Summary

Background

This study addresses a gap in the literature by examining how health literacy mediates the relationship between trust in public health authorities and distrust in health systems, particularly in the Turkish context. While previous studies have examined trust and health literacy separately, few have examined trust in health systems regarding health literacy and trust in public health authorities. Therefore, the primary purpose of this study is to determine the mediating role of health literacy in the relationship between trust in public health authorities and distrust in the health system.

Methods

This research was carried out across T眉rkiye. In this context, 924 people over 18 years of age participated in the study. As data collection tools, the trust in public health authorities scale, the distrust in health systems scale, and the health literacy scale adapted to Turkish by the researchers were used. Data were collected online via the survey technique. Translation and back-translation methods were used in the scale鈥檚 linguistic and cultural adaptation process. While content validity was assessed with expert opinions, structural validity was tested with confirmatory factor analysis (CFA). The reliability of the scale was measured with Cronbach鈥檚 alpha coefficient. In addition, the relationship between trust in public health authorities and distrust in health systems was examined using structural equation modeling (SEM), and the mediating role of health literacy in this relationship was analyzed with Bootstrap methods.

Results

As a result of CFA the adequacy indices of the scale were chi-square/degrees of freedom鈥�=鈥�2.940; GFI鈥�=鈥�0.946; AGFI鈥�=鈥�0.915; CFI鈥�=鈥�0.957; NFI鈥�=鈥�0.937; and RMSEA鈥�=鈥�0.069. The Cronbach鈥檚 alpha value of the scale was found to be 0.880. As a result of structural equation modeling, the total effect of Trust in Public Health Authorities on Distrust in Health Systems (尾=-0.257; p鈥�<鈥�0.001) and the impact on Health Literacy was found to be statistically significant (尾=-0.130; p鈥�<鈥�0.001). The indirect effects between Trust in Public Health Authorities and Distrust in Health Systems were calculated as -0.064 and 鈭掆��0.096/-0.037 with a 95% confidence interval, and since these values 鈥嬧�媎id not include the 0 range, the indirect effects were found to be statistically significant.

Conclusions

As a result of the research, it was concluded that health literacy mediates the relationship between trust in public health authorities and distrust in health systems. In this context, public health authorities can pay more attention to sharing information in clear, transparent, and plain language supported by regularly verifiable data to establish trust and communicate effectively with the public.

Introduction

Public health professionals are responsible for providing guidance and precautions on health issues. However, society must trust these experts and comply with recommendations [1] because community participation forms the basis of public health studies [2]. This can be achieved through society鈥檚 trust in the existing healthcare system. Healthcare system trust is essential for patient satisfaction, compliance with treatment, and continuity of communication with healthcare professionals [3]. In this context, studies investigating the trust of patients, healthcare providers, or public health experts in healthcare systems and experts are frequently reported in the literature [1, 4,5,6,7].

There is thought to be a relationship between trust in public health experts and health system distrust and health literacy. Many studies claim that trust in healthcare institutions is effective in complying with the recommendations of healthcare professionals [8,9,10,11,12]. Therefore, damage to trust in health institutions or health systems negatively affects public health efforts [13], which leads people to exhibit negative health-related behaviors or not comply with health recommendations [8, 9, 11]. Health literacy may also play a role in these situations and behaviors.

Health literacy refers to individual and systemic skills. This skill occurs during interactions with healthcare [14]. From a public health perspective, health literacy can be associated with the health of the healthcare system and society. Additionally, health literacy has become a significant public health issue [15, 16]. Health literacy encompasses the ability to effectively access, understand, and use health information, tools, or services [17]. Low health literacy can lead to distrust of healthcare institutions and the healthcare system. Additionally, low health literacy levels may be associated with adverse health outcomes [18]. For example, individuals with low health literacy are generally less likely to participate in preventive health measures, which may increase their likelihood of experiencing adverse health outcomes [19]. Low health literacy leads to poor health outcomes: some studies highlight its negative effects on chronic disease prevalence, hospitalization rates, and mortality [20]. On the other hand, health literacy can affect trust in health systems and healthcare professionals. Trust in healthcare systems varies greatly across the world, influenced by factors such as governance, transparency, and social norms [6, 7, 12]. This situation needs to be addressed in order to restore patient trust in healthcare institutions [21]. This can increase patient participation. Therefore, improving health literacy and addressing system-level issues is vital [14,15,16]. Research shows that individuals with low health literacy have a high degree of trust in healthcare providers. However, this trust can decrease when communication is inadequate or problematic [22]. On the other hand, individuals with high health literacy can question health information. This can lead to a more cautious approach towards healthcare providers. This is especially true when an individual鈥檚 research conflicts with the recommendations of healthcare professionals. Such skepticism can sometimes undermine an individual鈥檚 trust in the healthcare system [23]. At this stage, restoring and maintaining trust can help improve public health [24]. However, the relationship between health literacy and distrust in the health system cannot be drawn because studies do not find any relationship between health literacy and distrust [18]. Distrust in health systems often stems from perceived inequalities or inefficiencies [25]. Of course, trust in health professionals also impacts this situation. Trust in public health authorities refers to trust in their abilities and power [26].

Ensuring transparency in communication with health system officials plays an essential role in ensuring trust [11, 13]. Improving health literacy and addressing system-level issues are also vital [14,15,16, 27]. Measuring the validity and reliability of trust in public health experts is highly important for examining and increasing adherence to the recommendations of public health authorities [1]. Apart from this, although there is a growing field of study on health literacy and trust in public health systems, especially in developing countries, limited research has addressed these relationships as a whole [15, 17, 26, 28, 29]. There is a complex relationship between trust in public health authorities, health literacy, and distrust of health systems. Increasing health literacy can eliminate distrust, allowing people to interact more effectively with health systems. However, increasing health literacy can also make some people more cautious of health professionals. Therefore, research is needed to better understand the complex relationship between health literacy and trust in public health professionals and the healthcare system. This research aims to determine the relationships between the variables in question. This research aims to address this gap by examining the mediating role of health literacy in the relationship between trust in public health authorities and distrust in health systems in Turkey. In this context, the following research question was addressed in the study: 鈥淗ow does health literacy moderate the relationship between trust in public health authorities and distrust in health systems in Turkey?鈥�

Method

Purpose of the research

Adapting the Turkish version of the Trust in Public Health Authorities Scale is crucial to accurately measuring public perceptions and trust levels toward health authorities. Cultural differences, language barriers, and the unique dynamics of the local health system can make it challenging to apply this scale in another language directly. A scale adapted to Turkish would enable more effective communication with target audiences and increase the effectiveness of public health policies. Furthermore, determining trust levels is critical in evaluating the success of public health campaigns and reducing community resistance. Validation of the Turkish version of the Trust in Public Health Authorities Scale could provide a reliable tool for future research and public health initiatives in Turkey. In this context the purpose of this study is to establish the validity and reliability of the trust in public health authorities scale in Turkey and to determine the mediating role of health literacy in the relationship between trust in public health authorities and distrust in the health system.

Hypotheses of the research

The hypotheses of the research are presented below.

H1

Trust in public health authorities negatively affects distrust in health systems.

H2

Trust in public health authorities positively affects health literacy.

H3

Health literacy negatively affects distrust in health systems.

H4

Health literacy mediates the relationship between trust in public health authorities and distrust in health systems.

Research design and model

This descriptive, cross-sectional study uses a methodological approach to adapt a scale to another language and test the proposed mediation model. In the scale adaptation process, performing scale validity and reliability analyses is a critical step. A large data set is usually needed for these analyses, often collected with a cross-sectional method. To evaluate the effects of the study on the target population, the measurement results of the scale must be described in detail. This process is carried out with a descriptive approach. The independent variable of the research is Trust in Public Health Authorities (TiPHA), the dependent variable is Distrust in Health Systems (DiHS), and the mediator variable is Health Literacy (HL). The models established for the relationships between these variables are shown in Figs.听1 and 2.

Fig. 1

Simple effect model

Fig. 2

Mediated effect model

Sampling

The research universe consists of individuals over 18 who live in the Republic of Turkey and have internet access. The research sample consists of participants over 18 who have internet access and agree to participate in the research. The convenience sampling method was preferred for sample selection. Convenience sampling is a non-probability sampling method that allows researchers to collect data from easily accessible and cost- and time-efficient participants. In this method, randomness is not ensured in the sampling process. Therefore, the generalizability of the results is limited. Researchers usually work with easily accessible groups such as students, colleagues, or online platform users [30]. The number of samples was calculated as 384 at the 95% confidence level and 5% confidence interval. However, since the study involves scale adaptation and is based on structural equation modelling, the sample size during the scale adaptation phase should be at least five times [31], 10 times [32], and 15 times [33] the number of items. Considering that the number of items in the Trust in Public Health Authorities Scale is 14, it aims to reach at least 140 people. At the end of the data collection process, the data obtained from 409 participants were analysed, and a confirmatory factor analysis was performed. In the second phase of the study, which aimed to determine the mediating role, a mediation analysis was conducted using data obtained from 515 participants whose characteristics were similar to those of the sample. In both stages of the research, participants were individuals living in the community and having access to the internet. The first-stage data were collected between November 2022 and March 2023; the secondstage data were collected in November 2023. In both stages, the data were collected via Google Forms. Google Forms are set up only to receive one response from the same email, allowing each participant to provide only one response. Participants were recruited through institutional networks, social media platforms, and email invitations. These channels were chosen to maximize reach and ensure diversity among participants. The convenience sampling technique was chosen because of its efficiency in recruiting a large number of participants within the time constraints of the study. Participants were not asked questions that would reveal personal information during the survey. Thus, participant confidentiality was ensured. Participants were informed about the study in the first part of the prepared surveys. In addition, before the survey, the participants were given necessary explanations about data use, withdrawal rights, and potential risks, and informed consent was obtained.

Data collection tools

Three scales were used in this study: the Trust in Public Health Authorities Scale, the Distrust in Health Systems Scale, and the Health Literacy Scale. These scales were selected based on their validated use in previous studies and suitability for the Turkish context. The Trust in Public Health Authorities Scale has demonstrated robust psychometric properties in various settings, making it ideal for measuring institutional trust [1]. The Distrust in Health Systems Scale provides insights into negative perceptions of health systems, which is critical for understanding the dynamics of public trust [4]. Finally, the Health Literacy Scale is widely recognized for its comprehensive assessment of individuals鈥� ability to access, understand, and use health information [34]. An online survey of 4 parts was applied to achieve the research purpose.

Sociodemographic Information Form: In this section, some questions were asked, such as residence area, gender, age, marital status, income, occupation, education level, and health service usage status, to understand the demographic characteristics of the participants.

Trust in Public Health Authorities Scale: The scale developed by Holroyd to measure trust in public health authorities consists of 14 questions and two subscales: 鈥渂eneficence鈥� and 鈥渃ompetence.鈥� [1]. The Cronbach鈥檚 alpha reliability coefficient of the scale is 0.86. The Cronbach鈥檚 alpha reliability coefficient was calculated as 0.92 for the beneficence dimension and 0.87 for the competence dimension. The Cronbach鈥檚 alpha of the study was determined to be 0.797 for the Beneficence dimension, 0.796 for the Competence dimension, and 0.880 for the overall scale.The scale is a 4-point Likert type (鈥淪trongly agree,鈥� 鈥淎gree,鈥� 鈥淒isagree,鈥� 鈥淪trongly disagree鈥�). Higher scores indicate higher confidence. The authors made the Turkish adaptation of the scale. This scale was used in the study鈥檚 first and second phases. Sample items are presented below:

鈥淧ublic Health Authorities do whatever they have to do to protect the public鈥檚 health.

Public Health Authorities base their recommendations on the best information available.

Public Health Authorities care about everyone, regardless of who has more or less money.鈥�

Distrust in Health Systems Scale: This scale was developed by Rose et al. to measure individuals鈥� levels of distrust in health systems [4]. The scale was adapted to Turkish by Ye艧ildal et al. [5]. The scale consists of 10 items and a single dimension with a 5-point Likert scale. It is accepted that as the values from the scale approach 5, distrust in health systems increases, and as they decrease toward 1, distrust decreases. In the Turkish validity and reliability study of the scale, the Cronbach鈥檚 alpha value was calculated as 0.79. For this research, the Cronbach鈥檚 alpha coefficient of the scale was calculated as 0.755. This scale was used in the second stage of the study. Sample items are presented below:

鈥淚 believe that medical experiments were performed on me without my knowledge.

I believe that my medical records were kept confidential.

Many people die every day from errors in the healthcare system.鈥�

Health literacy scale-short form: This scale was developed by Duong et al. [34]. Y谋lmaz and Eskici adapted the Turkish scale [35]. The scale includes 4-point Likert-type answer options ranging from 1 (very difficult) to 4 (very easy) and consists of 12 items and a single dimension. The Cronbach鈥檚 alpha coefficient of the scale was calculated as 0.856. For this research, the Cronbach鈥檚 alpha coefficient of the scale was calculated as 0.855. A higher score on the scale indicates better health literacy. This scale was used in the second stage of the study. Sample items are presented below:

鈥淭o be able to access information about the treatment of diseases that concern you.

To understand the information leaflets on medicines (medicine information sheet),

Evaluate the advantages and disadvantages of different treatment options related to your disease.鈥�

Adaptation work

The field research of this study was carried out in two stages. The Turkish Adaptation of the Trust in Public Health Authorities Scale was carried out in the first stage. The translation process of the scale, which is the study鈥檚 first phase, was carried out in 4 stages.

The scale was initially translated by seven bilingual translators proficient in both the source and target languages. Each translator provided an independent scale translation and worked individually to adapt the scale to Turkish. The experts involved in this phase included academics specializing in English and public health. Once the initial translations were completed, the researchers compared the translations and reached a consensus on the most accurate and culturally appropriate wording. The draft version of the scale was then back-translated into the original language by two independent bilingual translators. These back-translators were professionals fluent in both languages 鈥嬧�媋nd were not involved in the initial translation process to avoid potential bias. The back-translation was performed without referencing the original scale to ensure neutrality and accuracy. After the back-translation, the researchers reviewed the scale to verify its accuracy and cultural relevance. During the evaluation phase, a public health expert provided feedback on the suitability of the scale for Turkish. The pilot study aimed to identify items that might be confusing, inappropriate, or unclear to the participants. The feedback from the participants was carefully evaluated. However, none of the items were identified as confusing or inappropriate.

Following the pilot study, the scale was applied to two different sample groups. There are 409 participants in the first of these samples and 515 participants in the second. Data obtained from the first participants were used for scale adaptation, and data obtained in the second stage were used for mediation testing.

Analysis of data

Translation and back-translation methods were used in the scale鈥檚 linguistic and cultural adaptation process. While content validity was assessed with expert opinions, structural validity was tested with confirmatory factor analysis (CFA). The reliability of the scale was measured with Cronbach鈥檚 alpha coefficient. The following values 鈥嬧�媤ere taken into consideration in the interpretation of Cronbach鈥檚 Alpha values: 鈥溾墺.9鈥� Excellent, 鈮�.8鈥� Good, 鈮�.7鈥� Acceptable, 鈮�.6鈥� Questionable, 鈮�.5鈥� Poor, and 鈮�.5鈥� Unacceptable鈥� [36]. In addition, the relationship between trust in public health authorities and distrust in health systems was examined using SEM, and the mediating role of health literacy in this relationship was analyzed with Bootstrap methods [37, 38].

Kurtosis and skewness values were considered when performing the study鈥檚 normality test. The skewness and kurtosis indices being close to 0 within the limits of 卤鈥�2 is considered evidence of normal distribution [39]. The fact that the index obtained from the analysis was in this range showed that the data were normally distributed. A comprehensive data-cleaning process was applied to address missing or incorrect responses in the survey. Responses with more than 10% missing data were excluded from the analysis. Data analysis was carried out in three stages. In the first stage, the validity and reliability of the Turkish form of the scale were examined with CFA. CFA was used to test whether the original structure of the scale was preserved and whether the factor loadings of the items were within acceptable limits [40]. It was stated that the factor loadings should be 0.30 and above. Therefore, items with factor loadings below 0.30 were removed from the scale. In addition, in cases where two items shared a common variance, this situation was modeled by adding the error term covariance to the CFA model [41].

The analyses were performed using the SPSS and AMOS programs, and fit indices (such as 蠂虏/df, RMSEA, CFI, GFI, and TLI) were evaluated [42]. Table听1 was taken into consideration in the evaluation of the goodness of fit indices [43, 44].

Table 1 Fit indices

In the second stage, correlation analysis was performed to evaluate the relationships between the scales. This analysis examined whether the adapted scale exhibited a significant relationship with the other two scales. The Pearson correlation coefficient was calculated using the SPSS program. The following values were taken into consideration in the interpretation of correlation coefficients: 0.00-0.10 negligible, 0.11-0.39 weak, 0.40-0.69 moderate, 0.70-0.89 strong, 0.90鈥夆垝鈥�1.00 very strong [45]:

Structural Equation Modeling is a powerful statistical method for analyzing direct and indirect effects between variables. Within the scope of this study, SEM is considered an appropriate method for testing hypotheses [40]. While hypotheses H1 and H2 examine the direct effects of trust in public health authorities on distrust of the health system and health literacy, hypothesis H3 addresses the relationship between health literacy and distrust of the health system. In addition, hypothesis H4 tests the mediation role of health literacy between trust in public health authorities and distrust of the health system. After analyzing the direct effects within the framework of the SEM model, the bootstrap method can be used to evaluate the mediation effect. This method effectively tests whether the mediation is statistically significant by determining the confidence intervals of the indirect effects [46].

In the last stage, the mediation effect between a scale and the other two scales was tested with Hayes鈥� process macro [47]. This analysis was carried out to understand how the mediator variable changed the effect of the independent variable on the dependent variable. Confidence intervals were calculated using the bootstrap method, and the statistical significance of the mediation effect was evaluated. In the model created with the bootstrap process, 5,000 samples were taken to determine the importance of the direct and indirect impact [48].

Results

When the gender distribution of the 409 participants in the first stage was examined, 273 (66.7%) participants were women, and 136 (33.3%) were men. The average age of the participants was 37.09; 233 participants (57%) were below the average age, and 176 participants (43%) were above the average age. A total of 262 (64.1%) of the participants were married, 134 were single (32.8%), and 13 (3.2%) were widowed/divorced. While 273 (66.7%) of the participants reported that their income was intermediate, only 28 (6.8%) stated that they had a high income level. A total of 158 (38.6%) of the participants were undergraduate graduates.

When the demographic distribution of the participants in the second stage was examined, 407 (79%) of the participants were women, and 108 (21%) were men. The average age of the participants was 32.37; 284 participants (55.1%) were above the average age, and 231 participants (44.9%) were below the average age. A total of 321 (62.3%) of the participants were married, 178 were single (34.6%), and 16 (3.1%) were widowed/divorced. While 283 (55.0%) of the participants reported that their income was medium, only 13 (2.5%) stated that they had a high income level. A total of 196 (38.1%) of the participants were undergraduates, and 159 (30.9%) were high school graduates.

During the adaptation process, confirmatory factor analysis was conducted on data obtained from 409 participants. As a result of the analysis, CFA was performed again by making covariance between items 10 and 11 to improve the goodness of fit. The model [path graph] of the scale is given in Fig.听3.

Fig. 3

Path diagram for the trust in public health authorities scale

According to Table听2, when the 2nd, 4th, and 7th items under the Beneficence dimension are removed from the scale due to their low factor loadings, the following solutions are obtained: chi-square/sd鈥�=鈥�2.940; GFI鈥�=鈥�0.946; AGFI鈥�=鈥�0.915; CFI鈥�=鈥�0.957; NFI鈥�=鈥�0.937; and RMSEA鈥�=鈥�0.069. This shows that the model is at an acceptable level. In the confirmatory factor analysis, the path coefficients for all items under the Beneficence and Competence dimensions were statistically significant (p鈥�<鈥�0.001). The path coefficients obtained for the items vary between 1.736 and 0.484. The Cronbach鈥檚 alpha of the study was determined to be 0.797 for the Beneficence dimension, 0.796 for the Competence dimension, and 0.880 for the overall scale.

Table 2 Confirmatory factor analysis results

Determine the mediating role of health literacy in the relationship between trust in public health authorities and distrust in health systems, the relationships between the variables were first determined. The findings regarding the correlation results are shown in Table听3.

Table 3 Correlation analysis results

As shown in Table听3, there are significant relationships between the dependent, independent, and mediator variables within the scope of the research. When the findings in the correlation matrix are examined, there is a positive, weakly significant relationship between trust in public health authorities and distrust in health systems (r鈥�=鈥�0.162, p鈥�<鈥�0.01), and a negative, moderately significant relationship with health literacy (r=-0.405, p鈥�<鈥�0.01) is available. A significant negative, weakly relationship was detected between distrust in health systems and health literacy (r=-0.273, p鈥�<鈥�0.01). Within the scope of the research, the effect of Trust in Public Health Authorities on Distrust in Health Systems is shown in Fig.听4.

Fig. 4

Path diagram for the effect of trust in public health authorities on distrust in health systems

As shown in Fig.听4, trust in public health authorities appears to have a negative effect on distrust in health systems. The goodness of fit values for the model are at an acceptable level (Chi-square/df鈥�=鈥�4.023; GFI鈥�=鈥�0.941; AGFI鈥�=鈥�0.905; CFI鈥�=鈥�0.937; NFI鈥�=鈥�0.919; RMSEA鈥�=鈥�0.077).

In the second model of the research, a path analysis based on the bootstrap method was conducted to test whether health literacy has a mediating role in the relationship between trust in public health authorities and distrust in health systems. The model for the mediating role is shown in Fig.听5 below.

Fig. 5

PATH model for the mediator role

According to Table听4, the total effect of Trust in Public Health Authorities on Distrust in Health Systems was statistically significant (尾=-0.257; p鈥�<鈥�0.001). A negative path coefficient was obtained between Trust in Public Health Authorities and Health Literacy (尾=-0.130; p鈥�<鈥�0.001). The indirect effects between Trust in Public Health Authorities and Distrust in Health Systems was obtained were 鈭掆��0.064 and 鈭掆��0.096/-0.037, respectively, with a 95% confidence interval. The indirect effect is statistically significant since these values do not include the range 0. Indirect effects refer to the impact of trust in public health authorities on distrust of health systems through health literacy. The goodness of fit values for the obtained model are at an acceptable level (chi-square/df鈥�=鈥�3.150; GFI鈥�=鈥�0.886; AGFI鈥�=鈥�0.861; CFI鈥�=鈥�0.879; NFI鈥�=鈥�0.833; RMR鈥�=鈥�0.085; RMSEA鈥�=鈥�0.065).

Table 4 Analysis results regarding the mediating role

According to Fig.听6, the findings obtained from the research show that hypotheses H1 and H4 are accepted and that hypotheses H2 and H3 are rejected.

Fig. 6

Trust in public health authorities predicts distrust in health systems through health literacy

Discussion and conclusion

The research was conducted in two stages to demonstrate the validity and reliability of the trust in public health authorities scale in Turkey and to determine the mediating role of health literacy in the relationship between trust in public health authorities and distrust in the health system. The findings obtained as a result of the CFA performed to show that the model has an acceptable fit (chi-square/sd鈥�=鈥�2.940; GFI鈥�=鈥�0.946; AGFI鈥�=鈥�0.915; CFI鈥�=鈥�0.957; NFI鈥�=鈥�0.937; and RMSEA鈥�=鈥�0.069.). In addition, the scale鈥檚 Cronbach Alpha reliability coefficient results were close to the original. This result confirms the validity and reliability of the Turkish version of the scale.

This study identified negative relationships between trust in public health officials, distrust in health systems, and health literacy. The negative relationship between trust in public health authorities and health literacy is particularly striking. In other words, individuals with high health literacy have been found to trust public health authorities less. This can be explained by the fact that these individuals have higher expectations for the health system and experience disappointment when their expectations are unmet. The study by Bertram et al. [49] shows that people with high levels of health literacy have higher expectations regarding the quality of healthcare services. At the same time, when these expectations are not met, their trust levels decrease. This finding may also be because people with high health literacy approach institutions skeptically. There is no direct evidence that individuals with high health literacy have a more critical or skeptical attitude toward healthcare institutions and professionals. However, it is known that the health literacy level affects how individuals perceive and use healthcare services [50]. Therefore, people with extensive health literacy should be encouraged to seek information and advice from public health authorities. Studies have shown that individuals with high levels of health literacy are more likely to trust public health authorities. For example, a survey by Gani et al. emphasizes that individuals with high health literacy trust health professionals significantly more than individuals with low health literacy. This suggests a direct relationship between health literacy and trust in health professionals [51]. Similarly, a study by Chen et al. reveals that university students with low health literacy are less likely to use official health information sources, and these individuals tend to develop a potential distrust of health authorities [52]. In contrast, people with a high level of health literacy can communicate effectively with health officials. On the other hand, people with low levels of health literacy have excessive trust in health authorities. Both conditions can make people sensitive to information and advice from healthcare providers [53].

Our study also reveals a negative relationship between distrust in health systems and health literacy. One reason could be that distrust in health systems leads to a reluctance to use health-related information sources. Some studies in the literature confirm this observation. According to Turhan et al., there is a negative relationship between distrust of health systems and health literacy [54]. Similarly, another study reported a negative relationship between health literacy and distrust in health systems [55]. In particular, the relationship between health literacy and health service utilization can directly affect trust in the health system [28]. Another study suggested that increased health literacy will increases trust in the health system [56]. Therefore, the level of health literacy affects trust in health systems and health authorities [18].

Although the study found a negative correlation between health literacy and trust in public health authorities, it is essential to emphasize that this relationship is correlational, not causal. Previous research suggests that individuals with high levels of health literacy may evaluate health information more critically, leading to skepticism towards public health authorities. This interpretation seems to be consistent with the findings of studies such as Turhan et al. [54] and 脺st眉nba艧 & 脰zt眉rk [55] that observed similar trends. However, further research is needed to understand better the causal mechanisms underlying this relationship.

The observed negative relationship between health literacy and trust in public health authorities contradicts the positive results reported by some studies. Existing research suggests that people with higher health literacy have greater trust in healthcare professionals and the healthcare system. For example, Lee et al.鈥檚 study [57] found that people with higher health literacy levels trust doctors and the healthcare system more. Similarly, De Gani et al.鈥檚 study [58] found that people with higher health literacy levels, particularly regarding the coronavirus, have the highest trust in healthcare professionals and authorities. This inconsistency may be due to cultural or systemic factors specific to the Turkish healthcare environment. Examining these contextual differences may contribute to better understanding and reconciling the contradictory findings in the literature. The current study鈥檚 findings underscore the need for specific public health strategies that consider the different effects of health literacy.

The study also indicated a negative direct relationship between trust in public health authorities and distrust in health systems. This situation shows trust in public health authorities increases, and distrust in health systems decreases. On the other hand, if the correlation between two variables is positive and the regression is negative, this indicates a complex relationship. In particular, the effect of different factors or variables may have produced such a result. At the same time, the mediation effect explaining this relationship was also revealed. The fact that the confidence intervals do not include zero indicates that these effects are statistically significant. Indirect effects express the impact of trust in public health authorities on distrust in health systems through health literacy. However, the findings do not provide definitive evidence that health literacy has a causal effect on trust in public health authorities and health literacy. Therefore, health literacy鈥檚 direct or indirect effects on increasing trust need to be further analyzed. On the other hand, the relationship between health literacy and trust in public health authorities may be shaped by an individual鈥檚 previous healthcare experiences and health-related attitudes. Future studies may examine how health literacy moderates these effects. In particular, longitudinal studies and experimental designs contribute to a more precise explanation of causal relationships. In addition to this, the negative coefficient in the relationship with health literacy indicates that this mediation has a complex dynamic. Moreover, the mediating role of health literacy is not limited to the dynamics of trust and distrust but also affects individuals鈥� health behaviors. In a study by Hsu et al., critical health literacy influenced health behaviors significantly. This study shows that individuals better equipped to understand and evaluate health information are more likely to exhibit positive health behaviors [59]. In this context, health literacy stands out as an essential mediator that affects trust dynamics and shapes health behaviors and health-related outcomes. The findings reveal that trust in public health authorities significantly impacts trust in health systems, both directly and indirectly. However, it is also noteworthy that these effects have a very complex structure. The mediation model used in the study provides a comprehensive framework for understanding the relationships between variables.

However, the limited literature on this topic limits the understanding of the subject. Therefore, the results of this study play an essential role in understanding the relationships among health literacy, public health authorities, and trust in the health system.

In conclusion policies and practices aimed at increasing health literacy must be developed without the context of research findings. Various measures should be taken to improve health literacy in society and ensure access to reliable health information through educational programs and accessible resources. Additionally, various communication strategies and social participation methods should be developed to increase communication and trust between the public and public health officials. Additionally, public health authorities can take more excellent care by regularly sharing information in clear, transparent, and plain language, supported by verifiable data, to build trust and communicate effectively with the public.

This research has several limitations. In addition, crossover studies have limitations in confirming the causal relationships between variables. The limitation of this research is that other variables, such as age, gender, and education level, are not considered in the relationships between variables. The analysis assumes that participants will read and understand the research statements and respond unbiasedly to omic or political factors when the research is conducted, which may also affect the results. In addition, using convenience sampling in the study may lead to biases such as limited generalizability and selection bias inherent to this method. The use of convenience sampling may also limit the generalizability of the findings to other populations, as this sampling method may not fully represent the broader population. In addition, the study鈥檚 cross-sectional design prevents the establishment of causal relationships between variables. Another significant limitation is that potential confounding factors such as socioeconomic status, which may affect trust and health literacy, were not considered. Future studies may address these limitations by using more representative sampling methods and longitudinal designs. This way, it may be possible to generalize the findings to a broader population and better understand the causal mechanisms.