitive evaluation of their life satisfaction at a given moment, offering a window into subjective well-being that is distinct from clinically diagnosed mental health conditions such as depression or anxiety disorders.

What sets this study apart methodologically is the rigor with which the authors treated the ordinal nature of the happiness scale. Because a numerical happiness rating does not guarantee that the distance between, say, a four and a five is identical to the distance between an eight and a nine, the researchers estimated their models using several complementary statistical approaches. These included ordinary least squares regression, ordered logit models, probit models, and both fixed-effects ordinary least squares and panel ordered logit specifications that exploit the repeated observations of the same individuals over time. The consistency of results across linear, ordinal, and binary frameworks strengthens confidence that the findings are not artifacts of any particular modeling assumption.

The first core finding is stark in its simplicity: throughout the entire study period, healthcare workers reported consistently lower levels of happiness than workers in other occupations. This disparity existed before the state of emergency was formally declared, during the acute crisis period while the emergency was in force, and after it was lifted. In other words, being a healthcare worker in Japan during the early months of the pandemic was associated with a persistent happiness penalty relative to the rest of the workforce, regardless of the prevailing policy environment.

The second finding is more surprising and arguably more consequential. The largest gap between healthcare workers and other workers did not occur during the emergency itself but after it ended. In the fixed-effects specification, the interaction between the fifth survey wave and healthcare worker status indicates that the well-being gap widened by 0.26 points on the eleven-point scale relative to the first wave. While 0.26 points may appear modest, in the context of an eleven-point self-report scale and a difference-in-differences style estimation, it represents a meaningful and statistically detectable divergence in subjective well-being during a period when one might expect relief and recovery.

The probit analysis sharpens this picture further. Using a binary definition of relative happiness, with scores of six or above on the eleven-point scale classified as relatively happy, the model estimates that healthcare workers were 13.8 percentage points less likely than other workers to report being relatively happy in the fifth wave, a difference significant at the one percent level. That such a substantial gap emerged precisely when the emergency measures were lifted suggests that the end of a formal crisis declaration does not mark the end of the psychological burden carried by frontline medical staff.

Several mechanisms could plausibly underlie this pattern, though the authors are careful to note that their data identify associations rather than definitive causal pathways. During the emergency period, healthcare workers may have experienced a sense of shared national purpose, heightened social appreciation, and clear institutional directives that framed their hardship as temporary and collective. Once the emergency was lifted and society at large began returning to something resembling normal life, the sustained workload, infection risk, and emotional strain of frontline work may have become more acutely felt by comparison. The contrast between society’s reopening and the continued demands placed on medical personnel could have deepened the perceived inequity of their situation, a phenomenon consistent with the social psychology of relative deprivation.

The timing also raises questions about the rhythm of institutional support. Public health emergencies mobilize resources, public sympathy, and media attention toward healthcare workers, but these tend to dissipate quickly once the acute phase passes. If the happiness gap is widest after the emergency is lifted, then the period of transition out of emergency measures may be precisely when healthcare organizations should intensify rather than withdraw psychological and material support for their staff. The authors suggest that targeted support for frontline workers during these transitions could be valuable from a healthcare organization perspective, although they caution that because their study measures self-reported happiness rather than clinical mental health outcomes, burnout, or turnover intentions, broader implications should be interpreted cautiously.

Methodologically, the panel structure of the data deserves emphasis. By observing the same individuals across five waves, the fixed-effects models control for all time-invariant individual characteristics, such as stable personality traits, baseline disposition toward optimism or pessimism, and enduring occupational selection effects. This means the widening happiness gap cannot be easily explained by healthcare workers simply being different kinds of people from the start. Instead, the divergence reflects changes over time in the relative well-being of the same workers, changes that align temporally with the lifting of the state of emergency. The use of panel ordered logit estimation further guards against the possibility that the linear models misrepresent the underlying scale of the happiness variable.

The study also benefits from its grounding in a carefully documented survey infrastructure. The surveys were conducted by the research firm INTAGE Inc. as part of an urgent academic response to the early stage of the pandemic in Japan, a period when standard ethical review timelines were difficult to satisfy given the rapidly evolving public health situation. Formal ethics approval from the Ethics Committee of the Graduate School of Economics at Osaka University was granted retrospectively on October 14, 2020, after data collection had concluded, and the authors transparently acknowledge this in the published paper. Informed consent was obtained from all participants at the time of the survey, and all procedures followed the Ethical Guidelines for Medical and Health Research Involving Human Subjects enforced by Japan’s Ministry of Health, Labour and Welfare and Ministry of Education, Culture, Sports, Science and Technology. The research was funded by the Japan Society for the Promotion of Science under grants 18KK0048, led by Yoshiro Tsutsui, and 20H05632, led by Fumio Ohtake.

The limitations of the research are worth considering alongside its contributions. Happiness measured on an eleven-point scale is a subjective construct, and while it correlates with objectively measurable outcomes in large samples, it does not substitute for clinical assessment of depression, anxiety, post-traumatic stress, or burnout syndrome. The survey window covers only the first three months of the pandemic in 2020, leaving open the question of whether the post-emergency happiness gap persisted, widened, or narrowed in the months and years that followed as Japan navigated subsequent waves of infection. The study also identifies healthcare workers as a broad category, and the experiences of physicians, nurses, care workers, and auxiliary staff may differ in ways that a single occupational flag cannot capture.

Nevertheless, the findings arrive at a moment when health systems worldwide are still absorbing the lessons of the COVID-19 pandemic. Studies from many countries have documented elevated rates of psychological distress among healthcare workers during the crisis, but fewer have tracked self-reported well-being across the specific transition from emergency measures to their lifting. By showing that the well-being penalty for healthcare workers peaked after the formal crisis ended, this Japanese panel study adds an important temporal dimension to the international evidence base. It suggests that the aftermath of a declared emergency, when public attention shifts elsewhere, may be a critical window for intervention, and that policies designed to protect the well-being of the healthcare workforce should extend well beyond the duration of the emergency measures themselves.

As pandemics and other public health emergencies remain a recurring threat, the question of how to sustain the morale and well-being of medical workers through every phase of a crisis, including its quiet aftermath, is likely to become an increasingly central concern for health policy researchers and administrators alike. This study offers a rigorous, data-driven starting point for that conversation, demonstrating with multiple complementary statistical methods that the end of the emergency was, for Japan’s healthcare workers, not the end of their burden but perhaps the moment it weighed most heavily on their reported happiness.

Subject of Research: Changes in self-reported happiness among Japanese healthcare workers across the transition from the COVID-19 state of emergency to its lifting in 2020.

Article Title: Subjective well-being of healthcare workers after Japan’s COVID-19 state of emergency

Article References: Yamamura, E., Ohtake, F., & Tsutsui, Y. (2026). Subjective well-being of healthcare workers after Japan’s COVID-19 state of emergency. Discover Social Science and Health. https://doi.org/10.1007/s44155-026-00473-9

Image Credits: AI Generated

DOI: 10.1007/s44155-026-00473-9

Keywords: COVID-19, subjective wellbeing, happiness, healthcare workers, state of emergency, Japan, panel data, ordered logit, fixed effects, public health, occupational health, mental health