Prostate cancer patients who visited the Department of Urology at Guangdong Provincial Hospital of Traditional Chinese Medicine from September to October 2025 were enrolled. Inclusion criteria included: 1) pathological diagnosis of prostate cancer; 2) being in the treatment decision-making phase, where all participants are newly diagnosed and have not received any prior treatment at enrollment; 3) maintaining clear consciousness. Exclusion criteria included: 1) distant metastasis; 2) severe cognitive or psychiatric disorders. A total of 251 valid questionnaires were collected. This study was approved by the Ethics Committee of Guangdong Provincial Hospital of Traditional Chinese Medicine (Approval No.: ZE2025-360-01), and all patients signed informed consent forms.

2.2.1. General Information Questionnaire

Designed by the researchers themselves, these variables include age, marital status, educational level, per capita monthly household income, method of medical expense payment, and Clinical Cancer Stage (with Clinical Cancer Stage based on the International Union Against Cancer’s TNM classification criteria).

2.2.2. Decision Conflict Scale (DCS)

This scale was developed by O’Connor [6] and translated into Chinese by Wang Lu et al. [7]. It consists of three dimensions: information and values, decision support and decision effectiveness, and decision uncertainty, with a total of 16 items. The Likert 5-level scoring system is adopted (0 = strongly agree, 4 = strongly disagree, with question 11 being scored in reverse). Convert the raw score to a standard score ranging from 0 to 100. The higher the score, the more severe the decision-making conflict. A score of less than 25 indicates no decision conflict, while a score of more than 37.5 indicates a high level of decision conflict. In this study, the Cronbach’s α coefficient of this scale was 0.831.

2.2.3. Generalized Anxiety Scale (GAD-7)

Compiled by Spitzer et al. [8], it consists of 7 items and is scored from 0 to 3 points, with a total score ranging from 0 to 21 points. A score of ≥5 indicates mild anxiety, while a score of ≥10 indicates moderate to severe anxiety. In this study, the Cronbach’s α coefficient of this scale was 0.85.

2.2.4. Patient Health Questionnaire Depression Scale (PHQ-9)

This scale was developed by Kroenke et al. [9] and consists of 9 items. It is scored on a scale of 0 to 3, with a total score ranging from 0 to 27. A score of ≥5 indicates mild depression, and ≥10 indicates moderate to severe depression. The Cronbach’s α coefficient of this scale is 0.87.

Data analysis was completed using SPSS27.0 statistical software. The measurement data were expressed as mean ± standard or median (quartile), and the Mann-Whitney U rank sum test was used for comparison between the two groups. Counting data were expressed as n (%), and the χ² test was used for comparison between groups. Correlation analysis was conducted using Spearman’s rank correlation analysis; Using the occurrence of high decision conflict as the dependent variable (high decision conflict = 1; non-high decision conflict = 0), statistically significant variables identified in the univariate analysis were included in a binary logistic regression model. All categorical variables were entered as dummy variables, with reference groups defined as follows: age (<65 years), highest education level (primary school or below), marital status (widowed), employment status (employed), alcohol consumption (never drinking), and cancer clinical stage (T1 stage); total anxiety scores and total depression scores were included directly as continuous variables. Prior to model inclusion, multiple collinearity tests were performed using the variance inflation factor (VIF) and tolerance criteria, with VIF < 5 and tolerance > 0.2 serving as thresholds for no significant collinearity, thereby eliminating potential interference from collinearity on regression results. A difference was considered statistically significant when P < 0.05.

A total of 251 tumor patients were included in this study, among whom 82.07% were aged 65 or above. The educational attainment was mainly junior high school (29.88%) and senior high school/technical secondary school (26.69%). The marital status was mostly married or cohabiting (89.24%). The employment status is mainly composed of retirees (88.45%). The proportion of families with an annual income of less than 100,000 yuan is 47.81%. Among smoking statuses, former smokers accounted for the highest proportion (44.62%). The majority of those who drink alcohol do not drink at all (58.96%). The tumor stages T2, T4 and unclear accounted for a relatively high proportion, which were 20.72%, 20.72% and 40.64%, respectively. Patients whose staging has been marked as “unknown” primarily include those who have not completed the full set of imaging-based staging examinations after initial pathological diagnosis, or those who first visited our hospital following a biopsy diagnosis at another institution and have not yet undergone comprehensive systematic staging evaluation; thus, their clinical T-stage cannot be determined at this stage. For details, please refer to Table 1.

Table 1.Demographic sociological characteristics of the research subjects.

The total score standard of decision conflict for this group of patients was 34.69 ± 19.33 points. The scores of each dimension were similar. The dimension of information and values (35.99 ± 19.57) was slightly higher, and the dimension of decision uncertainty (34.56 ± 21.81) had a slightly greater degree of dispersion, suggesting that the overall decision conflict was at a medium level, and individual differences were relatively obvious. See Table 2.

Table 2.Scores of the decision conflict scale for cancer patients.

Among the patients, the proportion of those without anxiety was the highest, reaching 52.59%, while the proportions of those with mild and moderate to severe anxiety decreased successively. The overall depressive mood was mainly non-depressive, accounting for 70.52%. Among the people with depressive mood, the proportion of severe depression was relatively high, while the detection rates of mild, moderate and moderate-severe depression were relatively low. It can be seen that the detection rate of anxiety in the research subjects was higher. Severe depression was the main prominent manifestation of depressive mood, as shown in Table 3.

Table 3.Scores of the anxiety and depression scale for prostate cancer patients.

Spearman correlation analysis showed that the total scores of anxiety and depression were significantly positively correlated with the total score of decision conflict (rₛ = 0.631, 0.677, both P < 0.001), indicating that the more severe the anxiety and depression of the patients, the higher the level of treatment decision conflict, as shown in Table 4.

Table 4.Correlation analysis of anxiety, depression and decision conflict in prostate cancer patients.

The level of decision conflict was divided into the high decision conflict group and the non-high decision conflict group with a critical value of 37.5. The results of univariate analysis showed that there were statistically significant differences in age, highest educational attainment, marital status, employment status, alcohol consumption, clinical stage of cancer, and levels of anxiety and depression between the two groups (all P < 0.05). However, there was no statistically significant difference in annual household income and smoking status between the two groups (P > 0.05). The high decision-making conflict group is characterized by lower age, lower educational attainment, higher rates of divorce/separation and widower, higher rates of non-retirement and alcohol consumption, higher rates of advanced tumors, and more severe anxiety and depression, as shown in Table 5.

Table 5.Univariate analysis of decision-making conflicts in prostate cancer patients.

The multicollinearity test prior to inclusion in the regression model revealed that all variables had VIF values <3 and tolerance coefficients >0.3. Specifically, the VIF values for the total anxiety score and total depression score were 2.14 and 2.09, respectively, both significantly below the critical threshold of 5, indicating no significant multicollinearity among the variables. Thus, the selection of model variables was appropriate. The univariate meaningful variables were included in the binary Logistic regression. The results showed that the total score of depression was an independent risk factor for decision conflict (OR = 1.252, 95%CI: 1.074 - 1.459, P = 0.003). Age, educational background, marriage, employment, alcohol consumption, tumor stage and total anxiety score did not enter the final independent influence model (all P > 0.05), suggesting that depressive mood is the core independent factor affecting the conflict in treatment decisions of tumor patients, as shown in Table 6.

Table 6.Binary logistic regression analysis of the influencing factors of decision-making conflicts in cancer patients.

Note: “-” indicates no exact value.

The results of this study show that the total criterion score of decision conflict in 251 prostate cancer patients was (34.69 ± 19.33) points, and the incidence of high decision conflict (DCS > 37.5 points) was 31.1%, which was at a medium level overall and basically consistent with the results of similar studies at home and abroad. The study by Steginga S K et al. [10] shows that approximately 63% of prostate cancer patients have moderate to severe decision-making conflicts; Wang et al.’s investigation of 192 patients with bladder cancer in the decision-making stage also found that the incidence of high decision conflict was 77.5%, which was similar to the results of this study [11]. This result confirms the high preference sensitivity of prostate cancer treatment decisions: there is no significant difference in 10-year tumor-specific survival rates among the current mainstream regimens such as radical prostatectomy, external radiotherapy, and active monitoring, but there are significant differences in quality of life outcomes such as urinary incontinence, erectile dysfunction, and intestinal adverse reaction [12]. Patients need to make individualized value trade-offs between “survival benefits” and “functional preservation”, and this dilemma is the core cause of the high incidence of decision-making conflicts.

From the perspective of dimension scores, the scores of the information and values dimension are relatively the highest, suggesting that the asymmetry of disease information and the ambiguity of personal value orientation are the main sources of decision-making conflicts. The diagnosis and treatment of prostate cancer involve a large number of professional medical terms and risk probability information, making it difficult for ordinary patients to accurately understand the long-term outcomes of different regimens. Meanwhile, the treatment outcome is highly bound to male self-identity and quality of life. Most patients have not established a clear value ranking in the early stage of diagnosis, which further aggravates the decision-making predicament. In addition, the degree of dispersion of the decision-making uncertainty dimension is the greatest, reflecting that there are significant individual differences in the decision-making adaptability of different patients, providing a target for the implementation of stratified intervention in clinical practice.

The core finding of this study is that after controlling for confounding factors such as demographic characteristics, disease stage, and lifestyle habits, depression is an independent risk factor for high decision conflict in prostate cancer patients (OR = 1.252, 95% CI: 1.074 - 1.459), while the independent effect of anxiety is not statistically significant. Previous studies have mostly combined anxiety and depression as “psychological distress” for overall analysis, masking the differences in their effects. This study, however, has clarified the specific predictive value of depression in decision-making conflicts, suggesting that there are essential differences in the mechanisms by which the two emotional states affect the decision-making process.

Spearman correlation analysis shows that both anxiety and depression are significantly positively correlated with the total score of decision conflict, which is consistent with the conclusions of most previous studies [11][13], indicating that the higher the degree of emotional distress, the more severe the decision conflict. However, when included in the multivariate logistic regression model, anxiety’s independent effect became insignificant, suggesting that depression may serve as a key mediating variable linking anxiety and decision conflict. As no formal mediation test was conducted in this study, it remains uncertain whether anxiety’s impact on decision conflict is fully mediated by depression; the exact pathway of interaction and causal direction require further validation through longitudinal studies and mediation analyses. From the perspective of psychopathology, the core of anxiety is excessive worry about the uncertainty of the future, which is a common emotional response in the stage of cancer diagnosis [14]. For newly diagnosed prostate cancer patients, regardless of the level of decision-making conflicts, concerns about tumor progression and treatment side effects are widespread. Therefore, anxiety is more of an accompanying baseline emotion and it is difficult to specifically distinguish the severity of decision-making conflicts.

In contrast, depression has an independent predictive effect on decision-making conflicts. The core symptoms of depression include anhedonia, reduced energy, a sense of hopelessness, and decreased cognitive flexibility, which can impair decision-making ability through multiple pathways. First, individuals in a depressive state have negative cognitive biases, tend to magnify the risks and negative outcomes of treatment plans, and find it difficult to objectively weigh the pros and cons of different plans, thus falling into a decision-making dilemma [15]. Secondly, the energy exhaustion and decision-making fatigue accompanied by depression will intensify self-depletion, reduce the patient’s ability to process and integrate information, making it impossible for them to effectively handle complex diagnostic and therapeutic information, and eventually lead to “decision paralysis” [16]. Thirdly, the treatment decisions for prostate cancer involve choices that are highly related to male identity, such as sexual function and urination function. When patients are constantly struggling between “saving their lives” and “maintaining the quality of life”, they are very likely to feel helpless and have a sense of worthlessness, which in turn aggravates depressive moods and forms a vicious cycle.

Univariate analysis showed that age, educational background, marital status, employment status, frequency of alcohol consumption, and clinical stage of cancer were all associated with high decision conflict. However, none of the above factors became independent risk factors in the multivariate analysis, suggesting that depression may play a significant role in the association between these demographic, behavioral, and clinical factors and decision conflict.

In terms of alcohol consumption frequency, the proportion of patients in the high decision-making conflict group who drink alcohol is significantly higher. When facing difficult treatment predicaments and accompanying negative emotions, some patients will drink alcohol to relieve psychological pain and avoid decision-making pressure. At the same time, long-term alcohol consumption may also damage the cognitive function and emotional regulation ability of the prefrontal lobe, further reducing decision-making efficiency and forming a mutually reinforcing cycle. However, after controlling for the level of depression, the effect of alcohol consumption was no longer significant, suggesting that depression may serve as a key intermediate variable in the association between alcohol intake and decision-making conflict. This inference is based solely on indirect evidence from multivariate regression results and has not been validated by standardized mediation analysis; the causal pathways and underlying mechanisms between these factors require further investigation.

In terms of disease staging, patients with T3 and T4 stages exhibited a higher prevalence of decision-making conflict, which correlates with more complex treatment regimens and greater prognostic uncertainty in advanced-stage patients. However, installment payments were also excluded from the regression equation, suggesting that the impact of disease severity and demographic characteristics on decision-making conflict may partially operate through depressive mood. This mechanistic hypothesis requires further validation in subsequent studies. Among demographic characteristics, younger age, lower education levels, divorce/widowed status, and non-retired status were associated with higher decision-making conflict: younger patients demonstrated stronger demands for preserving sexual and social functioning, leading to more intense psychological conflicts during decision-making trade-offs; patients with lower education levels exhibited relatively poorer health literacy and weaker comprehension of medical information, resulting in more pronounced information asymmetry; patients with poor marital status or remaining employed faced greater pressures from family and social roles, accompanied by heightened decision-making concerns [17][18]. Nevertheless, these demographic factors’ effects were attenuated in the multivariate model, further indicating that emotional factors—particularly depression—are the core intrinsic drivers of decision-making conflict.

The results of this study have clear guiding value for clinical decision support in prostate cancer. Firstly, depression screening should be incorporated into the routine assessment process before treatment decision-making, and simple scales such as PHQ-9 should be used to quickly identify high-risk patients. For patients with high depression scores, clinical medical staff should not urge them to make treatment choices immediately. Instead, they should first collaborate with psychological specialists for emotional intervention. After the depressive symptoms are relieved and cognitive functions recover, decision-making counseling should be gradually provided to prevent patients from making hasty and regretful decisions in a state of emotional exhaustion.

Secondly, the decision support model should transform from information notification to value clarification. This study shows that the dimensions of information and values are the core sources of decision-making conflicts, and patients with depression often have a sense of ambiguous values and hopelessness. Specialist nurses or decision counselors can guide patients through one-on-one structured interviews to sort out their core personal demands, such as prioritizing survival time, prioritizing the preservation of sexual function, and prioritizing the avoidance of urinary incontinence, etc. This helps patients screen out plans that meet their own wishes within a clear value framework and reduce self-consumption in their hearts.

In addition, for high-risk groups with high alcohol consumption frequency, advanced tumors, and insufficient social support, a multi-disciplinary collaborative decision support system should be established. The urology department, psychology department, and nursing team should jointly intervene, starting from multiple dimensions such as emotion regulation, disease popularization, and social support, to reduce the level of decision-making conflicts for patients and improve the quality of decision-making.

This study has certain limitations. First, the research is a cross-sectional design, which can only reveal the correlations among variables but cannot clarify the causal sequence and direction of action between depression and decision-making conflicts. In the future, longitudinal tracking studies can be conducted to explore the dynamic impact of depressive emotions on the decision-making process, decision satisfaction, and decision regret. Second, the absence of clinical staging information for some patients in this study may underestimate the true effect size of disease stage on decision-making conflicts, thereby affecting the stability of multivariate logistic regression results to some extent. Third, although the sample included patients with different clinical stages and those with undetermined initial diagnosis stages during the decision-making period, no stratified subgroup analysis was conducted based on staging; thus, the findings are more applicable to the overall population of prostate cancer patients at the initial diagnosis stage, while their specificity in specific prostate cancer subgroups requires further validation through stratified studies.