This cross-sectional study was conducted at the author’s Hospital of Wenling City between July 2023 and January 2024 and focused on patients with IBD. The majority of patients were recruited through hospital-based patient groups and were diagnosed according to standard guidelines, including endoscopic, pathological, and surgical findings, and received long-term standard treatment. Exclusion criteria comprised patients with concomitant infectious enteritis such as bacterial dysentery, amoebic dysentery, and intestinal tuberculosis, as well as those with other intestinal diseases like ischemic enteritis and radiation enteritis. Additionally, patients with concurrent heart, lung, liver, kidney, or mental disorders, and individuals under the age of 18 were not included in the study. A total of 18 patients were excluded due to concomitant enteritis conditions, and 25 patients were excluded due to comorbid heart, lung, liver, kidney, or mental disorders. The study received ethical approval from the Ethics Committee of the author’s Hospital, and informed consent was obtained from all participants.

The questionnaire was conceptualized to capture three interrelated domains: knowledge, attitudes, and decisional conflict, based on the understanding that patient education (knowledge) influences perceptions (attitude), which in turn affects confidence in decision-making (decisional conflict). This conceptual alignment allowed for a coherent interpretation of results across domains. The design of this questionnaire was based on “Expert consensus on biological therapy for IBD in China (2021)” (16, 17) and previous literature (18). Modifications were made based on feedback from three specialists (one with expertise in small intestine diseases and two in IBD). Experts recommended the following: (1) listed the specific types of biologics; (2) investigated women’s fertility status; and (3) investigated patients’ awareness of the side effects of biologics. A preliminary survey showed good internal consistency with a Cronbach’s α of 0.874, based on 23 valid responses.

The final questionnaire was in Chinese and consisted of four sections: demographic data, knowledge dimension, attitude dimension, and decisional conflict scale (Supplementary materials). The demographic section gathered information on age, height, weight, gender, place of residence, educational attainment, average monthly family income, smoking and drinking habits, type and duration of IBD, history of intestinal surgery, current and past medications, current use of biologics, history of switching biologics, parental status, and continuation of biologic therapy during pregnancy or lactation. The knowledge dimension comprised six questions, with points awarded as follows: 2 points for “very familiar,” 1 point for “heard of,” and 0 points for “not sure,” applicable to questions 1–5. Question 6 was not scored, making the total possible score range from 0 to 10 points. The attitude dimension included six questions, scored using a five-point Likert scale. Possible scores ranged from 6 to 30. The decisional conflict scale consisted of 16 questions, also scored on a five-point Likert scale, with total scores ranging from 0 to 100. This score was calculated by summing all item scores, dividing by 16, and then multiplying by 25. Clinically significant decisional conflict was defined as a total score > 37.5 (19). According to Bloom’s cutoff, a well-established benchmark in educational environments (20), knowledge, or attitude of respondents is classified based on their scoring percentages. Scores below 60% of the total were deemed insufficient or lower level. Scores between 60 and 80% were considered moderate. Scores exceeding 80% were regarded as sufficient or higher level.

The questionnaires were distributed through a combination of online and offline channels to ensure broad participation. The electronic questionnaire was imported into the Questionnaire Star platform¹ and distributed through IBD doctor-patient communication WeChat groups, the gastroenterology outpatient clinic, and inpatient wards, effectively reaching participants familiar with digital tools. For electronic distribution, researchers introduced the purpose, content, and significance of the study before the survey. To ensure data completeness, electronic questionnaires could only be submitted after all items were filled out, and each IP address was allowed to answer the survey only once to avoid duplicate submissions. In contrast, the paper questionnaires were targeted at inpatients who were unable to use online scanning methods, ensuring inclusivity and comprehensive participation. For paper distribution, researchers also explained the purpose and content of the study. Trained staff distributed the questionnaires in person, and participants completed them voluntarily. The completed questionnaires were collected on-site.

The researchers ensured the integrity and internal consistency of the paper and electronic questionnaires, and incomplete or logically contradictory answers were considered invalid. This dual-channel distribution model (online and offline) was designed to maximize accessibility, minimize selection bias, and ensure the representativeness of the sample by including both digitally literate patients and those without access to online tools.

A single population proportion formula, n = [(Z_α/2)²*P(1-P)]/d², was used to calculate the sample size. Since there were no KAP studies on inflammatory bowel disease in the Chinese population, the sample size for this study was calculated based on an expected proportion of patients’ understanding of biologic mechanisms at 50%, with a confidence level of 95% and a margin of error of 5%, requiring a sample size of 384 individuals. Considering a 10% dropout rate, the adjusted sample size required was 426 individuals. This calculation ensured sufficient statistical power to detect meaningful associations between patient characteristics and their knowledge, attitudes, and decisional conflict regarding biologics.

The statistical analysis was structured to first describe the sample characteristics, then explore bivariate associations, and finally identify independent predictors through multivariate modeling. This hierarchical approach allowed for both descriptive insights and hypothesis-driven testing. Statistical analysis was performed using STATA 18.0 (Stata Corporation, College Station, TX, United States). Initially, continuous variables were tested for normality. Those that were normally distributed were presented as mean ± standard deviation (SD) and comparisons between groups were made using the t-test and one-way analysis of variance (ANOVA). Non-normally distributed variables were reported as median (range) and compared using the Mann–Whitney U test or the Kruskal-Wallis analysis of variance. Categorical variables were presented as n (%). The relationships between the knowledge and attitude dimensions were assessed using Spearman correlation analysis. Variables with p < 0.05 in the univariate logistic regression analysis were selected for the multivariate regression analysis. Multivariate logistic regression analysis was utilized with adequate knowledge and higher attitude as the dependent variables. A two-sided p < 0.05 was considered statistically significant. The alpha value of 0.05 was selected based on conventional statistical standards in clinical research. For multiple comparisons, we applied the Bonferroni correction. Specifically, since three main multivariate analyses were conducted, the corrected significance level was set at 0.0167.

This methodology was chosen over alternative approaches because the KAP framework allows for a structured, validated, and population-specific assessment of patient knowledge, attitudes, and decisional conflict. This is particularly important in the Chinese IBD context, where cultural and healthcare system factors influence treatment perceptions, and where comparable KAP data are lacking. The cross-sectional design combined with a targeted questionnaire enabled efficient, large-scale data collection and facilitated direct comparison with existing international studies.

A total of 456 questionnaires were collected. After excluding 5 cases due to incomplete completion of the KAP section, 37 cases with illogical responses, and 6 questionnaires completed in less than 120 s, 405 valid questionnaires remained for analysis. Among the participants, 45.90% were aged 30–49, and 64.00% were male (Table 1). Among the medications previously or currently used by participants, the most common was mesalazine (87.2%), followed by steroids (26.4%) and azathioprine (21.0%). Traditional Chinese Medicine was used by 16.8%, while Sulfasalazine (4.2%) and Methotrexate (2.0%) were the least common. With respect to the biologics currently in use by the participants, adalimumab was the most commonly used (26.7%), followed by infliximab (25.9%) and vedolizumab (20.7%). Notably, 10.9% of participants were not using any biologics, and tofacitinib was not used by any participants.

The mean scores of knowledge and attitude were 6.03 ± 2.98 (possible range: 0–11) and 24.36 ± 3.32 (possible range: 8–40). Among them, 60 (14.8%) patients exhibited significant decisional conflict. Overall, younger age, higher education level, higher household income, diagnosis of CD, prior use of biologics, and lower decisional conflict were consistently associated with higher knowledge and higher attitude scores toward biologic therapy. Detailed subgroup analyses are presented in Table 1. Participants aged 18–29 and 30–49 had higher knowledge scores compared to those aged ≥ 50 years (p < 0.001). Rural residents had higher attitude scores (p = 0.034). Higher education levels positively impacted both knowledge and attitude scores (p < 0.001, p = 0.001). Income levels influenced both domains, with higher incomes linked to higher scores (p < 0.001). Non-smokers and non-drinkers showed higher attitude scores, and non-smokers also had higher knowledge scores (p = 0.041, p = 0.024, and p = 0.025, respectively). Participants with CD had higher knowledge and attitude scores than those with other IBD conditions (p < 0.001). Changing biologics positively influenced knowledge and attitude (both p < 0.001). Patients with lower decisional conflict scores showed higher knowledge and attitude scores (p < 0.001; Table 1).

The majority of respondents (60.2%) were very familiar with biologics as a treatment option for IBD, while 32.8% had heard of them, and a small percentage (6.9%) were not sure. However, familiarity with the risks associated with biologics, such as increased infection and tumor risk, was low, with only 25.2% very familiar and 31.4% unsure. Additionally, only 1.5% were familiar with Tofacitinib, highlighting a significant gap in knowledge about specific biologics (Supplementary Table 1; Table 2).

Among the participants, 81.5% agreed or strongly agreed that biologics would improve their condition, and 88.4% were willing to accept biologic therapy if recommended by their doctor. However, concerns about side effects were noted, with 13.1% expressing resistance or hesitancy toward these treatments. Additionally, financial considerations played a critical role, with only 68.1% feeling confident in their ability to afford biologic therapy, even with insurance coverage (Supplementary Table 2).

In terms of decision conflict, the majority of participants (23% strongly agree and 50.4% agree) reported having a clear understanding of their treatment options. Furthermore, 89.9% (27.4% strongly agree and 62.5% agree) felt no external pressure in making their decision, and 84.7% (22.2% strongly agree and 62.5% agree) expressed satisfaction with their choice (Supplementary Table 3).

In an additional subgroup analysis, patients who reported using Traditional Chinese Medicine (TCM) had lower attitude scores compared to those who did not use TCM (23.19 ± 3.49 vs. 24.59 ± 3.24, p = 0.001). No statistically significant differences were observed for knowledge scores (5.54 ± 3.15 vs. 6.13 ± 2.93, p = 0.139) or decisional conflict (26.93 ± 13.33 vs. 23.87 ± 13.32, p = 0.085; Supplementary Table 4).

Spearman correlation analysis showed that knowledge scores were positively correlated with attitude scores (r = 0.554, p < 0.001). In the multivariate analysis for knowledge, age ≥50 years (OR = 0.46, 95% CI [0.22–0.97], p = 0.04), UC (OR = 0.53, 95% CI [0.31–0.93], p = 0.026), never having used biologics (OR = 0.16, 95% CI [0.06–0.45], p = 0.001), and clinically significant decisional conflict (OR = 0.13, 95% CI [0.06–0.29], p < 0.001) were each associated with lower odds of having adequate knowledge.

In the multivariate analysis for attitude, higher knowledge score was associated with higher attitude (OR = 1.32, 95% CI [1.19–1.46], p < 0.001). Monthly household income of 10,000–20,000 RMB was also positively associated with higher attitude (OR = 2.20, 95% CI [1.01–4.78], p = 0.048). In contrast, never having used biologics (OR = 0.16, 95% CI [0.06–0.46], p = 0.001) and clinically significant decisional conflict (OR = 0.08, 95% CI [0.03–0.23], p < 0.001) were associated with lower odds of higher attitude (Table 3).