Abstract
Background:
Despite significant achievements in healthcare quality improvement for ischemic stroke and transient ischemic attack (TIA), the advancements and challenges of patient adherence to secondary prevention medications remain unclear.
Objectives:
We aimed to investigate adherence rates of secondary prevention medication and identified the key determinants in Chinese patients with ischemic stroke or TIA.
Design:
This is an observational study.
Methods:
Using the China National Stroke Registry (CNSR) database from 2007 to 2018, this observational study included patients with ischemic stroke or TIA who were admitted to the hospital within 7 days of symptom onset. The study outcome was the patient adherence to secondary prevention medications, which was defined as the consistent use of prescribed antithrombotic, lipid-modulating, antidiabetic, and antihypertensive medications post-discharge using an “all-or-none” approach. We calculated adherence rates in 3 and 12 months. Logistic regression models were used to evaluate influencing factor patterns and challenges in the improvement of patient adherence.
Results:
A total of 12,873 patients from CNSR I and 15,099 patients from CNSR III were included. Patient adherence rates for secondary prevention medications increased from 66.97% in CNSR I to 80.76% in CNSR III (p < 0.0001) in 3 months, and from 35.08% to 59.81% (p < 0.0001) in 12 months. Patient age, the National Institute of Health stroke scale score at admission, disease diagnosis, the Trial of Org 10172 in Acute Stroke Treatment classification, family income per capita, alcohol consumption, dyslipidemia history, hypertension history, diabetes history, and heart disease history appeared to exhibit a significant association with adherence.
Conclusion:
In spite of the remarkable progress in patient adherence to secondary prevention of stroke from 2007 to 2018, challenges remain in sustaining quality improvement initiatives, necessitating further improvements by addressing disease severity, lifestyle, medical history, and socioeconomic factors.
Introduction
Ischemic stroke, accounting for the majority of stroke cases, not only inflicts immediate neurological deficits but also carries a substantial risk of recurrence, which can lead to severe consequences for individuals and their families.1–3 Transient ischemic attack (TIA) is a temporary disruption of blood flow to the brain that causes short-lived neurological symptoms, serving as a warning sign for potential future stroke.4,5 The growing recognition of stroke recurrence as a critical public health concern underscores the urgent need for effective prevention strategies to mitigate this risk and improve long-term outcomes for stroke survivors.6–8 Secondary prevention of ischemic stroke and TIA is a cornerstone for reducing the likelihood of subsequent cerebrovascular events, managing these conditions, which typically involves a multifaceted pharmacological regimen tailored to address various risk factors, including antithrombotic, lipid-modulating, antidiabetic, and antihypertensive medications.7,9,10
Despite the proven efficacy of secondary prevention medications, their success hinges largely on patient adherence to prescribed medications.11,12 Nonadherence not only diminishes the potential benefits of these medications but also increases the risk of stroke recurrence, disability, and morbidity. 13 Therefore, understanding patients’ adherence to secondary prevention medications and obstacles impeding adherence is pivotal in formulating targeted interventions and ultimately reducing stroke recurrence while enhancing patient outcomes.14,15 While healthcare systems and providers have made strides in enhancing the quality of care and reducing the recurrence rate of stroke through continuous quality improvement initiatives in China, it remains unclear whether these measures have exerted any influence on perceptions and behaviors of patients with ischemic stroke or TIA.16–18 In addition, the patterns of factors influencing patient adherence to secondary prevention medications and the existing challenges remain insufficiently understood.19,20
To bridge this knowledge gap and advance quality improvement, this study investigated adherence rates of secondary prevention medications and identified the key determinants in patients with ischemic stroke or TIA, utilizing the China National Stroke Registries (CNSR I and III).
Methods
Data source and study population
The source population for the observational study was the entire population of the CNSR I and CNSR III, which are nationwide prospective registries of patients with acute cerebrovascular events used to evaluate healthcare quality. The CNSR I recruited 21,902 patients with acute cerebrovascular events within 14 days of the index event from 132 hospitals between September 2007 and August 2008. The CNSR III recruited 15,166 patients with ischemic stroke or TIA within 7 days of the symptom onset between August 2015 and March 2018. The comprehensive details have already been published elsewhere.21,22 This study encompassed patients who were admitted to the hospital within 7 days of symptom onset and enrolled in the CNSR I or CNSR III, with a diagnosis of ischemic stroke or TIA. The diagnosis was made in accordance with the criteria set forth by the World Health Organization and confirmed by magnetic resonance imaging or brain computed tomography.8,23 The study conformed to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) reporting guidelines. 24
Outcomes
At 3 months post-enrollment, patients underwent face-to-face interviews, while at 6 and 12 months, trained research coordinators contacted them via telephone. During these interactions, data pertaining to adherence to prescribed secondary prevention medications, survival status, and instances of stroke recurrence were systematically collected. Verification of vascular events was obtained from the respective hospitals, whereas suspected cases of recurrent cerebrovascular events that did not result in hospitalization were evaluated by an independent endpoint adjudication committee. Stroke recurrence can manifest in various forms, including cerebral infarction, intracerebral hemorrhage, and subarachnoid hemorrhage.
The study outcome was the patient adherence to secondary prevention medications in 3 and 12 months using an “all-or-none” approach. 25 In this context, patients were deemed adherent solely when they consistently complied with all prescribed medications. Any deviation from this full adherence led to their classification as non-adherent. Specifically, 3-month adherence was defined as patient-reported consistent use of all discharge-prescribed antithrombotic (including antiplatelet agents and anticoagulants), lipid-modulating, antidiabetic, and antihypertensive medications at the 3-month follow-up assessment, while 12-month adherence indicated their continuous use of these medications during follow-up visits at 3, 6, and 12 months after enrollment. In the classification and definition of secondary prevention medications, we chose to use the term “antithrombotic agents” rather than classifying “antiplatelet agents” and “anticoagulants” as two separate categories, considering the dynamic nature of patients’ conditions and the corresponding adjustments required in antithrombotic strategies.26,27 Furthermore, we chose to employ the term “lipid-modulating agents” rather than specifically “statins,” recognizing that some patients may switch to alternative medications due to intolerance to statins.9,10
Variables
The study incorporated a series of key variables, including patient demographics such as age and gender, lifestyle factors including smoking (current or past) and alcohol consumption (current or past), and medical history encompassing previous stroke, TIA, dyslipidemia, diabetes, hypertension, and heart disease. In addition, final diagnosis (categorized as either ischemic stroke or TIA), the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification for ischemic stroke, and clinical severity upon admission, assessed using the National Institutes of Health Stroke Scale (NIHSS) score, were considered. We also included family income per capita, with thresholds set at 3000 Chinese yuan (CNY) for CNSR I and 2300 CNY for CNSR III monthly based on the information collection. Treatment modalities administered, including intravenous thrombolysis and endovascular therapy, were also included as variables in the analysis. Endovascular therapy refers to arterial thrombolysis, mechanical thrombectomy, or stent therapy, regardless of whether intravenous thrombolysis was administered.
Statistical analysis
We used descriptive statistics to characterize patient categories. Accordingly, baseline characteristics were presented as either the mean with standard deviations for normally distributed continuous variables or the median with interquartile range (IQR) for skewed continuous variables, while categorical variables were summarized by the number of patients with percentages. Regarding adherence, patients who were lost to follow-up or had incomplete adherence data were categorized as non-adherent. Regarding other variables, our analysis revealed no missing data. Patients who died during the corresponding follow-up period were excluded from the analysis owing to the inaccessibility of adherence data. During the study period from 2007–2008 (CNSR I) to 2015–2018 (CNSR III), the adherence rates to secondary prevention medications (including adherence rates for specific drug categories) and the recurrence rates of stroke in 3 and 12 months were calculated. Also, we calculated the prescription rates of secondary prevention medications for patients at discharge.
To investigate the influencing factor patterns of patient adherence within 3- and 12-month timeframes in 2007–2008 and 2015–2018, logistic regression analysis was performed, integrating the aforementioned variables, with odds ratios (OR) and corresponding 95% confidence intervals (CI) estimated. Furthermore, data from two registries were merged to form a unified dataset for further analysis, in which study period (CNSR I or CNSR III) and interaction terms of study period and covariates were additionally included in the logistic regression model. An interaction test was conducted to evaluate how the study period modified the associations between influencing factors and patient adherence. We also carried out sensitivity analyses to ensure the robustness of our findings. First, we classified patients with an adherence rate of 75% or above as adherent. To elaborate further, given the four medication categories that constituted our focus, a patient was considered adherent if they adhered to at least three of these categories. Second, in light of the potential bias stemming from categorizing patients lost to follow-up as non-adherent, we excluded such patients from our analysis.
To further elucidate the impact of these influencing factors on adherence, we integrated statistically significant factors derived from any database analysis into logistic regression models to explore their associations with adherence to the four drug categories (antithrombotic, lipid-modulating, antidiabetic, and antihypertensive medications) over the 3- and 12-month periods. Specifically, in our analysis of adherence over a 3-month period, we included four variables: age, NIHSS score at admission, final diagnosis, TOAST classification for ischemic stroke, family income per capita, history of diabetes, and history of heart disease. For the 12-month analysis, we incorporated the following variables: NIHSS score at admission, TOAST classification for ischemic stroke, family income per capita, alcohol consumption history, history of dyslipidemia, history of diabetes, and history of hypertension. All analyses were performed using SAS statistical software, version 9.4 (SAS Institute, Inc.; Cary, NC, USA), and statistical significance was defined as a two-tailed p value of <0.05.
Results
Baseline characteristics
After excluding patients enrolled more than 7 days after symptom onset or with missing data of time from onset to enrollment, 12,873 patients (7917 (61.50%) were men; median (IQR) age, 67 (57–75) years; 11,757 ischemic stroke and 1116 TIA) from CNSR I and 15,099 patients (10,311 (68.29%) were men; median (IQR) age, 63 (54–70) years; 14,083 ischemic stroke and 1016 TIA) from CNSR III were included (Table 1, Supplemental Figure 1). Upon discharge, the proportions of patients prescribed antithrombotic, lipid-modulating, antidiabetic, and antihypertensive medications increased from CNSR I to CNSR III (p < 0.0001 for all; Figure 1).
Baseline characteristics of the study population.
Median with interquartile range.
Family income per capita >3000 CNY for CNSR I and >2300 CNY for CNSR III monthly.
CNSR, China National Stroke Registry; CNY, Chinese yuan; NA, not applicable; TIA, transient ischemic attack; TOAST, Trial of Org 10172 in Acute Stroke Treatment.
Prescription rate of secondary prevention medications at discharge.
Adherence and stroke recurrence rates
Over the 3-month period, patient adherence rates for secondary prevention medications increased from 66.97% in CNSR I to 80.76% in CNSR III (p < 0.0001), and over 12 months, these rates rose further from 35.08% to 59.81% (p < 0.0001). For specific medication categories, adherence rates also increased from CNSR I to CNSR III over both 3- and 12-month periods (Figure 2). Regarding stroke recurrence in CNSR I and III, the incidence rate decreased significantly from 12.68% to 5.86% over 3 months (p < 0.0001) and from 16.05% to 8.97% over 12 months (p < 0.0001).
Patient adherence rate of secondary prevention medications in 3 and 12 months.
Influencing factor patterns of adherence in 3 months
The logistic regression analysis revealed that higher NIHSS score at admission (OR (95% CI) 1.011 (1.004, 1.019), p = 0.0017 in CNSR I; OR (95% CI) 1.021 (1.010, 1.033), p = 0.0003 in CNSR III), diagnosis of ischemic stroke (OR (95% CI) 1.626 (1.413, 1.870), p < 0.0010 in CNSR I; OR (95% CI) 1.322 (1.103, 1.585), p = 0.0026 in CNSR III), undetermined etiology (OR (95% CI) 1.263 (1.145, 1.393), p < 0.0001 in CNSR I; OR (95% CI) 0.793 (0.711, 0.884), p < 0.0001 in CNSR III; with reference to large artery atherosclerosis (LAA)), and higher family income per capita (OR (95% CI) 1.273 (1.170, 1.385), p < 0.0001 in CNSR I; OR (95% CI) 1.176 (1.082, 1.278), p = 0.0001 in CNSR III) were positively associated with 3-month medication adherence in both CNSR I and III. Moreover, previous diabetes and heart disease were negatively associated with medication adherence in CNSR I. Regarding CNSR III, older age, cardioembolism, and small artery occlusion were negatively associated with adherence. In the combined dataset, there were interactions of study period with the association of undetermined etiology with adherence (p < 0.0001; Figure 3). The sensitivity analysis yielded results that were comparable to the primary findings (Supplemental Figures 2 and 3).
Factors influencing patient adherence to secondary prevention medications in 3 months.
Influencing factor patterns of adherence in 12 months
In 12 months, higher NIHSS score at admission (OR (95% CI) 1.021 (1.014, 1.029), p < 0.0001 in CNSR I; OR (95% CI) 1.015 (1.006, 1.025), p = 0.0009 in CNSR III), undetermined etiology (OR (95% CI) 1.812 (1.648, 1.992), p < 0.0001 in CNSR I; OR (95% CI) 0.899 (0.824, 0.979, p = 0.0149 in CNSR III; with reference to LAA), higher family income per capita (OR (95% CI) 1.430 (1.316, 1.554), p < 0.0001 in CNSR I; OR (95% CI) 1.127 (1.053, 1.206), p = 0.0006 in CNSR III), and a history of diabetes (OR (95% CI) 0.796 (0.719, 0.881), p < 0.0001 in CNSR I; OR (95% CI) 0.873 (0.805, 0.946), p = 0.0009 in CNSR III) were associated with medication adherence in both CNSR I and III. In addition, cardioembolism and a history of hypertension were statistically significantly associated with medication adherence in CNSR I. In CNSR III, small artery occlusion, alcohol consumption, and a history of dyslipidemia indicated a significant impact on patient adherence. In the combined dataset, significant interactions were observed between the study period and the relationship of ischemic stroke diagnosis and undetermined etiology with adherence (p < 0.0001 for both; Figure 4). The sensitivity analyses reported results that were similar to the main findings (Supplemental Figures 4 and 5).
Factors influencing patient adherence to secondary prevention medications in 12 months.
The impact of influencing factors on adherence by medication categories
Patient age significantly influenced adherence to antithrombotic and antihypertensive medications. Higher NIHSS score at admission was associated with adherence to antithrombotic, lipid-modulating, and antihypertensive medications. Also, diagnosis of ischemic stroke and alcohol consumption history significantly influenced adherence to antithrombotic and lipid-modulating medications. The history of dyslipidemia, diabetes, and hypertension was associated with adherence to corresponding secondary prevention drug categories. Moreover, a significant impact of TOAST classification and higher family income per capita on adherence to all four medication categories was observed (Supplemental Tables 1 and 2).
Discussion
Focusing on secondary prevention medication adherence and its determinants among Chinese patients with ischemic stroke or TIA, this observational study encompassed a series of progressively analyses: Initially, to evaluate the effectiveness of healthcare quality improvement efforts spanning from 2007 to 2018, we examined patient adherence to secondary prevention medications and the recurrence rates of stroke during the periods of 2007–2008 (CNSR I) and 2015–2018 (CNSR III). Subsequently, we explored the potential factors influencing medication adherence at both 3- and 12-month intervals, with the aim of pinpointing areas for potential enhancement strategies. Finally, we conducted a detailed examination of how statistically significant influencing factors impact adherence to specific medication categories. Based on these insights, we outlined key priorities for enhancing future clinical practice and shaping health policy.
Over the decade spanning from 2007–2008 to 2015–2018, China witnessed a substantial increase in patient adherence to pharmacological interventions for the secondary prevention of stroke. This encouraging progression can most likely be attributed to several pivotal factors. First, the continuous improvement in healthcare quality may have positively influenced patients’ awareness regarding treatment adherence, driven by effective measures implemented by healthcare institutions.14,19,28,29 Second, the revision of clinical practice guidelines, particularly the heightened emphasis on the paramount significance of secondary prevention, has profoundly influenced medical practice and patient engagement.6,30 Third, the advancement of news media, coupled with its increased focus on popular science of health regarding stroke, has significantly contributed to raising public awareness of health issues. 31 Last, the advancement of socioeconomic conditions and the effective execution of national pharmaceutical policies through these years, which have significantly improved medication accessibility and affordability, has likely contributed to this advancement as well.32,33
Regarding influencing factors of adherence, the severity of illness, as indicated by the NIHSS score at admission, predominantly affects patient adherence to antithrombotic and lipid-modulating medications, which was similar to a previous study. 34 Patients with milder strokes, as evidenced by lower NIHSS scores, may underestimate the long-term risks linked to their condition and the significance of secondary prevention measures. 35 This underscores the necessity for healthcare professionals to prioritize patient education, particularly for those with milder conditions. 36 Furthermore, tailored follow-up plans and reminder systems can further ensure patients with mild stroke or TIA adhere to their treatment, boosting overall adherence and improving long-term health outcomes. 37 Regarding the TOAST classification, despite the presence of certain findings with statistical significance, clinical experience fails to provide support. This discrepancy may potentially be attributed to the distribution of sample sizes. Hence, large-scale studies in the future are warranted to further delve into the association between the TOAST classification and adherence to secondary prevention medications.
The significant impact of disease diagnosis on medication adherence in 2007–2008, but not in 2015–2018, mirrors the evolving landscape of stroke prevention and management. During the earlier period, there might have been a lack of sufficient awareness and consensus concerning the optimal utilization of secondary prevention medications for various stroke-related diagnoses. 38 However, with advancements in scientific research and clinical practice, both healthcare professionals and patients developed a heightened awareness of the significance of secondary prevention medications to TIA patients, which probably played a role in reducing the impact of disease diagnosis on medication adherence during the subsequent period.39–41 However, it remains imperative to persistently monitor and guarantee that every patient with stroke-related diagnoses, encompassing TIA, receives appropriate secondary prevention care. 42
Family per capita income consistently stands out as a key factor that significantly influences patient adherence, especially to lipid-modulating and antihypertensive drugs. From the patients’ perspective, financial limitations can restrict their capacity to afford these typically long-term medications, even when they are aware of their crucial role in preventing secondary health issues. 43 Therefore, healthcare providers could deliver more tailored counseling on the long-term cost-effectiveness of these medications, and highlight how consistent use can prevent more costly and debilitating health problems down the line. 44 Furthermore, this finding calls for initiatives to enhance medication accessibility for low-income populations. Despite continuous reform and optimization of pharmaceutical policies, which have significantly improved drug accessibility and affordability, economic challenges still persist. 45 Governments might consider providing subsidies for these essential secondary preventive medications or incorporating them into public health initiatives. 32 In addition, efforts to increase public awareness about the importance of these medications should be stepped up, particularly in economically disadvantaged areas. 46
The significant negative correlation between alcohol consumption and secondary prevention medication adherence in 2015–2018 was a concerning finding. Alcohol consumption can have multiple detrimental effects on health, and in the context of stroke prevention, it may interfere with patients’ ability to adhere to their medication regimens. 47 Patients with alcohol use may face challenges such as forgetfulness, lack of motivation, or competing priorities related to their alcohol use, which can lead to missed doses or complete nonadherence to secondary prevention drugs. 48 Healthcare providers should assess the extent of alcohol consumption and its potential impact on the patient’s lifestyle and treatment plan. 49 This could involve developing tailored strategies to address specific barriers to adherence, such as setting up reminder systems, providing education on the interaction between alcohol and medications, and offering support for alcohol cessation or reduction. 47
This study revealed a complex and dynamic relationship between patients’ medical history and their adherence to secondary prevention medications over time. A history of dyslipidemia was positively associated with secondary prevention adherence from 2015 to 2018. Further analysis demonstrated a negative association with lipid-modulating medication adherence in 2007–2008 but a positive association in 2015–2018. A previous study also reported similar findings. 34 This shift indicates that lipid control treatment has garnered wider attention in patients with a history of dyslipidemia and has achieved notable success. 50 However, for patients without a history of dyslipidemia, attention needs to be paid to enhancing their adherence to lipid-modulating medications. By contrast, during both the 2007–2008 and 2015–2018 periods, a history of diabetes exhibited a negative correlation with adherence to secondary prevention medications and antidiabetic medications. This seems to imply that we ought to show more concern for and come up with more strategies to improve the secondary prevention adherence of patients who have a history of diabetes. Also, these results highlight the persistent challenges in attaining optimal medication adherence for blood glucose management. 9 Moreover, healthcare system-related factors, such as restricted access to affordable medications, insufficient follow-up care, and inadequate patient education initiatives, could also contribute to the suboptimal adherence rates. 13
Beyond patient-related factors, a range of strategies can be implemented to bolster patient adherence. At the healthcare facility level, this entails setting up dedicated stroke follow-up clinics or secondary prevention management centers, creating individualized patient profiles tailored to secondary prevention needs, and assembling a multidisciplinary team comprising neurologists, pharmacists, nurses, rehabilitation specialists, and other experts for comprehensive patient care. 51 On the community front, initiatives such as organizing community health education campaigns, establishing a structured community follow-up network, and facilitating patient support groups can play a pivotal role. 52 From a policy perspective, enhancing medical insurance reimbursement rates or introducing dedicated funds, instituting medication adherence assessment metrics, and tightening regulations on drug effectiveness and safety are crucial steps.45,53 In the realm of technological innovation, the development of intelligent medication reminder applications and the widespread adoption of electronic prescription and medication delivery services hold significant promise.
This study bears several limitations that warrant consideration. First, the analysis centers on the general landscape and deficiencies in accordance with the guideline recommendations, rather than delving into the specifics of individual patient cases. It is plausible that certain patients may no longer necessitate particular medications due to evolving medical conditions or re-assessments, data for which was not captured. Nonetheless, this limitation does not compromise our findings. Second, baseline characteristics differed between patients from CNSR I and CNSR III. However, we controlled for these variables, data sources, and their interactions in the analysis. Third, the datasets from the two registries lack information concerning the underlying reasons for patient non-adherence. Consequently, future research endeavors should explore this facet to attain a more comprehensive understanding of the issue. Fourth, deceased patients were excluded due to the unavailability of their adherence information, consequently resulting in failing to account for the impact of non-adherence on mortality and the increased likelihood of non-adherence among patients with more severe conditions. Future studies should delve into the association between patient adherence and mortality, taking potential covariates into account.
Conclusion
From 2007 to 2018, a decade of quality improvement initiatives led to substantial improvements in patient adherence to secondary prevention of stroke, reflecting significant progress in healthcare quality of ischemic stroke and TIA. However, patient medication adherence remains suboptimal and requires further improvement, particularly by accounting for factors such as disease severity, lifestyle, medical history, and socioeconomic status. Looking ahead, sustaining healthcare quality improvement demands ongoing commitment, innovation, and collaboration from policymakers, healthcare professionals, and patients across the healthcare ecosystem.
Supplemental Material
sj-docx-1-tan-10.1177_17562864251406061 – Supplemental material for Advancements and challenges of adherence to secondary prevention medications among patients with ischemic stroke or transient ischemic attack: the healthcare quality improvement in China, 2007–2018
Supplemental material, sj-docx-1-tan-10.1177_17562864251406061 for Advancements and challenges of adherence to secondary prevention medications among patients with ischemic stroke or transient ischemic attack: the healthcare quality improvement in China, 2007–2018 by Yuan Shen, Xinya Li, Xue Xia, Meng Gao, Xue Tian, Qin Xu, Xiaoli Zhang, Ruobing Tian, Xia Meng and Anxin Wang in Therapeutic Advances in Neurological Disorders
Supplemental Material
sj-docx-2-tan-10.1177_17562864251406061 – Supplemental material for Advancements and challenges of adherence to secondary prevention medications among patients with ischemic stroke or transient ischemic attack: the healthcare quality improvement in China, 2007–2018
Supplemental material, sj-docx-2-tan-10.1177_17562864251406061 for Advancements and challenges of adherence to secondary prevention medications among patients with ischemic stroke or transient ischemic attack: the healthcare quality improvement in China, 2007–2018 by Yuan Shen, Xinya Li, Xue Xia, Meng Gao, Xue Tian, Qin Xu, Xiaoli Zhang, Ruobing Tian, Xia Meng and Anxin Wang in Therapeutic Advances in Neurological Disorders
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
