Abstract Cholangiocarcinoma (CCA), strongly linked to Opisthorchis viverrini infection through risky dietary practices, remains a major public health burden in Northern Thailand. This study investigated dietary behaviors associated with CCA and their determinants guided by the Health Belief Model (HBM) among residents of Phrae Province, an endemic region in Northern Thailand. A cross-sectional study was conducted with 461 residents aged 20 and above, using a structured questionnaire. Participants were chosen through a multi-stage sampling process. Data analysis involved descriptive statistics and multiple logistic regression, with results expressed as adjusted odds ratios (AORs) and 95% confidence intervals (CIs). Overall, 67.7% of participants reported consuming at least one food type associated with CCA risk, including papaya salad with raw fermented fish sauce (45.1%) and raw fish larb (36.0%), and three additional fermented or raw fish preparations. Factors significantly linked to risky dietary behaviors included being male (AOR=2.17, 95% CI=1.15-4.08, P-value=0.016), having only primary school education (AOR=5.15, 95% CI=2.47-10.73, P-value<0.001), no history of liver fluke screening (AOR=2.09, 95% CI=1.09-3.99, P-value=0.025), moderate knowledge of CCA (AOR=3.73, 95% CI=1.97-7.06, P-value <0.001), moderate perceptions based on the Health Belief Model (AOR=4.51, 95% CI=2.27–8.96, P-value<0.001), and moderate attitudes toward CCA prevention (AOR=2.09, 95% CI=1.10-4.42, P-value=0.043). These findings underscore the need for gender-sensitive health education, expanded liver fluke screening, and culturally informed behavior change programs targeting high-risk groups in endemic communities of northern Thailand.

Keywords: Cholangiocarcinoma, Dietary behavior, Health belief model, Opisthorchiasis, Risk factors

Citation:  Tanthanapanyakorn, P. and Maoon, K. 2026. Factors associated with risky dietary behaviors for cholangiocarcinoma among residents of an endemic area in Northern Thailand: A cross-sectional study. Natural and Life Sciences Communications. 25(4): e2026080.

Graphical Abstract:

INTRODUCTION

Opisthorchis viverrini infection and cholangiocarcinoma (CCA) represent a critical public health challenge in Thailand, strongly associated with premature mortality (Laoprom et al., 2021). The International Agency for Research on Cancer (IARC) classifies O. viverrini as a Group 1 biological carcinogen and a definitive etiological agent of CCA (IARC, 1994). Infection occurs primarily through consumption of raw or undercooked freshwater fish containing infective metacercariae, found in traditional dishes such as koi pla, pla som, pla jom, and pla ra (Prasopdee et al., 2023; Smout et al., 2024). Recurrent infections provoke chronic bile duct inflammation, periductal fibrosis, and eventual malignant transformation, increasing CCA risk approximately sixfold. Additional risk factors include advanced age, family history of cancer, alcohol intake, and repeated praziquantel treatment (Kamsa-ard et al., 2018; Prasopdee et al., 2023).

Thailand bears the highest global burden of cholangiocarcinoma, with incidence highest in the Northeast (85 per 100,000 annually) and second highest in the North (14.5 per 100,000 annually) (Kamsa-ard et al., 2018). In Northern Thailand, liver and bile duct cancers predominantly affect males, with registry data from Lampang Cancer Hospital (2018-2022) recording age-standardized death rates of 34.3 per 100,000 in males and 12.0 per 100,000 in females (Lampang Cancer Hospital, 2020). Phrae Province represents a critical hotspot, reporting the highest male incidence (43.9 per 100,000), second-highest female incidence (15.8 per 100,000), and highest liver and bile duct cancer mortality in the region (42.54 per 100,000), with the highest death rates concentrated in Nong Muang Khai, Song, and Mueang Phrae districts (Department of Medical Services, Ministry of Public Health, 2025).

The national prevalence of liver fluke infection is 11.8%, with the highest burden in the Northeast and the second highest in the North (5.2%) (Health Information System Development Office, Ministry of Public Health, 2025). Provinces with the greatest prevalence, Nan, Lamphun, Phrae, and Lampang, are traversed by major river systems that sustain cyprinid fish, the second intermediate host harboring infective metacercariae. Shared ecological conditions, together with longstanding cultural practices involving the consumption of raw or partially cooked fish, amplify the risk of transmission. These environmental and behavioral determinants substantially contribute to the elevated cancer burden observed in Northern Thailand (Namsanor et al., 2020; Charoensuk et al., 2022; Khuntikeo et al., 2023).

Although nationwide public health efforts have contributed to reducing the prevalence of liver fluke infection, CCA incidence and mortality remain unacceptably high in many endemic communities. Persistently high mortality imposes substantial healthcare costs, diminishes quality of life, and places considerable socioeconomic burden on affected families (Chavengkun et al., 2016; Lausatianragit et al., 2019; Chuangchaiya et al., 2020; Perakanya et al., 2022). Effective prevention requires precise identification of modifiable behavioral risk factors, particularly dietary practices. Studies in Northeastern Thailand have demonstrated associations between sex, age, and consumption of raw or fermented freshwater fish, including pla ra, pla jom, and pla som, with liver fluke infection and CCA, partly attributed to carcinogenic N-nitroso compounds (Chuangchaiya et al., 2020; Perakanya et al., 2022; Busabong et al., 2025). Personal characteristics, knowledge, attitudes, and disease perceptions have also been identified as key determinants of risky dietary behaviors (Saenna et al., 2017; Jaikaew et al., 2024; Chompo et al., 2025). To address this gap, understanding the psychosocial and behavioral determinants of raw fish consumption is essential, yet evidence from Northern Thailand remains limited.

However, evidence from Northern Thailand remains limited, particularly in high-mortality provinces such as Phrae. Previous investigations have largely focused on infection prevalence and clinical outcomes, with few studies examining behavioral and psychosocial determinants in high-burden northern communities (Jaikaew et al., 2024; Busabong et al., 2025). Sociocultural dietary practices and environmental contexts in Northern Thailand also differ markedly from those in the Northeast, potentially influencing exposure pathways and risk perception. Therefore, this study aimed to determine the prevalence of risky dietary behaviors associated with CCA and to identify their sociodemographic and cognitive determinants, including knowledge, HBM-based perceptions, and preventive attitudes among residents of an endemic area in Northern Thailand. The findings are intended to inform liver fluke control strategies, support health authorities in prioritizing high-risk populations, and guide future research in endemic areas.

MATERIALS AND METHODS

Study design

Findings are reported in accordance with the STROBE Statement (Von Elm et al., 2007). This cross-sectional study examined factors associated with risky dietary behaviors related to cholangiocarcinoma among residents of an endemic area in Northern Thailand. A conceptual framework grounded in the Health Belief Model (HBM) guided the selection and analysis of study variables (Rosenstock et al., 1988; Stretcher and Rosenstock, 1997; Jones et al., 2015). The HBM posits that health behavior is shaped by individual beliefs and motivational processes. Six core constructs were applied: perceived severity, perceived susceptibility, perceived benefits, perceived barriers, self-efficacy, and cues to action (Rosenstock et al., 1988). Each construct was operationalized through validated questionnaire items to assess its relationship with risky dietary practices.

The framework was informed by the HBM and a comprehensive review of literature on dietary behaviors associated with cholangiocarcinoma risk. Factors were categorized into four domains: (1) personal factor (sex, age, occupation, education level, monthly income, marital status, family history of CCA, history of liver fluke examination, history of praziquantel treatment, and history of OV infection); (2) knowledge regarding CCA; (3) attitudes toward CCA prevention; and (4) perceptions related to CCA based on HBM constructs. The study proposed that these four domains were associated with risky dietary consumption behaviors among residents of an endemic area in Northern Thailand.

Setting

The study was conducted in Song District, Phrae Province, Northern Thailand, a predominantly rural endemic area for liver fluke infection and CCA. Song District was selected based on its exceptionally high cholangiocarcinoma mortality rate, the highest in Northern Thailand (Khuntikeo et al., 2016). The district comprises eight subdistricts where traditional consumption of raw and partially cooked freshwater fish remains prevalent, making it an appropriate setting for investigating behavioral risk factors associated with CCA.

Participants

The study population comprised 353,979 individuals aged 20 years and older with a registered domicile in Phrae Province, Thailand (Phrae Provincial Public Health Office, 2025).

The sample size was determined using Daniel’s formula for finite populations, as the total population size was known (Daniel, 1999). The confidence level was set at 95% (Z = 1.96), the allowable error at 5% (e = 0.05), and the expected population proportion at 0.50 to yield the maximum required sample size. Based on these parameters, the calculated sample size was 384 participants. To compensate for potential non-response and incomplete data, and to enhance data reliability, an additional 20% was added, resulting in a final sample size of 461 participants.

Inclusion and exclusion criteria

Eligible participants were residents aged 20 years and older with a registered domicile in Phrae Province who had resided in the study area for at least six months. Participants were required to be mentally competent, able to communicate independently, and capable of accurately completing the questionnaire or interview. All participants voluntarily agreed to participate and provided written informed consent before data collection. Individuals were excluded if they declined or withdrew consent, were unable to provide complete or reliable information, had cognitive or sensory impairments affecting communication, had severe illness or psychiatric conditions that could impair reliable participation, or were non-Thai citizens without permanent residence in Phrae Province.

Sampling technique

A multi-stage sampling technique was employed. Phrae Province was selected purposively based on its highest CCA mortality rate in Northern Thailand. Song District was then selected through simple random sampling using the lottery method. According to the Civil Registration Statistics Database (September 2025), the district comprised 33,668 residents aged 20 years and older. Stratified random sampling proportional to population size was subsequently applied across eight subdistricts. The total sample of 461 participants was allocated proportionally as follows: Huai Mai (n = 87), Tao Poon (n = 86), Hua Mueang (n = 70), Sa-iap (n = 60), Ban Nun (n = 54), Ban Klang (n = 39), Thung Nao (n = 37), and Dan Chumphon (n = 28). Finally, eligible participants were recruited in accordance with the inclusion criteria and provided voluntary informed consent.

Measurement tool

The research instrument was a structured questionnaire developed by the researchers, grounded in the study’s conceptual framework and an extensive review of relevant literature on CCA and risky dietary behaviors (Rosenstock et al., 1988; Stretcher and Rosenstock, 1997; Saenna et al., 2017; Jaikaew et al., 2024; Chompo et al., 2025). The questionnaire was designed to correspond with the study variables and consisted of five parts.

Part 1: Personal characteristics questionnaire. This section comprised ten items in a closed-ended format, including sex, age, occupation, education level, monthly income, marital status, history of liver fluke examination, family relationship with CCA patients, history of OV infection, and history of praziquantel treatment.

Part 2: Knowledge of the CCA questionnaire. This section contained 12 dichotomous items (Yes/No). Positive statements were scored 1 for correct and 0 for incorrect responses, while negative statements were reverse-scored. Total scores were categorized into three levels based on Bloom’s criteria: high (80–100 %), moderate (60–79 %), and low (<60%) (Bloom, 1956).

Part 3: Perception of the CCA questionnaire. This section assessed health beliefs based on the HBM, including perceived susceptibility, perceived severity, perceived benefits, perceived barriers, self-efficacy, and cues to action. It consisted of 24 items rated on a five-point Likert scale ranging from strongly agree to strongly disagree, with both positive and negative statements. Mean scores were interpreted using Best’s criteria (Best, 1977) as high (3.68–5.00), moderate (2.34–3.67), and low (1.00–2.33).

Part 4: Attitudes toward the CCA prevention questionnaire. This section included 12 Likert-scale items with five response levels. Both positively and negatively worded statements were scored accordingly. Mean scores were interpreted using Best’s criteria (Best, 1977) as high (3.68–5.00), moderate (2.34–3.67), and low (1.00–2.33).

Part 5: Risky dietary consumption behavior questionnaire. This section consisted of 10 items measured on a five-point Likert scale: regularly, often, sometimes, rarely, and never. Positive and negative items were scored inversely. Mean scores were classified into three levels using Best’s criteria (Best, 1977): Inappropriate behavior (3.68–5.00), moderate behavior (2.34–3.67), and Appropriate behavior (1.00–2.33).

Three experts reviewed the questionnaire for content validity, comprising a professional nurse, a public health specialist, and a family medicine specialist. Each item was evaluated using the Index of Item-Objective Congruence (IOC), with values ranging from 0.86 to 0.93 across sections. Item clarity and relevance were also checked through face validity assessment. A pilot study with 30 participants of similar characteristics to the study population was conducted before the main data collection. Cronbach's alpha exceeded 0.80 for the perception, attitude, and behavior scales. For the knowledge scale, the Kuder–Richardson 20 (KR-20) coefficient was above 0.80, meeting the threshold for acceptable reliability.

Data collection

Before data collection, a formal letter was submitted to the Phrae Provincial Public Health Office to obtain permission. The research team then coordinated with provincial and district public health offices, subdistrict health-promoting hospitals, and village health volunteers to recruit participants. Research assistants received standardized training covering study objectives, eligibility criteria, questionnaire administration, informed consent procedures, and data recording to ensure procedural consistency. Data were collected in Song District, Phrae Province, between February and March 2026. After eligible participants were identified, research assistants introduced the study and obtained written informed consent. Face-to-face interviews were conducted using a structured questionnaire, taking approximately 50 minutes per participant. Those who declined participation or did not meet eligibility criteria were excluded.

Ethical considerations

This study was approved by the Research Ethics Committee of Valaya Alongkorn Rajabhat University under the Royal Patronage (REC No. 0114/2025; COA No. 0001/2026) on February 6, 2026, and was conducted in accordance with the Declaration of Helsinki. All participants received information regarding the study objectives, procedures, and their rights before data collection.

Bias

Potential sources of bias were addressed across all study phases. Selection bias was minimized through clearly defined eligibility criteria and multi-stage stratified random sampling proportional to population size. Information bias was reduced through a structured, validated questionnaire and standardized training for research assistants. Although self-reported responses introduced the possibility of recall and reporting bias, face-to-face interviews with clear question explanations helped mitigate these limitations.

Statistical analysis

Data were analyzed using IBM SPSS Statistics version 29.0.1 (IBM Corp.) at a significance level of 0.05. Descriptive statistics, including frequency, percentage, mean, and standard deviation (SD), summarized participant characteristics and study variables. Normality of continuous variables was confirmed using the Kolmogorov-Smirnov test (P-value > 0.05). Risky dietary consumption behavior was initially classified into three levels: appropriate, moderate, and inappropriate. For logistic regression, the moderate and inappropriate categories were combined into a single binary outcome representing risky dietary behavior, with appropriate behavior as the reference group. Pearson’s correlation coefficient was used to assess the association between perceptions of CCA based on the HBM and risky dietary consumption behaviors.

Binary logistic regression (BLR) examined crude associations between independent variables and risky dietary consumption behavior, while multiple logistic regression (MLR) identified factors independently associated with risky dietary behavior after controlling for potential confounders. Results were expressed as Crude Odds Ratios (COR) and Adjusted Odds Ratios (AOR) with 95% confidence intervals (CI). Before analysis, key assumptions were verified: the dependent variable was dichotomous, observations were independent as each participant contributed data only once, and multicollinearity was assessed using the Variance Inflation Factor (VIF), with all values within acceptable limits. Variables were entered into the model using the enter method.

RESULTS

Personal factors

Table 1 summarizes participants’ personal characteristics. Most were female (54.9%) and aged 20-59 years (69.9%), with the majority having completed secondary school education or higher (62.5%). The largest occupational groups were farmers or agricultural workers (43.4%) and general laborers (32.5%). Most reported a monthly income of lower 10,000 Thai Baht (68.3%) and were married (69.6%). Regarding liver fluke-related history, nearly half (48.4%) had never undergone liver fluke screening, 11.9% had received praziquantel treatment, and small proportions reported a family history of CCA (4.1%) or previous Opisthorchis viverrini infection (2.4%).

Table 1.  Descriptive characteristics of participants’ personal factors (n = 461).

Note. Data are expressed as mean ± standard deviation or n (%). SD = Standard deviation, CCA = Cholangiocarcinoma,  OV= Opisthorchis viverrini

Knowledge, perception, and attitude levels regarding CCA

Table 2 presented the distribution of knowledge, perceptions, and attitudes toward CCA among participants. Most participants demonstrated a high level of CCA knowledge (64.0%), and a large proportion reported high HBM-based perceptions (67.7%). In contrast, attitudes toward CCA prevention were predominantly moderate (52.5%), with only 47.5% reporting high attitudes. This pattern suggested that despite relatively strong knowledge and disease perception, attitudes toward preventive behaviors remained less favorable, indicating a gap between cognitive awareness and behavioral intention that may require targeted health education interventions.

Risky dietary behaviors associated with CCA

Table 3 presented the distribution of risky dietary behaviors related to CCA among participants. Overall, 67.7% of participants reported consuming at least one risky food item. Papaya salad with raw fermented fish sauce was the most commonly reported item (45.1%), followed by raw fish larb (36.0%) and three additional raw or fermented fish preparations.

Association between the perception of HBM regarding CCA and risky dietary consumption behavior

Table 4 presented the association between HBM-based perceptions of CCA and risky dietary consumption behavior. All perception components showed statistically significant negative correlations with risky dietary behavior (P-value < 0.05). Self-efficacy had the strongest correlation (r = -0.335), followed by cues to action and perceived barriers. Overall perception scores showed a moderate negative correlation (r = -0.376), indicating that higher CCA perception was associated with lower engagement in risky dietary behaviors.

Table 2. Distribution of knowledge, perception, and attitude levels regarding CCA among participants (n = 461).

Note: CCA=Cholangiocarcinoma, HBM= Health Belief Model.

Table 3.  Distribution of risky dietary behaviors associated with CCA among participants (n = 461).

Note: CCA=Cholangiocarcinoma *Answer more than one choice

Table 4. Relationship between perception of the CCA and risky dietary consumption behavior (n=461).

Note: r, Correlation Coefficient *Statistically significant at P-value < 0.05, HBM= Health Belief Model.

Factors associated with risky dietary consumption behaviors related to CCA

MLR results are presented in Table 5. Cutoff points and reference categories for all independent variables were defined based on theoretical rationale, response distributions, and relevance to public health outcomes. All variables were assessed for multicollinearity before inclusion in the model and met the assumptions of logistic regression. Male sex was associated with 2.17 times higher odds of risky dietary consumption than female sex (AOR = 2.17, 95% CI = 1.15-4.08, P-value = 0.016). Participants with primary school education were 5.15 times more likely to report risky dietary behaviors compared with those with secondary school education or higher (AOR = 5.15, 95% CI = 2.47-10.73, P-value < 0.001).

Those who had never undergone liver fluke examination were 2.09 times more likely to engage in risky dietary behaviors than those previously screened (AOR = 2.09, 95% CI = 1.09-3.99, P-value = 0.025). Moderate knowledge of CCA was associated with 3.73 times higher odds compared with high knowledge (AOR = 3.73, 95% CI = 1.97-7.06, P-value < 0.001). Participants with moderate HBM-based CCA perception had 4.51 times higher odds than those with high perception (AOR = 4.51, 95% CI = 2.27-8.96, P-value < 0.001). Moderate attitudes toward CCA prevention were also associated with increased odds of risky dietary behaviors (AOR = 2.09, 95% CI = 1.10-4.42, P-value = 0.043).

Table 5. Factors associated with risky dietary consumption behaviors related to cholangiocarcinoma among participants (n = 461).

Note: 1.00 = Reference group; COR = Crude Odds Ratio, derived from BLR; AOR = Adjusted Odds Ratio, derived from MLR; CI = Confidence interval; HBM = Health Belief Model; CCA = Cholangiocarcinoma. *Statistically significant at P-value < 0.05.

DISCUSSION

The present study found that 67.7% of participants reported consuming at least one type of food associated with CCA risk, indicating that risky dietary practices remain common among residents in endemic areas of Northern Thailand. The most frequently consumed high-risk foods were papaya salad prepared with raw fermented fish sauce (45.1%), followed by raw fish larb (36.0%), raw fermented fish (Pla Som) (34.1%), fermented raw fish (Pla Jom) (32.5%), and raw fish salad (Koi Pla/Sa Pla) (30.8%). For instance, Chavengkun et al. (2016) reported that 78% of participants consumed raw fish among populations at risk of CCA in Nakhon Ratchasima Province. In addition, evidence from Charoensuk et al. (2022) indicated that in rural, endemic communities in Northeastern Thailand, approximately 60–90 % of individuals consume raw fish, reflecting the persistence of high-risk dietary practices. Similarly, Perakanya et al. (2022) found that 49.2% of respondents consumed raw fish in endemic communities of Sakon Nakhon Province.

Sociocultural and environmental factors likely explain the persistence of risky dietary behaviors in this population. Raw and fermented fish dishes are deeply embedded in local culinary traditions and regularly served at family and community gatherings, where social norms make refusal difficult. Taste preference and familiarity further reinforce consumption, as these dishes are widely regarded as desirable within local food culture (Satitvipawee et al., 2025). Low cost and easy access to freshwater fish in rural areas also sustain their continued use (Perakanya et al., 2022; Busabong et al., 2025). Limited personal perception of susceptibility to liver fluke infection may lead residents to underestimate health risks, reducing motivation to change established dietary habits.

Male sex was independently associated with risky dietary consumption behaviors related to CCA, a finding consistent with previous studies reporting higher-risk dietary patterns among men in endemic communities (Kamsa-ard et al., 2018; Homsana et al., 2024). Several mechanisms may account for this association. Men more frequently attend social gatherings where raw fish salad and fermented fish are served, and sociocultural norms in rural settings may discourage refusal of these dishes. Lower risk perception among men may also reduce motivation to adopt preventive behaviors (Chuangchaiya et al., 2020). Outdoor and agricultural work further increases exposure to freshly caught fish consumed raw (Chompo et al., 2025). Gender-sensitive interventions targeting men are therefore needed in endemic areas.

Lower educational attainment was independently associated with risky dietary consumption behaviors, consistent with prior evidence that individuals with less formal education are more prone to consuming raw or fermented freshwater fish linked to Opisthorchis viverrini infection and CCA (Kamsa-ard et al., 2018). This association likely reflects limited health literacy and reduced access to accurate information on dietary risks. Raw fish and fermented fish dishes remain deeply rooted in rural food culture, and those with lower education may have fewer opportunities to engage with health education or prevention programs (Kosonsasitorn, 2022; Khuntikeo et al., 2023; Jaikaew et al., 2024).

Absence of liver fluke screening was independently associated with risky dietary consumption behaviors, a pattern reported in prior studies where unscreened individuals showed a greater tendency toward consuming raw or fermented freshwater fish linked to Opisthorchis viverrini infection and CCA (Boondit et al., 2020; Perakanya et al., 2022). Those who have never been screened may have limited awareness of personal infection risk and tend to underestimate the health consequences of raw fish consumption (Songserm et al., 2020). Screening contacts also provides opportunities for health counseling on liver fluke transmission and CCA prevention, and unscreened individuals may have fewer such interactions with healthcare services.

Participants with moderate knowledge of CCA were significantly associated with risky dietary consumption behaviors. Previous studies reported that individuals with limited knowledge of liver fluke infection and CCA are more likely to consume raw or fermented freshwater fish dishes associated with Opisthorchis viverrini infection (Chavengkun et al., 2016; Sornlorm et al., 2019; Suwannahitatorn et al., 2019). This pattern may reflect the knowledge-attitude-practice gap widely documented in health behavior research, wherein factual awareness alone is insufficient to change deeply embedded cultural practices without accompanying motivational and self-efficacy support (Charoensuk et al., 2022). In rural communities where raw fish salad and fermented fish remain dietary staples, moderate knowledge may not translate into behavior change (Wang et al., 2021). Limited exposure to continuous health education or community-based prevention programs may further sustain these practices.

Moderate HBM-based CCA perception was independently associated with risky dietary consumption behaviors. Prior studies link lower perceived susceptibility, severity, benefits, or self-efficacy to continued consumption of raw or fermented freshwater fish (Nakbun et al., 2018; Suwannahitatorn et al., 2019; Satitvipawee et al., 2025). Moderate perception likely reflects insufficient recognition of personal risk, reducing motivation to avoid high-risk foods. Higher perception levels were negatively associated with risky behaviors, indicating that stronger disease beliefs support safer dietary practices (Charoensuk et al., 2022; Satitvipawee et al., 2025; Kyi et al., 2026). Where raw and fermented fish remain culturally entrenched, dietary change is unlikely without adequate self-efficacy and risk awareness (Rattaporn et al., 2026).

Moderate attitudes toward CCA prevention were independently associated with risky dietary consumption behaviors, consistent with prior studies linking less favorable prevention attitudes to continued raw or fermented fish consumption (Chavengkun et al., 2016; Lausatianragit et al., 2019; Suwannahitatorn et al., 2019; Satitvipawee et al., 2025). Individuals with moderate attitudes may not fully appreciate the value of preventive behaviors such as avoiding raw fish or adopting safer food preparation (Suwannahitatorn et al., 2019). In rural communities where these dishes remain culturally accepted, moderate attitudes likely reflect insufficient motivation to change established dietary practices despite some disease awareness (Nakbun et al., 2018).

Several limitations should be acknowledged. First, the cross-sectional design precludes causal inference, as the observed associations cannot confirm temporal or directional effects. Second, the study was conducted in Song District, Phrae Province, which may limit generalizability to populations with different sociocultural and environmental contexts. Third, self-reported data introduce the possibility of recall and social desirability bias, particularly for dietary and health-related behaviors. Finally, psychosocial measures, including knowledge, perceptions, and attitudes, reflect a single point in time and may not capture changes in beliefs or behaviors over time.

Several clinical and public health implications emerge from these findings. Community-based education on the risks of raw or fermented freshwater fish consumption should prioritize high-risk groups, particularly men, individuals with lower education, and those not previously screened for liver fluke infection. Integrating health education with routine screening may strengthen knowledge, risk perception, and preventive attitudes simultaneously. Village health volunteers and local healthcare workers are well-positioned to support behavior change at the community level. Sustained reduction in risky dietary practices will require coordinated efforts among public health authorities, healthcare facilities, community leaders, and educational institutions.

Future research should employ longitudinal designs to clarify causal relationships and examine how knowledge, HBM perceptions, and attitudes influence dietary behaviors and liver fluke infection risk over time. Qualitative and mixed-methods approaches would help explore sociocultural beliefs and contextual factors sustaining raw or fermented fish consumption, informing culturally appropriate interventions. Randomized controlled trials are also needed to evaluate the effectiveness of health education, behavior change strategies, and expanded liver fluke screening in reducing risky dietary practices.

CONCLUSION

These findings from Northern Thailand, a region with distinct sociocultural practices compared to the more extensively studied Northeast, point to the value of region-specific public health strategies. Many residents still reported consuming foods linked to liver fluke transmission and CCA risk, despite overall appropriate dietary behavior. Culturally appropriate health education, expanded liver fluke screening, and targeted behavior change programs addressing knowledge, perceptions, and attitudes remain priorities for reducing the burden of cholangiocarcinoma in endemic communities.

ACKNOWLEDGEMENTS

The authors would like to express their sincere gratitude to the provincial and local public health officers in Phrae Province for their valuable support and coordination in facilitating data collection. We also extend our appreciation to the community leaders and village health volunteers for their cooperation and assistance throughout the study process. Finally, we are deeply grateful to all participants for their time, willingness, and active collaboration in this research. No financial support was received for this study.

AUTHOR CONTRIBUTIONS

Phannathat Tanthanapanyakorn: Conceptualization (Lead), Methodology (Lead), Formal Analysis (Lead), Validation (Lead), Data Curation (Supporting), Investigation (Supporting), Writing – Original Draft (Equal), Writing – Review & Editing (Supporting); Kunlaphatcha Maoon: Conceptualization (Supporting), Methodology (Supporting), Formal Analysis (Supporting), Resources (Lead), Data Curation (Lead), Investigation (Lead), Supervision (Lead), Project Administration (Lead), Writing – Original Draft (Equal), Writing – Review & Editing (Lead), Providing Final Approval of the Manuscript (Lead).

CONFLICT OF INTEREST

The authors declare that they have no conflicts of interest.

REFERENCES

Best, J.W. 1977. Research in education. Prentice Hall.

Bloom, B.S. 1956. Taxonomy of educational objectives: Handbook I: The cognitive domain. New York (NY): David McKay Co.

Boondit, J., Suwannahitatorn, P., Siripattanapipong, S., Leelayoova, S., Mungthin, M., Tan-Ariya, P., Piyaraj, P., Naaglor, T., and Ruang-Areerate, T. 2020. Epidemiological survey of Opisthorchis viverrini infection in a lightly infected community, eastern Thailand. American Journal of Tropical Medicine and Hygiene. 102: 838-843. https://doi.org/10.4269/ajtmh.19-0864

Busabong, W., Songserm, N., Woradet, S., and Sripa, B. 2025. Influence of knowledge, attitude and self-efficacy on Opisthorchis viverrini and cholangiocarcinoma prevention: A one health approach in high-risk areas of Thailand. Tropical Medicine and International Health. 30: 99-107. https://doi.org/10.1111/tmi.14073

Charoensuk, L., Ribas, A., Chedtabud, K., and Prakobwong, S. 2022. Infection rate of Opisthorchis viverrini metacercariae in cyprinoid fish from markets and its association to human opisthorchiasis in northeast Thailand. Acta Tropica. 225: 106216. https://doi.org/10.1016/j.actatropica.2021.106216

Chavengkun, W., Kompor, P., Norkaew, J., Kujapun, J., Pothipim, M., Ponphimai, S., Kaewpitoon, S.J., Padchasuwan, N., and Kaewpitoon, N. 2016. Raw fish consuming behavior related to liver fluke infection among populations at risk of cholangiocarcinoma in Nakhon Ratchasima province, Thailand. Asian Pacific Journal of Cancer Prevention. 17: 2761-2765.

Chompo, P., Sripa, B., Suwannatrai, A.T., Tangkawattana, P., and Tangkawattana, S. 2025. Knowledge gaps and risk behaviors among cat owners in Opisthorchis viverrini endemic areas: Implications for public health interventions. Open Veterinary Journal. 15: 4322-4336. https://doi.org/10.5455/OVJ.2025.v15.i9.39

Chuangchaiya, S., Navanesan, S., Jaichuang, S., Rahim, M.A.F.A., and Idris, Z.M. 2020. Current prevalence of Opisthorchis viverrini infection and associated risk factors in Nakhon Phanom province, northeastern Thailand. Tropical Biomedicine. 37: 986-999.

Daniel, W.W. 1999. Biostatistics: A foundation for analysis in the health sciences. 7^th ed. John Wiley & Sons.

Department of Medical Services, Ministry of Public Health. 2025. Cancer incidence and mortality in Health Region 1, 2018-2022. Lampang Cancer Hospital.

Health Information System Development Office, Ministry of Public Health. 2025. Provincial mortality (13 causes): Trends and rankings by province, 2018-2023 [Internet]. Nonthaburi (Thailand): Ministry of Public Health; [cited 2026 Mar 18]. Available from: https://www.hiso.or.th/healthoutcome/index.php?code=36

Homsana, A., Southisavath, P., Kling, K., Hattendorf, J., Vorasane, S., Paris, D.H., Probst-Hensch, N., Sayasone, S., and Odermatt, P. 2024. Burden and risk factors of suspected cholangiocarcinoma in high Opisthorchis viverrini endemic rural communities in southern Lao PDR. PLoS Neglected Tropical Diseases. 18: e0012617. https://doi.org/10.1371/journal.pntd.0012617

International Agency for Research on Cancer (IARC). 1994. Schistosomes, liver flukes, and Helicobacter pylori. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. 61: 1-241.

Jaikaew, J., Songserm, N., Charoenbut, P., Thongchai, C., and Chada, W. 2024. Environmental management for Opisthorchis viverrini and cholangiocarcinoma prevention in a high-risk area of Thailand: The KALMeFS model. Asian Pacific Journal of Cancer Prevention. 25: 4305-4311. https://doi.org/10.31557/APJCP.2024.25.12.4305

Jones, C.L., Jensen, J.D., Scherr, C.L., Brown, N.R., Christy, K., and Weaver, J. 2015. The health belief model as an explanatory framework in communication research: Exploring mediation. Health Communication. 30: 566-576. https://doi.org/10.1080/10410236.2013.873363

Kamsa-ard, S., Kamsa-ard, S., Luvira, V., Suwanrungruang, K., Vatanasapt, P., and Wiangnon, S. 2018. Risk factors for cholangiocarcinoma in Thailand: A systematic review and meta-analysis. Asian Pacific Journal of Cancer Prevention. 19: 605-614.

Khuntikeo, N., Loilome, W., Thinkhamrop, B., Chamadol, N., and Yongvanit, P. 2016. A comprehensive public health conceptual framework and strategy to combat cholangiocarcinoma in Thailand. PLoS Neglected Tropical Diseases. 10: e0004293. https://doi.org/10.1371/journal.pntd.0004293

Khuntikeo, N., Thinkhamrop, B., Crellen, T., Eamudomkarn, C., Petney, T.N., Andrews, R.H., and Sithithaworn, P. 2023. Epidemiology and control of Opisthorchis viverrini infection: Implications for cholangiocarcinoma prevention. Recent Results in Cancer Research. 219: 27-52. https://doi.org/10.1007/978-3-031-35166-2_3

Kosonsasitorn, C. 2022. Cholangiocarcinoma in Lampang hospital. Lampang Medical Journal. 34: 84-91.

Kyi, P.P.N., Siewchaisakul, P., and Wungrath, J. 2026. Dietary diversity and social determinants among Myanmar university students, Chiang Mai province. Natural and Life Sciences Communications. 25: e2026042. https://doi.org/10.12982/NLSC.2026.042

Laoprom, N., Prathummang, S., Chuangchaiya, S., Navanesan, S., Munajat, M.B., Suwannatrai, A.T., and Idris, Z.M. 2021. Opisthorchis viverrini metacercarial infection in cyprinid fish in Nakhon Phanom province, northeastern Thailand. Tropical Biomedicine. 38: 25-30. https://doi.org/10.47665/tb.38.2.033

Lampang Cancer Hospital. 2020. Trends in cancer incidence and mortality in Northern Thailand, 1993-2017 [Internet]. Lampang (Thailand): Lampang Cancer Hospital; [cited 2026 Mar 18]. Available from: https://www2.lpch.go.th/lpch/uploads/20200617093948752932.pdf

Lausatianragit, W., Jaroenprasert, S., Lausatianragit, K., Pramual, P., and Boonprasong, S. 2019. Factors related to the consumption of raw fish among people in Sisaket Province, 2016. Journal of Health Science of Thailand. 28: 974-985.

Namsanor, J., Kiatsopit, N., Laha, T., Andrews, R.H., Petney, T.N., and Sithithaworn, P. 2020. Infection dynamics of Opisthorchis viverrini metacercariae in cyprinid fishes from two endemic areas in Thailand and Lao PDR. American Journal of Tropical Medicine and Hygiene. 102: 110-116. https://doi.org/10.4269/ajtmh.19-0432

Nakbun, S., Thongkrajai, P., and Nithikathkul, C. 2018. Risk factors for Opisthorchis viverrini infection in Nakhon Phanom Province, Thailand. Asian Biomedicine. 12: 45-51. https://doi.org/10.1515/abm-2018-0030

Perakanya, P., Ungcharoen, R., Worrabannakorn, S., Ongarj, P., Artchayasawat, A., Boonmars, T., and Boueroy, P. 2022. Prevalence and risk factors of Opisthorchis viverrini infection in Sakon Nakhon Province, Thailand. Tropical Medicine and Infectious Disease. 7: 313. https://doi.org/10.3390/tropicalmed7100313

Phrae Provincial Public Health Office. 2025. Annual report: Population data of Phrae Province.

Prasopdee, S., Rojthongpond, T., Chitkoolsamphan, Y., Pholhelm, M., Yusuk, S., Pattaraarchachai, J., Butthongkomvong, K., Kulsantiwong, J., Phanaksri, T., Kunjantarachot, A., et al. 2023. Update on the risk factors for opisthorchiasis and cholangiocarcinoma in Thailand. Parasites, Hosts and Diseases. 61: 463-470. https://doi.org/10.3347/PHD.23032

Rattaporn, W., Sillabutra, J., Kitidamrongsuk, P., Kamsa-Ard, S., and Satitvipawee, P. 2026. Predictors of liver fluke infection-preventive behaviors in Khon Kaen Province: A structural equation model. Asian Pacific Journal of Cancer Prevention. 27: 931-939. https://doi.org/10.31557/APJCP.2026.27.3.931

Rosenstock, I.M., Stretcher, V.J., and Becker, M.H. 1988. Social learning theory and the health belief model. Health Education Quarterly. 15: 175-183. https://doi.org/10.1177/109019818801500203

Saenna, P., Hurst, C., Echaubard, P., Wilcox, B.A., and Sripa, B. 2017. Fish sharing as a risk factor for Opisthorchis viverrini infection: Evidence from two villages in northeastern Thailand. Infectious Diseases of Poverty. 6: 66. https://doi.org/10.1186/s40249-017-0281-7

Satitvipawee, P., Sillabutra, J., Kitidamrongsuk, P., Rattaporn, W., Pattanasin, S., Kamsa-Ard, S., and Kamsa-Ard, S. 2025. Prevalence and factors affecting prevention of Opisthorchis viverrini infection in Khon Kaen province, northeast Thailand. Asian Pacific Journal of Cancer Prevention. 26: 3785-3795. https://doi.org/10.31557/APJCP.2025.26.10.3785

Smout, M.J., Laha, T., Chaiyadet, S., Brindley, P.J., and Loukas, A. 2024. Mechanistic insights into liver-fluke-induced bile-duct cancer. Trends in Parasitology. 40: 1183-1196. https://doi.org/10.1016/j.pt.2024.10.012

Songserm, N., Woradet, S., Sripa, B., and Ali, A. 2020. Sustainable prevention of cholangiocarcinoma through community participation in a high-incidence area in Thailand. Asian Pacific Journal of Cancer Prevention. 21: 777-782. https://doi.org/10.31557/APJCP.2020.21.3.777

Sornlorm, K., Loahasiriwong, W., Sithithaworn, P., and Thinkhamrop, W. 2019. Influence of geographic, knowledge, and behavioral factors on Opisthorchis viverrini infection in Northeast Thailand. Asian Pacific Journal of Tropical Medicine. 12: 499-506. https://doi.org/10.4103/1995-7645.271289

Stretcher, V.J. and Rosenstock, I.M. 1997. The health belief model. In: Glanz K, Lewis FM, Rimer BK, editors. Health behavior and health education: theory, research, and practice. San Francisco (CA): Jossey-Bass. p. 31-43.

Suwannahitatorn, P., Webster, J., Riley, S., Mungthin, M., and Donnelly, C.A. 2019. Uncooked fish consumption among those at risk of Opisthorchis viverrini infection in central Thailand. PLoS One. 14: e0211540. https://doi.org/10.1371/journal.pone.0211540

Von Elm, E., Altman, D.G., Egger, M., Pocock, S.J., Gøtzsche, P.C., Vandenbroucke, J.P., and STROBE Initiative. 2007. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Lancet. 370: 1453-1457. https://doi.org/10.1016/S0140-6736(07)61602-X

Wang, Y.C., Grundy-Warr, C., Namsanor, J., Kenney-Lazar, M., Tang, C.J.Y., Goh, L.Y.W., Chong, Y.C., Sithithaworn, P., Ngonkum, S., and Khuntikeo, N. 2021. Masculinity and misinformation: Social dynamics of liver fluke infection risk in Thailand. Parasitology International. 84: 102382. https://doi.org/10.1016/j.parint.2021.102382         

OPEN access freely available online

Natural and Life Sciences Communications

Chiang Mai University, Thailand. https://cmuj.cmu.ac.th