Long-term Medication Adherence Attitudes and Behaviors of Stroke Patients in Rural Areas of China and Related Factors: Based on Follow-up Data from a County in Hebei Province

LONG Yutong¹, LU Shiyu², TAN Jie²,³, YANG Bolu²,³, DUAN Jingying¹, YANG Tongde², YAN Lijing²,³,⁴, GONG Enying¹,⁵, SHAO Ruitai¹,⁵

¹School of Population Medicine and Public Health, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China
²Global Health Research Center, Duke Kunshan University, Kunshan 215316, China
³School of Public Health, Wuhan University, Wuhan 430072, China
⁴Duke Global Health Institute, Duke University, Durham 27710, USA
⁵State Key Laboratory of Respiratory Health and Multimorbidity, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China

Corresponding authors: GONG Enying, Professor; E-mail: gongenying@cams.cn
SHAO Ruitai, Professor; E-mail: shaoruitai@sph.pumc.edu.cn

Abstract

Background: Good medication adherence can significantly reduce the risk of recurrence in stroke patients and is crucial for secondary prevention. However, existing research primarily focuses on short-term medication use and adherence behaviors among rural stroke patients, while studies on long-term medication adherence behaviors remain limited.

Objective: To analyze the long-term medication behaviors, adherence attitudes, and related factors among stroke survivors in rural China.

Methods: Based on the System-integrated Technology-enabled Model of Care for stroke management in rural China (SINEMA) study conducted in 2017 across 5 towns and 60 villages in Nanhe County, Hebei Province, this study included all stroke survivors who had previously participated in the 2017 SINEMA study, consented to participate in the follow-up survey conducted from May to July 2024, and completed the survey. The study investigated medication use including antihypertensive drugs, statins, antiplatelet drugs, and hypoglycemic drugs. The Maastricht Utrecht Adherence in Hypertension-16 (MUAH-16) scale was employed to assess medication adherence attitudes, specifically focusing on dimensions of positive medical and medication attitudes. The Morisky Green Levine scale was used to measure medication adherence behaviors among patients taking the four categories of drugs, with a score of 0 defined as high adherence. Descriptive analysis was applied to present patient characteristics, medication adherence status, and relevant factors identified through multivariate logistic regression analysis.

Results: A total of 912 participants were included with a mean age of (72.0±7.9) years; 46.1% (420/912) were female. Regarding medication use, 772 cases (84.6%) reported taking antihypertensive drugs, 547 cases (59.9%) took antiplatelet drugs, 427 cases (46.8%) used statins, and 203 cases (22.3%) took hypoglycemic drugs. Among patients currently taking these four drug types, adherence rates were 77.3% (157/203) for hypoglycemic drugs, 71.5% (552/772) for antihypertensive drugs, 71.2% (392/547) for antiplatelet drugs, and 71.2% (304/427) for statins. The survey showed that 49.2% (449/912) of patients held positive attitudes toward medical care and medication use. Multivariate logistic regression analysis revealed that among patients taking antihypertensive medications, those with higher educational attainment (secondary school or above: OR=1.87, 95%CI=1.13-3.09) and positive attitudes toward medical care and medication (OR=1.53, 95%CI:1.08-2.17) had higher medication adherence, while patients capable of independently completing work tasks (OR=0.56, 95%CI=0.32-0.99) and those visiting village clinics ≥1 time per month (OR=0.68, 95%CI=0.53-0.88) had lower adherence (P<0.05). For patients taking antiplatelet drugs, higher educational attainment (secondary school or above: OR=1.79, 95%CI=1.09-2.96) was associated with better adherence, whereas patients with hemorrhagic stroke (OR=0.55, 95%CI=0.31-0.98) had lower adherence (P<0.05). Among patients on statins, those visiting village clinics ≥1 time per month (OR=0.67, 95%CI=0.46-0.98) showed lower adherence (P<0.05). For patients taking antidiabetic medications, male patients (OR=0.21, 95%CI=0.06-0.73), those enrolled in the chronic and major disease insurance program (OR=0.34, 95%CI=0.15-0.79), and individuals visiting village clinics ≥1 time per month (OR=0.34, 95%CI=0.13-0.89) all had significantly lower medication adherence (P<0.05).

Conclusion: Stroke survivors in rural areas with longer disease histories tend to exhibit relatively high medication adherence; however, nearly one-third still fail to consistently follow treatment recommendations. The determinants of adherence vary across different medication types, highlighting the need for tailored interventions to enhance medication adherence and secondary prevention behaviors in this population.

Keywords: Stroke; Multiple medications; Medication Adherence; Rural

Introduction

Stroke represents a major global burden of non-communicable diseases and remains the leading cause of disability and mortality in China. The prevalence and incidence of stroke in China continue to rise. Research based on Chinese populations indicates an annual prevalence of 1,329.5 per 100,000 and an annual incidence of 442.1 per 100,000. Although case fatality rates have declined with improved healthcare, they remain high at 35.8% (95%CI=26.1%-46.1%), demonstrating that the characteristic of high incidence persists. Secondary prevention medications are critical for improving stroke prognosis and reducing recurrence and related complications. Existing research and domestic and international stroke secondary prevention guidelines emphasize the effectiveness of rational use of antihypertensive drugs, antiplatelet drugs, statins, and hypoglycemic drugs in reducing stroke recurrence. However, studies have also found that poor medication adherence is widespread and represents an important barrier to long-term stroke management. Rural areas, with relatively scarce medical resources and limited continuous health management support, face more prominent issues with rational drug use and adherence behaviors among stroke patients. Previous research has primarily focused on short-term medication behaviors, with fewer studies examining long-term medication adherence, which is particularly important for controlling stroke recurrence. Research on long-term medication behaviors, adherence, and related factors for stroke secondary prevention drugs remains limited. Stroke patients often have multiple chronic conditions, which not only increase the complexity of medication management but may also weaken adherence to long-term therapy. Additionally, patients' attitudes and beliefs toward treatment significantly influence their long-term medication behaviors, with negative attitudes (such as concerns about adverse drug reactions or doubts about treatment efficacy) potentially reducing adherence. Therefore, understanding long-term medication behaviors in stroke patients requires comprehensive consideration of the complexity of multimorbidity and the influence of psychological and social support factors.

This study builds upon a long-term follow-up survey of stroke patient management to understand the current status of long-term medication use among stroke patients in rural Hebei Province, analyze adherence to antihypertensive drugs, antiplatelet drugs, statins, and hypoglycemic drugs, explore related factors, and provide evidence for optimizing stroke medication management in rural areas.

1.1 Study Population

This study was conducted in Nanhe County, Hebei Province, located in a "stroke high-incidence area" in China with a stroke burden twice the national average. Before national poverty alleviation, Nanhe County was designated as a "provincial-level impoverished county," with residents' per capita annual disposable income of 11,030 yuan, only half of the national average. Based on the System-integrated Technology-enabled Model of Care for stroke management in rural China (SINEMA), a cluster-randomized controlled trial conducted in 2017 across 5 towns and 60 villages in this region, this study's subjects comprised all stroke survivors who had previously participated in the 2017 SINEMA study and consented to participate in the follow-up survey conducted from May to July 2024. The SINEMA study developed and evaluated a system-integrated mobile health intervention model to improve secondary prevention effects for stroke patients in rural China. The intervention period was 1 year, with follow-up conducted in 2018, 2022, and 2023. This study was approved by the Ethics Review Committee of the Chinese Academy of Medical Sciences (approval number: CAMS & PUMC-IEC-2024-047), and all participants signed informed consent forms.

1.2 Data Collection

This study employed face-to-face questionnaires and physical examinations for data collection. The questionnaire covered sociodemographic characteristics (including age, sex, education level, marital status, annual household income, work capacity), enrollment in chronic disease major illness insurance (the chronic disease insurance package only applies to individuals participating in the health insurance system with serious chronic diseases, allowing reimbursement for outpatient services at county hospitals), lifestyle factors (such as smoking status), self-reported number of comorbid chronic diseases, stroke type and recurrence history, disability level (assessed using the modified Rankin Scale), and medication use.

The validated Maastricht Utrecht Adherence in Hypertension-16 (MUAH-16) scale was used to assess medication adherence attitudes. This scale includes four dimensions: positive medical and medication attitudes, lack of self-discipline, medication resistance, and active health management. This study focused on the "positive medical and medication attitudes" dimension, with scores dichotomized based on previous research and population distribution (16 points defined as positive attitude, <16 points as not positive). Participants self-reported current use of stroke secondary prevention medications (including antihypertensive drugs, antiplatelet drugs, statins, and hypoglycemic drugs). For each medication class, the Morisky Green Levine scale was used to assess medication adherence behaviors. This 4-item scale has been validated for assessing patient medication adherence with strong validity and predictive power, where a score of 0 indicates high adherence. The study also evaluated medication usage rates, adherence proportions, stroke recurrence rates, and disability indicators [modified Rankin Scale (mRS) score ≥3]. All surveys were conducted by uniformly trained investigators from neighboring counties. All data for this study were collected during a single survey using digital face-to-face interview systems and on-site supervision to ensure standardized and high-quality data collection.

1.3 Statistical Analysis

This study primarily employed descriptive statistics and multivariate logistic regression models to assess factors associated with high adherence behaviors among users of various medication classes. In regression models, all analyses followed the methods in the original SINEMA study statistical analysis plan (SAP), using intention-to-treat principles and controlling for cluster structure (village-level). Descriptive analysis was first conducted for all survey participants. Continuous variables were expressed as (x-±s) if normally distributed, with group comparisons using independent t-tests; non-normally distributed variables were expressed as M(QR) with group comparisons using rank-sum tests. Categorical variables were expressed as frequencies and percentages, with group comparisons using χ² tests. For each medication type, descriptive analysis compared variable distributions across different adherence behaviors and analyzed differences in medication attitudes across adherence statuses. Multivariate logistic regression analysis examined the influence of medication adherence attitudes on adherence behaviors. All analyses were completed using STATA 18.0 software, with statistical significance set at P<0.05.

Results

2.1 Main Demographic and Health Characteristics of the Survey Population

This study included 912 participants with a mean age of (72.0±7.9) years; 46.1% (420/912) were female. Educational attainment was primary school or below for 71.8% (655/912) of participants, and most (74.9%, 683/912) were married. Annual household income was below 5,000 yuan for 37.7% (344/912) of participants. Ischemic stroke was the predominant type (86.2%, 784/912). The median time since first stroke onset was 13 (10, 18) years, with 16.2% (148/912) having a stroke recurrence history in the past 2 years, and 38.0% (347/912) experiencing moderate to severe disability. Most patients had comorbid chronic conditions, primarily hypertension (90.0%), dyslipidemia (53.7%), diabetes (24.1%), and heart disease (22.9%). Regarding medication use, 772 cases (84.6%) self-reported taking antihypertensive drugs, 547 cases (59.9%) took antiplatelet drugs, 427 cases (46.8%) took statins, and 203 cases (22.3%) took hypoglycemic drugs. Demographic and health characteristics of patients taking the four medication classes are shown in Table 1 [TABLE:1].

2.2 Patient Medication Adherence Behaviors, Attitudes, and Related Factors

Among patients taking antihypertensive drugs, 71.5% (552/772) showed high adherence. No statistically significant differences were observed in sex, marital status, annual household income, work capacity, chronic disease major illness insurance enrollment, smoking status, proportions of self-reported hypertension, dyslipidemia, diabetes, or heart disease, stroke type, stroke recurrence in the past 2 years, or time since first recurrence between patients with different antihypertensive adherence levels (P>0.05). However, significant differences were found in education level, frequency of village clinic visits in the past year, and disability level (P<0.05). The proportion of patients with positive medical and medication attitudes was significantly higher among those with high antihypertensive adherence compared to those with low adherence (P<0.05), as shown in Table 2 [TABLE:2].

Among patients taking antiplatelet drugs, 71.2% (392/547) showed high adherence. No statistically significant differences were observed in sex, marital status, education level, annual household income, work capacity, chronic disease major illness insurance enrollment, frequency of village clinic visits in the past year, smoking status, proportions of self-reported hypertension, dyslipidemia, diabetes, or heart disease, stroke type, stroke recurrence in the past 2 years, time since first recurrence, disability level, or positive medical and medication attitudes between patients with different antiplatelet adherence levels (P>0.05), as shown in Table 2 [TABLE:2].

Among patients taking statins, 71.2% (304/427) showed high adherence. No statistically significant differences were observed in sex, marital status, education level, annual household income, work capacity, chronic disease major illness insurance enrollment, frequency of village clinic visits in the past year, smoking status, proportions of self-reported hypertension, dyslipidemia, diabetes, or heart disease, stroke type, stroke recurrence in the past 2 years, time since first recurrence, disability level, or positive medical and medication attitudes between patients with different statin adherence levels (P>0.05), as shown in Table 2 [TABLE:2].

Among patients taking hypoglycemic drugs, 77.3% (157/203) showed high adherence. No statistically significant differences were observed in sex, education level, annual household income, work capacity, chronic disease major illness insurance enrollment, smoking status, proportions of self-reported hypertension, dyslipidemia, diabetes, or heart disease, stroke type, stroke recurrence in the past 2 years, time since first recurrence, disability level, or positive medical and medication attitudes between patients with different hypoglycemic adherence levels (P>0.05). However, significant differences were found in marital status and frequency of village clinic visits in the past year (P<0.05), as shown in Table 2 [TABLE:2].

Logistic Regression Analysis

Using low adherence as the reference, multivariate logistic regression analysis was performed with medication adherence behavior for the four drug classes as the dependent variable. Based on previous relevant studies and variables showing statistical significance in univariate analysis (P<0.05), potential related factors were included as independent variables. Results showed that among patients taking antihypertensive medications, those with higher education (secondary school or above: OR=1.87, 95%CI=1.13-3.09) and positive medical and medication attitudes (OR=1.53, 95%CI:1.08-2.17) had higher medication adherence, while patients capable of independently completing work tasks (OR=0.56, 95%CI=0.32-0.99) and those visiting village clinics ≥1 time per month (OR=0.68, 95%CI=0.53-0.88) had lower adherence (P<0.05).

Among patients taking antiplatelet drugs, higher education (secondary school or above: OR=1.79, 95%CI=1.09-2.96) was associated with higher adherence, while patients with hemorrhagic stroke (OR=0.55, 95%CI=0.31-0.98) had lower adherence (P<0.05).

Among patients taking statins, those visiting village clinics ≥1 time per month (OR=0.67, 95%CI=0.46-0.98) had lower adherence (P<0.05).

Among patients taking hypoglycemic drugs, male patients (OR=0.21, 95%CI=0.06-0.73), those enrolled in chronic disease major illness insurance (OR=0.34, 95%CI=0.15-0.79), and individuals visiting village clinics ≥1 time per month (OR=0.34, 95%CI=0.13-0.89) all had significantly lower medication adherence (P<0.05), as shown in Table 3 [TABLE:3].

Discussion

Based on long-term follow-up data from Nanhe County in rural Hebei Province, this study found that stroke patients demonstrated relatively high overall adherence to secondary prevention medications, but with significant differences across drug classes and heterogeneous influencing factors. Compared to previous studies focusing on short-term stroke patients, this study included mostly patients with longer disease courses, who showed higher adherence levels, possibly reflecting the promotion of self-management behaviors through long-term follow-up and medical care. Regarding medication categories, usage rates of antihypertensive and antiplatelet drugs were significantly higher than those of statins and hypoglycemic drugs. Although overall adherence was relatively high, male patients among hypoglycemic drug users showed lower adherence, suggesting potential gender differences in long-term self-management. The study further confirmed that patients' education level, disease control status, and attitudes toward medication were associated with adherence behaviors, particularly indicating that stroke patients with lower education levels in rural areas need improvement in medication behaviors.

Compared to previous research, this study focused on patients who had suffered stroke for a relatively long time, representing a unique population that may have established long-term medication behaviors and differs substantially from studies focusing on newly diagnosed patients. Previous research predominantly examined medication behaviors in newly diagnosed or short-term stroke patients, mainly within 1-2 years of onset. For example, a study in Hunan targeting elderly stroke survivors (average disease course 49.37 months) found non-adherence rates as high as 61.4%, while a study in Harbin on ischemic stroke patients (average disease course 3.77 years) reported non-adherence rates of 65.48%. In contrast, this study focused on patients requiring long-term management after stroke (average disease course up to 13 years). Long-term stroke patients may face more challenges in medication adherence, as beliefs about medication, disease perception, and health literacy may significantly impact adherence during prolonged disease management. The lower non-adherence rates found in this study compared to previous research may be related to the characteristics of this population, including an average 13-year stroke history, 14.8% with previous stroke recurrence, and 38% with disability status. These long-term stroke patients may have developed better medication adherence behaviors, resulting in higher adherence rates compared to short-term newly diagnosed patients in previous studies.

In rural Hebei, secondary prevention drugs are commonly used among long-term stroke patients, with relatively high overall adherence but nearly 30% still non-adherent, and differences exist in usage rates and adherence across drug classes. Education level, clinic visit frequency, and patient attitudes toward medication are key influencing factors that show heterogeneity across different drugs. It is recommended that future rural stroke management implement targeted interventions for key populations such as those with low education levels, combined with personalized education and integration of primary care resources to improve long-term adherence and reduce stroke recurrence, thereby improving patient outcomes.

Compared to previous studies, this study further demonstrates that medication adherence issues are more prominent among populations with lower education levels and warrant greater attention. Education level is typically associated with health literacy, disease awareness, and appreciation for long-term regular medication use. Consistent with previous research, this study also found that patients' medication attitudes, particularly perceptions of medication necessity and concerns about adverse reactions, were significantly associated with antihypertensive medication adherence behaviors. Previous studies have emphasized the negative impact of medication concerns on adherence, while perception of medication necessity was associated with higher adherence.

Village clinic visit frequency was identified as a positive factor across multiple drug classes, suggesting that primary care services play an active role in long-term self-management for stroke patients. This study also found potential differences in medication usage proportions and adherence behaviors across drug classes. Previous research has found differences in medication persistence among different drugs in high-risk stroke populations in China. Specifically, 4.7-year persistence for hypoglycemic drugs (39.9%) was lower than for antithrombotic drugs (59.8%) and lipid-lowering drugs (43.9%). This study also found that over 70% of survey subjects had two or more comorbidities, with 22.3% of patients simultaneously taking hypoglycemic drugs. These complex comorbidity issues increase the burden of disease management for stroke patients, potentially raising concerns about drug efficacy and cross-medication adverse reactions, leading to potentially low adherence. Future research should pay attention to multiple risk factor control and polypharmacy behaviors in stroke patients.

This study's focus on medication behaviors among rural stroke patients with long disease courses has certain uniqueness. The study covered four commonly used secondary prevention drug classes for rural stroke patients, providing valuable supplementary evidence for examining adherence in this population. The study employed standardized assessment tools and statistical methods, ensuring result reliability. However, several limitations exist: First, measurement of medication use and adherence relied on patient self-reporting, which may introduce recall bias and social desirability bias, potentially overestimating actual adherence levels. Second, this was a cross-sectional study that could only establish associations between factors of interest, and despite controlling for some confounding factors, unobserved variables could not be excluded. Third, survey subjects came from a population previously enrolled in stroke studies, requiring caution in generalizing results, with external validity needing further verification. Finally, influencing factors from healthcare providers' perspectives were not covered, which future research could address through multi-level investigations.

The findings further demonstrate the necessity of strengthening secondary prevention medication management among long-term stroke patients in rural areas. Future intervention strategies should focus more on improving patients' understanding of different drug therapies and combine health education to improve medication attitudes and behavioral motivation.

Author Contributions: LONG Yutong, YAN Lijing, GONG Enying, and SHAO Ruitai were responsible for conceptualization and overall framework design. LONG Yutong, LU Shiyu, TAN Jie, YANG Bolu, and YANG Tongde were responsible for data collection, organization, and entry. All authors approved the manuscript. LONG Yutong, GONG Enying, and SHAO Ruitai take overall responsibility for the article and provided supervision.

Conflict of Interest: The authors declare no conflicts of interest.

References

[1] FEIGIN V L, BRAININ M, NORRVING B, et al. World stroke organization: global stroke fact sheet 2025[J]. Int J Stroke, 2025, 20(2): 132-144. DOI: 10.1177/17474930241308142.

[2] ZHAO Y, HUA X, REN X W, et al. Increasing burden of stroke in China: a systematic review and meta-analysis of prevalence, incidence, mortality, and case fatality[J]. Int J Stroke, 2023, 18(3): 259-267. DOI: 10.1177/17474930221135983.

[3] WANG R X, ZHU S J, LIANG J G, et al. Effects of health education based on the patient version of guidelines on medication compliance in secondary prevention among patients with ischemic stroke[J]. Chinese Journal of Stroke, 2025, 20(2): 171-180.

[4] WANG R X, ZHU S J, LIANG J G, et al. Effects of health education based on the patient version of guidelines on medication compliance in secondary prevention among patients with ischemic stroke[J]. Chinese Journal of Stroke, 2025, 20(2): 171-180.

[5] AL ALSHAIKH S, QUINN T, DUNN W, et al. Predictive factors of non-adherence to secondary preventative medication after stroke or transient ischaemic attack: a systematic review and meta-analyses[J]. Eur Stroke J, 2016, 1(2): 65-75. DOI: 10.1177/2396987316647187.

[6] JAMISON J, SUTTON S, MANT J, et al. Barriers and facilitators to adherence to secondary stroke prevention medications after stroke: analysis of survivors and caregivers views from an online stroke forum[J]. BMJ Open, 2017, 7(7): e016814. DOI: 10.1136/bmjopen-2017-016814.

[7] BUSHNELL C D, OLSON D M, ZHAO X, et al. Secondary preventive medication persistence and adherence 1 year after stroke[J]. Neurology, 2011, 77(12): 1182-1190. DOI: 10.1212/WNL.0b013e31822f0423.

[8] DE GEEST S, SABATÉ E. Adherence to long-term therapies: evidence for action[J]. Eur J Cardiovasc Nurs, 2003, 2(4): 323. DOI: 10.1016/s1474-5151(03)00091-4.

[9] SJÖLANDER M, ERIKSSON M, GLADER E L. The association between patients' beliefs about medicines and adherence to drug treatment after stroke: a cross-sectional questionnaire survey[J]. BMJ Open, 2013, 3(9): e003551.

[10] CHAKRABARTI S. Treatment attitudes and adherence among patients with bipolar disorder: a systematic review of quantitative and qualitative studies[J]. Harv Rev Psychiatry, 2019, 27(5): 290-302. DOI: 10.1097/HRP.0000000000000228.

[11] GALLACHER K I, JANI B D, HANLON P, et al. Multimorbidity in stroke[J]. Stroke, 2019, 50(7): 1919-1926. DOI: 10.1161/strokeaha.118.020376.

[12] GONG E Y, GU W B, SUN C, et al. System-integrated technology-enabled model of care to improve the health of stroke patients in rural China: protocol for SINEMA-a cluster-randomized controlled trial[J]. Am Heart J, 2019, 207: 27-39. DOI: 10.1016/j.ahj.2018.08.015.

[13] YAN L L, GONG E Y, GU W B, et al. Effectiveness of a primary care-based integrated mobile health intervention for stroke management in rural China (SINEMA): a cluster-randomized controlled trial[J]. PLoS Med, 2021, 18(4): e1003582. DOI: 10.1371/journal.pmed.1003582.

[14] JI R J, LIU G F, SHEN H P, et al. Persistence of secondary prevention medications after acute ischemic stroke or transient ischemic attack in Chinese population: data from China National Stroke Registry[J]. Neurol Res, 2013, 35(1): 29-36. DOI: 10.1179/1743132812Y.0000000107.

[15] CABRAL A C, CASTEL-BRANCO M, CARAMONA M, et al. Developing an adherence in hypertension questionnaire short version: MUAH-16[J]. J Clin Hypertens (Greenwich), 2018, 20(1): 118-124. DOI: 10.1111/jch.13137.

[16] MORISKY D E, GREEN L W, LEVINE D M. Concurrent and predictive validity of a self-reported measure of medication adherence[J]. Med Care, 1986, 24(1): 67-74. DOI: 10.1097/00005650-198601000-00007.

[17] BANKS J L, MAROTTA C A. Outcomes validity and reliability of the modified Rankin scale: implications for stroke clinical trials: a literature review and synthesis[J]. Stroke, 2007, 38(3): 1091-1096. DOI: 10.1161/01.STR.0000258355.23810.c6.

[18] GONG E Y, YAN L L, MCCORMACK K, et al. System-integrated technology-enabled model of care (SINEMA) to improve the health of stroke patients in rural China: Statistical analysis plan for a cluster-randomized controlled trial[J]. Int J Stroke, 2020, 15(2): 226-230. DOI: 10.1177/1747493019869707.

[19] CAO W J, KADIR A A, WANG J, et al. Medication non-adherence and associated factors among older adult stroke survivors in China[J]. Front Pharmacol, 2022, 13: 1054603. DOI: 10.3389/fphar.2022.1054603.

[20] GAST A, MATHES T. Medication adherence influencing factors-an (updated) overview of systematic reviews[J]. Syst Rev, 2019, 8(1): 112. DOI: 10.1186/s13643-019-1014-8.

[21] FOLEY L, LARKIN J, LOMBARD-VANCE R, et al. Prevalence and predictors of medication non-adherence among people living with multimorbidity: a systematic review and meta-analysis[J]. BMJ Open, 2020, 10(11): e044987.

[22] RUKSAKULPIWAT S, LIU Z J, YUE S H, et al. The association among medication beliefs, perception of illness and medication adherence in ischemic stroke patients: a cross-sectional study in China[J]. Patient Prefer Adherence, 2020, 14: 235-247. DOI: 10.2147/PPA.S235107.

[23] JAMISON J, GRAFFY J, MULLIS R, et al. Barriers to medication adherence for the secondary prevention of stroke: a qualitative interview study in primary care[J]. Br J Gen Pract, 2016, 66(649): e568-76. DOI: 10.3399/bjgp16X685609.

[24] YI X Y, CHEN H, YU M, et al. Persistence of drug therapy is associated with ischemic stroke and other vascular events in high-risk stroke population[J]. Front Neurol, 2022, 13: 925061. DOI: 10.3389/fneur.2022.925061.

(Received: July 10, 2025; Revised: September 2, 2025)
(Editor: MAO Yamin)