Structural Equation Modeling of Healthcare Professionals’ Knowledge, Attitudes, and Practices on Electroacupuncture for Post-Stroke Hemiplegia Authors: Mingming Song, Lidan Deng, Yunfei Zhu, Hongting Du, Suqin Guo Published: 2026 ABSTRACT AbstractBackgroundStroke remains a leading cause of disability worldwide, and hemiplegia is a common sequela that greatly affects patients’ quality of life. Electroacupuncture, which combines traditional acupuncture with modern electrotherapy, has been applied in stroke rehabilitation, but its use among healthcare professionals is uneven. This study aimed to assess healthcare professionals’ knowledge, attitudes, and practices (KAP) regarding electroacupuncture in post-stroke hemiplegia management.MethodsA cross-sectional survey was conducted at Wuxi Secondary Chinese Medicine Hospital during 2023–2024 using a validated questionnaire covering demographics and KAP domains.ResultsOf 311 valid responses, 200 participants (64.31%) were physicians, and 188 (60.45%) had prior electroacupuncture experience. Mean knowledge, attitude, and practice scores were 17.77±7.21, 41.05±4.63, and 33.03±11.39, respectively, indicating insufficient knowledge, favorable attitudes, and moderate practice. Structural equation modeling (SEM) analysis revealed that knowledge was positively associated with practice both directly and indirectly through attitude. Notably, the direct associations of knowledge with attitude (β = 0.37 vs β = 0.26) and practice (β = 0.58 vs β = 0.55) were stronger among participants with prior electroacupuncture experience compared to the inexperienced group.ConclusionHealthcare professionals demonstrated insufficient knowledge yet generally favorable attitudes towards electroacupuncture. While knowledge significantly improved attitudes and practices, these relationships were markedly stronger among participants with prior clinical experience. Targeted, competency-based educational interventions combining theory with clinical mentorship must be prioritized to promote evidence-based integration in stroke rehabilitation.Keywords: electroacupuncture, stroke, hemiplegia, healthcare professionals, knowledge, attitude, practice, rehabilitation, cross-sectional study Background Stroke remains one of the leading causes of mortality and disability worldwide, with ischemic stroke accounting for approximately 62.4% of all cases.1 Global statistics reveal approximately 13.7 million new stroke cases annually with over 2.7 million deaths from ischemic stroke.2 Patients with ischemic stroke commonly suffer from motor and sensation dysfunction, which often leads to hemiplegia, causing significant disability and reducing quality of life.3 Electroacupuncture, which integrates traditional Chinese acupuncture with modern electrotherapy, has emerged as a promising adjunctive treatment for post-stroke rehabilitation. A growing body of evidence supports electroacupuncture’s efficacy in enhancing motor function recovery through multiple mechanisms, including promoting neuroplasticity, improving neurovascular coupling, reducing neuroinflammation, and facilitating cerebral blood flow.4 Clinical trials have demonstrated that electroacupuncture can significantly improve muscle contraction, increase recovery as measured by the Orgogozo scale, enhance independence on the Barthel scale, and decrease disability on the Modified Rankin scale in post-stroke hemiplegic patients.3 Furthermore, integrating electroacupuncture into standard rehabilitation protocols has been shown to effectively alleviate muscle spasticity, accelerate the recovery of motor functions, and significantly enhance patients’ overall activities of daily living compared to conventional rehabilitation alone.4 The World Health Organization has recognized and recommended acupuncture as an alternative and complementary strategy for treating ischemic stroke and improving post-stroke sequelae.1 Despite substantial evidence supporting electroacupuncture’s effectiveness, its integration into mainstream rehabilitation protocols remains variable and inconsistent. The Knowledge, Attitude, and Practice (KAP) theory posits that human behavior change progresses through three sequential stages: knowledge acquisition, attitude formation, and practice implementation. This theoretical framework is particularly relevant for understanding healthcare professionals’ adoption of electroacupuncture, as cognitive change from knowledge acquisition must first influence perceptions before translating into clinical behavior.5,6 Previous studies in Western healthcare systems have revealed substantial knowledge gaps and skepticism among clinicians regarding acupuncture therapies, highlighting the critical need for education and training programs.7 In China, where acupuncture has been used for thousands of years, particularly in the treatment of neurological disorders, the landscape of healthcare professionals’ perspectives on electroacupuncture for post-stroke hemiplegia presents a unique research opportunity.8 However, there remains a notable gap in the literature regarding empirical studies that systematically evaluate the knowledge, attitudes, and practices of healthcare professionals in this specific context. Furthermore, previous research has rarely explored how prior clinical experience with electroacupuncture might influence these professional perspectives. This study aimed to comprehensively assess the knowledge, attitudes, and practices of Chinese healthcare professionals regarding electroacupuncture for post-stroke hemiplegia. By assessing knowledge, attitudes, and practices, this study provides evidence to inform educational interventions and policy recommendations for optimizing electroacupuncture in post-stroke rehabilitation. Additionally, this research provided insights into the current level of knowledge, attitudes, and practices among healthcare professionals in one hospital, which may help inform local educational strategies and future studies on electroacupuncture in stroke rehabilitation. Methods Patient and Public Involvement Statement Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research. Study Design and Participants This cross-sectional study was conducted at Wuxi Secondary Chinese Medicine Hospital between November, 2023 and December, 2024, involving healthcare professionals. The study included healthcare professionals directly involved in patient care, such as physicians, clinical pharmacists, nurses, and technicians. Individuals not engaged in clinical work, including hospital administrative and logistics staff, were excluded. Moreover, the participants were required to possess a certain level of comprehension ability and be capable of understanding the questionnaire and providing responses independently. The study protocol was reviewed and approved by the Ethics Committee of Wuxi Secondary Chinese Medicine Hospital (No. 2023WZ01). Informed consent was embedded on the first page of the questionnaire, and participants were required to actively click to confirm their agreement before proceeding to the survey. Questionnaire Introduction The finalized questionnaire version (provided as Supplementary Material 1), administered in Chinese, comprised four sections: demographic characteristics, knowledge, attitudes, and practices. The questionnaire was developed based on an extensive review of previous KAP studies and relevant clinical guidelines related to post-stroke rehabilitation and electroacupuncture. In addition, several KAP studies conducted specifically among healthcare professionals were consulted to ensure that our operationalization of the three constructs (knowledge, attitudes, and practices) aligned with measurement approaches commonly used in professional clinical populations.9,10 To ensure content validity, an expert panel consisting of specialists in rehabilitation medicine, acupuncture, and traditional Chinese medicine reviewed the draft questionnaire for clarity, relevance, and comprehensiveness. Minor revisions were made based on their feedback before conducting the pilot test. To improve conceptual clarity, we explicitly defined the three KAP components according to established KAP theory. Knowledge (K) This dimension assesses cognitive understanding of EA principles, mechanisms, and therapeutic rationale-not technical proficiency. For example, items focus on the physiological effects of EA on muscle tone (eg, “Post-stroke hemiplegia can lead to increased muscle tone... Acupuncture stimulation... reducing muscle tone and alleviating muscle spasms.”) and the impact of treatment parameters (eg, “The effectiveness of electroacupuncture treatment for post-stroke hemiplegia is related to the intensity of the electrical current”). Attitudes (A) This measures evaluative beliefs and perceptions toward EA. Items cover its perceived clinical effectiveness and safety (eg, “I believe electroacupuncture combined with rehabilitation techniques is an effective and safe intervention”), and address perceived structural barriers (eg, “I believe that the learning curve for electroacupuncture therapy is long, making it relatively costly for doctors to learn”). Practice (P) This is defined as self-reported clinical behaviors or behavioral intentions. Items assess actions like “I can develop personalized electroacupuncture treatment plans...” and “I am proficient in operating electroacupuncture devices”. Crucially, we emphasize in the Discussion that Practice scores for the subgroup without prior EA experience must be interpreted as intended behaviors rather than actual performance, consistent with the K-A-P behavioral model. The knowledge section included 13 items, which assessed participants’ conceptual understanding of electroacupuncture mechanisms, therapeutic principles, and indications, rather than their practical proficiency. Each item was rated on a three-point scale: “Very aware” (2 points), “Heard of it” (1 point), and “Unclear” (0 points), yielding a total possible score ranging from 0 to 26. The inclusion of the “Unclear” option was intended to reduce random guessing and capture uncertainty more accurately, consistent with practices in previous KAP study.11 The attitude section consisted of 10 items, assessed using a five-point Likert scale from “strongly disagree” (1 point) to “strongly agree” (5 points), with a total score range of 10 to 50. The practice section included 9 items, also evaluated using a five-point Likert scale, ranging from “never” (1 point) to “always” (5 points), producing a total score range of 9 to 45. Following published KAP studies,12,13 where scores ≥ 80% of the total are classified as “good” or “adequate”. Conversely, scores below this threshold were categorized as insufficient knowledge, negative attitudes, or poor practices. For participants without prior electroacupuncture experience, the practice items were designed to capture intended or potential behaviors rather than actual performance, which enables comparison of behavioral intentions between groups and aligns with the KAP framework commonly used in health behavior studies. A pilot study was conducted in which 48 questionnaires were distributed, 1 case was excluded for refusal to provide informed consent, 1 case was excluded due to logical errors in the responses, indicating inattentive participation, so 46 were deemed valid for analysis. The Cronbach’s α coefficient for the pilot version was 0.949, suggesting excellent internal consistency. In the formal survey, the overall Cronbach’s α reached 0.9577, further confirming the reliability of the instrument. This study was mainly conducted by distributing questionnaires through WeChat Moments forwarding on the Wenjuanxing platform and offline QR code scanning. Statistical Methods Data were analyzed using STATA version 17.0 (StataCorp LLC, College Station, TX, USA). Continuous variables were expressed as means ± SD and categorical variables as frequencies (%). Group comparisons of KAP scores across demographic and professional characteristics were performed using appropriate parametric or non-parametric tests. Logistic regression was applied with dichotomized practice (80% cut-off) as the dependent variable, and knowledge and attitude scores as independent variables. Correlations among KAP scores were assessed using Spearman’s rank correlation, and direct/indirect associations were further examined using structural equation modeling (SEM) stratified by prior electroacupuncture experience. Statistical significance was set at P < 0.05. Results Basic Information on the Population A total of 323 questionnaires were collected, of which 311 were valid, yielding a response rate of 96.28%. Participants included a majority of females (59.16%), physicians (64.31%), and those aged ≤ 35 years (51.45%). Overall mean scores were 17.77±7.21 for knowledge, 41.05±4.63 for attitude, and 33.03±11.39 for practice. Differences in KAP scores were observed across education, profession, years of stroke-related experience, medical background, hospital type, and prior use of electroacupuncture (all P < 0.001). Further details are presented in Table 1. Table 1Demographic Characteristics and KAP ScoresVariablesN (%)Knowledge ScoreAttitude ScorePractice ScoreMean ± SDPMean ± SDPMean ± SDPTotal scoreN=31117.77±7.2141.05±4.6333.03±11.39Gender0.1960.1290.115Male127(40.84)18.26±7.5341.40±4.8433.77±11.80Female184(59.16)17.41±6.9740.80±4.4732.51±11.00Age (years old) [range:20~89]0.9300.0590.792≤35160(51.45)17.69±6.9740.60±4.7933.15±10.70>35151(48.55)17.84±7.4741.51±4.4132.88±12.00Education<0.001<0.001<0.001Bachelor’s degree or below165(53.05)15.56±7.0339.71±4.5529.20±11.70Master’s degree or above146(46.95)20.25±6.5842.55±4.2437.34±9.26Profession as healthcare worker<0.001<0.001<0.001Physician200(64.31)19.37±6.9341.84±4.3835.76±10.30Nurse55(17.68)14.38±7.6239.21±4.6026.92±11.70Medical technician56(18.01)15.33±5.8940.01±4.8429.25±11.40Professional title0.1310.0870.082Junior67(21.54)16.22±7.7940.92±5.0430.88±12.30Intermediate130(41.8)18.68±6.5741.00±4.1933.55±11.10Associate senior67(21.54)17.44±7.6141.61±4.5132.64±11.80Senior25(8.04)19.72±7.8142.2±4.9737.36±11.30No title22(7.07)15.77±6.0538.63±5.2232.68±6.67Years of work experience0.4330.4320.058≤5 years66(21.22)17.71±7.0941.01±5.5633.68±10.405–10 years86(27.65)17.87±6.9940.67±4.1032.50±11.2011–15 years78(25.08)18.88±6.8241.80±4.3035.60±10.30≥16 years81(26.05)16.61±7.8040.74±4.6130.56±12.70Years of experience in stroke-related diagnosis<0.001<0.001<0.001≤5 years139(44.69)15.15±7.5339.87±4.9030.28±11.805–10 years81(26.05)18.37±6.1141.35±4.2933.09±10.9≥10 years91(29.26)21.20±5.9842.56±3.9937.15±9.78Current department0.9110.4560.387Non-Rehabilitation180(57.88)17.72±7.5741.17±4.6433.25±11.70Rehabilitation131(42.12)17.82±6.7040.87±4.6132.70±10.80Medical background<0.001<0.001<0.001Traditional Chinese medicine136(43.73)20.75±5.8142.87±4.4238.77±6.93Western medicine93(29.9)13.98±7.1139.03±4.0225.98±12.00Integrated Chinese and Western Medicine62(19.94)17.87±7.0240.37±4.0232.91±11.20Other20(6.43)14.65±7.7240.1±5.8026.95±11.90Type of hospital<0.001<0.001<0.001Other159(51.13)15.30±6.8539.59±4.2429.14±11.40Tertiary public hospital152(48.87)20.33±6.6742.57±4.5337.08±9.81Previous use of electroacupuncture<0.001<0.001<0.001Yes188(60.45)21.14±5.5842.40±4.2639.07±6.74No123(39.55)12.59±6.2938.97±4.3923.78±10.80Number of cases per year in use electroacupuncture for post-stroke hemiplegia (for healthcare professionals who answered “Yes” in Question 11):<0.001<0.001<0.001Less than 10 cases155(49.84)14.03±6.8739.34±4.5926.85±11.9010–30 cases32(10.29)19.78±5.6042.09±3.8538.06±5.4130–80 cases35(11.25)20.40±5.9641.11±3.1638.08±7.17More than 80 cases89(28.62)22.50±4.8843.61±4.1439.97±6.44Abbreviation: SD, means standard deviation. Demographic Characteristics and KAP Scores Abbreviation: SD, means standard deviation. Distribution of Responses to Knowledge, Attitude, and Practice The distribution of knowledge dimensions showed that the three questions with the highest number of participants choosing the “Unclear” option were “The effectiveness of electroacupuncture treatment for post-stroke hemiplegia is related to the duration of needle retention”. (K13) with 16.72%, “The effectiveness of electroacupuncture treatment for post-stroke hemiplegia is related to the intensity of the electrical current”. (K11) with 12.86%, and High-frequency electroacupuncture stimulation at tendon acupuncture points in spastic muscle groups can repeatedly stimulate peripheral receptors, regulate the excitability of neurons along the neural conduction pathway, reduce the excitability of spastic muscles, and enhance the excitability of hypotonic muscles. (K7) with 12.22% (Supplementary Table S1). Responses to the attitude dimension showed that 21.86% strongly agreed and 27.65% agreed that the learning curve for electroacupuncture therapy is long, making it relatively costly for doctors to learn (A8), 20.9% strongly agreed and 26.7% agreed that electroacupuncture therapy requires more time and longer treatment courses than Western medicine treatments (A9) (Supplementary Table S2). Responses to the practice dimension reflected actual behaviors among participants with prior electroacupuncture experience, while for those without experience, the items captured potential or intended behaviors. Specifically, 18.65% reported never being proficient in acupuncture techniques (P6), 15.76% never developing personalized electroacupuncture treatment plans and courses based on patients’ physical conditions (P1), and 15.11% never being proficient in operating electroacupuncture devices (P3) (Supplementary Table S3). Univariate and Multivariate Analysis for Practice The 80% of the highest scores for the practice dimensions were used as the cut-off value for each dimension to divided the groups, and the number of participants below the cut-off value were 168 (54.02%). Multivariate logistic regression showed that knowledge score (OR = 1.19, 95% CI: [1.11, 1.28], P < 0.001), attitude score (OR = 1.21, 95% CI: [1.10, 1.34], P < 0.001), and never use the electroacupuncture (OR = 0.12, 95% CI: [0.04, 0.39], P < 0.001) were independently associated with practice (Table 2). Table 2Univariate and Multivariate Logistic Regression Analysis for Practice DimensionVariablesUnivariate AnalysisMultivariate AnalysisOR (95% CI)POR (95% CI)PKnowledge score1.30(1.23,1.38)<0.0011.19(1.11,1.28)<0.001Attitude score1.34(1.25,1.44)<0.0011.21(1.10,1.34)<0.001GenderMaleFemale0.53(0.33,0.84)0.0070.62(0.27,1.40)0.254Age (years old) [range:20~89]≤35>351.40(0.89,2.20)0.135EducationBachelor’s degree or belowMaster’s degree or above4.82(2.98,7.79)<0.0011.27(0.42,3.84)0.669Profession as healthcare workerPhysicianNurse0.15(0.07,0.32)<0.0011.01(0.20,5.02)0.987Medical technician0.25(0.13,0.48)<0.0013.41(0.75,15.5)0.112Professional titleJuniorIntermediate2.04(1.10,3.78)0.0231.58(0.54,4.61)0.402Associate senior2.10(1.04,4.24)0.0373.46(0.87,13.6)0.076Senior5.25(1.91,14.4)0.0014.82(0.77,29.9)0.091No title0.45(0.13,1.50)0.1970.48(0.07,2.92)0.426Years of work experience≤5 years5–10 years1.14(0.59,2.18)0.68611–15 years1.77(0.91,3.44)0.090≥16 years1.09(0.56,2.12)0.779Years of experience in stroke-related diagnosis≤5 years5–10 years1.70(0.96,2.98)0.0641.01(0.36,2.84)0.978≥10 years3.91(2.23,6.84)<0.0010.66(0.20,2.09)0.481Current departmentNon-RehabilitationRehabilitation0.57(0.36,0.91)0.0190.49(0.19,1.25)0.137Medical backgroundTraditional Chinese medicineWestern medicine0.14(0.07,0.25)<0.0010.71(0.23,2.16)0.558Integrated Chinese and Western Medicine0.44(0.24,0.82)0.0102.27(0.75,6.80)0.143Other0.12(0.04,0.40)<0.0010.48(0.08,2.60)0.396Type of hospitalOtherTertiary public hospital4.59(2.84,7.42)<0.0011.32(0.55,3.15)0.530Previous use of electroacupunctureYesNo0.04(0.02,0.09)<0.0010.12(0.04,0.39)<0.001Number of cases per year in use electroacupuncture for post-stroke hemiplegia (for healthcare professionals who answered “Yes” in Question 11):Less than 10 cases10–30 cases5.93(2.63,13.3)<0.0010.89(0.27,2.93)0.86130–80 cases7.76(3.45,17.4)<0.0011.94(0.50,7.49)0.334More than 80 cases9.63(5.27,17.6)<0.0010.76(0.23,2.44)0.653Notes: The practice dimension was determined based on a cut - off value of 80% of the full score. Among them, 168 participants (accounting for 54.02%) scored less than or equal to 80%, and 143 participants (accounting for 45.98%) scored more than 80%. Univariate and Multivariate Logistic Regression Analysis for Practice Dimension Notes: The practice dimension was determined based on a cut - off value of 80% of the full score. Among them, 168 participants (accounting for 54.02%) scored less than or equal to 80%, and 143 participants (accounting for 45.98%) scored more than 80%. Correlations Between KAP Among participants with prior experience in using electroacupuncture for post-stroke hemiplegia, significant positive correlations were observed between knowledge and attitude (r = 0.4158, P < 0.001), knowledge and practice (r = 0.6249, P < 0.001), and attitude and practice (r = 0.4567, P < 0.001). In contrast, among those without such experience, the correlations followed a similar trend but were weaker, see Table 3 for more details. Table 3Correlation AnalysisTotalKnowledgeAttitudePracticeKnowledge1Attitude0.543 (P<0.001)1Practice0.708 (P<0.001)0.5394 (P<0.001)1Previously used electroacupuncture in the treatment of post-stroke hemiplegiaKnowledge1Attitude0.416 (P<0.001)1Practice0.625 (P<0.001)0.4567 (P<0.001)1Previously not used electroacupuncture in the treatment of post-stroke hemiplegiaKnowledge1Attitude0.361 (P<0.001)1Practice0.413 (P<0.001)0.2797 (P= 0.0017)1 Correlation Analysis SEM Analysis The fit of the SEM model yielded good indices demonstrating good model fit (Supplementary Table S4). SEM analysis showed that knowledge directly affected attitude (β = 0.37, P < 0.001) and practice (β = 0.58, P < 0.001), attitude directly affected practice (β = 0.35, P = 0.001), and knowledge indirectly affected practice through attitude (β = 0.13, P = 0.002) in those patients who previously used electroacupuncture in the treatment of post-stroke hemiplegia (Table 4 and Figure 1A), and knowledge also had direct effects on attitude (β = 0.26, P < 0.001) and practice (β = 0.55, P < 0.001), attitude had a direct effect on practice (β = 0.48, P = 0.024), as well as knowledge had an indirect effect on practice (β = 0.12, P = 0.044) in those patients who previously did not used electroacupuncture in the treatment of post-stroke hemiplegia (Table 4 and Figure 1B). Table 4Structural Equation Modeling Results Stratified by Prior Electroacupuncture ExperienceModel pathsPredictorTotal effectsDirect EffectIndirect effectβ (95% CI)Pβ (95% CI)Pβ (95% CI)PParticipants with prior electroacupuncture experienceAttitudeKnowledge0.37(0.28,0.47)<0.0010.37(0.28,0.47)<0.001PracticeAttitude0.35(0.14,0.55)0.0010.35(0.14,0.55)0.001PracticeKnowledge0.71(0.57,0.85)0.0020.58(0.42,0.73)<0.0010.13(0.04,0.21)0.002Participants without prior electroacupuncture experienceAttitudeKnowledge0.26(0.15,0.37)<0.0010.26(0.15,0.37)<0.001PracticeAttitude0.48(0.06,0.90)0.0240.48(0.06,0.90)0.024PracticeKnowledge0.68(0.40,0.96)<0.0010.55(0.25,0.85)<0.0010.12(0.00,0.25)0.044 Structural Equation Modeling Results Stratified by Prior Electroacupuncture Experience Structural equation models illustrating the relationships among knowledge, attitudes, and practices. (A) Participants with prior experience using electroacupuncture for post-stroke hemiplegia; (B) Participants without such experience. Discussion This study showed that healthcare professionals exhibited insufficient cognitive understanding of electroacupuncture (knowledge), yet expressed largely positive perceptions and willingness to support its use (attitudes), and demonstrated moderate levels of actual or intended behaviors related to electroacupuncture (practice). Targeted training and continuous professional education on electroacupuncture may enhance evidence-based practice and improve the integration of complementary therapies in post-stroke rehabilitation.14,15 These distinctions were explicitly incorporated into our interpretation of findings, with knowledge treated as the cognitive foundation, attitudes as the evaluative mediator, and practice as the behavioral outcome, consistent with KAP theory. While attitudes toward electroacupuncture were generally positive, significant deficiencies in knowledge and uneven patterns of clinical practice were evident. These findings, when viewed through the lens of broader clinical and educational systems, reflect a gap between theoretical endorsement and practical application, a phenomenon also noted in previous investigations on the implementation of complementary therapies in neurological rehabilitation.14,15 Previous surveys and anecdotal reports have indicated that the clinical implementation of electroacupuncture in stroke rehabilitation remains limited and inconsistent. For example, studies have shown that although electroacupuncture is widely recognized as beneficial, its actual integration into standard rehabilitation programs is highly variable, often depending on institutional resources, regional traditions, and the availability of trained practitioners.8,16,17 This variability provides an important baseline for understanding the current state of practice and highlights the need for more structured strategies to improve implementation.18,19 Beyond its established role in stroke recovery, the clinical significance of electroacupuncture is further underscored by its proven benefits in managing a variety of other medical conditions. For instance, electroacupuncture has demonstrated substantial efficacy in treating chronic musculoskeletal pain,15 alleviating inflammatory pain,20 and improving symptoms of osteoarthritis.21 These broad therapeutic capabilities highlight its versatility and reinforce the strong rationale for firmly integrating electroacupuncture as a vital component of comprehensive rehabilitation processes. Knowledge emerged as a foundational determinant in shaping both attitudes and behaviors. Participants who had prior experience with electroacupuncture consistently demonstrated a more comprehensive understanding of its physiological mechanisms, therapeutic indications, and procedural considerations. In contrast, those without prior exposure showed widespread uncertainty or misconceptions across knowledge domains, particularly on technical details such as the effects of stimulation frequency, duration, and specific acupuncture points. This pattern aligns with earlier studies which have suggested that clinical exposure enhances both conceptual clarity and procedural familiarity, especially in interventions that require nuanced decision-making and hands-on skill.18,19 However, even among experienced professionals, misunderstandings persisted, suggesting that experience alone may not fully substitute for systematic training. The prevalence of incorrect responses to fundamental knowledge items points to a need for more structured and standardized educational content. These findings mirror concerns raised in previous evaluations of integrative medicine training programs, which often lack consistency in content depth and delivery modes across institutions.22,23 To improve knowledge, targeted educational interventions should be embedded within continuing professional development structures, ideally incorporating modular, competency-based training on electroacupuncture for stroke rehabilitation that combines theory with simulation or supervised clinical practice. Attitudes toward electroacupuncture were generally supportive across the sample, but these positive perceptions were more pronounced among those with clinical experience. Respondents with experience were more likely to agree with statements affirming electroacupuncture’s effectiveness, practicality, and contribution to long-term functional recovery. This group also reported more confidence in combining electroacupuncture with rehabilitation techniques, viewing the integration as both safe and beneficial. Similar patterns have been noted in studies examining clinician perceptions of acupuncture-based interventions in neurorehabilitation, where perceived efficacy was closely linked to familiarity and institutional endorsement.7,20,24 Nonetheless, a notable proportion of respondents, particularly those without prior experience, expressed reservations about the learning curve and perceived time burden of electroacupuncture. These concerns echo broader structural barriers commonly reported in studies on the integration of non-pharmacological therapies, including limited training opportunities, time constraints in clinical workflows, and inconsistent institutional support.16,17 These attitudinal barriers may not necessarily reflect resistance to the intervention itself but rather pragmatic concerns about feasibility in real-world settings. Addressing such issues requires not only improving educational access but also reducing implementation burdens. For example, clinical departments could establish electroacupuncture mentorship programs, integrate its use into standardized treatment pathways for stroke, and ensure that device operation and safety protocols are clear and accessible. Through institutional support and workflow integration, the gap between positive attitudes and practical application could be meaningfully narrowed. Practice behaviors related to electroacupuncture varied markedly across the two groups. Professionals with experience consistently reported greater engagement in essential practices such as individualized treatment planning, precise point localization, and regular consultation of literature or clinical guidelines. In contrast, those without experience frequently reported limited or no participation in these practices. This discrepancy is consistent with prior research indicating that direct clinical exposure tends to foster greater procedural confidence and willingness to adopt non-routine therapeutic modalities.21 While correlation and structural equation modeling both confirmed that knowledge and attitudes significantly influenced practice, the strength of these relationships was clearly moderated by whether the individual had real-world exposure. The group without prior experience demonstrated weaker linkages across KAP dimensions, suggesting that without experiential learning, even those with favorable attitudes may struggle to translate belief into action. Moreover, practical deficiencies were particularly evident in competencies such as device operation and accurate control of stimulation parameters, reinforcing the need for hands-on training. These findings are aligned with reports from rehabilitation departments in comparable healthcare settings, where the implementation of traditional techniques often lags due to insufficient operational familiarity among staff.25 The broader healthcare delivery environment also appears to play a role. Participants from tertiary public hospitals consistently demonstrated stronger knowledge and more active practice behaviors, possibly reflecting greater access to training resources, institutional support for integrative approaches, and interdisciplinary collaboration. These systemic advantages may account for some of the variation in scores between institutions, a trend previously observed in cross-institutional studies of complementary and alternative medicine implementation in neurological settings.26,27 Similarly, professionals with a background in Traditional Chinese Medicine performed better across all KAP domains, which likely reflects curricular differences and clinical orientation. The interconnection between knowledge, attitudes, and practices, stronger in experienced individuals, supports the underlying assumptions of the KAP model, yet the differential strengths of these relationships suggest that experience moderates the extent to which knowledge and attitude translate into action. This has implications for how future training and policy strategies are developed. It is not sufficient to simply disseminate information or foster favorable perceptions. Without practical opportunities to observe, practice, and receive feedback, the impact of knowledge and attitude may be limited. Therefore, implementation strategies should focus on embedding electroacupuncture into routine clinical workflows through institutional protocols, supervision systems, and quality assurance frameworks. Health systems could also establish certification requirements that mandate electroacupuncture competencies for professionals working in stroke rehabilitation units, ensuring accountability and consistency.7,27 Despite the valuable insights provided, this study has several limitations. First, the cross-sectional design prevents the establishment of causal relationships among knowledge, attitudes, and practices, capturing only a snapshot of these variables at a single point in time. Second, as the survey was conducted in a single hospital, the generalizability of the findings to other settings or regions may be restricted. Third, the reliance on self-reported questionnaires inherently introduces response bias, such as social desirability, potentially affecting the objective accuracy of the reported knowledge and practice behaviors. In participants without prior electroacupuncture experience, practice scores should be interpreted as indicative of potential or intended behaviors rather than actual clinical performance. Finally, as the survey was conducted entirely online, there may be risks of sampling bias and biased responses. Furthermore, although our KAP framework followed established principles, we acknowledge that variations exist across KAP studies in healthcare settings, and our operational definitions may differ from those used in some previous professional populations. To mitigate these limitations, we adopted a validated questionnaire with high internal consistency, applied consistency checks to exclude invalid responses, and achieved a high response rate, which help preserve the robustness of the findings despite these constraints. Conclusions In the current study, healthcare professionals demonstrated insufficient knowledge yet generally favorable attitudes and active engagement in the use of electroacupuncture for post-stroke hemiplegia. Significant associations were observed among knowledge, attitudes, and practices, though these relationships were markedly stronger in those with prior clinical experience, indicating that hands-on exposure is critical for translating knowledge into practice. Given that knowledge is the foundational driver of behavioral change, and as demonstrated by the significantly stronger KAP pathway coefficients in experienced clinicians (β = 0.37 vs 0.26 for the knowledge-attitude path; β = 0.58 vs 0.55 for the knowledge-practice path), targeted educational interventions must be a top priority. To maximize clinical integration, future training policies should mandate competency-based educational programs that combine theoretical instruction with structured clinical mentorship, thereby promoting the evidence-based application of electroacupuncture in post-stroke rehabilitation. Funding Statement This research was supported by an institutional project titled “Precise Acupuncture of the Gallbladder Meridian on the Lower Extremity Ankle Joint of Hemiplegic Patients after Stroke” (Project No. ZYYB11) from Wuxi No. 2 Chinese Medicine Hospital. Abbreviations KAP, Knowledge, Attitudes, and Practices; EA, Electroacupuncture; SEM, Structural Equation Modeling; SD, Standard Deviation; OR, Odds Ratio; CI, Confidence Interval. Data Sharing Statement The datasets generated and analyzed during the current study contain sensitive information regarding healthcare professionals and cannot be publicly shared due to privacy protection and the confidentiality commitment made to participants prior to data collection. These data include individual-level responses that could potentially allow identification of medical staff, and therefore cannot be deposited in a public repository. Requests for access to the anonymized data for reasonable academic purposes may be submitted to the corresponding authors, Dr. Mingming Song (727004985@qq.com) or Dr. Suqin Guo (80504724@qq.com), and will be considered in accordance with institutional and ethical requirements. Ethics Approval and Consent to Participate All procedures were performed in accordance with the ethical standards of the Declaration of Helsinki (2013 version) and its later amendments. The study protocol was reviewed and approved by the Ethics Committee of Wuxi Secondary Chinese Medicine Hospital (No. 2023WZ01). Electronic informed consent was obtained from all included participants through the online questionnaire system, where participants were required to read the consent statement and confirm their agreement before proceeding. Author Contributions Mingming Song and Lidan Deng are co-first authors. All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors declare that they have no competing interests. References Associated Data Data Availability Statement The datasets generated and analyzed during the current study contain sensitive information regarding healthcare professionals and cannot be publicly shared due to privacy protection and the confidentiality commitment made to participants prior to data collection. These data include individual-level responses that could potentially allow identification of medical staff, and therefore cannot be deposited in a public repository. Requests for access to the anonymized data for reasonable academic purposes may be submitted to the corresponding authors, Dr. Mingming Song (727004985@qq.com) or Dr. Suqin Guo (80504724@qq.com), and will be considered in accordance with institutional and ethical requirements.