Association Between eHealth Literacy and Online Health Information Seeking Behavior Among Cancer Patients: A Chain Mediating Role of Self‐Efficacy and Information Utility

ABSTRACT

Background

Searching for health information online helps cancer patients better understand their health conditions and the way to cope with the adverse effects of treatment. Although eHealth literacy, self‐efficacy and information utility are recognized as key factors that are associated with online health information seeking behavior (OHISB), the underlying mechanisms linking these factors in cancer patients remain understudied. This study aims to examine the mediating roles of self‐efficacy and information utility in the relationship between eHealth literacy and OHISB.

Methods

In this cross‐sectional study, 502 cancer patients were selected from a tertiary hospital in Shanghai from June to December 2024. Demographic and clinical characteristics were obtained from electronic health records. eHealth literacy, self‐efficacy, information utility and OHISB were assessed using self‐report questionnaires. Mediation analysis was performed using structural equation modeling, with indirect effects evaluated through bootstrapping.

Results

The prevalence of OHISB among cancer patients was 61.4%. There was a significant correlation between eHealth literacy, self‐efficacy, information utility and OHISB (r = 0.357, 0.499, 0.469, p < 0.01). eHealth literacy demonstrated both direct (β = 0.190, standard error [SE] = 0.050, 95% confidence interval [CI] [0.099, 0.295]) and indirect (β = 0.311, SE = 0.045, 95% CI [0.227, 0.404]) impacts on OHISB, mediated by self‐efficacy (β = 0.223, SE = 0.038, 95% CI [0.155, 0.304]) and information utility (β = 0.056, SE = 0.025, 95% CI [0.013, 0.110]). The findings indicated that self‐efficacy and information utility served as a sequential mediating factor in the association between eHealth literacy and OHISB, contributing 6.6% to the overall effect (β = 0.033, SE = 0.012, 95% CI [0.010, 0.057]).

Conclusions

eHealth literacy enhances the OHISB of cancer patients through both direct effects and a chain mediation pathway involving self‐efficacy and information utility. Targeted interventions should focus on improving eHealth literacy, strengthening self‐efficacy, and developing reliable information platforms to optimize patient‐centered health information services.

This study reveals that eHealth literacy enhances online health information seeking behavior among cancer patients both directly and indirectly through the chain mediation of self‐efficacy and information utility, highlighting its critical role in empowering cancer patients to effectively use online health information.

Abbreviations

AGFI

Adjusted Goodness‐of‐Fit Index

AI

artificial intelligence

CFI

Comparative Fit Index

CI

confidence interval

CMIS

Comprehensive Model of Information Seeking

eHL

eHealth literacy

GFI

Goodness‐of‐Fit Index

GSES

General Self‐Efficacy Scale

HINTS

Health Information National Trends Survey

IFI

Incremental Fit Index

IQR

interquartile range

KAB

knowledge, attitude, and behavior

OHISB

online health information seeking behavior

RMSEA

root mean square error of approximation

SE

standard error

TLI

Tucker–Lewis Index

1. Background

In 2025, global internet users have grown to 5.56 billion, with 1.108 billion being Chinese users [1, 2]. The digital revolution driven by the internet has profoundly transformed global patterns of health information seeking. According to Eurostat, approximately 52% of individuals searched for health‐related information online in 2022, with even higher proportions reported in the United States and China [3, 4, 5]. Online health information seeking behavior (OHISB) is defined as the active process of retrieving, browsing, evaluating, and selecting health‐related information through the Internet [6], enhancing the convenience and autonomy of information access [7]. This behavioral shift is prominent among cancer patients, who increasingly use the internet to obtain information on treatments, side effects, and psychological supports to meet their diverse needs [8, 9, 10].

Cancer remains a critical global health challenge, with an estimated 35 million new cases worldwide by 2050. In light of the substantial burden of cancer and the growing transitional care needs, higher levels of OHISB have been observed among cancer patients than among those with other chronic diseases [11, 12, 13]. Such heightened OHISB enhances their knowledge of the disease, facilitates decision‐making, improves treatment adherence, strengthens self‐management capabilities, and promotes the utilization of healthcare services [5, 14, 15, 16]. However, cancer patients still encounter significant obstacles to engaging in OHISB, such as limited eHealth literacy, persistent psychological distress, and considerable financial constraints [6, 11, 17, 18], which emphasizes the importance of identifying modifiable factors that can facilitate OHISB among cancer patients.

The Comprehensive Model of Information Seeking (CMIS), developed by J. David Johnson, provides a theoretical framework for understanding the determinants of health information seeking behavior [19]. It posits that antecedent factors (e.g., demographics, beliefs, and direct experience) influence information carrier factors (like information utility and characteristics), which subsequently determine information seeking behavior. In the context of digital health, eHealth literacy can be conceptualized as direct experience within the CMIS framework, as it reflects an individual's ability to find, comprehend, and apply online health information [20, 21]. Recent studies have shown that eHealth literacy is positively associated with OHISB. Mitsutake et al. revealed a significant association between eHealth literacy and online COVID‐19 information seeking among Japanese internet users [12]. A study conducted in Sweden found that people with lower eHealth literacy were less willing to utilize the national online health information portal [22]. A systematic review also reported that eHealth literacy was associated with OHISB among cancer patients [23]. While the association between eHealth literacy and OHISB has been widely recognized, mechanisms that mediate this relationship remain unclear.

Self‐efficacy refers to an individual's perception regarding their ability to perform the actions required to attain certain results [24], which is a core determinant of health‐related behaviors. Previous studies have identified the association between self‐efficacy and OHISB in cancer patients [17, 25]. Those with greater self‐efficacy prefer online health information, as they believe they can benefit from it and obtain valuable insights [26]. The Knowledge, Attitude, and Behavior (KAB) model suggests that knowledge influences behavior by changing one's beliefs [27]. According to the theory, eHealth literacy, representing digital knowledge and skills, may have both a direct effect on health behaviors and an indirect effect through self‐efficacy. A recent study demonstrated that eHealth literacy influenced self‐care behaviors through the mediating role of self‐efficacy among diabetic patients [28]. Individuals with higher eHealth literacy are proficient in using online platforms and digital tools, thereby enhancing their self‐efficacy in changing health‐related behaviors [29]. Based on the above discussion, this study proposes that self‐efficacy mediates the relationship between eHealth literacy and OHISB among cancer patients.

Information utility is defined as the information delivered by the medium in direct response to the specific requirements of a person [19]. It refers to the caliber of the informational content and its credibility and usability [30]. As a core component of CMIS, it influences behaviors related to seeking health information [31]. Xu et al. reported that lower levels of OHISB were correlated with reduced perceptions of information utility among patients with periodontitis [32]. Ma et al. found that information utility was associated with health information‐seeking behavior among cancer patients [33]. According to the CMIS, information utility is influenced by antecedent factors, including eHealth literacy. Higher eHealth literacy manifests as more discerning judgments regarding online sources. For instance, Sun et al. indicated that individuals' prior knowledge, such as eHealth literacy, improved their judgment of the credibility of online information [34]. Alhewiti et al. pointed out that individuals with higher eHealth literacy trust professional health websites rather than social media [35]. Therefore, this study hypothesizes that information utility mediates the association between eHealth literacy and OHISB.

In addition, the CMIS also emphasizes that information utility is influenced not only by eHealth literacy but also by self‐efficacy. Those with higher self‐efficacy are more confident in identifying, comprehending, and applying information, which enhances their perception of its value [24]. As an antecedent factor, self‐efficacy motivates individuals to seek high‐quality information, thereby improving information utility [19]. Empirical studies have substantiated that individuals with higher levels of self‐efficacy are more capable of recognizing information utility and engaging in health information seeking [32, 36]. Thus, this study posits that information utility also mediates the association between self‐efficacy and OHISB.

The unprecedented evolution of the internet and the limitations of traditional healthcare channels in providing timely health information underscore the critical need to investigate the underlying mechanisms influencing OHISB. As cancer patients increasingly rely on digital platforms to access treatment updates, side‐effect management strategies, and psychosocial support, understanding how eHealth literacy facilitates OHISB through self‐efficacy and information utility holds significant implications for optimizing healthcare services utilization. Grounded in the CMIS and KAB theoretical framework and intricate relationship among eHealth literacy, self‐efficacy, information utility, and OHISB (Figure 1), this study hypothesizes:

Figure 1.

The hypothetical sequential mediation model of self‐efficacy and information utility on eHealth literacy and online health information seeking behavior (OHISB).

eHealth literacy directly influences OHISB.

Self‐efficacy mediates the effect of eHealth literacy on OHISB.

Information utility mediates the effect of eHealth literacy on OHISB.

Self‐efficacy and information utility are sequential mediators between eHealth literacy and OHISB, with self‐efficacy positively influencing information utility.

2. Methods

2.1. Study Design and Population

This cross‐sectional study was conducted in accordance with STROBE guidelines. Between June and December 2024, a convenience sample consisting of cancer patients from the oncology center of a tertiary hospital in Shanghai was recruited. The recruitment strategies included displaying research posters in both the oncology outpatient clinic and inpatient ward, and providing a brief study description to patients by physicians or research nurses at the time of registration. The oncology center of this hospital has been designated as a key discipline in Shanghai and has consistently ranked at the top among comprehensive hospitals across the city. It manages approximately 68,000 outpatient visits and 13,000 inpatient discharges annually, with patients coming from all parts of the country. The inclusion criteria were: (1) age ≥ 18 years; (2) diagnosis of a malignant tumor; and (3) current or prior receipt of anticancer treatment. The exclusion criteria were: (1) any impairments in vision, hearing, or communication; or (2) terminal‐stage cancer with a life expectancy < 3 months.

The proposed model includes three latent variables, corresponding to eight, three, and three observed indicators, with 77 degrees of freedom in the model [37]. The effect size was set to Root mean square error of approximation (RMSEA) = 0.08, with a significance level of α = 0.05 and statistical power of 1 − β = 0.80. Based on this, the minimum sample size was 341 participants. Taking a 20% attrition rate into account, the minimum required sample size is 427 participants.

2.2. Ethical Consideration

This research was conducted according to the Declaration of Helsinki and obtained permission from the Ethics Committee of the participating hospital (RA‐2022‐491). All participants provided written informed consent after receiving complete information about the study's objective, significance, methods and impacts of the study. Participants were assured that their data would be stored on computers with account passwords only available to members of the research team. They could withdraw from the research at any time for any reason, and their data would be deleted after they withdrew.

2.3. Data Collection

All investigators received standardized training before administering the survey. Paper questionnaires were distributed by oncology nurses from the participating hospital. After being verified by two independent oncology nurses, incomplete questionnaires were returned to participants for supplementation. The demographic data were cross‐checked against electronic medical records to ensure data accuracy.

2.4. Measures

2.4.1. Demographic and Clinical Characteristics

The self‐report questionnaire covered both demographic and clinical information. Demographics included age, residence, education level, sex, marital status, employment status, and annual income. Clinical data, including chronic disease status, family history of cancer, cancer type, cancer stage, surgical treatment, radiotherapy, chemotherapy, biological immunotherapy, and catheter status, were extracted from electronic health records.

2.4.2. Online Health Information Seeking Behaviors

OHISB was assessed using items from the Health Information National Trends Survey (HINTS), a nationally representative survey administered by the United States National Cancer Institute [30]. Participants were asked whether they had ever searched for health information online (response options: yes/no). Those who answered “no” were assigned a value of 0. Those who answered “yes” were then asked to evaluate their most recent search experience based on four items: “It took a lot of effort to get the information you needed,” “You felt frustrated during your search for the information,” “You were concerned about the quality of the information,” and “The information you found was hard to understand.” Responses were recorded on a 4‐point Likert scale, where a score of 1 indicates “strongly agree” and a score of 4 indicates “strongly disagree”. The OHISB score was calculated as the mean of the four items, consistent with recommendations in scale development of DeVellis et al. [38].

2.4.3. eHealth Literacy

The eHealth Literacy Scale, translated and adapted by Guo et al., was used to assess individuals' capacity to seek, assess, and utilize health information online [39]. The Cronbach's α coefficient of the scale was 0.913, and the factor load coefficients ranged from 0.692 to 0.869 [39]. This scale comprises three aspects: application skills, evaluation competence, and decision‐making capacity regarding online health information and services. This eight‐item instrument employs a 5‐point Likert scale (1 = strongly disagree to 5 = strongly agree), with higher total scores reflecting greater proficiency of eHealth literacy.

2.4.4. Self‐Efficacy

The General Self‐Efficacy Scale (GSES), developed by Schwarzer et al. and translated into Chinese by Wang et al., has demonstrated good psychometric properties (Cronbach's α = 0.870, test‐retest reliability r = 0.830, and split‐half reliability r = 0.900) [40, 41]. In this study, self‐efficacy was assessed using items adapted from the GSES and HINTS [42, 43]. Participants responded to the following items: “If I try hard enough, I can manage to solve difficult problems.”; “Overall, how confident are you about your ability to take good care of your health?”; and “Overall, how confident are you that you could get advice or information about health or medical topics if you needed it?” Responses were evaluated on a 5‐point Likert scale, with higher scores reflecting higher levels of self‐efficacy.

2.4.5. Information Utility

Information utility refers to how well the information provided by a medium meets an individual's needs [19]. It encompasses the caliber, credibility, and usability of the information [30]. This measure of the variable was developed from its conceptual definition and adapted from the HINTS. Participants were asked the following questions: “How much of the health information you encounter online would you consider to be inaccurate or misleading?”, “To what extent do you trust health or medical information you find online?”, and “Have you made health‐related decisions based on information obtained online?” Responses were assessed using a 5‐point Likert scale, where higher scores indicate greater trust in online health information.

2.4.6. Statistical Analysis

Statistical analysis was performed using IBM SPSS version 25.0 (IBM, Armonk, United States) and IBM AMOS version 28.0 (IBM, Armonk, United States). The demographic and clinical characteristics were analyzed using descriptive statistics. Continuous variables were presented as mean ± standard deviation or as median with interquartile range, whereas categorical variables were described using frequencies and percentages. Differences in OHISB by demographic and clinical characteristics were assessed using the Mann–Whitney U test or the Kruskal–Wallis H test. Spearman's correlation analysis was conducted to examine the relationships among eHealth literacy, self‐efficacy, information utility, and OHISB. Structural equation modeling with maximum likelihood estimation was employed to identify the sequential mediating effects of self‐efficacy and information utility in the relationship between eHealth literacy and OHISB. The model fit was evaluated using the χ² to degrees of freedom ratio (χ²/df), Goodness‐of‐Fit Index (GFI), Adjusted Goodness‐of‐Fit Index (AGFI), Incremental Fit Index (IFI), Tucker–Lewis Index (TLI), Comparative Fit Index (CFI), and RMSEA, with χ²/df ratio < 5.0, GFI > 0.90, CFI > 0.90, IFI > 0.90, TLI > 0.90, RMSEA < 0.08 considered indicative of an acceptable model fit [44]. The significance of mediating effect was tested via 5000 bootstrap samples, with statistical significance determined by 95% confidence intervals (CI) excluding zero (p < 0.05, two‐tailed). Additionally, hierarchical regression analysis was performed to examine moderating effects in this study.

3. Results

3.1. Descriptive Results

Table 1 presents the distribution of OHISB by different demographic and clinical characteristics. A total of 532 cancer patients were enrolled in this study, and 502 of them completed the questionnaires (response rate = 94.4%). The prevalence of OHISB among cancer patients in this study was 61.4%. Among the 502 participants, more than half were younger than 65 years, 60.4% were female, 93.8% were married, and 80.3% resided in urban areas. Most participants (72.3%) held less than a bachelor's degree, only 21.1% were employed, and 48.6% reported an annual income exceeding ¥50,000 (approximately $7020.2 USD).

Table 1.

The distribution of OHISB by different demographic and clinical characteristics (n = 502).

Characteristics	n (%)	OHISB [Median (IQR)]	U/H	p
Age (years)			25336.0	< 0.001
< 65	273 (54.4)	2.0 (0.0, 2.5)
≥ 65	229 (45.6)	0.8 (0.0, 2.3)
Sex			29870.0	0.856
Female	303 (60.4)	1.8 (0.0, 2.5)
Male	199 (39.6)	2.0 (0.0, 2.3)
Marital status			7189.0	0.883
Married	471 (93.8)	1.8 (0.0, 2.5)
Unmarried or other	31 (6.2)	1.8 (0.0, 2.5)
Residence			15052.0	< 0.001
Rural areas	99 (19.7)	0.0 (0.0, 2.3)
Urban areas	403 (80.3)	2.0 (0.0, 2.5)
Education level			26.0	< 0.001
≤ Senior high school	363 (72.3)	1.8 (0.0, 2.3)
College	125 (24.9)	2.0 (1.1, 2.5)
≥ Postgraduate	14 (2.8)	2.4 (1.2, 2.8)
Employment status			15256.0	< 0.001
Employed	106 (21.1)	2.3 (1.5, 2.5)
Retired or other	396 (78.9)	1.8 (0.0, 2.3)
Annual incomea			23665.5	< 0.001
≤ ¥50000	258 (51.4)	1.0 (0.0, 2.3)
> ¥50000	244 (48.6)	2.0 (0.0, 2.5)
Comorbid chronic diseases			28109.5	0.131
No	205 (40.8)	2.0 (0.0, 2.5)
Yes	297 (59.2)	1.8 (0.0, 2.3)
Family history of cancer			26680.0	0.037
No	308 (61.4)	1.8 (0.0, 2.4)
Yes	194 (38.6)	2.0 (0.0, 2.5)
Cancer type			14955.0	0.429
Gastrointestinal Cancer	428 (85.3)	1.8 (0.0, 2.5)
Other	74 (14.7)	1.8 (0.0, 2.3)
Cancer stage			0.8	0.844
I	14 (2.8)	0.8 (0.0, 2.3)
II	92 (18.3)	1.8 (0.0, 2.4)
III	129 (25.7)	2.0 (0.0, 2.3)
IV	267 (53.2)	1.8 (0.0, 2.5)
Surgical treatment			25709.0	0.260
No	160 (31.9)	1.8 (0.0, 2.3)
Yes	342 (68.1)	2.0 (0.0, 2.5)
Radiotherapy			10049.0	0.482
No	455 (90.6)	1.8 (0.0, 2.5)
Yes	47 (9.4)	2.0 (0.0, 2.5)
Chemotherapy			20924.5	0.001
No	144 (28.7)	1.0 (0.0, 2.0)
Yes	358 (71.3)	2.0 (0.0, 2.5)
Biological Immunotherapy			30650.0	0.638
No	236 (47.0)	1.8 (0.0, 2.4)
Yes	266 (53.0)	1.8 (0.0, 2.5)
Catheter			23683.5	0.010
No	161 (32.1)	1.0 (0.0, 2.3)
Yes	341 (67.9)	2.0 (0.0, 2.5)

Abbreviations: IQR, interquartile range; OHISB, online health information seeking behavior.

^a One Chinese Yuan (¥) = approximately US $0.1404.

Among 502 participants, 59.2% had comorbid chronic diseases, and 38.6% reported a family history of cancer. The majority of them (85.3%) were diagnosed with gastrointestinal cancers, and 53.2% presented with stage IV disease. In terms of treatment history, 68.1% had undergone surgical treatment, 9.4% had received radiotherapy, 71.3% had received chemotherapy, 53.0% had received biological immunotherapy, and 67.9% had a catheter. There were significant differences in OHISB across age, education level, residence, employment status, annual income, family history of cancer, chemotherapy, and catheter use (p < 0.05).

3.2. Correlation Results

Spearman's correlation analysis was conducted to investigate the relationships between eHealth literacy, self‐efficacy, information utility, and OHISB (Table 2). There was a significant positive correlation between eHealth literacy, self‐efficacy, information utility, and OHISB (r = 0.357, 0.499, 0.469, p < 0.01). Self‐efficacy was associated with information utility and OHISB (r = 0.554, 0.714, p < 0.01). Information utility was correlated with OHISB (r = 0.565, p < 0.01).

Table 2.

The connections between eHealth literacy, self‐efficacy, information utility and OHISB.

Variables	eHealth literacy	Self‐efficacy	Information utility	OHISB
eHealth literacy	1
Self‐efficacy	0.357a	1
Information utility	0.499a	0.554a	1
OHISB	0.469a	0.714a	0.565a	1
Score	23.5 ± 9.4b	8.4 ± 3.5b	7.7 ± 3.5b	1.8 (0.0, 2.5)c

Abbreviation: OHISB, online health information seeking behavior.

^a p < 0.01,

^b Mean ± Standard deviation,

^c Median (Interquartile range).

3.3. Mediation Model Testing

Structural equation modeling was conducted to examine the mediating effects of self‐efficacy and information utility on the association between eHealth literacy and OHISB. The model demonstrated an acceptable fit indices: χ²/df = 3.072, GFI = 0.980, AGFI = 0.969, IFI = 0.986, TLI = 0.979, CFI = 0.986, and RMSEA = 0.064 (Table 3).

Table 3.

Model fit standards and model fit indices.

Item	χ²/df	GFI	AGFI	IFI	TLI	CFI	RMSEA
Model fit standards	< 5.0	> 0.90	> 0.90	> 0.90	> 0.90	> 0.90	< 0.08
Model fit indices	3.072	0.980	0.969	0.986	0.979	0.986	0.064

Abbreviations: AGFI, Adjusted Goodness‐of‐Fit Index; CFI, Comparative Fit Index; GFI, Goodness‐of‐Fit Index; IFI, Incremental Fit Index; RMSEA, Root Mean Square Error of Approximation; TLI, Tucker‐Lewis Index.

The mediation analysis revealed that eHealth literacy was positively associated with self‐efficacy (β = 0.477, p < 0.001), information utility (β = 0.347, p < 0.001), and OHISB (β = 0.201, p < 0.001). Self‐efficacy had a direct effect on both information utility (β = 0.428, p < 0.001) and OHISB (β = 0.495, p = 0.001). Information utility positively predicted OHISB (β = 0.169, p < 0.001) as well. All pathways are shown in Figure 2.

Figure 2.

The mediating effects of self‐efficacy and information utility between eHealth literacy and OHISB. OHISB, online health information seeking behavior; ^a p < 0.01.

To assess the indirect effects, 95% CI was computed utilizing percentile bootstrapping and bias‐corrected percentile bootstrapping with 5,000 resamples. An indirect effect was considered statistically significant if the upper and lower limits of the CI did not include zero. The findings revealed that both self‐efficacy and information utility acted as a mediator in the association between eHealth literacy and OHISB. Moreover, self‐efficacy and information utility contributed to a chain mediating effect in the relationship between eHealth literacy and OHISB, as presented in Table 4.

Table 4.

Bootstrap analysis of the mediating model.

Model Path	Standardize β	Standard error	Bias‐corrected 95% CI		Effect proportion
			Lower	Upper
1. eHL→Se→OHISB	0.223	0.038	0.155	0.304	44.5%
2. eHL→IU → OHISB	0.056	0.025	0.013	0.110	11.2%
3. eHL→Se→IU → OHISB	0.033	0.012	0.010	0.057	6.6%
Total indirect effect	0.311	0.045	0.227	0.404	62.1%
Total direct effect	0.190	0.050	0.099	0.295	37.9%
Total effect	0.501	0.039	0.419	0.576	100.0%

Abbreviations: CI, confidence interval; eHL, eHealth literacy; IU, information utility; OHISB, online health information seeking behavior; Se, self‐efficacy.

A hierarchical regression analysis was conducted to examine the potential moderating effects of several factors (e.g., age, education, residence, employment status, annual income, family history of cancer, chemotherapy, and catheter use) on the relationships between the independent variables (eHealth literacy, self‐efficacy, and information utility) and OHISB. The results revealed that annual income had a significant positive moderating effect on the association between eHealth literacy and OHISB (β = 0.328, p = 0.004). Both age (β = 0.099, p = 0.032) and employment status (β = 0.294, p = 0.012) were positive moderators of the association between information utility and OHISB. Detailed results are presented in Supporting Information S1: File 1.

4. Discussion

This study supported a sequential mediation model in which eHealth literacy influenced OHISB through the mediating roles of self‐efficacy and information utility. The findings identified a significant positive effect that eHealth literacy had on OHISB (supporting Hypothesis 1). In addition, both self‐efficacy and information utility mediated the association between eHealth literacy and OHISB (supporting Hypothesis 2 and 3), and they were sequential mediators between eHealth literacy and OHISB, with self‐efficacy positively influencing information utility (supporting Hypothesis 4).

In this study, the prevalence of OHISB among cancer patients was 61.4%, which is consistent with the findings of Yip et al. and Mitsutake et al. [11, 12]. However, it is higher than that reported among patients with diabetes [13], a difference that may be attributed to the greater perceived risk associated with cancer. According to the Common‐Sense Model of Self‐Regulation, individuals construct cognitive representations of health threats that guide their self‐management behaviors. Zhang et al. also found that a higher perceived risk among cancer patients was associated with more active OHISB [45]. This study demonstrated that eHealth literacy influenced OHISB among cancer patients. This finding supports Hypothesis 1 and is consistent with prior research conducted in Sweden and Minnesota [22, 46]. One possible explanation for this is that people with greater access to digital devices (such as smartphones, computers, and tablets) tend to develop higher eHealth literacy [46]. Those with strong eHealth literacy skills are more proficient in using digital resources, understanding online health information, and evaluating the credibility of sources [12]. Consequently, in an era of rapid digital transformation, improving eHealth literacy among cancer patients is essential for enhancing their ability to seek health information and facilitating effective self‐management.

Findings from this research indicated that self‐efficacy mediated the relationship between eHealth literacy and OHISB, supporting Hypothesis 2. Self‐efficacy, the belief in one's own capability to achieve success in particular circumstances, influences the coping strategies and overall well‐being of individuals diagnosed with cancer. Consistent with previous research, this study also revealed that eHealth literacy was related to self‐efficacy [28, 47]. Higher eHealth literacy is conducive to more effective seeking, comprehension, and evaluation of online health information, thereby enhancing individuals' self‐efficacy in health information utilization. Park et al. found that health anxiety is negatively correlated with self‐efficacy. In contrast, individuals with higher eHealth literacy are better equipped to manage health anxiety, which in turn enhances their self‐efficacy [48]. A study by Ji et al. suggested that eHealth literacy improves patients' knowledge, which may contribute to greater self‐efficacy [28]. Higher self‐efficacy is associated with improved psychological adjustment, better quality of life, and more health‐promoting behaviors [25, 49, 50]. Tierney et al. demonstrated that individuals with greater self‐efficacy tend to be more confident and resilient when tackling tasks, which partially explains why self‐efficacy promotes OHISB [51]. In accordance with the findings of Zhang et al. and Ji et al. [28, 47], this study confirmed the mediating influence of self‐efficacy between eHealth literacy and OHISB, emphasizing its pivotal role in health‐related behaviors.

More importantly, this study demonstrated that eHealth literacy had an indirect effect on OHISB via information utility, supporting Hypothesis 3. Consistent with this finding, de Oliveira Collet et al. found that individuals with higher eHealth literacy are better able to filter online content critically and, as a result, seek online health information more frequently [52]. According to the dual processing model [53], users with higher ability tend to apply systematic strategies such as assessing content accuracy and scientific validity, while those with lower ability often depend on heuristic cues like the visual features. Higher eHealth literacy enables individuals to evaluate the information utility in complex digital environments, rather than relying on the peripheral route. Additionally, Masilamani et al. argued that information credibility was related to individuals' willingness to use online information, which stimulated proactive information‐seeking behavior correspondingly [54]. This process is also influenced by information sources, highlighting the importance of providing professional and accessible health information. As information utility increases, cancer patients may experience lower decision‐making anxiety in the context of information overload, thereby engaging more actively in health information‐seeking behavior [55]. Owing to commercial and ideological incentives, a large amount of health misinformation spreads online, hindering the public from accessing reliable health information [56]. Therefore, the development of patient‐oriented information platforms aimed at enhancing information utility is of critical importance. In contrast to the results of this study, Lim et al. found that the impact of eHealth literacy on the evaluation of information credibility was not significant. One possible reason is that 80% of the participants were college students or graduates, and 12% were postgraduates. The relatively homogeneous educational background may have resulted in low variance in eHealth literacy, thereby weakening its influence on credibility evaluation.

This study further shed light on the underlying mechanism between eHealth literacy and OHISB, confirming the chain mediating roles of self‐efficacy and information utility in the process, supporting Hypothesis 4. Adequate eHealth literacy provides cancer patients with a supportive informational environment, effectively enhancing their self‐efficacy and information utility. According to John W. Atkinson's Expectancy‐Value Theory [57], both expectancy and task value contribute to enhancing motivation and ultimately lead to action. This may explain the positive influence of self‐efficacy and information utility on OHISB observed in this study. Ma et al. also found that the interplay between self‐efficacy and utility value predicted individuals' achievement [58]. Those with greater self‐efficacy demonstrated an enhanced ability to overcome obstacles, sustain effort, and proactively develop strategies to access high‐quality and value information, thus further reinforcing their information‐seeking behaviors. As self‐efficacy improves, patients' perception of information utility strengthens, and this positive psychological transformation manifests as more proactive information‐seeking behavior. Cui et al. also reported that self‐efficacy influenced health information‐seeking behavior via information utility [36]. This finding is consistent with the theoretical assumptions of the CMIS model that self‐efficacy functions as an antecedent factor [19]. However, in the present study, self‐efficacy also serves as a mediating variable. This discrepancy may be attributed to the integration of the KAB model into the original CMIS framework. Previous research has suggested that eHealth literacy can indirectly influence health information seeking behavior by enhancing self‐efficacy, and the results of this study further support this view [28, 29]. In addition, as a foundational framework, the CMIS model may not have captured certain critical pathways, a limitation acknowledged by the author. The significance of this study lies in identifying the potential relationship between self‐efficacy and information utility, and validating their sequential mediating roles in the association between eHealth literacy and OHISB.

Furthermore, the results showed that annual income moderated the relationship between eHealth literacy and OHISB. This finding is consistent with prior research suggesting that individuals with higher socioeconomic status tend to engage in proactive health behaviors and accumulate greater health capital [59]. In contrast to Wilson et al., this study revealed that age strengthened the link between information utility and OHISB, which underscores the importance of age‐stratified health strategies [60]. Employment status also positively moderated the relationship between information utility and OHISB. The workplace environment facilitates the accumulation of social capital, broadens access to information resources, and enhances the utility of information, thereby actively promoting information seeking behaviors [61]. Taken together, these results highlight the importance of targeted health information dissemination. Public health initiatives should prioritize improving access to trustworthy digital health resources among lower income and unemployed populations to narrow disparities in OHISB.

4.1. Limitations

To begin with, this study was conducted at a single tertiary hospital in Shanghai with convenience sampling, and the sample contained a high proportion of gastrointestinal cancer patients. These factors may restrict the broader applicability of our results. Future investigations should incorporate more heterogeneous samples to enhance external validity. Secondly, this study adopted a cross‐sectional design, limiting insights into how OHISB evolves during cancer treatment. Longitudinal studies are recommended to explore the dynamic changes in OHISB. Thirdly, limitations in our assessment of confounders (e.g., social support), information sources (e.g., professional medical websites vs. social media), and lifestyle measures should be addressed in future research to reduce bias in the estimated serial mediation effects.

4.2. Implications for Clinical Practice

Considering the rapid internet advancement and the increasing burden of cancer, the findings from this survey offer implications for clinical practice. Guiding cancer patients to seek and utilize health information has become a critical issue in health management. The positive impact of eHealth literacy on OHISB emphasizes that enhancing eHealth literacy among cancer patients helps promote proactive health behaviors and effective self‐care, which correspondingly alleviates healthcare systems' pressure and improves patients' quality of life. Individuals with lower eHealth literacy benefit most from targeted interventions [62]; accordingly, resources should be prioritized for these groups to mitigate the digital divide. Different populations should receive differentiated interventions. For example, small‐group, face‐to‐face sessions covering basic computer terminology and health‐website navigation can be offered to older adults, whereas e‐learning modules focused on rigorous methods for evaluating online health information are more suitable for younger and middle‐aged adults. This stratified, needs‐based approach is likely to maximize gains in eHealth literacy and promote OHISB. This study also revealed the chain mediating relationship among eHealth literacy, self‐efficacy, and perceived information utility, which suggests that clinical interventions should not only focus on eHealth literacy but also on the confidence of cancer patients and the value of online health information. Health policymakers should integrate eHealth literacy into health promotion strategies, develop targeted training programs, and create supportive ecosystems that facilitate digital health information seeking. To provide coping strategies for the adverse effects of anticancer therapy and the latest information on cancer treatment, it is recommended that a professional and comprehensible health information platform be established. With the rapid development of artificial intelligence (AI), its safety has become paramount. The National Computer Network Emergency Response Technical Team has introduced the AI Safety Governance Framework 2.0, which underscores the need to improve model transparency and interpretability, strengthen data and privacy protection mechanisms, and reinforce cybersecurity defenses to prevent information leakage. For health education agents, it is essential to develop a scientifically rigorous and standardized knowledge corpus to ensure that patients receive accurate and reliable support from medical large language models.

5. Conclusions

This research emphasizes the critical role of eHealth literacy in empowering cancer patients through improved OHISB. eHealth literacy enhances OHISB both directly and indirectly through the mediation pathway of self‐efficacy and information utility. These findings create a theoretical foundation for further research into the interactive mechanisms of health information seeking behavior and offer practical guidance for developing targeted interventions. The study recommends that healthcare professionals need to adopt multidimensional interventions to strengthen eHealth literacy and self‐efficacy of cancer patients, and establish authoritative and reliable information access platforms to improve information utility, thereby providing directions for creating multimodal health information support system for cancer patients.

Author Contributions

Tianchun Zhou: methodology, software, data curation, formal analysis, writing – original draft, conceptualization, visualization. Mengmeng Lyu: conceptualization, methodology, software, formal analysis, validation, writing – review and editing, data curation. Lu Chen: investigation, validation, data curation. Mengting Ji: conceptualization, writing – review and editing, project administration, resources, funding acquisition, supervision, validation.

Ethics Statement

This research was conducted according to the Declaration of Helsinki and its subsequent revisions. This research obtained permission from the Ethics Committee of the participating hospital (RA‐2022‐491).

Consent

All participants provided written informed consent after receiving complete information about the study's objective, significance, methods and impacts of the study. Participants were assured that their data would be stored on computers with account passwords only available to members of the research team. They could withdraw from the research at any time for any reason, and their data would be deleted after they withdrew.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Moderating effects on the relationship between eHealth literacy and OHISB.

Table S2: Moderating effects on the relationship between self‐efficacy and OHISB.

Table S3: Moderating effects on the relationship between Information utility and OHISB.

Zhou T., Lyu M., Chen L., and Ji M., “Association Between eHealth Literacy and Online Health Information Seeking Behavior Among Cancer Patients: A Chain Mediating Role of Self‐Efficacy and Information Utility,” Health Care Science 5 (2026): 299–309. 10.1002/hcs2.70079.

Data Availability Statement

The data that support the findings of this study can be obtained by contacting the corresponding author, upon reasonable request.