Mental health literacy among resident physicians in Türkiye: a neglected competency in medical training? Authors: Zeynep Aydin, Arda Borlu, Hasan Durmuş, Fevziye Çeti̇nkaya Published: 2025 ABSTRACT BackgroundMental health literacy (MHL) is an essential but often overlooked competency in medical training. Deficits in MHL among physicians may compromise both patient care and physician well-being.MethodsA descriptive, cross-sectional study was conducted with 305 resident physicians at Erciyes University Faculty of Medicine. Data were collected using the 22-item Mental Health Literacy Scale (MHLS). Group differences were analyzed using t-tests and ANOVA, with effect sizes and 95% confidence intervals (CIs) reported.ResultsThe mean MHLS score was 16.86 ± 2.88 (range: 0–22). Female residents scored higher than males (mean difference: 0.9; 95% CI: 0.2–1.6; d = 0.31). Unmarried residents had higher scores than married peers (mean difference: 0.8; 95% CI: 0.1–1.5; d = 0.28). Internal medicine residents achieved higher total scores compared to surgical residents (mean difference: 1.2; 95% CI: 0.2–2.2; η² ≈ 0.02). Psychiatry residents, including child and adolescent psychiatry, scored significantly higher in the resource subscale only (mean difference: 1.0; 95% CI: 0.2–1.8; d = 0.32). Other variables such as income level, chronic illness, or having a relative with mental illness showed no significant associations.ConclusionsResident physicians’ mental health literacy differed across demographic and specialty-related subgroups, with female, unmarried, and childless residents showing higher levels. Variation across specialties and training years suggests that both educational and experiential factors may play a role. These findings underscore the importance of integrating structured, specialty-inclusive educational strategies into residency training to address identified gaps.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12909-025-08243-y. Introduction Mental health is a fundamental component of overall health and well-being. According to the World Health Organization (WHO), mental health is not merely the absence of mental disorders but a state in which individuals realize their potential, cope with normal life stresses, work productively, and contribute to their communities [1]. Mental health literacy (MHL), a subdomain of health literacy, is defined as an individual’s knowledge and beliefs about mental health that aid in the recognition, management, or prevention of mental disorders [2, 3]. It is a multifaceted concept that includes: the ability to recognize specific mental disorders,knowledge and beliefs about risk factors and causes,knowledge and beliefs about self-help strategies,knowledge and beliefs about available professional help,attitudes that facilitate recognition and appropriate help-seeking, and.the ability to access mental health-related information [3]. the ability to recognize specific mental disorders, the ability to recognize specific mental disorders, knowledge and beliefs about risk factors and causes, knowledge and beliefs about risk factors and causes, knowledge and beliefs about self-help strategies, knowledge and beliefs about self-help strategies, knowledge and beliefs about available professional help, knowledge and beliefs about available professional help, attitudes that facilitate recognition and appropriate help-seeking, and. attitudes that facilitate recognition and appropriate help-seeking, and. the ability to access mental health-related information [3]. the ability to access mental health-related information [3]. Mental disorders are prevalent worldwide and pose a significant public health burden. A 2021 systematic review reported a high prevalence of mental disorders in the Eastern Mediterranean Region [4]. Globally, mental illnesses account for approximately 13% of the total disease burden, and this rate is projected to rise in the coming decade. Stigmatization of mental illness remains a major barrier to seeking help [5], and in many developing countries, limited access to mental health services exacerbates the issue [6]. Given the widespread nature of mental health disorders, it is essential for healthcare professionals—particularly physicians involved in direct patient care—to possess adequate mental health literacy. This competency enables early recognition of mental health problems, appropriate management, and provision of effective psychological support during routine medical care [7]. However, studies have shown that even physicians may harbor stigmatizing attitudes toward individuals with mental illnesses. For instance, a 2023 study by Alibrahim et al. among resident physicians in Syria reported significant stigma in attitudes toward mental illness [8]. Similarly, Hemingway et al. found that resident physicians held negative perceptions, particularly toward patients with recurrent psychiatric admissions [9]. Resident physicians themselves are at increased risk for mental health problems, including stress, anxiety, depression, and burnout, due to long working hours, high responsibilities, and emotionally demanding work environments. In Brazil, 19% of residents reported symptoms of depression, 16% anxiety, and 17.7% stress; burnout syndrome was observed in 63% of participants [10]. During the COVID-19 pandemic, these numbers rose dramatically, with 67.7% reporting moderate to severe depression and 52.8% reporting anxiety symptoms [11]. In Türkiye, undergraduate medical education lasts six years, and according to the National Core Curriculum, psychiatry clerkship is mandatory during the clinical training period [12]. This rotation is typically completed in the fifth year of medical school and usually lasts between four to six weeks, depending on the institution [13]. After graduation, physicians are required to take the national Medical Specialty Examination (TUS) to enter residency training. The content and duration of residency programs are regulated by the Ministry of Health and the Specialty Council, with psychiatry residency lasting four years, while surgical specialties generally require five to six years [14, 15]. Following graduation, physicians can directly pursue psychiatry, surgery, or other specialties depending on their TUS ranking, allowing for early specialization immediately after medical school [14]. Considering these findings, enhancing mental health literacy among resident physicians is not only important for improving patient care but also crucial for protecting physicians’ own mental well-being. Early recognition of symptoms and timely access to support services are key to maintaining professional competence and psychological resilience. As residents are still in training, they represent a critical population for interventions aimed at improving mental health awareness and literacy. Despite the importance of this issue, no studies have yet been identified that specifically assess the mental health literacy levels of resident physicians. This study aims to evaluate the mental health literacy of resident physicians at Erciyes University Faculty of Medicine in Kayseri, Türkiye and to explore the relationship between MHL and various demographic, professional, and personal factors. The findings are expected to provide valuable insights for developing targeted educational strategies and improving the quality of mental health services at both the individual and systemic levels. Materials and methods Study design This study employed a descriptive, cross-sectional design to assess the mental health literacy (MHL) levels of resident physicians at Erciyes University Faculty of Medicine. Study setting and duration The study was conducted among resident physicians affiliated with clinical and basic medical science departments at Erciyes University Faculty of Medicine. Data collection took place between January 1, 2024, and March 1, 2024, using a structured 43-item questionnaire. The questionnaire comprised sociodemographic items (e.g., age, gender), questions related to contact with individuals diagnosed with mental illness, and the 22-item Mental Health Literacy Scale (MHLS). Data were collected through in-person visits to the departments where the residents were actively working. Study population and sampling The study population included all resident physicians officially registered at Erciyes University Faculty of Medicine as of January 2024. According to official records obtained from the faculty dean’s office, there were 759 resident physicians at that time. Residents who voluntarily agreed to participate and provided informed consent were included in the study. Participation was entirely voluntary, and no coercive measures were applied. As data collection was conducted face-to-face, all questionnaires were completed in full. Therefore, there were no missing responses at the item level, and the final analytic sample consisted of 305 residents. Data collection instruments Data were collected using a 43-item questionnaire developed by the researchers. The instrument included three sections: Sociodemographic data (e.g., age, gender, department, years of residency),Experience with individuals with mental illness,The Mental Health Literacy Scale (MHLS). Sociodemographic data (e.g., age, gender, department, years of residency), Sociodemographic data (e.g., age, gender, department, years of residency), Experience with individuals with mental illness, Experience with individuals with mental illness, The Mental Health Literacy Scale (MHLS). The Mental Health Literacy Scale (MHLS). Mental Health Literacy Scale (MHLS) Originally developed by Jung (2016) and adapted into Turkish by Göktaş et al. (2019), the MHLS has undergone psychometric validation in the Turkish context [16, 17]. The scale includes 22 items grouped into three subscales: MHLS-1: Knowledge of symptoms and signs of mental health problems (items 1–10),MHLS-2: Beliefs and attitudes toward mental health (items 11–18),MHLS-3: Awareness of available mental health resources (items 19–22). MHLS-1: Knowledge of symptoms and signs of mental health problems (items 1–10), MHLS-1: Knowledge of symptoms and signs of mental health problems (items 1–10), MHLS-2: Beliefs and attitudes toward mental health (items 11–18), MHLS-2: Beliefs and attitudes toward mental health (items 11–18), MHLS-3: Awareness of available mental health resources (items 19–22). MHLS-3: Awareness of available mental health resources (items 19–22). Items 1 to 18 are scored using a 6-point Likert scale (strongly agree, agree, undecided, disagree, strongly disagree, don’t know), while items 19 to 22 are binary (yes/no). Responses of “strongly agree,” “agree,” or “yes” are assigned 1 point, and all other responses receive 0 points. Items 11 to 18 are reverse-scored. The total score ranges from 0 to 22, with higher scores indicating higher levels of mental health literacy. For transparency, example items from each subscale are provided in Supplementary Table 1. Data collection procedure Following informed consent and a brief explanation of the study’s purpose, questionnaires were distributed and completed in person at the participants’ workplaces. Participation was entirely voluntary. Statistical analysis Data were analyzed using IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp., Armonk, NY, USA). Normality was evaluated using skewness and kurtosis values for the MHLS total and subscale scores. Skewness values ranged from − 1.383 to − 0.176, and kurtosis values ranged from − 0.843 to 1.573. While the knowledge-focused subscale showed a mild deviation, all values were within the acceptable ± 2 range, supporting the use of parametric tests.Therefore, Independent Samples t-tests were used to compare the mean MHLS scores between two groups, while One-Way Analysis of Variance (ANOVA) was applied for comparisons across more than two groups. Effect sizes (Cohen’s d for t-tests and eta-squared for ANOVA) were reported alongside p-values. For multiple subgroup comparisons, Tukey’s HSD and LSD post-hoc tests were applied to minimize the risk of Type I error. A p-value of < 0.05 was considered statistically significant. Ethics approval and consent to participate This study was conducted in accordance with the principles of the Declaration of Helsinki. Ethics approval was obtained from the Erciyes University Social and Human Sciences Research Ethics Committee (Approval No: 533, dated 26 December 2023). Written informed consent was obtained from all participants prior to data collection. Permission to use the Turkish version of the MHLS was obtained from the original adaptation authors. Consent for publication Not applicable. Availability of data and materials The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. Results Out of a total of 759 resident physicians officially registered at Erciyes University Faculty of Medicine, 305 voluntarily participated in the study, yielding a response rate of approximately 40.2%. Non-participation was mainly due to workload and time constraints, lack of interest or perceived relevance, absence during the data collection period (e.g., leaves, rotations, exams), and the voluntary nature of the study. The mean age of the participants was 30.05 ± 4.75 years. Among them, 61.3% were female, and 55.1% were married. The majority (70.2%) were aged between 24 and 30 years. While 11.8% of the participants reported having a chronic illness, 46.9% stated that their income exceeded their expenses. Detailed sociodemographic characteristics are presented in Table 1. Table 1Sociodemographic characteristics of the resident physiciansN (305)%GenderFemale18761.3Male11838.7Age groups24–30 years21470.231–35 years6421.036 years and older278.9Marital statusMarried16855.1Not married13744.9Chronic illnessYes3611.8No26988.2Income statusIncome = Expenses11537.7Income < Expenses4715.4Income > Expenses14346.9TOTAL305100.0 Table 1Sociodemographic characteristics of the resident physiciansN (305)%GenderFemale18761.3Male11838.7Age groups24–30 years21470.231–35 years6421.036 years and older278.9Marital statusMarried16855.1Not married13744.9Chronic illnessYes3611.8No26988.2Income statusIncome = Expenses11537.7Income < Expenses4715.4Income > Expenses14346.9TOTAL305100.0 Sociodemographic characteristics of the resident physicians As shown in Table 2, most participants (75.7%) were from internal medicine specialties, and 74.8% had between 1 and 5 years of professional experience. Nearly half (45.6%) had been in residency for one year.Table 2Professional and mental health-related characteristics of the resident physiciansn (305)%Type of SpecialtyBasic Medical Sciences206.6Internal Medicine23175.7Surgical Medicine5417.7Physician Experience1–5 years22874.86–10 years5217.011 years and above258.2Residency Experience1-year resident13945.62–3 years resident10634.84 years and above6019.7Experience Working in Psychiatry Outpatient/Inpatient ClinicsDuring medical education20667.5Post-graduation278.9No experience7023.0Other20.7Presence of a Relative with a Mental IllnessYes10333.8No20266.2Close Relationship with a Patient Having a Mental Health Problem*Parent216.9Child10.3Sibling82.6Spouse31.0Relative6220.3Friend82.6Living in the Same Household with the Patient*Yes123.9No11236.7To Address Mental Health Issues**I did nothing185.9I tried to treat myself with non-medication methods3712.1I started medication on my own123.9I sought help from a healthcare professional6120.0The Mental Health Professional Preferred/Chosen First by ParticipantsFamily Physician82,7Psychiatrist23878,0Psychologist5217,0Other72,3Institutions/Organizations Participants Prefer/Would Prefer to Receive Mental Health Services FromFamily Health Center62,0Public Hospital3411,1Private Hospital5919,3University Hospital7223,6Private Practice/Clinic13143,0Other31,0Participants’ Perceived Causes of Mental Illnesses**Genetic/Organic Cause17657.7Poor Social Environment20065.6Economic Reasons8628.2Personality Traits10735.1Stressful Living Conditions12741.6Traumatic Life Events8226.9Lack of Spirituality144.6TOTAL300100.0*Participants who answered “Yes”**Participants were allowed to select more than one option Table 2Professional and mental health-related characteristics of the resident physiciansn (305)%Type of SpecialtyBasic Medical Sciences206.6Internal Medicine23175.7Surgical Medicine5417.7Physician Experience1–5 years22874.86–10 years5217.011 years and above258.2Residency Experience1-year resident13945.62–3 years resident10634.84 years and above6019.7Experience Working in Psychiatry Outpatient/Inpatient ClinicsDuring medical education20667.5Post-graduation278.9No experience7023.0Other20.7Presence of a Relative with a Mental IllnessYes10333.8No20266.2Close Relationship with a Patient Having a Mental Health Problem*Parent216.9Child10.3Sibling82.6Spouse31.0Relative6220.3Friend82.6Living in the Same Household with the Patient*Yes123.9No11236.7To Address Mental Health Issues**I did nothing185.9I tried to treat myself with non-medication methods3712.1I started medication on my own123.9I sought help from a healthcare professional6120.0The Mental Health Professional Preferred/Chosen First by ParticipantsFamily Physician82,7Psychiatrist23878,0Psychologist5217,0Other72,3Institutions/Organizations Participants Prefer/Would Prefer to Receive Mental Health Services FromFamily Health Center62,0Public Hospital3411,1Private Hospital5919,3University Hospital7223,6Private Practice/Clinic13143,0Other31,0Participants’ Perceived Causes of Mental Illnesses**Genetic/Organic Cause17657.7Poor Social Environment20065.6Economic Reasons8628.2Personality Traits10735.1Stressful Living Conditions12741.6Traumatic Life Events8226.9Lack of Spirituality144.6TOTAL300100.0*Participants who answered “Yes”**Participants were allowed to select more than one option Professional and mental health-related characteristics of the resident physicians *Participants who answered “Yes” **Participants were allowed to select more than one option Regarding experience with mental health services, 67.5% reported exposure to psychiatry clinics during their undergraduate education, while only 8.9% had experience after graduation. A total of 33.8% of participants had a relative diagnosed with a mental illness. The most commonly cited perceived causes of mental illness were poor social environment (65.6%), genetic/organic factors (57.7%), and stressful living conditions (41.6%). When asked about how they addressed personal mental health concerns, 20% reported seeking help from a healthcare professional, while 12.1% tried self-help methods without medication. The majority (78.0%) indicated that a psychiatrist would be their first choice for mental health support. Table 3 presents the scores of the Mental Health Literacy Scale (MHLS) and its subdimensions. The mean total MHLS score was 16.86 ± 2.88, and the scores of the subdimensions are shown in the table. Table 3Resident physicians’ MHL scale and subscale scoresMHLScores obtained from the scaleThe lowest and highest possible scores obtainable from the scaleMin-MaxMean ± SD (Standard Deviation)Total Score of the MHL Scale0–229–2216,86 ± 2,88Knowledge-Focused MHL0–105–109,29 ± 0,97Belief-Focused MHL0–80–84,86 ± 2,35Resource-Focused MHL0–40–42,69 ± 1,19 Table 3Resident physicians’ MHL scale and subscale scoresMHLScores obtained from the scaleThe lowest and highest possible scores obtainable from the scaleMin-MaxMean ± SD (Standard Deviation)Total Score of the MHL Scale0–229–2216,86 ± 2,88Knowledge-Focused MHL0–105–109,29 ± 0,97Belief-Focused MHL0–80–84,86 ± 2,35Resource-Focused MHL0–40–42,69 ± 1,19 Resident physicians’ MHL scale and subscale scores The internal consistency of the MHLS was also assessed. Cronbach’s α for the total scale was 0.641, indicating acceptable reliability. For the subscales, Cronbach’s α values were 0.352 for the knowledge-focused subscale and 0.764 for the belief-focused subscale. The resource-focused subscale, which includes binary items, demonstrated a KR-20 of 0.638. The relatively low Cronbach’s α for the knowledge-focused subscale (0.352) may be attributed to the limited number of items (10) and their heterogeneous content. Previous research has shown that subscales with fewer items or a broader content range tend to produce lower α values. Despite this, the subscale still provides meaningful information on knowledge-related aspects of mental health literacy. Table 4 presents the comparison of MHLS total and subscale scores across demographic and professional variables. Female residents scored on average 0.89 points higher than males on the MHLS total scale (95% CI: 0.20–1.58, p = 0.009, d = 0.31). Unmarried residents demonstrated higher total scores compared to married residents (MD = 0.80, 95% CI: 0.16–1.44, p = 0.015; d ≈ 0.28). Participants aged 31–35 years showed higher knowledge subscale scores than both younger and older groups (Table 5, MD ≈ 0.40–0.50, p = 0.008, η² ≈ 0.02), while residents with 6–10 years of experience also outperformed their peers in knowledge (Table 5, MD ≈ 0.40–0.60, p = 0.007, η² ≈ 0.03). Internal medicine residents achieved higher total scores compared to those in surgical specialties (Table 4, MD = 1.21, 95% CI: 0.18–2.24, p = 0.022, η² ≈ 0.02).Table 4Comparison of MHLS total scores by independent variables (n = 305)CharacteristicsMean ± SDt/FPEffect SizeGender Female (n = 187)17,20 ± 2,76t = 2.650.009Cohen’s d = 0.31 Male (n = 118)16,31 ± 2,98Marital Status Married (n = 168)16,50 ± 2,83t=−2.450.015Cohen’s d = 0.28 Not married (n = 137)17,30 ± 2,88Children Yes (n = 187) 16,38 ± 3,98 t=−2.27 0.023 Cohen’s d = 0.24 No (n = 118)17,16 ± 2,78Relative with Mental Illness Yes (n = 103)16,66 ± 2,99t=−0.830.406Cohen’s d = 0.10 No (n = 202)16,96 ± 2,82Age Groups 24–30 (n = 214)16,98 ± 2,75F = 1.150.317η²=0.008 31–35 (n = 64)16,76 ± 3,21 ≥36 (n = 27)16,11 ± 2,99Years of Practice 1–5 (n = 228) 16,88 ± 2,85F = 0.920.399η²=0.006 6–10 (n = 52) 17,09 ± 2,95 ≥ 11 (n = 25) 16,16 ± 2,96Specialty Type Basic Sci (n = 20)16,85 ± 2,99abF = 3.880.022η²=0.025 Internal (n = 231)17,09 ± 2,75a Surgical (n = 54)15,88 ± 3,21bResidency Duration 1 year (n = 139)17,01 ± 2,82F = 1.890.153η²=0.012 2–3 yrs (n = 106)17,02 ± 2,88 ≥ 4 yrs (n = 60)16,21 ± 2,94Psychiatry Experience During edu (n = 206)16,85 ± 2,74F = 2.290.079η²=0.022 After grad (n = 27)18,00 ± 3,06 No exp (n = 70)16,38 ± 3,10 Other (n = 2)18,50 ± 3,53 Specialization in Psychiatry Yes (n = 13) 18,30 ± 2,21 t = 1.59 0.064 η²=0.005 No (n = 292)16,79 ± 2,89*Independent t-test was used for two-group comparisons; One-Way ANOVA was applied for comparisons across more than two groups. Effect sizes are reported as Cohen’s d for t-tests and eta-squared (η²) for ANOVA. Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05Table 5Comparison of knowledge subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizeGender Female (n = 187)9,41 ± 0,84 t = 2.40 0.016 Cohen’s d = 0.29 Male (n = 118)9,11 ± 1,12Age Groups 24–30 (n = 214) 9,22 ± 1,00a F = 4.85 0.008 η²=0.031 31–35 (n = 64) 9,62 ± 0,67b ≥36 (n = 27) 9,11 ± 1,18abYears of Practice 1–5 (n = 228)9,23 ± 0,98b F = 5.07 0.007 η²=0.032 6–10 (n = 52)9,67±,0,64a ≥ 11 (n = 25)9,08 ± 1,22ab*Knowledge-focused subscale: Independent t-test was used for two-group comparisons; ANOVA for more than two groups. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Table 4Comparison of MHLS total scores by independent variables (n = 305)CharacteristicsMean ± SDt/FPEffect SizeGender Female (n = 187)17,20 ± 2,76t = 2.650.009Cohen’s d = 0.31 Male (n = 118)16,31 ± 2,98Marital Status Married (n = 168)16,50 ± 2,83t=−2.450.015Cohen’s d = 0.28 Not married (n = 137)17,30 ± 2,88Children Yes (n = 187) 16,38 ± 3,98 t=−2.27 0.023 Cohen’s d = 0.24 No (n = 118)17,16 ± 2,78Relative with Mental Illness Yes (n = 103)16,66 ± 2,99t=−0.830.406Cohen’s d = 0.10 No (n = 202)16,96 ± 2,82Age Groups 24–30 (n = 214)16,98 ± 2,75F = 1.150.317η²=0.008 31–35 (n = 64)16,76 ± 3,21 ≥36 (n = 27)16,11 ± 2,99Years of Practice 1–5 (n = 228) 16,88 ± 2,85F = 0.920.399η²=0.006 6–10 (n = 52) 17,09 ± 2,95 ≥ 11 (n = 25) 16,16 ± 2,96Specialty Type Basic Sci (n = 20)16,85 ± 2,99abF = 3.880.022η²=0.025 Internal (n = 231)17,09 ± 2,75a Surgical (n = 54)15,88 ± 3,21bResidency Duration 1 year (n = 139)17,01 ± 2,82F = 1.890.153η²=0.012 2–3 yrs (n = 106)17,02 ± 2,88 ≥ 4 yrs (n = 60)16,21 ± 2,94Psychiatry Experience During edu (n = 206)16,85 ± 2,74F = 2.290.079η²=0.022 After grad (n = 27)18,00 ± 3,06 No exp (n = 70)16,38 ± 3,10 Other (n = 2)18,50 ± 3,53 Specialization in Psychiatry Yes (n = 13) 18,30 ± 2,21 t = 1.59 0.064 η²=0.005 No (n = 292)16,79 ± 2,89*Independent t-test was used for two-group comparisons; One-Way ANOVA was applied for comparisons across more than two groups. Effect sizes are reported as Cohen’s d for t-tests and eta-squared (η²) for ANOVA. Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Comparison of MHLS total scores by independent variables (n = 305) *Independent t-test was used for two-group comparisons; One-Way ANOVA was applied for comparisons across more than two groups. Effect sizes are reported as Cohen’s d for t-tests and eta-squared (η²) for ANOVA. Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Table 5Comparison of knowledge subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizeGender Female (n = 187)9,41 ± 0,84 t = 2.40 0.016 Cohen’s d = 0.29 Male (n = 118)9,11 ± 1,12Age Groups 24–30 (n = 214) 9,22 ± 1,00a F = 4.85 0.008 η²=0.031 31–35 (n = 64) 9,62 ± 0,67b ≥36 (n = 27) 9,11 ± 1,18abYears of Practice 1–5 (n = 228)9,23 ± 0,98b F = 5.07 0.007 η²=0.032 6–10 (n = 52)9,67±,0,64a ≥ 11 (n = 25)9,08 ± 1,22ab*Knowledge-focused subscale: Independent t-test was used for two-group comparisons; ANOVA for more than two groups. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Comparison of knowledge subscale scores by independent variables (n = 305) *Knowledge-focused subscale: Independent t-test was used for two-group comparisons; ANOVA for more than two groups. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Residents in their first year of training scored higher on the belief subscale compared to those with four or more years (Table 6, MD ≈ 0.90, p = 0.040, d ≈ 0.38). Psychiatry residents, including child and adolescent psychiatry, scored significantly higher in the resource subscale only (Table 7, MD = 0.96, 95% CI: 0.34–1.58, p = 0.010, d = 0.32).Table 6Comparison of belief subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizeMarital Status Married (n = 168)4,55 ± 2,45t=−2.870.009Cohen’s d = 0.30 Not married(n = 137)5,25 ± 2,16Children Yes (n = 187)4.51 ± 2,53t=−2.140.044Cohen’s d = 0.24 No (n = 118)5.09 ± 2,20Residency Duration 1 year (n = 139)5,13 ± 2,18aF = 3.250.040η²=0.021 2–3 yrs (n = 106)4,88 ± 2,41ab ≥ 4 yrs (n = 60)4,21 ± 2,51b*Belief-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05Table 7Comparison of resource subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizePsychiatry Experience During edu (n = 206)2,68 ± 1,16abF = 2.840.038η²=0.018 After grad (n = 27)3,25 ± 1,05a No exp (n = 70)2,48 ± 1,24b Other (n = 2)3,00 ± 1,41abSpecialization in Psychiatry Yes (n = 13)3,61 ± 1,12t = 2.200.010Cohen’s d = 0.82 No (n = 292)2,65 ± 1,17*Resource-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Table 6Comparison of belief subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizeMarital Status Married (n = 168)4,55 ± 2,45t=−2.870.009Cohen’s d = 0.30 Not married(n = 137)5,25 ± 2,16Children Yes (n = 187)4.51 ± 2,53t=−2.140.044Cohen’s d = 0.24 No (n = 118)5.09 ± 2,20Residency Duration 1 year (n = 139)5,13 ± 2,18aF = 3.250.040η²=0.021 2–3 yrs (n = 106)4,88 ± 2,41ab ≥ 4 yrs (n = 60)4,21 ± 2,51b*Belief-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Comparison of belief subscale scores by independent variables (n = 305) *Belief-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Table 7Comparison of resource subscale scores by independent variables (n = 305)CharacteristicsMean ± SDt/FpEffect SizePsychiatry Experience During edu (n = 206)2,68 ± 1,16abF = 2.840.038η²=0.018 After grad (n = 27)3,25 ± 1,05a No exp (n = 70)2,48 ± 1,24b Other (n = 2)3,00 ± 1,41abSpecialization in Psychiatry Yes (n = 13)3,61 ± 1,12t = 2.200.010Cohen’s d = 0.82 No (n = 292)2,65 ± 1,17*Resource-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Comparison of resource subscale scores by independent variables (n = 305) *Resource-focused subscale: Independent t-test and ANOVA as appropriate. Effect sizes reported as Cohen’s d or η². Tukey’s HSD post-hoc test was performed to identify between-group differences following significant ANOVA results. Groups sharing different superscripts (a, b) differ significantly at p < 0.05 Other variables, including income status, presence of chronic illness, having a relative with a mental illness, or personally experiencing a mental health problem, did not show statistically significant differences in MHLS scores (all p > 0.05, trivial effect sizes). Effect sizes are reported in Tables 4, 5, 6 and 7 together with p-values. The significant differences corresponded to small-to-moderate magnitudes (e.g., gender: d = 0.31; marital status: d = 0.28; knowledge subscale differences by age: η² = 0.02). Reporting mean differences with 95% confidence intervals strengthens the interpretability of the parametric test results and places emphasis on the magnitude of observed differences rather than solely on statistical significance. No statistically significant differences were found in MHLS scores based on income level, presence of chronic illness, having a relative with a mental illness, or having personally experienced a mental health problem (p > 0.05). To address potential confounding, a multivariable linear regression analysis was conducted including gender, marital status, age, specialty type, and psychiatry work experience as predictors of MHLS total scores (N = 305). As shown in Table 8, gender (β = − 0.12, p = 0.034), marital status (β = 0.13, p = 0.022), and specialty type (β = − 0.11, p = 0.049) remained significant independent predictors after adjustment. Age and psychiatry experience were not independently associated with MHLS scores (p > 0.05). These findings indicate that the associations observed in univariate analyses largely persisted when controlling for overlapping demographic and professional characteristics. Table 8Multivariable linear regression of MHLS total scoresFactorB (95% Cl)St. Er.T p Constant18.18(16.31–20.061)0.9519.12< 0.001Gender (Male vs. Female)−0.72 (−1.38 – −0.05)0.33−2.130.034Marital status (Unmarried vs. Married)0.75 (0.11–1.39)0.322.300.022Specialty (Surgical vs. Internal medicine)−0.67 (−1.344 – −0.003)0.34−1.970.049Age−0.009 (−0.07–0.06)0.035−0.260.794Psychiatry experience (Yes vs. No)−0.13 (−0.50–0.23)0.18−0.700.48-R2: 0.051 Table 8Multivariable linear regression of MHLS total scoresFactorB (95% Cl)St. Er.T p Constant18.18(16.31–20.061)0.9519.12< 0.001Gender (Male vs. Female)−0.72 (−1.38 – −0.05)0.33−2.130.034Marital status (Unmarried vs. Married)0.75 (0.11–1.39)0.322.300.022Specialty (Surgical vs. Internal medicine)−0.67 (−1.344 – −0.003)0.34−1.970.049Age−0.009 (−0.07–0.06)0.035−0.260.794Psychiatry experience (Yes vs. No)−0.13 (−0.50–0.23)0.18−0.700.48-R2: 0.051 Multivariable linear regression of MHLS total scores -R2: 0.051 Discussion This study assessed the mental health literacy (MHL) of 305 resident physicians and examined its relationship with demographic, professional, and individual factors. Resident physicians represent a critical group due to their early involvement in patient care and the formative impact of residency on future professional attitudes. The mean total MHL score was 16.86 ± 2.88 (range: 0–22), with subscale scores of 9.29 ± 0.97 (knowledge), 4.86 ± 2.35 (belief), and 2.69 ± 1.19 (resource). Comparable studies highlight the significance of these results. In Türkiye, family physicians had a mean MHL score of 15, while family health workers scored higher across knowledge and resource subscales, likely due to closer community interaction [18]. Among urban residents, the mean MHL score was 14.76 ± 3.67 [19]. Although direct statistical comparisons were not conducted, the similarity suggests that resident physicians’ MHL is not substantially higher than that of the general population. This finding underscores that mental health remains underrepresented within medical education and aligns with reviews calling for stronger MHL and stigma-reduction components in health professions curricula [20]. Similar patterns have been observed internationally. For example, in the United Arab Emirates, only 38.6% of school nurses correctly identified psychosis and 49.35% recognized depression with suicidal ideation [21]. Pediatric hospital staff also demonstrated limited recognition of PTSD (47%), depression (53.7%), and psychosis (54.3%) [22]. In Türkiye, Öztaş and Aydoğan reported higher MHL scores among professionals with advanced education, and among nurses, dietitians, and social workers compared to midwives or health officers [23]. Likewise, a meta-analysis from China found insufficient awareness overall (81%), with psychiatric hospital staff showing higher awareness (88%) than non-psychiatric staff (68%) [24]. Overall, these findings highlight the persistent gap in mental health literacy among healthcare professionals and the need for structured, multi-level educational interventions across all clinical settings [25]. Together, these findings reflect a consistent global trend: MHL remains inadequate among healthcare professionals outside psychiatric specialties. This underscores the need for structured and standardized mental health education across all medical disciplines, supporting the relevance of the current study. In our study, female residents demonstrated significantly higher total MHL and knowledge subscale scores than males. This finding is consistent with previous studies among family physicians and health-related university students [18, 26], as well as reports of higher MHL among women in various professional groups [23]. Possible explanations include greater openness to mental health issues, more frequent help-seeking behaviors, and broader social support networks [27, 28]. Recent educational initiatives, including digital MHL modules for medical students, suggest that engagement levels and help-seeking attitudes may differ by gender [29]. However, not all studies confirm this association. For example, Göktaş et al. found no gender difference among Turkish university students [17], and another study among non-mental health professionals also reported no differences [23]. These inconsistencies imply that gender-related differences in MHL may be shaped by both cultural norms and the professional context of medical training. Unmarried residents scored higher in both total MHL and the belief subscale compared to married participants, while residents without children had higher scores than those with children. Although these differences may reflect greater flexibility and fewer competing responsibilities, alternative explanations are possible. For instance, Aykan et al. reported that healthcare workers with young children experienced elevated stress and burnout during the COVID-19 pandemic [30], which may partly explain lower scores among parents. Longitudinal data also show increased physician burnout and diminished work-life integration during recent years, highlighting the impact of family and workload factors on mental health engagement [31]. Nonetheless, these associations should be interpreted cautiously, as causal relationships cannot be established. They may reflect contextual or generational factors rather than direct effects of marital or parental status. Future studies should explore these patterns in longitudinal designs and assess whether modular training could better support physician-parents. Residents aged 31–35 scored higher on the knowledge subscale compared to those aged 24–30, suggesting that mental health knowledge may increase with age and cumulative clinical exposure. However, residents aged ≥ 36 had lower scores than the 31–35 group, indicating that this relationship is not linear. Similar findings have been reported, with Öztaş and Aydoğan showing a weak but significant positive correlation between age and MHL scores [23]. These findings suggest that while experience may enhance literacy up to a point, reduced exposure to updated educational content could lead to stagnation or decline among older cohorts [29]. Such patterns may also suggest that while professional experience initially strengthens literacy, increasing rigidity in practice or reliance on outdated knowledge over time could limit further gains. Systematic reviews confirm that without continuous education, stigmatizing or static attitudes toward mental illness can persist among clinicians [25]. This emphasizes the need for ongoing curricular renewal and lifelong professional development to minimize generational disparities. Interestingly, residents with one year of experience scored higher on the belief subscale than those with four years of experience. This may reflect more idealistic attitudes at the start of training, which could be eroded over time by workload and burnout. This pattern is consistent with research documenting gradual declines in empathy and idealism during medical training [32]. The decline may also relate to elements of the “hidden curriculum” in clinical training, whereby implicit norms and role-modeling shape attitudes to mental health [33]. Structured, stage-specific interventions—including stress management and professional identity development programs—could mitigate these effects and sustain empathy throughout residency [34]. In multivariable analysis, gender, marital status, and specialty type remained significant independent predictors of MHL scores after adjustment, whereas age and psychiatry experience did not. This suggests that structural and professional characteristics may have a stronger influence than chronological age or exposure alone. Residents in internal medicine specialties had significantly higher total MHL scores than those in surgical specialties. This contrast likely reflects differences in clinical practice patterns—internal medicine emphasizes ongoing patient care and psychosocial interaction, while surgical training is more procedure-focused and time-pressured [35]. Cultural norms within surgical fields, including lower perceived relevance of psychosocial dialogue, may further contribute and can be addressed by explicit educational strategies [33]. Within internal medicine, psychiatry residents showed higher scores on the resource subscale, likely due to direct exposure to mental health services and referral systems. However, this effect was limited to the resource domain rather than the overall scale. Taken together, psychiatry training seems to enhance resource-related literacy specifically, while internal medicine as a whole provides an advantage in total MHL. These findings highlight how clinical context and specialty orientation shape different facets of mental health literacy. In light of these findings, targeted educational interventions may help address gaps in MHL, particularly among surgical residents. Short psychiatry mini-rotations or clerkships have demonstrated measurable improvements in attitudes toward psychiatry and reductions in stigma [36]. Complementarily, online or blended MHL modules tailored to medical trainees have been shown to enhance knowledge and well-being [29]. Integrating case-based anti-stigma content within residency curricula is also supported by reviews emphasizing the effectiveness of education combined with contact-based approaches and their sustained impact over time [25]. Maintaining empathy, openness, and professionalism requires longitudinal and structured interventions rather than single-session efforts [37]. Evidence suggests that programs integrating stress management, resilience, and identity development can alleviate burnout and enhance awareness of mental health issues among physicians [38]. Continuous faculty development and mentorship are also important to reinforce these competencies and counter the hidden curriculum’s negative influence [33]. Embedding reflective practice and structured feedback within psychiatry or interdisciplinary rotations can further strengthen recognition and referral capacities [39, 40]. Finally, regular renewal of training curricula and integration of mental health literacy content across all specialties—especially those with limited patient continuity—may foster equitable understanding, reduce stigma, and improve both patient care and physician well-being [40]. Such multi-level strategies promote consistent and equitable development of MHL across disciplines. Interventions should extend beyond early training stages and continue throughout residency to prevent declines in empathy and optimism. Longitudinal, stage-specific programs addressing stress management, resilience, and professional identity are especially valuable for maintaining positive attitudes and sustaining MHL gains. Clinical exposure also plays a critical role. Residents with prior psychiatry experience or rotations scored higher in the resource subscale, consistent with previous findings that targeted training improves knowledge and attitudes [18]. Even brief interdisciplinary exposure or collaboration with mental health professionals can increase awareness of referral pathways and service structures when accompanied by structured reflection and supervision. Overall, these findings underscore the urgent need for integrated, specialty-inclusive, and longitudinal educational strategies within residency training. Addressing the influence of the hidden curriculum and ensuring continuous curricular renewal are critical to sustaining positive attitudes and preventing declines in MHL over time. Embedding mental health content throughout medical education could better equip physicians to recognize and address patients’ mental health needs while also supporting their own well-being and professional growth. Limitations This study has several limitations that should be considered when interpreting the findings. First, it was conducted in a single university hospital, which may limit the generalizability of the results to other institutions or healthcare settings. Moreover, as this study was carried out within the specific cultural and institutional context of Türkiye, the generalizability of the findings may be further constrained. Future research should aim to include multi-center, nationally representative, and cross-cultural samples to enhance external validity and to validate these findings across different healthcare environments. Second, the cross-sectional design of this study precludes any causal inference. Therefore, differences in MHL scores between groups (e.g., belief scores among senior residents) should be interpreted with caution, as they may not solely reflect developmental changes over time but could also be influenced by alternative factors such as increased workload, burnout, or systemic challenges encountered at later stages of residency. Third, although the MHLS demonstrated overall acceptable reliability in our sample (Cronbach’s α = 0.641 for the total scale; α = 0.764 for the belief-focused subscale; KR-20 = 0.638 for the resource-focused subscale), the knowledge-focused subscale yielded a relatively low internal consistency (Cronbach’s α = 0.352). This may be related to the limited number and heterogeneity of items, a factor known to reduce reliability coefficients. Nevertheless, the subscale was retained in accordance with the validated structure of the instrument, and future adaptations may consider revising or expanding this dimension. Finally, the response rate of approximately 40% (305 out of 759 eligible residents) raises the possibility of selection bias, as those with greater interest in mental health or more flexible schedules may have been more likely to participate. Despite this, the achieved sample size was sufficient for statistical analyses, and the findings provide valuable insights that can inform future research and educational strategies. Taken together, these limitations highlight the need for larger, longitudinal, multi-center, and cross-cultural studies to further strengthen the evidence base and refine the tools used to assess mental health literacy in physicians. Conclusion and recommendations This study assessed the mental health literacy (MHL) levels of resident physicians employed at a university hospital and explored the associations between MHL and various demographic, individual, and professional variables. The results revealed that residents’ overall MHL levels were comparable to those reported among the general population and primary healthcare providers. However, considering their clinical responsibilities, residents are expected to demonstrate higher levels of MHL. Certain subdomains of MHL were significantly higher among female participants, older age groups, unmarried individuals, those without children, residents in internal medicine specialties, individuals with shorter durations of residency, those who had completed psychiatry rotations, and those specializing in psychiatry. An increase in years of professional experience was associated with higher scores in the knowledge subscale, suggesting that mental health-related knowledge improves with cumulative clinical exposure. At the same time, declining scores in later stages of training may indicate the influence of the hidden curriculum or increasing rigidity in practice. These findings imply that MHL development is shaped by both experiential and structural factors, including specialty area, psychiatric exposure, and broader educational context. In light of these findings, it is recommended that MHL be systematically integrated into undergraduate and postgraduate medical education. Comprehensive, practice-oriented educational modules tailored to the needs of resident physicians—particularly those in surgical or non-psychiatric specialties—are warranted. Psychiatry rotations should be promoted across all disciplines to enhance familiarity with mental health services, referral systems, and communication with patients experiencing psychiatric problems. Addressing the hidden curriculum and ensuring curricular renewal will be crucial to sustaining positive attitudes and preventing declines in MHL during training. Moreover, sociodemographic characteristics such as parental status should be considered to ensure training accessibility and engagement. Flexible learning modalities, including online and asynchronous formats, may support participation among residents with caregiving responsibilities. Finally, given the demanding nature of residency, institutional support mechanisms that address both MHL and physician well-being are essential. There remains a notable gap in the literature regarding the MHL levels of resident physicians. As one of the few studies focusing on this group—positioned at an early stage of professional development and receptive to intervention—this research provides valuable insights and emphasizes the need for further empirical studies in this area. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.