Medical mistrust, health system distrust, and their associations with healthcare delay and cancer screening attitudes in primary care

Abstract

Aim

To examine the associations of medical mistrust and health system distrust with healthcare delay behavior and cancer screening attitudes among adults attending primary care services.

Method

This descriptive and cross-sectional study was conducted with 568 adults who attended a family health center in eastern Türkiye. Data were collected using a Personal Information Form, the Medical Mistrust Scale, the Health System Distrust Scale, the Healthcare Demand Procrastination Scale, and the Attitude Scale for Cancer Screening. Data were analyzed using Pearson correlation and multiple linear regression analyses.

Results

Medical mistrust and health system distrust were positively associated with healthcare delay behavior and negatively associated with cancer screening attitudes (p < 0.001). Medical mistrust showed the strongest independent association with both healthcare delay behavior (β = 0.59) and cancer screening attitudes (β = −0.79).

Conclusion

The findings indicate that trust in healthcare is closely associated with healthcare utilization and preventive health attitudes. These associations highlight the potential relevance of trust-centered strategies, improved access, and patient-centered health service delivery in primary care.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12913-026-15484-0.

Introduction

Timely and effective use of health services is one of the key factors in maintaining and improving individual health. Primary care services play a particularly important role in early diagnosis, treatment, and preventive care. However, individuals’ level of trust in the health system and healthcare professionals has been associated with their use of health services [1, 2]. Low levels of trust have been associated with avoiding healthcare, delayed care seeking, and poorer health outcomes [3].

Medical mistrust refers to individuals’ doubts about the competence, ethical behavior, and intentions of healthcare professionals to act in the patient’s best interest [4]. In contrast, health system distrust reflects broader negative perceptions related to healthcare institutions, policies, and service delivery processes [5]. Recent studies have shown that these forms of distrust are associated with lower use of health services, greater health inequalities, and greater vulnerability among disadvantaged groups [6, 7].

These relationships can be interpreted within established models of healthcare utilization and preventive health behavior. Andersen’s Behavioral Model of Health Services Use proposes that healthcare utilization is shaped by predisposing characteristics, enabling resources, and perceived or evaluated healthcare needs [8]. Within this framework, mistrust may represent an attitudinal barrier associated with individuals’ willingness to engage with healthcare services even when care is needed. Similarly, the Health Belief Model emphasizes the role of perceived benefits and perceived barriers in preventive health behaviors [9]. In the context of cancer screening, mistrust toward healthcare professionals or institutions may increase perceived barriers or reduce confidence in the expected benefits and reliability of screening. Taken together, these theoretical perspectives provide a basis for expecting medical mistrust and health system distrust to be associated with delayed healthcare seeking and less favorable attitudes toward cancer screening.

Medical mistrust and health system distrust may also be associated with delayed or avoided healthcare seeking. Healthcare demand procrastination refers to postponing or avoiding professional care despite the presence of a health concern [10]. Evidence from previous research indicates that healthcare avoidance is multifactorial and may reflect unfavorable evaluations of healthcare professionals and organizations, low perceived need for care, and structural barriers such as cost and time constraints [11, 12]. In their systematic review of 124 studies, Benkert et al. (2019) further demonstrated that medical mistrust has been examined at interpersonal and systemic levels and has been frequently associated with behavioral responses, including reduced engagement with healthcare. Importantly, much of this evidence has been derived from cross-sectional studies and predominantly US-based populations, highlighting the need to examine these relationships in different healthcare and cultural contexts [13]. Consistent with this evidence, more recent research has also linked lower trust in healthcare with delayed care seeking [14]. Such delays may have important health consequences, particularly when timely assessment, diagnosis, or treatment is required [15].

Delaying healthcare may be associated with adverse treatment-related outcomes and lower participation in preventive services. This is particularly important in the context of cancer screening. Cancer screening is an effective public health strategy that reduces mortality through early detection [16]. However, distrust in the health system and healthcare professionals has been associated with lower participation in cancer screening [17].

Although previous research has established associations between mistrust and healthcare engagement, medical mistrust and health system distrust represent related but distinct dimensions of trust that may operate at interpersonal and institutional levels [3, 11, 18]. However, evidence remains limited regarding how these dimensions are simultaneously associated with healthcare delay behavior and cancer screening attitudes within the same primary care population, particularly in understudied cultural and geographic settings such as eastern Türkiye. Accordingly, the contribution of the present study lies not in proposing previously unknown relationships, but in extending existing evidence by examining these four interrelated constructs concurrently within a specific primary care context. Understanding these associations may help identify context-relevant patterns that can inform the development and testing of future trust-centered interventions aimed at reducing delays in healthcare seeking and improving engagement with cancer screening. Therefore, this study aimed to examine the associations of medical mistrust and health system distrust with healthcare delay behavior and attitudes toward cancer screening among adults attending primary care services.

Based on the theoretical framework and previous empirical evidence, it was hypothesized that: (H1) higher medical mistrust would be associated with higher health system distrust; (H2) higher medical mistrust and health system distrust would be associated with greater healthcare delay behavior; (H3) higher medical mistrust and health system distrust would be associated with less favorable attitudes toward cancer screening; and (H4) greater healthcare delay behavior would be associated with less favorable attitudes toward cancer screening.

Research Questions

1. Is medical mistrust significantly associated with health system distrust?

2. Are medical mistrust and health system distrust significantly associated with healthcare delay behavior?

3. Are medical mistrust and health system distrust significantly associated with attitudes toward cancer screening?

4. Is healthcare delay behavior significantly associated with attitudes toward cancer screening?

Methods

Study design

This descriptive and cross-sectional study was conducted to examine the associations of medical mistrust and health system distrust with healthcare delay behavior and attitudes toward cancer screening among adults attending primary care services. The study protocol and statistical analysis plan were not prospectively registered.

Setting and time

The study was conducted in a primary care center located in eastern Türkiye, and the data collection process was completed between February and April 2026.

Population and sample

The study population consisted of 9,852 registered individuals aged 30–70 years at the primary care center. The age range of 30–70 years was selected to reflect the target age groups covered by Türkiye’s national population-based cancer screening program, in which cervical cancer screening begins at age 30, breast cancer screening targets women aged 40–69 years, and colorectal cancer screening targets women and men aged 50–70 years [19]. The minimum required sample size was calculated using the sample size formula for a known finite population. Based on a population size of 9,852, a 95% confidence level, a 5% margin of error, and an expected proportion of 50% to provide the most conservative sample size estimate, the minimum required sample size was calculated as 370 participants. To increase the precision of the estimates and account for potential incomplete or unusable responses, data collection continued beyond the minimum required sample size, and the study was completed with 568 participants. A non-probability convenience sampling method was used, whereby eligible individuals attending the primary care center during the study period were invited to participate. The study was reported in accordance with the STROBE guideline [20].

Inclusion and exclusion criteria

Individuals aged between 30 and 70 years, attending the relevant primary care center, able to read and understand Turkish, and providing written informed consent were included in the study. Individuals with cognitive impairment limiting communication, diagnosed severe psychiatric illness, or any physical or mental condition preventing active participation in the data collection process were excluded from the study.

Data collection instruments

Research data were collected using the Participant Information Form, Medical Mistrust Scale, Health System Distrust Scale, Healthcare Demand Procrastination Behavior Scale, and Cancer Screening Attitude Scale. In this study, healthcare delay behavior refers to the construct measured by the Healthcare Demand Procrastination Behavior Scale (HDPBS).

Participant information form

This form was developed by the researchers based on the literature to collect sociodemographic and clinical information. The form consisted of nine questions including variables such as age, gender, marital status, education level, income level, employment status, place of residence, family type, and frequency of primary care visits.

Medical Mistrust Scale

The “Medical Mistrust Scale,” developed by Thompson et al. [18], was adapted into Turkish by Şengül and Bulut [21]. The scale consists of 17 items and a multidimensional structure. The items are rated on a 4-point Likert scale ranging from “1-Strongly disagree” to “4-Strongly agree.” Scale scores are calculated using the arithmetic mean method. Higher scores indicate higher levels of distrust toward the healthcare system and healthcare professionals. In the adaptation study by Şengül and Bulut (2020), Cronbach’s alpha values for the subdimensions were reported as 0.625 for MMS1 and 0.669 for MMS2. In the present study, the Cronbach’s alpha coefficient was calculated as α = 0.897.

Health System Distrust Scale

The “Health Care System Distrust Scale,” developed by Rose et al. (2004) [22], was adapted into Turkish by Yeşildal et al. [23]. The scale consists of 10 items and a single dimension. Items are rated on a 5-point Likert scale ranging from “1-Strongly disagree” to “5-Strongly agree.” Scale scores are calculated using the arithmetic mean method. Scores approaching 5 indicate higher distrust in the healthcare system, whereas scores approaching 1 indicate lower distrust. In the study conducted by Yeşildal et al. (2020), Cronbach’s alpha was reported as 0.789. In the present study, Cronbach’s alpha was found to be 0.767.

Healthcare Demand Procrastination Behavior Scale

The Healthcare Demand Procrastination Behavior Scale (HDPBS), developed by Söyler et al. (2022), consists of 11 items and three subdimensions and was used to determine individuals’ healthcare demand procrastination behaviors [24]. The subdimensions include “self-remedy seeking” (3 items), “avoidance” (4 items), and “inaction” (4 items). Responses are rated on a 5-point Likert scale ranging from “1-Strongly disagree” to “5-Strongly agree.” In the study conducted by Söyler et al. (2022), the Cronbach’s alpha coefficient of the scale was reported as 0.85. In the present study, Cronbach’s alpha was found to be 0.79.

Cancer Screening Attitude Scale

The Cancer Screening Attitude Scale was developed by Yıldırım Öztürk et al. (2020), and its Turkish validity and reliability study was conducted [25]. The scale consists of 24 items rated on a 5-point Likert scale ranging from “1-Strongly disagree” to “5-Completely agree.” Negatively worded items (items 9, 12, and 14–24) are reverse-scored in accordance with the original scoring procedure before calculation of the total score. The total score ranges from 24 to 120, with higher scores indicating more favorable attitudes toward cancer screening. In the original validation study, Cronbach’s alpha coefficient was reported as 0.95. In the present study, Cronbach’s alpha was 0.816.

Data collection procedure

Data were collected by the researchers using face-to-face interviews. Participants were informed about the aim of the study, and written informed consent was obtained. During the data collection process, anonymity was ensured and each participant was assigned a code number. No clinical intervention or biological sampling was performed within the scope of the study. Participants were informed that their responses would be kept confidential, would not be shared with healthcare personnel in an individually identifiable form, and would have no effect on the healthcare services they received. They were encouraged to provide honest responses, and no identifying information was recorded on the questionnaires. These procedures were intended to reduce concerns about disclosure and minimize potential social desirability bias.

Before recruitment, individuals who clearly did not meet the predefined eligibility criteria were identified with the assistance of primary care center staff and were not invited to participate in data collection. The same predefined inclusion and exclusion criteria were applied throughout the recruitment period to promote consistency in participant eligibility assessment. Eligible individuals attending the primary care center during the study period were invited to participate, and data were collected face-to-face by the researchers. A total of 603 individuals initiated the questionnaire. Of these, 14 discontinued participation before completing the questionnaire, and 21 returned questionnaires with missing responses. These 35 questionnaires were excluded from the analyses. Consequently, the final analytical sample consisted of 568 participants with complete data. No imputation procedure was applied; analyses were conducted using complete questionnaire data only.

Data analysis

Study data were analyzed using IBM SPSS Statistics 26.0. Continuous variables were presented as mean ± standard deviation, whereas categorical variables were presented as frequency (n) and percentage (%). The normality of scale scores was assessed using skewness and kurtosis values, and parametric test assumptions were found to be satisfied.

Independent samples t test was used for variables with two groups, and one-way analysis of variance (ANOVA) was used for variables with three or more groups when comparing scale scores according to sociodemographic characteristics. Following statistically significant omnibus ANOVA tests, Games–Howell post-hoc comparisons were performed to identify between-group differences because homogeneity of variance was not consistently satisfied. Because the sociodemographic comparisons were secondary and exploratory analyses involving multiple comparisons, no additional study-wide multiplicity adjustment was applied; therefore, these findings were interpreted cautiously and were not considered confirmatory analyses. Relationships among the Medical Mistrust Scale (MMS), Health System Distrust Scale (HSDS), Healthcare Demand Procrastination Behavior Scale (HDPBS), and Cancer Screening Attitude Scale (CSAS) were evaluated using Pearson correlation analysis.

Multiple linear regression analyses were conducted to examine the associations of medical mistrust and health system distrust with healthcare delay behavior and cancer screening attitudes. In the first regression model, HDPBS was included as the dependent variable, while MMS and HSDS were entered as independent variables. In the second regression model, CSAS was included as the dependent variable, with MMS and HSDS as independent variables. Multicollinearity was assessed using tolerance and variance inflation factor (VIF) values. Regression assumptions were further evaluated by examining residual distributions, studentized residuals, and Cook’s distance. Because evidence of heteroscedasticity was observed, HC3 heteroscedasticity-robust standard errors were used for coefficient inference. Unstandardized coefficients (B), standardized coefficients (β), robust standard errors (SE), 95% confidence intervals (CI), coefficients of determination (R²), and F statistics were reported. Statistical significance was accepted as p < 0.05.

As a sensitivity analysis, hierarchical multiple regression was additionally performed to assess the robustness of the associations after adjustment for selected sociodemographic characteristics. Gender and income status were entered in the first block, followed by medical mistrust and health system distrust in the second block. Changes in explained variance (ΔR²) and F-change statistics were examined to assess the additional contribution of the trust variables.

Because all primary study variables were assessed using self-report instruments at a single time point, potential common method variance was additionally examined using Harman’s single-factor test. All measurement items were entered simultaneously into an unrotated exploratory factor analysis, and the proportion of total variance accounted for by the first factor was examined. This analysis was used as a diagnostic assessment rather than as a definitive test of common method bias.

Results

A total of 603 participants initiated the questionnaire. Fourteen participants discontinued the questionnaire before completion, and 21 questionnaires were excluded because of missing responses. Consequently, 568 participants with complete data were included in the final analyses, representing 94.2% of the questionnaires initiated.

When the sociodemographic characteristics of the participants were examined, the majority were female (62.85%), married (74.12%), had an income lower than their expenses (54.23%), had an associate degree education (41.37%), and were employed in an income-generating job (61.62%). Most participants lived in the city center (51.06%), had a nuclear family structure (60.04%), and most frequently visited primary care centers once every six months (51.76%). The sociodemographic characteristics of the participants are presented in Table 1.

Table 1.

Distribution of sociodemographic characteristics of the participants

Variables		n	%
Age	30–35	120	21.13
	36–45	74	13.03
	46–55	189	33.27
	56–64	84	14.79
	≥ 65	101	17.78
Gender	Female	357	62.85
	Male	211	37.15
Marital Status	Married	421	74.12
	Single	147	25.88
Income Status	Income lower than expenses	308	54.23
	Income equal to expenses	172	30.28
	Income higher than expenses	88	15.49
Education Level	Primary	131	23.06
	High school	55	9.68
	Associate degree	235	41.37
	Bachelor’s degree	87	15.32
	Postgraduate	60	10.56
Employment Status	Yes	350	61.62
	No	218	38.38
Place of Residence	City	290	51.06
	District	206	36.27
	Town/Village	72	12.68
Family Type	Nuclear family	341	60.04
	Extended family	227	39.96
Frequency of Primary Care Visits	Once a month	72	12.68
	Once every three months	114	20.07
	Once every six months	294	51.76
	Once a year	88	15.49
Total		568	100

Comparisons of Healthcare Demand Procrastination Behavior Scale (HDPBS) scores according to sociodemographic variables showed statistically significant differences across all variables (p < 0.05). Healthcare delay behavior decreased with increasing age; the highest mean scores were observed in the 30–35 age group, whereas the lowest scores were found among individuals aged 65 years and older. Higher healthcare demand procrastination scores were identified among males, single individuals, participants reporting equal income and expenses, those with higher education levels, employed individuals, residents of city centers, and individuals attending primary care centers once a year (p < 0.05).

Significant differences were also found in Cancer Screening Attitude Scale (CSAS) scores according to sociodemographic variables (p < 0.05). Cancer screening attitude scores increased with age. Female participants, married individuals, those with lower income levels, lower educational attainment, and unemployed participants had significantly higher cancer screening attitude scores. In addition, individuals who attended primary care centers more frequently demonstrated more positive attitudes toward cancer screening. Detailed comparisons of HDPBS and CSAS scores according to sociodemographic characteristics are presented in Table 2.

Table 2.

Comparison of HDPBS and CSAS total scores according to sociodemographic variables

Sociodemographic Variables		SCALES
		HDPBS (Mean ± SD)	F/t-p	CSAS (Mean ± SD)	F/t-p
Age	30–35	49.35 ± 1.49 ᵃ	F:1272.42-p:<0.001	67.38 ± 2.24 ᵃ	F:333.98- p:<0.001
	36–45	45.68 ± 4.05 ᵇ		71.36 ± 1.67 ᵇ
	46–55	27.95 ± 5.04 ᶜ		72.87 ± 1.69 ᶜ
	56–64	22.67 ± 4.08 ᵈ		74.17 ± 1.72 ᵈ
	≥ 65	18.02 ± 3.54 ᵉ		75.26 ± 1.33 ᵉ
Gender	Female	28.30 ± 11.88	t:-10.96- p:<0.001	73.12 ± 2.62	t:9.57- p:<0.001
	Male	38.89 ± 10.66		70.46 ± 3.49
Marital Status	Married	26.14 ± 8.21	t:-56.04- p:<0.001	73.41 ± 2.12	t:18.11- p:<0.001
	Single	49.67 ± 1.53		68.46 ± 3.07
Income Status	Income lower than expenses	24.62 ± 9.61 ᵃ	F:237.69- p:<0.001	73.70 ± 2.54 ᵃ	F:109.99- p:<0.001
	Income equal to expenses	43.04 ± 9.57 ᵇ		70.26 ± 3.29 ᵇ
	Income higher than expenses	37.76 ± 7.04 ᶜ		70.27 ± 2.28 ᵇ
Education Level	Primary	17.70 ± 2.19 ᵃ	F:759.69- p:<0.001	75.38 ± 1.08 ᵃ	F:355.91- p:<0.001
	High school	25.91 ± 2.68 ᵇ		73.31 ± 1.73 ᵇ
	Associate degree	30.91 ± 7.35 ᶜ		72.34 ± 1.85 ᶜ
	Bachelor’s degree	48.99 ± 1.60 ᵈ		70.40 ± 2.22 ᵈ
	Postgraduate	50.65 ± 0.66 ᵉ		65.63 ± 1.57 ᵉ
Employment Status	Yes	35.71 ± 10.66	t:8.52- p:<0.001	71.23 ± 3.07	t:-8.97- p:<0.001
	No	26.64 ± 13.28		73.57 ± 2.98
Place of Residence	City	37.17 ± 11.51 ᵃ	F:120.92- p:<0.001	71.01 ± 3.41 ᵃ	F:73.76- p:<0.001
	District	31.01 ± 10.81 ᵇ		72.51 ± 2.48 ᵇ
	Town/Village	15.86 ± 0.94 ᶜ		75.54 ± 1.19 ᶜ
Family Type	Nuclear family	33.33 ± 12.38	t:2.56-p:0.011	71.91 ± 3.42	t:-2.09-p: 0.037
	Extended family	30.58 ± 12.58		72.47 ± 2.92
Frequency of Primary Care Visits	Once a month	15.86 ± 0.94 ᵃ	F:410.43- p:<0.001	75.54 ± 1.19 ᵃ	F:178.65- p:<0.001
	Once every three months	22.82 ± 3.54 ᵇ		74.28 ± 1.65 ᵇ
	Once every six months	34.33 ± 9.50 ᶜ		71.69 ± 2.17 ᶜ
	Once a year	50.80 ± 0.71 ᵈ		68.03 ± 3.81 ᵈ

Values are presented as mean ± SD. For variables with three or more categories, Games–Howell post-hoc comparisons were performed following significant omnibus ANOVA tests. Within each outcome and sociodemographic variable, means with different superscript letters (a–e) indicate statistically significant pairwise differences (p < 0.05). No post-hoc comparison was required for two-category variables. Sociodemographic comparisons were exploratory and were interpreted cautiously because no study-wide multiplicity adjustment was applied

Given the exploratory nature and number of sociodemographic comparisons, these findings should be interpreted cautiously because no adjustment for multiple testing was applied.

According to Pearson correlation analysis, statistically significant relationships were found among all scale scores (p < 0.001). A strong positive correlation was identified between medical mistrust and health system distrust (r = 0.80). Medical mistrust was strongly positively correlated with healthcare delay behavior (r = 0.90) and strongly negatively correlated with cancer screening attitudes (r = − 0.84). Similarly, health system distrust showed a strong positive correlation with healthcare delay behavior (r = 0.85) and a negative correlation with cancer screening attitudes (r = − 0.72). In addition, healthcare delay behavior was strongly negatively associated with cancer screening attitudes (r = − 0.82). The Pearson correlation coefficients among the study variables are shown in Table 3.

Table 3.

Correlations between study variables (n = 568)

Scales	Mean ± SD	1	2	3	4	95% CI
1. MMS	47.18 ± 9.90	-	0.80**	0.90**	− 0.84**	0.77–0.83; 0.89–0.92; −0.87–−0.82
2. HSDS	30.77 ± 4.84	-	-	0.85**	− 0.72**	0.83–0.88; −0.75–−0.67;
3. HDPBS	32.23 ± 12.53	-	-	-	− 0.82**	−0.85–−0.79
4. CSAS	72.13 ± 3.24	-	-	-	-	-

Note. MMS = Medical Mistrust Scale; HSDS = Health System Distrust Scale; HDPBS = Healthcare Demand Procrastination Behavior Scale; CSAS = Cancer Screening Attitude Scale. n = 568. The 95% confidence intervals for Pearson correlation coefficients were calculated using Fisher’s z transformation. **p < 0.001

Multiple linear regression analysis demonstrated that Medical Mistrust Scale (MMS) and Health System Distrust Scale (HSDS) scores together explained 86.5% of the variance in HDPBS scores (R² = 0.865, F = 1811.00, p < 0.001). Medical mistrust showed a strong positive association with healthcare delay behavior (β = 0.59, p < 0.001), whereas health system distrust showed a moderate positive association (β = 0.35, p < 0.001).

In the second regression model examining cancer screening attitudes, MMS and HSDS scores explained 71.4% of the variance in CSAS scores (R² = 0.714, F = 708.80, p < 0.001). Medical mistrust showed a strong negative association with cancer screening attitudes (β = −0.79, p < 0.001), whereas health system distrust showed a smaller but significant negative association (β = −0.12, p = 0.006).

Regression diagnostics indicated no problematic multicollinearity (VIF = 2.77 and tolerance = 0.36 for both predictors). No highly influential observations were identified based on Cook’s distance (all values < 1), and no studentized residuals exceeded an absolute value of 3. Residual distributions showed no substantial departures from normality, although heteroscedasticity was detected. Therefore, HC3 heteroscedasticity-robust standard errors were used for coefficient inference. The results of the multiple linear regression analyses are presented in Table 4.

Table 4.

Predictors of HDPBS and CSAS according to regression analysis

Variables	B	SE	95% CI for B	β	t	P	Tolerance	VIF	R	R 2	F	p
Model 1: Constant (HDPBS)	—33.30	1.25	—	—	-26.66	< 0.001	—	—	0.93	0.87	1811.00	< 0.001
MMS	0.78	0.03	0.72–0.84	0.59	26.83	< 0.001	0.36	2.77
HSDS	0.94	0.07	0.81–1.07	0.35	13.99	< 0.001	0.36	2.77
Model 2: Constant (CSAS)	86.09	0.47	—	—	183.41	< 0.001	—	—	0.85	0.71	708.80	< 0.001
MMS	-0.25	0.01	−0.27–−0.22	-0.79	− 18.92	< 0.001	0.36	2.77
HSDS	-0.08	0.03	−0.13–−0.02	-0.12	-2.75	0.006	0.36	2.77

Note. MMS = Medical Mistrust Scale; HSDS = Health System Distrust Scale; HDPBS = Healthcare Demand Procrastination Behavior Scale; CSAS = Cancer Screening Attitude Scale. SE represents HC3 heteroscedasticity-robust standard errors. CI = confidence interval; VIF = variance inflation factor

As a sensitivity analysis, hierarchical regression models were additionally performed to assess whether the observed associations remained after adjustment for selected sociodemographic characteristics. Gender and income status were entered in the first block, followed by medical mistrust and health system distrust in the second block. For healthcare delay behavior, the addition of the trust variables substantially increased the explained variance from R² = 0.459 to R² = 0.910 (ΔR² = 0.450, F-change = 1403.33, p < 0.001). Medical mistrust (β = 0.58, p < 0.001) and health system distrust (β = 0.26, p < 0.001) both remained independently associated with greater healthcare delay behavior. For cancer screening attitudes, the explained variance increased from R² = 0.289 to R² = 0.731 after the trust variables were added (ΔR² = 0.443, F-change = 462.67, p < 0.001). Medical mistrust remained strongly and negatively associated with cancer screening attitudes (β = −0.75, p < 0.001), whereas the independent association with health system distrust was attenuated and was no longer statistically significant after adjustment (β = −0.03, p = 0.428). VIF values in the adjusted models remained below conventional thresholds, indicating no problematic multicollinearity.

In practical terms, holding health system distrust constant, each one-point increase in the Medical Mistrust Scale score was associated with an estimated 0.78-point higher HDPBS score. Similarly, holding health system distrust constant, each one-point increase in medical mistrust was associated with an estimated 0.25-point lower CSAS score. These coefficients represent statistical associations within the observed sample and should not be interpreted as causal effects or as established clinically meaningful changes.

The diagnostic assessment of common method variance showed that the first unrotated factor accounted for 81.48% of the total variance. This finding indicates substantial shared variance among the self-reported measures and suggests that common method variance cannot be excluded when interpreting the magnitude of the observed associations. Accordingly, the particularly strong correlations observed among the study constructs should be interpreted cautiously.

Discussion

The findings suggest that medical mistrust and health system distrust are substantially associated with healthcare-seeking behaviors and attitudes toward cancer screening among adults attending primary care services.

The sociodemographic findings showed that the sample mainly consisted of women, married individuals, and individuals with low income levels. This finding is consistent with studies indicating that healthcare utilization is associated with socioeconomic factors among individuals using primary care services. Indeed, trust in healthcare systems has been reported to be closely associated with individuals’ socioeconomic conditions, experiences in accessing healthcare services, previous contact with healthcare institutions, and perceptions regarding the fairness of the healthcare system [1, 26, 27].

Correlation analysis demonstrated strong positive relationships between medical mistrust, health system distrust, and healthcare delay behavior, whereas strong negative relationships were found between these variables and attitudes toward cancer screening. These findings indicate that trust in healthcare systems is closely associated with both healthcare-related attitudes and health behaviors. Previous studies have widely reported that medical mistrust is associated with healthcare avoidance, delayed healthcare seeking, lower healthcare utilization, and reduced use of healthcare services in general [7, 11, 26, 28]. In addition, distrust in healthcare systems has been associated with lower participation in preventive healthcare services such as cancer screening programs [29–31]. In this context, the negative relationships identified between medical mistrust, health system distrust, healthcare delay behavior, and cancer screening attitudes are consistent with previous literature.

However, previous evidence has not been entirely uniform regarding the strength and consistency of these relationships. The association between mistrust and healthcare engagement may vary according to the dimension of trust assessed, the type of healthcare service, population characteristics, previous healthcare experiences, and structural barriers to care. In some settings, factors such as accessibility, affordability, perceived need, and prior experiences with healthcare may coexist with or outweigh trust-related factors in shaping healthcare utilization. Thus, the strong associations observed in the present study should be interpreted within the specific primary care and sociocultural context of the study rather than as universally consistent relationships across populations and healthcare systems.

One of the important findings of the study was the strong positive associations of medical mistrust and health system distrust with healthcare delay behavior. In the regression analysis, medical mistrust showed the strongest independent association with healthcare delay behavior. These findings indicate that higher levels of medical mistrust are associated with greater healthcare delay behavior. Previous studies have shown that delayed healthcare seeking may lead to negative outcomes, particularly in diseases where early diagnosis is critical, and that healthcare avoidance and delay behaviors may be associated with risk perception, fear, distrust, low perceived control, and previous negative healthcare experiences [14, 32, 33]. Accordingly, the present findings support a close association between medical mistrust and decisions regarding timely healthcare seeking, without establishing a causal relationship. Given the cross-sectional design, the direction of these relationships cannot be determined. Although greater mistrust may be associated with delayed healthcare seeking, the reverse pathway is also plausible; repeated delays, unmet healthcare needs, or previous negative experiences with healthcare services may contribute to greater mistrust. Moreover, unmeasured factors such as health literacy, health anxiety, perceived healthcare need, previous negative healthcare experiences, and structural barriers to care may be associated with both mistrust and healthcare-seeking behavior. Therefore, the observed associations should not be interpreted as evidence of a unidirectional or causal pathway.

The sensitivity analyses further supported the robustness of the principal findings after adjustment for gender and income. Medical mistrust remained strongly associated with both healthcare delay behavior and cancer screening attitudes, while health system distrust remained independently associated with healthcare delay behavior. However, the association between health system distrust and cancer screening attitudes was attenuated and was no longer statistically significant after adjustment. This finding suggests that the observed relationship between health system distrust and cancer screening attitudes may partly reflect sociodemographic differences, whereas the associations involving medical mistrust appeared comparatively robust. Although the high R² values observed in the primary models warrant caution regarding potential model overfitting, several diagnostic and sensitivity analyses supported the stability of the observed associations. Multicollinearity diagnostics remained within acceptable limits, no highly influential observations were identified, and the principal association of medical mistrust with both outcomes remained substantial after adjustment for selected sociodemographic characteristics. Nevertheless, given the cross-sectional single-center design and the unusually high proportion of explained variance, replication in independent samples and prospective validation are needed to establish the generalizability and stability of these models.

The study also found that health system distrust was significantly and positively associated with healthcare delay behavior. This finding indicates that more negative perceptions of healthcare institutions, healthcare policies, and service delivery processes are associated with greater healthcare delay behavior. Previous studies have similarly reported associations between distrust in healthcare systems and individuals’ perceptions of healthcare institutions, healthcare delivery processes, and the adequacy of healthcare systems, which may be related to healthcare decision-making [2, 15, 34, 35]. In the context of preventive healthcare services, greater health system distrust may also be associated with less favorable perceptions of the necessity, reliability, and benefits of cancer screening.

Another notable finding of the study was that medical mistrust and health system distrust represent distinct yet complementary constructs operating at different levels. While medical mistrust is more closely related to individuals’ relationships and clinical experiences with healthcare professionals, health system distrust reflects broader structural and institutional perceptions. Previous studies have emphasized that although these concepts are often used interchangeably, they operate at different levels and may be associated with health behaviors through different pathways [4, 28]. Accordingly, strategies aimed at promoting favorable attitudes toward cancer screening may need to address both individual level trust based communication and system level factors related to transparency, accessibility, and quality of healthcare services.

The magnitude of several correlations observed in the present study also warrants cautious interpretation. Medical mistrust, health system distrust, healthcare delay behavior, and cancer screening attitudes are conceptually related constructs reflecting interconnected perceptions and behavioral orientations toward healthcare. Their strong associations may therefore partly reflect genuine conceptual proximity within the broader domain of healthcare trust and engagement. At the same time, because all constructs were assessed through self-report instruments within the same survey and at the same time point, shared method variance may have contributed to the magnitude of these relationships. Consistent with this possibility, the diagnostic single-factor analysis revealed substantial shared variance among the measurement items. Therefore, the observed correlations should not be interpreted as evidence that the constructs are interchangeable or as evidence of causal relationships. Rather, they indicate closely related but theoretically distinct dimensions that should be examined using longitudinal, multi-method, and independently measured outcomes in future studies.

The statistical strength of these associations should not be equated directly with clinical or practical significance. Although several correlation coefficients and regression estimates were large and statistically significant, the present study did not assess clinically defined thresholds, objectively verified healthcare utilization, actual cancer screening participation, or patient outcomes. Therefore, the magnitude of the observed statistical associations indicates strong relationships within this sample but does not by itself establish that corresponding differences are clinically meaningful. The practical significance of these findings should be evaluated in future prospective studies using objective behavioral and clinical outcomes.

The contribution of the present study should be interpreted primarily as a contextual extension of existing evidence rather than as the identification of previously unknown relationships. Previous research has established associations between mistrust and healthcare engagement; however, the present study examined interpersonal-level medical mistrust and broader health system distrust concurrently in relation to both healthcare delay behavior and cancer screening attitudes within the same primary care population in eastern Türkiye. Examining these constructs together provides an integrated perspective on how different dimensions of distrust coexist with delayed healthcare seeking and less favorable attitudes toward preventive services in this specific setting. These context-specific findings may help inform future prospective and intervention studies designed to evaluate whether trust-centered strategies can improve timely healthcare seeking and engagement with cancer screening.

Sociodemographic findings further showed that healthcare delay behavior was higher among younger individuals. This pattern may reflect differences in perceived health risks and tendencies to postpone healthcare utilization across age groups. In contrast, older individuals demonstrated more positive attitudes toward cancer screening, which may be related to greater health awareness or more frequent contact with healthcare services. Additionally, individuals attending primary care centers more frequently had more positive attitudes toward cancer screening, highlighting an association between primary care engagement and preventive health attitudes. Recent studies have also demonstrated that regular cancer screening is effective in reducing mortality and increasing early diagnosis [36, 37]. More broadly, the relevance of trust to healthcare engagement is consistent with the foundational work of Hall et al., who conceptualized trust in physicians and medical institutions as a multidimensional construct with important implications for healthcare relationships [38]. In the context of health services research, these findings suggest that trust may be an important correlate of healthcare utilization decisions and engagement with preventive services.

From a primary care perspective, the observed associations suggest several practical areas that may warrant attention, although the cross-sectional design does not establish that these strategies would directly change healthcare-seeking or screening behaviors. Primary care professionals may support trust by using clear and consistent communication, shared decision-making, transparent explanations of diagnostic and treatment processes, and sufficient opportunities for patients to express concerns or uncertainty. At the community level, outreach activities and culturally appropriate health education may help engage individuals who are reluctant to use preventive services, while patient-navigation approaches may facilitate access to recommended cancer screening by helping individuals understand and move through referral and screening pathways. At the health-system level, greater transparency regarding service procedures, waiting processes, eligibility criteria, and continuity of care may contribute to a more predictable and trustworthy care environment. Implementation of these approaches may nevertheless be constrained by limited consultation time, workforce shortages, high patient volume, resource limitations, and differences in local service capacity. Therefore, these strategies should be regarded as potential directions for intervention development and evaluation rather than as interventions proven effective by the present study. Future pragmatic and intervention studies are needed to determine their feasibility, acceptability, cost implications, and effectiveness in primary care settings.

Limitations

This study has several limitations. First, although relationships among variables were identified, causal inferences cannot be made due to the cross-sectional design. The internal consistency of the Health System Distrust Scale in the present study was acceptable (Cronbach’s α = 0.767), although lower than that observed for some of the other study measures. This level of reliability may introduce some measurement error and should be considered when interpreting the magnitude and precision of associations involving health system distrust. Future studies may further evaluate the psychometric performance of this measure across different primary care populations and sociocultural settings. Second, the study was conducted in a single center using convenience sampling, which may limit generalizability. In addition, because the number and characteristics of eligible individuals who declined participation were not systematically recorded, a response rate could not be calculated and potential non-response bias could not be fully assessed. Individuals attending the primary care center during the study period may also differ systematically from those who did not seek primary care, particularly in their healthcare-seeking behaviors and attitudes toward healthcare. Therefore, selection bias cannot be excluded. Third, the use of self-report measures may have resulted in social desirability and recall bias. A further limitation concerns potential common method variance. In addition, restricting eligibility to adults aged 30–70 years was consistent with the study’s focus on age-relevant cancer screening attitudes but limits the applicability of the findings to younger adults and individuals older than 70 years. In addition, although 568 of the 603 initiated questionnaires were complete and included in the final analysis, 14 participants discontinued participation and 21 questionnaires contained missing responses. The total number of otherwise potentially eligible individuals who declined participation before questionnaire initiation was not systematically recorded. Therefore, an overall participation rate could not be calculated, and the possibility of non-response bias cannot be excluded.

Because the principal variables were measured using self-report instruments within the same survey and at a single time point, shared measurement characteristics may have inflated some of the observed associations. The diagnostic single-factor analysis identified substantial shared variance, and common method variance therefore cannot be excluded. Nevertheless, this diagnostic finding does not establish that the observed associations are entirely attributable to methodological artifact. The constructs were assessed using separate validated instruments developed to measure theoretically distinct domains. Accordingly, the findings should be interpreted as evidence of associations rather than causal or fully independent effects, and replication using longitudinal designs, multi-method assessments, and objective healthcare utilization or screening data is warranted. The external validity of the findings should also be considered in relation to the study setting and sampling strategy. Participants were recruited through convenience sampling from a single primary care center in eastern Türkiye; therefore, the sample should not be considered representative of all primary care users in Türkiye. The demographic composition of the sample and the specific sociocultural and healthcare context of the region may have shaped perceptions of medical and health system trust as well as healthcare-seeking attitudes. Because participants were recruited from primary care attendees rather than through population-based probability sampling, direct comparison with regional or national population demographics would not provide a valid assessment of representativeness. In particular, the relatively high proportions of women, married participants, and individuals reporting income below expenses in the present sample may reflect both the characteristics of primary care users attending the study center and the local context. These characteristics should therefore be considered when interpreting the findings. Cultural norms, socioeconomic conditions, previous experiences with healthcare institutions, and local patterns of healthcare access may also shape both trust and healthcare-seeking attitudes, and these contextual influences may differ across regions of Türkiye and other healthcare systems.

Accordingly, the present findings are most appropriately interpreted within the context of the population studied rather than generalized directly to other regions or healthcare systems. Future multicenter studies using probability-based sampling across geographically and socioeconomically diverse populations are needed to examine the transferability of these associations.

In addition, actual cancer screening behaviors were not assessed; instead, attitudes toward cancer screening were evaluated. Furthermore, variables such as health literacy, health anxiety, and previous healthcare experiences were not included in the study and may have influenced the findings. Future studies should use longitudinal and multicenter designs with larger and more representative samples.

Conclusion

This study demonstrated that medical mistrust and health system distrust are significant factors associated with delayed health services utilization and less favorable attitudes toward cancer screening among adults attending primary care services. Higher levels of distrust were associated with increased healthcare delay behavior and reduced positive attitudes toward participation in cancer screening programs.

These findings highlight the close relationship between trust, health services utilization, and engagement with preventive health services. Trust centered strategies that strengthen patient provider relationships, enhance transparency within healthcare systems, and facilitate timely access to care may be relevant for addressing healthcare delay and unfavorable attitudes toward cancer screening. However, prospective and intervention studies incorporating objectively verified healthcare utilization and actual screening participation are needed to determine whether such strategies translate into behavioral improvements. From a health services research perspective, developing trust-centered strategies within primary care settings may represent an important approach to promoting equitable access to care and strengthening the effectiveness of preventive health services.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors have materially participated in the research and manuscript preparation. All authors have approved the final manuscript. Mehmet Uçar: Investigation, Conceptualization, Investigation, Writing – original draft, Writing – review & editing. Muhammet Faruk Yiğit: Conceptualization, Investigation, Writing – original draft, Writing – review & editing. Mehmet Kaplan: Conceptualization, Investigation, Writing – original draft, Writing – review & editing.

Funding

No funds, grants, or other support was received.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Before the study was initiated, ethical approval was obtained from the Scientific Research and Publication Ethics Committee of Muş Alparslan University (Approval No: 228433; Approval Date: 12.01.2026). Written institutional permission was also obtained from Van Tuşba Primary Care Center where the study was conducted (Approval No: 829455; Approval Date: 03.02.2026). Before participation, all individuals were informed in detail about the purpose, scope, and voluntary nature of the study. Written informed consent was obtained from participants who agreed to participate. Participants were informed that they could withdraw from the study at any time without providing any reason. All stages of the study were conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.