Correlation between uncertainty stress and depression among healthcare professionals in China: a nationwide cross-sectional survey


In 2008, the WHO ranked depression as the third-largest global burden of disease, with projections suggesting it will become the leading cause by 2030.1 Over the years, multiple studies have demonstrated a high prevalence of depression among healthcare professionals (HCPs).2–4 Before the COVID-19 pandemic, most investigations of depression among HCPs were limited in scope. Behera surveyed 257 primary care workers in Indonesia and found that 21 (9.5%) participants were experiencing major depression.5 Similarly, Knuth et al surveyed 181 primary care workers in Brazil and found that 28.2% of them had experienced depressive episodes.6 The COVID-19 pandemic has exacerbated this trend,7 drawing more attention to the psychological well-being of HCPs. A meta-analysis revealed that the global prevalence of depression among HCPs during the COVID-19 pandemic was 21.7% (95% CI 18.3% to 25.2%).8

Depression is significant among Chinese doctors, with reported prevalence rates ranging from 25.2% to 57.6%.9–11 Given these disparities, it is evident that the prevalence of depression among HCPs far exceeds that of the general population, which averages approximately one in ten people.1 The above findings suggest that depression has reached epidemic proportions among HCPs, necessitating a pressing need for a deeper understanding of the underlying mechanisms and the development of effective interventions.

Scholz defined uncertainty from a psychological perspective as ‘incomplete information or knowledge of a situation, or possible alternatives, or their probability of occurring, or their results unknown to the subject’.12 Yang defined uncertainty stress (US) as stress resulting from an individual’s sense of uncertainty.13 In addition, US is widely thought to be associated with various adverse health outcomes, including self-harm and suicidal thoughts.14 Currently, US assessments are extensively used in surveys related to community populations, mental health among college students, addiction and other related areas.14–19 For instance, Zhu et al surveyed 2534 Chinese residents and found that high US was negatively related to self-rated health (r=−0.256, p<0.01) and positively associated with increased negative emotions (r=0.646, p<0.01).17 Indeed, medical science is inherently characterised by uncertainty, and HCPs continually face uncertainty in clinical decision-making.20 Such uncertainty is exacerbated by epidemics such as COVID-19.21 Additionally, HCPs in China face the risk of medical violence.22 Collectively, these factors can contribute to US. In terms of past research, previous studies on US among HCPs have been qualitative in nature and lack quantitative results.

Stress plays a pivotal role in the development of depression,23–25 encompassing factors such as stressful life events23 and work stress. Although US is a significant stressor,26 few studies have directly confirmed the link between US and depression. Zhang et al proposed that US partially mediates the association between well-intentioned childhood experiences and depressive symptoms (indirect effect=−0.47, 95% bootstrap CI −0.55 to –0.39).18 However, there is a paucity of research on the relationship between US and depression in HCPs.

Hence, we investigated the prevalence of US and depression among HCPs. Furthermore, we explored the correlations between US and depression in this population. We also put forth a key hypothesis that US is directly and positively associated with depression.


Design, setting and participants

Our research plan for this study was as follows:

Step 1: given the significant geographic and economic disparities across China’s regions, we selected three provinces through purposive sampling: Zhejiang Province (in eastern China with a gross domestic product (GDP) of RMB 7351.6 billion in 2021, ranked 4th nationally),27 Guizhou Province (in western China with a GDP of RMB 1958.6 billion in 2021, ranked 22nd nationally)28 and Henan Province (in central China with a GDP of RMB 5888.7 billion in 2021, ranked 12th nationally).29

Step 2: we used quota sampling to recruit HCPs. In each province, we selected 18–20 primary medical institutes. At each institute, we obtained 10–30 samples. Six to eight secondary medical institutes per province were selected and 30–50 samples per institute were obtained. Three to five tertiary medical institutes per province were selected and 70–90 samples per institute were obtainted. We invited one to three key HCPs from each medical institute to act as coordinators for the online survey, using communication tools such as WeChat Group and Ding Talk, which are Chinese internet-based communication tools widely used for work purposes. Due to the impact of the COVID-19 pandemic, some medical institutes collected fewer questionnaires than anticipated.

Inclusion criteria

  1. Individuals aged 18–65 years.

  2. HCPs who have obtained the necessary qualifications and certificates.

  3. In-service personnel at medical institutes.

  4. Volunteers for this study who provided informed consent.

Exclusion criteria

  1. Participants who were on leave or sick leave for more than 2 weeks at the time of the questionnaire survey.

  2. Hospital administrative staff.

The questionnaire survey commenced on 29 September 2022 and concluded on 18 January 2023 after the WHO declared the COVID-19 outbreak a public health emergency of international concern. We conducted our survey in parallel with a large survey of HCPs with a sample size of 3000. We also estimated the sample size based on the prevalence estimate of US (0.45) and a margin of error of 0.013. The result was 2870, which met the minimum standards. Ultimately, 2996 questionnaires were collected, of which 2976 were valid.


The questionnaire encompassed the following information.

Sociodemographic measures

Sociodemographic data included age, sex, marital status, education level, the level of the medical institute, the location of institutes across regions, specialisation, professional title and annual income level.

Uncertainty stress (independent variable)

We assessed US using the Life Stress Questionnaire30 compiled by Professor Yang Ting Zhong at Zhejiang University. This questionnaire is widely used in surveys involving community populations and college students’ mental health. Respondents rated their US over the past 4 weeks on a 5-point Likert scale (1=never, 2=almost never, 3=sometimes, 4=often and 5=frequent). The scale included four phrases: ‘Life is subtle, destiny is unpredictable’, ‘Values of confusion’, ‘The world is changing too fast to keep up with’ and ‘I do not know how to achieve my goals’. The total scores ranged from 4 to 20 (with cut-off scores of 0–12 indicating low US and 13–20 denoting high US). The Cronbach’s α coefficient for this scale was 0.81.

Depression (outcome variable)

We assessed depression using the Patient Health Questionnaire-9 (PHQ-9), which is used to diagnose both major and subthreshold depressive disorders in the general population.31 32 This scale comprises nine items assessing the severity of depressive symptoms over the past 2 weeks. We used a 4-point severity scale (0=not at all to 3=nearly every day). The total scores ranged from 0 to 27, with cut-off scores as follows: 0–4=minimal or none, 5–9=mild, 10–14=moderate, 15–19=moderately severe and 20–27=severe. The Cronbach’s α coefficient for this scale was 0.84. We categorised depression as a binary variable, either present or absent, with a PHQ-9 score ≥5 indicating the presence of depression.

Statistical analyses

We imported all data into SPSS V.26.0 and ensured data accuracy through double entry and verification. We performed all statistical analyses using SPSS V.26.0, with statistical significance set at p<0.05. We included missing data in the analysis and marked them as ‘NA’ (not applicable) in SPSS V.26.0, with missing data accounting for less than 1%. We began with a descriptive analysis of sociodemographic traits, US and depression. We also described the proportions of high US and depression across different sociodemographic traits. Subsequently, we employed a χ2 test to compare the differences in the proportions of high US and depression based on sociodemographic traits. To account for sociodemographic factors, we employed a binary logit model to analyse the impact of sociodemographic traits on US (model 1). We analysed another binary logit model to assess the relationship between US and depression while controlling for sociodemographic traits (model 2). Finally, we performed multiple linear regression analysis to examine the relationship between US and depression while controlling for sociodemographic traits (model 3).

Patient and public involvement

The patients and the public were not involved in the design, conduct, reporting or dissemination of our research plans.


Sociodemographic traits, US and depression of HCPs

The analysis encompassed 2976 HCPs across China, including doctors, nurses and other specialists. Approximately one in five HCPs has experienced high levels of US (26.54%), whereas more than seven in ten exhibit symptoms of depression (71.63%). We found significant differences (p<0.05) in the proportion of US among HCPs across education levels (χ2=12.34, p=0.006), regions (χ2=50.11, p<0.001) and annual income levels (χ2=8.06, p=0.045) (table 1).

Table 1

Univariate analysis of US and depression across sociodemographic traits among HCPs (N=2976)

We also noted significant disparities in the prevalence of depression among HCPs across education levels (χ2=59.93, p<0.001), regions (χ2=19.87, p<0.001), levels of medical institutes (χ2=15.85, p<0.001) and professional title (χ2=21.33, p<0.001) (table 1).

US and depression: a correlational analysis

Table 2 presents the results of the regression analyses. In model 1, the findings imply that HCPs with graduate degrees, lower annual incomes or those in central China and at primary and secondary institutes are more likely to experience US. HCPs with graduate degrees were 83% more likely to experience high US than those with high school degrees or below (OR: 1.83; 95% CI 1.07 to 3.11; p<0.05). In comparison to HCPs in eastern China, those in central China were 75% more likely to experience high US (OR: 1.75; 95% CI 1.36 to 2.24, p<0.01). HCPs from primary institutes were 33% more likely to experience high US than those from tertiary institutes (OR: 1.33; 95% CI 1.03 to 1.72; p<0.05). The probability of high US among those from secondary institutes was 30% greater (OR: 1.30; 95% CI 1.01 to 1.68; p<0.05). Additionally, HCPs with annual incomes of less than ¥50 000 had an 85% greater likelihood of experiencing high US than those with annual incomes exceeding ¥150 000 (OR: 1.85; 95% CI 1.26 to 2.73; p<0.01). The probability of high US among those with annual incomes between 50 000 and 99 999 was 49% higher (OR: 1.49; 95% CI 1.10 to 2.03; p<0.05) (table 2).

Table 2

Results of regression analysis of US and depression among HCPs

Adjusted logistic regression modelling (model 2) indicated that HCPs with high US were 5.02 times more likely to experience depression than those with low US (adjusted OR: 5.02; 95% CI 3.88 to 6.50; p<0.01). Furthermore, multiple linear regression revealed that the likelihood of depression increased by 1.32 points for every 1-point increment in US (B: 1.32; 95% CI 1.25 to 1.39; p<0.01) (model 3) (table 2).


Previous studies on US among HCPs have primarily focused on stress resulting from changes in healthcare systems and diagnostic uncertainty.33 In recent years, the academic community has shown increased concern about the US related to pandemics, including concerns about acquiring knowledge about novel pandemics, securing protective resources,21 and coping with personal and financial situations.34 Unfortunately, previous studies have not quantitatively measured the prevalence of US among HCPs. In our study, we found that 26.54% of HCPs had experienced US. During the same period, the prevalence of US among the general Chinese population was 45.3%.35 Before the outbreak of COVID-19, 11.4% of urban residents reported US.30 US among HCPs during the COVID-19 pandemic was lower than among other kinds of citizens during the same period. Hummel et al surveyed medical and non-medical professionals in eight European countries, with medical professionals exhibiting less mental stress than their non-medical counterparts.36 Indeed, HCPs with more knowledge and real-time information on COVID-19 may have experienced reduced uncertainty during the pandemic.37 However, they were impacted by long-term factors, including the uncertainty of making a diagnosis and a changing medical environment, potentially making them more susceptible to US than the general public in the long run.

US was prevalent among HCPs across various kinds of medical institutes (table 1). Primary and secondary hospitals offer limited promotional opportunities and further education prospects, leading to uncertainty regarding career development.38 Tertiary hospitals providing tertiary care present more challenging patient cases, resulting in heightened stress stemming from uncertainty. Moreover, the growing trend of ‘physicians performing more endoscopic procedures and surgeons performing more minimally invasive procedures’ poses significant challenges to doctors in tertiary hospitals.39 In addition, HCPs in China face the extremely serious threat of medical violence.22 40 This pervasive threat may have further exacerbated the overall US experienced by HCPs in China. However, HCPs in primary and secondary hospitals had a greater likelihood of experiencing high levels of US (table 2). This may be because almost every HCP in primary and secondary hospitals must perform nucleic acid testing for COVID-19 for residents, which significantly increases the uncertainty of occupational exposure. The significant regional disparities observed may be linked to variations in medical resource distribution and the varying severity of the COVID-19 pandemic across different regions.

Moreover, the relatively higher US among graduates aligns with common observations. HCPs with graduate degrees often experience significant stress while performing scientific tasks. Further, HCPs with annual incomes below ¥50 000 demonstrated somewhat higher levels of US, a finding consistent with that of a previous study,41 indicating that this group may experience heightened financial concerns.

Depression among HCPs has gradually become a research hotspot with numerous studies highlighting the psychological challenges that they face.2–4 9 42 Chinese HCPs have reported a notably high prevalence of psychological problems. For instance, Zhou et al investigated 606 first-line hospital staff members and 1099 individuals from the general population, revealing depression prevalence rates of 57.6% and 47.6%, respectively (p<0.001, OR=1.54).9 Additionally, a meta-analysis confirmed a global prevalence of depression among HCPs during the COVID-19 pandemic at 21.7% (95% CI 18.3% to 25.2%).8 In this study, the share of depression (PHQ-9: ≥5) stood at 71.63%, surpassing both the global prevalence rate and the findings of prior domestic studies. This heightened prevalence may be attributed to changes in COVID-19 prevention policies and the rapid increase in infection rates during the survey period, which may have affected the mental health of HCPs. These results underscore the severity of depression among Chinese PHPs. Among sociodemographic variables, the relationship between education and depression is of particular interest, with varying findings in the literature. Bracke et al examined 28 288 individuals across 21 European countries and found that academic qualifications had positive effects on mental health in most European states, although the benefits were limited or absent in other nations.43 In contrast, Shi and Yang, using data from the China Family Panel Studies, found that individuals with a master’s degree or higher were more susceptible to depression.44 Our study suggests that the probability of depression increases with a higher education level among Chinese HCPs, which is in line with the work of Shi and Yang, mentioned earlier. A rich body of literature has substantiated the relationship between professional title and mental health.45–47 The present study revealed that HCPs with intermediate professional titles had a significantly greater proportion of depression than those with junior professional titles, which is consistent with past research.45 47 This association may be attributed to HCPs with intermediate professional titles often having more frequent night shifts and greater responsibilities in medical aid tasks. Hence, special attention should be directed toward the mental health of HCPs in central China, those with higher education levels, and those with intermediate professional titles.

Our study also revealed a robust correlation between US and depression in HCPs. Notably, the US of HCPs appeared more closely related to depression compared with previous research involving university students. This phenomenon can be attributed to the unique circumstances within hospitals and the nature of HCPs’ work. Hospitals serve as windows that reflect broader societal and public health issues, including the unequal distribution of medical resources and economic disparities. Furthermore, uncertainties associated with making a diagnosis, the ongoing COVID-19 pandemic, and the threat of medical violence contributed to the negative psychological well-being of HCPs.

The relationship between US and depression can also be explained physiologically. As proposed by Peters et al, individuals experiencing uncertainty and threats enter a state of heightened alertness in response to environmental changes, invoking stress to swiftly reduce uncertainty. With repeated exposure to challenging environments, one’s stress response mechanism may become overactivated, resulting in harm to both the body and brain (overload). This overload can lead to neuroendocrine and cardiovascular changes as well as negative emotions. These pathophysiological shifts impair the brain’s ability to cope with uncertainty and create a negative feedback loop.48 This concept is supported by previous research indicating that individuals residing in unstable, insecure environments are at greater risk of experiencing depression and cognitive impairment.49 50

In light of the abovementioned evidence, it is imperative to address the issue of US among HCPs; alleviating US may ameliorate depression among them. At the government level, efforts should focus on strengthening the oversight of medical violence, balancing the distribution of medical resources and improving HCPs’ incomes and personal security. Hospital management should pay more attention to stress relief for HCPs with graduate degrees and low incomes. At the individual level, raising awareness of the presence of US, ensuring access to the latest therapeutic guidelines and showing concern for occupational exposure may aid in relieving US and depression.

It is important to note that this study was a cross-sectional survey; as such, we could not establish causal relationships. Future research should employ interventions to track changes throughout respondents’ lives. We did not conduct qualitative research or follow-up with HCPs through structured interviews on US, which may provide insights into the direction of interventions. There may have been selection bias due to the sampling method. We gathered the information based on questionnaires, thus recall bias is inevitable. Finally, the conclusions drawn are specific to HCPs and may have limited external validity as we primarily focused on an Asian-dominant population.

This post was originally published on