Burnout among doctors in China: a systematic review and meta-analysis

Abstract

Objectives

The current study was performed to systemically examine the prevalence of burnout, risk factors among Chinese doctors, and possible treatment strategies.

Methods

Two authors independently conducted literature searches in PubMed, Embase, Web of Science, Cochrane Library, China National Knowledge Infrastructure (CNKI) and Chinese Scientific and Technical Papers and Citation (CSTPC) databases from the year of 1974 (when “burnout” was first defined), to May 27th, 2023, without language restriction. All published studies investigating burnout, and/or its 3 dimensions among practicing doctors in China, were included. Quality assessment followed the Joanna Briggs Institute (JBI) Critical Appraisal Checklists.

Results

A total of 133 studies comprising 193,866 Chinese doctors were included in the current study. The pooled prevalence of burnout and severe burnout among Chinese doctors were 61% and 12%, respectively. The corona virus disease 19 (COVID-19) pandemic had a significant impact on severe burnout (14%). Emergency physicians had the highest prevalence of burnout (91%), while neurologists experienced the highest prevalence of emotional exhaustion (69%) and depersonalization (59%), whereas the lowest personal accomplishment levels were detected among anesthesiologists (65%). Additionally, 27 risk factors were demonstrated to be associated with burnout among Chinese doctors. Of which, personal psychological status was the greatest predictor of burnout among Chinese doctors (OR 3.88, 95% CI 3.75–4.01).

Conclusions

The overall prevalence of burnout is high among Chinese doctors, and it varies across different medical specialties. Personal psychological status was the greatest predictor of burnout among Chinese doctors. Regular psychological counseling, workload alleviation and income increase are recommended coping strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25034-8.

Introduction

The vast population of 1.3 billion people and the growing healthcare demands in China have led to an overwhelming workload for Chinese doctors. With a physician-to-population ratio of 1:735, significantly lower than in Western countries (1:280–1:640) [1]. China faces challenges in meeting the healthcare needs of its population. For instance, there are only 0.4 pediatricians per 1,000 people, compared to 1.9 in the United States and 0.93 in Japan [2]. Similarly, the country has just 0.47 anesthesiologists per 1,000 individuals, far below the European standard of 2.4 per 1,000 [3]. China has been undergoing major healthcare reforms, transitioning from a physician-centered to a patient-centered model [4]. This shift is not merely from disease treatment to health promotion but also from life sustenance to enhancing quality of life. This new paradigm necessitates that healthcare providers frequently endure work overloads and additional shifts [5]. Hospitals require staff to manage complex tasks, adapt to changes, and cope with patient expectations, ethical dilemmas, and end-of-life decisions [6], all of which contribute to heightened burnout risk among Chinese physicians [7].

Burnout, first defined by Herbert J. Freudenberger in 1974 and later expanded by Maslach and Jackson in 2001, is characterized by three dimensions: emotional exhaustion (EE), depersonalization (DP), and low personal accomplishment (PA) [8]. EE describes a state of emotional depletion at work; DP refers to a detached, negative, and dehumanizing attitude towards patients or colleagues; Low PA denotes a reduced sense of personal achievement and competence in one's job. Burnout not only negatively impacts physicians' physical and mental health—manifesting as anxiety, depression, insomnia, cognitive decline, and increased risk of substance abuse—but also undermines job performance, organizational commitment, and professional efficacy. It is strongly associated with poorer patient care, higher medical error rates, and deteriorating doctor-patient relationships [2]. In June 2019, the World Health Organization (WHO) recognized burnout as a syndrome stemming from chronic workplace stress [9].

Burnout is a significant concern among doctors worldwide, with prevalence rates ranging from 20 to 70% [10–16]. While Zheng et al. [17, 18] provided valuable insights into burnout among Chinese doctors, their data were collected before the COVID-19 pandemic, leaving a gap in understanding the pandemic's impact on burnout. The current study addresses this gap by incorporating recent literature and updated data collected during and after the pandemic, providing a timely reflection of burnout in the post-pandemic era. The current study adds several novel contributions: (1) it investigates the prevalence of burnout across different medical specialties; (2) examines the impact of the COVID-19 pandemic on burnout rate; (3) explores regional differences in burnout across eastern, central, and western China, as regional disparities in healthcare infrastructure, socioeconomic conditions, and work environments may contribute to variations in burnout levels; (4) identifies China-specific risk factors for burnout and its dimensions. Methods.

Protocol and registration

The systematic review and meta-analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Guidelines [19] and were registered at PROSPERO (registration number CRD42021236568).

Search strategy

Three authors (YTS, WW, YTY) independently searched databases including PubMed, Embase, Web of science, Cochrane Library, CNKI and CSTPC from 1974 (when “burnout” was first defined), to May 27th, 2023, with no language restrictions. This date was defined a priori as the cut-off for the literature search. We also cross-checked recent publications during the revision process and found no additional eligible studies beyond our original cut-off date. Key search terms included burnout, emotional exhaustion, depersonalization, personal accomplishment, and relevant medical specialties (e.g., Anesthesiologists, Cardiologists, Surgeons). Additional studies were identified through manual searching of reference lists. A detailed search strategy is provided in the appendix1.

Inclusion and exclusion criteria

We will include the following studies: (1) Population: doctors certified by the Ministry of Health of the People’s Republic of China, currently practicing full-time at hospitals in mainland China. Doctors must have completed their probationary period (ranging from six months to one year) to be considered full-time employees; (2) Intervention/Exposure: studies examining burnout among healthcare providers working in hospitals, with a focus on the occupational stress and workload factors contributing to burnout; (3) Comparison: studies that report burnout in doctors compared to other relevant healthcare providers or baseline levels. However, if no direct comparison is available, we will still include studies that focus on burnout in doctors; (4) Outcomes: studies reporting the prevalence of burnout, severe burnout, or any of its 3 dimensions (EE, DP, low PA) as dichotomous data; (5) Study design: observational studies, including cohort studies, cross-sectional studies, and case–control studies, that assess burnout in doctors working in mainland China.

Excluded studies were as follows: (1) publications such as review articles, case reports, guidelines, conference abstracts, and letters; (2) studies that did not differentiate between doctors and other healthcare workers such as nurses or administrative staff; (3) duplicate publications; (4) studies lacking information on the outcomes of interest; (5) non-clinicians, including administrative staff, support staff, pharmacy staff, and others who do not engage directly in patient care in hospitals.; (6) studies with sample sizes smaller than 100 participants, as they may lack sufficient power to detect significant burnout outcomes.

Definitions

Burnout was defined by high EE or high DP or low PA, while severe burnout was determined as high EE along with high DP or low PA [10]. All definitions were based on the MBI scale [8], a 22-item questionnaire with 3 subscales assessing domains of burnout: EE (9 items), DP (5 items), and low PA (8 items). Response is scored on a scale from 0 (never) to 6 (every day). Scores are classified as low (EE = 0–16, DP = 0–6, low PA ≥ 39), moderate (EE = 7–26, DP = 7–12, low PA = 32–38), or high (EE ≥ 27, DP ≥ 13, low PA ≤ 0–31). The Maslach Burnout Inventory Human Services Survey (MBI-HSS) [3] includes similar subscales. The 15-item Chinese Maslach Burnout Inventory (CMBI) [20], uses a 1–6 scoring system. The Chinese version of the Maslach Burnout Inventory-General Scale (MBI-GS) [21] measures three dimensions of burnout: exhaustion (5 items), cynicism (4 items), and reduced professional efficacy (6 items) on a 7-point Likert scale (0 = never to 6 = every day).

Study selection

After removing duplicate records using EndNote X7 (Thomson Scientific, USA) reference management software, two authors (YTS and YCL) independently screened the titles and abstracts of potential studies and retrieved the full texts for further assessment. Any disagreement about inclusion were resolved through consensus, with input from the third author (YTY) when necessary.

Data extraction

Data were independently extracted by YTS and WW into an Excel spreadsheet (Microsoft office). Extracted information included study characteristics (first author, year of data collection, publication year, sample size, participation rate, region) and participant demographics (mean age, female proportion, specialty). Burnout-related data (number of doctors with burnout, severe burnout rates, high EE, high DP, low PA) and potential risk factors for burnout and its dimensions were also recorded. Any discrepancies were discussed and reconciled with the author (YTY).

Evaluation of study quality

Two authors (YTS and WW) independently assessed methodological quality using the Joanna Briggs Institute (JBI) Critical Appraisal Checklists for analytical observational studies [22]. Disagreements were resolved through discussion among the authors. The JBI Checklist, which contains 8 questions for cross-sectional studies, assigns a score of 1 for ‘yes’ responses and 0 for ‘no’, ‘unclear’, or ‘not applicable’ responses. Scores of 4–6 suggest moderate quality, while scores of 7 or above indicate high quality. This checklist was included as Table S1 in the Supplementary material.

Statistical analysis

The primary outcomes addressed the pooled-prevalence of burnout, severe burnout, and 3 dimensions (high EE, high DP, low PA), which were presented as percentages with 95% confidence interval (CIs) using a random effect model. These analyses were conducted with the STATA statistical software package version 15.0 (Stata Corp 2007, College Station, TX). For secondary outcomes, we conducted a qualitative synthesis of the risk factors associated with burnout. When two or more studies reported the same risk factors, a meta-analysis was performed. Adjusted odds ratios (ORs) from multivariate analyses, with their 95% CIs, were extracted. The ORs were then transformed into their natural logarithm (log [OR]), and the 95% CIs were converted to standard errors (SEs) using a random-effects model to aggregate the risk factor data. Heterogeneity was quantified using the Q statistic and an I² index, with P < 0.10 and I² > 50% indicating significant heterogeneity. Publication bias was assessed visually with funnel plots. Sensitivity analyses were employed to assess the robustness of our findings. Furthermore, exploratory subgroup analyses were performed to investigate potential sources of heterogeneity. Meta-analyses were executed using Review Manager 5.3 (The Cochrane Collaboration, Denmark).

Results

Literature search

A total of 2,091 records were identified through a database search, which included 80 studies from PubMed, 225 from Web of science, 92 Embase, 16 from Cochrane, 1,110 from CNKI, 568 from CSTPC, covering the period from 1974 (when “burnout” was first defined), to May 27th, 2023. After removing 1,139 duplicate records, 657 were excluded on the basis of their titles and abstracts. Following full-text review, 162 articles were further excluded. Ultimately, 133 studies met the inclusion criteria. Of these, 74 studies provided odds ratios (ORs) and 95% confidence intervals (CIs) from multivariate analysis to predict burnout risk factors. The PRISMA flow diagram (Fig. 1) illustrates the literature selection process, including the number of articles at each stage. According to the JBI checklist scores, all included studies were observational with a low to moderate risk of bias (Table S2 and Figure S1).

Fig. 1.

PRISMA flow diagram of study selection

Study characteristics

The characteristics of the included studies are detailed in Table 1. A total of 193,866 Chinese physicians from 17 medical specialties were included, with 53.2% being female. Participation rates in the cross-sectional studies ranged from 12.3% to 100%, with the number of participants varying from 108 to 22,213. The majority of studies (80%) were conducted prior to the COVID-19 pandemic, and most were multi-center (117 of 133). The most frequently represented specialties were clinicians (68 studies), followed by primary care physicians (16 studies), with additional representation from specialties such as neurology, anesthesiology, and oncology. All studies employed the Maslach Burnout Inventory (MBI) to assess burnout, using different versions of the scale, including the CMBI, MBI-GS, and MBI-HSS. Geographically, the studies were distributed across China, with the majority conducted in eastern China (64 studies), followed by central China (11 studies) and western China (21 studies).

Table 1.

Characteristics of included studies

Study	Collection year	Marrige ratio %	N	Female%	Medical specialties	Scale	Multicenter	Region
Wen 2016 [23]	2013	73.5	1,537	42.7	N/D	3	46	NW
Wu 2013 [24]	2010	80.9	1,202	53.8	N/D	3	7	EA
Wang 2014 [25]	2008	78.1	457	59.5	N/D	4	21	EA
Qiao 2016 [21]	2016	68.2	512	70.1	Infectious disease specialists	3	4	CA
Wu 2012 [26]	2008	65.3	2,721	62.0	N/D	3	12	NW
Zhou 2017 [27]	2014	82	6,804	54.0	Neurologists	2	Yes	NW
Zhang 2020 [28]	2016	82.1	2,617	47.4	N/D	2	Yes	NW
Li 2018 [3]	2015	79	2,873	55.0	Anesthesiologists	4	211	EA
Ma 2019 [1]	2015	78.7	1,620	43.9	Oncologists	4	Yes	NW
Liu 2020 [4]	2018	73.9	1,052	79.1	N/D	3	Yes	WA
Zhao 2021 [29]	2019	95.4	1,248	18.9	Primary care physicians	4	16	NW
Lu 2017 [30]	2012	44.8	500	44.8	Rehabilitation professionals	3	24	EA
Rui 2016 [31]	2011	71.2	395	41.7	Anesthesiologists	4	30	NW
Ye 2019 [32]	2017	77.0	678	80.7	Pediatricians and obstetricians	3	11	NW
Wang 2020 [33]	2017	90.3	699	59.9	N/D	4	47	EA
Gan 2019 [34]	2014	93.3	1,015	35.0	N/D	4	Yes	CA
Xiao 2014 [35]	2012	80.4	205	39.0	Emergency physicians	3	3	EA
Zhang 2021 [36]	2019	N/A	2,693	35.6	Primary care physicians	3	Yes	EA
Pu 2017 [37]	2014	81.8	5,558	54.0	Neurologists	2	30	NW
Wu 2011 [38]	2008	71.1	900	100.0	N/D	3	60	EA
Sui 2019 [39]	2015	85.2	1,392	56.3	N/D	4	Yes	EA
Cheng 2019 [40]	2015	72.3	2,576	39.5	N/D	2	6	NW
Zheng 2018 [41]	N/A	93.1	202	0.0	Orthopedic surgeons	4	Yes	NW
Hu 2021 [42]	2019	72.2	2,411	68.7	Intensivists	2	30	NW
Song 2021 [43]	2017	N/A	3,506	51.2	N/D	2	9	EA
Pang 2021 [44]	2016	82.8	285	44.9	Oncologists	4	No	EA
Liang 2021 [2]	2019	91.0	700	6.7	Pediatricians and obstetricians	3	29	NW
Ma 2020 [20]	2000	77.5	2,530	45.5	N/D	1	Yes	NW
Pei 2020 [45]	2019	82.6	N/A	46.9	N/D	3	305	NW
Zhao 2018 [46]	2017	65.1	393	46.8	Interventional medical staffs	4	Yes	EA
Sun 2023 [47]	2019	74.1	8,738	74.1	N/D	3	22	NW
Yu 2022 [48]	NA	73.6	144	72.9	Stomatologists	3	9	EA
Zhou 2015 [49]	2014	80.1	1,611	66.4	N/D	4	8	EA
Du 2015 [50]	2014	89.2	759	50.1	Primary care physicians	3	Yes	WA
Chen 2018 [51]	2017	N/A	310	N/A	Anesthesiologists	4	30	WA
Shen 2019 [52]	2010	78.7	602	16.9	N/D	1	Yes	NW
Liu 2017 [53]	2017	N/A	518	57.8	N/D	2	3	WA
Wang 2017 [54]	2014	82.2	1,095	56.2	Neurologists	2	180	NW
Lin 2018 [55]	2018	N/A	324	45.7	N/D	4	21	EA
Li 2022 [56]	2020	90.4	426	57.3	Primary care physicians	2	30	EA
Ma 2022 [57]	2020	N/A	2,232	79.8	N/D	2	N/A	WA
Yin 2012 [58]	2011	38.26	460	41.5	Resident doctors	3	Yes	EA
Liu 2012 [59]	2011	N/A	266	40.9	N/D	4	6	EA
Cheng 2012 [60]	2010	78.6	653	49.3	N/D	1	N/A	WA
Liu 2013 [61]	2010	81.5	734	47.5	Primary care physicians	1	Yes	NW
Zhu 2013 [62]	2013	80.0	352	50.9	Oncologists	3	3	EA
Zhang 2015 [63]	2013	82.5	160	64.4	Primary care physicians	3	14	EA
Chang 2015 [64]	2014	N/A	516	46.7	Primary care physicians	4	12	EA
Liu 2017 [65]	2016	51.2	924	83.1	N/D	4	Yes	EA
Yu 2019 [66]	2017	79.5	1,202	7.1	Neurologists	4	Yes	NW
Ran 2020 [67]	2019	79.1	1,279	66.5	Primary care physicians	3	N/A	CA
Wang 2017 [68]	2014	82.2	1,139	56.4	Neurologists	2	180	WA
Wang 2021[69]	2018	N/A	133	61.0	Anesthesiologists	4	17	WA
Yao 2021 [11]	2019	81.2	6,520	58.1	Psychiatrists	4	41	NW
Yan 2021 [70]	2019	83.3	15,243	30.1	Emergency physicians	4	10	NW
Wang 2021 [10]	2019	92.1	1,813	37.1	Intensivists	2	Yes	NW
Fu 2022 [66]	2021	72.7	1,602	86.6	N/D	3	3	CA
Huo 2021 [71]	2020	74.9	N/A	81.2	N/D	3	Yes	NW
Wang 2021 [72]	2020	75.4	500	81.0	N/D	3	Yes	WA
Wang 2022 [73]	2019	89.3	711	72.2	Endocrinologists	4	31	NW
Che 2022 [74]	2021	74.3	6,631	59.9	Anesthesiologists	4	Yes	NW
Liu 2014 [75]	2013	32.3	300	30.3	N/D	3	N/A	EA
Zhai 2019 [76]	2018	81.2	245	45.7	N/D	4	3	CA
Sun 2017 [77]	2014	N/A	379	15.8	Primary care physicians	3	4	CA
Wu 2014 [78]	2013	68.7	399	30.3	N/D	3	N/A	EA
Liu 2012 [79]	2011	75.7	819	49.1	N/D	3	Yes	WA
Wu 2021 [80]	2019	82.5	22,213	57.6	N/D	4	144	NW
Luo 2021 [81]	2020	N/A	1,982	72.5	N/D	3	21	EA
Yin 2021 [82]	2019	N/A	169	N/A	N/D	1	4	EA
Li 2022 [83]	2020	82.5	861	75.4	N/D	3	19	EA
Wu 2021 [84]	N/A	N/A	140	66.4	Pediatricians	3	Yes	WA
Xu 2021 [85]	2020	74.4	500	50.6	N/D	3	4	EA
Gao 2021 [86]	2020	75.0	1,044	78.3	N/D	3	N/A	EA
Shan 2021 [87]	2019	81.7	N/A	72.9	N/A	4	5	EA
Jin 2021 [88]	NA	42.0	264	43.2	N/D	3	N/A	WA
Zhen 2021 [89]	2019	N/A	370	61.8	Pediatricians	1	N/A	EA
Su 2022 [90]	2020	69.6	1,463	84.7	Oncologists	4	N/A	EA
Jin 2021 [91]	2020	60.3	262	67.2	Anesthesiologists	4	50	EA
Chen 2021 [92]	2019	70.5	234	58.5	Stomatologists	3	Yes	WA
Yu 2020 [93]	2019	76.4	182	70.9	N/D	1	N/A	N/A
Wang 2020[94]	2019	N/A	1,369	N/A	Intensivists	2	Yes	NW
Zhang 2020 [95]	2020	88.8	1,308	46.6	N/D	3	10	EA
Gu 2020 [96]	2019	N/A	244	63.5	N/D	1	2	WA
Li 2005 [97]	2003	N/A	281	N/A	N/A	2	N/A	WA
Ren 2007 [98]	2006	71.1	256	46.9	N/D	4	3	EA
Zhang 2022 [99]	2021	N/A	311	N/A	Tuberculosis	3	N/A	EA
Wang 2008 [100]	2007	N/A	646	39.9	N/D	1	20	EA
Yuan 2022 [101]	2020	N/A	1,655	68.4	Gastroenterologists	1	331	NW
Cui 2013 [102]	2012	35.8	510	54.9	N/D	3	Yes	NW
Cheng2012 [103]	2010	N/A	611	49.2	N/D	1	N/A	WA
Jiang 2009 [104]	2008	N/A	461	44.5	N/D	1	Yes	EA
Huang2011 [105]	2010	83.5	243	44.4	N/D	1	Yes	EA
Liu 2012 [106]	2010	81.6	1,569	47.0	N/D	1	Yes	NW
Dai 2020 [107]	2018	44.7	891	71.5	N/D	3	N/A	EA
Zhou 2021 [108]	2020	97.3	2,272	35.0	Primary care physicians	4	Yes	EA
Li 2009 [109]	2008	N/A	342	44.4	Infectious disease specialists	1	Yes	CA
Huang2015 [110]	2014	69.9	775	61.2	N/D	3	15	CA
Jiang 2011 [111]	2010	N/A	160	N/A	Psychiatrists	2	N/A	EA
Zhang 2016 [112]	2014	69.6	112	47.3	Psychiatrists	3	4	EA
Liu 2012 [113]	2011	N/A	372	N/A	Primary care physicians	3	Yes	EA
Dai 2010 [114]	2009	N/A	131	35.9	Stomatologists	4	4	EA
Wu 2021 [115]	2020	77.2	399	N/A	Obstetricians	4	N/A	EA
Zhang 2022 [116]	2021	75.6	160	N/A	N/D	4	N/A	EA
Wang 2019 [117]	2017	75.9	469	41.4	N/D	1	3	EA
Yu 2012 [118]	2011	N/A	235	56.0	Stomatologists	3	N/A	EA
Chen 2009 [119]	2007	50.6	108	51.9	Resident doctors	2	2	EA
Chen 2015 [120]	2014	N/A	4,884	77.7	N/D	4	46	EA
Huang2019 [121]	2018	N/A	743	61.0	N/D	3	Yes	WA
Liu 2022[122]	2020	66.1	342	76.3	N/D	3	N/A	EA
Yu 2012 [123]	2011	84.4	237	56.0	Stomatologists	3	N/A	EA
Wei 2019 [124]	2018	N/A	151	44.4	N/D	3	6	EA
Zheng 2019 [125]	2017	85.6	3,236	63.8	N/D	4	30	NW
Wang 2013 [126]	2011	N/A	4,674	40.8	N/D	3	Yes	NW
Wu 2019 [127]	2017	80.2	499	8.4	Primary care physicians	3	18	EA
Li 2017 [128]	2016	N/A	1,047	84.3	N/D	3	N/A	WA
Jiang 2010 [129]	2009	N/A	461	45.5	N/D	1	11	NW
Han 2018 [130]	2017	N/A	136	22.8	Intensivists	2	4	WA
Chen 2019 [131]	2018	95.3	316	20.8	Primary care physicians	3	Yes	EA
Liu 2012 [132]	2011	N/A	447	N/A	N/D	3	N/A	WA
Ma 2017 [133]	2016	N/A	2,584	80.8	Primary care physicians	4	29	EA
Gu 2018 [134]	2016	84.6	292	39.3	Primary care physicians	3	Yes	EA
Liu 2015 [135]	2014	65.5	415	44.8	N/D	4	3	EA
Song 2018 [136]	2017	84.7	203	46.3	Oncologists	3	N/A	EA
Li 2019 [137]	2018	N/A	265	35.1	N/D	3	N/A	CA
Zhang 2022 [138]	2020	72.6	2,113	78.6	N/D	3	3	CA
Tian 2019 [139]	2014	81.7	5,369	53.6	Neurologists	2	Yes	NW
Zhong2019 [140]	2015	81.6	5,590	53.7	Neurologists	2	Yes	NW
Zhang 2019 [141]	2018	N/A	131	36.6	Oncologists	3	2	EA
Zhang 2018 [142]	2017	N/A	446	N/A	N/D	4	N/A	EA
Xi 2011 [143]	2010	N/A	271	66.4	N/D	1	5	WA
Zhao 2019 [144]	2015	81.1	592	62.2	N/D	4	10	EA
Luo 2013 [145]	2011	N/A	917	16.1	Primary care physicians	1	N/A	CA
Zhang 2016 [146]	2015	N/A	1,098	47.9	Resident doctors	1	Yes	EA

N/A Not applicable, N/D Not described, EA Eastern area, CA Central area, WA Western area, NW Nationwide, MBI Maslach Burnout Inventory (MBI), CMBI Chinese Maslach Burnout Inventory, MBI-GS Maslach Burnout Inventory-General Scale, MBI-HSS the Maslach Burnout Inventory Human Services Survey. 1= CMBI;2 = MBI;3 = MBI-GS;4 = MBI-HSS

Primary outcomes

Overall burnout prevalence

A total of 88 studies involving 137,277 Chinese doctors, reported on burnout prevalence using dichotomous data (Figure S2). As shown in Table 2, the pooled prevalence of burnout across all studies was 61% (95% CI: 56–66%), with notable variability across specialties, regions, and study methodologies. Prevalence estimates varied by time period, with 62% (95% CI: 56–68%) before the COVID-19 pandemic and 51% (95% CI: 40–61%) during the pandemic. Prevalence was highest among emergency physicians (91%), orthopedic surgeons, intensivists, and infectious disease specialists (around 77%), while psychiatrists (41%) and stomatologists (18%) had much lower rates. Regionally, burnout was most prevalent in western China (66%) and least in central China (49%). The prevalence was slightly higher in Chinese-language studies (62%) compared to English-language studies (57%). Additionally, studies using the CMBI scale reported the highest prevalence (70%), followed by the MBI (64%), MBI-GS (57%), and MBI-HSS (54%).

Table 2.

Subgroup analysis of burnout among Chinese doctors

	Burnout	Severe burnout	High emotional exhaustion	High depersonalization	Low personal achievement
Overall prevalence	0.61(0.56,0.66)	0.12(0.10,0.14)	0.45(0.40,0.51)	0.41(0.37,0.45)	0.50(0.43,0.57)
Pre-COVID19	0.62(0.56,0.68)	0.11(0.09,0.13)	0.45(0.39,0.50)	0.42(0.38,0.46)	0.51(0.44,0.58)
During-COVID19	0.51(0.40,0.61)	0.14(0.05,0.24)	0.50(0.32,0.69)	0.36(0.19,0.53)	0.45(0.16,0.73)
Medical specialties
Clinicians	0.62(0.53,0.71)	0.10(0.08,0.12)	0.46(0.37,0.55)	0.41(0.36,0.47)	0.50(0.41,0.59)
Primary care physicians	0.47(0.24,0.69)	0.15(0.02,0.28)	0.37(0.27,0.46)	0.35(0.23,0.47)	0.50(0.30,0.71)
Anesthesiologists	0.58(0.57,0.59)	0.06(0.04,0.09)	0.58(0.39,0.77)	0.39(0.25,0.53)	0.65(0.54,0.76)
Neurologists	0.53(0.51,0.55)	-	0.67(0.54,0.79)	0.59(0.49,0.69)	0.60(0.52,0.69)
Infectious disease specialists	0.77(0.73,0.80)	-	0.34(0.13,0.54)	0.34(0.03,0.66)	0.38(0.13,0.63)
Oncologists	0.49(0.44,0.54)	0.01(0.00,0.02)	0.17(0.01,0.33)	0.23(0.09,0.38)	0.30(0.03,0.58)
Pediatricians and obstetricians	0.64(0.61,0.66)	0.18(0.15,0.22)	0.59(0.16,1.01)	0.53(0.29,0.78)	0.47(0.32,0.63)
Emergency physicians	0.91(0.91,0.92)	-	-	-	-
Orthopedic surgeons	0.77(0.75,0.80)	-	-	-	-
Intensivists	0.77(0.72,0.82)	0.39(0.37,0.40)	0.59(0.43,0.75)	0.49(0.23,0.74)	0.62(0.27,0.97)
Stomatologists	0.18(0.15,0.22)	-	0.05(0.03,0.07)	0.29(0.25,0.34)	0.50(0.45,0.55)
Residents	0.69(0.57,0.81)	0.05(0.04,0.06)	-	-	-
Psychiatrists	0.41(0.40,0.42)	0.02(0.01,0.03)	0.27(0.26,0.28)	0.35(0.33,0.36)	0.22(0.20,0.23)
Scale
CMBI	0.70(0.65,0.75)	0.13(0.08,0.17)	0.26(0.22,0.30)	0.39(0.33,0.46)	0.42(0.33,0.51)
MBI-HSS	0.54(0.37,0.72)	0.08(0.05,0.11)	0.50(0.40,0.60)	0.39(0.33,0.44)	0.58(0.50,0.67)
MBI-GS	0.57(0.48,0.65)	0.10(0.08,0.12)	0.42(0.29,0.55)	0.41(0.32,0.50)	0.45(0.32,0.58)
MBI	0.64(0.57,0.72)	0.20(0.11,0.29)	0.61(0.52,0.70)	0.48(0.31,0.64)	0.55(0.43,0.67)
Region
Nationwide	0.60(0.51,0.70)	0.15(0.11,0.20)	0.50(0.40,0.59)	0.44(0.39,0.49)	0.51(0.38,0.64)
Eastern area	0.61(0.55,0.66)	0.10(0.07,0.12)	0.45(0.38,0.53)	0.40(0.33,0.48)	0.54(0.43,0.64)
Western area	0.66(0.54,0.78)	0.09(0.06,0.11)	0.44(0.22,0.66)	0.39(0.24,0.54)	0.44(0.28,0.60)
Central area	0.49(0.30,0.67)	0.19(0.07,0.31)	0.32(0.20,0.43)	0.37(0.21,0.52)	0.43(0.24,0.62)
Publication language
English	0.57(0.48,0.67)	0.13(0.08,0.18)	0.47(0.37, 0.57)	0.38(0.31,0.45)	0.51(0.43,0.59)
Chinese	0.62(0.56,0.68)	0.11(0.09,0.13)	0.45(0.37, 0.52)	0.42(0.36,0.48)	0.50(0.41,0.58)

Under the proposal of China State Statistical Bureau, all provinces are divided into three areas on the basis of economic development and geographical position when statistical analysis is conducted. The eastern area refers to developed areas, including include 11 provinces or municipalities (i.e. Beijing, Fujian, Jiangsu, Hebei, Hainan, Guangdong, Liaoning, Shanghai, Shandong Tianjin and Zhejiang). The central area refers to developing areas, including eight provinces (i.e. Anhui, Jilin, Jiangxi, Heilongjiang, Henan, Hubei Hunan and Shanxi). The western area refers to underdeveloped areas, including 11 provinces or autonomous regions (i.e. Chongqing, Gansu, Guizhou, Inner Mongolia, Ningxia, Qinghai, Sichuan, Shaanxi, Tibet, Xinjiang and Yunnan)

Prevalence of severe burnout

Severe burnout, defined as high EE along with high DP or low PA, was reported in 58 studies involving 46,833 Chinese physicians (Figure S3). The overall prevalence of severe burnout was 12% (95% CI: 10–14%). Prevalence increased slightly during the COVID-19 pandemic (14%) compared to the pre-pandemic period (11%). The highest rates were found among intensivists (39%), Primary care physicians (15%), and pediatricians/obstetricians (18%). Regionally, severe burnout was more common in central China (19%) and less so in western China (9%). As with overall burnout, the scale used influenced the reported prevalence, with MBI studies showing the highest rate (20%) (Table 2).

Prevalence of high EE

Data on high EE were reported in 91 studies involving 128,937 Chinese doctors (Figure S4). The pooled prevalence of high EE was 45% (95% CI: 40–51%), with a slight increase during the COVID-19 pandemic (50%) compared to pre-pandemic levels (45%). Neurologists reported the highest rates of EE (67%), followed by intensivists and pediatricians (59%), and anesthesiologists (58%). In contrast, oncologists (17%) and stomatologists (5%) had much lower prevalence rates. Regional differences were evident, with western China showing a higher prevalence (44%) compared to central China (32%). The MBI scale again reported the highest prevalence of EE (61%), emphasizing the importance of scale selection in estimating burnout severity.

Prevalence of high DP

A total of 90 studies, involving 128,741 Chinese doctors, assessed the prevalence of high DP (Figure S5). The pooled prevalence of high DP was 41% (95% CI: 37–45%), with slightly higher rates before the pandemic (42%) compared to during the pandemic (36%). Neurologists, pediatricians, and intensivists reported the highest rates of DP (49–53%), while oncologists (23%) and stomatologists (29%) had lower rates. Regionally, the prevalence was fairly consistent across China, with western China showing a prevalence of 39%, eastern China 40%, and central China 37%. The MBI scale once again yielded the highest DP prevalence (48%).

Prevalence of low PA

A total of 91 studies involving 128,905 Chinese doctors assessed the prevalence of low PA (Figure S6), with a pooled prevalence of 50% (95% CI: 43–57%), with a slight decrease during the pandemic (45%) compared to pre-pandemic levels (51%). The highest rates of low PA were reported by anesthesiologists (65%), intensivists (62%), and neurologists (60%). In contrast, psychiatrists (22%), oncologists (30%), and infectiologists (38%) had relatively lower rates. Regionally, low PA was most common in eastern China (54%), followed by western China (44%) and central China (43%). Studies using the MBI-HSS scale reported the highest prevalence (58%).

Risk factors of burnout

A synthesis of data from 74 studies involving 107,242 Chinese doctors identified key risk factors for burnout and its three dimensions. In total, 27 risk factors were associated with overall burnout (Figure S7), 26 with high EE (Figure S8), 26 with high DP (Figure S9), and 23 with low PA (Figure S10). A meta-analysis was conducted when three or more studies provided the same effect measures of outcomes related to the burnout or its three dimensions (Document S1). Psychological well-being consistently emerged as one of the most significant predictors across all burnout dimensions. Other critical factors influencing overall burnout included job stress (OR = 2.79), workload (OR = 1.91), and job satisfaction (OR = 2.25) (as depicted in Fig. 2). Specifically, emotional exhaustion was strongly associated with job stress (OR = 2.53) and hospital type (OR = 1.85) (Fig. 3A). Depersonalization was primarily linked to psychological factors such as self-regulation (OR = 2.31) and professional values (OR = 2.84). (Fig. 3B). Finally, low personal accomplishment was most influenced by workload (OR = 1.76) and income (OR = 0.59) (Fig. 3C).

Fig. 2.

Risk factors of burnout after sensitivity analysis

Fig. 3.

Risk factors of EE (3A), DP (3B), PA(3C) after sensitivity analysis

Discussion

The present study encompassing 133 studies and 193,866 Chinese doctors demonstrated that: (1) 61% of Chinese doctors experienced burnout, with 12% reporting severe burnout. High levels of EE were reported by 45%, DP by 41%, and low PA by 50%, (2) the COVID-19 pandemic has exerted a greater impact on severe burnout (14%) among Chinese doctors, particularly in the domain of EE, as compared to that before the COVID-19 (11%), (3) emergency physicians had the highest overall burnout prevalence, while intensivists have the highest rate of severe burnout. Neurologists experienced the highest levels of EE and DP, with pediatricians/obstetricians and intensivists following. Anesthesiologists reported the lowest personal achievement levels, (4) The Western region reports the highest overall burnout prevalence (66%) and the highest prevalence of EE (44%), while the Central and eastern China shows the higher prevalence of severe burnout and lowest PA. DP prevalence among healthcare workers was consistent across regions, (5) the risk factors such as psychological status, age, educational background, marital status, years in practice, professional title, type of hospital, professional values, workload, job stress, on-call nights, vacation time, medical disputes, doctor-patient relationships, income, and job satisfaction were all identified as independently associated with burnout. Self-regulation was specifically associated with DP, and regional factors were particularly associated with low PA.

The 61% pooled prevalence of burnout observed in our study is notably higher than the 49% found among French doctors [147], 56.6% among American physicians [148], 26% among non-consultant doctors in Ireland [149], and 2.6% among healthcare professionals in Ecuador [150]. Additionally, the 14% pooled prevalence for severe burnout in our study exceeds the 5% reported in French doctors [147] but falls below the 12% seen among General Practitioners in Europe [151] and the 11.7% among Yemen physicians [152].

To properly interpret the higher prevalence of burnout among Chinese physicians, it is important to consider several unique systemic and cultural factors, including heavy workloads, income disparities, hierarchical work culture, and patient distrust. These factors are as follows: 1) China’s ongoing healthcare reforms, aimed at improving accessibility and quality, have increased demand for services, leading to overwhelming patient loads, long working hours, and chronic stress, further exacerbated by a shortage of healthcare workers. Patient-centered care emphasizes greater involvement with patients, focusing on individualized care and communication. This approach often requires more time for consultations, coordination among different care providers, and personalized follow-up, all of which can contribute to increased workload. Furthermore, as the model often involves multidisciplinary teams, healthcare providers may experience additional challenges related to interprofessional collaboration, decision-making, and managing patient expectations, all of which can add to the overall burden [4]. 2) Traditional Chinese work culture emphasizes endurance and resilience, fostering a "work until you drop" mentality, which intensifies emotional exhaustion and depersonalization among doctors [153]. 3) The rigid hierarchical structure within China’s medical system limits junior doctors' autonomy and their ability to voice concerns, which deepens job dissatisfaction and contributes to burnout. In contrast, Western countries tend to offer more autonomy to physicians and have less pronounced hierarchical pressures, which can help reduce burnout [152]. 4) the majority of Chinese patients pay out of pocket for medical expenses, adding financial strain on both physicians and patients [2]. 5) Informational asymmetry between doctors and patients, coupled with a lack of patient education, fosters distrust and resentment towards healthcare providers, further contributing to emotional strain. In comparison, countries with stronger patient education systems may experience less public scrutiny, reducing stress for physicians. 6) Significant income disparities across medical titles in China also contribute to the high burnout rates.

The prevalence of burnout among physicians varies across different specialties [29, 31, 41, 42]. Emergency medicine doctors in China experience the highest rate of burnout at 91%, aligning with a similar study from the US [154]. The demanding nature of emergency medicine, requiring physicians to handle critical situations in a highly stressful environment, may explain this high prevalence. In the present study, the prevalence of severe burnout among intensivists was 39%, which is higher than rates reported in Western countries [155, 156]. The high patient mortality in intensive care units potentially contributes to the severe burnout among intensivists. Anesthesiologists, often referred to as the guardians of surgery, are tasked with vital patient care and increasingly diverse roles including diagnosis, treatments for conditions such as shock and respiratory failure, as well as managing emergencies [14]. The notable prevalence of low PA among anesthesiologists in China may stem from their underappreciation by the general public and hospital management compared to surgeons. Our study showed an overall burnout rate of 77% among Chinese orthopedists, nearly double the rate among North American pediatric orthopedists (38%) [157]. According to statistics by prosthesis manufacturers, there were over 200,000 arthroplasty procedures in 2012 and an annual increase of 15%−25% [158]. In an investigation of Chinese orthopedic surgeons [159], approximately 80% of participants complained no support from their hospitals. Obstetricians and pediatricians in our study showed a burnout rate of 64%, exceeding Iran's 44% [160], but akin to rates among US obstetricians [24]. The two-child policy in China [32] has led to more high-risk births and heavier workloads, contributing to burnout among these specialists. According to data from China’s Sixth National Population Census, 50.3% of the population resides in rural areas [161]. Village doctors, essential to rural healthcare, play an irreplaceable role in providing sustainable basic services which including prevention, treatment, management, health education and health care, etc. and improving the health level of rural residents [36,

162]. Our study showed that village doctors face higher severe burnout and lower PA. The reasons maybe due to patients can choose their hospitals freely in China and preferences for overcrowded secondary or tertiary hospitals over primary ones. This trend is evidenced by the 58.9% bed utilization rate in primary hospitals versus 104.5% in tertiary hospitals [23]. It is essential to allocate resources to address burnout, tailoring interventions to the unique factors prevalent in different regions and specialties.

Burnout prevalence in China varied significantly by region, with western China reporting the highest overall rate (66%) and the highest rate of emotional exhaustion (44%). This discrepancy can be attributed to differences in healthcare infrastructure and resources. Western China faces more challenges, including higher patient loads, limited resources, and shortage of healthcare professionals, all of which contribute to higher burnout rates [4]. In contrast, central and eastern China experienced lower overall burnout rates but higher rates of severe burnout and Low personal accomplishment. The relatively stronger healthcare infrastructure and more structured systems in these regions likely provide better institutional support, alleviating some of the pressures associated with burnout. However, the increased availability of medical professionals and more evenly distributed resources in these areas may reduce the intensity of general burnout, but it could also result in higher expectations, greater workloads, more intense competition, and greater job demands, potentially leading to more severe burnout and low PA [21, 24]. Depersonalization prevalence among healthcare workers was consistent across regions, indicating it is a widespread issue throughout China [23].

However, our study indicates that burnout and low PA were less frequent during the COVID-19 outbreak in China. The swift governmental response with new safety policies likely mitigated burnout [163]. Physicians may have felt a greater sense of achievement during the pandemic, witnessing the recovery of COVID-19 patients and the disease's decline.

Burnout prevalence varied according to the measurement tool used. Studies utilizing the Chinese version of the Maslach Burnout Inventory (CMBI) reported the highest overall burnout rate (70%). The CMBI [20], a comprehensive and culturally adapted tool, effectively captures the three core burnout dimensions: emotional exhaustion (EE), depersonalization (DP), and low personal accomplishment (PA), making it highly sensitive for assessing burnout in China. In contrast, scales like the MBI-GS and MBI-HSS may not assess all dimensions as thoroughly, potentially underestimating burnout prevalence. The original Maslach Burnout Inventory (MBI) [8], a well-established tool, reported the highest rate of severe burnout (20%) as well as the highest prevalence of EE (61%) and DP (48%). Its detailed assessment of EE and DP makes it particularly effective in detecting these core components of burnout. Meanwhile, the MBI-HSS [3], which focuses on healthcare workers, reported the highest prevalence of low PA (58%), highlighting its sensitivity to burnout's impact on physicians' perception of their professional efficacy.

The risk factors of burnout among doctors in China include poor psychological status, being aged 35–44, having a lower level of education, being unmarried, experiencing long working hours, having a low job satisfaction, taking less vacation, having frequent nights on call, being prone to medical disputes, experiencing a deterioration of the physician–patient relationship, earning a low income, managing a tremendous workload, perceiving a low professional value, achieving a low curative rate of patients, holding a junior professional title, and working in tertiary teaching hospitals. Recognizing burnout as a systemic healthcare issue, rather than an individual one, and urging governments, professional medical organizations, healthcare institutions, and media outlets to promote a deeper understanding of the medical landscape is crucial for effectively addressing this challenge.

Poor psychological health is the greatest predictor of burnout (OR = 3.88). Swider et at [164] noted that personality traits, such as character, temperament, and coping abilities, can influence how an individual responds to work-related stress, thereby affecting susceptibility to burnout and turnover. Educating Chinese physicians about burnout, by incorporating it into the physician's pledge to "attend to my own health," could raise awareness and encourage proactive self-care [165]. Offering activities to improve psychological health, such as self-care workshops, yoga, massage, mindfulness, meditation, Tao Te Ching, stress management, and resilience training is beneficial. Physicians aged 35–45, often considered the ‘golden period’ for professional development, are more vulnerable to burnout, potentially due to their relative inexperience with complex cases [29]. Additionally, unmarried physicians exhibit a higher prevalence of burnout [166], highlighting the importance of family support, which is crucial for emotional well-being and stress management. Family support systems may buffer the effects of stress, helping to mitigate burnout. Physicians with lower educational levels may be less aware of their professional value and may lack advanced coping strategies to manage challenging cases [166]. Therefore, ongoing medical education and training programs are essential to equip physicians with the necessary skills to handle complex cases and manage stress effectively [167].

In China, the average time spent with each new and return patient is 16.37 and 10.89 min, respectively, compared to 49.1 and 18.2 min in the U.S. [68]. Ma et al. [1] found that despite working fewer hours, Chinese doctors experience higher rates of burnout, which can negatively impact the patient experience and exacerbate the doctor-patient relationship. This deterioration increases the risk of workplace violence, reduces patient respect, and further contributes to burnout and diminished care quality. Institutions should implement effective, ongoing monitoring, feedback, and assessment tools, including burnout evaluations as part of routine mental health check-ups during physicians' physical exams, to proactively intervene and address burnout. Additionally, the fee-for-service payment model, requiring Chinese doctors to negotiate approvals before conducting procedures and examinations, complicates doctor-patient communication about stressful financial matters, adding to stress [168]. According to the theory of effort-reward imbalance, the mismatch between income and workloads is a major contributor toward job burnout. The average compensation of Chinese oncologists in 2015 was merely 1.95 times the gross domestic product per capita, whereas this proportion is significantly higher worldwide, and even more so in the U.S. China, with 20% of the world’s population, only reported 3% of the world’s total health expenditure [25]. Policymakers should thus consider initiatives to balance income and workload.

Low job satisfaction and a lack of professional fulfillment are key contributors to burnout. Doctors who perceive low professional value or struggle with a low curative rate often feel ineffective, negatively impacting both their mental health and job satisfaction [3]. Additionally, doctors with junior professional titles may face challenges in career progression and recognition, which can further contribute to feelings of inadequacy and burnout [58]. Working in large, high-pressure environments like tertiary teaching hospitals, characterized by high patient loads, limited resources, and intense performance demands, significantly strains healthcare professionals' mental and physical health, with a lack of work-life balance and insufficient institutional support further exacerbating burnout [46].

Long working hours leave doctors with insufficient time for sleep, personal relationships, family activities, depleting their physical and mental energy. In 2011, the United States implemented a limit of up to 80 h per week for medical residents, while Germany, following the European Working Time Directive, capped doctors' working hours at 48 h per week. To date, there are no regulations in China that cap physician work hours. Therefore, it is urgent to reduce work hours among doctors in China. A study in the USA found that female doctors experience burnout 1.6 times more frequently than their male counterparts, due to the combined pressures of work and home responsibilities, slower career progression, and gender bias. However, studies on Dutch dentists and European family doctors [151] indicate that burnout is also prevalent among males. Additionally, the media’s inadequate and biased reporting on medical disputes and the limitations of the healthcare system may further increase burnout risk.

The strengths of the present study lie in its large sample size, and comprehensive summary of individual, occupational, organizational, and social factors related to burnout and risk factors. However, several limitations should be noted: 1) The included studies utilized a cross-sectional design, which limits causal inferences. 2) Burnout scales were based on self-reporting, which may introduce recall/report bias and subjectivity. 3) Physicians in some departments may have been more likely to participate in the survey, potentially causing a clustering effect that could affect the interpretation of results. 4) The response rates for the included studies ranged from 12.3% to 100%, introducing potential non-responder bias and affecting representativeness. 5) While the existing literature provides valuable insights into burnout in healthcare settings, there is a clear need for more focused and large-scale studies on the link between the patient-centered care model and burnout, particularly within the Chinese healthcare context. These gaps in knowledge present opportunities for future research to provide a clearer understanding of the complex relationship between patient-centered care and burnout, and to explore potential interventions to support healthcare providers in this model.

Conclusions

The overall prevalence of burnout is high among Chinese doctors, and it varies across different specialties. Personal psychological status was the greatest predictor of burnout among Chinese doctors. Regular psychological counseling, workload alleviation and income increase are recommended coping strategies.

Abbreviations

CNKI

China National Knowledge Infrastructure

CSTPC

Chinese Scientific and Technical Papers and Citation

EE

Emotional exhaustion

DP

Depersonalization

PA

Personal accomplishment

COVID-19

The corona virus disease 19

MBI

Maslach Burnout Inventory

MBI-HSS

Maslach Burnout Inventory Human Services Survey

CMBI

Chines13e Maslach Burnout Inventory

MBI-GS

Maslach Burnout Inventory-General Scale

ORs

Odds ratios

SEs

Standard errors

CIs

Confidence intervals

Authors’ contributions

YTY proposed this study. YTS, WW, ZHG were responsible for searching data, summarizing the results on tables, preparing Tables 1, 2 and Table S1-S10 and preparing the first draft of the manuscript. YCL and YTY edited the manuscript and tables, prepared Figs. 1, 2 and 3, and read and approved the final manuscript. All authors read and approved the final manuscript. The manuscript is an original study and has not been published or submitted for publication.

Funding

None.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.