Analysis of the current status and correlates of knowledge-attitude-practice (KAP) regarding cardiovascular and cerebrovascular diseases (CCVD) in adolescents and young adults Authors: Tongyan Sun, Hua Zhang Published: 2026 ABSTRACT AbstractObjectiveTo investigate knowledge-attitude-practice (KAP) regarding cardiovascular and cerebrovascular diseases (CCVD) among adolescents and young adults, and to identify associated factors to provide evidence for targeted prevention.MethodsA cross-sectional survey was conducted through digital platforms from August to October 2025, recruiting 636 participants from North, East and Southwest China through convenience sampling. Data were collected via a self-designed CCVD KAP questionnaire, S-CVI was 0.93, and Cronbach's α was 0.92. The data were analyzed using SPSS 27.0.ResultsThe overall KAP score was (74.90 ± 20.83), with a scoring rate 63.47%. Among the three dimensions, the attitude subscale scored the highest, followed by knowledge and practice. Multiple linear regression analysis showed that the total score positively correlated with female gender, better self-rated health, and overweight; negatively with East and Southwest China residence, higher-maternal education, and unknown family history. For knowledge and attitude, exercise and sleep were positively correlated, smoking and drinking negatively. Higher education positively correlated with knowledge and attitude but negatively with practice.ConclusionAwareness of CCVD was moderate, with positive attitude but suboptimal behavioral adherence. These findings support targeted interventions to address the knowledge-practice gap, particularly among those with low education, unhealthy lifestyles, or uncertain family history.Keywords: Adolescents, Young adults, Cardiovascular and cerebrovascular diseases (CCVD), Knowledge-attitude-practice (KAP) Introduction Cardiovascular and cerebrovascular diseases (CCVD), including hypertension, coronary heart disease, and stroke, are the leading cause of global mortality (Martin et al., 2025; Palaniappan et al., 2026) and the predominant cause of death in both urban and rural China (National Center for Cardiovascular Diseases, 2024). The rising younger-onset CCVD trend is concerning: 18.37% of adolescents exhibit clustering of cardiovascular metabolic risk factors (Wang et al., 2023), and over 10% of myocardial infarctions in China occur in those aged 20–35 years, corroborated by expert consensus (Chinese Society of Cardiology, 2025). The global prevalence of pediatric hypertension has nearly doubled over the past two decades to 4.28% (Ruan et al., 2025). This trend is clinically significant, as adolescent hypertension tracks into adulthood and promotes early target organ damage, notably left ventricular hypertrophy and arterial stiffening (Khoury and Urbina, 2021). Given the increasing prevalence of subclinical pediatric cardiovascular disease, timely early intervention, screening, and health education are imperative (Wang, 2025). The Healthy China Initiative (2019–2030) has set a target of reducing CCVD mortality to 190.70 per 100,000 by 2030 (Healthy China Action Promotion Committee, 2019), necessitating the extension of prevention strategies from high-risk adults to adolescents and young adults. Despite growing research attention, the evidence base remains limited, with existing studies being relatively sparse and heterogeneous in geographic and age coverage (Sun et al., 2023), mostly confined to localized or narrow age range samples. Moreover, although young individuals generally hold favorable attitudes toward CCVD prevention, prior studies have revealed persistent gaps in factual knowledge and behavioral adherence. Accordingly, this study applies the KAP framework to systematically characterize CCVD-related knowledge, attitudes, and practices among adolescents and young adults, and to comprehensively analyze their multifaceted associated factors. Methods Study design and population A cross-sectional survey was conducted from August to October 2025 via the Questionnaire Star platform, with QR codes and links distributed through WeChat and QQ. A total of 661 participants were recruited from North, East, and Southwest China using convenience sampling. After excluding 25 invalid responses, 636 valid questionnaires were retained (effective rate: 96.22%). Inclusion criteria were as follows: age ≤ 35 years, voluntary participation, and the ability to complete the survey independently. Exclusion criteria comprised severe physical or mental illness or CCVD-related intervention participation within the past three months. This study was approved by the Ethics Committee of Tianjin Medical University (approval No. TMUHMEC20250007). Electronic informed consent was obtained from all participants or their legal guardians prior to data collection. For participants under 18 years of age, additional parental consent and adolescent assent were also secured. Measures Demographic data encompassed 16 variables: age, gender, education level, only child status, residential status, residential region, father's education level, mother's education level, father's occupational type, mother's occupational type, household monthly income, household size, family history of CCVD, self-rated health status, height, and weight. Education level was classified as lower (junior high school or below), medium (senior high school, technical secondary school, or junior college diploma), or higher (bachelor's degree or above). Household size was categorized as small (1–2 members), medium (3–4 members), or large (5 members or more). Household monthly income was divided into low (<2600 CNY), medium (2600–5000 CNY), upper-middle (5000–8000 CNY), and high (>8000 CNY) categories. Family history of CCVD was recorded as yes, no, or unknown (participants unable to report whether a family history of CCVD existed). Body mass index (BMI), derived from self-reported height and weight, served as an objective anthropometric measure and was not incorporated into the KAP scale, which is designed to assess subjective, self-reported constructs. BMI was classified as underweight (<18.50 kg/m2), normal weight (18.50–23.90 kg/m2), overweight (24.00–27.90 kg/m2), or obese (≥28.00 kg/m2). Additional lifestyle variables included adequate sleep (≥7 h/day) and regular exercise (≥3 times/week). The questionnaire was based on the KAP model, comprising 41 items across three dimensions: knowledge (14 items), attitude (15 items), and practice (12 items), using single-choice and multiple-choice questions. For the knowledge dimension one point was assigned for each correct answer and zero otherwise. The attitude dimension was assessed using a Likert scale with predefined scoring criteria. For the practice dimension single-choice items were scored on item-specific ordinal scales with varying ranges while multiple-choice items were scored by summing the number of correct options selected. The maximum scores for the knowledge, attitude, and practice dimensions were 48, 34, and 37, respectively. The scoring rate for each dimension was calculated as the mean actual score divided by the maximum possible score multiplied by 100%. The questionnaire was developed through Chinese and international literature review based on the KAP framework, expert consultation, and pre-survey. Content validity was evaluated by a panel of six experts (two attending physicians, one epidemiologist, and three senior specialty nurses), all holding senior titles. Using a single-round Delphi process, the experts rated item representativeness of each item on a 4-point scale. The I-CVI ranged from 0.83 to 1.00, and the S-CVI was 0.93, both exceeding recommended thresholds and confirming satisfactory content validity. A pre-survey was conducted among thirty adolescents and young adults. Exploratory factor analysis (EFA) was performed on the final sample (N = 636) using Principal Axis Factoring (PAF) with Promax rotation. A forced three-factor solution was prespecified per the KAP framework, with scree-plot inspection as supplementary support, accounting for 36.50% of the total variance and aligning with the theoretical dimensions. Although below the 50% threshold, this conservative estimate reflects PAF's extraction of only common variance and our theory-driven retention prioritizing conceptual coherence over statistical optimization. The solution demonstrated clean loadings with no cross-loadings, supported by strong sampling adequacy (KMO = 0.93; Bartlett's P < 0.01). Item loadings ranged from 0.45 to 0.70, confirming structural validity. Of the 41 original items, 13 multiple-choice items were excluded due to violation of the continuous-data assumption, leaving 28 items for EFA with no further eliminations; these 13 items were treated as categorical covariates in subsequent analyses. Cronbach's α was 0.92 overall, with subscale values of 0.86 (knowledge), 0.87 (attitude), and 0.72 (practice), all exceeding 0.70. Statistical methods Qualitative data were described using the number of cases, and comparisons between groups were performed using t-tests, one-way ANOVA. Multiple linear regression analysis was used to explore associated factors, the significance level α = 0.05. Variable selection was guided by the KAP theoretical framework and prior literature (Yang et al., 2024; Li et al., 2021; Jacobs Jr et al., 2022; Zhang et al., 2021; Chen and Wang, 2006; Lim et al., 2021). Variables were selected across the following domains: demographic characteristics (age, gender, only child), socioeconomic factors (education, parental education, parental occupation, household income), family structure (household size), geographic factors (residential region, urban-rural status), health status (BMI, self-rated health), health behaviors (exercise, sleep, smoking, drinking), and family history of CCVD. All variables were included based on their documented associations with health literacy and KAP in previous studies. Multicollinearity was assessed using the variance inflation factor (VIF), and residual independence was examined via the Durbin-Watson statistic. All VIF values were below the recommended threshold of 5, and Durbin-Watson statistics ranged from 1.69 to 1.93, suggesting no significant autocorrelation. Normality, homoscedasticity, influential observations, and linearity were evaluated using P—P plots, scatterplots of standardized residuals versus predicted values, Cook's distance, and partial regression plots, respectively, with all assumptions being satisfactorily met. Given the exploratory design aiming to generate preliminary hypotheses, corrections for multiple comparisons were not implemented. The observed associations in this exploratory study are hypothesis-generating and require confirmation in future research. All statistical analyses were performed using SPSS Statistics for Windows, version 27.0 (IBM Corp., Armonk, NY, USA). Results Basic characteristics of the research subjects Among the respondents, 70.59% were aged 18–35 years, 57.55% were female. Characteristics of the study participants are presented in Table 1, Table 2. Table 1Distribution of demographic characteristics among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariableCategoryFrequencyProportion (%)Age<18 years old18729.4018–35 years old44970.59Gendermale27042.45female36657.55Education levellower educational level16325.56medium educational level16425.78higher educational level30948.58Only childyes28845.28no34854.72Residential statusurban37859.43rural25840.57Residential regionNorth China20131.60East China21734.12Southwest China21834.28Father's education leveljunior high school and below15524.37senior high school or technical secondary school15223.90college diploma13721.54bachelor's degree and above19230.19Mother's education leveljunior high school and below15424.21senior high school or technical secondary school18629.25college diploma10616.67bachelor's degree and above19029.87Father's occupational typemanual labor34554.25mental labor26141.04others304.72Mother's occupational typemanual labor29346.07mental labor30648.11others375.82Household monthly incomelow income304.72medium income8012.58upper-middle income15925.00high income36757.70Household sizesmall household6510.22medium household44069.18large household13120.60Family historyyes18128.46no25640.25unknown19931.29Total636100.00 Distribution of demographic characteristics among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. Table 2Distribution of lifestyle-related behaviors among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariableCategoryFrequencyProportion (%)Health statusvery healthy21633.96relatively healthy19730.97general health8813.84relatively unhealthy8413.21poor health518.02Body mass indexunderweight17327.20normal weight38159.91overweight436.76obese396.13Exercise behaviornever exercise589.121–2 times/week17627.673–4 times/week17627.67≥5 times/week22635.53Smoking behavior≥4 times/week8613.52<4 times/week12119.03never smoke42967.45Drinking behavior≥3 times/week8212.89<3 times/week19130.03never drink alcohol36357.08Sleep behaviormore than 9 h16626.107–8 h27843.716–7 h11818.55less than 6 h7411.64total636100.00 Distribution of lifestyle-related behaviors among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. KAP scores for CCVD The overall KAP score regarding CCVD was (74.90 ± 20.83) points, with a scoring rate of 63.47%. The score for the knowledge dimension was (29.26 ± 9.99) points, with a scoring rate of 60.96%. The score for the attitude dimension was (23.51 ± 8.23) points, with a scoring rate of 69.15%. And the score for the practice dimension was (22.13 ± 5.74) points, with a scoring rate of 59.81%. Univariate analysis of KAP regarding CCVD For the total KAP score, all variables except household size demonstrated statistically significant associations (P < 0.05). Most demographic and lifestyle variables were also significantly associated with dimension-specific scores, though the association patterns varied across dimensions. In terms of score characteristics, participants from North China, those residing in urban areas, and those with a confirmed family history of CCVD generally exhibited significantly higher scores across all three dimensions. Participants whose parents had a junior high school education or below, or whose parents were employed in “other” occupations, tended to score higher than participants whose parents were engaged in manual or mental labor occupations. Health status, BMI, and health behaviors were also found to be significantly associated with KAP scores (P < 0.05). Additional information is available in Table 3, Table 4. Table 3Univariate analysis of factors associated with total knowledge-attitude-practice score regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariablescoreF/tPAge4.35<0.01<1880.40 ± 21.1118–3572.62 ± 20.30Gender−3.86<0.01Male71.23 ± 20.80Female77.61 ± 20.46Only child−2.76<0.01Yes72.41 ± 19.98No76.96 ± 21.32Educational level15.16<0.01Lower educational level80.80 ± 19.03Medium educational level68.42 ± 18.02Higher educational level75.24 ± 17.88Residential status4.80<0.01Urban78.07 ± 21.16Rural70.26 ± 19.47Residential region143.94<0.01North China91.42 ± 17.91East China70.99 ± 16.08Southwest China63.57 ± 17.88Father's occupational type19.43<0.01Manual labor77.16 ± 20.07mental labor70.04 ± 20.64others91.23 ± 18.82Mother's occupational type17.21<0.01Manual labor73.78 ± 19.09Mental labor73.68 ± 21.54Others93.92 ± 19.28Father's education level26.69<0.01Junior high school and below84.62 ± 16.56Senior high school or technical secondary school78.99 ± 20.52College diploma68.94 ± 18.50Bachelor's degree and above68.08 ± 21.97Mother's education level23.10<0.01Junior high school and below84.97 ± 16.80Senior high school or technical secondary school76.07 ± 19.86College diploma71.12 ± 18.80Bachelor's degree and above67.71 ± 22.42Health status71.97<0.01Very healthy84.81 ± 19.41Relatively healthy79.91 ± 18.62General health71.40 ± 16.92Relatively unhealthy55.18 ± 12.56Poor health52.14 ± 6.62Household size1.960.14Small household73.60 ± 15.61Medium household75.96 ± 21.18Large household72.01 ± 21.72Household monthly income6.74<0.01Low income76.40 ± 18.57Medium income81.23 ± 17.72Upper-middle income78.40 ± 19.84High income71.89 ± 21.55Family history72.71<0.01Yes80.32 ± 18.05No81.42 ± 20.23Unknown61.59 ± 17.56Body mass index10.83<0.01Underweight75.50 ± 18.21Normal weight72.42 ± 21.05Overweight80.12 ± 23.38Obese90.82 ± 18.99 Univariate analysis of factors associated with total knowledge-attitude-practice score regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. Table 4Univariate analysis of factors associated with knowledge-attitude-practice scores regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariableknowledgeattitudepracticescoreF/tPscoreF/tPscoreF/tPAge1.770.785.19<0.015.31<0.01<1830.42 ± 11.0825.94 ± 7.2624.04 ± 6.0218–3528.78 ± 9.4722.49 ± 8.4021.34 ± 5.43Gender−3.83<0.01−4.64<0.01−0.710.48Male27.51 ± 9.6521.77 ± 8.3821.94 ± 5.59Female30.55 ± 10.0524.79 ± 7.8922.27 ± 5.85only child−1.770.08−3.42<0.01−2.050.04Yes28.50 ± 9.4422.29 ± 7.9821.62 ± 5.47No29.89 ± 10.3924.51 ± 8.3122.56 ± 5.93Educational level10.28<0.0118.89<0.0122.90<0.01Lower educational level30.92 ± 10.4325.21 ± 7.0724.66 ± 5.52Medium educational level26.60 ± 8.4820.49 ± 7.5221.33 ± 5.10Higher educational level29.80 ± 10.8424.21 ± 8.7521.23 ± 5.79Residential status3.56<0.016.27<0.012.260.02Urban30.38 ± 10.5725.15 ± 8.0022.54 ± 6.07Rural27.63 ± 8.8321.10 ± 7.9721.53 ± 5.17Residential region69.10<0.01294.25<0.0153.20<0.01North China35.41 ± 10.9630.76 ± 3.4225.25 ± 6.58East China28.51 ± 7.6320.75 ± 7.1621.72 ± 4.85Southwest China24.34 ± 7.9619.56 ± 8.0319.67 ± 4.21Father's occupational type9.44<0.01104.42<0.0111.10<0.01Manual labor30.29 ± 9.8024.51 ± 7.8222.36 ± 5.60Mental labor27.41 ± 9.6721.28 ± 9.4021.35 ± 5.60Others33.63 ± 11.9631.27 ± 2.6826.33 ± 6.61Mother's occupational type9.42<0.01108.98<0.018.38<0.01Manual labor28.66 ± 9.2823.48 ± 7.8921.63 ± 5.12Mental labor29.02 ± 10.2022.58 ± 8.5022.08 ± 5.91Others36.05 ± 11.3531.30 ± 2.7326.57 ± 7.10Father's education level15.73<0.0137.78<0.0113.60<0.01Junior high school and below32.55 ± 9.4127.97 ± 6.1424.10 ± 5.45Senior high school or technical secondary school31.30 ± 9.7824.74 ± 7.7622.95 ± 5.64College diploma26.87 ± 8.9221.05 ± 7.6321.02 ± 5.22Bachelor's degree and above26.70 ± 10.2620.68 ± 8.7120.69 ± 5.86Mother's education level15.81<0.0132.29<0.018.68<0.01Junior high school and below33.29 ± 9.0627.95 ± 6.3223.74 ± 5.87Senior high school or technical secondary school29.68 ± 9.8123.82 ± 7.6422.57 ± 5.31College diploma27.82 ± 8.8521.86 ± 7.6821.44 ± 5.37Bachelor's degree and above26.40 ± 10.4020.52 ± 8.8720.79 ± 5.89Health status104.69<0.01136.19<0.0159.18<0.01Very healthy32.55 ± 10.4627.57 ± 6.7624.68 ± 5.99Relatively healthy31.18 ± 9.8925.78 ± 6.9922.68 ± 5.38General health28.74 ± 8.7121.98 ± 7.3220.68 ± 4.67Relatively unhealthy22.10 ± 5.6615.00 ± 5.2318.08 ± 3.90Poor health20.63 ± 2.0113.78 ± 4.2317.73 ± 2.47Household size1.760.185.75<0.010.450.64Small household29.46 ± 8.4721.42 ± 6.7322.72 ± 5.01Medium household29.67 ± 10.1224.17 ± 8.1722.11 ± 5.85Large household27.79 ± 10.1522.31 ± 8.8221.91 ± 5.70Household monthly income5.05<0.017.40<0.014.57<0.01Low income29.70 ± 10.0023.97 ± 7.1422.73 ± 6.22Medium income32.40 ± 8.8725.19 ± 7.4723.64 ± 5.33Upper-middle income30.11 ± 9.8625.45 ± 7.3322.84 ± 5.77High income28.18 ± 10.1222.26 ± 8.6321.45 ± 5.68Family history53.71<0.0166.87<0.0147.94<0.01Yes31.98 ± 9.1425.08 ± 6.6223.26 ± 5.44No31.40 ± 10.4126.47 ± 7.4223.55 ± 6.01Unknown24.04 ± 8.0218.26 ± 8.0719.29 ± 4.50Body mass index6.80<0.0140.40<0.012.960.04Underweight29.33 ± 9.2723.95 ± 7.3822.32 ± 5.22Normal weight28.35 ± 10.0022.34 ± 8.4921.72 ± 5.65Overweight31.63 ± 10.4925.58 ± 8.3322.91 ± 6.68Obese35.23 ± 10.2330.67 ± 3.9124.92 ± 6.86Exercise behavior4.34<0.0118.78<0.01Never exercise26.47 ± 8.8221.50 ± 6.001–2 times/week29.11 ± 8.4826.09 ± 6.553–4 times/week31.16 ± 9.8824.98 ± 7.90≥5 times/week28.62 ± 11.1720.86 ± 9.22Smoking behavior76.82<0.01227.50<0.01≥4 times/week22.11 ± 5.3115.99 ± 5.76< 4 times/week23.29 ± 7.1516.27 ± 6.44Never smoke32.58 ± 9.8627.29 ± 6.40Drinking behavior99.69<0.01135.77<0.01≥3 times/week22.21 ± 5.0716.12 ± 5.53< 3 times/week24.31 ± 6.6819.43 ± 8.19Never drink alcohol33.46 ± 10.1927.32 ± 6.41Sleep behavior74.37<0.0178.57<0.01More than 9 h29.69 ± 10.6523.18 ± 7.777–8 h32.70 ± 9.8927.21 ± 6.516–7 h25.39 ± 7.6020.31 ± 8.64Less than 6 h21.58 ± 4.2815.42 ± 6.08 Univariate analysis of factors associated with knowledge-attitude-practice scores regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. Multiple linear regression analysis of KAP regarding CCVD Multiple linear regression analyses were performed to identify factors associated with KAP scores. To ensure robustness, two modeling strategies were conducted: Model 1, based on univariate significance screening, and Model 2, which incorporated all relevant variables guided by the KAP theoretical framework and prior literature (Yang et al., 2024; Chen et al., 2006). The results derived from the two approaches were generally consistent. For the attitude dimension, all candidate variables met the univariate screening threshold, rendering the results from Models 1 and 2 essentially consistent; therefore, only one model was fitted. Model 2 was designated as the primary inferential model, as it avoids the potential exclusion of important confounders that may arise from univariate significance screening. Unstandardized coefficients (B) with 95% confidence intervals are reported to emphasize the magnitude and precision of the associations. For total score, female gender, good health, and overweight were positive correlates; medium education, rural residence, East or Southwest China, paternal college education, maternal education of senior high school or above, poor health, and unknown family history were negative. For knowledge, higher education, paternal senior high or technical secondary education, medium income, overweight, regular exercise, and adequate sleep were positive; East or Southwest China, higher maternal education, unknown family history, smoking, and drinking were negative. For attitude, higher education, overweight, very good health, regular exercise, and adequate sleep were positive; rural residence, East or Southwest China, higher maternal education, large household size, poor health, unknown family history, smoking, and drinking were negative. For practice, good health was positive; Southwest China and unknown family history were negative. Regular exercise, smoking, drinking, and sleep were excluded as components of the practice score. Regression analysis results are presented in Table 5, Table 6. Table 5Multiple linear regression analysis of factors associated with total knowledge-attitude-practice score regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariableM1M2B95%CIB95%CIΔB90.1883.68, 96.6990.4183.89, 96.930.22Age<1818–35−2.69−6.00, 0.61−2.63−5.94, 0.680.06Gendermalefemale3.361.01, 5.713.491.14, 5.850.13Only childyesno0.66−1.75, 3.070.90−1.53, 3.320.24Educational levellow educational levelmedium educational level−4.01−7.51, −0.51−3.91−7.41, −0.410.10high educational level1.56−2.07, 5.201.63−2.02, 5.270.07Residential statusurbanrural−3.04−5.48, −0.60−2.79−5.24, −0.340.25Residential regionNorth ChinaEast China−10.70−13.95, −7.46−10.61−13.89, −7.340.09Southwest China−15.03−18.31, −11.74−14.94−18.23, −11.650.08Father's occupational typemanual labormental labor−0.72−3.27, 1.82−0.66−3.20, 1.890.07others−2.75−9.17, 3.67−2.84−9.25, 3.57−0.09Mother's occupational typemanual labormental labor0.47−2.05, 2.980.40−2.12, 2.93−0.07others4.52−1.24, 10.294.45−1.31, 10.21−0.07Father's education leveljunior high school and belowsenior high school or technical secondary school−0.96−4.55, 2.63−1.15−4.75, 2.45−0.19college diploma−4.78−8.58, −0.98−4.95−8.75, −1.15−0.17bachelor's degree and above−2.53−6.59, 1.53−2.62−6.68, 1.45−0.09Mother's education leveljunior high school and belowsenior high school or technical secondary school−3.96−7.41, −0.51−3.93−7.38, −0.480.03college diploma−5.79−9.85, −1.72−5.87−9.93, −1.80−0.08bachelor's degree and above−4.18−8.24, −0.11−4.17−8.24, −0.11<0.01Health statusgeneral healthvery healthy6.843.01, 10.676.822.99, 10.65−0.02relatively healthy3.860.13, 7.593.980.25, 7.710.12relatively unhealthy−8.39−12.91, −3.86−8.33−12.85, −3.810.06poor health−10.03−15.21, −4.84−9.99−15.17, −4.800.04Household sizemedium householdsmall household−0.23−4.14, 3.67−0.23large household−2.56−5.46, 0.34−2.56Household monthly incomeupper-middle incomelow income−0.69−6.47, 5.09−0.95−6.74, 4.84−0.26medium income2.57−1.38, 6.522.50−1.45, 6.45−0.07high income−0.07−2.94, 2.80−0.22−3.09, 2.65−0.15Family historynoyes1.55−1.39, 4.491.40−1.54, 4.34−0.15unknown−8.90−11.95, −5.85−8.88−11.93, −5.830.03Body mass indexnormal weightunderweight−0.78−3.50, 1.94−0.61−3.34, 2.120.17overweight6.942.28, 11.607.192.52, 11.850.25obese3.27−1.75, 8.293.52−1.51, 8.550.25Notes: For the total score outcome in Table 5, Model 1 had an overall F-statistic of 23.11 with an adjusted R2 of 0.53; Model 2 presented an F-statistic of 24.51 and an adjusted R2 of 0.53, revealing that the selected predictors collectively explained 53.00% of the variance in total score. Notably, global F-tests indicated that all constructed regression models reached statistical significance at the level of P < 0.01.Abbreviations: M1, Model 1 (variables selected based on univariate screening, P < 0.05); M2, Model 2 (variables selected based on the knowledge-attitude-practice theoretical framework and prior literature); ΔB, change in unstandardized regression coefficient from Model 1 to Model 2 (i.e., B2–B1). Multiple linear regression analysis of factors associated with total knowledge-attitude-practice score regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. Notes: For the total score outcome in Table 5, Model 1 had an overall F-statistic of 23.11 with an adjusted R2 of 0.53; Model 2 presented an F-statistic of 24.51 and an adjusted R2 of 0.53, revealing that the selected predictors collectively explained 53.00% of the variance in total score. Notably, global F-tests indicated that all constructed regression models reached statistical significance at the level of P < 0.01. Abbreviations: M1, Model 1 (variables selected based on univariate screening, P < 0.05); M2, Model 2 (variables selected based on the knowledge-attitude-practice theoretical framework and prior literature); ΔB, change in unstandardized regression coefficient from Model 1 to Model 2 (i.e., B2–B1). Table 6Multiple linear regression analysis of factors associated with knowledge-attitude-practice scores regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025.VariableknowledgeattitudepracticeM1M2MM1M2B95%CIB95%CIΔBB95%CIB95%CIB95%CIΔBconstant28.1023.78, 32.4227.8923.55, 32.22−0.2226.6523.95, 29.3525.3623.23, 27.5025.3423.15, 27.54−0.02Age<1818–351.09−0.66, 2.831.09−0.05−2.02, 0.08−0.48−1.59, 0.64−0.48−1.60, 0.64<0.01Gendermalefemale0.030.56, 1.940.63−0.62, 1.880.600.04−0.02, 1.490.03−0.77, 0.820.03Only childyesno0.03−0.26, 1.260.29−0.52, 1.100.29−0.52, 1.10<0.01Educational levellow educational levelmedium educational level0.03−1.22, 2.360.33−1.51, 2.160.30−0.01−1.32, 0.91−1.44−2.62, −0.27−1.45−2.63, −0.27<0.01high educational level0.141.15, 4.412.170.27, 4.072.030.181.74, 4.03−1.93−3.15, −0.71−1.93−3.16, −0.71<0.01Residential statusurbanrural−0.04−2.00, 0.55−0.75−2.02, 0.52−0.71−0.09−2.29, −0.76−0.18−1.00, 0.64−0.18−1.00, 0.64<0.01Residential regionNorth ChinaEast China−0.09−3.62, −0.27−2.15−3.85, −0.44−2.06−0.25−5.30, −3.23−1.56−2.66, −0.47−1.56−2.66, −0.46<0.01Southwest China−0.20−5.98, −2.33−4.21−6.04, −2.38−4.01−0.20−4.51, −2.32−2.93−4.03, −1.82−2.93−4.04, −1.82<0.01Father's occupational typemanual labormental labor−0.03−1.96, 0.69−0.60−1.92, 0.73−0.56−0.01−1.02, 0.57−0.01−0.87, 0.85−0.01−0.87, 0.85<0.01others−0.05−5.72, 0.94−2.25−5.59, 1.09−2.200.03−0.74, 3.27−0.33−2.49, 1.84−0.33−2.49, 1.84<0.01Mother's occupational typemanual labormental labor0.05−0.35, 2.280.94−0.38, 2.250.89−0.01−0.92, 0.670.38−0.47, 1.230.38−0.47, 1.23<0.01others0.04−1.14, 4.851.89−1.10, 4.891.850.01−1.40, 2.201.88−0.07, 3.821.88−0.07, 3.82<0.01Father's education leveljunior high school and belowsenior high school or technical secondary school0.090.21, 3.952.090.21, 3.962.00−0.03−1.76, 0.50−0.47−1.68, 0.74−0.47−1.69, 0.74<0.01college diploma0.01−1.85, 2.160.18−1.83, 2.180.17−0.07−2.63, −0.22−1.18−2.46, 0.10−1.18−2.46, 0.10<0.01bachelor's degree and above0.04−1.27, 3.000.86−1.27, 2.990.82−0.02−1.57, 1.01−0.79−2.15, 0.58−0.79−2.16, 0.58Mother's educational leveljunior high school and belowsenior high school or technical secondary school−0.12−4.39, −0.81−2.59−4.38, −0.79−2.47−0.10−2.82, −0.66−0.32−1.48, 0.84−0.31−1.48, 0.850.01college diploma−0.08−4.29, −0.10−2.29−4.39, −0.19−2.21−0.08−3.08, −0.54−0.86−2.23, 0.51−0.86−2.23, 0.52<0.01bachelor's degree and above−0.12−4.78, −0.56−2.58−4.70, −0.47−2.46−0.09−2.88, −0.320.02−1.35, 1.380.02−1.35, 1.39<0.01Health statusgeneral healthvery healthy−0.01−2.28, 1.74−0.21−2.22, 1.80−0.200.120.93, 3.352.341.05, 3.632.351.05, 3.64<0.01relatively healthy−0.05−2.96, 0.96−0.95−2.91, 1.01−0.900.07−0.01, 2.351.350.10, 2.611.350.09, 2.61relatively unhealthy−0.07−4.46, 0.37−2.05−4.47, 0.36−1.98−0.08−3.37, −0.47−1.04−2.57, 0.49−1.04−2.57, 0.49<0.01poor health−0.06−4.84, 0.64−2.22−4.97, 0.53−2.16−0.09−4.39, −1.09−1.27−3.01, 0.48−1.27−3.02, 0.49<0.01Household sizemedium householdsmall household−0.04−2.35, 0.10large household−0.06−2.13, −0.31Household monthly incomeupper-middle incomelow income−0.04−4.72, 1.31−1.78−4.80, 1.24−1.75−0.05−3.59, 0.040.41−1.54, 2.360.41−1.54, 2.37<0.01medium income0.070.10, 4.212.110.05, 4.162.040.01−1.07, 1.410.82−0.51, 2.150.82−0.52, 2.15<0.01high income0.01−1.32, 1.700.19−1.31, 1.700.18−0.02−1.25, 0.56−0.10−1.06, 0.87−0.10−1.06, 0.87<0.01Family historynoyes0.05−0.35, 2.721.06−0.48, 2.611.01−0.01−1.08, 0.790.34−0.65, 1.330.34−0.65, 1.33<0.01unknown−0.10−3.83, −0.60−2.15−3.76, −0.53−2.05−0.15−3.55, −1.61−1.84−2.87, −0.82−1.84−2.87, −0.82<0.01Exercise behaviornever exercise1–2 times/week0.03−1.56, 3.060.79−1.52, 3.100.750.110.66, 3.433–4 times/week0.191.90, 6.594.341.99, 6.694.150.110.62, 3.44≥5 times/week0.232.46, 7.244.902.51, 7.294.670.04−0.72, 2.16Smoking behaviornever smoke≥4 times/week−0.16−6.89, −2.51−4.28−6.97, −2.58−4.12−0.15−4.91, −2.28< 4 times/week−0.17−6.28, −2.11−4.77−6.37, −2.19−4.61−0.17−4.86, −2.34Drinking behaviornever drink alcohol≥3 times/week−0.15−6.71, −2.36−4.61−6.79, −2.44−4.46−0.12−4.18, −1.56< 3 times/week−0.17−5.35, −2.10−3.86−5.50, −2.22−3.69−0.09−2.52, −0.54Sleep behaviorless than 6 h6–7 h0.06−0.64, 3.911.54−0.74, 3.821.480.05−0.42, 2.327–8 h0.212.04, 6.574.231.97, 6.504.020.171.45, 4.18more than 9 h0.191.87, 6.624.191.81, 6.574.000.080.13, 3.00Body mass indexnormal weightunderweight−0.01−1.69, 1.15−0.22−1.64, 1.20−0.210.01−0.63, 1.08−0.82−1.74, 0.09−0.83−1.75, 0.09<0.01overweight0.070.34, 5.172.640.23, 5.062.580.070.92, 3.831.12−0.45, 2.681.12−0.45, 2.69<0.01obese0.04−1.12, 4.091.44−1.16, 4.041.400.04−0.18, 2.960.34−1.36, 2.030.34−1.36, 2.03<0.01Notes: Separate regression analyses for the three KAP dimensions are presented in Table 6. For the knowledge dimension, Model 1 yielded an F-statistic of 14.63 and an adjusted R2 of 0.45, while Model 2 had an F-statistic of 14.31 and an adjusted R2 of 0.45, explaining 45.00% of the variance in knowledge scores. The hierarchical regression model for the attitude dimension obtained an overall F-statistic of 37.67 and an adjusted R2 of 0.71, which represented the optimal explanatory performance across all dimensions. For the practice dimension, Model 1 had an F-statistic of 9.96 and an adjusted R2 of 0.29, and Model 2 showed an F-statistic of 9.61 and an adjusted R2 of 0.29, with predictors explaining 29.00% of the variance in practice scores. Notably, global F-tests indicated that all constructed regression models reached statistical significance at the level of P < 0.01.Abbreviations: M1, Model 1 (univariate screening); M2, Model 2 (theory-driven selection); M, single model for attitude dimension (all variables entered); ΔB, change in B coefficient from M1 to M2. Multiple linear regression analysis of factors associated with knowledge-attitude-practice scores regarding cardiovascular and cerebrovascular diseases among adolescents and young adults in North China, East China, and Southwest China, August to October 2025. Notes: Separate regression analyses for the three KAP dimensions are presented in Table 6. For the knowledge dimension, Model 1 yielded an F-statistic of 14.63 and an adjusted R2 of 0.45, while Model 2 had an F-statistic of 14.31 and an adjusted R2 of 0.45, explaining 45.00% of the variance in knowledge scores. The hierarchical regression model for the attitude dimension obtained an overall F-statistic of 37.67 and an adjusted R2 of 0.71, which represented the optimal explanatory performance across all dimensions. For the practice dimension, Model 1 had an F-statistic of 9.96 and an adjusted R2 of 0.29, and Model 2 showed an F-statistic of 9.61 and an adjusted R2 of 0.29, with predictors explaining 29.00% of the variance in practice scores. Notably, global F-tests indicated that all constructed regression models reached statistical significance at the level of P < 0.01. Abbreviations: M1, Model 1 (univariate screening); M2, Model 2 (theory-driven selection); M, single model for attitude dimension (all variables entered); ΔB, change in B coefficient from M1 to M2. Discussion The overall KAP score fell within the moderate range, indicating that recent health education efforts in China have yielded preliminary positive outcomes among adolescents and young adults. The dimension scores ranked in descending order as attitude, knowledge, and practice, a pattern consistent with the findings of Yang (Yang et al., 2024). This hierarchy reflects a generally proactive orientation toward disease prevention, while concurrently revealing a discernible gap between health awareness and actual behavioral engagement. Knowledge scores were lower than attitude scores, suggesting that participants possessed limited CCVD-specific knowledge. This pattern may reflect the broad and non-specific nature of current health education, which often lack sufficient depth. Moreover, health information delivered through fragmented media channels may increase awareness, but tends to match adolescents' cognitive inclination toward immediate and tangible content rather than systematic reasoning, which may explain why attitudes were more favorable than knowledge in this population (Yang et al., 2024). The practice dimension scored the lowest, underscoring a critical concern. Despite adequate recognition of healthy behaviors, the persistence of unhealthy habits may stem from developmental psychological characteristics, environmental constraints, inadequate health services, and peer influence compounded by emotional lability (Gui et al., 2025). Higher educational attainment was positively associated with elevated scores in both health knowledge and attitude domains, a relationship associated with enhanced access to health information and greater proactive awareness of disease prevention within this demographic. This observation aligns with the established consensus identifying educational level as a primary determinant of health literacy. Consistent with prior investigations, data from surveys encompassing 25 provinces in China indicate that the overall health literacy among Chinese residents remains lower than that of developed nations, with significant disparities observed across geographic regions, urban-rural divides, and socioeconomic strata (Li et al., 2021). Within this sample, participants from East and Southwest China had significantly lower KAP scores than those from North China. Given the convenience sampling design, these regional variations warrant cautious interpretation. The observed differences may relate to cross-regional variations in medical resources, health education coverage, and public health awareness, though digital health education may help address such gaps in resource-limited settings (Shakya et al., 2025). However, as none of these factors were directly measured, these explanations remain speculative. Consistent with this observation, inadequate resource allocation and limited health education have been linked to lower health awareness in some regions (Lin, 2022). Furthermore, the observed regional patterns may reflect sampling-related factors, including recruitment approaches, socioeconomic composition, and online platform usage, rather than true geographic disparities. Our data support only the presence of regional differences within this specific sample, and region-specific recommendations require confirmation in population-based studies. Higher maternal education was unexpectedly associated with lower knowledge and attitude scores among offspring, though this should not be interpreted as causal (Xu and Xu, 2024). One possible explanation is that demanding careers may reduce parent-child health communication and prioritize academic achievement over health discussions (Wang and Zhu, 2024). However, as this pathway was not assessed, the interpretation remains speculative. Residual confounding from unmeasured variables such as parenting styles and family health culture, as well as selection bias, may also contribute. Participants from larger households had lower attitude scores, which may partly reflect the widespread perception that CCVD predominantly affects older adults. Middle-income groups demonstrated superior CCVD knowledge scores, plausibly attributable to greater financial stability and more effective use of public health resources (Gui et al., 2025), whereas upper-middle-income individuals tended to forgo public health education in favor of private services. A positive CCVD family history was associated with higher practice scores, whereas uncertain family history correlated with lower knowledge and attitude scores, confirming that risk perception drives health management. The high proportion of “unknown” responses suggests insufficient awareness of familial CCVD susceptibility rather than actual absence of affected relatives. Excellent self-rated health was associated with higher attitude and practice scores, whereas fair or poor ratings corresponded to lower scores consistent with self-efficacy theory (Leung et al., 2025), which posits that better health is associated with greater self-management confidence and positive health behaviors. Regular exercise and sufficient sleep were positively associated with KAP scores, whereas smoking and excessive drinking were negatively associated. This suggests that healthier lifestyles correspond to stronger health awareness and greater engagement with CCVD-related knowledge (Lim et al., 2021). Overweight status was positively associated with CCVD knowledge and attitude, possibly reflecting health-information seeking motivated by weight concerns rather than superior health literacy. However, practice scores did not increase correspondingly, indicating a persistent knowledge-practice gap (Upadhayay and Bhaumik, 2025). As health anxiety and information-seeking behavior were not directly assessed, this interpretation warrants caution. These findings highlight the need for prevention strategies that target modifiable lifestyle factors and bridge the knowledge-practice gap. This study has several strengths and limitations. A key strength is its multi-regional design covering North, East, and Southwest China, with a relatively balanced sample that enables preliminary geographic comparisons. Additionally, our findings yield valuable practical implications for health promotion initiatives targeting adolescents and young adults. Several limitations warrant consideration. First, the use of convenience sampling through online platforms may constrain the generalizability of the findings, as individuals with higher educational attainment and greater health related interest are likely overrepresented, thereby introducing potential educational and self selection biases. To mitigate this concern, neutral recruitment language was adopted, and statistical adjustments for educational level and socioeconomic status were applied across all regression models. Second, the geographic coverage restricted to three regions limits the external validity of the results when extrapolated to the broader Chinese population. Furthermore, the online survey modality may inherently exclude individuals without internet access. Third, the cross sectional design precludes any causal inferences. Fourth, self-reported health behavior data are prone to recall bias and social desirability bias. To mitigate these effects, anonymous questionnaire administration, reverse-scored items, and a minimum completion time threshold were implemented. Finally, no correction for multiple comparisons was performed. Consequently, isolated significant findings should be interpreted with prudence and await replication in independent samples. Conclusions This cross-sectional survey reveals a moderate level of CCVD-related KAP among adolescents and young adults. Key associated factors include educational attainment, health behaviors, family history of CCVD, and geographic region. Based on these observed associations, several targeted interventions merit consideration: community- and school-based health education for lower-educated groups; interventions targeting the cognition-behavior gap for higher-educated groups; and enhanced resource allocation for regions with lower scores. Broader strategies should include smoking and alcohol cessation programs, promotion of regular exercise and adequate sleep, and family-based health management. However, given the cross-sectional design, these findings require cautious interpretation and verification in future prospective cohort studies with nationally representative samples. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process During the preparation of this work, the author used deepseek to improve language, grammar, and readability. After using this tool, the author reviewed and edited the content as needed and takes full responsibility for the content of the published article. CRediT authorship contribution statement Tongyan Sun: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Hua Zhang: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization, Methodology. Funding This work was supported by the Tianjin Municipal Education Commission Social Science Project (2022SK206). Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments Acknowledgments The authors would like to express their sincere gratitude to all individuals and institutions who provided support and assistance during the conduct of this research. Ethics declaration Informed consent and patient details Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed. Studies in human This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the 天津医科大学伦理委员会. (Approval No. TMUHMEC20250007) Data availability Data will be made available on request. References Associated Data Data Availability Statement Data will be made available on request.