Factors Associated With Utilization of Electroconvulsive Therapy During Psychiatric Hospitalization Among Children and Adolescents in China : The Journal of ECT

Secondary Logo

Journal Logo

Original Study

Factors Associated With Utilization of Electroconvulsive Therapy During Psychiatric Hospitalization Among Children and Adolescents in China

Geng, Feng MD∗,†; Wang, Song MD; Tian, Yanghua MD; Jiang, Feng MD‡,§; Conrad, Rachel MD∥,¶; Liu, Tingfang PhD#; Liu, Yuanli PhD∗∗; Mo, Daming MD††; Liu, Huanzhong MD, PhD; Tang, Yi-lang MD, PhD‡‡,§§

Author Information
The Journal of ECT ():10.1097/YCT.0000000000000898, January 17, 2023. | DOI: 10.1097/YCT.0000000000000898
  • Open
  • PAP



Electroconvulsive therapy (ECT) is well established as an effective treatment for certain psychiatric disorders, including major depressive disorder (DD), bipolar disorder (BD), schizophrenia (SZ), and catatonia, and is considered an essential procedure in psychiatry, even during the COVID-19 pandemic.1–4 However, the use of ECT in children and adolescents (C/A) patients is controversial as compared with adults. Further research is needed to evaluate concerns of potential age-related adverse effect profiles related to the use of ECT among C/A patients. Potential ethical issues may arise, such as incorporating the child into the decision-making process at a developmentally appropriate level.5 Although overall data on the safety and efficacy of ECT for C/A patients are limited, existing studies support its use in appropriate patients. A recent retrospective chart review and analysis of all patients 18 years and younger who received ECT at a university hospital from 1985 through 2016 found that ECT is a safe and effective treatment for C/A patients with certain severe psychiatric illnesses.6 Electroconvulsive therapy outcomes and adverse effects were similar to those reported in adults, particularly among patients aged 15 to 18 years. A study in Canada demonstrated that 77% of adolescent patients with treatment-resistant depression showed significant improvement after ECT.7

While several studies have found that ECT in C/A patients was overall safe with few adverse effects, ECT is underused in many countries.5,8 Despite data suggesting safety and efficacy, a recent analysis using the Nationwide Inpatient Sample database from the years 2002 to 2017 in the United States found that ECT has not been optimally used in C/A patients.9 Overall, ECT is much more commonly used in both C/A patients and adult patients in China as compared with other countries. The variation in the use of ECT may be to both stigma and regulation. Studies conducted on multiple continents demonstrate that both patients and clinicians have negative opinions about ECT.5 Clinicians' self-reports reflect their hesitation to recommend ECT. A survey of 151 C/A psychiatrists in Belgium showed that only 38% of them would recommend ECT for an adolescent.10 Although 66% of C/A psychiatrists surveyed in Turkey considered ECT safe for adolescents, only 34% had prescribed ECT for their patients.11 One retrospective study in India showed that merely 22 C/A patients received ECT treatment over a period of 10 years among all patients in a C/A psychiatric hospital and the most common reason was catatonia.12 In another study, C/A patients were only 0.2% of the 7469 psychiatric patients who received ECT in Australia.13

One reason for the underutilization of ECT in many countries may be a result of stigma due to negative portrayals in the popular press and alarming misrepresentation of the procedure in movies.14,15 Attitudes toward ECT may vary between geographic regions based on the degree of misinformation about ECT in the media, and the use of ECT may be lower in certain countries because of both patient and clinician media exposure.

Regulation is another barrier specific to the use of ECT among C/A patients in some countries.5 For example, in certain states within the United States, the use of ECT is illegal in children. In other states, the use of ECT in C/A patients requires the consultation of 2 or 3 psychiatrists. These regulations create barriers to access for patients for whom ECT may be safe and effective, perpetuate stigma, and broadly impact standard clinical practice.

The practice parameter of the American Academy of Child and Adolescent Psychiatry suggests the following 3 eligibility criteria for ECT in adolescent patients: diagnosis, symptom severity, and lack of treatment response to an appropriate pharmacotherapy (lack of response to 2 or more trials of pharmacotherapy) accompanied by other appropriate therapies.16 Electroconvulsive therapy could be a valuable tool in the treatment of severe psychiatric conditions in C/A when used judiciously.5

Two recent retrospective studies focused on the use of ECT in adolescent psychiatric patients in China. One reported that 42.6% of adolescent patients at the National Clinical Research Center of Mental Disorders received ECT during a period of 8 years.17 Older age, high risk of aggression, use of antipsychotics, and use of antidepressants were significantly associated with use of ECT.17 Another study, based on samples at the same site, demonstrated that 49.2% of adolescents with SZ received ECT during a period of 10 years. Male sex and high risk of suicide were independent factors associated with the use of ECT.18 The high rates of ECT in these 2 studies may not be representative of typical use in China because of the unique patient population and resources at the National Clinical Research Center of Mental Disorders. Complex treatment-resistant patients from other cities and patients who need ECT are often transferred to this tertiary referral center when locally available treatment has been ineffective.

As a part of a nationwide survey, we collected clinical data on patients discharged from 41 tertiary psychiatric hospitals in China, and we specifically aimed to examine the demographic and clinical correlates of the utilization of ECT in C/A patients during hospitalization.


Setting and Subjects

This study was a part of a large research project (the National Survey for the Evaluation of Psychiatric Hospital Performance), which gathers clinical data from 41 tertiary psychiatric hospitals in 29 provinces in Mainland China.19 Our data analysis included all patients who had a psychiatric hospitalization and were discharged from these hospitals between March 19 and 31, 2019. Data were retrieved from semistructured interviews and discharge medical records, which included age, sex, primary reasons for admission (multiple choices), primary diagnosis according to the International Classification of Diseases, Tenth Revision, Global Assessment of Functioning (GAF) score on admission, length of stay (LOS), voluntary or involuntary status, history of prior psychiatric hospitalizations, history of aggressive behaviors, and either self-injury or suicide attempts during the hospitalization. The International Classification of Diseases, Tenth Revision, diagnoses at discharge were collapsed into 4 hierarchical groups: SZ, BDs, DDs, and all other diagnoses.

The study protocol was approved by the ethics committee of Chaohu Hospital of Anhui Medical University (no. 201903-kyxm-02).

Statistical Analysis

The data were analyzed using SPSS version 22.0 software (IBM Corp, Armonk, NY). Summary statistics were used to describe the data. Sociodemographic and clinical characteristics in various subgroups were conducted using the Mann-Whitney U test as appropriate. A χ2 test was used to compare categorical variables. Logistic regression was used to identify independent demographic and clinical correlates of ECT with the “enter” method in the whole sample. All the tests were 2-sided, and statistical significance was defined as P < 0.05.


A total of 196 C/A patients including 85 boys (43.4%) and 111 girls (56.6%) who met the study criteria were included in the analyses. The mean age was 15.28 ± 1.79 years, and the mean LOS was 32.53 ± 22.42 days. The 3 most common discharge diagnoses were DDs (31.1%), SZ (29.6%), and BDs (16.3%). Among 196 C/A patients, 30 (15.3%) received ECT, and the mean number of treatments was 7.03 ± 4.642 times, ranging from 1 to 18 times. During the same period, 2484 adult patients were discharged and 363 patients (14.6%) of them received ECT.

Between ECT and non-ECT groups, we found significant differences in sex, age, LOS, GAF total on admission, self-injury behaviors during hospitalization, and the ratio of marked improvement on discharge (Table 1).

TABLE 1 - Clinical and Demographic Features of the Study Sample
Variables The Whole Sample (N = 196) Non-ECT Group (n = 166) ECT Group (n = 30) χ2 P
Female 111 (56.6%) 100 (60.2%) 11 (36.7%) 5.749 0.016
First admission (yes) 136 (69.4%) 113 (68.1%) 23 (76.7%) 0.884 0.347
Primary reason of admission
 Psychotic symptoms 92 (46.9%) 75 (45.2%) 17 (56.6%) 1.346 0.246
 Depressive symptoms 84 (43.3%) 74 (44.6%) 11 (36.7%) 0.648 0.421
 Manic symptoms 27 (13.8%) 23 (13.9%) 4 (13.3%) 0 1
 Aggressive behavior 28 (14.3%) 23 (13.9%) 5 (16.7%) 0.164 0.686
 Self-injurious and suicidal behavior 34 (17.3%) 26 (15.7%) 8 (26.7%) 2.146 0.143
 Others 7 (3.6%)
Involuntary admission (yes) 63 (32.1%) 51 (30.7%) 12 (40%) 1.003 0.317
Aggressive behavior during hospitalization 62 (31.6%) 52 (31.3%) 10 (33.3%) 0.047 0.828
Self-injury/suicidal behavior during hospitalization 24 (12.2%) 15 (9%) 9 (30%) 10.392 0.001
Discharge Diagnosis
 SZ 58 (29.6%) 47 (28.3%) 11 (36.7%) 0.963 0.81
 BD 61 (31.1%) 53 (31.9%) 8 (26.7%)
 BD 32 (16.3%) 27 (16.3%) 5 (16.7%)
 Others 45 (23%) 39 (23.5%) 6 (20%)
Marked improvement on discharge 103 (52.6%) 81 (48.8%) 22 (73.3%) 6.135 0.013
Mean Mean Mean Z P
Age, y 15.28 ± 1.79 (6–17) 15.16 ± 1.83 (6–17) 15.93 ± 1.36 (13–17) −2.319 0.02
Mean LOS, d 32.53 ± 22.42 31.6 ± 23.25 37.33 ± 16.59 −2.788 0.005
GAF on admission 47.67 ± 18.76 49.2 ± 18.51 36.77 ± 17.92 −2.999 0.003
GAF on discharge 72.07 ± 15.97 71.9 ± 16.36 72.57 ± 13.85 −0.105 0.916

Multiple logistic regression analysis revealed that ECT use was significantly associated with an older age, male sex, lower GAF scores on admission, and SIB during hospitalization (Table 2). The full model containing all predictors was statistically significant (omnibus χ2 = 35.737, P < 0.01). The model as a whole explained between 16.7% (Cox and Snell R2) and 29.0% (Nagelkerke R2) of the variance and correctly classified 87.2% of cases.

TABLE 2 - Factors Associated With ECT Use in Youth Inpatients
B SE Wald df OR 95% CI for OR
Lower Upper
Age 0.417 0.173 5.816 1 1.517* 1.081 2.128
Sex (reference female) 1.252 0.513 5.950 1 3.496* 1.279 9.557
GAF on admission −0.036 0.013 7.397 1 0.964† 0.939 0.990
Involuntary admission −0.323 0.546 0.351 1 0.724 0.248 2.109
First admission 0.316 0.549 0.330 1 1.371 0.467 4.024
Primary reason of admission
 Psychotic symptoms (reference no) 0.526 0.621 0.717 1 1.692 0.501 5.717
 Depressive symptoms (reference no) −0.615 0.790 0.607 1 0.540 0.115 2.542
 Manic symptoms (reference no) −0.624 0.935 0.444 1 0.536 0.086 3.352
 Aggressive behaviors (reference no) −0.033 0.675 0.002 1 0.968 0.258 3.635
 Self-injurious and suicidal behavior (reference no) 0.834 0.637 1.715 1 2.303 0.661 8.025
Aggressive behaviors during hospitalization −0.703 0.562 1.563 1 0.495 0.164 1.491
Self-injury/suicidal behavior during hospitalization 1.940 0.638 9.242 1 6.960† 1.992 24.312
Diagnosis (reference SZ) 0.751 3
 BD 0.795 1.067 0.554 1 2.214 0.273 17.930
 DD 0.189 0.857 0.049 1 1.208 0.225 6.482
 Other −0.015 0.682 0.001 1 0.985 0.259 3.745
 Constant −7.798 2.942 7.026 1 0
*P < 0.05.
P < 0.01.
CI indicates confidence interval.


Based on a nationwide epidemiological survey, we examined the factors associated with the utilization of ECT in C/A patients during hospitalization in China. We found that the overall utilization rate in C/A patients was 15.3%, which is comparable with the rate in adults (14.6%) in the same survey. We found significant differences between ECT and non-ECT groups in age, sex, LOS, GAF scores on admission, treatment outcome, and SIB during hospitalization. SIB during hospitalization was most significantly associated with the use of ECT and encompasses both suicidal self-injury (SSI) and non-SSI.

The frequency of ECT use in C/A patients in our study was lower than in 2 previous studies in China, which were 42.6% and 49.2%.17,18 This is likely due to variation in patient populations and the practice preferences of different centers. The previous 2 studies were based on data from the National Clinical Research Center of Mental Disorders, where patients are often referred from other centers because of either treatment-resistant psychiatric illness or a specific indication for ECT. In a third study, conducted at a tertiary psychiatric hospital in Guangzhou, China, 17.8% of 13,831 patients received ECT during psychiatric hospitalization, although neither the patients' age nor the proportion of C/A patients were reported in that study.20

The number of ECT sessions in our study was similar to those reported in adult patients in China and the recommended number of sessions (8–12 sessions) according to the most recent expert consensus for ECT use among adult patients in China.21 The most recent Expert Consensus on Electroconvulsive Therapy was published in 2019, and it recommends 8 to 12 sessions in a course of treatment while considering other clinical factors including age, diagnosis, and the severity of symptoms, as well as history of prior ECT treatment. This consensus extends to patients ranging from ages 13 to 70 years when considered medically appropriate. The 2019 expert consensus differs from other guidelines, which recommend individualized treatment frequency and no cutoffs for a lower age limit. The youngest patients who received ECT in our survey were 13 years, which is consistent with the recommendations from the most recent expert consensus in China.21

The frequency of ECT use was not statistically significant among diagnostic groups in our sample. Schizophrenia was the most common diagnosis among patients who received ECT (36.7%), followed by DDs (26.7%) and BDs (16.7%). Patients with SZ were more likely to receive ECT, although this was not statistically significant. Nearly one fifth of patients with SZ (19.1%), 15.6% of patients with BDs, and 13.1% of patients with DD received ECT.

An older age was associated with the use of ECT in our results, which is similar to a previous study in China.17 As mentioned earlier, the expert consensus in China suggests that ECT should not be considered for patients below the age of 13 years,21 which is specific to China but not endorsed by international guidelines. Although there were no long-term effects of ECT on brain development and risk of neurocognitive impairment, the potential effect on learning may be considered in younger patients.21–23 However, the adverse effects on learning are not supported in the previous literature.

Male patients were more likely to receive ECT. One previous study of the use of ECT among adolescent patients with SZ demonstrated that ECT use was independently and positively associated with the male sex.18 Our study found that more females were diagnosed with a DD and slightly more males were diagnosed with SZ. This is consistent with a prior study, which showed that girls were significantly more likely to be diagnosed with depression by the age of 12 years old and the odds ratio (OR) further increased during adolescence (OR, 2.37 at age 12 years; OR, 3.02 at age 13–15 years).24 This may partly explain findings in our study that more male C/A patients received ECT. The hospitals in our survey also often treated patients with severe psychiatrical illness or patients referred from smaller local hospitals, which may have contributed to a higher rate of ECT use among patients with SZ.

Patients who received ECT had significantly lower GAF scores on admission, which is consistent with available ECT guidelines indicating that ECT is appropriate for severely impaired patients.5,16

SIB during hospitalization was most significantly associated with the use of ECT (OR, 6.960; P < 0.01), while aggressive behavior was not. Aggressive, self-injurious, and suicidal behaviors need immediate intervention and treatment during hospitalization. These were listed as indications for ECT treatment in many guidelines.5,16,21 There are several possible explanations for this finding. First, psychiatrists in China may prefer to reserve ECT for patients with SIB. Second, other interventions including medications, restraint, and seclusion may be preferred for patients with aggressive behavior. In fact, we previously reported that the rate of restraints and seclusion in this sample was as high as 29.1%,25 which calls for urgent and careful examination of the practice.


A few limitations of the study need to be mentioned. First, as a retrospective study, the diagnosis and consent details of each patient could not be independently verified. According to the Mental Health Law that came into effect in 2013, informed consent is required for all patients before ECT can be initiated.26 Similarly, we were unable to confirm the use of ECT for each patient was appropriate or whether it reflected overuse based on the existing protocols or guidelines. Cross-national studies are necessary to fully understand the pattern of ECT use. Second, some potentially relevant demographic data including family income, insurance coverage, and clinical data including structured symptom assessment and ECT treatment parameters (dosing, electrode placement, etc) were not available. Third, SSI is a widely accepted indication for ECT, while non-SSI often fails to respond to ECT. Unfortunately, our survey did not have data to distinguish them.


Based on a representative national data set, we demonstrated the current status of ECT utilization in C/A psychiatric inpatients in China. Compared with previous studies in Western countries, our results presented a relatively frequent rate of ECT use (15.3%) in C/A patients, and we found that the overall rate of ECT in C/A inpatients was comparable with that in adult inpatients from the same study sample. This finding is unprecedented in the literature, suggesting a shift in the understanding of, and attitude toward ECT use in C/A patients, leading to similar access to and acceptability of ECT in this patient population. Moreover, some factors including age, sex, severity, and SIB during hospitalization were significantly associated with ECT use in this sample. While ECT may be effective and appropriate for C/A patients, more research is needed. Developing more specific and more operational criteria to guide ECT for C/A patients, with a focus on the long-term effects of ECT in these patients, will support the optimal use of ECT in C/A psychiatric patients.

List of 41 provincial psychiatric hospitals in study: Sichuan Mental Health Center (Mianyang Third Peoples' Hospital), Hebei Mental Health Center (Hebei Sixth Peoples' Hospital), Nanjing Brain Hospital (Jiangsu Mental Health Center), Hefei Fourth Peoples' Hospital (Anhui Mental Health Center), Ningxia Mental Health Center, Harbin First Specialized Hospital, Urumqi Fourth People's Hospital, Liaoning Mental Health Center (Liaoning Third Peoples' Hospital), Inner Mongolia Autonomous Region Mental Health Center, Heilongjiang Neuropsychiatric Hospital (The Third Hospital of Heilongjiang Province), Tianjin Anding Hospital, Taiyuan Psychiatric Hospital (Shanxi Mental Health Center), Jiangxi Psychiatric Hospital, Chengdu Fourth People's Hospital (Chengdu Mental Health Center), Guangzhou Huiai Hosptial (Guangzhou Mental Health Center), Hangzhou Seventh People's Hospital, Zhejiang Mental Health Center, Chongqing Mental Health Center, Guizhou Mental Health Center (Guizhou Second People's Hospital), Nanning Fifth People's Hospital, Hainan Mental Health Center, Henan Psychiatric Hospital, Xi'an Mental Health Center, The Eighth Hospital of Shijiazhuang, Jilin Neuropsychiatric Hospital, Guangxi Brain Hospital, Zhengzhou Eighth People's Hospital (Zhengzhou Mental Health Center), Changchun Sixth Hospital (Changchun Psychological Hospital), Shanghai Mental Health Center, Shengyang Mental Health Center, Hunan Brain Hospital, Yunnan Psychiatric Hospital, The Sixth Hospital of Peking University, Shandong Mental Health Center, Beijing Anding Hospital, Chaohu Hospital of Anhui Medical University, Beijing Huilongguan Hospital, People's Hospital of Wuhan University (Hubei Mental Health Center), and Wuhan Mental Health Center.


1. Tang YL, Jiang W, Ren YP, et al. Electroconvulsive therapy in China: clinical practice and research on efficacy. J ECT. 2012;28:206–212.
2. Maixner DF, Weiner R, Reti IM, et al. Electroconvulsive therapy is an essential procedure. Am J Psychiatry. 2021;178:381–382.
3. Ying YB, Jia LN, Wang ZY, et al. Electroconvulsive therapy is associated with lower readmission rates in patients with schizophrenia. Brain Stimul. 2021;14:913–921.
4. Ghaziuddin N, Yaqub T, Shamseddeen W, et al. Maintenance electroconvulsive therapy is an essential medical treatment for patients with catatonia: a COVID-19 related experience. Front Psych. 2021;12:670476.
5. Benson NM, Seiner SJ. Electroconvulsive therapy in children and adolescents: clinical indications and special considerations. Harv Rev Psychiatry. 2019;27:354–358.
6. Pierson MD, Mickey BJ, Gilley LB, et al. Outcomes of youth treated with electroconvulsive therapy: a retrospective cohort study. J Clin Psychiatry. 2021;82:19 m13164.
7. Zhand N, Courtney DB, Flament MF. Use of electroconvulsive therapy in adolescents with treatment-resistant depressive disorders: a case series. J ECT. 2015;31:238–245.
8. Shoirah H, Hamoda HM. Electroconvulsive therapy in children and adolescents. Expert Rev Neurother. 2011;11:127–137.
9. Trivedi C, Motiwala F, Mainali P, et al. Trends for electroconvulsive therapy utilization in children and adolescents in the United States from 2002 to 2017: a nationwide inpatient sample analysis. J ECT. 2021;37:100–106.
10. De Meulenaere M, De Meulenaere J, Ghaziuddin N, et al. Experience, knowledge, and attitudes of child and adolescent psychiatrists in Belgium toward pediatric electroconvulsive therapy. J ECT. 2018;34:247–252.
11. Bilginer Ç, Karadeniz S. Knowledge, attitudes, and experience of child and adolescent psychiatrists in Turkey concerning pediatric electroconvulsive therapy. Asian J Psychiatr. 2019;46:74–78.
12. Jacob P, Gogi PK, Srinath S, et al. Review of electroconvulsive therapy practice from a tertiary child and adolescent psychiatry centre. Asian J Psychiatr. 2014;12:95–99.
13. Chanpattana W. A questionnaire survey of ECT practice in Australia. J ECT. 2007;23:89–92.
14. Euba R, Crugel M. The depiction of electroconvulsive therapy in the British press. J ECT. 2009;25:265–269.
15. McDonald A, Walter G. Hollywood and ECT. Int Rev Psychiatry. 2009;21:200–206.
16. Ghaziuddin N, Kutcher SP, Knapp P, et al. Practice parameter for use of electroconvulsive therapy with adolescents. J Am Acad Child Adolesc Psychiatry. 2004;43:1521–1539.
17. Zhang QE, Wang ZM, Sha S, et al. Common use of electroconvulsive therapy for Chinese adolescent psychiatric patients. J ECT. 2016;32:251–255.
18. Wang S, Yang C, Jia J, et al. Use of electroconvulsive therapy in adolescents with schizophrenia in China. Child Adolesc Psychiatry Ment Health. 2018;12:49.
19. Geng F, Jiang F, Conrad R, et al. Factors associated with involuntary psychiatric hospitalization of youths in China based on a nationally representative sample. Front Psych. 2020;11:607464.
20. Ma Y, Rosenheck R, Fan N, et al. Rates and patient characteristics of electroconvulsive therapy in China and comparisons with the United States. J ECT. 2019;35:251–257.
21. The Committee of Electroconvulsive Therapy and Neuromodulation, The Committee of Anesthesiology, and The Committee of Sleep Medicine of the Chinese Medical Doctor Association. Expert consensus on modified electroconvulsive therapy (2019) [in Chinese]. J Transl Med. 2019;8:129–134.
22. de la Serna E, Flamarique I, Castro-Fornieles J, et al. Two-year follow-up of cognitive functions in schizophrenia spectrum disorders of adolescent patients treated with electroconvulsive therapy. J Child Adolesc Psychopharmacol. 2011;21:611–619.
23. Ghaziuddin N, Laughrin D, Giordani B. Cognitive side effects of electroconvulsive therapy in adolescents. J Child Adolesc Psychopharmacol. 2000;10:269–276.
24. Salk RH, Hyde JS, Abramson LY. Gender differences in depression in representative national samples: meta-analyses of diagnoses and symptoms. Psychol Bull. 2017;143:783–822.
25. Geng F, Jiang F, Conrad R, et al. Elevated rates of restraint and seclusion in child and adolescent psychiatric inpatients in China and their associated factors. Child Psychiatry Hum Dev. 2021;52:939–944.
26. Phillips MR, Chen H, Diesfeld K, et al. China's new mental health law: reframing involuntary treatment. Am J Psychiatry. 2013;170:588–591.

electroconvulsive therapy; children and adolescents; China

Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc.