Skip to main content

The determinants of mental health inequalities between Chinese migrants and non-migrants during the Shanghai 2022 lockdown: a Blinder-Oaxaca decomposition



The mental health inequality between migrants and non-migrants was exacerbated by the COVID-19 pandemic. Identifying key determinants of this inequality is essential in promoting health equity.


This cross-sectional study recruited Shanghai residents by purposive sampling during the city-wide lockdown (from April 29 to June 1, 2022) using an online questionnaire. Migration statuses (non-migrants, permanent migrants, and temporary migrants) were identified by migration experience and by household registration in Shanghai. Mental health symptoms (depression, anxiety, loneliness, and problematic anger) were assessed by self-report scales. The nonlinear Blinder-Oaxaca decomposition was used to quantify mental health inequality (i.e., differences in predicted probabilities between migration groups) and the contribution of expected correlates (i.e., change in predicted probability associated with variation in the correlate divided by the group difference).


The study included 2738 participants (771 [28.2%] non-migrants; 389 [14.2%] permanent migrants; 1578 [57.6%] temporary migrants). We found inequalities in depression (7.1%) and problematic anger (7.8%) between permanent migrants and non-migrants, and inequalities in anxiety (7.3%) and loneliness (11.3%) between temporary migrants and non-migrants. When comparing permanent migrants and non-migrants, age and social capital explained 12.7% and 17.1% of the inequality in depression, and 13.3% and 21.4% of the inequality in problematic anger. Between temporary migrants and non-migrants, age and social capital also significantly contributed to anxiety inequality (23.0% and 18.2%) and loneliness inequality (26.5% and 16.3%), while monthly household income (20.4%) and loss of monthly household income (34.0%) contributed the most to anxiety inequality.


Significant inequalities in depression and problematic anger among permanent migrants and inequalities in anxiety and loneliness among temporary migrants were observed. Strengthening social capital and economic security can aid in public health emergency preparedness and promote mental health equity among migrant populations.


The COVID-19 pandemic created an increased global burden of mental disorders. An estimated 53.2 million cases of major depression (an increase of 27.6%) and 76.2 million cases of anxiety (an increase of 25.6%) resulted from the COVID-19 pandemic globally [1]. Addressing mental health inequality is a key goal for global mental health and sustainable development, especially during the pandemic [2]. Following the definition of health inequality, mental health inequality is the avoidable difference in mental health status and determinants between disadvantaged and more advantaged social groups [3].

It is well-established that migration is related to disadvantaged circumstances and mental health outcomes. Compared to non-migrants, migrants encounter greater exposure to stressors and to the social determinants known to exacerbate mental health problems, and therefore, may experience worse mental health [4]. Chinese citizens who are living in another city other than the city where they have household registration (hukou in Chinese, managed at the municipal level) are classified as internal migrants [5]. Since the hukou affiliation is linked with the provision of public social services (e.g., housing, education, job benefits, health insurance, employment) [6], and marginalization, isolation, and discrimination [7], internal migrants without a hukou (temporary migrants) are more vulnerable to mental health problems than migrants with a hukou (permanent migrants) [8]. Individuals with higher socioeconomic status (SES), including higher education and income, are likely to acquire a hukou in their migration destination, but they may face additional stressors such as purchasing property [9]. Thus, we hypothesized that differences in key social determinants, including SES, may lead to heterogeneity among internal migrants in mental health.

The current literature is limited since the majority of evidence on mental health inequalities between internal migrants and non-migrants relies on studies with simple classification based on their current place of usual residence, regardless of their migration experience [10, 11]. Our previous work among middle-aged and elderly migrants in China demonstrated that different types of internal migrants have heterogenous mental health outcomes. We found a significantly lower incidence of depression in permanent migrants than non-migrants [12]. However, there remains limited evidence from younger migrants, which includes a larger proportion of internal migrants than the middle-aged and elderly [13], and it is unclear whether mental health inequality among people with different migration statuses was exacerbated by the scarcity of public service resources during pandemic-era lockdowns.

In addition, although previous studies document the association between social determinants and mental health (e.g., education, employment, social capital, physical health, and household factors) [14, 15], few studies have quantified and compared the contributions of these factors to mental health inequality. Identifying the determinants of differences in health status is critical for population-based mitigation strategies, and in turn contribute to achieving the 2030 Agenda for Sustainable Development Goal to leave no one behind [16].

Shanghai is the largest city in China and has one of the largest internal migrant populations − 10.5 million - which accounts for 42.1% of its total population [17]. We conducted an online cross-sectional study among Shanghai residents with different migration statuses (non-migrants, permanent migrants, and temporary migrants) during the sudden city-wide COVID-19 lockdown in 2022. This study aimed to (1) compare the mental health conditions among Chinese adult migrants and non-migrants during the lockdown, and assess possible mental health inequalities between groups, and (2) identify the major social determinants of mental health inequality observed between migrants and non-migrants.


Study design and participants

A cross-sectional survey was conducted in Shanghai during the COVID-19 lockdown in 2022 (from April 29 to June 1). An unexpected, unprecedented and complete citywide lockdown was implemented among the entire population of 24.9 million from March 28, 2022 to June 1, 2022. In this survey, Chinese adult citizens (≥ 18 years old) living in 16 administrative districts of Shanghai were recruited by purposive sampling. International migrants were not included in this study, as the hukou system only applies to Chinese citizens. The survey company Wenjuanxing randomly recruited individuals via WeChat with the aim for 200 persons per district. Sampling was concluded when the quota or the end of the lockdown was reached. Participants completing the survey received a 6 Chinese Yuan incentive (approximately 1 USD). Network IP address was checked and three filter questions with predetermined options were included to enhance the data quality of the online questionnaire. The questionnaire was brief and completed in less than 15 min. A total of 3,230 individuals provided valid responses. Ten participants with data missing on sex and 407 students (who are not included in China Migrants Dynamic Survey definition of migrants [18] and had stable supplies from universities or at home during the lockdown) were excluded from our study.

Additionally, we excluded 75 elderly participants (aged 60 and above) from the analyses since we were comparing migrant to non-migrants who were working age and living in Shanghai. Elders may not represent a valid group for these study comparisons. The final sample consist of 2,738 participants. All respondents completed the consent form and explicitly agreed to participate. The study was approved by the Institutional Review Board of the NYU Shanghai (2022-008-NYUSH).


Mental health outcomes

Mental health outcomes included depression, anxiety, loneliness, and problematic anger, which are the most common mental health problems reported during the pandemic [19].

Depression symptoms in the past two weeks were measured by the Patient Health Questionnaire (PHQ-9), one of the most widely used screening instruments. It has nine items with each scoring 0–3 and a total score ranging from 0 to 27, where score of 10 or above indicates possible major depression [20]. Thus, the indicator of depression was coded as “0” = no (scoring 0–9), “1” = yes (scoring 10 or above). The Cronbach’s alpha in the current study was 0.89.

Anxiety symptoms in the past two weeks were measured by the seven-item Generalized Anxiety Disorder Scale (GAD-7). Participants answered on a four-point Likert scale (0 = not at all, 3 = nearly every day). A summed score of 0–21 was generated to illustrate the severity of symptoms, with a higher score indicating worse severity and a cut-off of 10 for moderate to severe anxiety [21]. Therefore, anxiety was coded as “0” = no (scoring 0–9), “1” = yes (scoring 10 or above). The Cronbach’s alpha in the study was 0.92.

Loneliness was assessed by the 3-item version of UCLA Loneliness Scale. The UCLA Loneliness Scale was used widely in Chinese migrants. Participants were asked about the frequency they feel lack of companionship, left out, and isolated from others. Each item was rated on 3-point scale (1 = hardly ever or never, 3 = almost always), giving a total score of 3–9, which was dichotomized into “0” = not lonely (scoring 3 to 5) and “1” = lonely (scoring 6 or above) [22]. The Cronbach’s alpha in the current study was 0.88.

The Dimensions of Anger Reactions (DAR-5) was applied to assess anger experiences in the past four weeks. It is composed of four anger response parameter items (including frequency, intensity, duration, and antagonism) as well as one item for social relationship impairment. Each item was answered on a 5-point Likert scale, giving a score range from 5 to 25. A score of 12 or above was coded as “1” = problematic anger, or “0”= no otherwise [23]. The Cronbach’s alpha in the current study was 0.79.

Migration status

In this study, migration status was ascertained by both migration experience and hukou status to capture the heterogeneity within the migrant population by asking: “Were you born in Shanghai” and for participants not born in Shanghai, “Do you have a Shanghai hukou”? The former condition was used to separate non-migrants and migrants, where non-migrants were those born in Shanghai with a Shanghai hukou. The latter condition was employed to further divide migrants into “permanent migrants” and “temporary migrants”, defining permanent migrants as those who changed their hukou to Shanghai, and temporary migrants as those who live in Shanghai without a Shanghai hukou.

Social determinants

Based on previous studies [5, 15], we collected the following information as social determinants of mental health, including:

  1. 1)

    Individual demographics: age, sex, education level [secondary or lower, high school, college or higher], marital status [single, married/cohabitating, widowed/divorced], and employment status [yes/no]).

  2. 2)

    Household factors: living arrangement [living alone or not], monthly household income [≤ 4,000 Chinese Yuan, 4,001–8,000, 8,001–15,000, 15,001–30,000, ≥ 30,001], loss of monthly household income during the lockdown [none, < 50%, ≥ 50%], and partner violence) [yes/no]. Intimate partner violence was screened by the Hurt, Insult, Threaten, And Scream Scale (HITS) in married/partnered participants. The HITS screening for four types of common domestic violence on a 5-point Likert scale, and a cut-off point of 10 was used to identified possible violence (0 = no, 1 = yes) [24]. The Cronbach’s alpha in the current study was 0.95.

  3. 3)

    Health-related factors: COVID-19 infection in lockdown [yes/no], chronic disease [yes/no], self-rated health [relatively bad or not], past diagnosis of mental illness [yes/no], hazardous drinking[yes/no], and smoking during the lockdown [yes/no]. Hazardous drinking was evaluated with two items from the alcohol use disorders identification test. The frequency and quantity of drinking was assessed and each item was scored 0–4. A score above 0 on either item indicates a hazardous drinking [25].

  4. 4)

    Social capital was the sum of trustworthy network connections that empower individuals and consisting of structural, cognitive, bonding, bridging, and linking components. Bonding capital reflects individuals’ capacity to bond with others within their own community. Bridging capital indicates individual’s ability to connect with people from different communities or social identities. In this study, social capital was assessed by subscales (bonding social capital and bridging social capital) from the Chinese version of Revised Personal Social Capital Scale (PSCS-R). Responses to each question were scored on a 5-point Likert scale (1 = none, 5 = all). The Cronbach’s alpha in the current study was 0.80. Items were summed to create a total score ranging from 8 to 40, with higher scores indicating better social capital. This scale was divided into quartiles in this study. The subscales of PSCS-R are available in appendix (Table S1).

Statistical analysis

The participants’ characteristics were described by mean ± standard deviation (SD) and frequency (proportion) of mental health outcomes and social factors. To examine the unadjusted differences in mental health outcomes among the participants with different migration statuses, we used Chi-square tests for categorical variables and Dunnett-t tests for continuous variables.

Bivariate and multivariable logistic regressions were employed to assess the unadjusted and adjusted associations (adjusting for individual demographics, household factors, health-related factors, and social capital) between migration statuses and mental health, with non-migrants as the reference group.

When significant migration-status-based group difference in mental health outcome was found in in the adjusted analysis, determinants of inequality were further identified by Blinder-Oaxaca (BO) decomposition [26]. The BO decomposition is a statistical tool that quantifies the extent to which an individual determinant explains a health inequality. The method has been widely used in epidemiology and health-related studies, but with few applications to mental health [27,28,29].

We used the nonlinear BO method in this study as the mental health outcomes were binary variables [30]. The BO decomposition analyses included two phases. Firstly, we calculated the predicted probability of mental health problems by estimating a logistic regression model for each group and then the difference in predicted probabilities between the migrant groups (i.e., permanent, or temporary migrants) and non-migrants was used to quantify the mental health inequality.

Secondly, we quantified the contribution of each social determinant, including individual demographics, household factors, health-related factors, and social capital to the inequality. In this step, we calculated the change in the predicted probability (absolute contribution) from replacing one social determinant of the reference group (non-migrants) with that of a migrant group (permanent or temporary migrants), while fixing the distribution of other determinants in the model. The proportion of explanation (relative contribution) was calculated by dividing the change in predicted probability by the group-difference in predicted probabilities. The sign of the proportion of explanation indicates the direction of the contribution, where a negative sign indicates a negative contribution, meaning that removing the intergroup differences in this factor will increase the inequality, and vice versa. BO decomposition’s statistical method was elaborated in the appendix (p3).

R 4.1.1 and STATA 11 (Stata Corp, College Station, TX) were used to perform statistical analysis. All analyses used a significance level of 0.05.

Role of the funding source

The funding source played no role in the study’s design, data collection, data analysis, data interpretation, report writing, or the decision to submit the manuscript for publication.


Descriptive statistics across three migration-status-based groups are presented in Table 1. A total of 2,738 Shanghai residents were included in the data analysis, including 771 (28.2%) non-migrants, 389 (14.2%) permanent migrants, and 1578 (57.6%) temporary migrants. Both permanent and temporary migrant groups were younger and presented a lower rate of chronic diseases yet had a higher rate of hazardous drinking than non-migrants (permanent:36.0% and temporary: 29.7% vs. non-migrants: 25.3%). Compared to non-migrants, permanent migrants on average had higher rates of completing college or higher level of education (permanent: 88.2% vs. non-migrants: 82.1%), employment (92.5% vs. 91.1%), more monthly household income (≥ 30,001 Chinese Yuan: 20.3% vs. 12.6%), and greater social capital (P75-P100: 30.3% vs. 23.5%). On the contrary, temporary migrants were characterized by lower education levels than non-migrants (college or higher level, temporary: 47.9% vs. non-migrants: 82.1%), lower employment (80.9% vs. 91.1%), much lower household income (≤ 4000 Chinese Yuan: 10.8% vs. 6.2%), and less social capital (P75-P100: 14.5% vs. 23.5%). Notably, temporary migrants endured significantly greater loss of income (> 50%: 51.9% vs. 14.9%) and a higher prevalence of COVID-19 infection (5.5% vs. 3.5%) than non-migrants during the lockdown.

Table 1 Characteristics of participants with different migration statuses

The total prevalence of depression, anxiety, problematic anger, and loneliness was 27.3%, 21.0%, 37.4%, and 33.0%, respectively, but participants with different migration statuses showed different prevalences of mental health problems. In general, mental health problems were more prevalent among both permanent and temporary migrants than non-migrants. In comparison with non-migrants (Table 1, depression: 20.6%, anxiety: 16.2%, problematic anger: 32.8%, loneliness: 26.1%), temporary migrants had statistically significantly higher prevalences of depression (30.5%), anxiety (23.5%), problematic anger (38.8%), and loneliness (37.4%), while permanent migrants only had statistically significantly higher prevalences of depression (27.8%) and problematic anger (40.6%). Statistically significant differences in loneliness were only found in temporary migrants (37.4%) when compared to permanent migrants (28.8%). Similar associations were observed in bivariate logistic regression analysis. After adjusting for covariates, permanent migrants continued to have a higher prevalence of depression (adjusted OR 1.60 [95% confidence interval (CI) 1.16–2.20]) and problematic anger (1.45 [95% CI 1.10–1.92]) than non-migrants. Temporary migrants had an increased odds of anxiety (1.31 [95% CI 1.02–1.74]) and loneliness (1.47 [95% CI 1.16–1.85]) in adjusted models (Fig. 1).

Fig. 1
figure 1

The adjusted associations between migration status and mental health

Note: The non-migrants were set as reference group. The associations were adjusted by sets of covariates, which including individual demographics (age, gender, education level, marital status, and employment status), household factors (living arrangement, monthly household income, loss of monthly household income, and partner violence), health-related factors (COVID-19 infection in lockdown, chronic disease, self-rated health, past diagnosis of mental illness, hazardous drinking, and smoking during the lockdown), and social capitals. aOR, adjusted odds ratio

Inequalities in mental health among migrants and non-migrants were further quantified by nonlinear BO decompositions (Figs. 2 and 3). Specifically, significant differences of 7.2% (27.8% minus 20.6%, p = 0.008) and 7.8% (40.6% minus 32.8%, p = 0.019) in predicted probabilities of depression and problematic anger were observed between permanent migrants and non-migrants (reference group). Significant anxiety and loneliness inequalities were observed between temporary migrants and non-migrants as 7.3% (23.5% minus 16.2%, p = 0.049) and 11.3% (37.4% minus 26.1%, p = 0.013) respectively.

Fig. 2
figure 2

Significant contributors to mental health inequality between permanent migrants and non-migrants

A: Absolute contribution for each contributor; B: Relative contribution (proportion of explanation, which equals the value of the absolute contribution divided by net difference) for each contributor. Permanent migrants had a 7.2% (27.8% minus 20.6%, p = 0.008) and 7.8% (40.6% minus 32.8%, p = 0.019) higher predicted probability of depression and problem anger than non-migrants

Fig. 3
figure 3

Significant contributors to mental health inequality between temporary migrants and non-migrants

A: Absolute contribution for each contributor; B: Relative contribution (proportion of explanation, which equals the value of the absolute contribution divided by net difference) for each contributor. Temporary migrants had 7.3% (23.5% minus 16.2%, p = 0.049) and 11.3% (37.4% minus 26.1%, p = 0.013) higher predicted probability in anxiety and loneliness than non-migrants

The quantified contribution of the social determinants to mental health inequalities between permanent migrants and non-migrants is shown in Fig. 2. In Fig. 2A, the magnitude of difference corresponding to each determinant indicates the absolute difference in predicted probabilities statistically explained by each factor (absolute contribution), while the proportion of explanation (relative contribution) was shown in Fig. 2B. For depression, the largest magnitude of the inequality was explained by age, followed by monthly household income, and social capital. Specifically, if permanent migrants had the same distribution in age and household income as non-migrants, the predicted probability of depression would decrease by 0.9% (SD = 0.3%; proportion of explanation: 12.7%) (p < 0.05), 0.7% (SD = 0.4%; 10.2%) (p < 0.05), respectively. On the contrary, the probability of depression would increase by 1.2% (SD = 0.3%; -17.1%) (p < 0.05) if permanent migrants reduced their social capital to the same level as non-migrants.

For problematic anger, similar patterns of contribution were observed for age and social capital. If permanent migrants had the same distribution in age, the probability of problematic anger would decrease by 1.0% (SD = 0.3%; proportion of explanation: 13.3%) (p < 0.05), and would increase by 1.7% (SD = 0.4%; -21.4%) (p < 0.05) if social capital is the same as non-migrants.

Significant determinants of mental health inequalities between temporary migrants and non-migrants are shown in Fig. 3. For anxiety, age, education level, employment status, loss of income, household income, self-rated health, and social capital significantly contributed to the inequality (p < 0.05). The predicted probability of anxiety would be lowered by 2.5% (SD = 0.7%; proportion of explanation: 34.0%), 1.7% (SD = 0.5%; 23.0%), and 1.4% (SD = 0.3%; 18.2%) if temporary migrants had the same distribution of loss of household income, age, and social capital as non-migrants. In contrast, the probability would increase by 1.8% (SD = 0.6%; -24.2%) and 1.5% (SD = 0.5%; -20.4%) if the two groups had the same distribution in education levels and monthly household income.

For loneliness, the predicted probability would lower by 3.0% (SD = 0.7%; 26.5%), 1.9% (SD = 0.3%; 16.3%), and 1.8% (SD = 0.8%; 16.1%) if temporary migrants had the same distribution of age, social capital, and loss of household income as non-migrants, and it would increase by 2.2% (SD = 0.7%; -19.4%) and 1.1% (0.4%, -10.0%) for education level and chronic diseases (p < 0.05).


The COVID-19 pandemic, a once-in-a-century public health emergency, exposed and amplified existing health inequalities. Unlike relative normalcy in 2021, the 2022 Shanghai lockdown was abrupt and lasted for months, and left residents inadequately prepared and deficient in essential supplies [31]. This was especially true among marginalized populations like migrants. This study demonstrated mental health inequality among Chinese citizens aged 18 to 59 (except for students) with varied migration statuses and estimated the effects of key social determinants that account for these inequalities. Mental health inequalities were found between non-migrants and internal migrants, but prominent mental health problems differed between permanent and temporary migrants. Several social determinants, including age, household income and income loss, education level, and employment status, and social capital were key contributors to mental health inequalities.

We observed higher estimated prevelances of mental health problems among permanent and temporary migrants compared to non-migrants. There were also differences in the source of mental health burdens among permanent and temporary migrants, where anxiety and loneliness showed higher prevalence in temporary migrants while depression and problematic anger were more prominent among permanent migrants when compared to non-migrants. These differences might result from the varied distribution of common causes of these mental health problems among permanent and temporary migrants. In this study, temporary migrants faced the greatest uncertainty during the lockdown. More than half (51.9%) of them experienced a loss in half their household income, and 19.1% became unemployed due to the lockdown. The major life changes and uncertainty are common causes of both loneliness and anxiety [32]. Meanwhile, temporary migrants also suffered more from financial concerns [33], stressful events, and employment difficulties [34], which are common correlates of anxiety. Overall, this study contributes to the literature on mental health inequality by demonstrating the importance of dividing migrants into groups that reflect their underlying vulnerabilities to mental health problems.

Moreover, we found several key social determinants contributed to mental health inequalities. Age and social capital were significant contributors, but the direction for the two indicators were opposite across the four types of mental health outcomes between non-migrants and migrants. Younger migrants had greater mental health problems in the current study and this aligns with prior studies during COVID-19 [35, 36]. This phenomenon might be explained by more social media use among the youth [37]. We found that younger participants faced more negative social media exposure, such as rumors or insults against Shanghai residents, in our survey (appendix Figure S1). In future public health emergencies, attention should be given to the younger populations, with increased dissemination of positive social media information to this group.

Social capital is an important determinant of inequalities in mental health [5, 38,39,40]. Consistent with previous studies on Chinese internal migrants [38], temporary migrants reported poorer social capital than non-migrants. Studies on international migrants also support that migration is related to poor social capital [39]. Poor social capital may be associated with marginalization due to registration status (hukou status for Chinese internal migrants) and difficulties for migrants in developing social networks and mutual trust in their migration destination [41, 42]. The difference in social capital was positively associated with both the inequality in anxiety and loneliness between temporary migrants and non-migrants. Contrary to previous studies [5, 40], better social capital was observed in permanent migrants compared to non-migrants. Generally, migrants with higher social capital were more likely to have local hukou and be permanent migrants [9]. Additionally, we found that the difference in social capital was negatively associated with both the inequality in depression and problematic anger between permanent migrants and non-migrants. In other words, lower reported social capital was associated with greater mental health inequalities. Therefore, a possible interventional pathway to reduce mental health inequality is to enhance the social network and social capital of migrant groups and provide bridges within and between migrants and local residents.

Contrary to expectations, household income was positively associated with inequality in depression between permanent migrants and non-migrants, suggesting higher income was related to higher risks for mental health problems. It was reported that household income is associated with mental health in an inverse-U shape, where extremely high income is positively related with risk of depression [43]. Another possible reason is that migrants with higher income have higher expectations regarding their quality of life, which is harder to achieve because of the lockdown [44]. This suggests that high-income groups also require attention for psychological counseling support. In addition, previous work showed that the contribution of income to health outcomes varies among different socioeconomic groups, suggesting that income might not be a reliable indicator [45]. Notably, temporary migrants endured the most loss of income during the lockdown, and compared to income, loss of household income contributed more to mental health inequalities. This finding supports previous research [46] showing that income loss is a more effective determinant of mental health than increased income, especially for those at a lower level of income. Thus, it is important to ensure financial security of temporary migrants during public health emergencies, governments and social organizations can provide assistance such as unemployment benefits to help temporary migrants cope with crises.

Among the strengths of this study was our nuanced approach to classifying migrants, separating permanent migrants from “non-migrants” by combining hukou status and migration experience, which reduced misclassification bias. Another strength is that we quantitatively assessed the social determinants of mental health inequalities. We identified key social determinants that if modified, may reduce mental health inequalities during public health emergencies, providing evidence to support targeted migration-policy.

However, there were some limitations for this study. Limited by the lockdown requirements, an online cross-sectional survey with purposive sampling was practical, though several limitations might be introduced by the study design. First, the cross-sectional design limited causal inference on the impact of the lockdown, but it is worth noting that the social determinants of mental health explored in this research, such as social capital, tended to exhibit relatively stable patterns before and after the lockdown. Moreover, the study incorporated variables related to changes induced by the lockdown, such as unemployment and income reduction. Second, this study was based on the online survey with a non-probabilistic sampling, which might influence representativeness and generalizability. We only included participants aged below 60 to increase the representativeness of migrants and people who exposed to social media. The age and sex distribution of the analytic sample aligns with census [47] and nationwide surveys [48] of internal Chinese migrants. We found a lower total prevalence of mental health problems than was observed in previous national studies [49, 50]. This could be because these studies were conducted during the early stage of the pandemic, when the virus was more toxic and the public knew little about it. Third, there will be inevitable reporting bias as self-report questionnaires were used although we used standardized and validated scales with excellent psychometric properties, and the psychometric properties of self-assessment instruments used in this study have been validated in Chinese population [51, 52]. Additionally, community-level determinants, such as supplies provided by communities in the lockdown, might impact mental health but were not included in this study.


Significant but heterogeneous inequalities in mental health were found between Shanghai residents with varied migration statuses during an unprecedented COVID-19 lockdown. Compared to non-migrants, mood disorders including depression and problematic anger were higher among permanent migrants while anxiety and loneliness among temporary migrants were of greater concern. Better social capital and economic security contributed significantly to reduce these inequalities. Vulnerability among migrants was seen during the pandemic, and interventions to improve the social determinants of health among the migrant community should be a priority for future policies, and could contribute to achieving Sustainable Development Goal 3 and 10 (improve health and wellbeing & reduce inequality). Population-based interventions aimed at strengthening social capital and economic security (such as unemployment benefits) would likely promote mental health equity between internal migrants and non-migrants during times of crisis, such as this public health emergency.

Data availability

No datasets were generated or analysed during the current study.


  1. Santomauro DF, Herrera AMM, Shadid J, Zheng P, Ashbaugh C, Pigott DM, Abbafati C, Adolph C, Amlag JO, Aravkin AY. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. Lancet. 2021;398:1700–12.

    Article  Google Scholar 

  2. Patel V, Saxena S, Lund C, Thornicroft G, Baingana F, Bolton P, Chisholm D, Collins PY, Cooper JL, Eaton J, et al. The Lancet Commission on global mental health and sustainable development. Lancet. 2018;392:1553–98.

    Article  PubMed  Google Scholar 

  3. Braveman P. Health disparities and health equity: concepts and measurement. Annu Rev Public Health. 2006;27:167–94.

    Article  PubMed  Google Scholar 

  4. Alahmad B, Kurdi H, Colonna K, Gasana J, Agnew J, Fox MA. COVID-19 stressors on migrant workers in Kuwait: cumulative risk considerations. BMJ Glob Health 2020, 5.

  5. Wang X, Zhou P, Liu Z. Neighborhood social capital and self-rated mental health: disparities between migrants and native residents in Beijing. Front Public Health. 2022;10:1055712.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Xu F, Crush J, Zhong T. Pathways to food insecurity: Migration, Hukou and COVID-19 in Nanjing, China. Popul Space Place. 2023;29:e2640.

    Article  PubMed  Google Scholar 

  7. Wang B, Li X, Stanton B, Fang X. The influence of social stigma and discriminatory experience on psychological distress and quality of life among rural-to-urban migrants in China. Soc Sci Med. 2010;71:84–92.

    Article  PubMed  Google Scholar 

  8. Huang L, Said R, Goh HC, Cao Y. The Residential Environment and Health and well-being of Chinese migrant populations: a systematic review. Int J Environ Res Public Health 2023, 20.

  9. Huang X, Dijst M, van Weesep J, Jiao Y, Sun Y. Residential choice among rural–urban migrants after Hukou Reform: evidence from Suzhou, China. Popul Space Place. 2017;23:e2035.

    Article  Google Scholar 

  10. Shao Y, Ying H, Li X, Tong L. Association between socioeconomic status and mental health among China’s migrant workers: a moderated mediation model. PLoS ONE. 2022;17:e0274669.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Li Z, Dai J, Wu N, Gao J, Fu H. The mental health and depression of rural-to-urban migrant workers compared to non-migrant workers in Shanghai: a cross-sectional study. Int Health. 2019;11:S55–63.

    Article  PubMed  Google Scholar 

  12. Chen F, Zheng M, Xu J, Hall BJ, Pan Y, Ling L, Chen W. Impact of migration status on incidence of depression in the middle-aged and elderly population in China: exploring healthy migrant and salmon bias hypotheses from a mental health perspective. J Affect Disord. 2022;315:182–9.

    Article  PubMed  Google Scholar 

  13. Hao Z. The pattern of Age-specific Migration rate of floating Population and its changes in China. J East China Normal Univ. 2023;55:185–201.

    Google Scholar 

  14. Jeste DV, Pender VB. Social Determinants of Mental Health: recommendations for Research, Training, Practice, and policy. JAMA Psychiatry. 2022;79:283–4.

    Article  PubMed  Google Scholar 

  15. Dykxhoorn J, Fischer L, Bayliss B, Brayne C, Crosby L, Galvin B, Geijer-Simpson E, Jones O, Kaner E, Lafortune L, et al. Conceptualising public mental health: development of a conceptual framework for public mental health. BMC Public Health. 2022;22:1407.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Transforming our world. the 2030 Agenda for Sustainable Development. Accessed 12 May 2023.

  17. National Bureau of Statistics of China. China Population Census Yearbook 2020 2022.

  18. Technical Document of the National Migrant Population Health and Family Planning Dynamic Monitor Survey. (2018). Accessed May 12 2023.

  19. COVID-19 pademic. triggers 25% increase in prevalence of anxiety and depression worldwide. Accessed July 11 2023.

  20. Levis B, Sun Y, He C, Wu Y, Krishnan A, Bhandari PM, Neupane D, Imran M, Brehaut E, Negeri Z, et al. Accuracy of the PHQ-2 alone and in Combination with the PHQ-9 for screening to detect Major Depression: systematic review and Meta-analysis. JAMA. 2020;323:2290–300.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Lan Y, Roberts H, Kwan MP, Helbich M. Daily space-time activities, multiple environmental exposures, and anxiety symptoms: a cross-sectional mobile phone-based sensing study. Sci Total Environ. 2022;834:155276.

    Article  CAS  PubMed  Google Scholar 

  22. Victor CR, Rippon I, Barreto M, Hammond C, Qualter P. Older adults’ experiences of loneliness over the lifecourse: an exploratory study using the BBC loneliness experiment. Arch Gerontol Geriatr. 2022;102:104740.

    Article  PubMed  Google Scholar 

  23. Forbes D, Alkemade N, Mitchell D, Elhai JD, McHugh T, Bates G, Novaco RW, Bryant R, Lewis V. Utility of the dimensions of anger Reactions-5 (DAR-5) scale as a brief anger measure. Depress Anxiety. 2014;31:166–73.

    Article  PubMed  Google Scholar 

  24. Sherin KM, Sinacore JM, Li XQ, Zitter RE, Shakil A. HITS: a short domestic violence screening tool for use in a family practice setting. Fam Med. 1998;30:508–12.

    CAS  PubMed  Google Scholar 

  25. Lu J, Yang Y, Cui J, Xu W, Wu C, Li J, Li X. Alcohol use disorder and its association with quality of life and mortality in Chinese male adults: a population-based cohort study. BMC Public Health. 2022;22:789.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Oaxaca RL. Male-female wage differentials in Urban Labor Markets. Int Econ Rev. 1973;14:693–709.

    Article  Google Scholar 

  27. Shah NS, Huang X, Petito LC, Bancks MP, Ning H, Cameron NA, Kershaw KN, Kandula NR, Carnethon MR, Lloyd-Jones DM, Khan SS. Social and Psychosocial Determinants of Racial and Ethnic Differences in Cardiovascular Health in the United States Population. Circulation. 2023;147:190–200.

    Article  PubMed  Google Scholar 

  28. Allen J, Cotter-Roberts A, Darlington O, Dyakova M, Masters R, Munford L. Understanding health inequalities in Wales using the Blinder-Oaxaca decomposition method. Front Public Health. 2022;10:1056885.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Novak P, Chu J, Ali MM, Chen J. Racial and ethnic disparities in Serious Psychological Distress among those with Alzheimer’s Disease and related dementias. Am J Geriatr Psychiatry. 2020;28:478–90.

    Article  PubMed  Google Scholar 

  30. Fairlie RW. An Extension of the Blinder-Oaxaca Decomposition Technique to Logit and Probit Models. 2006.

  31. Liu Y, Li G, Qi X, Wu B, Latkin CA, Tang W, Hall BJ. Prevalence and determinants of food insecurity during the 2022 COVID-19 related lockdown in Shanghai. Glob Public Health. 2023;18:2246066.

    Article  PubMed  Google Scholar 

  32. Buecker S, Denissen JJA, Luhmann M. A propensity-score matched study of changes in loneliness surrounding major life events. J Pers Soc Psychol. 2021;121:669–90.

    Article  PubMed  Google Scholar 

  33. Hou F, Liu H, Peng X, You L, Zhou Z, Xie H, Liu T. Gender disparities in depressive and anxiety symptoms among internal migrant workers in Shenzhen: a cross-sectional study. BMJ Open. 2020;10:e041446.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Wong DF, Song HX. The resilience of migrant workers in Shanghai China: the roles of migration stress and meaning of migration. Int J Soc Psychiatry. 2008;54:131–43.

    Article  PubMed  Google Scholar 

  35. Zlotnick C, Dryjanska L, Suckerman S. Health literacy, resilience and perceived stress of migrants in Israel during the COVID-19 pandemic. Psychol Health. 2022;37:1076–92.

    Article  PubMed  Google Scholar 

  36. Varma P, Junge M, Meaklim H, Jackson ML. Younger people are more vulnerable to stress, anxiety and depression during COVID-19 pandemic: a global cross-sectional survey. Prog Neuropsychopharmacol Biol Psychiatry. 2021;109:110236.

    Article  CAS  PubMed  Google Scholar 

  37. Why younger people. face a major mental crisis. Accessed May 12 2023.

  38. Li Q, Zhou X, Ma S, Jiang M, Li L. The effect of migration on social capital and depression among older adults in China. Soc Psychiatry Psychiatr Epidemiol. 2017;52:1513–22.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Tan ST, Low PTA, Howard N, Yi H. Social capital in the prevention and management of non-communicable diseases among migrants and refugees: a systematic review and meta-ethnography. BMJ Glob Health 2021, 6.

  40. Johnson CM, Rostila M, Svensson AC, Engström K. The role of social capital in explaining mental health inequalities between immigrants and Swedish-born: a population-based cross-sectional study. BMC Public Health. 2017;17:117.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Wang X, Liu Z. Neighborhood environments and inclusive cities: an empirical study of local residents’ attitudes toward migrant social integration in Beijing, China. Landsc Urban Plann. 2022;226:104495.

    Article  Google Scholar 

  42. Wang Z, Zhang F, Wu F. Affective neighbourly relations between migrant and local residents in Shanghai. Urban Geogr. 2017;38:1182–202.

    Article  CAS  Google Scholar 

  43. Zhang R, Zhang C, Xia J, Feng D, Wu S. Household Wealth and Individuals’ Mental Health: Evidence from the 2012–2018 China Family Panel Survey. Int J Environ Res Public Health 2022, 19.

  44. Blackmore R, Boyle JA, Fazel M, Ranasinha S, Gray KM, Fitzgerald G, Misso M, Gibson-Helm M. The prevalence of mental illness in refugees and asylum seekers: a systematic review and meta-analysis. PLoS Med. 2020;17:e1003337.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Chen R, Kessler RC, Sadikova E, NeMoyer A, Sampson NA, Alvarez K, Vilsaint CL, Green JG, McLaughlin KA, Jackson JS, et al. Racial and ethnic differences in individual-level and area-based socioeconomic status and 12-month DSM-IV mental disorders. J Psychiatr Res. 2019;119:48–59.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Thomson RM, Igelström E, Purba AK, Shimonovich M, Thomson H, McCartney G, Reeves A, Leyland A, Pearce A, Katikireddi SV. How do income changes impact on mental health and wellbeing for working-age adults? A systematic review and meta-analysis. Lancet Public Health. 2022;7:e515–28.

    Article  PubMed  PubMed Central  Google Scholar 

  47. the Seventh National Population Census. Accessed 28 May 2024.

  48. China Migrants Dynamic Survey. Accessed 28 May 2024.

  49. Shi L, Lu ZA, Que JY, Huang XL, Liu L, Ran MS, Gong YM, Yuan K, Yan W, Sun YK, et al. Prevalence of and risk factors Associated with Mental Health symptoms among the General Population in China during the Coronavirus Disease 2019 Pandemic. JAMA Netw Open. 2020;3:e2014053.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Li Y, Qin Q, Sun Q, Sanford LD, Vgontzas AN, Tang X. Insomnia and psychological reactions during the COVID-19 outbreak in China. J Clin Sleep Med. 2020;16:1417–8.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Wang W, Bian Q, Zhao Y, Li X, Wang W, Du J, Zhang G, Zhou Q, Zhao M. Reliability and validity of the Chinese version of the Patient Health Questionnaire (PHQ-9) in the general population. Gen Hosp Psychiatry. 2014;36:539–44.

    Article  PubMed  Google Scholar 

  52. Tong X, An D, McGonigal A, Park SP, Zhou D. Validation of the generalized anxiety Disorder-7 (GAD-7) among Chinese people with epilepsy. Epilepsy Res. 2016;120:31–6.

    Article  PubMed  Google Scholar 

Download references


The original study and data collection was funded by the NYU Shanghai Center for Global Health Equity. We thank Lu Haidong for assistance with the sample weighting.


This study was funded by the Natural Science Foundation of Guangdong Province for Distinguished Young Scholars (2022B1515020094).

Author information

Authors and Affiliations



BJH conceptualized the study and was responsible for the study design and data collection. WC wrote the grant application and formulated the statistical plan. MZ developed the statistical plan, DK completed the data cleaning and descriptive statistics (in R), and KW conducted Fairlie decomposition analysis (in Stata). WC, GL, and YZ conducted data validation. All authors contributed to manuscript revisions and gave critical inputs to the manuscript. All authors were granted complete access to the study data and had final responsibility for the manuscript’s submission for publication.

Corresponding authors

Correspondence to Wen Chen or Brian J. Hall.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of the NYU Shanghai (2022-008-NYUSH). All procedures were in accordance with the Declaration of Helsinki. All respondents completed the consent form and explicitly agreed to participate.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zheng, M., Kong, D., Wu, K. et al. The determinants of mental health inequalities between Chinese migrants and non-migrants during the Shanghai 2022 lockdown: a Blinder-Oaxaca decomposition. Int J Equity Health 23, 136 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: