Mental health inequalities in Türkiye: educational differences in depression risk

Abstract

Purpose

This study aims to identify the socio-demographic and health-related factors associated with depression among individuals in Türkiye and to examine whether the pattern of factors associated with depression varies across educational groups.

Methods

This study uses data from the 2022 Türkiye Health Survey conducted by the Turkish Statistical Institute. The sample consists of 22,742 individuals, including 13,070 with below-high school education and 9,672 with high school education or higher. Binary logistic regression models were applied to identify the factors associated with individuals’ likelihood of experiencing depression within the last 12 months according to educational level.

Results

Among individuals with lower educational attainment, female sex, older age, marital status, alcohol use, tobacco use, poorer self-rated general health, the presence of chronic disease, and a history of heart attack were significantly associated with depression. Among individuals with higher educational attainment, female sex, age, marital status, alcohol use, tobacco use, general health status, chronic disease, and limitations in daily activities due to health problems were significantly associated with depression.

Conclusions

Policies aimed at improving mental health outcomes should focus on reducing socioeconomic disadvantages, improving access to mental health services, and implementing targeted interventions for high-risk groups. Understanding how the factors associated with depression vary across educational levels may contribute to more effective and equitable mental health strategies.

Introduction

About 280 million individuals globally, or roughly 3.8% of the world’s population, suffer from depression, a serious public health concern [1]. In addition to impairing social functioning and increasing the risk of illness, negative psychological tendencies can also result in extreme behaviours like suicide [2]. Depression involves feeling unhappy, being unable to experience pleasure and joy, and lacking energy [3]. Negative emotions such as sadness and guilt are common, and constructive emotion regulation, such as efforts to feel better, often fails [4, 5].

Depression is a serious mental disorder associated with loss of self-worth, reduced quality of life, and impaired functioning [6]. It has also been linked to adverse educational outcomes, including reduced interest in school, poorer academic performance, increased absenteeism, and a higher risk of school dropout [7–13].

An individual’s level of education is one of the most commonly used indicators of socioeconomic status [14]. Low socioeconomic status is associated with an increased risk of psychiatric disorders, including depression, schizophrenia [15], anxiety disorders [16], and post-traumatic stress disorder [17].

Socioeconomic difficulties, such as limited access to healthcare services, are experienced by those with low levels of education [18]. As a result, people with depression and poor educational achievement are at risk on several fronts; for example, people with low educational attainment may have less access to depression screening and treatment. An accurate and appropriate assessment based on educational levels is the first step towards effective treatment. Understanding how education affects people’s responses to questions on depression is crucial, given the severity of depression as a public health concern and the vulnerability linked to low education [19].

A greater degree of knowledge could be a buffer against depression. It should come as no surprise that those with less education have greater rates of depression than people with more education [20]. Accordingly, several studies have demonstrated that lower levels of depression are linked to higher levels of education [14, 21]. There is probably a reciprocal relationship between depression and education. According to Fletcher [22], women who experienced depressive symptoms had lower rates of high school graduation, college enrolment, and four-year university enrolment. According to Bjelland, Krokstad [23], education can help prevent anxiety and sadness, although this benefit fades with time. Additionally, dropping out of school from elementary school to university has been linked to psychiatric diagnoses in general [24].

Overall health and well-being are correlated with educational achievement [19]. Since education is a resource in and of itself and the human capital it exhibits enables people to produce other resources, education is essential to well-being. Competence, or the capacity to handle circumstances well, is demonstrated by education [25]. Unlike external resources, such as employment or wealth, education is part of an individual [26]. Education plays a crucial role in determining one’s place in the stratification system and affects employment, work, earnings, and income. Emotional well-being can be maintained by overcoming poverty and financial adversity. However, years of education represent the cumulative information, skills, beliefs, and behaviours learnt in school, even if the benefits of formal education for mental health can extend well beyond the socioeconomic domain [25]. Gaining status through education can help people succeed more broadly and effectively accomplish basic objectives, such as emotional well-being [27]. These mechanisms may partly explain why educational attainment is consistently associated with better mental health outcomes.

Moreover, because resources and support are closely linked to mental health, education plays an important role in promoting psychological well-being [28]. Research has consistently shown a strong association between education and health [29], and low educational attainment has been identified as an important factor associated with depression, particularly in later life [30]. Lower educational attainment is generally associated with reduced human capital, lower perceived control, and decreased social engagement, all of which may increase vulnerability to depression [28].

Education has been associated with chronic depression in the majority of research with educational level-related factors [31]. Low educational attainment frequently surfaced as a factor associated with depression in Gilchrist and Gunn [32], which compared the findings of 17 observational studies on depression in primary health care. These investigations included two significant worldwide projects: The World Health Organisation (WHO) Collaborative Project on Psychological Problems in General Health Services [33] and the LIDO research [34]. Education is a long-term prognostic factor, despite studies reporting observation periods of less than a year [31]. Consequently, poorer education was associated with a longer duration of the current major depressive episode for at least 24 months, according to retrospective data from the STAR*D study [35]. Lower education was associated with worse outcomes at a 2-year follow-up, according to a prospective analysis by Swindle, Cronkite [36]. A four-year prospective cohort study yielded similar results [36, 37]. However, schooling has not proven to be a reliable indicator of how depression may develop in some other evaluations. This includes a meta-analysis by Licht-Strunk, van der Windt [38] and an analysis of 426 individuals with a lifetime history of major depression up to age 30 [39]; neither study found any conclusive predictors for the prognosis of late-life depression in primary care settings [38, 39].

The causal relationship between schooling and a long-term course of depression is less well understood [31]. The term “resources,” which Ross and Mirowsky [25] define as “things that help a person achieve their goals,” is frequently used in ideas linking education to psychological well-being. Consequently, professional career and economic status are greatly influenced by schooling. Additionally, the quantity of formal education reflects the accumulation of knowledge and skills, including the development of aptitudes and dispositions like analytical thinking, focus, adaptability, and persistence. As a result, problem-solving abilities for more difficult tasks are developed [31]. This gives one the ability and inspires them to mould their life. Therefore, low-educated people may be dominated by sentiments of helplessness, fatalism, and powerlessness, whereas higher levels of education strengthen the sense of personal control [25]. This suggests that educational success and emotional well-being interact at multiple levels, including material circumstances (such as the possibility of escaping poverty and financial hardship), aspects of status and respect related to social positioning, learnt skills and competencies, and perceived self-efficacy in handling difficult life situations [31].

Depression negatively impacts not only personal well-being but also learning, educational performance, and progress. Therefore, it is important to understand how this disorder affects relevant psychological processes and how it can be managed within the education system.

In Türkiye, previous studies have examined depression in relation to various demographic, socioeconomic, and health-related characteristics [40–44]. These studies have contributed to a better understanding of the prevalence of depression and its associated factors in the Turkish population. However, most of this evidence has focused on the overall determinants of depression rather than examining whether different factors are associated with depression across educational groups.

Educational attainment is an important indicator of socioeconomic position and may influence how demographic, socioeconomic, and health-related factors are associated with depression. Nevertheless, studies examining educational differences in the factors associated with depression remain limited in Türkiye. Understanding these differences may contribute to a better understanding of educational inequalities in mental health, facilitate the identification of vulnerable educational groups, and support the development of more effective strategies for the prevention of depression.

Therefore, examining the factors associated with depression according to educational level may help address an important gap in the existing literature and provide evidence to inform mental health policies and practices in Türkiye.

Against this background, this study aimed to identify factors associated with depression according to educational level using binary logistic regression. To our knowledge, this is the first known study to identify factors associated with depression among individuals with varying levels of education in Türkiye. In this study, the following research questions were addressed:

• RQ1. Is there a significant relationship between the independent variables and the experience of depression among individuals with below high school education?

• RQ2. Is there a significant relationship between the independent variables and the experience of depression among individuals with high school and above education?

Materials and methods

Data

This analysis uses survey data from the Turkish Statistical Institute’s (TurkStat) 2022 Türkiye Health Survey. The data were collected using stratified two-stage cluster sampling [45]. The study included 9,672 participants with a high school education or higher and 13,070 participants with an education level below high school. Figure 1 presents the sample selection process and illustrates how the dependent variable was defined according to participants’ responses to the depression question.

Fig. 1.

Sample selection process and determination of the dependent variable for identifying the factors associated with depression according to educational level

As shown in Fig. 1, the sample selection process was based on the availability of complete information for the variables included in the analyses. Cases with missing information on the study variables were not included in the final analytical sample. Consequently, the analyses were conducted using complete-case data, and no imputation procedure was required.

Outcome variables

The dependent variable in this study was whether respondents had experienced depression during the last 12 months. It was measured using the survey question, “Have you experienced depression during the last 12 months?”, with responses recoded as 1 for “Yes” and 0 for “No”.

Brief scales have become increasingly common in epidemiological research for assessing the physical and mental health of populations because they reduce respondent burden, facilitate administration and translation, and provide efficient indicators of global health [46]. Likewise, the adoption of single-item measures has grown substantially in recent years [47]. One such measure is self-rated mental health (SRMH), which is assessed using the question: “In general, how would you describe your mental health: Excellent, Very Good, Good, Fair, or Poor?” The earliest applications of a single-item SRMH measure date back to studies conducted during the 1970s among university students investigating personality characteristics and mental health help-seeking behaviors [48]. In 1981, the measure was incorporated into the National Institute of Health Diagnostic Interview Programme, which was developed based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) [49]. It was later adopted as part of the World Health Organization’s International Diagnostic Interview [50]. More recently, the single-item SRMH measure has been widely employed as an independent indicator of mental health in both small-scale and large-scale studies [47].

The SRMH measure has been used independently in several national epidemiological surveys, including the Canadian Community Health Survey and the Panel Study of Medical Expenditure [51]. Researchers have applied this measure to investigate a wide range of issues, including mental disorders [52], healthcare needs [53], healthcare utilization patterns [54], and adherence to treatment plans [55]. In addition, several studies have evaluated the association between SRMH and validated clinical measures used to diagnose mental health conditions [52, 56]. Recent evidence has also demonstrated that depression can be effectively assessed using a single self-report question, and the use of this approach has been documented in the current literature [40, 57–59].

Independent variables

The independent variables of the study are gender, age, marital status, employment status, general health status, presence or absence of a disease/health problem lasting 6 months or longer, having had a heart attack in the last 12 months, having had diabetes in the last 12 months, alcohol use, tobacco use, e-cigarette use, and whether treatment costs are covered by the Social Security Institution (SSI).

The selection of covariates was limited to variables available in the Türkiye Health Survey dataset. Variables identified in previous literature as potential confounders, such as household income, place of residence (urban/rural), geographic region, household composition, and access to mental health services, were not included because these variables were unavailable or were not collected in a form suitable for the present analyses. Consequently, the multivariable models were adjusted using all relevant sociodemographic, behavioral, and health-related variables available in the dataset.

Educational attainment was categorized into two groups: below high school and high school or above. This classification was adopted because high school completion is commonly regarded as an important educational threshold in the education–health relationship and is closely linked to socioeconomic resources, employment opportunities, health-related behaviors, and access to healthcare [18, 29]. The primary objective of this study was to compare the determinants of depression between individuals with lower educational attainment and those with at least a high school education. Therefore, educational attainment was analyzed using these two groups to facilitate a clear comparison between lower and higher educational attainment. Table 1 provides a detailed description of the coding and categorization of all variables included in the analyses.

Table 1.

Description and coding of the study variables

Variables	Description
Gender	Coded as 1 for female participants and 0 for male participants
Marital Status	Coded as 1 if the participant was married and 0 if unmarried.
Age	Represented by six dummy variables corresponding to the age categories. Each variable was coded as 1 if the participant belonged to the respective age category and 0 otherwise.
Work Status	Coded as 1 if the participant was employed and 0 if unemployed
Disease status	Coded as 1 if the participant had a disease or health problem lasting or expected to last for six months or longer, and 0 otherwise.
General health status	Represented by three dummy variables corresponding to the categories very good/good, moderate, and poor/very poor. Each variable was coded as 1 if the participant belonged to the respective category and 0 otherwise.
History of heart attack	Coded as 1 if the participant had experienced a heart attack and 0 otherwise.
Living with diabetes	Coded as 1 if the participant had diabetes and 0 otherwise.
Alcohol use	Coded as 1 if the participant consumed alcohol and 0 otherwise.
Tobacco use	Coded as 1 if the participant used tobacco products and 0 otherwise.
E-cigarette use	Coded as 1 if the participant used e-cigarettes and 0 otherwise.
Social Security Institution status	Coded as 1 if the participant’s medical expenses were covered by the Social Security Institution (SSI) and 0 otherwise.

To analyse the association between each variable and depression in the binary logistic regression model, both ordinal and nominal variables were established as dummy variables [60].

Statistical analysis

Utilising survey statistics, Stata 17 (Stata Corporation) effectively handled the complex sampling design and weights. A weighted analysis was conducted. The research employed binary logistic regression to identify factors associated with individuals’ experience of depression in Türkiye, based on educational attainment.

The primary reason for the binary dependent variable in the binary logit model is that the model is designed to explain a probability-based decision-making process. This model was developed to examine how individuals arrive at a choice or probability of occurrence between two opposing situations (e.g., “yes/no”). While classical linear regression models assume a continuous dependent variable, in the binary logit model, the dependent variable takes only two values, represented by the Bernoulli distribution. In this framework, the outcome for each observation is modelled as the probability 1 P(Y = 1|X) and linearised using a logit transformation. This approach guarantees that probabilities remain in the 0–1 range, eliminating the out-of-bounds estimation problem observed in linear regression [61, 62]. This structure is widely used, particularly in the social sciences, health research, and economics literature, because many behavioural outcomes are inherently binary. Therefore, the binary logit model is considered a powerful econometric tool that both theoretically preserves probability limits and is suitable for explaining two-state decision processes [62, 63].

In addition to estimating separate models for below high school and high school and above, an alternative empirical strategy could involve a pooled model with interaction terms between education and key explanatory variables. However, separate models were preferred in this study to allow for more context-specific interpretation of determinants [64–72]. This approach enables clearer identification of factors operating within each educational group/context, while acknowledging the limitations of formal statistical comparison.

The coefficients estimated in the logit model show the change that will occur in the z-score against a unit change in the independent variable. Therefore, the interpretation of the coefficients in the logit model is not meaningful. The sign of the coefficient gives the direction of the relationship between the independent variable and the probability, the size of the coefficient does not directly give the magnitude of the effect.

The marginal effect is the partial derivative of the regression equation with respect to each independent variable. In the linear regression model, the slope coefficient is equal to the marginal effect for each variable and are interpreted as the change in the dependent variable in response to a unit change in the independent variable. In nonlinear regression models such as the probit model, the slope is not constant at all points, but changes according to the observation points, that is, the marginal effect changes according to the observation points. In nonlinear regression models, the average marginal effects are interpreted, since the interpretation according to any point will not be meaningful [73].

Marginal effects can be calculated in two different ways, marginal effects for means (MEMs) and average marginal effects (AMEs). When calculating the marginal effects for the means, the mean of each variable is calculated and the marginal effects are obtained for these means. Since the mean for dummy variables is statistically invalid, the marginal effects for the means are not appropriate in models with dummy variables among the independent variables. When calculating the average marginal effects, marginal effects are calculated for all units in the data set, that is, for each observation point, and these marginal effects are averaged. Since there are categorical independent variables in the models used, average marginal effects were calculated [73].

Results

Descriptive statistics

The frequencies and percentages of the variables used in the study are shown in Table 2. The prevalence of depression during the last 12 months was 8.26% among participants with an education level below high school and 5.76% among those with a high school education or higher. Among participants with an education level below high school, 17.5% are aged 35–44. According to the study, 70.5% of individuals with a high school education level are married. Table 2 shows that 55.7% of individuals with a high school education or higher who participated in the study are employed. According to the study, 19.7% of individuals with a high school education or higher reported drinking alcohol.

Table 2.

Findings on associated with individuals’ depression rates according to their educational levels

Variables		Below high school		High school and above
		n	%	n	%
Gender	Men	5,816	44.5	5,155	53.3
	Women	7,254	55.5	4,517	46.7
Marital Status	Unmarried	3,853	29.5	3,840	39.7
	Married	9,217	70.5	5,832	60.3
Age	15–24	1,851	14.2	1,974	20.4
	25–34	1,296	9.9	2,781	28.8
	35–44	2,291	17.5	2,337	24.2
	45–54	2,566	19.6	1,335	13.8
	55–64	2,393	18.3	774	8.0
	65+	2,673	20.5	471	4.9
Work Status	Not working	9,385	71.8	4,288	44.3
	Working	3,685	28.2	5,384	55.7
Disease status	No	4,951	37.9	5,653	58.4
	Yes	8,119	62.1	4,019	41.6
General health status	Very good/good	6,807	52.1	7,397	76.5
	Moderate	4,670	35.7	2,010	20.8
	Poor/very poor	1,593	12.2	265	2.7
History of heart attack	No	12,609	96.5	9,572	99.0
	Yes	461	3.5	100	1.0
Living with diabetes	No	10,876	83.2	9,123	94.3
	Yes	2,194	16.8	549	5.7
Alcohol use	No	12,180	93.2	7,769	80.3
	Yes	890	6.8	1,903	19.7
Tobacco use	No	9,228	70.6	6234	64.5
	Yes	3,842	29.4	3438	35.5
E-cigarette use	No	12,950	99.1	9,477	98.0
	Yes	120	0.9	195	2.0
Social Security Institution status	No	1,044	8.0	587	6.1
	Yes	12,026	92.0	9,085	93.9

Model estimation

Table 3 shows that gender, age, marital status, general health status, disease status, history of heart attack, alcohol use, and tobacco use were significantly associated with depression among individuals with below-high-school education. For individuals with high-school education or higher, gender, marital status, age, employment status, general health status, disease status, alcohol use, and tobacco use were significantly associated with the likelihood of experiencing depression.

Table 3.

The results of the binary logistic regression model and marginal effects

Variables	Below high school				High school and above
	β	S.E	dy/dx	S.E	β	S.E	dy/dx	S.E
Gender (reference: men)
Women	0.690a	0.085	0.638a	0.080	0.519a	0.102	0.490a	0.096
Marital Status (reference: unmarried)
Married	-0.546a	0.079	-0.496a	0.071	-0.625a	0.111	-0.586a	0.103
Age (reference:15–24)
25–34	-0.003	0.186	-0.003	0.167	0.131	0.166	0.124	0.156
35–44	-0.095	0.167	-0.086	0.150	0.130	0.176	0.123	0.166
45–54	-0.051	0.160	-0.046	0.144	0.053	0.186	0.050	0.176
55–64	-0.221	0.159	-0.200	0.143	0.075	0.198	0.071	0.187
65+	-0.745a	0.155	-0.687a	0.141	-0.550b	0.243	-0.526b	0.233
Work Status (reference: not working)
Working	-0.089	0.097	-0.082	0.090	-0.443a	0.109	-0.417a	0.103
General health status (reference: very good/good)
Moderate	0.591a	0.092	0.552a	0.086	0.545a	0.104	0.515a	0.099
Poor/very poor	1.122a	0.107	1.027a	0.098	1.302a	0.173	1.203a	0.155
Disease status (reference: no)
Yes	3.329a	0.249	3.208a	0.246	3.407a	0.227	3.286a	0.224
History of heart attack (reference: no)
Yes	0.436a	0.133	0.393a	0.118	0.199	0.288	0.187	0.269
Living with diabetes (reference: no)
Yes	0.074	0.079	0.068	0.072	0.080	0.147	0.075	0.138
Alcohol use (reference: no)
Yes	0.559a	0.138	0.502a	0.121	0.423a	0.113	0.396a	0.105
Tobacco use (reference: no)
Yes	0.427a	0.081	0.388a	0.073	0.700a	0.101	0.657a	0.094
E-cigarette use (reference: no)
Yes	-0.602	0.441	-0.563	0.419	0.179	0.299	0.168	0.279
Social Security Institution status (reference: no)
Yes	-0.019	0.121	-0.017	0.111	-0.224	0.201	-0.210	0.188
Constant	-5.781a	0.281			-5.574a	0.299

S.E Standard Error

^ap < 0.01; ^bp < 0.05; ^cp < 0.10

The marginal effects of factors associated with individuals’ likelihood of experiencing depression according to their educational level are presented in Table 3. When examining the model established for individuals with an educational level below high school, a female individual is 63.8% more likely to experience depression than a male individual. Similarly, an individual aged 65 and over with an education level below high school is 68.7% less likely to experience depression than an individual aged 15–24 with the same education level. An individual with a high school education or below who is married is 49.6% less likely to experience depression than an individual with the same education level who is unmarried.

Table 3 shows that an individual with a lower secondary education level and moderate general health status is 55.2% more likely to experience depression than an individual with the same education level and very good/good general health status. An individual with a general health status of poor/very poor and an education level below high school has a 102.7% higher likelihood of experiencing depression compared to an individual with the same education level and a general health status of very good/good.

An individual with a high school education level and a disease/health problem lasting 6 months or longer/expected to last longer, has a 320.8% higher likelihood of experiencing depression than an individual with the same education level who does not have a disease/health problem lasting 6 months or longer/expected to last longer. An individual with a high school education or below who has had a heart attack in the last 12 months is 39.3% more likely to experience depression than an individual with the same level of education who has not had a heart attack in the last 12 months. An individual with a high school education or less who uses alcohol is 50.2% more likely to experience depression than an individual with the same level of education who does not use alcohol. An individual with a high school education or less who uses tobacco is 38.8% more likely to experience depression than an individual with the same level of education who does not use tobacco.

When examining the model established for individuals with a high school education or higher, women are 49% more likely to experience depression than men. Similarly, an individual aged 65 and above with a high school education or higher is 52.6% less likely to experience depression than an individual aged 15–24 with the same level of education. An individual with a high school education or higher who is married is 58.6% less likely to experience depression than an individual with the same level of education who is unmarried.

An individual with a high school education or higher who is employed is 41.7% less likely to experience depression than an individual with the same level of education who is not employed. Table 3 shows that an individual with a high school or higher education level and moderate general health status is 51.5% more likely to experience depression than an individual with the same education level and very good/good general health status. An individual with a high school or higher education level and poor/very poor general health is 120.3% more likely to experience depression than an individual with the same education level and very good/good general health.

An individual with a high school education or higher who has a disease or health problem lasting 6 months or longer/expected to last 6 months or longer, is 328.6% more likely to experience depression than an individual with the same level of education who does not have a disease or health problem lasting 6 months or longer/expected to last 6 months or longer. An individual with a high school education or higher who uses alcohol is 39.6% more likely to experience depression than an individual with the same level of education who does not use alcohol. An individual with a high school education or higher who uses tobacco is 65.7% more likely to experience depression than an individual with the same level of education who does not use tobacco.

To incorporate the independent variables into the binary logistic regression model, the study examined multicollinearity. Variance inflation factor (VIF) values of 5 or higher are considered to indicate moderate multicollinearity, whilst values of 10 or more are considered to indicate high multicollinearity [74]. Table 4 displays the VIF values. There are no variables in this study that lead to multicollinearity issues.

Table 4.

VIF values

Variables		VIF
		Below high school	High school and above
Gender	Men	ref.	ref.
	Women	1.40	1.18
Marital Status	Unmarried	ref.	ref.
	Married	1.47	1.55
Age	15–24	ref.	ref.
	25–34	1.93	2.24
	35–44	2.78	2.51
	45–54	2.95	2.06
	55–64	2.81	1.67
	65+	2.83	1.48
Work Status	Not working	ref.	ref.
	Working	1.51	1.41
Disease status	No	ref.	ref.
	Yes	1.81	1.53
General health status	Very good/good	ref.	ref.
	Moderate	1.71	1.13
	Poor/very poor	1.56	1.53
History of heart attack	No	ref.	ref.
	Yes	1.06	1,05
Living with diabetes	No	ref.	ref.
	Yes	1.23	1.18
Alcohol use	No	ref.	ref.
	Yes	1.11	1.10
Tobacco use	No	ref.	ref.
	Yes	1.25	1.13
E-cigarette use	No	ref.	ref.
	Yes	1.01	1.02
Social Security Institution status	No	ref.	ref.
	Yes	1.02	1.05

In binary logistic regression analysis, an odds ratio (OR) less than 1 indicates lower odds of the outcome relative to the reference group, whereas an OR greater than 1 indicates higher odds of the outcome. The estimated odds ratios (ORs) for the model are presented in Table 5.

Table 5.

Odds ratio results for the model

Variables	Below high school				High school and above
			95 % CI				95 % CI
	OR	S.E	Lower	Upper	OR	S.E	Lower	Upper
Gender (reference: men)
Women	1.994a	0.170	1.687	2.358	1.681a	0.171	1.377	2.052
Marital Status (reference: unmarried)
Married	0.579a	0.046	0.496	0.676	0.536a	0.059	0.431	0.665
Age (reference:15–24)
25–34	0.997	0.186	0.692	1.436	1.140	0.189	0.823	1.579
35–44	0.909	0.152	0.655	1.262	1.139	0.201	0.806	1.610
45–54	0.951	0.152	0.694	1.302	1.055	0.197	0.732	1.520
55–64	0.801	0.128	0.586	1.095	1.078	0.214	0.731	1.591
65+	0.475a	0.074	0.350	0.643	0.577b	0.140	0.358	0.929
Work Status (reference: not working)
Working	0.915	0.089	0.756	1.107	0.642a	0.070	0.519	0.795
General health status (reference: very good/good)
Moderate	1.806a	0.166	1.509	2.161	1.724a	0.180	1.406	2.115
Poor/very poor	3.071a	0.329	2.490	3.788	3.678a	0.635	2.622	5.160
Disease status (reference: no)
Yes	27.908a	6.936	17.147	45.422	30.173a	6.857	19.327	47.106
History of heart attack (reference: no)
Yes	1.546a	0.205	1.192	2.006	1.220	0.351	0.694	2.145
Living with diabetes (reference: no)
Yes	1.077	0.085	0.923	1.256	1.083	0.160	0.811	1.445
Alcohol use (reference: no)
Yes	1.749a	0.242	1.334	2.293	1.526a	0.172	1.223	1.903
Tobacco use (reference: no)
Yes	1.532a	0.124	1.308	1.795	2.013a	0.203	1.652	2.452
E-cigarette use (reference: no)
Yes	0.547	0.241	0.231	1.300	1.196	0.358	0.665	2.149
Social Security Institution status (reference: no)
Yes	0.982	0.118	0.775	1.244	0.799	0.161	0.539	1.186
Constant	0.003a	0.001	0.002	0.005	0.004a	0.001	0.002	0.007

a: p < 0.01; b: p < 0.05

The odds ratios (ORs) for the factors associated with experiencing depression according to educational level are presented in Table 5. When examining the model established for individuals with below-high-school education, women had higher odds of experiencing depression than men (OR = 1.994; 95% CI = 1.687–2.358). Married individuals had lower odds of experiencing depression than unmarried individuals (OR = 0.579; 95% CI = 0.496–0.676). Similarly, individuals aged 65 years and older had lower odds of experiencing depression than those aged 15–24 years (OR = 0.475; 95% CI = 0.350–0.643). Individuals reporting moderate general health had higher odds of experiencing depression than those reporting very good/good general health (OR = 1.806; 95% CI = 1.509–2.161). Likewise, individuals reporting poor/very poor general health had higher odds of experiencing depression than those reporting very good/good general health (OR = 3.071; 95% CI = 2.490–3.788). Individuals with a disease or health problem lasting or expected to last six months or longer had higher odds of experiencing depression than those without such a condition (OR = 27.908; 95% CI = 17.147–45.422). Individuals who had experienced a heart attack within the previous 12 months had higher odds of experiencing depression than those who had not (OR = 1.546; 95% CI = 1.192–2.006). Alcohol users had higher odds of experiencing depression than non-users (OR = 1.749; 95% CI = 1.334–2.293). Likewise, tobacco users had higher odds of experiencing depression than non-users (OR = 1.532; 95% CI = 1.308–1.795).

When examining the model established for individuals with a high school education or higher, women had higher odds of experiencing depression than men (OR = 1.681; 95% CI = 1.377–2.052). Married individuals had lower odds of experiencing depression than unmarried individuals (OR = 0.536; 95% CI = 0.431–0.665). Individuals aged 65 years and older had lower odds of experiencing depression than those aged 15–24 years (OR = 0.577; 95% CI = 0.358–0.929). Employed individuals had lower odds of experiencing depression than those who were not employed (OR = 0.642; 95% CI = 0.519–0.795). Individuals reporting moderate general health had higher odds of experiencing depression than those reporting very good/good general health (OR = 1.724; 95% CI = 1.406–2.115). Similarly, individuals reporting poor/very poor general health had higher odds of experiencing depression than those reporting very good/good general health (OR = 3.678; 95% CI = 2.622–5.160). Individuals with a disease or health problem lasting or expected to last six months or longer had higher odds of experiencing depression than those without such a condition (OR = 30.173; 95% CI = 19.327–47.106). Alcohol users had higher odds of experiencing depression than non-users (OR = 1.526; 95% CI = 1.223–1.903). Tobacco users had higher odds of experiencing depression than non-users (OR = 2.013; 95% CI = 1.652–2.452).

Discussion

Using binary logistic regression, this study identified factors associated with individuals’ experience of depression by educational level in Türkiye. The findings showed that gender, age, marital status, general health status, disease status, history of heart attack, alcohol use, and tobacco use were associated with the likelihood of experiencing depression among individuals with below-high-school education. Among individuals with high-school education or higher, gender, marital status, age, employment status, general health status, disease status, alcohol use, and tobacco use were associated with the likelihood of experiencing depression.

Women with a high school education level and women with a high school education level or higher are more likely to experience depression than men with the same education levels. Studies have also found that women are more likely to suffer from depression [75, 76]. Earlier, the gender ratio of depressive disorders (women: males) was greater than 1.7 for lifetime prevalence and greater than 1.4 for 12-month prevalence after the age of 18, according to numerous national and international studies [77]. Women were more likely than men to seek psychological care, according to a prior study done on a sample from Türkiye [78]. A previous study suggested that this difference may be explained by socioeconomic factors such as education, income, culture, diet, and exposure to abuse, all of which may contribute to the higher prevalence of depression among women [79]. Another study suggested that biological sex differences and fluctuations in ovarian hormones may also contribute to the higher prevalence of depression among women [80]. These findings may also reflect the influence of traditional gender roles in the Turkish context. Previous studies have suggested that gender inequality, male dominance within the family, and the relatively disadvantaged social and economic position of women remain important features of the Turkish social context, which may be associated with women’s greater vulnerability to adverse mental health outcomes [81, 82]. These gendered expectations may increase women’s psychosocial burden and chronic stress, which may in turn contribute to their greater vulnerability to depression regardless of educational attainment.

In both educational groups, individuals aged 65 years and older were significantly less likely to experience depression than those aged 15–24 years. Previous studies have also reported similar findings [83, 84]. One possible explanation is that stronger emotional relationships with adult children may be associated with higher life satisfaction and lower levels of depression among middle-aged and older adults, as suggested by previous research [85].

In both educational groups, married individuals were significantly less likely to experience depression than unmarried individuals. Previous studies have consistently reported that married individuals generally experience fewer depressive symptoms than unmarried individuals [86, 87]. One possible explanation is that marriage may be associated with emotional support and a stable family environment for mental health according to the marital protection hypothesis. In the Turkish context, where family relationships remain a central component of social life, these findings suggest that marriage was associated with lower odds of depression among the participants, which is consistent with the marital protection hypothesis in previous studies [40].

The study found that as their general health declines, people with a high school education or less and those with a high school education or more are more likely to suffer from depression. A previous study conducted in Türkiye found that health status is a very strong predictor of depressive symptoms [88]. Previous studies conducted in Türkiye support this finding. Aydın et al. (2025) found that a reciprocal relationship may exist between health/illness problems and depression in approximately 70% of women with possible major depression and 30% of men with possible major depression. Similarly, Koç [89] reported high levels of depression among older individuals suffering from chronic diseases. These findings are also consistent with the Stress Process Model, which identifies poor health as a major source of stress and suggests that such stressors are associated with a higher risk of depression [90].

Individuals with a high school education or higher who are employed are less likely to experience depression than individuals with the same level of education who are not employed. Similarly, previous studies have shown that income level is an important variable associated with depression levels [91, 92]. One possible explanation is that the protective effect of employment may be more pronounced among individuals with higher educational attainment. According to Bauldry [20], education can provide greater protection against depression, particularly by enhancing personal and socioeconomic resources. Similarly, Ross and Mirowsky [93] argued that individuals from disadvantaged backgrounds are more likely to lack a sense of mastery and self-efficacy, whereas higher educational attainment strengthens these psychological resources. Consequently, employment may be associated with better mental health outcomes among highly educated individuals by reinforcing these personal and socioeconomic advantages.

Individuals with a high school education level or below and those with a high school education level or above who have an illness/health problem lasting six months or longer/expected to last six months or longer are more likely to experience depression than individuals with the same education level who do not have an illness/health problem lasting six months or longer/expected to last six months or longer. Studies with similar findings exist [94, 95].

High school graduates who have had a heart attack within the last 12 months are more likely to suffer from depression than those who have not. Those with heart disease showed considerably greater levels of depression than those with other chronic diseases, according to a prior study done on a Turkish sample [96]. This finding is also consistent with the Stress Process Model, which identifies serious physical health conditions as major stressors that increase the risk of depression [90]. The association observed only among individuals with lower educational attainment may indicate that this group is more vulnerable to the psychological consequences of major health events because of more limited coping resources and fewer socioeconomic resources available to manage health-related stress.

Depression is more common in people with a high school education or less and in people with a high school education or more who drink than in people with the same education level who don’t drink. Other studies have revealed similar results [43, 94, 97]. However, women who received psychosocial assistance were more likely to drink alcohol, according to a prior study on a Turkish sample [98]. This finding may be explained by the Self-Medication Theory, which suggests that individuals may use alcohol to alleviate depressive emotions, although such coping behavior may ultimately increase the risk of depression over time [99].

People who smoke and have completed high school or less, as well as those who have completed high school or more, are more likely to suffer from depression than people who have completed the same amount of schooling but do not smoke. The study found that tobacco users are more likely than non-users to suffer from depression. Other research has revealed similar results [94, 97]. In the Turkish context, a previous study reported that anxiety, depression, loneliness, and feelings of being unloved were among the factors associated with smoking initiation among young people [100]. These findings further suggest a close relationship between tobacco use and poor mental health.

The observed differences between educational groups may reflect disparities in health literacy, socioeconomic resources, access to healthcare, employment opportunities, and coping strategies. Individuals with higher educational attainment may have better access to health information and mental health services and may also be more likely to recognize depressive symptoms and seek professional help. By contrast, those with lower educational attainment may experience greater socioeconomic disadvantage and chronic health problems while also facing barriers to accessing mental health care. In the Turkish context, educational inequalities are closely intertwined with employment opportunities, income, and health service utilization, which may contribute to the different patterns of depression observed across educational groups. Furthermore, differences in awareness and reporting of depressive symptoms by educational level may also have influenced the observed associations.

Conclusion

Separate analyses suggested that the pattern of associations with depression varied across educational groups. Women with a high school education or below are more likely to experience depression than men with the same education level. According to the study, as the age of individuals with a high school education or below increases, their likelihood of experiencing depression decreases. Married individuals with an education level below high school are less likely to experience depression than unmarried individuals with the same education level. According to the study, individuals with a lower than high school education level are more likely to experience depression as their overall health deteriorates. Individuals with a lower than high school education level who have a disease/health problem lasting 6 months or longer/expected to last longer than six months, are more likely to experience depression than individuals with the same education level who do not have a disease/health problem lasting 6 months or longer/expected to last longer than six months. Individuals with less than a high school education who have had a heart attack in the last 12 months are more likely to experience depression than individuals with the same level of education who have not had a heart attack in the last 12 months. Individuals with less than a high school education who use alcohol are more likely to experience depression than individuals with the same level of education who do not use alcohol. Individuals with less than a high school education who use tobacco are more likely to experience depression than individuals with the same level of education who do not use tobacco.

Women with a high school education or higher are more likely to experience depression than men with the same level of education. The likelihood of experiencing depression decreases as the age of individuals with a high school education or higher increases. Married individuals with a high school education or higher are less likely to experience depression than unmarried individuals with the same level of education. The likelihood of experiencing depression increases as the general health status of individuals with a high school education or higher deteriorates. Individuals with a high school education or higher who are employed are less likely to experience depression than individuals with the same level of education who are not employed. Individuals with a high school education or higher who have a disease or health problem lasting 6 months or longer, or expected to last that long, are more likely to experience depression than individuals with the same level of education who do not have a disease or health problem lasting 6 months or longer, or expected to last that long. Individuals with a high school education or higher who consume alcohol are more likely to experience depression than individuals with the same level of education who do not consume alcohol. Individuals with a high school education or higher who use tobacco are more likely to experience depression than individuals with the same level of education who do not use tobacco.

Mental health is becoming a higher priority for policymakers, as evidenced by its inclusion in the United Nations Sustainable Development Goals (SDGs). Because education indicates resourcefulness or the capacity to deal with events effectively, it is an essential tool on the route to well-being. The capacity to produce resources and turn things into resources is shaped by education [25].

The study’s conclusions promote the deployment of tailored, equity-oriented interventions and highlight the importance of addressing mental health from a socioeconomic perspective. Reducing mental health disparities requires addressing the underlying causes of socioeconomic disadvantage, increasing awareness, and expanding access to reasonably priced mental health care. Future longitudinal research is required to assess the long-term efficacy of interventions to promote mental health equity and to investigate the underlying relationships between socioeconomic position and mental health.

Improvements made specifically for those who are disadvantaged because of unique family structures and attitudes will have a favourable influence on many facets of life, including a may help reduce the frequency and chronicity of depression. Through empathy and relationship-building, culturally appropriate educational materials have helped alter knowledge and beliefs. Improving mental health literacy, controlling or preventing the early onset of illness, and including youth in the eradication of stigma are all promising outcomes of early intervention through research and policy in the educational system.

Limitations and future research

This study has several limitations. First, because secondary data from the Türkiye Health Survey were used, the analysis was restricted to the variables available in the dataset. Therefore, potentially important socioeconomic and contextual factors, such as household income, region of residence, urban or rural residence, household structure, access to mental health services, and previous mental health treatment, could not be included in the analysis. The absence of these variables may have resulted in residual confounding and should be considered when interpreting the findings. Second, the cross-sectional design of the study does not allow causal inferences to be drawn regarding the relationship between educational attainment and depression. Third, depression was assessed using a self-reported survey question rather than a standardized clinical diagnostic instrument. Consequently, the findings may be subject to reporting bias or misclassification. Future studies should incorporate a broader range of socioeconomic and contextual variables, use standardized measures of depression, and employ longitudinal study designs to better understand the mechanisms linking educational attainment and depression. In addition, future research should examine whether the associations between educational attainment and depression differ by gender through gender-stratified analyses or by testing interaction effects between gender and other explanatory variables. Such analyses may provide a more comprehensive understanding of gender-specific determinants of depression and contribute to the development of more targeted mental health policies and interventions.

Authors’ contributions

ÖA conceived and led the design and development of the study proposal. ŞÜ supervised data collection, led the data analysis and drafting the manuscript. ÖA, AKK, and ŞÜ made substantial contributions to the conceptualization and design of the study, data interpretations and writing the manuscript. All authors read and approved the final version of the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data underlying this study is subject to third-party restrictions by the Turkish Statistical Institute. Data are available from the Turkish Statistical Institute (bilgi@tuik.gov.tr) for researchers who meet the criteria for access to confidential data. The authors of the study did not receive any special privileges in accessing the data.

Declarations

Ethics approval and consent to participate

This study is based on secondary analysis of anonymized data obtained from the Türkiye Health Survey conducted by the Turkish Statistical Institute (TurkStat). The dataset does not contain any personally identifiable information. In Türkiye, surveys conducted by TurkStat are regulated under the Turkish Statistical Law No. 5429, and data collection is carried out as part of the Official Statistics Programme. In accordance with this legal framework and national regulations governing the use of secondary anonymized data, ethical approval from an institutional review board was not required for the present study [101].

All procedures performed in this study were in accordance with relevant guidelines and regulations, including the Declaration of Helsinki.

Informed consent to participate was obtained from all participants by TurkStat during the original data collection process [101].

TurkStat also conducts the Türkiye Health Survey within the scope of the Official Statistics Program put into effect by law. Since the Türkiye Health Survey is conducted within the scope of legal responsibility by the state, ethical approval is not required [102].

The data were obtained through the joint teamwork of both the Turkish Statistical Institute (TurkStat) and the European Union Statistical Office (SOEU) [101].

For this study, secondary data were employed. Official approval was received from the Turkish Statistical Institute to use the microdata set from the Türkiye Health Survey. The Türkiye Health Survey provides many indicators in the field of health, including the utilization of health services by individuals aged 15 and over, the degree of difficulty they experience in performing their daily activities, and their smoking and alcohol use habits. The Turkish Statistical Institute also received a “Letter of Undertaking” authorizing it to use the study’s data.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.