Effects of Clinical Pharmacist-Led Medication Management on Medication Adherence and Quality of Life in Stroke Patients Undergoing Intravenous Thrombolysis

Introduction

Acute ischemic stroke (AIS) is one of the leading causes of death and disability among adults worldwide [1]. Intravenous thrombolysis (IVT), as one of the most effective reperfusion therapies during the acute phase, has been shown to significantly improve neurological outcomes [2,3]. However, following thrombolysis, patients who have had AIS typically require long-term use of lipid-lowering, antiplatelet, anticoagulant, and other medications [4,5]. Poor adherence or inadequate lipid control may markedly impair long-term outcomes [6]. Low-density lipoprotein cholesterol (LDL-C) is closely associated with the risk of stroke recurrence [7], and evidence suggests that the lower the LDL-C level, the lower the recurrence rate [8]. Consequently, intensive lipid-lowering therapy is a critical component of secondary prevention [9]. Vascular and thrombotic risk is highly heterogeneous and influenced by genetic, metabolic, and clinical factors across different patient populations and disease settings [10–12]. Despite guideline recommendations that statins should be routinely prescribed after stroke, in real-world practice, patients often demonstrate poor medication adherence and insufficient follow-up monitoring, leading to suboptimal lipid control and reduced effectiveness of interventions.

Clinical medication management, with the active participation of clinical pharmacists, has emerged as a key strategy for optimizing drug therapy. In recent years, this approach has shown significant benefits in the management of chronic diseases, such as hypertension and diabetes [13–15]. Previous studies have demonstrated that clinical pharmacists can improve adherence and clinical outcomes through individualized medication education, behavioral reminders, adverse reaction monitoring, and drug–drug interaction management [16,17]. Nonetheless, the paradigm of precision medicine requires that evidence be tailored to specific high-risk subgroups, such as patients after IVT for AIS. This population often faces multiple medication regimens, poor treatment adherence, and neurological deficits that further compromise continuity of care after discharge, placing them at high risk for adverse outcomes and making optimal secondary prevention paramount. However, dedicated investigations into pharmacist-led support for this unique post-IVT cohort are lacking. Therefore, evaluating structured, professional support to bridge this gap is both urgent and necessary.

To address this evidence gap, this study focuses specifically on patients with AIS who have undergone IVT. This population provides a distinct methodological advantage: their relative homogeneity and standardized initial management help minimize confounding, offering a clearer assessment of the effect of structured post-discharge support.

Therefore, in the present study, we aimed to investigate the effects of a clinical pharmacist-led medication management model on medication adherence, lipid control, and functional outcomes in this defined cohort. By doing so, we seek to generate targeted evidence to inform and optimize post-discharge care strategies for this high-risk population.

Material and Methods

CLINICAL PHARMACIST-LED MEDICATION MANAGEMENT:

A treatment medication management team was established with clinical pharmacists as the core members, who were responsible for the management of medication use in the intervention group.

Pharmacists reviewed prescribing rules for statins and maintained the pre-prescription review system, which covered but was not limited to dosage, route of administration, drug–drug interactions, and contraindications. Upon enrollment, pharmacists performed pre-medication risk assessments, screened for contraindications, stratified risks, and set individualized lipid-lowering targets. For patients with atherosclerotic ischemic stroke or transient ischemic attack, the target values were set as LDL-C <1.8 mmol/L and non–HDL-C <2.6 mmol/L. Commonly prescribed medications included atorvastatin 20 mg once daily or rosuvastatin 10 mg once daily.

Before and after the first administration, patients received medication counseling, including of all secondary prevention medications. During hospitalization, pharmacists continuously monitored clinical symptoms and laboratory indices, assisted by an adverse drug reaction intelligent monitoring system and routine ward rounds, and actively reported any adverse drug reactions. Patients were guided to use the “Medicine Safety Assistant” WeChat mini-program, with personalized reminders set for dosing times and frequencies. At discharge, pharmacists provided education on secondary prevention of ischemic stroke, explained precautions for different drug categories, and emphasized follow-up schedules.

To ensure standardization across clinical pharmacists, the following measures were implemented: (1) all participating pharmacists underwent uniform training on the intervention protocol prior to study initiation; (2) standardized assessment tools (eg, medication review checklists) and patient education materials were used; and (3) regular case discussions and quality checks were conducted to ensure adherence to the study protocol.

This study was approved by the Ethics Committee of Jingjiang People’s Hospital (approval No. 2022-KY-017-01). Written informed consent was obtained from all patients or their legal representatives.

STUDY POPULATION:

This study was a retrospective cohort analysis of a prespecified subgroup derived from a prospectively established and randomized database. The source database included consecutive patients with AIS admitted to the Department of Neurology, Jingjiang People’s Hospital, between January and October 2023. At the point of admission, these patients were prospectively randomized in real-time using a computer-generated random number sequence (EpiCalc 2000). Patients assigned odd numbers were allocated to the control group (n=400), and those assigned even numbers, to the pharmacist intervention group (n=400). This randomization defined the original prospective cohorts.

For the purpose of the present study, we conducted a retrospective analysis focusing on a specific subgroup from this database: patients who underwent IVT. To obtain a homogeneous study population and minimize confounding related to differential reperfusion strategies and post-procedural complexity, we excluded patients who underwent mechanical thrombectomy. From the original 800 randomized patients, we identified and extracted the IVT subgroup (control: n=93; intervention: n=97) for comparative outcome analysis.

The inclusion criteria were as follows: age >18 years; first-ever AIS confirmed by clinical examination and computed tomography or magnetic resonance imaging; treatment with IVT; and written informed consent obtained from the patient or their legal representative. The exclusion criteria were as follows: underwent mechanical thrombectomy; history of recurrent AIS; use of statins or fibrates within 1 month prior to AIS onset; requirement for intensive statin therapy; presence of malignant tumor, chronic liver disease, active liver disease, hepatic insufficiency, alanine aminotransferase level >3 times the upper limit of normal, severe renal insufficiency (creatinine clearance <30 mL/min), or myopathy; life expectancy <3 months or inability to complete the study for other reasons; and considered unsuitable for participation by the investigators.

Both groups received treatment in accordance with AIS management guidelines, including antiplatelets, anticoagulants, statins, fibrinolytics, volume expansion, and neuroprotection as individualized combinations.

INTRAVENOUS THROMBOLYSIS:

According to the Chinese guidelines for the diagnosis and management of AIS, all patients met the indications for IVT and had no contraindications to IVT [18]. Whether the patient underwent IVT was jointly determined by the experienced neurologists and the patient’s caregivers. The dose of IVT was recombinant tissue plasminogen activator 0.9 mg/kg, and the maximum dose was 90 mg; 10% of the total dose was intravenously injected, and 90% of the dose was infused via micropump within 1 hour.

DATA COLLECTION:

Baseline demographic and clinical data were collected, including age, sex, smoking and alcohol history, past medical history (hypertension, diabetes, hyperlipidemia, coronary heart disease, atrial fibrillation, valvular heart disease), lipid profile at admission, National Institutes of Health Stroke Scale (NIHSS) score, baseline medication adherence, and length of hospital stay.

OUTCOME MEASURES:

The 3 primary outcomes were lipid profiles at 1 month after discharge, including changes in LDL-C level; medication adherence; and clinical outcomes after 3 months. (1) Lipid profiles at 1 month after discharge and the reduction in LDL-C level – defined as the difference between the level at admission and the level at 1 month after discharge – were assessed. (2) Medication adherence at 3 months after discharge was assessed using the medication adherence questionnaire (MAQ). The MAQ was assessed by the same dedicated clinical pharmacist for all patients in both groups. This assessing pharmacist was fully blinded to patient group allocation (intervention vs control), was not involved in any aspect of the inpatient medication management intervention, and served solely as an independent outcome assessor. The assessments were conducted at 2 time points: a baseline face-to-face interview at admission and a follow-up telephone interview at 3 months after discharge. At both times, the assessor had no information regarding the patient’s assignment during the index hospitalization. This instrument consists of 4 yes/no questions. A total score of 4 indicates high adherence, 2–3 indicates moderate adherence, and 0–1 indicates low adherence. Medication adherence and lipid parameters were evaluated as independent outcomes; (3) Clinical outcomes at 3 months were evaluated using the modified Rankin Scale (mRS). Scores on the mRS range from 0 to 6, with higher scores indicating worse neurological function; a score of 0–2 represents a favorable outcome, and a score of 6 indicates death.

STATISTICAL ANALYSIS:

All statistical analyses were performed using SPSS version 26.0. Continuous variables with normal distribution are expressed as mean±standard deviation (SD) and were compared using independent samples t tests. Non-normally distributed data are expressed as median (interquartile range, IQR) and were compared using Mann-Whitney U tests. Categorical variables are presented as numbers (percentages), and comparisons between groups were conducted using chi-square tests or Fisher exact tests as appropriate. A 2-sided P value <0.05 was considered statistically significant.

To address the potential for selection bias, we performed a comparative analysis of the baseline characteristics between the intervention and control groups, after excluding those patients who did not return for the 1-month assessment.

To rigorously address whether the observed LDL-C reduction was driven by the pharmacist intervention or by the preferential use of rosuvastatin, we conducted multivariable linear regression analysis. The model was specified as follows: dependent variable: delta LDL-C (baseline level minus follow-up level); independent variables: treatment group (intervention vs control), statin type (rosuvastatin vs others), baseline LDL-C level, age, and sex.

Then, we focused on the achievement of the maximum MAQ score, which reflects the absence of self-reported non-adherent behaviors. Covariates were selected using a dual strategy: based on clinical reasoning, the key variables of age, baseline LDL-C, baseline NIHSS score, history of diabetes, and history of hyperlipidemia were forced into the final model irrespective of their univariate significance. To account for other potential confounders, all remaining candidate variables (eg, sex, smoking status) were first screened in univariate analysis; those demonstrating a marginal association (P<0.10) with the outcome were then eligible for inclusion in the multivariable model.

Results

BASELINE CHARACTERISTICS:

The median age in the control group was 71 years (IQR, 59–81), with 59 men (63.4%), while the median age in the intervention group was also 71 years (IQR, 60–80), with 59 men (60.8%). The 2 groups were comparable at baseline, with no significant differences in smoking history, alcohol history, or comorbidities. Specifically, the median NIHSS score at admission was 3 (IQR, 2–5) in both groups (P=0.909), and baseline LDL-C levels were similar (control: 3.16 mmol/L [2.57–3.59] vs intervention: 3.01 mmol/L [2.62–3.57]). A full comparison of baseline characteristics is presented in Table 1.

TREATMENT CHARACTERISTICS:

In the control group, 77 patients (82.8%) received atorvastatin 20 mg/day, 11 (11.8%) received rosuvastatin 10 mg/day, 4 (4.3%) received simvastatin 20 mg/day, and 1 (1.1%) received pitavastatin 2 mg/day. In the intervention group, 75 patients (77.3%) received atorvastatin 20 mg/day, 20 (20.6%) received rosuvastatin 10 mg/day, 1 (1.0%) received simvastatin 20 mg/day, and 1 (1.0%) received pitavastatin 2 mg/day. There was no statistically significant difference in the choice of statin therapy between the groups (P=0.225; Table 2).

The median length of hospital stay was 9 days (IQR, 7–12) in the control group and 10 days (IQR, 7–13) in the intervention group, with no significant difference between the 2 groups (P=0.381).

LIPID LEVELS, MEDICATION ADHERENCE, AND CLINICAL OUTCOMES:

At 1 month after discharge, 78 patients (83.9%) in the control group and 88 patients (90.7%) in the intervention group returned for lipid re-examination. LDL-C levels were significantly lower in the intervention group, compared with the control group (2.11 [1.64–2.53] vs 2.33 [1.80–2.96] mmol/L; P=0.037). No significant differences were observed between the groups in total cholesterol, triglycerides, or high-density lipoprotein cholesterol (all P>0.05). The reduction in LDL-C was significantly greater in the intervention group, compared with the control group (0.98 [0.38–1.46] vs 0.70 [0.22–1.31] mmol/L; P=0.044; Table 3). As detailed in Table 4, an analysis conducted after excluding patients who did not return for the 1-month assessment demonstrated that the 2 groups remained well balanced in key baseline demographic and clinical variables, even within the cohort with complete follow-up data.

The 3-month MAQ score was significantly higher in the intervention group (3 [IQR, 3–4]) than in the control group (3 [IQR, 2–3]; P<0.001; Table 3). However, there was no significant difference in the median mRS score between the intervention and control groups (both 1 [IQR, 0–3]; P=0.643).

MULTIVARIABLE LINEAR REGRESSION ANALYSIS OF FACTORS ASSOCIATED WITH LDL-C REDUCTION:

After adjusting for baseline LDL-C, statin type, age, and sex, analysis revealed that inclusion in the pharmacist-led intervention group (β=0.239 mmol/L; P=0.016) and a higher baseline LDL-C level (β=0.724 mmol/L; P<0.001) were independently associated with a greater reduction in LDL-C. The variable for statin type (rosuvastatin vs others) was not a statistically significant predictor in this model (P=0.374; Table 5).

UNIVARIATE AND MULTIVARIATE PREDICTORS FOR ACHIEVING THE HIGHEST MAQ SCORE:

A significantly higher proportion of patients in the pharmacist-led management group achieved optimal medication adherence (MAQ score of 4) at 90 days, compared with in the control group (67.8% vs 32.2%; P=0.002). In the univariate screening of candidate variables, a history of hypertension was also associated with the outcome (P=0.026). Variables forced into the model based on clinical rationale (baseline LDL-C, NIHSS score, diabetes history, and hyperlipidemia history) were not significantly associated in univariate analysis (all P>0.05). After adjustment for these prespecified clinical variables and other covariates selected via univariate screening (history of hypertension, P < 0.10), multivariable logistic regression confirmed that assignment to the clinical pharmacist-led intervention group was independently associated with achieving optimal adherence (adjusted odds ratio [aOR], 3.03; 95% CI, 1.53–5.99; P=0.001). Additionally, a history of hypertension (aOR, 2.67; 95% CI, 1.09–6.51; P=0.031) and a history of hyperlipidemia (aOR, 8.18; 95% CI, 1.30–51.71; P=0.025) were independent predictors in the final model (Table 6).

Discussion

Findings from this retrospective study suggest that clinical pharmacist-led medication management may improve medication adherence and early lipid control in patients with AIS receiving IVT.

The results showed that the median 3-month medication adherence score in the intervention group was 3 (IQR, 3–4), which was significantly higher than that in the control group, 3 (IQR, 2–3; P<0.001). This finding indicates that systematic pharmacist interventions, including individualized counseling and regular follow-up, effectively enhanced patients’ awareness of the importance of secondary prevention and increased their willingness to persist with therapy. Achieving the maximum MAQ score represents optimal medication-taking behavior without reported non-adherence. Our findings suggest that pharmacist-led medication management was a favorable factor associated with this optimal behavior. Pharmacist-led medication management not only helped patients understand the mechanism of action and potential adverse effects of drugs, but also provided timely support when doubts or nonadherence occurred. These interventions played a crucial role in the early recovery phase following AIS. Interestingly, a history of hypertension or hyperlipidemia was independently associated with optimal adherence. This may suggest that patients with pre-existing chronic conditions possess greater health awareness, are more familiar with long-term medication routines, or have established more frequent contacts with the healthcare system, thereby facilitating adherence behavior after AIS.

Previous research has shown that patients that experience AIS face long-term treatment with multiple medications, such as statins, antiplatelets, and antihypertensives. This study focused on statin therapy and LDL-C as a key objective measure, not to diminish the critical role of antiplatelet medications in AIS management, but to measure the objective and quantifiable nature of LDL-C levels for assessing adherence and control. However, some patients discontinue or interrupt therapy due to insufficient medication knowledge, fear of adverse reactions, economic burden, or lack of confidence in treatment [19,20]. Without timely identification and intervention, such issues can result in poor lipid control, increased risk of recurrent stroke, and reduced quality of life. By offering individualized education and follow-up, clinical pharmacists fill the gaps left by traditional follow-up practices, where medical and nursing staff may have limited capacity or pharmacological expertise. Thus, pharmacist-led management uniquely contributes to improving adherence and ensuring treatment continuity.

In addition, patients in the intervention group achieved significantly lower LDL-C levels at 1 month after discharge, compared with the control group (P=0.037). Moreover, the reduction in LDL-C was significantly greater in the intervention group (0.98 [0.38–1.46] vs 0.70 [0.22–1.31] mmol/L; P=0.044), suggesting its positive effect on early lipid control. LDL-C is a well-established risk factor for atherosclerotic progression and recurrent ischemic stroke [21]. Its control is directly linked to the long-term risk of recurrent cerebrovascular events [22]. By reinforcing patients’ understanding of statin therapy, correcting misconceptions, and addressing adverse events or adherence barriers, pharmacists promoted more precise and individualized lipid-lowering treatment, which may reduce the risk of recurrent stroke due to poor lipid control. Although the LDL-C reduction was modest (2.11 vs 2.33 mmol/L), lower LDL-C levels are associated with reduced long-term vascular risk. The observed difference may therefore reflect earlier lipid control relevant to secondary prevention rather than short-term functional outcomes.

When interpreting our findings, we note the potential influence of the higher proportion of rosuvastatin use in the intervention group. To isolate the intervention effect from that of statin potency, we performed multivariable linear regression analysis. The results showed that after adjusting for baseline LDL-C and statin type (rosuvastatin vs others), the pharmacist-led intervention remained independently associated with a greater reduction in LDL-C (β=0.239 mmol/L; P=0.016) while statin type was not a statistically significant predictor (P=0.374). This indicates that the observed improvement in LDL-C was primarily attributable to the intervention itself–achieved through patient education, enhanced follow-up, and medication management–rather than the specific statin selected.

In addition to considering its overall effect on adherence, considering stroke etiology is also important in the interpretation of medication management. The clinical relevance of specific medications can vary according to stroke etiology. In patients with cardiogenic embolism, anticoagulation is the cornerstone of secondary prevention, and the role of statin therapy may be perceived as less critical. Accordingly, the present study was not intended to assess the relative importance of statins across stroke subtypes. Rather, it focused on medication adherence as a marker of effective post-discharge medication management. From this perspective, improved adherence reflects the contribution of pharmacist-led interventions in enhancing treatment continuity and patient engagement, independent of any single drug class. The improvement in medication adherence observed in this study was based on questionnaire assessment rather than inferred from lipid outcomes.

It is noteworthy that although improvements were observed in medication adherence and LDL-C control, there was no significant difference in mRS scores at 3 months between the 2 groups. This suggests that pharmacist-led medication management may not substantially affect short-term functional recovery. Functional recovery after stroke is influenced by multiple factors, including baseline stroke severity, time to thrombolysis, lesion location, rehabilitation intensity, and multiple factors beyond post-discharge medication use [23,24]. Moreover, neurological improvement often requires longer-term, comprehensive interventions [25]. Therefore, the benefits of pharmacist-led management for functional outcomes require verification over a longer evaluation period (eg, 6 months or more). In the short term, its value lies mainly in enhancing adherence, thereby laying a solid foundation for long-term recovery.

According to our study design, we aimed to evaluate a homogeneous IVT cohort; therefore, we excluded patients who underwent mechanical thrombectomy, which better highlighted the role of medication management in secondary prevention, and demonstrated its core value in improving adherence and biochemical control in this population.

The WeChat-based “Medicine Safety Assistant” mini-program itself is a potential confounder. While the platform served as the primary tool for delivering reminders and facilitating follow-up, its inherent features (eg, medication reminders, information access) may have independently contributed to improved adherence. Therefore, the observed effects should be attributed to the integrated intervention package – consisting of both the clinical pharmacist’s expertise (education, counseling, dose adjustment) and the structured digital support – rather than to the pharmacist’s action alone. Future randomized controlled trials that isolate the effect of the pharmacist’s input from that of the digital tool are warranted.

Several limitations should be acknowledged. First, the study’s retrospective and non-randomized design can introduce residual confounding despite the existence of generally comparable baseline characteristics between groups. Second, the 1-month follow-up reflects early post-discharge medication management and does not capture long-term lipid control or stroke recurrence. Third, medication adherence was assessed using a self-reported questionnaire, which can be subject to reporting bias. Patients with poor medication adherence may have been less likely to return for follow-up, which could introduce selection bias, particularly in lipid re-examination outcomes. Fourth, LDL-C was used as a surrogate marker of lipid management; the observed reduction was modest and may have been influenced in part by differences in statin potency between groups. Fifth, functional outcomes assessed by mRS were exploratory and should be interpreted cautiously given the short follow-up period. Finally, the assessment of recurrent stroke rates was beyond the scope of this short-term feasibility/process study, and future larger-scale trials with longer follow-up durations are warranted to conclusively determine the effect on clinical recurrence.

Conclusions

Clinical pharmacist-led medication management may be associated with significantly improved medication adherence and early lipid control in patients with AIS receiving IVT. Although no significant differences were observed in short-term functional outcomes within 3 months, the long-term benefits of such interventions remain promising. Strengthening the role of clinical pharmacists in stroke centers and incorporating pharmaceutical care into routine post-stroke management may provide patients with more comprehensive secondary prevention strategies and improved long-term outcomes.