Psychiatric Symptomatology and Medication Adherence in Type 2 Diabetes: Cross-Cultural Insights from Greece

Introduction

According to the World Health Organization, diabetes is a chronic disease that is caused either by hereditary causes or by an acquired deficiency of the pancreas in the production of insulin, or by the ineffectiveness of the insulin that is produced. Such a deficiency leads to an increase in the concentration of glucose in the blood, which then affects many organs of the body, especially the blood vessels and nerves (Banday, Sameer & Nissar, 2020). Cognitive, emotional, anxiety, and psychotic disorders, as well as personality disorders, often coexist with diabetes mellitus (DM). Patients with DM show an increased risk of developing depressive symptoms, which seem to be related, among other things, to biological mechanisms that link metabolic disorders with changes in the structure and function of the brain. Particularly in patients with type 2 diabetes mellitus (T2DM), mental disorders occur with high frequency (Kalra, Jena & Yeravdekar, 2018).

According to a meta-analysis, thirty-nine studies (n = 17,486) were included. The pooled prevalence of depression was 35% (95% CI: 30%–41%). Significant risk factors included younger age (≤60 years), female sex, single status, unemployment, physical inactivity, anxiety, low social support, poor medication adherence, diabetes-related complications, physical disability, insulin-based treatment, and fasting glucose ≥126 mg/dL. (Yang et al., 2025). At the same time, clinical studies show that people with DM experience increased levels of anxiety and psychological burden. Management of the disease requires significant lifestyle changes, such as strict adherence to medication (oral or insulin therapy), the adoption of an appropriate diet, regular physical activity, and frequent monitoring of glucose levels. This constant burden and chronic stress that accompany DM further contribute to the development of depressive symptoms. The prevalence of depression appears to be up to twice as high in patients with T2DM compared to the general population, with 15–30% of diabetics meeting diagnostic criteria for depression (Yang et al., 2025; Xu & Chen, 2025). On the other hand, the presence of depression can worsen the course of diabetes, increase the risk of acute and chronic complications, and negatively affect the mental health of patients.

The chronic nature of the disease and its complications often lead to feelings of pessimism, low self-esteem, and loss of control. In addition, the gradual decline in functionality, especially in elderly patients, can lead to dependency, social isolation, and, in some cases, suicidal ideation. Improving the quality of life of patients requires effective treatment of depression and strengthening self-care skills, which also contributes to reducing health costs (Liu, 2025). In the context of Primary Health Care, there are guidelines for the early detection and assessment of depression in people with chronic diseases, such as T2DM. The assessment should include symptoms of anxiety and depression, dietary habits, and the patient's cognitive status (Sebastian et al., 2023). Effective management of the disease requires a high degree of compliance with medication, diet, and physical activity, as well as regular medical monitoring (Dai et al, 2023).

Mental health is a critical factor in medication adherence in patients with type 2 diabetes. The presence of depressive and anxiety symptoms has been associated with reduced adherence to treatment instructions, negatively affecting both medication intake and the adoption of healthy self-care behaviors (Alli et al., 2024). Patients with impaired mental health often exhibit reduced motivation, difficulties in decision-making, and lower self-efficacy, which leads to missed doses and irregular medication intake (Hisan et al., 2025). As a result, glycemic control worsens, and the risk of complications increases, making clear the need for a holistic approach to disease management (Hao et al., 2025).

Despite the documented association between mental health and medication adherence in patients with type 2 diabetes mellitus, significant research gaps remain. Specifically, a limited number of studies have investigated in depth the reasons for non-adherence through specialized and multidimensional assessment tools (Kwakye et al., 2024). Furthermore, the lack of culturally adapted and psychometrically tested tools in the Greek language limits the understanding of the factors that influence patient behavior in the Greek population. In particular, the lack of translation and cultural adaptation of the Medication Adherence Reasons Scale (MARS) constitutes a significant research gap, given that this scale has been used internationally to understand the reasons for non-adherence (Unni et al., 2014).

Purpose

This study aims to investigate medication adherence and its association with psychological symptomatology in patients with type 2 diabetes mellitus at a community level. At the same time, the study aims to translate and culturally adapt the Medication Adherence Reasons Scale (MARS) into Greek, as well as to initially evaluate its psychometric properties in the Greek population.

Methodology

Ethical considerations

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Approval for the study was obtained from the Scientific Committee of the 5th Health Region of Thessaly and Central Greece (approval date: 27/09/2023). All participants were informed about the purpose of the study and provided written informed consent before participation. Confidentiality and anonymity were strictly maintained, and participants were informed of their right to withdraw at any time without any consequences.

Declaration of Generative AI and AI-Assisted Technologies in the Writing Process

No AI tools were used in the preparation of this article.

Results

Table 1 presents the demographic and clinical characteristics of the sample (N=107). Regarding gender, the sample was almost evenly distributed, with 49.5% men and 50.5% women. Most participants were retired (61.7%), while smaller percentages corresponded to individuals with full (28.0%) and part-time employment (9.3%). In relation to educational level, more than half of the participants had completed primary education (55.1%), while lower percentages were recorded for secondary (20.6%), upper secondary (18.7%), and tertiary education (5.6%). Most of the sample was married (76.6%), while 19.6% were widowed and 3.7% divorced. Regarding accommodation, most participants lived with more than two people (38.3%) or with two people (32.7%), while smaller percentages lived alone (15.0%) or with one person (14.0%). Regarding comorbidity, almost half of the participants reported hypertension (48.6%), while smaller percentages had coronary heart disease (20.6%), other diseases (29.0%), and stroke (1.9%). Regarding symptoms, polyuria was the most frequently reported symptom (65.4%), while a significant percentage presented polydipsia (41.1%) and blurred vision (41.1%). In addition, more than half of the participants reported anxiety or nervousness (54.2%), while tachycardia occurred in approximately half of the sample (50.5%). In contrast, symptoms such as nausea/abdominal pain (5.6%), hunger (3.7%), and changes in behavior (7.5%) were recorded at lower rates. Regarding complications of diabetes mellitus, 15.0% of participants reported diabetic retinopathy, while much lower rates were recorded for diabetic nephropathy (0.9%) and erectile dysfunction (3.7%). The majority reported other types of complications (80.4%). In relation to treatment, most participants were following combination therapy (61.7%), while lower rates were receiving oral medication (25.2%) or insulin (12.1%). Finally, most patients (75.7%) stated that they were following a special diet for diabetes mellitus.

Responses to the individual items of the MARS scale showed a strong concentration at low values (mainly in category 0) for most questions, indicating limited reporting of reasons for non-adherence to medication. In contrast, some items, such as MARS12 and MARS18, showed a greater dispersion of responses, while MARS20 showed an inverse distribution with a high concentration at the higher values. The analytical results of the frequencies and percentages for each item of the scale are presented in Table 2.

A Principal Component Analysis (PCA) was conducted to examine the underlying structure of the Medication Adherence Reasons Scale (MARS). The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy was 0.50, indicating a marginally acceptable level for factor analysis. Bartlett’s test of sphericity was statistically significant (p < .001), supporting the suitability of the data for dimensionality reduction. The PCA revealed a two-component solution with eigenvalues greater than 1. The first component had an eigenvalue of 7.28 and accounted for 36.4% of the total variance, while the second component had an eigenvalue of 3.02, explaining an additional 15.1% of the variance. Together, the two components accounted for 51.5% of the total variance. The first component was characterized by high loadings from several items (e.g., MARS1, MARS3, MARS5, MARS7, MARS8, and MARS11), suggesting a dominant dimension underlying multiple adherence-related reasons. The second component was defined by items such as MARS13, MARS18, and MARS20, indicating a secondary dimension that reflects distinct aspects of medication-adherence behavior. One item (MARS12) demonstrated cross-loading across both components. Detailed results of factor analysis are depicted in Table 3 A & B.

Notably, certain items (e.g., MARS10 and MARS14) exhibited very high uniqueness values, indicating low shared variance with the extracted components and suggesting limited contribution to the overall factor structure. Overall, the findings support a preliminary two-factor structure of the scale, although the relatively low KMO value indicates that the factor solution should be interpreted with caution.

Table 4, A & B, presents the correlations between SCL-90 and MARS. The correlations between psychiatric symptomatology and individual items of the Medication Adherence Reasons Scale (MARS) are presented in Tables 3A and 3B. Overall, most correlations were of low to moderate magnitude, suggesting a relatively weak but statistically significant relationship between certain dimensions of mental health and reasons for non-adherence to medication. More specifically, interpersonal sensitivity showed positive and statistically significant correlations with several items of the MARS scale (e.g., MARS1, MARS4, MARS5, MARS7, MARS8, MARS9), suggesting that individuals with increased interpersonal sensitivity are more likely to report more reasons for non-adherence. Similarly, depression was positively correlated with specific items of the scale (such as MARS4, MARS6, MARS16, and MARS18), which reinforces the hypothesis that depressive symptoms are associated with reduced adherence to treatment. Of particular interest is the dimension of paranoid ideation, which showed the strongest positive correlations with multiple items of the scale (indicatively MARS1, MARS3, MARS5, MARS7, MARS8, and MARS9), suggesting that individuals with increased levels of paranoid thoughts may experience more barriers or reasons for avoiding medication. In addition, phobic anxiety symptomatology and the obsessive-compulsive dimension also showed statistically significant, mainly positive, correlations with some items of the MARS, although of smaller magnitude. In contrast, dimensions such as somatization and hostility showed limited or non-statistically significant correlations with most items of the scale. Finally, overall psychopathology indices (GSI, PST, and PSI) showed mainly low to moderate positive correlations with selected MARS items (particularly with MARS16), suggesting that overall mental health burden is associated with increased reasons for non-adherence to medication.

Discussion

This study investigated medication adherence in patients with diabetes mellitus and its relationship to psychiatric symptomatology, utilizing the Medication Adherence Reasons Scale (MARS). Findings indicate, on the one hand, high levels of self-reported adherence and, on the other hand, a statistically significant, although generally modest, association between specific dimensions of psychopathology and reasons for non-adherence to treatment. Furthermore, factor analysis revealed a preliminary bifactorial structure of the scale, which nevertheless requires careful interpretation.

Regarding medication adherence, the strong concentration of responses at low MARS scores suggests that participants reported limited reasons for non-adherence. This finding may reflect high levels of treatment adherence, which have been reported in other studies among chronically ill populations, particularly in older individuals or those with long-term experience managing their disease (Sahoo et al., 2022; Stewart, Moon & Horne, 2023). However, this interpretation should be made with caution, as the use of self-reported instruments may be affected by social desirability bias, with participants underestimating non-adherence behaviors. In addition, cultural factors, such as strong trust in the physician and the healthcare system, may contribute to shaping such responses (Latkin et al., 2017).

According to the results of the present study, the scale structure through Principal Component Analysis revealed a two-factor solution, which explains more than 50% of the total variance, a finding that suggests a relatively satisfactory but not fully established conceptual structure of the tool. The first factor seems to gather generalized reasons for non-adherence, while the second captures more specialized and differentiated aspects of adherence behavior, which is consistent with the theoretical approach of the multifactorial nature of medication adherence. Similarly, the revised Medication Adherence Reasons Scale (MAR-Scale) has revealed a multidimensional structure, confirming that non-adherence is not a unidimensional phenomenon but is associated with different cognitive, behavioral, and practical factors (Unni, Olson & Farris, 2014). However, the marginally acceptable value of the KMO index in the present study indicates limited suitability of the sample for factor analysis, which raises issues regarding the stability of the factor structure and requires careful interpretation of the results. Furthermore, the appearance of high uniqueness in some items indicates that they do not contribute substantially to the extracted factors, a finding that may reflect the complexity and heterogeneity of the reasons for non-adherence, as has been pointed out in previous studies that emphasize the need for further optimization and cultural adaptation of measurement tools (Unni & Farris, 2015). Our findings reinforce the need for further investigation of the psychometric adequacy of the scale in larger and more heterogeneous samples, as well as for confirmatory factor analysis.

Of particular interest are the findings regarding the relationship between psychiatric symptomatology and medication non-adherence (Seghatoleslam et al., 2022). Paranoid ideation emerged as the dimension with the strongest and most consistent positive correlations with multiple items of the MARS scale. This finding suggests that individuals with increased levels of suspicion and mistrust may experience more barriers to adherence, possibly due to reduced trust in health professionals or questioning the necessity of treatment (Shukla, Schilt-Solberg & Gibson-Scipio, 2025). This association has important clinical implications, as it highlights the role of cognitive and emotional processes in the adoption of health behaviors. Depression also showed positive associations with individual components of nonadherence, confirming previous findings linking depressive symptoms to reduced adherence to treatment. Possible mechanisms include reduced activation, fatigue, lack of motivation, and feelings of futility, which may negatively affect an individual's ability to adhere to consistent treatment regimens. Recognition and management of depression in patients with diabetes mellitus, therefore, becomes critical to improving adherence and clinical outcomes (Osborn & Egede, 2012; Yang et al., 2023).

At the same time, interpersonal sensitivity emerged as an important factor associated with increased reasons for non-compliance. Individuals with high interpersonal sensitivity tend to be particularly vulnerable to criticism and rejection, which may affect their relationship with health professionals and, by extension, their compliance with treatment. This finding highlights the importance of the quality of the therapeutic relationship and communication in enhancing compliance, supporting the need for interventions that focus on the interpersonal dimension of care (Zolnierek & DiMatteo, 2009). Other dimensions of psychopathology, such as phobic anxiety symptomatology and the obsessive-compulsive dimension, showed weaker but statistically significant associations with some MARS items, while somatization and hostility did not appear to be significantly related to compliance. Overall psychopathology indices also showed low to moderate correlations, suggesting that generalized mental burden may influence, to some extent, compliance behavior (DiMatteo et al., 2000).

The findings of the study have important clinical implications. Integrating mental health assessment into daily clinical practice for patients with diabetes mellitus may help identify individuals at high risk for non-adherence. Systematic screening for depressive symptoms and suspicious features could be a key element of a holistic care approach. Furthermore, the role of nurses is crucial in developing therapeutic relationships of trust, providing personalized education, and enhancing patients' active participation in the management of their disease (Alotaibi et al., 2025).

Limitations

The present study presents some limitations that should be considered when interpreting the findings. The relatively small sample size and the use of a convenience sample limit the generalizability of the results. Furthermore, the cross-sectional nature of the study does not allow for causal conclusions. The use of self-reported instruments may be affected by bias, while the low value of the KMO index indicates limitations in the factor analysis. Future research should focus on using larger and more representative samples, as well as on applying confirmatory factor analysis to verify the structure of the MARS scale. Furthermore, conducting longitudinal studies could contribute to understanding the causal relationship between mental health and treatment adherence, while interventional studies could evaluate the effectiveness of psychosocial interventions in improving adherence.

Practical Value

The Greek version of the MARS offers clinicians a practical instrument for identifying individualized barriers to medication adherence among patients with type 2 diabetes. The observed associations between psychiatric symptoms and non-adherence highlight the need for integrated psychological assessment in diabetes management and support the development of targeted interventions to improve adherence and health outcomes.

Conclusions

The findings of this study clearly highlight the multidimensional nature of medication adherence in patients with diabetes, underlining that health behavior is not solely a result of biomedical parameters, but is significantly shaped by psychological and interpersonal factors. Even though participants reported high levels of adherence, analysis of individual reasons for non-adherence revealed that specific dimensions of psychiatric symptomatology, such as paranoid ideation, depression, and interpersonal sensitivity, are associated with increased barriers to treatment adherence. These findings reinforce the view that mental health is a critical and often underestimated factor in the effective management of chronic diseases.

In particular, the strong association of paranoid ideation with multiple elements of non-compliance highlights the role of trust in the therapeutic relationship and suggests that cognitive processes related to suspicion and mistrust may act as substantial barriers to the adoption of therapeutic guidelines. At the same time, the effect of depression and interpersonal sensitivity confirms that emotional and social factors affect the ability of the individual to respond to the demands of chronic treatment. Understanding these mechanisms becomes particularly important for the design of interventions that are not limited to medication but incorporate psychosocial dimensions of care. At the same time, the preliminary emergence of a bifactorial structure of the MARS scale suggests that the reasons for non-compliance are not unidimensional, but reflect different aspects of the patient's behavior, which reinforces the need for more targeted and individualized interventions. However, the limited psychometric power of the analysis suggests that further investigation is required before definitive conclusions can be drawn.

At a clinical level, the results of the study highlight the need for a holistic, patient-centered approach to diabetes management, which systematically incorporates mental health assessment and emphasizes the therapeutic relationship between health professionals and patients. The role of nurses is particularly critical in this context, as they can contribute to the early recognition of psychological difficulties, enhance trust, and provide personalized education and support. Ultimately, the integration of psychosocial interventions into routine care may be a key factor in improving compliance and, by extension, clinical outcomes in patients with chronic diseases.