Association between childhood trauma and 26-week prognosis in adolescents with major depressive disorder: a cohort study Authors: Zhichun Liu, Xianlong Zhang, Zihe Xiao, Qing Zhang, Yingying Yang, Lei Xia, Huanzhong Liu Published: 2026 ABSTRACT BackgroundMajor depressive disorder (MDD) is a common, severe, and recurrent condition. As an essential factor in the occurrence and development of adolescent depression, childhood trauma may also have a specific predictive effect on the responsiveness of antidepressant treatment and play a particular guiding role in the treatment of depression in adolescents.MethodsWe conducted a 26-week follow-up survey on 150 patients. A self-administered questionnaire was used to capture general sociodemographic information about the patients. The Hamilton Rating Scale for Depression (HAMD), the Childhood Trauma Questionnaire (CTQ), the Insomnia Severity Index Scale (ISI), and the Positive and Negative Suicide Ideation Scale (PANSI) were used to assess the patients’ clinical symptoms.ResultsWe found the prevalence of emotional abuse was 52.0%, physical abuse was 20.0%, sexual abuse was 10.7%, emotional neglect was 61.3%, and physical neglect was 62.0% among adolescents with MDD. Logistic regression analysis showed that childhood trauma, particularly emotional abuse and emotional neglect (P < 0.05), was associated with lower odds of a positive treatment response at the 26-week follow-up.ConclusionThe results showed that having experienced childhood trauma, especially emotional abuse or neglect, was associated with a lower likelihood of a positive treatment response in adolescents with depression. This suggests that families, schools and society should better screen for and prevent traumatic childhood events, and provide support to adolescents exposed to trauma, to help improve outcomes with adolescent depression.Clinical trial numberNot applicable. Introduction Major depressive disorder (MDD) is common in childhood and adolescence. A systematic review and meta-analysis reported that approximately one in five children and adolescents worldwide experience depression or depressive symptoms [1]. Among children aged 9 to 17, 5% meet the diagnostic criteria for depression, as well as 3% of adolescents with a mood disorder [2]. Depressive disorders are also one of the leading causes of suicidal ideation and behavior in adolescents, with some studies indicating that about 25% of adolescents have seriously considered suicide and about 9% have made at least one suicide attempt [3]. Childhood trauma, an adverse life event experienced by an individual in childhood that exceed his or her ability to cope [4]. A study of childhood trauma showed that the incidence of emotional abuse was 16.3%, emotional neglect 18.4%, sexual abuse 12.7%, physical abuse 22.6%, and physical neglect 16.3%, with a higher proportion of girls than boys [5]. Studies have found that people who have experienced childhood trauma are more likely to develop anxiety-like or depression-like behaviors, while the prevalence of childhood trauma is also high among people with depression, with about half (45% to 63%) of people with depressive disorders having experience at least one form of childhood trauma [6]. Epidemiological studies have shown that people who have experienced childhood trauma are three times more likely to develop depression in later life than people who have not, and that the greater the degree of trauma experienced, the more likely the patient is to be accompanied by more co-morbidities and even more severe depressive symptoms [7]. A survey undertaken by the Centers for Disease Control and Prevention in 2003 showed that the more negative life events patients experienced in childhood, the more severe mental health problems they would experience in adulthood [8], and the more suicidal tendencies they would exhibit in adolescence or adulthood [9]. One study found that the more traumas adolescents experienced in childhood, the more severe their depressive symptoms were [10]. Children who were sexually abused and physically neglected during childhood had a 1.3 times higher risk of depression than those who were not abused, and they had an earlier onset and longer duration of illness. Children who experienced physical neglect were also nearly 1.3 times more likely to develop depression than those who did not [11]. Adolescents who had experienced emotional were at increased risk for depression and anxiety than healthy controls. Emotional neglect and abusive experiences that occurred early in life also significantly predicted the risk of depression in adolescence [12]. Childhood trauma has been unanimously recognized as a significant risk factor for the occurrence and development of adolescent depression. Besides increasing the susceptibility to depressive disorders, it may also have a significant impact on treatment outcomes. For instance, a study on the treatment of adolescent depression found that adolescents without a history of trauma showed more pronounced differences in treatment efficacy, while those with a history of trauma demonstrated weaker or less obvious treatment differences [13]. A longitudinal study has shown that individuals with a history of trauma have a slower rate of symptom improvement during psychotherapy compared to those without such experiences, although their long-term outcomes may eventually converge [14]. Additionally, research on interventions targeting trauma has indicated that addressing trauma-related processes can help improve depressive symptoms and comorbid issues in adolescents, further supporting the clinical significance of trauma in treatment planning [15]. However, the findings of previous studies have also been inconsistent. A previous study found that depressed patients with childhood traumatic experiences responded worse to antidepressant treatment than those without traumatic experiences [16]. Another study also said that depressed patients with childhood trauma who were treated with medication did not show as much improvement in depressive symptoms as those without childhood trauma [17]. However, some studies have also found that the presence or absence of traumatic childhood experiences did not significantly affect the efficacy of treating depression with fluoxetine or nortriptyline [18]. Although there has been an increasing number of related studies, the existing research mainly focuses on the relationship between childhood trauma and the occurrence of depression, while paying less attention to its impact on prognosis. Secondly, the results of different studies regarding whether childhood trauma predicts poorer treatment outcomes are inconsistent, and the research subjects are mostly adult populations, while longitudinal studies targeting the adolescent population are still relatively scarce. Therefore, it is of great significance to explore the relationship between childhood trauma and the prognosis of adolescent depression. In our study, we conducted a 26-week longitudinal follow-up of adolescents with MDD to investigate whether childhood trauma was associated with lower odds of a positive treatment response. We further explored which subtypes of childhood trauma might serve as risk factors for poor treatment response, with the aim of providing evidence to support individualized treatment strategies for adolescent depression. Method Participants The enrollments of this study were from the psychological counseling outpatient clinic or inpatient wards of The Fourth Affiliated Hospital of Anhui Medical University, Hefei, China. The patients were enrolled from February 2021 to January 2022, and we conducted a 26-week follow-up survey. The inclusion criteria were: (1) meet the diagnostic criteria for depressive disorder in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5); (2) Han Chinese, aged between 13 and 18 years old. The exclusion criteria were: (1) meet the diagnostic criteria for other mental disorders in the DSM-5, such as bipolar disorder and schizophrenia; (2) combining severe neurological disorders or substance abuse; (3) combining severe liver, kidney, cardiovascular diseases, or infectious diseases; (4) patients who could not cooperate with the assessment and who did not sign the informed consent form; (5) received psychological or physical therapy during the course of treatment; (6) poor medication adherence and interruptions in treatment affect the assessment of prognosis. We enrolled 150 patients at baseline, of whom 39 patients were dropped out during follow-up. Among them, 4 patients received psychological treatment due to their condition, 3 patients received physical therapy, 25 patients failed to complete the follow-up on time after discharge, and 7 patients experienced interruptions in medication during the treatment. In the end, a total of 111 patients completed the follow-up study. Before the study started, we had been approved by the Ethics Committee of The Fourth Affiliated Hospital of Anhui Medical University (202009-kyxm-04), and all patients were informed of the relevant information of the experiment by their physicians before enrollment, and signed the informed consent form after obtaining the consent of themselves and their guardians. Assessments At baseline, all participants completed a self-administered questionnaire to collect socio-demographic and clinical information, including gender, age, BMI, education level, relationship with classmates, relationship with teachers, age at onset, and duration of illness of mental illness. In addition, participants completed a series of standardized psychological scales to assess childhood trauma, insomnia symptoms, and suicidal ideation. All clinician-rated assessments were conducted independently by two psychiatrists who had received standardized training. Mental health status Childhood Trauma Questionnaire (CTQ) The scale contains 28 entries that can be categorized into 5 dimensions: emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect. The scale is scored on a 5-point scale, with total scores ranging from 25 to 125; the larger the total score, the more traumatized the patient is. It has been shown that the Chinese version of the CTQ also has positive psychometric properties, with a Cronbach alpha coefficient of 0.83 [19, 20]. Insomnia Severity Index (ISI) This scale consists of 7 entries including falling asleep, sleep maintenance, early waking up, sleep satisfaction, impact on life, and level of worry about insomnia. Each entry has a score of 0 to 4, and the total score is 0 to 28, with the higher the score, the more severe the insomnia [21]. Positive and Negative Suicide Ideation (PANSI) The scale consists of 14 entries, with scores ranging from 1 to 5 for each question, and the higher the total score, the higher the degree of suicidal ideation in the subject. The Cronbach’s alpha coefficient for this scale was 0.92, indicating that the scale has good internal consistency [22, 23]. Assessment of depressive symptoms and prognosis Hamilton Rating Scale for Depression (HAMD) This scale is the most widely used clinical scale for assessing the severity of depression in depressed patients, and needs to be evaluated and scored by a physician with specialized training. The higher the final score, the more serious the patient’s depressive symptoms [24]. The treatment regimen (type and dosage of medication used) for all enrolled patients during the follow-up period will be chosen by their treating physicians to optimize the treatment regimen according to the needs of the patient’s own condition, independent of this study. We will collect data and assess the scale at baseline and assess the severity of depression using the HAMD again at 26 weeks. The prognosis will be reflected by the rate of depression scale score reduction. Patients with a reduction of ≥ 50% in HAMD scores from baseline were defined as having a positive treatment response, whereas those with a reduction of < 50% were defined as having no significant treatment response [25, 26]. Statistical analysis All data in this study were analyzed statistically with the software SPSS 23.0. We used univariate analysis, correlation analysis, and regression analysis to further explore the differences between the two groups of patients. Demographic and clinical data were described as mean ± standard deviation (X ± S) for continuous data and as frequency (%) for categorical data. First, we compared patients who completed the follow-up with those who were lost to follow-up, as well as patients who showed a positive treatment response with those who did not. Then, the correlation between the reduction rate of HAMD score and the demographic and clinical data was analyzed. Finally, we used binary logistic regression analysis to identify the risk factors related to prognosis, including the significant factors identified in the previous analysis as independent variables. Considering that the collinearity between the CTQ total score and its factor scores may affect the results of statistical analysis, we designed two regression models. In Model 1, the total score of CTQ was included as an independent variable along with covariates in the analysis, while in Model 2, the scores of the five factors of CTQ were analyzed along with covariates. In this study, the difference was considered statistically significant at P < 0.05. Results Incidence of childhood trauma A total of 150 patients were enrolled at baseline, of which 78 (52.0%) had experienced emotional abuse, 30 (20.0%) had experienced physical abuse, 16 (10.7%) had experienced sexual abuse, 92 (61.3%) had experienced emotional neglect, and 93 (62.0%) had experienced physical neglect. Socio-demographic and clinical characteristics of patients who completed and incompleted follow-up In this study, 39 patients fell out during follow-up, and 111 patients finally completed follow-up, with a loss of follow-up rate of 26%. The results of the univariate analysis showed that there were no statistically significant differences between the follow-up group and the loss group on the factors of gender, age, CTQ total score, ISI score, PANSI score, and baseline HAMD score (Table 1). Table 1Socio-demographic and clinical characteristics of patients who completed and incompleted follow-upCompleted(N = 111)Incompleted(N = 39)T/Z/X2 P Gender n (%)2.8680.090 Male39 (35.1%)8 (20.5%) Female72 (64.9%)31 (79.5%)Age (years) Mean (SD)15.3 (1.6)15.2 (1.5)-0.6380.523BMI (kg/m2) Mean (SD)20.8 (4.6)20.0 (2.8)-0.2760.782Only child n (%)55 (49.5%)17 (43.6%)0.4110.522Education level n (%)0.1390.710 Middle school39 (35.1%)15 (38.5%) High school72 (64.9%)24 (61.5%)Relationship with classmates n (%)2.6190.270 Good19 (17.1%)7 (17.9%) Average71 (64.0%)20 (51.3%) Bad21 (18.9%)12 (30.8%)Relationship with teachers n (%)4.5800.101 Good20 (18.0%)2 (5.1%) Average73 (65.8%)32 (82.1%) Bad18 (16.2%)5 (12.8%)Age at onset (years)13.8 (2.0)13.9 (1.4)-0.5470.584Duration of illness (months)23.4 (18.6)17.6 (13.8)-1.6700.095CTQ total score Mean (SD)53.3 (10.7)50.4 (12.8)-1.3690.173 Emotional abuse12.6 (4.3)11.3 (4.7)-1.7450.081 Physical abuse7.0 (2.5)7.7 (3.3)-1.1830.237 Sexual abuse6.0 (2.0)5.3 (0.7)-1.9430.052 Emotional neglect16.9 (4.3)15.9 (5.5)-1.0420.297 Physical neglect10.8 (3.2)10.2 (3.4)-0.8870.375ISI score Mean (SD)12.5 (5.2)13.8 (5.2)1.4160.159PANSI score Mean (SD)46.0 (11.2)49.7 (11.7)1.7300.086Baseline HAMD score Mean (SD)28.4 (8.2)28.2 (7.9)-0.1690.865BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Table 1Socio-demographic and clinical characteristics of patients who completed and incompleted follow-upCompleted(N = 111)Incompleted(N = 39)T/Z/X2 P Gender n (%)2.8680.090 Male39 (35.1%)8 (20.5%) Female72 (64.9%)31 (79.5%)Age (years) Mean (SD)15.3 (1.6)15.2 (1.5)-0.6380.523BMI (kg/m2) Mean (SD)20.8 (4.6)20.0 (2.8)-0.2760.782Only child n (%)55 (49.5%)17 (43.6%)0.4110.522Education level n (%)0.1390.710 Middle school39 (35.1%)15 (38.5%) High school72 (64.9%)24 (61.5%)Relationship with classmates n (%)2.6190.270 Good19 (17.1%)7 (17.9%) Average71 (64.0%)20 (51.3%) Bad21 (18.9%)12 (30.8%)Relationship with teachers n (%)4.5800.101 Good20 (18.0%)2 (5.1%) Average73 (65.8%)32 (82.1%) Bad18 (16.2%)5 (12.8%)Age at onset (years)13.8 (2.0)13.9 (1.4)-0.5470.584Duration of illness (months)23.4 (18.6)17.6 (13.8)-1.6700.095CTQ total score Mean (SD)53.3 (10.7)50.4 (12.8)-1.3690.173 Emotional abuse12.6 (4.3)11.3 (4.7)-1.7450.081 Physical abuse7.0 (2.5)7.7 (3.3)-1.1830.237 Sexual abuse6.0 (2.0)5.3 (0.7)-1.9430.052 Emotional neglect16.9 (4.3)15.9 (5.5)-1.0420.297 Physical neglect10.8 (3.2)10.2 (3.4)-0.8870.375ISI score Mean (SD)12.5 (5.2)13.8 (5.2)1.4160.159PANSI score Mean (SD)46.0 (11.2)49.7 (11.7)1.7300.086Baseline HAMD score Mean (SD)28.4 (8.2)28.2 (7.9)-0.1690.865BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Socio-demographic and clinical characteristics of patients who completed and incompleted follow-up BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Socio-demographic and clinical characteristics of patients with and without a positive treatment response at the 26-week follow-up At the 26-week follow-up, 30 patients (27.0%) showed a positive treatment response, whereas 81 patients (73.0%) did not show a positive treatment response. The rate of positive treatment response was significantly higher in male patients than in female patients (38.5% vs. 20.8%, X² = 3.986, P = 0.046). Significant differences were also observed between the positive treatment response and non-response groups in CTQ total score, emotional abuse score, physical abuse score, sexual abuse score, emotional neglect score, physical neglect score, and PANSI score (Table 2). Table 2Socio-demographic and clinical characteristics of the groups with and without a positive treatment responsePositivetreatment response (N = 30)Non-response (N = 81)T/Z/X2 P Gender n (%)3.986 0.046 Male15 (38.5%)24 (61.5%) Female15 (20.8%)57 (79.2%)Age (years) Mean (SD)15.5 (1.3)15.2 (1.7)-0.5260.599BMI (Kg/m2) Mean (SD)21.2 (4.9)20.6 (4.5)-0.5050.614Only child n (%)13 (23.6%)42 (76.4%)0.6350.425Education level n (%)0.0590.809 Middle school10 (25.6%)29 (74.4%) High school20 (27.8%)52(72.2%)Relationship with classmates n (%)1.1390.566 Good7 (36.8%)12 (61.2%) Average18 (25.4%)53 (74.6%) Bad5 (23.8%)16 (76.2%)Relationship with teachers n (%)0.6600.719 Good6 (30.0%)14 (70.0%) Average18 (24.7%)55 (75.3%) Bad6 (33.3%)12 (66.7%)Age at onset (years)14.3 (1.6)13.6 (2.1)-1.5010.133Duration of illness (months)18.8 (12.7)25.2 (20.2)-1.3990.162Type of antidepressant use n (%)0.1610.923 Only one of SSRIs20 (26.0%)57 (74.0%) Two of SSRIs4 (30.8%)9 (69.2%) SSRIs combined with trazodone6 (28.6%)15 (71.4%)Dose of medication (mg/day)a Mean (SD)44.1 (23.1)41.8 (22.8)-0.4970.619CTQ total score Mean (SD)40.6 (6.3)57.9 (7.8)10.909 < 0.001 Emotional abuse7.7 (2.2)14.4 (3.3)-7.311 < 0.001 Physical abuse6.1 (1.8)7.3 (2.6)-2.458 0.014 Sexual abuse5.0 (0.2)6.3 (2.3)-3.970 < 0.001 Emotional neglect12.8 (3.7)18.4 (3.5)-5.714 < 0.001 Physical neglect9.0 (2.7)11.4 (3.1)-3.514 < 0.001 ISI score Mean (SD)11.3 (5.1)12.9 (5.2)1.4620.146PANSI score Mean (SD)40.1 (9.5)48.2 (11.1)3.564 0.001 Baseline HAMD score Mean (SD)28.0 (8.3)28.6 (8.2)-0.3960.692a: Dose equivalent to fluoxetineBMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Table 2Socio-demographic and clinical characteristics of the groups with and without a positive treatment responsePositivetreatment response (N = 30)Non-response (N = 81)T/Z/X2 P Gender n (%)3.986 0.046 Male15 (38.5%)24 (61.5%) Female15 (20.8%)57 (79.2%)Age (years) Mean (SD)15.5 (1.3)15.2 (1.7)-0.5260.599BMI (Kg/m2) Mean (SD)21.2 (4.9)20.6 (4.5)-0.5050.614Only child n (%)13 (23.6%)42 (76.4%)0.6350.425Education level n (%)0.0590.809 Middle school10 (25.6%)29 (74.4%) High school20 (27.8%)52(72.2%)Relationship with classmates n (%)1.1390.566 Good7 (36.8%)12 (61.2%) Average18 (25.4%)53 (74.6%) Bad5 (23.8%)16 (76.2%)Relationship with teachers n (%)0.6600.719 Good6 (30.0%)14 (70.0%) Average18 (24.7%)55 (75.3%) Bad6 (33.3%)12 (66.7%)Age at onset (years)14.3 (1.6)13.6 (2.1)-1.5010.133Duration of illness (months)18.8 (12.7)25.2 (20.2)-1.3990.162Type of antidepressant use n (%)0.1610.923 Only one of SSRIs20 (26.0%)57 (74.0%) Two of SSRIs4 (30.8%)9 (69.2%) SSRIs combined with trazodone6 (28.6%)15 (71.4%)Dose of medication (mg/day)a Mean (SD)44.1 (23.1)41.8 (22.8)-0.4970.619CTQ total score Mean (SD)40.6 (6.3)57.9 (7.8)10.909 < 0.001 Emotional abuse7.7 (2.2)14.4 (3.3)-7.311 < 0.001 Physical abuse6.1 (1.8)7.3 (2.6)-2.458 0.014 Sexual abuse5.0 (0.2)6.3 (2.3)-3.970 < 0.001 Emotional neglect12.8 (3.7)18.4 (3.5)-5.714 < 0.001 Physical neglect9.0 (2.7)11.4 (3.1)-3.514 < 0.001 ISI score Mean (SD)11.3 (5.1)12.9 (5.2)1.4620.146PANSI score Mean (SD)40.1 (9.5)48.2 (11.1)3.564 0.001 Baseline HAMD score Mean (SD)28.0 (8.3)28.6 (8.2)-0.3960.692a: Dose equivalent to fluoxetineBMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Socio-demographic and clinical characteristics of the groups with and without a positive treatment response a: Dose equivalent to fluoxetine BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Correlation analysis of factors related to treatment response Spearman correlation analysis of the HAMD reduction rates of the treated patients with demographic and clinical data revealed that higher education levels, and older age of onset were positively correlated with greater HAMD reduction rates. The higher the total score of the CTQ, the emotional abuse score, the sexual abuse score, the emotional neglect score, and the PANSI score, the smaller the HAMD reduction rate (Table 3). Table 3Correlation analysis of factors related to treatment responseHAMD score reduction rate R P Gender0.1800.059Age0.1630.086BMI-0.0510.596Only child-0.0740.442Education level0.198 0.037 Relationship with classmates0.0560.560Relationship with teachers0.1560.102Age at onset0.197 0.038 Duration of illness-0.1640.085CTQ total score-0.491 <0.001 Emotional abuse-0.484 <0.001 Physical abuse-0.1510.115Sexual abuse-0.248 0.009 Emotional neglect-0.424 <0.001 Physical neglect-0.1700.074ISI score-0.1260.188PANSI score-0.227 0.016 Baseline HAMD score0.0350.719BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Table 3Correlation analysis of factors related to treatment responseHAMD score reduction rate R P Gender0.1800.059Age0.1630.086BMI-0.0510.596Only child-0.0740.442Education level0.198 0.037 Relationship with classmates0.0560.560Relationship with teachers0.1560.102Age at onset0.197 0.038 Duration of illness-0.1640.085CTQ total score-0.491 <0.001 Emotional abuse-0.484 <0.001 Physical abuse-0.1510.115Sexual abuse-0.248 0.009 Emotional neglect-0.424 <0.001 Physical neglect-0.1700.074ISI score-0.1260.188PANSI score-0.227 0.016 Baseline HAMD score0.0350.719BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Correlation analysis of factors related to treatment response BMI: Body mass index; CTQ: Childhood trauma questionnaire; ISI: Insomnia severity index; PANSI: Positive and negative suicide ideation scale; HAMD: Hamilton depression scale Binary logistic regression analysis of factors related to treatment response Binary logistic regression analyses were conducted to identify factors associated with treatment response at the 26-week follow-up (Table 4). In Model 1, higher CTQ total scores were associated with a lower likelihood of a positive treatment response (OR = 0.691, 95% CI: 0.498–0.770, P < 0.001). In Model 2, after controlling for covariates, emotional abuse (OR = 0.449, 95% CI: 0.283–0.680, P < 0.001) and emotional neglect (OR = 0.652, 95% CI: 0.479–0.888, P = 0.007) remained independently associated with a lower likelihood of a positive treatment response. However, in the multivariate analysis, sexual abuse and PANSI scores were not significantly associated with treatment response outcomes (P > 0.05). Table 4Binary logistic regression analysis of factors related to treatment responseModel 1aModel 2bOR P 95%CIOR P 95%CILowerUpperLowerUpperEducation level Middle schoolRef---Ref---High school8.4820.0610.90879.2016.0560.1080.67454.422Age at onset1.3040.2510.8292.0511.3530.2510.8072.269CTQ total score0.691 <0.001 0.4980.770---- Emotional abuse----0.449 <0.001 0.2970.680 Sexual abuse0.2620.2830.0233.023 Emotional neglect----0.652 0.007 0.4790.888PANSI score1.0120.7950.9271.1051.0130.7870.9211.115a: Model 1 incorporates total CTQ scores into the regression equation b: Model 2 incorporates the factor scores of the CTQ into the regression equation Table 4Binary logistic regression analysis of factors related to treatment responseModel 1aModel 2bOR P 95%CIOR P 95%CILowerUpperLowerUpperEducation level Middle schoolRef---Ref---High school8.4820.0610.90879.2016.0560.1080.67454.422Age at onset1.3040.2510.8292.0511.3530.2510.8072.269CTQ total score0.691 <0.001 0.4980.770---- Emotional abuse----0.449 <0.001 0.2970.680 Sexual abuse0.2620.2830.0233.023 Emotional neglect----0.652 0.007 0.4790.888PANSI score1.0120.7950.9271.1051.0130.7870.9211.115a: Model 1 incorporates total CTQ scores into the regression equation b: Model 2 incorporates the factor scores of the CTQ into the regression equation Binary logistic regression analysis of factors related to treatment response a: Model 1 incorporates total CTQ scores into the regression equation b: Model 2 incorporates the factor scores of the CTQ into the regression equation Discussion This study mainly explored the factors influencing the prognosis of adolescents with depression and provided some theoretical support for the future treatment of adolescents with depression disorder. Our research revealed significant gender differences in treatment response among adolescents with depression. Although the correlation analysis did not show a statistically significant association between gender and treatment response rates, the p-value approached 0.05. Further studies are needed to validate the potential for better prognosis in males compared to females. Previous studies have shown that gender is related to the severity of depressive symptoms in adolescents, with girls scoring higher than boys, meaning that girls’ depressive symptoms are more severe [27]. This may be due to the differences in psychological characteristics and cognition between adolescent boys and girls, because adolescent girls are more emotionally sensitive than boys, and are more susceptible to the influence of external factors than boys, which can lead to greater mood swings [28]. A four-year cohort study also found that girls tend to experience greater interpersonal stress and exhibit stronger emotional reactivity to such stressors, leading to higher rates of depression and poorer disease outcomes among females [29]. In addition, the different body fat content and hepatic blood flow of male and female students, as well as the differences in hormone levels and metabolic enzyme activities, may cause differences in the metabolism of antidepressants between male and female patients, thus affecting the therapeutic effect [30]. We also found that patients in the positive treatment response group had significantly lower baseline suicidal ideation scores compared with those in the non-response group. This finding is consistent with previous studies showing that higher baseline suicidal ideation scores are associated with poorer treatment response and lower likelihood of improvement following treatment [31–33]. Although our study cannot determine a causal relationship between suicidal ideation and treatment outcomes, the results suggest that suicidal ideation may serve as an important clinical indicator of treatment response in adolescents with MDD. Although our findings are generally consistent with previous studies in adults showing that childhood trauma predicts poorer treatment outcomes and lower likelihood of a positive treatment response in depression [34, 35], this correlation may differ among teenagers. Adult studies typically link trauma to chronic and difficult-to-treat depression [36]. In contrast, studies of adolescents, such as the TADS trial, have found that a history of trauma moderates treatment response in adolescents, particularly during the acute treatment phase, although long-term differences are less consistent [13]. These differences may be related to the ongoing development of the emotional regulation and stress response systems during adolescence [37], and adolescents tend to have shorter disease courses and are more influenced by family and environment [38, 39]. Therefore, childhood trauma during adolescence may delay the short-term recovery rather than inevitably leading to a chronic course as it does in adults. More importantly, our study found that childhood trauma experiences, particularly emotional abuse and emotional neglect, were independently associated with a lower likelihood of a positive treatment response at the 26-week follow-up. In contrast, physical abuse, sexual abuse, and physical neglect were not significantly associated with treatment response outcomes, suggesting that different subtypes of childhood trauma may exert distinct effects on treatment response in adolescent depression. Our research results are consistent with previous studies. Previous research has shown that childhood abuse is closely associated with an increased risk of developing depression and poorer treatment outcomes [8, 18, 40, 41]. Previous studies have also found that individuals with childhood trauma experiences tend to exhibit higher stress sensitivity and poorer emotional regulation abilities, which may reduce their responsiveness to antidepressant treatment [34, 42, 43]. These findings may partly explain the relatively lower likelihood of a positive treatment response among patients with histories of emotional abuse and emotional neglect in this study. Several previous studies have further suggested that childhood trauma may influence treatment responsiveness and that patients with trauma histories could respond differently to psychotherapy and pharmacological treatment [17, 44]. However, because treatment strategies were not experimentally controlled or compared in the present study, our findings do not allow conclusions regarding the effectiveness of specific interventions. Future longitudinal and interventional studies are needed to determine whether trauma-focused psychotherapy or combined treatment approaches may improve treatment response outcomes in adolescents with MDD who have experienced childhood trauma. The potential mechanism by which childhood trauma affects the alleviation of depressive symptoms may be related to the dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis and the persistent activation of inflammation. Previous studies have shown that childhood maltreatment is associated with elevated inflammatory markers, such as interleukin-6 (IL-6), and dysfunction of the HPA axis in patients with depression [45, 46], Altered cortisol regulation may further affect brain regions involved in emotion regulation, including the amygdala, medial prefrontal cortex, and hippocampus, which are closely related to the pathophysiology of depressive disorders [47–50]. In addition, early traumatic experiences may impair stress-response systems and reduce responsiveness to antidepressant treatment, with changes in cortisol (COR) and adrenocorticotropic hormone (ACTH) being associated with poorer treatment outcomes [51–54]. However, biological indicators related to these mechanisms were not measured in the present study. Future research incorporating molecular and neurobiological markers may help further clarify the mechanisms linking childhood trauma to poorer treatment response in adolescents with MDD. The most innovative aspect of our study is that we used a longitudinal experimental design to clarify the relationship between childhood traumatic experiences and prognosis in adolescent patients with MDD. Of course, there are still some shortcomings in our study: (1) This study did not set a control group of healthy people, so it is impossible to determine whether there are differences in childhood traumatic experiences between healthy people and depressed patients; (2) Some of the questionnaires used in this study are self-assessment questionnaires, such as the CTQ, so they are easily affected by the emotional changes at that time during the patients’ filling in the questionnaires, resulting in a recall bias in the filling results; (3) The sample size of this study is comparatively small, and the study subjects are all from within Anhui province, so the results are not nationally representative; (4) In this study, we did not bother to measure biomarkers such as COR and ACTH, which may have made the results less convincing; (5) The loss of follow-up rate of this study is still relatively high, which is 26%, so what we need to do next is to further expand the sample size and try hard to reduce the loss rate; (6) This study was conducted in a natural treatment environment, and the treatment plans during the follow-up period were not standardized. Differences in drug types, dosage adjustments, and treatment compliance may have affected the relief effect. Although we attempted to partially explain these factors by including treatment-related variables in the analysis and excluding patients who received psychological or physical therapy during the follow-up period, we still could not eliminate the residual confounding factors. Conclusion This study longitudinally examined the association between childhood trauma and treatment response outcomes in adolescents with MDD over a 26-week follow-up period. The findings suggest that childhood traumatic experiences, particularly emotional abuse and emotional neglect, were associated with a lower likelihood of a positive treatment response in adolescents with MDD. These results highlight the clinical importance of routinely assessing childhood trauma experiences in adolescent patients with depression. Further longitudinal and mechanism-based studies are needed to better clarify how different subtypes of childhood trauma influence treatment response and clinical outcomes in adolescents with depression.