Latent profile analysis of uncertainty in illness and eating self-efficacy in patients with gastric cancer and its associated factors: a cross-sectional study

Abstract

Background:

Patients with gastric cancer frequently experience substantial psychological burden throughout diagnosis and treatment. Uncertainty in illness and eating self-efficacy are two key determinants of psychological adjustment and rehabilitation. However, prior research has predominantly relied on variable-centered approaches, with limited attention to within-population heterogeneity and its correlates among patients with gastric cancer.

Objective:

This study aimed to identify latent profiles of uncertainty in illness and eating self-efficacy among patients with gastric cancer and to examine factors associated with profile membership.

Methods:

We conducted a cross-sectional survey and recruited 426 patients with gastric cancer via convenience sampling from two tertiary hospitals in Zhejiang Province, China, between September and December 2025.

Results:

Two latent profiles were identified: a Low Uncertainty in Illness–High Eating Self-efficacy profile and a High Uncertainty in Illness–Low Eating Self-efficacy profile. Multivariable logistic regression showed that, using the high uncertainty in illness–low eating self-efficacy class as the reference, health information avoidance, interoceptive sensitivity, male sex, retirement status, lower income, and urban residence were significantly associated with a greater likelihood of belonging to the High Uncertainty in Illness–Low Eating Self-efficacy profile, whereas a history of alcohol consumption was associated with a greater likelihood of belonging to the Low Uncertainty in Illness–High Eating Self-efficacy profile.

Conclusion:

These findings suggest that clinicians should deliver stratified, individualized psychological support and nutrition counseling to facilitate patients’ psychological adjustment and rehabilitation.

1 Introduction

Gastric cancer is among the most common malignancies worldwide (1, 2). In recent years, advances in early detection and multimodal therapy have improved survival outcomes for some patients (3, 4); nonetheless, the intrinsic complexity of the disease, the protracted treatment course, and persistent uncertainty regarding prognosis and recurrence continue to impose substantial psychological strain (5). A diagnosis of gastric cancer signifies not only physiological disruption but also heightened concerns about disease trajectory (6, 7), doubts about treatment effectiveness, and anxiety regarding future quality of life. Accumulating evidence indicates that psychological factors play a nontrivial role in cancer patients’ illness adaptation, treatment adherence, and recovery (8, 9). Therefore, elucidating the psychological characteristics of this population and their determinants is of considerable theoretical and clinical importance for developing tailored psychological interventions and improving overall quality of life.

Uncertainty in illness refers to the cognitive difficulty individuals experience when constructing meaning from illness-related events (10, 11). Importantly, uncertainty in illness is not merely a negative emotional experience; rather, it is a complex cognitive state shaped jointly by the stimulus frame, cognitive capacity, and structure providers (12). For patients with gastric cancer, sources of uncertainty are multifaceted and layered. On the one hand, tumor heterogeneity across stages leads to substantial variability in treatment selection and prognostic estimation, translating medical complexity into cognitive ambiguity (13, 14). On the other hand, surgery, chemotherapy, targeted therapy, and other modalities entail adverse effects and uncertain efficacy, further intensifying psychological burden (15, 16). Prior studies have linked uncertainty in illness among cancer patients to anxiety, depression, and other negative affective states, and have shown that it can meaningfully influence adaptive coping and quality of life (17, 18). Nonetheless, most existing work treats uncertainty in illness as a unidimensional construct, overlooking the possibility of heterogeneous distribution patterns across patient subgroups—thereby constraining our understanding of psychological diversity in gastric cancer.

Eating self-efficacy denotes an individual’s belief and confidence in their capability to perform specific dietary behaviors (19). Surgery often results in reduced gastric capacity, impaired digestion, and malabsorption, requiring patients to re-establish adaptive dietary patterns (20). In this transition, eating self-efficacy influences whether patients can implement dietary regimens recommended by healthcare teams, thereby affecting nutritional status, weight management, and overall recovery (21, 22). High eating self-efficacy enables patients to maintain proactive coping when confronting dietary difficulties, seek nutritional support, and sustain healthy eating practices; conversely, low eating self-efficacy may engender helplessness and frustration, ultimately undermining needed behavior change (23). Notably, eating self-efficacy is not an isolated psychological construct but is closely intertwined with illness cognition, emotional state, and social support (24). Evidence suggests that cancer patients with lower self-efficacy often report higher uncertainty in illness (18, 25), implying a potential intrinsic linkage between these constructs. However, systematic empirical research examining uncertainty in illness and eating self-efficacy jointly remains limited.

According to stress-and-coping models (26, 27), uncertainty in illness may indirectly shape coping selection and enactment by influencing both primary and secondary appraisal processes (28). When patients experience high uncertainty while processing illness-related information, cognitive ambiguity may erode confidence in behavioral capability, thereby weakening eating self-efficacy (25). From an information-processing perspective (29), elevated uncertainty in illness may indicate difficulties in organizing and integrating health information; the resulting depletion of cognitive resources may spill over into dietary self-management, limiting patients’ ability to use nutritional guidance to build and strengthen eating self-efficacy. Accordingly, we posit that uncertainty in illness and eating self-efficacy may exhibit joint patterning within the gastric cancer population, rather than a simple linear association. Conventional variable-centered approaches are ill-suited to detect such latent categorical structures, underscoring the need for more refined analytic techniques to reveal distinct psychological profiles.

Latent profile analysis (LPA), a person-centered statistical method, classifies individuals into qualitatively distinct latent classes based on response patterns across observed variables (30, 31). Importantly, LPA is a model-based approach that assumes the population may contain unobserved subgroups, making it particularly suitable for detecting hidden heterogeneity in psychological characteristics. In the present study, this is especially relevant because uncertainty in illness and eating self-efficacy may co-occur in different ways across patients with gastric cancer, rather than following a single average pattern. Compared with traditional cluster analysis, LPA is grounded in stronger statistical theory and offers several methodological advantages. Specifically, LPA relies on probabilistic modeling rather than purely distance-based classification, provides objective model-fit indices to determine the optimal number of classes, and estimates posterior probabilities that reflect the uncertainty of class assignment (32). In addition, LPA allows for the inclusion of covariates to examine predictors of class membership, which is highly relevant to the present study given our interest in sociodemographic, clinical, lifestyle, and psychological correlates of latent profile membership. In recent years, LPA has been widely applied in psychology and health sciences to identify subgroups characterized by distinct constellations of psychological and behavioral features—for example, patterns of psychological adaptation, symptom clustering, and quality-of-life trajectories among patients with cancer (33–35).

Despite this progress, several important gaps remain in the current literature:

• Most existing studies concerning uncertainty in illness and self-efficacy in oncology populations have adopted variable-centered approaches, which emphasize average relationships between variables and may overlook meaningful within-population heterogeneity.

• Although uncertainty in illness and eating self-efficacy are both highly relevant to the adjustment and rehabilitation of patients with gastric cancer, little is known about whether these two constructs form distinct joint profiles at the individual level.

• The factors associated with such potential profiles remain unclear, particularly with respect to cognitive and perceptual processes such as health information avoidance and interoceptive sensitivity. This lack of person-centered evidence limits a more precise understanding of psychological heterogeneity in gastric cancer and constrains the development of targeted supportive interventions.

The present study aimed to use LPA to (1) identify latent classes reflecting joint profiles of uncertainty in illness and eating self-efficacy among patients with gastric cancer and (2) determine factors associated with class membership. Specifically, we considered sociodemographic variables, clinical baseline characteristics, lifestyle factors, and psychological variables including health information avoidance and interoceptive sensitivity as potential predictors. Health information avoidance, as a cognitive tendency, reflects motivational avoidance of health-threatening information and may influence uncertainty in illness and eating self-efficacy by restricting access to disease-related knowledge (36). Interoceptive sensitivity refers to individuals’ perception and awareness of internal bodily signals and is particularly relevant for patients with gastric cancer who must monitor bodily responses closely to adjust dietary behaviors (37, 38). Thus, our findings are expected to advance understanding of psychological heterogeneity in gastric cancer and provide evidence to support stratified, individualized psychological and nutritional care in clinical practice.

2 Methods

2.1 Study design and sample

This cross-sectional study was conducted in the outpatient clinics and inpatient units of the Departments of Medical Oncology and Gastrointestinal Surgery at two tertiary hospitals in Zhejiang Province, China. Participants were recruited by convenience sampling between 12 September and 3 December 2025. Data were collected using an electronic questionnaire administered through the Credamo survey platform.

Prior to recruitment, we coordinated with department heads to obtain support and to identify suitable data-collection periods that would minimize disruption to routine clinical care. During recruitment, trained research staff approached potentially eligible patients in outpatient waiting areas and hospital wards, briefly introduced the study background and aims, and invited participation. Written informed consent was obtained from all participants.

To enhance study rigor, prespecified inclusion and exclusion criteria were applied.

Inclusion criteria were: (1) primary gastric cancer confirmed by endoscopic biopsy and histopathology; (2) aged 18–85 years; (3) able to read and write Chinese sufficiently to understand and complete the questionnaire independently or with assistance from research staff; (4) aware of their gastric cancer diagnosis; (5) clear consciousness and able to communicate effectively; and (6) willingness to participate with provision of written informed consent.

Exclusion criteria were: (1) concurrent primary malignancy at another site or metastatic gastric cancer; (2) history of severe psychiatric disorders (e.g., schizophrenia, bipolar disorder, or major depressive disorder); (3) evident cognitive impairment precluding comprehension of the questionnaire; (4) end-stage disease, unstable vital signs, or current intensive care; (5) receipt of systematic psychotherapy within the past 6 months or concurrent participation in other psychological intervention studies; and (6) inability to complete the survey due to visual, hearing, or other physical impairments.

Latent profile analysis. Simulation work by Nylund-Gibson and Choi (39) indicates that when the sample size exceeds 300, model fit indices in LPA can more accurately recover the true number of classes. In addition, each latent class should include at least 25–50 individuals to ensure stable and reliable estimation of class-specific parameters. Multivariable logistic regression. The minimum required sample size was estimated using G^*Power 3.1 (40), with f² = 0.15, α = 0.05, 1 − β = 0.95, and 14 predictors, yielding a minimum sample size of 194 participants. Thus, we set the target sample size at no fewer than 400 valid respondents.

Across the two participating hospitals, 536 patients with gastric cancer who met preliminary screening criteria were approached. Of these, 62 were excluded for failing to meet inclusion criteria or meeting exclusion criteria, including 33 outside the age range, 18 with other primary tumors, 11 unaware of their diagnosis, 9 with a psychiatric history, 7 with cognitive impairment, and 4 currently participating in other studies. Among the remaining 474 eligible patients, 48 declined participation, primarily due to lack of interest in the questionnaire, concerns about privacy, or fatigue/physical discomfort. The final analytic sample comprised 426 participants, yielding an effective response rate of 79.47%.

2.2 Measures tools

Eating Self-Efficacy Scale. Eating self-efficacy was assessed using the scale developed by Glynn and Ruderman (41), which includes 25 items encompassing two domains: eating in response to negative emotions and eating in socially acceptable contexts. The scale has been applied in Chinese older adults to assess confidence in healthy eating and the ability to differentiate healthy from unhealthy dietary behaviors (42), demonstrating satisfactory cultural applicability and reliability. In this study, items were rated on a 5-point Likert scale ranging from 1 (“no difficulty controlling eating”) to 5 (“extreme difficulty controlling eating”). Total scores range from 25 to 125, with higher scores indicating stronger eating self-efficacy. In the present sample, Cronbach’s α was 0.984, indicating good internal consistency.

Mishel Uncertainty in Illness Scale. Uncertainty in illness was measured using the Mishel Uncertainty in Illness Scale (43), designed to assess illness-related uncertainty in adults. A Chinese version translated and validated by Ye, She (44) has demonstrated adequate cultural adaptation and reliability in Chinese patients with malignant tumors. The scale comprises 20 items across three dimensions: ambiguity (8 items), lack of clarity (7 items), and unpredictability (5 items). Responses are scored on a 5-point Likert scale (1 = “strongly disagree” to 5 = “strongly agree”). Total scores range from 20 to 100, with higher scores indicating greater uncertainty in illness. In this study, Cronbach’s α was 0.981, indicating good internal consistency.

Health Information Avoidance Scale. Health information avoidance was assessed using the Information Avoidance Scale developed by Howell and Shepperd (45), originally designed to measure the tendency to avoid learning information. The scale includes 8 items and is unidimensional. A Chinese version translated using back-translation by Sun and Wang (46) has shown good cultural applicability and reliability among Chinese older adults. Participants rated each item on a 5-point Likert scale (1 = “strongly disagree” to 5 = “strongly agree”). An example item is: “I would rather not know the dietary requirements for patients with gastric cancer.” Total scores range from 8 to 40, with higher scores indicating stronger health information avoidance. In the present study, Cronbach’s α was 0.968, indicating good internal consistency.

Interoceptive Sensitivity scale. Interoceptive sensitivity was measured using the Multidimensional Assessment of Interoceptive Awareness (MAIA) developed by Mehling, Price (47). The MAIA comprises 32 items across eight dimensions: Noticing, Not Distracting, Not Worrying, Attention Regulation, Emotional Awareness, Self-Regulation, Body Listening, and Trusting. A Chinese version translated by Lin, Hsu (48) has demonstrated cultural applicability in Chinese adults. Items were rated on a 5-point Likert scale (1 = “strongly disagree” to 5 = “strongly agree”). An example item is: “When I am tense I notice where the tension is located in my body.” Total scores range from 32 to 160, with higher scores indicating greater interoceptive sensitivity. In this study, Cronbach’s α was 0.987, indicating good internal consistency.

2.3 Statistical analysis

All analyses were conducted using SPSS 27.0 and Mplus 8.3. Reliability and construct validity were evaluated for each scale using Cronbach’s α. In descriptive analyses, continuous variables are presented as means and standard deviations, and categorical variables as frequencies and percentages. Skewness and kurtosis were examined to assess approximate normality. Pearson correlation analyses were performed to evaluate associations among continuous variables and to preliminarily characterize distributions and interrelationships.

Latent profile analysis models were specified using continuous indicator variables. In the primary analysis, all individual item scores from the Mishel Uncertainty in Illness Scale (20 items) and the Eating Self-Efficacy Scale (25 items) were used as indicators, yielding a total of 45 item-level continuous indicators. We applied an iterative model-comparison strategy by fitting models with 1 to 5 latent classes and determining the optimal number of classes based on a comprehensive evaluation of multiple statistical criteria.

Lower values of the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and sample-size adjusted BIC (aBIC) indicate better relative model fit while accounting for model complexity. The Lo–Mendell–Rubin adjusted likelihood ratio test (LMR) and bootstrap likelihood ratio test (BLRT) were used to compare a model with k classes to a model with k − 1 classes; a significant p value suggests that the model with one additional class provides a statistically better fit. Entropy was used only as an indicator of classification accuracy, with higher values reflecting clearer separation between profiles, but it was not treated as an index of overall model fit. Final model selection also considered class size and substantive interpretability.

To ensure estimation stability and avoid local maxima, models were estimated using 500 random starting values for the initial stage and 100 optimizations for the final stage; convergence was confirmed by verifying that the best log-likelihood value was replicated across multiple random starts.

Given that the use of 45 item-level indicators increases the number of freely estimated parameters relative to the sample size, we conducted a sensitivity analysis using subscale-level (dimension-level) scores as indicators. Specifically, three subscale scores from the Uncertainty in Illness Scale (ambiguity, lack of clarity, and unpredictability) and two subscale scores from the Eating Self-Efficacy Scale (negative affect eating and socially acceptable eating) were entered as five continuous indicators. The same model-comparison strategy was applied to evaluate whether the optimal class solution and profile characteristics remained consistent with those obtained from the item-level analysis. Methodological simulation research has demonstrated that when latent classes are well separated and class proportions are balanced, sample sizes between 200 and 500 generally provide adequate statistical power to recover two- to three-class solutions reliably, even with a moderate-to-large number of indicators (49, 50).

After selecting the optimal LPA model, independent-samples t tests or one-way ANOVA were used to compare latent profile scores across groups defined by sociodemographic and clinical characteristics.

Subsequently, latent class membership was exported as a categorical variable, and multivariable logistic regression was conducted to identify predictors of class membership. Latent class membership served as the dependent variable, and variables significant in univariable analyses were entered as independent variables. For outcomes with more than two classes, the clinically most relevant or largest class was selected as the reference group. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated for other classes relative to the reference.

2.4 Ethical considerations

The study protocol was approved by the Medical Ethics Committee of Taizhou First People’s Hospital. All procedures were conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants before data collection. Questionnaire data were collected anonymously or in de-identified form and were used solely for research purposes. To protect privacy, questionnaire data were collected anonymously or de-identified and were used only for research purposes.

3 Results

3.1 Sample characteristics

A total of 426 patients with gastric cancer were included in the analysis. Their sociodemographic, clinical, and lifestyle characteristics are presented in Table 1. Overall, the sample was predominantly male and married, and a considerable proportion reported a history of smoking and alcohol use.

3.2 Descriptive statistics and correlations analysis

Descriptive statistics and correlation coefficients for key variables are shown in Table 2. Mean scores of core variables were all close to the midpoint (M = 3.000). Skewness ranged from −0.013 to 0.354 and kurtosis from −0.621 to −0.451, within acceptable limits (|skewness| < 3; |kurtosis| < 8), indicating approximate normality. Correlation analyses showed that uncertainty in illness was strongly negatively correlated with eating self-efficacy. Uncertainty in illness was positively correlated with health information avoidance and interoceptive sensitivity. Eating self-efficacy was negatively correlated with health information avoidance and interoceptive sensitivity. Health information avoidance was positively correlated with interoceptive sensitivity.

3.3 Latent profile analysis of uncertainty in illness and eating self-efficacy

Using item-level indicators from the uncertainty in illness and eating self-efficacy measures, LPA models with one to five classes were estimated; fit indices are presented in Table 3. With increasing numbers of classes, AIC, BIC, and aBIC decreased, indicating improved fit. The LMR test indicated that the two-class model fit significantly better than the one-class model (p < 0.001), whereas the LMR tests for the three-, four-, and five-class models were not significant (p > 0.05). The BLRT was significant for all models (p < 0.001). Therefore, the two-class solution was selected as the optimal model.

Model estimation for the two-class solution converged successfully, and the best log-likelihood value was replicated across multiple sets of random starting values, confirming that the solution represented a global rather than a local maximum. Although this ratio is relatively modest, the exceptionally high entropy (0.984) and balanced class proportions (53.2% vs. 46.8%) indicate favorable conditions for stable parameter estimation, consistent with simulation evidence that well-separated classes with balanced proportions can be reliably recovered at this sample size (49).

To further verify the robustness of the class solution, a sensitivity analysis was conducted using five subscale-level indicators (three uncertainty in illness dimensions: ambiguity, lack of clarity, and unpredictability; and two eating self-efficacy dimensions: negative affect eating and socially acceptable eating). Consistent with the primary analysis, the two-class model was supported as the optimal solution. The two-class model showed high entropy (0.903), significant LMR and BLRT results (both p < 0.001), and relatively balanced class proportions (46.5 and 53.5%). Although AIC, BIC, and aBIC continued to decrease with additional classes, the LMR test was not significant for the three-class model (p = 0.141) or for models with more classes (p > 0.05), indicating that more complex class solutions did not provide a statistically superior fit. Critically, the profile characteristics were consistent with those obtained from the item-level analysis: one class was characterized by lower uncertainty in illness and higher eating self-efficacy, whereas the other exhibited the opposite pattern. Class proportions were also comparable. These convergent results across indicator levels support the stability and trustworthiness of the two-class solution.

Based on the selected solution, patients were categorized into two latent profiles with distinct psychological patterns. Class 1 included 227 participants (53.20%), characterized by lower uncertainty in illness and higher eating self-efficacy, and was labeled the “Low Uncertainty in Illness–High Eating Self-efficacy” group. Class 2 included 199 participants (46.80%), characterized by higher uncertainty in illness and lower eating self-efficacy, and was labeled the “High Uncertainty in Illness–Low Eating Self-efficacy” group. The profile patterns are illustrated in Figure 1, demonstrating a clear crossover pattern and supporting heterogeneous joint distributions of these two psychological constructs among gastric cancer patients.

Figure 1

3.4 Univariable comparisons between latent profiles

To examine differences across profiles, chi-square tests were used for categorical variables and independent-samples t tests for continuous variables (Table 4).

Significant differences were observed between profiles in gender (χ² = 8.395, p = 0.004), education background (χ² = 6.240, p = 0.044), employment status (χ² = 7.985, p = 0.046), monthly income (χ² = 7.988, p = 0.018) and place of residence (χ² = 5.354, p = 0.021). But marital status (χ² = 1.282, p = 0.733) and age (t = −0.808, p = 0.420) did not differ significantly. Clinically, TNM stage (χ² = 7.596, p = 0.022), smoking history (χ² = 5.784, p = 0.016) and alochol (χ² = 4.061, p = 0.044) differed significantly.

For psychological variables, the High Uncertainty in Illness–Low Eating Self-efficacy group reported significantly higher health information avoidance (t = −5.502, p < 0.001) and higher interoceptive sensitivity (t = −4.178, p < 0.001) than the Low Uncertainty in Illness–High Eating Self-efficacy group.

3.5 Multivariable logistic regression

Variables that were significant in univariable analyses were entered into a multivariable logistic regression model to identify independent predictors of profile membership. Using Class 2 (High Uncertainty in Illness–Low Eating Self-efficacy) as the reference category, the model estimated predictors of membership in Class 1 (Low Uncertainty in Illness–High Eating Self-efficacy). Results are shown in Table 5.

Health information avoidance was a significant predictor (B = −0.530, p < 0.001, OR = 0.589, 95% CI: 0.457–0.758). For each one-unit increase in health information avoidance, the odds of belonging to Class 1 decreased by 41.1%. Interoceptive sensitivity was also significant (B = −0.335, p = 0.015, OR = 0.715, 95% CI: 0.546–0.937), indicating a 28.5% reduction in the odds of belonging to Class 1 per one-unit increase.

Gender remained significant (B = −0.558, p = 0.011, OR = 0.572, 95% CI: 0.372–0.880). Using female as the reference, male had 42.8% lower odds of belonging to Class 1, suggesting male were more likely to belong to the High Uncertainty in illness–Low Eating Self-efficacy group. For education, none of the levels differed significantly from the reference category (bachelor’s degree or above) (all p > 0.050). For employment status, compared with employed participants, retirees had significantly lower odds of belonging to Class 1 (B = −0.574, p = 0.020, OR = 0.563, 95% CI: 0.347–0.914), whereas differences for unemployed participants and students were not significant.

Regarding monthly income level, compared with ≥ 8,001 RMB, participants earning ≤ 4,000 RMB had significantly lower odds of belonging to Class 1 (B = −0.686, p = 0.020, OR = 0.504, 95% CI: 0.282–0.899), while 4,001–8,000 RMB did not differ significantly from the reference. For residence, urban residents had significantly lower odds of belonging to Class 1 compared with rural residents (B = −0.445, p = 0.043, OR = 0.641, 95% CI: 0.417–0.985).

After controlling for covariates, TNM stage was not a significant independent predictor (stage I: p = 0.136; stage II: p = 0.210; reference = stage III). Smoking history showed a marginal association but did not reach statistical significance (B = −0.435, p = 0.062, OR = 0.647, 95% CI: 0.410–1.022). Alcohol use history was a significant predictor (B = 0.672, p = 0.015, OR = 1.958, 95% CI: 1.137–3.373); compared with those without alcohol use, those with alcohol use had 95.8% higher odds of belonging to Class 1.

4 Discussion

4.1 Latent profile model of uncertainty in illness and eating self-efficacy

Using LPA, we identified two qualitatively distinct subgroups of patients with gastric cancer characterized by different joint patterns of uncertainty in illness and eating self-efficacy: a low uncertainty in illness–high eating self-efficacy group and a high uncertainty in illness–low eating self-efficacy group. The two profiles were relatively balanced in size (53.20% vs. 46.80%), indicating that nearly half of patients experience the combined psychological burden of elevated uncertainty in illness and diminished eating self-efficacy. Notably, the two-profile structure remained stable in the subscale-level sensitivity analysis. Even when uncertainty in illness and eating self-efficacy were represented by five dimension-level indicators rather than aggregated overall measures, the optimal latent structure still consisted of two classes. This finding suggests that the identified profiles were not merely a by-product of using broad total scores. Instead, the heterogeneity observed in this sample appears to reflect a more global pattern of psychological adaptation versus maladaptation. In other words, the dimensions of illness uncertainty tended to cluster together, and the dimensions of eating self-efficacy also tended to move in parallel, rather than forming multiple cross-cutting subgroup configurations. This pattern may be particularly plausible in gastric cancer, where illness-related ambiguity and eating-related challenges are closely intertwined in everyday adjustment.

Within stress-and-coping framework, uncertainty in illness may indirectly shape coping choices and enactment through its influence on primary and secondary appraisal processes (28). When patients perceive substantial ambiguity regarding illness-related information, such cognitive uncertainty may erode confidence in one’s behavioral capacity (10), thereby undermining eating self-efficacy. The inverse configuration observed across the two profiles provides empirical support for this mechanism. Information-processing theory offers an additional explanatory lens: high uncertainty in illness may reflect difficulties in organizing and integrating illness-related information (51), and the consequent depletion of cognitive resources may spill over into dietary self-management, limiting patients’ ability to capitalize on nutritional guidance to build and consolidate eating self-efficacy.

Notably, we did not identify a transitional profile characterized by moderate uncertainty in illness and moderate eating self-efficacy, which may relate to the distinctive context of gastric cancer. As a life-threatening malignancy, gastric cancer diagnosis and treatment can generate profound psychological impact (52), potentially fostering more polarized response patterns rather than intermediate states. Our findings differ from some LPA studies in other cancer populations that have identified three or more adjustment profiles (53). Such discrepancies may stem from heterogeneity across cancer types in clinical characteristics, treatment modalities, and functional sequelae, underscoring the need to validate these patterns in broader oncology populations.

4.2 Determinants of optimal profile membership

Both Health Information Avoidance and Interoceptive sensitivity emerged as significant predictors of latent profile membership. Patients with stronger Health Information Avoidance were more likely to belong to the high uncertainty in illness–low eating self-efficacy profile, a finding with clear theoretical implications. Health Information Avoidance reflects motivational tendencies to avoid health-threatening information (54). When patients habitually avoid illness-related information, they are less able to acquire sufficient knowledge to understand disease trajectory, treatment options, and rehabilitation requirements; this information deficit can translate directly into heightened cognitive uncertainty. At the same time, information avoidance restricts access to nutritional guidance and dietary self-management knowledge (55), weakening the knowledge base needed to develop eating self-efficacy. This aligns with prior evidence suggesting that information avoidance plays a maladaptive role in psychological adjustment among patients with cancer (56, 57).

Higher Interoceptive sensitivity also predicted membership in the high uncertainty in illness–low eating self-efficacy profile, a result that may initially appear counterintuitive. Interoceptive sensitivity refers to an individual’s capacity to perceive and attend to internal bodily signals (58); in principle, greater interoceptive sensitivity could facilitate monitoring of bodily responses and adaptive dietary adjustments (59). Postoperative patients commonly experience symptoms such as dumping syndrome, reflux, and abdominal distension (60); those with high Interoceptive sensitivity may over-attend to and amplify such sensations, perpetuating worry and uncertainty about illness status.

Male patients were more likely to belong to the high uncertainty in illness–low eating self-efficacy profile. Men may be more inclined to adopt suppressive or avoidant coping styles and may be less likely to seek emotional support or health information proactively (61, 62), contributing to higher uncertainty in illness. In addition, within traditional Chinese cultural contexts, meal preparation and dietary management are often socially ascribed to women; consequently, men may face greater challenges in postoperative dietary adjustment due to limited prior knowledge and skill acquisition, manifesting as lower eating self-efficacy.

Employment status also warranted attention. Compared with employed patients, retirees were more likely to belong to the high uncertainty in illness–low eating self-efficacy profile. This association may reflect characteristics of retirement populations, including older age, greater comorbidity burden, and functional decline, all of which increase the complexity of disease management (63). Retirement may also entail reduced workplace-based social networks, limiting access to health information and social support (64). By contrast, employed individuals may possess more social resources, stronger information-seeking capacity, and a more active life orientation, which may help sustain lower uncertainty in illness and higher eating self-efficacy.

Monthly personal income represented a key socioeconomic determinant of profile membership. Lower-income patients were more likely to belong to the high uncertainty in illness–low eating self-efficacy profile, highlighting the role of socioeconomic status in cancer adjustment. Limited financial resources may constrain access to high-quality healthcare services, nutritional supplements, and professional dietary counseling (65), while heightening concerns regarding treatment costs and quality of life, thereby increasing uncertainty in illness. Additionally, low-income patients may face constrained food choices, making it more difficult to purchase and prepare recommended foods and further weakening eating self-efficacy.

Residence type showed an intriguing pattern: urban residents were more likely than rural residents to belong to the high uncertainty in illness–low eating self-efficacy profile, contrary to some expectations. Urban residency is often assumed to confer better access to medical resources and information channels and thus better psychological adjustment (66). Our findings may instead reflect potential adverse features of urban lifestyles. Urban residents may experience faster-paced living, greater occupational stress, and more complex social demands, increasing psychological burden. Moreover, although urban residents may be exposed to larger volumes of health information, the complexity and inconsistency of such information may paradoxically exacerbate uncertainty in illness. Rural residents, despite relatively fewer medical resources, may benefit from tighter family support networks and simpler daily routines that facilitate adjustment.

The predictive role of alcohol use history also merits careful interpretation. Patients with an alcohol use history were more likely to belong to the low uncertainty in illness–high eating self-efficacy profile. This association may reflect differences in personality traits or coping styles, or residual confounding by socioeconomic status or social support. Importantly, these results should not be construed as endorsing alcohol consumption; alcohol use is a recognized risk factor for gastric cancer and may adversely affect postoperative recovery. Smoking showed a near-significant association with profile membership, suggesting that the relationship between smoking and adjustment typologies may be complex and warrants further investigation.

4.3 Practical and clinical implications

The pronounced heterogeneity in uncertainty in illness and eating self-efficacy among patients with gastric cancer indicates that clinicians should move beyond “one-size-fits-all” approaches toward stratified and personalized supportive care. Patients classified into the high uncertainty in illness–low eating self-efficacy profile should be prioritized for more intensive and comprehensive psychological care and nutritional support.

To address elevated uncertainty in illness, clinical teams should strengthen informational support by proactively assessing patients’ information needs and providing clear, comprehensible explanations regarding staging, treatment options, prognosis, and recovery trajectories, thereby helping patients develop a coherent cognitive framework for understanding their illness. Given that Health Information Avoidance is a key predictor of uncertainty in illness, clinicians should assess avoidance tendencies and adopt gradual, non-threatening communication strategies to facilitate engagement with essential health information.

To improve low eating self-efficacy, nutrition teams should deliver structured dietary counseling and skills training, including individualized meal plans, demonstrations of food preparation techniques, recipe guidance, and stepwise goal setting. In practice, peer-support groups can be established in which well-recovered gastric cancer survivors share dietary self-management experiences to provide vicarious learning opportunities. Healthcare professionals should also provide positive feedback and encouragement to reinforce confidence in dietary behavior change.

At admission or during follow-up, brief assessments could flag patients at elevated risk—particularly men, retirees, low-income patients, urban residents, and those with strong Health Information Avoidance—and enable early provision of preventive psychological support and dietary counseling. For patients with high Interoceptive sensitivity, clinicians may consider mind–body strategies such as mindfulness training to cultivate nonjudgmental awareness of bodily sensations and reduce excessive symptom monitoring and catastrophizing.

4.4 Limitations and future directions

This study employed a cross-sectional design, which can delineate associations but does not support causal inference. Future work should adopt longitudinal designs with repeated measurements to characterize trajectories over time and clarify reciprocal mechanisms.

Participants were recruited via convenience sampling from two tertiary hospitals in Zhejiang Province, potentially limiting representativeness and generalizability. Future studies should broaden recruitment across regions and healthcare settings to enhance external validity. Additionally, all variables were assessed via self-report questionnaires, raising concerns about common method and social desirability biases. Future research should incorporate multi-source assessments (e.g., clinician-rated evaluations, caregiver reports, objective physiological indicators). For eating self-efficacy, integrating dietary diaries and objective nutritional status indicators would help validate self-reported measures.

Although the predictors included demographic, clinical, lifestyle, and psychological domains, other influential factors may have been omitted, such as social support, coping styles, quality of patient–clinician communication, and prior mental health status. Future studies could extend the predictor set to build more comprehensive models. Moreover, we did not stratify patients by treatment phase (e.g., preoperative, early postoperative, recovery period), despite likely phase-specific differences in psychological status; targeted phase-based investigations are warranted.

From a methodological standpoint, the primary LPA employed 45 item-level indicators, which resulted in a relatively large number of freely estimated parameters relative to the sample size. Although the sample-to-parameter ratio for the two-class model was modest, several factors mitigate concerns about estimation reliability, including the high entropy value, balanced class proportions, successful replication of the best log-likelihood across multiple random starts, and—most critically—the convergent results obtained from a sensitivity analysis using five subscale-level indicators with a substantially more favorable sample-to-parameter ratio. Nonetheless, future studies with larger samples could further strengthen confidence in item-level LPA solutions and potentially support the identification of more fine-grained latent profiles. Additionally, alternative approaches such as factor mixture modeling or the use of parceled indicators may offer complementary strategies for balancing model complexity with sample size constraints.

Finally, while the two-profile solution exhibited high classification precision, it did not further differentiate more granular subgroups. This may reflect the sample size or the selection of indicators. Larger samples and inclusion of additional adaptation-relevant variables (e.g., anxiety, depression, quality of life) may reveal more nuanced latent structures and enable a more comprehensive typology of psychological characteristics in gastric cancer.

5 Conclusion

This study identified two latent profiles of uncertainty in illness and eating self-efficacy among patients with gastric cancer: a low uncertainty in illness–high eating self-efficacy profile and a high uncertainty in illness–low eating self-efficacy profile. Health information avoidance, interoceptive sensitivity, sex, employment status, income, residence, and alcohol use history were associated with profile membership. These findings highlight psychological heterogeneity in gastric cancer and may help inform more individualized supportive care.

References

• 1.

  WongMCSHuangJChanPSFChoiPLaoXQChanSMet al. Global incidence and mortality of gastric Cancer, 1980-2018. JAMA Netw Open. (2021) 4:e2118457. doi: 10.1001/jamanetworkopen.2021.18457,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2.

  MorganEArnoldMCamargoMCGiniAKunzmannATMatsudaTet al. The current and future incidence and mortality of gastric cancer in 185 countries, 2020-40: a population-based modelling study. EClinicalMedicine. (2022) 47:101404. doi: 10.1016/j.eclinm.2022.101404,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3.

  LordickFSiewertJR. Recent advances in multimodal treatment for gastric cancer: a review. Gastric Cancer. (2005) 8:78–85. doi: 10.1007/s10120-005-0321-z,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4.

  GuanWLHeYXuRH. Gastric cancer treatment: recent progress and future perspectives. J Hematol Oncol. (2023) 16:57. doi: 10.1186/s13045-023-01451-3,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5.

  GhiglieriCDempsterMWrightSGraham-WisenerL. Psychosocial functioning in individuals with advanced oesophago-gastric cancer: a mixed methods systematic review. BMC Palliat Care. (2023) 22:164. doi: 10.1186/s12904-023-01288-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6.

  BurzCPopVSilaghiCLupanISamascaG. Prognosis and treatment of gastric Cancer: a 2024 update. Cancers (Basel). (2024) 16:1708. doi: 10.3390/cancers16091708,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7.

  GongYQLuCJXiaoYRZhangJYXuZLiJet al. Epidemiological trends of early-onset gastrointestinal cancers from 1990 to 2021 and predictions for 2036: analysis from the global burden of disease study 2021. Ann Med. (2025) 57:2555518. doi: 10.1080/07853890.2025.2555518,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8.

  ZhuHYangLYinHYuanXGuJYangY. The influencing factors of psychosocial adaptation of Cancer patients: a systematic review and Meta-analysis. Health Serv Insights. (2024) 17:11786329241278814. doi: 10.1177/11786329241278814,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9.

  JiangHDongYZongWZhangXJXuHJinF. The relationship among psychosocial adaptation, medication adherence and quality of life in breast cancer women with adjuvant endocrine therapy. BMC Womens Health. (2022) 22:135. doi: 10.1186/s12905-022-01722-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10.

  Johnson WrightLAfariNZautraA. The illness uncertainty concept: a review. Curr Pain Headache Rep. (2009) 13:133–8. doi: 10.1007/s11916-009-0023-z,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11.

  McCormickKM. A concept analysis of uncertainty in illness. J Nurs Scholarsh. (2002) 34:127–31. doi: 10.1111/j.1547-5069.2002.00127.x,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12.

  EpsteinRM. Facing epistemic and complex uncertainty in serious illness: the role of mindfulness and shared mind. Patient Educ Couns. (2021) 104:2635–42. doi: 10.1016/j.pec.2021.07.030,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13.

  KhalighiSReddyKMidyaAPandavKBMadabhushiAAbedalthagafiM. Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment. NPJ Precis Oncol. (2024) 8:80. doi: 10.1038/s41698-024-00575-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14.

  De PalmaMHanahanD. The biology of personalized cancer medicine: facing individual complexities underlying hallmark capabilities. Mol Oncol. (2012) 6:111–27. doi: 10.1016/j.molonc.2012.01.011,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15.

  KudererNMDesaiALustbergMBLymanGH. Mitigating acute chemotherapy-associated adverse events in patients with cancer. Nat Rev Clin Oncol. (2022) 19:681–97. doi: 10.1038/s41571-022-00685-3,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16.

  MaoJJPillaiGGAndradeCJLigibelJABasuPCohenLet al. Integrative oncology: addressing the global challenges of cancer prevention and treatment. CA Cancer J Clin. (2022) 72:144–64. doi: 10.3322/caac.21706,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17.

  WangTSunJGuDShenSZhouYWangZ. Dyadic effects of social support, illness uncertainty on anxiety and depression among lung cancer patients and their caregivers: a cross-sectional study. Support Care Cancer. (2023) 31:402. doi: 10.1007/s00520-023-07876-3,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18.

  WangHZhangH. The impact of illness uncertainty on self-efficacy, personal control, and negative emotions in elderly persons with colon cancer. Geriatr Nurs. (2025) 65:103480. doi: 10.1016/j.gerinurse.2025.103480,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19.

  GbenroMOJrMartinganoAJPerskyS. Exploring the impact of genetic beliefs about specific eating behaviors on dietary self-efficacy. J Behav Med. (2022) 45:497–502. doi: 10.1007/s10865-022-00290-w,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20.

  Utrilla FornalsACostas-BatlleCMedlinSMenjón-LajusticiaECisneros-GonzálezJSaura-CarmonaPet al. Metabolic and nutritional issues after lower digestive tract surgery: the important role of the dietitian in a multidisciplinary setting. Nutrients. (2024) 16:246. doi: 10.3390/nu16020246,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21.

  ZhuDQNormanIJWhileAE. Nurses' self-efficacy and practices relating to weight management of adult patients: a path analysis. Int J Behav Nutr Phys Act. (2013) 10:131. doi: 10.1186/1479-5868-10-131,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22.

  TangHWangRLiuWXiaoHJingHSongFet al. The influence of nutrition literacy, self-care self-efficacy and social support on the dietary practices of breast cancer patients undergoing chemotherapy: a multicentre study. Eur J Oncol Nurs. (2023) 64:102344. doi: 10.1016/j.ejon.2023.102344,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23.

  AynehchiASaleh-GhadimiSDehghanP. The association of self-efficacy and coping strategies with body mass index is mediated by eating behaviors and dietary intake among young females: a structural-equation modeling approach. PLoS One. (2023) 18:e0279364. doi: 10.1371/journal.pone.0279364,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24.

  VerstuyfJPatrickHVansteenkisteMTeixeiraPJ. Motivational dynamics of eating regulation: a self-determination theory perspective. Int J Behav Nutr Phys Act. (2012) 9:21. doi: 10.1186/1479-5868-9-21,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25.

  EyniSMousaviSE. Intolerance of uncertainty, cognitive fusion, coping self-efficacy and self-perceived burden in patients diagnosed with cancer. Psychooncology. (2023) 32:800–9. doi: 10.1002/pon.6125,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26.

  WheatonB. Stress, personal coping resources, and psychiatric symptoms: an investigation of interactive models. J Health Soc Behav. (1983) 24:208–29.

  • Pubmed Abstract
  • Google Scholar

• 27.

  HillsonJMKuiperNA. A stress and coping model of child maltreatment. Clin Psychol Rev. (1994) 14:261–85.

  • Google Scholar

• 28.

  WonghongkulTMooreSMMusilCSchneiderSDeimlingG. The influence of uncertainty in illness, stress appraisal, and hope on coping in survivors of breast cancer. Cancer Nurs. (2000) 23:422–9. doi: 10.1097/00002820-200012000-00004. PubMed 11128121,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29.

  TurnerKLMakhijaMV. The role of individuals in the information processing perspective. Strateg Manag J. (2012) 33:661–80. doi: 10.1002/smj.1970

  • CrossRef
  • Google Scholar

• 30.

  YangQZhaoALeeCWangXVorderstrasseAWoleverRQ. Latent profile/class analysis identifying differentiated intervention effects. Nurs Res. (2022) 71:394–403. doi: 10.1097/nnr.0000000000000597,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31.

  GabrielASCampbellJTDjurdjevicEJohnsonRERosenCC. Fuzzy profiles: comparing and contrasting latent profile analysis and fuzzy set qualitative comparative analysis for person-centered research. Organ Res Methods. (2018) 21:877–904. doi: 10.1177/1094428117752466

  • CrossRef
  • Google Scholar

• 32.

  MorganGB. Mixed mode latent class analysis: an examination of fit index performance for classification. Struct Equ Model Multidiscip J. (2015) 22:76–86. doi: 10.1080/10705511.2014.935751

  • CrossRef
  • Google Scholar

• 33.

  YangDHuCZhouZHeLHuangSWanMet al. The impact of perceived stigma on appearance anxiety in postoperative rhinoplasty patients: a variable-centered and person-centered perspective. Acta Psychol. (2025) 260:105660. doi: 10.1016/j.actpsy.2025.105660,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34.

  ParkJHChunMBaeSHWooJChonEKimHJ. Latent profile analysis for assessing symptom clusters in women with breast cancer. J Cancer Surviv. (2026) 20:218–25. doi: 10.1007/s11764-024-01648-6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35.

  ChengWSunSFangJZhangNZhangYZhangBet al. Identifying psychosocial distress phenotypes in postoperative patients with breast cancer: a latent profile analysis. Asia Pac J Oncol Nurs. (2025) 12:100792. doi: 10.1016/j.apjon.2025.100792,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36.

  ChenMHuangXWuYSongSQiX. A model for predicting factors affecting health information avoidance on WeChat. Digit Health. (2025) 11:20552076251314277. doi: 10.1177/20552076251314277,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37.

  TiemannAOrtmannJRuboMMeyerAHMunschSVögeleCet al. The relevance of cardiac and gastric interoception for disordered eating behavior. J Eat Disord. (2025) 13:114. doi: 10.1186/s40337-025-01284-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38.

  FarrisSGDerbyLKibbeyMM. Getting comfortable with physical discomfort: a scoping review of interoceptive exposure in physical and mental health conditions. Psychol Bull. (2025) 151:131–91. doi: 10.1037/bul0000464,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39.

  Nylund-GibsonKChoiAY. Ten frequently asked questions about latent class analysis. Transl Issues Psychol Sci. (2018) 4:440. doi: 10.1037/tps0000176

  • CrossRef
  • Google Scholar

• 40.

  FaulFErdfelderELangA-GBuchnerA. G* power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. (2007) 39:175–91. doi: 10.3758/BF03193146,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41.

  GlynnSMRudermanAJ. The development and validation of an eating self-efficacy scale. Cogn Ther Res. (1986) 10:403–20. doi: 10.1007/bf01173294

  • CrossRef
  • Google Scholar

• 42.

  JiaoW. Older adults' exposure to food media induced unhealthy eating during the COVID-19 omicron lockdown? Exploring negative emotions and associated literacy and efficacy on Shanghainese. Foods. (2024) 13:1797. doi: 10.3390/foods13121797,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43.

  MishelMH. The measurement of uncertainty in illness. Nurs Res. (1981) 30:258–63.

  • Pubmed Abstract
  • Google Scholar

• 44.

  YeZSheYLiangMKnobfTDixonJHuQet al. Revised Chinese version of Mishel uncertainty in illness scale: development, reliability and validity. Chin Gen Pract. (2018) 21:1091–7.

  • Google Scholar

• 45.

  HowellJLShepperdJA. Establishing an information avoidance scale. Psychol Assess. (2016) 28:1695–708. doi: 10.1037/pas0000315,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46.

  SunLWangY. Mapping health information avoidance among older adults in China: informational, social and psychological dynamics in the social media landscape. Health Commun. (2026):1–19. doi: 10.1080/10410236.2026.2616279,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47.

  MehlingWEPriceCDaubenmierJJAcreeMBartmessEStewartA. The multidimensional assessment of interoceptive awareness (MAIA). PLoS One. (2012) 7:e48230. doi: 10.1371/journal.pone.0048230,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48.

  LinFLHsuCCMehlingWYehML. Translation and psychometric testing of the Chinese version of the multidimensional assessment of interoceptive awareness. J Nurs Res. (2017) 25:76–84. doi: 10.1097/jnr.0000000000000182,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49.

  TeinJ-YCoxeSChamH. Statistical power to detect the correct number of classes in latent profile analysis. Struct Equ Model Multidiscip J. (2013) 20:640–57. doi: 10.1080/10705511.2013.824781,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50.

  WurptsICGeiserC. Is adding more indicators to a latent class analysis beneficial or detrimental? Results of a Monte-Carlo study. Front Psychol. (2014) 5:920. doi: 10.3389/fpsyg.2014.00920,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51.

  KuangK. Reconceptualizing uncertainty in illness: commonalities, variations, and the multidimensional nature of uncertainty. Ann Int Commun Assoc. (2018) 42:181–206. doi: 10.1080/23808985.2018.1492354

  • CrossRef
  • Google Scholar

• 52.

  SchütteKSchulzCMiddelberg-BispingK. Impact of gastric cancer treatment on quality of life of patients. Best Pract Res Clin Gastroenterol. (2021) 50:101727. doi: 10.1016/j.bpg.2021.101727,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53.

  YangDSheTGuiGLiLZhouZLiuLet al. Psychological capital and death anxiety in pancreatic cancer patients: a latent profile analysis. Front Psych. (2025) 16:1627422. doi: 10.3389/fpsyt.2025.1627422,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54.

  ChasiotisAWedderhoffORosmanTMayerA-K. The role of approach and avoidance motivation and emotion regulation in coping via health information seeking. Curr Psychol. (2021) 40:5235–44. doi: 10.1007/s12144-019-00488-3

  • CrossRef
  • Google Scholar

• 55.

  ZhangLLiHHuangTYangMYuXLiuY. Nutritional self-management in chronic diseases: a conceptual analysis. Front Public Health. (2025) 13:1680903. doi: 10.3389/fpubh.2025.1680903,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 56.

  ZhuRZhaoHYunYZhaoYWangWWangLet al. Research on health information avoidance behavior and influencing factors of cancer patients-an empirical analysis based on structural equation modeling. BMC Public Health. (2024) 24:3617. doi: 10.1186/s12889-024-21113-4,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57.

  ZerbinatiLFolesaniFCarusoRBelvederi MurriMNanniMGRighettiSet al. Maladaptive coping styles moderate the relationship between information on cancer treatment and psychosocial symptoms: an Italian multicenter study. Front Psychol. (2024) 15:1338193. doi: 10.3389/fpsyg.2024.1338193,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58.

  Solano DuránPMoralesJPHuepeD. Interoceptive awareness in a clinical setting: the need to bring interoceptive perspectives into clinical evaluation. Front Psychol. (2024) 15:1244701. doi: 10.3389/fpsyg.2024.1244701,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59.

  LoucksEBKronishIMSaadehFBScarpaciMMProulxJAGutmanRet al. Adapted mindfulness training for Interoception and adherence to the DASH diet: a phase 2 randomized clinical trial. JAMA Netw Open. (2023) 6:e2339243. doi: 10.1001/jamanetworkopen.2023.39243,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 60.

  ChenHJingSLiZCaoLGuanWChenXet al. Impact of distal or pylorus preserving gastrectomy on postoperative quality of life in T1 stage middle third gastric cancer patients. Sci Rep. (2025) 15:8632. doi: 10.1038/s41598-025-90866-8,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61.

  LysovaADimEE. I thought about killing myself, but a part of me insisted on getting help: coping experiences of male survivors of intimate partner violence. J Fam Violence. (2025):1–12. doi: 10.1007/s10896-025-00847-8

  • CrossRef
  • Google Scholar

• 62.

  AhmadSJafreeSR. Influence of gender identity on the adoption of religious-spiritual, preventive and emotion-focused coping strategies during the COVID-19 pandemic in Pakistan. Ann Med. (2023) 55:2291464. doi: 10.1080/07853890.2023.2291464,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 63.

  FabbriEZoliMGonzalez-FreireMSaliveMEStudenskiSAFerrucciL. Aging and multimorbidity: new tasks, priorities, and frontiers for integrated gerontological and clinical research. J Am Med Dir Assoc. (2015) 16:640–7. doi: 10.1016/j.jamda.2015.03.013

  • CrossRef
  • Google Scholar

• 64.

  LaraJO’BrienNGodfreyAHeavenBEvansEHLloydSet al. Pilot randomised controlled trial of a web-based intervention to promote healthy eating, physical activity and meaningful social connections compared with usual care control in people of retirement age recruited from workplaces. PLoS One. (2016) 11:e0159703. doi: 10.1371/journal.pone.0159703,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 65.

  VaiciurgisVTClancyAKCharltonKEStefoska-NeedhamABeckEJ. Food provision to support improved nutrition and well-being of people experiencing disadvantage - perspectives of service providers. Public Health Nutr. (2024) 27:e36. doi: 10.1017/s1368980024000132,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66.

  ChenXOromHHayJLWatersEASchofieldELiYet al. Differences in rural and urban health information access and use. J Rural Health. (2019) 35:405–17. doi: 10.1111/jrh.12335,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar