Knowledge, Attitudes, and Practices Regarding Autoimmune Thyroiditis Among Patients in Jilin: A Cross-Sectional Endocrinology Department Study.

Save

Objectives

To investigate the knowledge, attitudes, and practices (KAP) related to the diagnosis and treatment of autoimmune thyroiditis (AITD) among patients with AITD in an endocrinology department.

Methods

This cross-sectional study was conducted from November 2024 to September 2025 in the Department of Endocrinology of the authors’ hospital. A total of 474 patients with autoimmune thyroiditis were enrolled by convenience sampling. Data were collected using a self-designed, expert-validated structured KAP questionnaire, administered either on site in paper form or online via the Wenjuanxing platform.

The questionnaire assessed knowledge, attitudes, and practices regarding AITD, and scores of at least 70% of the maximum possible score were defined as indicating sufficient knowledge, positive attitudes, or adequate practices, respectively.

Results

Among 474 participants, mean knowledge, attitude, and practice scores were 5.79±4.19, 36.15±4.08, and 45.38±10.67, respectively. Higher income, longer disease duration, female gender, urban residence, and marital status correlated with better knowledge and attitudes; smoking, drinking, uncertain family history, and self-employment correlated with lower scores. Knowledge and attitude were independently associated with practice.

Structural equation modeling showed that knowledge was significantly and positively associated with both attitude (β=0.451) and practice (β=0.238); attitude was positively associated with practice (β=0.411) and partially mediated the association between knowledge and practice (indirect effect β=0.185; all P<0.001).

Conclusion

Patients with AITD followed at the author’s Hospital had insufficient knowledge but generally positive attitudes and moderate practice behaviors. Future studies should design and test interventions to improve AITD-related KAP. Interventions that strengthen disease-specific knowledge and address psychological concerns may further reinforce positive attitudes and promote more comprehensive and consistent self-management, thereby potentially improving clinical outcomes.

Introduction

Autoimmune thyroiditis (AITD), particularly Hashimoto’s thyroiditis (HT), represents one of the most common autoimmune diseases globally and is a principal cause of hypothyroidism. Its high prevalence, estimated at 5–10% worldwide and exceeding 20% in some populations, establishes it as a significant public health concern.1–3 The disease disproportionately affects women, with incidence rising with age, and imposes a multifaceted clinical burden.2–4 Chronic lymphocytic infiltration can lead to goiter and progressive hypothyroidism, which in turn is linked to dyslipidemia, atherosclerosis, and increased cardiovascular risk. Furthermore, AITD profoundly impacts quality of life through persistent symptoms like depression, cognitive dysfunction, and fatigue, even in biochemically euthyroid patients.5–9 For women of reproductive age, it poses additional risks including infertility and adverse pregnancy outcomes.4 The chronic, relapsing nature of AITD necessitates lifelong management, generating substantial direct and indirect healthcare costs and societal impact.1,8,10

Effective management of AITD extends beyond standard levothyroxine replacement therapy for hypothyroidism. While crucial for restoring euthyroidism, pharmacological treatment alone often fails to resolve all patient symptoms, highlighting a gap between biochemical control and patient-perceived well-being.11,12 This underscores the critical role of patient self-management, which is integral to long-term disease control. Optimal outcomes depend on the patient’s ability to adhere to medication regimens, engage in regular laboratory monitoring, recognize symptoms of under- or overtreatment, and make informed lifestyle modifications related to diet, stress, and exercise.11,13,14 Consequently, patient education and empowerment have become central pillars of modern AITD care.11,13,15,16

Despite the recognized importance of patient engagement, significant gaps in knowledge, attitudes, and practices (KAP) persist among individuals with AITD. Existing studies indicate that patients often misunderstand core aspects of their condition, demonstrate poor adherence to medical advice, and have low awareness of how lifestyle factors influence thyroid health.11–13 The Knowledge, Attitude, and Practice (KAP) model provides a classical theoretical framework for systematically assessing these patient-side factors, positing that knowledge shapes attitudes, which in turn drive health behaviors.17–19 However, while the KAP model has been widely applied to other chronic diseases, its application to AITD remains limited. The existing literature on this topic is sparse, often focusing on knowledge alone rather than an integrated KAP assessment, and rarely models the interplay between these components.20

Therefore, this study was designed to comprehensively investigate the knowledge, attitudes, and practices related to AITD among patients in a specialized endocrinology setting in China. By identifying specific deficits and their interrelationships, this research aims to provide an evidence base for developing targeted educational interventions. The findings are expected to inform the optimization of chronic disease management models and enhance doctor-patient collaboration, ultimately improving patient outcomes in AITD.

To our knowledge, this is among the first studies to apply an integrated KAP assessment with structural equation modeling to examine how knowledge and attitudes jointly relate to self-management practices in patients with AITD in China.

Materials and Methods

Study Design and Participants

This cross-sectional study was conducted from November 2024 to September 2025 at the authors’ Hospital, enrolling patients with AITD. Participants were enrolled using convenience sampling of eligible patients attending the endocrinology outpatient clinics during the study period. Patients diagnosed with autoimmune thyroiditis according to established clinical criteria were invited to participate if they met the eligibility criteria and provided informed consent.

The study was approved by the Institutional Review Board of the Second Hospital of Jilin University (Research Review No. (407) of 2024). All participants were informed about the study protocol and provided written informed consent to participate in the study.

Recruitment and questionnaire administration were conducted in the Department of Endocrinology at the Second Hospital of Jilin University. Eligible patients were approached by trained study staff in the endocrinology clinical area (such as the outpatient waiting area and clinic area) during the study period. Participants completed the questionnaire either on site in paper form or online via the Wenjuanxing platform using a QR code.

The inclusion criteria were: 1) age between 18 and 60 years (this age restriction was used to reduce heterogeneity related to older age, including age-related cognitive decline and comorbidity burden, which could confound questionnaire-based KAP assessment); and 2) meeting the diagnostic criteria for AITD, defined here as Hashimoto’s thyroiditis as the prototypical form, clinically suspected based on diffuse thyroid enlargement and firm texture, and confirmed by laboratory tests showing positive serum thyroid peroxidase antibodies (TPOAb) and/or thyroglobulin antibodies (TgAb). The exclusion criteria were: 1) thyroid cancer, history of thyroidectomy, or acute/subacute thyroiditis; 2) secondary thyroid dysfunction caused by pituitary or hypothalamic diseases; 3) other systemic autoimmune diseases, such as systemic lupus erythematosus, rheumatoid arthritis, Sjögren’s syndrome, or psoriasis; 4) severe cardiovascular or cerebrovascular diseases, hepatic or renal insufficiency, malignant tumors, or acute infection; 5) receipt of iodine-131 therapy or diagnostic tests, or use of iodine-containing contrast agents or drugs within the past six months; 6) current or recent use of glucocorticoids, immunosuppressants, biologics, or sex hormones (estrogen/progestin) that could affect immune or endocrine evaluation; 7) pregnant or lactating women, minors (<18 years), or older adults (>60 years); and 8) neurological disorders, history of mental illness (eg, use of antidepressants/anxiolytics or receipt of psychological therapy), or consciousness impairment that could affect emotional or quality-of-life assessments.

Sample Size Calculation

The sample size was determined a priori based on the number of items in the KAP questionnaire (Appendix). The final instrument included 11 knowledge items, 9 attitude items, and 12 practice items, yielding a total of 32 scored items. Following the commonly used rule of thumb in questionnaire-based psychometric research and multivariable analysis that at least 5–10 participants are needed per item, the minimum required sample size was estimated to be 160–320 participants (32 × 5–10).

To allow for incomplete questionnaires and exclusions related to data quality control, we increased the target sample size by approximately 50%, resulting in a planned minimum sample of about 300 participants. A total of 474 valid questionnaires were ultimately included in the final analysis, exceeding this minimum requirement and supporting the stability of the planned regression and structural equation modeling analyses.

Data Collection and Instrument

Data were collected using a self-designed, structured questionnaire. The initial item pool was developed based on a review of the KAP literature, including methodological references on KAP survey design and relevant studies on thyroid-related knowledge, attitudes, and practices. The draft questionnaire was then reviewed by three senior endocrinologists for clinical relevance, content coverage, clarity of wording, and appropriateness for patients with AITD.

Revisions were made based on their suggestions before pilot testing.

A pilot survey was subsequently conducted in 30 patients with AITD to assess comprehensibility and preliminary internal consistency. Based on participant feedback, minor wording adjustments were made to improve clarity and readability. The pilot questionnaire showed satisfactory reliability, with an overall Cronbach’s α of 0.847.

In the formal study, the overall internal consistency of the instrument was high (Cronbach’s α = 0.9298). The knowledge, attitude, and practice subscales also demonstrated good reliability, with Cronbach’s α values of 0.9303, 0.8192, and 0.8962, respectively. In addition, the Kaiser–Meyer–Olkin measure was 0.9321, indicating excellent sampling adequacy for the questionnaire data.

The questionnaire comprised four sections:1)Demographics: Collected information on age, gender, education, income, and clinical history.2)Knowledge Dimension: Contained 11 items. Correct answers were scored 1, while incorrect or “Not sure” responses were scored 0 (total score: 0–11).3)Attitude Dimension: Included 9 items rated on a 5-point Likert scale from “strongly disagree” (1) to “strongly agree” (5) (total score: 9–45). Attitude Item 4 and Item 5 were negatively worded and were reverse-coded prior to analysis (1↔5, 2↔4, 3 unchanged) so that higher scores consistently indicated a more positive attitude.4)Practice Dimension: Consisted of 12 items rated on a 5-point Likert scale from “never” (1) to “always” (5) (total score: 12–60).

For analysis, KAP scores were converted to percentages, with a score ≥70% considered sufficient.19 Percentage scores were calculated as (raw score/maximum possible score) × 100 for each dimension.

The questionnaire was administered both online via the Wenjuanxing platform and in paper form. Data quality was ensured by excluding questionnaires that were incomplete, completed in under 90 seconds, showed invariant response patterns (ie, selecting the same option for all items), or contained implausible demographic data.

Statistical Analysis

Statistical analyses were conducted using R version 4.3.2, with structural equation modeling performed in Stata 18.0. A two-sided P-value <0.05 was considered statistically significant.

Descriptive statistics were used to summarize participant characteristics and KAP scores. Data were presented as mean ± standard deviation (SD) or median (interquartile range) for continuous variables, and as frequencies and percentages (n, %) for categorical variables. For descriptive and comparative analyses, continuous or multi-category demographic variables were categorized into clinically relevant groups; when adjacent categories had small cell counts, they were combined to ensure stable estimates and to avoid sparse data in subgroup comparisons.

Bivariate analyses were performed to compare KAP scores across different demographic groups using independent-samples t-tests, one-way ANOVA, or their non-parametric equivalents (Wilcoxon–Mann–Whitney U-test, Kruskal–Wallis H-test) as appropriate. Correlations between KAP dimensions were assessed using Spearman’s rank correlation coefficient.

To identify factors associated with KAP levels, multivariable logistic regression models were constructed. For these analyses, KAP scores were dichotomized at the sample median for each dimension and used as dependent variables (knowledge: ≥6 vs ≤5; attitude: ≥37 vs ≤36; practice: ≥45 vs ≤44). Because the median practice score was 45, this approach ensured that all participants were assigned to one of the two groups, with no scores left unclassified.

Variables with P<0.05 in univariate analyses were entered into the multivariable models. Candidate variables were then selected using a backward stepwise procedure based on model fit and statistical significance, and the final models retained variables that remained significant after adjustment. This approach explains why some variables significant in univariate analyses are not shown in the multivariable column in Tables S4 and S5.

Finally, a structural equation model (SEM) was employed to test the hypothesized mediating effect of attitude on the relationship between knowledge and practice. The SEM was estimated using maximum likelihood estimation. Model fit was evaluated using the following indices: root mean square error of approximation (RMSEA) <0.08, standardized root mean square residual (SRMR) <0.08, and both the Tucker–Lewis index (TLI) and comparative fit index (CFI) >0.90.

The chi-square test of model fit was also examined and is reported with the fit indices.

Results

Characteristics of the Participants

A total of 500 questionnaires were collected. After excluding 26 responses for ineligibility (n=6), rapid completion (n=4), uniform responses (n=13), or implausible data (n=3), 474 valid questionnaires were included in the final analysis. The internal consistency of the instrument was high, with a Cronbach’s α of 0.9298 for the overall scale.

The subscales for knowledge, attitude, and practice also demonstrated strong reliability (Cronbach’s α = 0.9303, 0.8192, and 0.8962, respectively). The Kaiser-Meyer-Olkin measure of 0.9321 indicated excellent sampling adequacy.

The majority of the 474 participants were female (80.2%) and of Han ethnicity (92.2%). Most resided in urban areas (67.1%) and held an associate degree or higher (52.3%). Over half (55.7%) had been diagnosed with AITD within the past year.

Additional demographic and clinical characteristics are detailed in Table 1.Table 1Characteristics of the ParticipantsN=474n (%)KnowledgePAttitudePPracticePMean±SDMean±SDMean±SDTotal score474 (100.0)5.79±4.1936.15±4.0845.38±10.67Age (years)0.0010.021<0.001<35149 (31.4)6.85±3.9136.74±3.9848.34±10.8535–4497 (20.5)5.96±4.1436.37±4.5844.13±10.2545–5494 (19.8)5.05±4.2636.21±3.3945.48±10.68≥55134 (28.3)5.00±4.2435.27±4.1542.93±10.06Gender<0.001<0.0010.003Male94 (19.8)4.45±4.3434.49±5.0542.51±11.83Female380 (80.2)6.12±4.0836.56±3.7046.09±10.25Body mass index (kg/m2)0.0310.0380.034<18.519 (4.0)5.05±4.5236.42±3.5847.47±12.4318.5–23.9196 (41.4)6.37±4.0936.76±3.6346.90±10.3623.9–27.9132 (27.8)5.78±4.2435.52±4.3643.90±10.09≥28.0127 (26.8)5.01±4.1335.82±4.4144.26±11.19Place of residence0.168<0.0010.001Rural156 (32.9)5.41±4.1834.97±4.1543.12±10.73Urban318 (67.1)5.97±4.1836.72±3.9346.49±10.48Education level<0.001<0.001<0.001Junior high school or below142 (29.9)4.68±4.2435.00±4.1642.95±10.89Senior high school/ technical secondary school84 (17.7)4.50±4.0535.80±3.5044.24±9.67Associate/Bachelor’s degree or above248 (52.3)6.86±3.9236.92±4.0647.16±10.57Ethnicity0.2080.5400.377Han437 (92.2)5.72±4.2036.11±4.0345.51±10.62Ethnic minority37 (7.8)6.62±4.0136.54±4.7243.89±11.23Employment status<0.0010.093<0.001Employed222 (46.8)6.56±4.0336.57±4.1447.10±10.69Self-employed52 (11.0)3.92±3.8435.56±4.4040.54±9.99Unemployed200 (42.2)5.42±4.2635.83±3.9044.73±10.40Monthly income per capita (RMB)<0.001<0.0010.005<5000206 (43.4)5.02±4.1635.31±4.1243.42±10.655000–10000109 (23.0)6.62±4.0636.94±3.5747.40±10.37>10,00054 (11.4)7.35±3.7237.74±4.3146.91±11.10Prefer not to disclose105 (22.2)5.63±4.2836.14±4.0446.33±10.26Marital status0.7140.0830.044Married383 (80.8)5.75±4.1535.99±4.0744.90±10.60Single/divorced/widowed91 (19.2)5.93±4.3536.81±4.0747.41±10.78Smoking habit<0.001<0.001<0.001Never359 (75.7)6.31±4.0936.60±3.6146.40±10.01Used to38 (8.0)4.89±4.4335.42±5.1645.89±10.98Currently smoke77 (16.2)3.81±3.8734.38±4.9740.38±12.11Drinking alcohol0.001<0.0010.018Never319 (67.3)6.03±4.2036.39±3.7845.96±10.31Used to92 (19.4)6.16±3.8436.58±3.9845.80±10.22Currently drinking63 (13.3)4.00±4.2434.29±5.1341.84±12.46Family members with autoimmune thyroiditis0.0040.1990.431Yes85 (17.9)6.55±3.9136.86±3.8445.02±10.88No264 (55.7)6.02±4.0836.03±4.3645.92±10.67Not sure125 (26.4)4.78±4.4335.91±3.5944.47±10.54Duration of autoimmune thyroiditis<0.0010.0060.197Less than 1 year264 (55.7)5.01±4.2435.68±3.9845.38±10.771–2 years58 (12.2)7.12±4.0436.00±4.1347.53±11.46More than 2 years152 (32.1)6.64±3.8737.01±4.1244.57±10.12Regular physical examinations0.0100.0160.022Yes334 (70.5)6.11±4.1236.44±3.9446.11±10.14No140 (29.5)5.03±4.2535.45±4.3343.64±11.68Abbreviation: SD, standard deviation.

Overall Knowledge, Attitude, and Practice (KAP) Scores

The mean knowledge, attitude, and practice scores were 5.79±4.19 (range 0–11), 36.15±4.08 (range 9–45), and 45.38±10.67 (range 12–60), respectively, indicating insufficient knowledge but sufficient attitudes and practices overall (Table 1).

Item-Level Analysis of KAP Dimensions

Correct response rates were lowest for foundational concepts. Fewer than half of the participants correctly identified the definition of AITD (48.3%) or its etiology (45.2%), with over 50% selecting “not sure” for these items. Knowledge regarding management was comparatively better; 66.9% recognized the need for timely treatment once thyroid dysfunction occurs.

However, only 49.8% understood that early-stage disease often requires monitoring rather than immediate therapy (Table S1).

Participants generally displayed positive attitudes. Over 95% agreed that AITD should be taken seriously, and 92.0% supported monitoring immune indicators even with normal thyroid function. However, some ambivalence was noted: 28.1% agreed or strongly agreed with the fatalistic view that follow-up is unnecessary because the disease is incurable.

Additionally, nearly two-thirds (64.6%) reported anxiety about potential complications (Table S2).

Adherence to physician-directed behaviors was high, with 81.2% reporting regular medical follow-ups. In contrast, proactive self-management was less common. Fewer than two-thirds of participants consistently engaged in emotional regulation (63.9%) or regular physical activity (57.6%).

Active learning was also limited, with only 50.8% frequently seeking information about their condition. Participation in educational activities was low (44.5%) (Table S3).

Correlations

Spearman correlation analysis showed significant positive associations among knowledge, attitude, and practice scores. The correlation coefficient between knowledge and attitude was 0.413 (P<0.001), between knowledge and practice was 0.368 (P<0.001), and between attitude and practice was 0.386 (P<0.001), indicating moderate relationships. These results suggest that higher knowledge is associated with more positive attitudes and better practice behaviors (Table 2).Table 2Correlation AnalysisSpearmanKnowledgeAttitudePracticeKnowledge1.000Attitude0.413 (P<0.001)1.000Practice0.368 (P<0.001)0.386 (P<0.001)1.000

Multivariable Analyses

In multivariable logistic regression for knowledge (high vs low), a monthly income >10,000 RMB (OR=2.724, 95% CI: 1.260–5.890, P=0.011), disease duration >2 years (OR=2.002, 95% CI: 1.263–3.172, P=0.003), and disease duration 1–2 years (OR=2.136, 95% CI: 1.103–4.135, P=0.024) were positively associated with higher knowledge scores. Current smoking (OR=0.481, 95% CI: 0.243–0.953, P=0.036), current drinking (OR=0.364, 95% CI: 0.179–0.743, P=0.006), and being unsure about family history of autoimmune thyroiditis (OR=0.458, 95% CI: 0.245–0.858, P=0.015) were negatively associated with knowledge (Table S4).

For attitude, higher knowledge was independently associated with more positive attitudes (OR=1.185, 95% CI: 1.123–1.252, P<0.001). Female gender (OR=1.908, 95% CI: 1.005–3.624, P=0.048), urban residence (OR=2.596, 95% CI: 1.556–4.332, P<0.001), monthly income >10,000 RMB (OR=3.622, 95% CI: 1.613–8.135, P=0.002), and being single/divorced/widowed (OR=1.804, 95% CI: 1.008–3.229, P=0.047) were also associated with higher attitude scores (Table S5).

For practice, both knowledge (OR=1.111, 95% CI: 1.052–1.173, P<0.001) and attitude (OR=1.150, 95% CI: 1.079–1.225, P<0.001) were independently associated factors of better practice. Self-employment was associated with lower practice scores (OR=0.396, 95% CI: 0.175–0.895, P=0.026) (Table 3).Table 3Practice (Cutoff Based on the Median: ≥45 Vs ≤44)Practice (cutoff based on the median: ≥45 vs ≤44)Univariate analysisMultivariable analysisOR (95%CI)POR (95%CI)PKnowledge1.182 (1.128,1.239)<0.0011.111 (1.052,1.173)<0.001Attitude1.219 (1.153,1.289)<0.0011.150 (1.079,1.225)<0.001Age (years)<3535–440.633 (0.376,1.060)0.0830.679 (0.378,1.218)0.19445–540.647 (0.383,1.089)0.1020.892 (0.481,1.653)0.716≥550.445 (0.275,0.714)0.0010.656 (0.351,1.224)0.185GenderMaleFemale2.082 (1.315,3.338)0.0021.476 (0.793,2.748)0.220Body mass index (kg/m2)<18.518.5–23.90.811 (0.291,2.106)0.67323.9–27.90.391 (0.138,1.037)0.065≥28.00.631 (0.222,1.674)0.365Place of residenceRuralUrban1.756 (1.194,2.593)0.0041.261 (0.764,2.080)0.365Education levelJunior high school or belowSenior high school / technical secondary school1.422 (0.826,2.454)0.2050.988 (0.520,1.879)0.971Associate / bachelor’s degree or above2.170 (1.429,3.317)<0.0010.735 (0.373,1.450)0.374EthnicityHanEthnic minority0.986 (0.502,1.942)0.966Employment statusEmployedSelf-employed0.325 (0.168,0.607)0.0010.396 (0.175,0.895)0.026Unemployed0.605 (0.411,0.890)0.0110.749 (0.441,1.270)0.283Monthly income per capita (RMB)<50005000–100002.268 (1.412,3.683)0.0011.657 (0.944,2.909)0.079>10,0001.986 (1.084,3.707)0.0281.275 (0.613,2.655)0.516Prefer not to disclose1.194 (0.745,1.914)0.4621.078 (0.636,1.829)0.780Marital statusMarriedSingle/divorced/widowed1.552 (0.978,2.487)0.064Smoking habitNeverUsed to1.118 (0.571,2.233)0.7471.865 (0.816,4.261)0.139Currently smoke0.391 (0.229,0.652)<0.0010.757 (0.390,1.468)0.410Drinking alcoholNeverUsed to1.070 (0.672,1.709)0.776Currently drinking0.674 (0.388,1.159)0.156Family members with autoimmune thyroiditisYesNo0.929 (0.567,1.515)0.767Not sure0.808 (0.464,1.403)0.450Duration of autoimmune thyroiditisLess than 1 year1–2 years1.389 (0.783,2.504)0.266More than 2 years0.822 (0.550,1.225)0.336Regular physical examinationsYesNo0.683 (0.459,1.015)0.060Abbreviations: OR, odds ratio; CI, confidence interval.

SEM Analysis

The SEM demonstrated an acceptable model fit (RMSEA=0.058, SRMR=0.056, TLI = 0.897, CFI=0.904). The TLI was slightly below the prespecified threshold, whereas the RMSEA, SRMR, and CFI supported an acceptable fit. The path analysis showed that knowledge had significant direct associations with both attitude (β=0.451, P<0.001) and practice (β=0.238, P<0.001).

Attitude also had a significant association with practice (β=0.411, P<0.001). Furthermore, knowledge showed a significant indirect association with practice, mediated by attitude (β = 0.185, P < 0.001) (Table S6-S7, Table 4 and Figure 1).Figure 1Structural equation model.A diagram showing a structural equation model with paths between knowledge, attitude and practice.The diagram shows a structural equation model with three components: Knowledge, Attitude and Practice, each in an oval. Knowledge is at the top center, with arrows to Attitude (left) and Practice (right), showing direct paths with coefficients of 0.45 and 0.24, respectively.

Attitude, at the bottom left, points to Practice with a coefficient of 0.41. Each component has indicators: Knowledge (k1-k11) with coefficients 0.7-0.79, Attitude (a1-a9) with coefficients 0.8-0.9 and Practice (p1-p6) with coefficients 0.55-0.77. Indicators link to error terms, epsilon 1 to epsilon 35, with values 0.36-0.92.

The diagram visually represents these relationships and error terms.Table 4Path AnalysisModel pathsTotal effectsDirect effectIndirect effectβ (95%CI)Pβ (95%CI)Pβ (95%CI)PAttitudeKnowledge0.451 (0.373, 0.530)<0.0010.451 (0.373, 0.530)<0.001PracticeKnowledge0.423 (0.340, 0.506)<0.0010.238 (0.143, 0.333)<0.0010.185 (0.133, 0.238)<0.001Attitude0.411 (0.320, 0.501)<0.0010.411 (0.320, 0.501)<0.001Abbreviation: CI, confidence interval.

Discussion

This cross-sectional study investigated the knowledge, attitudes, and practices (KAP) of patients with autoimmune thyroiditis (AITD), showing that knowledge was significantly associated with both attitudes and practices, with attitudes partially mediating the knowledge-practice relationship in the SEM. Our findings highlight a critical disconnect: while patients exhibit moderate compliance with physician-led directives, they lack the proactive engagement essential for optimal chronic disease self-management. This suggests that current educational paradigms may foster adherence but fall short of true patient empowerment, a cornerstone of modern chronic care models that prioritize enhancing a patient’s confidence and problem-solving skills over simple instruction-following.21,22

A key finding was the nature of the patients’ knowledge deficit. The most significant gaps were not in treatment protocols but in understanding the fundamental pathophysiology of AITD—specifically, its nature as a chronic, systemic autoimmune process.23 This lack of foundational knowledge likely contributes directly to the psychological burden observed; nearly two-thirds of participants reported anxiety about complications. This anxiety is not merely a reaction to a diagnosis but is increasingly understood to be linked to the underlying autoimmune inflammatory state, which can impair quality of life through persistent symptoms like fatigue and cognitive dysfunction, even in biochemically euthyroid individuals.24 This complex interplay underscores that effective management must address the patient’s understanding of the autoimmune process itself, not just its hormonal consequences.

Comparable patterns have also been reported in other populations. In China, a cross-sectional study of patients with thyroid nodules and thyroid cancer similarly found that knowledge, attitudes, and practices were interrelated, and that sociodemographic differences were associated with KAP variation.25 Outside China, studies from Saudi Arabia have likewise reported suboptimal public knowledge and awareness of thyroid disorders, despite generally favorable perceptions of the importance of thyroid health.26 Taken together, these findings suggest that insufficient disease-related knowledge, coupled with more positive attitudes than actual self-management behaviors, may represent a broader pattern across different settings. Differences in the level and distribution of KAP across studies may reflect variations in health literacy, information sources, public awareness campaigns, and access to specialist endocrine care within different healthcare systems.

The associations between sociodemographic factors and KAP scores align with established literature on health disparities,25,26 identifying vulnerable populations—such as newly diagnosed, rural, or lower-income patients—who require tailored support. The path analysis from our structural equation model provides a useful framework for interpreting the observed relationships among knowledge, attitude, and practice. To bridge the gap from knowledge to practice, educational efforts must evolve.

It is insufficient to merely provide information; interventions must be designed to build self-efficacy—the patient’s confidence in their ability to manage their condition.27 This can be achieved by framing education around the “why” of the disease (the autoimmune mechanism) and then equipping patients with practical, actionable skills. Modern digital health tools are particularly well-suited for this, offering platforms for goal-setting, remote monitoring of disease impact, and facilitating shared decision-making between patients and clinicians, thereby moving beyond passive data collection to active self-management support.28,29 However, these path coefficients should be interpreted as statistical associations rather than causal effects, given the cross-sectional design of the present study.

Building on these findings, future research should employ longitudinal and interventional designs to validate the identified KAP pathways. Key priorities include evaluating whether targeted educational programs—addressing specific knowledge gaps and maladaptive attitudes like fatalism—can produce sustained improvements in KAP scores, symptom burden, and quality of life. It is also critical to investigate barriers to behavior change within high-risk subgroups, such as newly diagnosed or socioeconomically disadvantaged patients.

Furthermore, testing the efficacy and scalability of low-cost delivery models, including digital health platforms and nurse- or peer-led education, is essential for broader implementation. Finally, cross-cultural validation of these KAP-SEM models will be vital for developing context-specific interventions and confirming the generalizability of our findings.

This study has several limitations. The foremost is its constrained generalizability, a consequence of multiple design choices. The research was conducted at a single tertiary hospital in one region of China, using a convenience sample.

This setting likely attracts patients with higher health awareness and better access to specialist care than those in community or primary-care settings. This selection bias was further amplified by stringent exclusion criteria that, while enhancing internal validity, systematically omitted patients with significant comorbidities, psychiatric histories, or those on immunosuppressive therapies. Consequently, our findings may not be applicable to the broader, more clinically complex AITD population encountered in real-world practice.

Moreover, the absence of a formal, hypothesis-driven power calculation means that while the sample size was large, the precision of some estimates may be limited, and the potential for Type II errors in subgroup analyses cannot be dismissed. Second, our reliance on self-reported data from structured questionnaires introduces susceptibility to recall and social desirability biases. The fixed-response format, though advantageous for standardization, may have failed to capture the full nuance of patients’ beliefs and self-management strategies.

This measurement limitation is compounded by the incomplete assessment of crucial confounders, such as health literacy and access to healthcare resources, which could influence the observed relationships between KAP variables and clinical status. In addition, knowledge, attitude, and practice were measured using self-reported questionnaires collected at a single time point, which may introduce common method variance. Although we used quality-control procedures during data collection, common method variance cannot be fully excluded and may have inflated the observed associations among KAP dimensions.

Finally, and most fundamentally, the cross-sectional design precludes any inference of causality. The directional pathways proposed in the SEM, while theoretically grounded, remain correlational. Therefore, longitudinal and experimental studies are imperative to confirm whether improving knowledge can indeed modify attitudes and lead to sustained changes in self-management behaviors and clinical outcomes.

Conclusion

In conclusion, patients with AITD followed at the authors’ Hospital had insufficient knowledge but generally positive attitudes and moderate levels of practice behaviors toward AITD. While most participants endorsed the importance of disease management and follow-up, engagement in proactive behaviors such as active information seeking, systematic symptom recording, and participation in educational activities remained suboptimal.

Interventions that strengthen disease-specific knowledge and address psychological concerns may help reinforce positive attitudes and promote more comprehensive and consistent self-management, thereby potentially contributing to improved clinical outcomes. Future studies should design and test targeted KAP-based interventions for patients with AITD.

Related articles

PLOS

Adversarial robustness of deep learning-based thyroid nodule segmentation in ultrasound

Deep learning-based segmentation models are increasingly integrated into clinical imaging workflows, yet their robustness to adversarial perturbations remains incompletely characterized, particularly for ultrasound images. We evaluated adversarial attacks and inference-time defenses for thyroid nodule segmentation in B-mode ultrasound. Two black-box adversarial attacks were developed: (1) Structured Speckle Amplification Attack (SSAA), which injects boundary-targeted noise, and (2) Frequency-Dom...

PLOS

Frequencies of thyroid-related diagnoses and dyslipidemia among Japanese women with menstrual-related disorders aged under 40 years: A nationwide cross-sectional study using health insurance claims data

Women with endometriosis and dysmenorrhea are increasingly recognized as having systemic comorbidities beyond pelvic disease, but large-scale data on endocrine and metabolic comorbidities across a broad menstrual-related disorders population remain limited. Using the Japan Medical Data Center Claims Database, we conducted a nationwide cross-sectional study of women aged under 40 years enrolled in 2023. Women with confirmed diagnoses of endometriosis (ICD-10: N80) and/or dysmenorrhea (N94.4 or N9...