Multidimensional Interventions on Supporting Disease Management for Hospitalized Patients with Heart Failure: The Role of Clinical and Community Pharmacists

Background: existing trials on the role of clinical pharmacists in managing chronic disease patients have focused on variety of interventions, including preparing patients for the transition from hospital to home. However, little quantitative evidence is available regarding the effect of multidimensional interventions on supporting disease management for hospitalized patients with heart failure (HF). The present paper reviews the effects of inpatient, discharge and/or after-discharge interventions performed on hospitalized HF patients by multidisciplinary teams, including pharmacists. Methods: articles were identified through search engines in three electronic databases following the PRISMA Protocol. Randomized controlled trials (RCTs) or non-randomized intervention studies conducted in the period 1992–2022 were included. In all studies, baseline characteristics of patients as well as study end-points were described in relation to a control group i.e., usual care and a group of subjects that received care from a clinical and/or community pharmacist, as well as other health professionals (Intervention). Study outcomes included all-cause hospital 30-day re-admission or emergency room (ER) visits, all-cause hospitalization within >30 days after discharge, specific-cause hospitalization rates, medication adherence and mortality. The secondary outcomes included adverse events and quality of life. Quality assessment was carried out using RoB 2 Risk of Bias Tool. Publication bias across studies was determined using the funnel plot and Egger’s regression test. Results: a total of 34 protocols were included in the review, while the data from 33 trials were included in further quantitative analyses. The heterogeneity between studies was high. Pharmacist-led interventions, usually performed within interprofessional care teams, reduced the rates of 30-day all-cause hospital re-admission (odds ratio, OR = 0.78; 95% CI 0.62–0.98; p = 0.03) and all-cause hospitalization >30 days after discharge (OR = 0.73; 95% CI 0.63–0.86; p = 0.0001). Subjects hospitalized primarily due to heart failure demonstrated reduced risk of hospital admission within longer periods, i.e., from 60 to 365 days after discharge (OR = 0.64; 95% CI 0.51–0.81; p = 0.0002). The rate of all-cause hospitalization was reduced by multidimensional interventions taken by pharmacists: reviews of medicine lists and/or their reconciliation at discharge (OR = 0.63; 95% CI 0.43–0.91; p = 0.014), as well as interventions that were based mainly on patient education and counseling (OR = 0.65; 95% CI 0.49–0.88; p = 0.0047). In conclusion, given that HF patients often have complex treatment regimens and multiple comorbid conditions, our findings highlight the need for greater involvement from skilled clinical and community pharmacists in disease management.


Introduction
The prognosis for patients with heart failure (HF) has improved considerably in the last few decades; however, it remains unsatisfactory. Five-year mortality rates are high, ranging from 20 to 67%; quality of life (QoL) is also markedly reduced [1]. The 2 of 23 prevalence of HF appears to be up to 2% of adults, representing about 5 per 1000 personyears [1]. After initial diagnosis, HF patients are hospitalized once every year on average. Due to population growth and ageing, the absolute number of hospital admissions for HF is expected to increase considerably in the future (potentially by as much as 50% in the next 25 years) [1]. This phenomenon is also supported by the fact that HF patients, especially those with HF with preserved ejection fraction (HFpEF), exhibit a large number of comorbidities, including hypertension (HTN), renal dysfunction, diabetes mellitus and/or obesity affecting the entire cardiovascular system [2]. Acute coronary syndromes (ACS), rapid arrhythmias and severe bradycardia/conduction disturbance can also be specific causes of heart failure, hospitalization and urgent emergency visits [2,3]. About 20% of patients with HF are affected by chronic obstructive pulmonary disease (COPD), which has a major impact on symptoms and outcomes [4]. Medication and dietary nonadherence have also been recognized as contributing factors in up to 30% of hospitalized HF patients [5].
Clinical and community pharmacists are recognized as important actors in the heath care system. They contribute to the management of chronic diseases, including hypertension, diabetes mellitus, hyperlipidemia or anticoagulation therapy [6]. Many studies have focused on the role of pharmacists in the setting of heart failure. Such approaches should be multifactorial due to the incidences of comorbidities in HF, e.g., ischemic heart failure or hypertension, polypharmacy, potential drug-drug interactions and the presence of patientrelated specific determinants of pharmacotherapy, such as age, renal or hepatic function, etc. [7,8]. Clinical pharmacists are capable of performing in-patient medication reviews (MRs) and reconciliations at discharge and, thus, assisting the physician in optimization of pharmacotherapy. Other tasks could include drug therapy monitoring and education about the disease, therapy and lifestyle modifications. In some diseases, patient education and drug counseling is crucial for medication adherence. This can in turn correspond with a patient's clinical condition, thereby determining the hospital readmission [9].
A disease management program led by a HF specialist nurse, as a part of multidisciplinary team, is strongly recommended in the updated clinical guidelines [2]. The question about the role of clinical pharmacists in such multidisciplinary teams remains open. A number of prospective and retrospective studies have been conducted with the purpose of evaluating the benefits of pharmacist-led interventions on in-patient and after discharge management of heart failure. The results of individual clinical trials are diverse and ambiguous. Although recent systematic reviews have evaluated the role of the pharmacist in multidisciplinary approaches for patients hospitalized for heart failure and resulting potential improvements in their outcomes after discharge [10][11][12][13], the results of pooled quantitative analysis aiming to compare the benefits of such activities are scarce.
The study presents a systematic review and quantitative analysis of data from randomized and non-randomized controlled trials by evaluating the effects of various interventions taken by pharmacists for patients suffering from HF. Pharmacists could work in co-operation with other health care providers. Different end-points were also analyzed. The primary outcomes included those associated with rates of hospitalization and death, as well as medication adherence. The secondary outcomes were quality of life, prescription of relevant medications, medication errors and adverse events. Where possible, a cut-off point of 30 days was used when assessing the risk of hospital re-admission. The contribution of such additional factors (variables) was also taken into account; variables included he follow-up period and type of pharmacist-led intervention, as well as key clinical characteristics, such as diagnosis on admission, comorbidities, age, New York Heart Association functional class (NYHA-FC) and left ventricular ejection fraction (LVEF%).

Materials and Methods
The systematic review was conducted in accordance with the Cochrane guidelines, including the statement on Preferred Reporting Items for Systematic Review and Metaanalysis (PRISMA) (Supplementary Table S1).

Selection of Studies
For the purposes of study selection, the following PICO criteria were defined: randomized controlled trials or non-randomized intervention studies performed on adult patients from any country who were hospitalized with diagnosis of HF (P-population). The intervention (I) must have targeted the management of heart failure, e.g., medication review, education and counseling in cooperation with other health professionals; a health service intervention might have also aimed to prepare patients for the transition from hospital to home (reconciliation at discharge, education, monitoring, or supporting the patient in the post-discharge phase in hospital or at home). The comparator (C) was the Usual Care Group (control group), while outcomes (O) (see Section 2.5) were reported for a minimum follow-up of 30 days. Baseline characteristics for patients and study end-points were described in relation to a control group, as well as a group of subjects that received care from a clinical and/or community pharmacist. Pharmacists could work together with other health professionals. Trials that did not fulfil the defined criteria were not included. The study selection was completed independently by two reviewers (M.J.-S., M.W.-N.). Observational retrospective studies, review papers and conference posters were not included.

Extraction of Data and Assessment of Study Quality
The extracted data included population characteristics (age, sex, admission diagnosis, comorbidities, criteria for HF, NYHA classification, values of LVEF%, if possible), descriptions of the intervention, study end-points and items needed for the assessments of study quality. Standardized forms for extracted data were used. The quality of an individual study was estimated in accordance with the revised version of Cochrane Collaboration's RoB 2 Risk of Bias Tool [14,15]. Two reviewers (M.J.-S., M.W.-N.) assessed the risk of bias independently and any differences were solved by discussion to reach consensus.

Analysis of Study Data
As the overall effect size, the odds ratio (OR) with 95% confidence interval (CI) was adjusted for dichotomous data. The calculations concerned the individual study outcomes reported as a proportion of hospitalized or deceased patients according to intention-to-treat analysis. The percentage of adherent patients, as defined by each study, was also considered. Due to expected high heterogeneity between studies, a random-effects model was proposed. Heterogeneity was estimated using the χ 2 test (I 2 ). Another measure used was Cochran's Q statistics. The heterogeneity between two sub-groups of patients was designated by statistically pronounced Q values (p < 0.05). In order to evaluate the influence of each study on the overall effect size, sensitivity analysis was conducted using the leave-one-out method, i.e., removing one study each time and repeating the analysis. Bias across the studies was estimated using Duval and Tweedie's trim and fill method with Egger's test. The analyses were conducted using STATISTICA 13.1 software and R Core Team (RoB 2 Risk of Bias Tool) [15]. A p-value < 0.05 was considered statistically evident.

Study Outcomes
Study end-points (outcomes) included all-cause hospital 30-day re-admission or emergency room (ER) visits, hospitalization within a period longer than 30 days after dis-charge, as defined by each study (e.g., 60-, 90-, 180-or 365-day period), specific-cause hospitalization rates, mortality (primary), time to first unplanned hospitalization, adverse events, medication adherence, quality of life and health literacy (secondary end-points).

Search Results and Characteristics of Individual Studies
The search included 2019 titles. During the process of study selection, 91 papers were indicated to be relevant to the review question and full-text articles were assessed for the purposes of the present review. Finally, the systematic review included 34 papers, as presented in the PRISMA flow chart in Figure 1. One study assessed the potential advantages of pharmacist-driven intervention, albeit only in relation to underutilization of cardiovascular medications [16]. Ultimately, 33 studies were included as part of a further meta-analysis. The number of participants was 12,048, while the mean ages was 66.7 years (62.8-70.6) (Intervention) and 64.1 years (61.2-67.0) (Usual care). The proportion of women was 48%.
Study end-points (outcomes) included all-cause hospital 30-day re-admission or emergency room (ER) visits, hospitalization within a period longer than 30 days after discharge, as defined by each study (e.g., 60-, 90-, 180-or 365-day period), specific-cause hospitalization rates, mortality (primary), time to first unplanned hospitalization, adverse events, medication adherence, quality of life and health literacy (secondary end-points).

Search Results and Characteristics of Individual Studies
The search included 2019 titles. During the process of study selection, 91 papers were indicated to be relevant to the review question and full-text articles were assessed for the purposes of the present review. Finally, the systematic review included 34 papers, as presented in the PRISMA flow chart in Figure 1. One study assessed the potential advantages of pharmacist-driven intervention, albeit only in relation to underutilization of cardiovascular medications [16]. Ultimately, 33 studies were included as part of a further meta-analysis. The number of participants was 12,048, while the mean ages was 66.7 years (62.8-70.6) (Intervention) and 64.1 years (61.2-67.0) (Usual care). The proportion of women was 48%.

Randomization-Bias
50% of trials demonstrated unclear risk of bias (some concerns). These protocols did not report the methods used to generate random sequence. Moreover, the process of allocation of participants was not described. The randomized allocation of subjects to the study groups was not performed in 29.4% protocols (high risk of bias).

Deviations from the Intended Interventions-Bias
In some protocols (open trials), the subjects knew about the intervention (29.4%). In some trials, no information was provided about blinding (44.2%). In 44.2% of studies, it was not clear whether data were analyzed according to the intention-to-treat principle. In total, 5.9% of trials did not provide a consort flow diagram; thereby, it was not possible to follow the participants in a study. Overall, the risk of bias in 64.7% of studies was judged as unclear, while 32.4% of studies were highly biased.

Missing Data (Outcomes)-Bias
26.5% of protocols demonstrated unclear risk of bias (some concerns). Some trials presented high proportion of missing data (ranging from 10 to 30%), while the proportion of missing data varied between study groups (32.3%).

Measurement of Outcomes-Bias
In most cases, the information about blinding was not provided and/or it was not known whether the research evaluations were determined by the participants' knowledge of the study groups to which they belonged (61.8%). In 38.2% of the included studies, it was stated that the outcome assessors knew about the intervention or that the trials were open. This phenomenon could have determined a final outcome e.g., where patients were asked to report their quality of life or adherence to treatment. In 71% of papers, the risk of bias was assessed as unclear (some concerns).

Selection of the Result-Bias
In 23.5% of papers, the study result being evaluated was likely to have been adjusted from multiple data analyses. This was true for the measurement of parameters such as quality of life or adherence (different scales, dichotomous or continuous data). A total of 76.5% of protocols demonstrated low risk of bias.
In the case of n = 2 (5.9%) studies, the risk of bias was assessed as low, while n = 17 (50%) protocols were judged as unclear (some concerns); n = 15 (44.2%) studies were highly biased. Figure 2 summarizes the results of bias assessments. Further analyses were performed to assess publication bias. The analyses concerned outcomes such as all-cause mortality and all-cause hospitalization rates. The sub-group analyses were performed by baseline diagnosis in the intervention arm, type of pharmacists intervention and follow-up. In most cases, the absence of publication bias was designated by a symmetrical plot and a non-significant (NS) result of Egger's test, as demonstrated in Table 1. The results of sensitivity analysis are presented in Supplementary Figure S1. Statistical significance was not influenced by any single study included in the meta-analysis according to the all-cause hospitalization rate. In relation to the all-cause mortality parameter, the analyses were repeated after removing one study [17].
Further analyses were performed to assess publication bias. The analyses concerned outcomes such as all-cause mortality and all-cause hospitalization rates. The sub-group analyses were performed by baseline diagnosis in the intervention arm, type of pharmacists intervention and follow-up. In most cases, the absence of publication bias was designated by a symmetrical plot and a non-significant (NS) result of Egger's test, as demonstrated in Table 1. HF-heart failure; NS-non significant. * trim and fill method included the following procedures: (1) trim (removing) the smaller studies causing funnel plot asymmetry; (2) using the trimmed funnel plot to estimate the true 'center' of the funnel; and (3) replacing the omitted studies and their missing 'counterparts' around the center (filling). The first column from the right provides an estimate of the number of missing (omitted) studies. Non-significant Egger's test results indicate the absence of publication bias across the studies for the particular set of variables, e.g., tested parameter; all-cause mortality; admission diagnosis in the intervention arm for all patients; pharmacist intervention: all types; and follow-up more than 30 days after discharge.

Study Characteristics
The review included 34 prospective intervention studies (including 26 randomized controlled trials) ( Table 2). In 18 papers, all subjects were hospitalized for heart failure, as defined by each protocol. The inclusion criteria were mainly subjects with HFrEF where LVEF was below 40 or 45%; in some papers, the baseline diagnosis was defined as congestive heart failure. Patients with HF and preserved ejection fraction (HFpEF) were also included. Only 30% of papers included in the review provided data regarding the patients' classification according to NYHA-FC. In the remaining trials (n = 16 studies), heart failure was listed as one of the comorbidities. This group also experienced related conditions, such as arterial hypertension (HTN), cerebrovascular diseases (CVD) with stroke, arrhythmias, acute coronary syndrome (ACS), coronary artery disease (CAD), diabetes mellitus (DM), dyslipidemia or chronic obstructive pulmonary disease (COPD). Table 2 summarizes the pharmacist-led interventions focused on HF patients; subjects assigned to the usual care group received the services typically provided by an individual center, as defined for each protocol. In 28/34 studies, clinical pharmacists took care of HF patients in co-operation with other health care providers (e.g., general practitioners, medical specialists, nurses, and/or community pharmacists). In 19 studies (55.8%), clinical pharmacists' activities focused on in-patient medicine management and consultation to identify drug-related problems (DRPs), reviews of medicine lists, and reconciliation at discharge. Pharmacists assessed patient knowledge about the disease, treatment, possible side-effects and lifestyle changes that could help control symptoms and added information where necessary; they also provided patient motivation after discharge. In 14/34 studies, pharmacists mainly prepared patients for the transition from hospital to home through counselling and education. In six/34 studies, the patient-oriented activities could also have been performed by community pharmacists. They received notes about the hospitalization and any medication-related issues identified or needed upon discharge; post-discharge medication reviews were performed in order to optimize medical treatment. They could also conduct home-based visits. In one study, pharmacist transitions coordinators worked with in-patient and community-based pharmacists. Table 3 summarizes the educational interventions that were performed by pharmacists in the reviewed protocols.
In n = 22 studies (64.7%), the beneficial effects of pharmacists led-interventions for hospitalized patients were assessed for at least a 30-day follow-up period; a more pronounced parameter was the rate of re-hospitalizations. Some authors (n = 14 studies; 41.2%) did not consider 30-day follow-ups in their observations and reported outcomes for longer periods, i.e., 60, 90, 180 or 365 days after discharge. In general, the median follow-up period was 90 days (IQR 30-180 days) after discharge.

Synthesis and Analysis of Study Data
The most frequently reported data in individual trials was included in the metaanalysis' these data included all-cause hospital 30-day re-admission rates, all-cause >30-day hospitalization rates, all-cause mortality and medication adherence. The effects of pharmacistled intervention on the time taken up by the first unplanned hospitalization, health literacy, quality life, adverse events or clinical outcomes for individual studies are given in Table 2.
The rate for hospitalizations of patients admitted with diagnoses such as heart failure, ACS, HTN, dyslipidemias or diabetes was insignificantly decreased in the Intervention Group as compared to the Usual Care Group (OR = 0.76; 95% CI 0.57-1.02; p = 0.06) for 30-day and (OR = 0.85; 95% CI 0.70-1.03; p = 0.1) >30-day follow-up periods.
Further analyses based on the random-effects model identified an insignificant reduction in 30-day re-admission rates, primarily among patients primarily hospitalized for heart failure. In this sub-group of patients, the risk of >30-day hospitalization was significantly decreased (p = 0.0002) (Figure 3a). Figure 3b demonstrates the influence of the type of intervention led by pharmacists. The activities that focused on in-patient medication reviews and/or reconciliation at discharge and included mainly education and counseling tended to decrease rates of hospitalization (p < 0.05).  (30) HF (100%) -Identification of potential prescription errors in the discharge medication in order to discuss them with the cardiologist. Verbal and written information about (side)effects of, and changes in hospital drug therapy upon hospital discharge were provided. A discharge medication list (including dose adjustments, discontinued medication) was approved by the physician, faxed to the community pharmacy and given as written information to the patient with the instruction to hand it over to their GP.
Number of medication discrepancies after discharge (p = 0.01); medication adherence (NS)   Medication reconciliation was provided in order to optimize a patient's medication regimen within the multidisciplinary team; consultations on the appropriateness of prescribed medications also occurred.
All-cause hospital re-admission (p = 0.045); ↑ in beta-blocker prescriptions and ↓ in medications that should be avoided in HF. In-patient and discharge pharmacist counseling and education, as well as a post-discharge follow-up phone call, was provided. Upon discharge, the pharmacist helped ensure the patient had a follow-up appointment with their cardiologist within one week. In-patient and discharge advice relating to the clinically relevant drug-drug interactions was provided.

Mills
Recommendations were sent to GPs (discharge letters).
Incidence of drug-drug, all-cause hospitalization (NS); mortality (NS) Verbal counseling was provided to the study patients; advice on medication reviews, education about disease, medication adherence, lifestyle, non-drug treatment and self-monitoring was also provided. Follow-up consisted of telephone contact by the local research coordinator monthly for 6 months after discharge and aimed to reinforce education and adherence relating to the disease.
All-cause (NS) and cardiac-related hospitalization (p = 0.02); total length of hospital stay (days) (p < 0.005); medication adherence (NS) Initial medication reconciliation was provided; in-patient pharmacists completed a summary of the encounter with the use of a standardized template, which included the discharge medication list, allergy information, vaccination history, principal hospital diagnosis, active problem list, recent laboratory results, post-discharge appointments, primary care and nurse management, etc. Pharmacists provided notes directly to the community pharmacist about the hospitalization and any medication-related issues identified or requiring intervention upon discharge.
(a) (b) Figure 3. Forest plot comparing intervention vs. usual care for dichotomous data concerning hospitalization rate (odds ratio, OR) in studies where 100% of patients were hospitalized from heart failure. (a)-all cause hospitalization, including 30-day re-admission, according to the number of days before the next admission to hospital. Pharmacist-led interventions caused a significant reduction in the risk of hospitalization within >30 days after discharge. (b)-all cause hospitalization according to type of pharmacist-led intervention. The first sub-group (in-patient/at discharge MR) received services that were focused on in-patient medication reviews and/or reconciliation at discharge, while the second group was prepared for the transition from hospital to home, mainly by counselling and education. Both types of intervention significantly reduced the risk of hospital admission. Days of follow-up are shown in brackets [3,18,19,22,26,33,34,37,40,[43][44][45]47].
In 11 studies, the efficacy of interventions taken by multidisciplinary teams was assessed in relation to all-cause hospitalization rates (30-day, or >30 days after discharge). The additional analysis aimed to compare the effects of care with clinical pharmacists as a part of Figure 3. Forest plot comparing intervention vs. usual care for dichotomous data concerning hospitalization rate (odds ratio, OR) in studies where 100% of patients were hospitalized from heart failure. (a)-all cause hospitalization, including 30-day re-admission, according to the number of days before the next admission to hospital. Pharmacist-led interventions caused a significant reduction in the risk of hospitalization within >30 days after discharge. (b)-all cause hospitalization according to type of pharmacist-led intervention. The first sub-group (in-patient/at discharge MR) received services that were focused on in-patient medication reviews and/or reconciliation at discharge, while the second group was prepared for the transition from hospital to home, mainly by counselling and education. Both types of intervention significantly reduced the risk of hospital admission. Days of follow-up are shown in brackets [3,18,19,22,26,33,34,37,40,[43][44][45]47]. Table 3. Summary of educational interventions according to reviewed papers.

Stage Intervention/Tool Details References
Prior to discharge and at discharge Patient education (1) the disease and symptoms (e.g., early recognition of worsening symptoms, knowing when to call the physician, how to respond to escalating signs); (2) self-monitoring (e.g., salt and fluid restriction, daily weighing); (3) exercise alternating with rest periods; (4) rationale for medical therapy, proven benefits, indications and contraindications and adverse reactions; (5) secondary lifestyle modifications (e.g., blood pressure and glycaemia control, alcohol and smoking cessation, etc.) [16,18,20,30,34,37,40,42,45,46] Medication schedule (1) education on proper medication use; an illustrated schedule showing the discharge regimen (e.g., proper use of diuretics in relation to peripheral edema); (2) information on action to take if doses of medication are missed; (3) identification and avoidance of medication errors [4,19,20] Adherence aids (1) pillbox with the patient practiced filling; (2) personalized illustrated medication schedule [27,37,41] Materials (1) booklets and leaflets (e.g., written, audiovisual or available for downloading from the websites); (2) interactive photo CD program [18,20,31,33,34,37,40,45,46] Event diary (1) symptom monitoring diary cards (including monitoring body weight) [34,40] After discharge Systematic phone calls (1) reinforcement of education; (2) management with barriers to medication adherence [30,45,46] In 11 studies, the efficacy of interventions taken by multidisciplinary teams was assessed in relation to all-cause hospitalization rates (30-day, or >30 days after discharge). The additional analysis aimed to compare the effects of care with clinical pharmacists as a part of such a multidisciplinary team. The risk of hospitalization was more significantly reduced (OR = 0.80; 0.65-0.99; p = 0.037; T 2 = 0.025; I 2 = 49.11%) in the Intervention Group compared to the Usual Care Group (n = 11 studies). The included patients were hospitalized due to heart failure and other conditions. For a more homogenous sub-group of patients, i.e., 100% of subjects hospitalized for HF, the result was non-significant (n = 5 studies) (OR = 0.86; 0.65-1.14; p = 0.28; T 2 = 0.06; I 2 = 50.26%). Due to study heterogeneity (followup, primarily diagnosis) this finding needs to be verified with a higher number of protocols.
Due to the limited number of studies accompanied with pronounced heterogeneity, it was not possible to perform further sub-group analyses addressing specific-cause hospitalization rates e.g., associated with cardiac-related or DRP issues. Figure 4 demonstrates the rate of death in hospitalized patients with heart failure receiving pharmacist-led interventions compared to usual care. These activities only demonstrated slight beneficial impacts on patients and mortality. Non-RCT studies demonstrated lower risk of mortality in patients diagnosed from HF compared to RCT protocols (OR = 0.42; 95% CI 0.19-0.91; p = 0.03 vs. OR = 0.91; 95% CI 0.74-1.10; p = 0.32) (heterogeneity Chi2, p = 0.06).
The risk of death among patients admitted with heart failure, ACS, HTN, dyslipidemias or diabetes was non-significantly decreased in the Intervention Group compared to the Usual Care Group (OR = 0.92; 95% CI 0.56-1.51 p = 0.74). Due to the limited number of studies, the comparative analysis concerning follow-up period, i.e., 30-day vs. >30-day mortality, could not be performed.
pitalization rates e.g., associated with cardiac-related or DRP issues. Figure 4 demonstrates the rate of death in hospitalized patients with heart failure receiving pharmacist-led interventions compared to usual care. These activities only demonstrated slight beneficial impacts on patients and mortality. Non-RCT studies demonstrated lower risk of mortality in patients diagnosed from HF compared to RCT protocols (OR = 0.42; 95% CI 0.19-0.91; p = 0.03 vs. OR = 0.91; 95% CI 0.74-1.10; p = 0.32) (heterogeneity Chi2, p = 0.06).
The risk of death among patients admitted with heart failure, ACS, HTN, dyslipidemias or diabetes was non-significantly decreased in the Intervention Group compared to the Usual Care Group (OR = 0.92; 95% CI 0.56-1.51 p = 0.74). Due to the limited number of studies, the comparative analysis concerning follow-up period, i.e., 30-day vs. >30-day mortality, could not be performed. Figure 5 demonstrates the insignificant influence of another covariate (age) for allcause hospitalization rates.  A meta-analysis for medication adherence was performed for dichotomous measures (n = 5 studies). The findings indicate the percentage of patients taking more than 80% of the prescribed doses (>80% compliance) [19,39] or prescription filling [46]. In one study, a score of BMQ-Regimen Screen ≥1 indicated potential non-adherence [20]. In five studies, all patients were hospitalized for HF. In general, medication adherence was not improved in studies where interventions were performed within multidisciplinary teams (OR = 2.62; 95%CI 0.68; 10.01; p > 0.05) rather than usual care.
A meta-analysis for medication adherence was performed for dichotomous measures (n = 5 studies). The findings indicate the percentage of patients taking more than 80% of the prescribed doses (>80% compliance) [19,39] or prescription filling [46]. In one study, a score of BMQ-Regimen Screen ≥1 indicated potential non-adherence [20]. In five studies, all patients were hospitalized for HF. In general, medication adherence was not improved in studies where interventions were performed within multidisciplinary teams (OR = 2.62; 95% CI 0.68; 10.01; p > 0.05) rather than usual care.
The authors hope to analyze the impact of pharmacist-led interventions in potentially reducing adverse drug events (adverse drug reactions), as well as improving the health literacy or quality of life. Unfortunately, it was not possible to perform the pre-defined calculations due to the limited number of studies and pronounced heterogeneity (follow-up, admission diagnosis). Figure 6 summarizes the role of pharmacists in an interprofessional care team, according to trial protocols reviewed in the present paper.

Discussion
The present study reviews prospective studies on the effectiveness of pharmacist-led interventions for heart failure; it is recognized as baseline diagnosis at admission against a range of outcomes and performs a pooled analysis. The most frequent outcome assessed in the studies was the risk of all-cause readmission. This risk was based on a cut-off point of 30 days, where possible; this cut-off point was selected due to the hospital readmissions reduction programs implemented in some countries, including the USA, which penalize hospitals with excess in-patient rehospitalizations within 30 days of index in-patient stays for targeted conditions [49].

Hospitalization Rate and Mortality
The pooled analysis of data from 26 studies, which compared patients receiving a number of pharmacist-led interventions to those receiving usual care, revealed reductions in the rates of all-cause 30-day hospital re-admission and >30-day hospitalization. The sub-group of patients primarily hospitalized for heart failure alone received some benefits, which manifested as a reduced risk of hospitalization more than 30 days after discharge. However, this was not the case among the more heterogeneous group of patients, i.e., those hospitalized for HF and other conditions. These patients displayed a non-significant reduction in hospitalization rate for the 30-day and longer follow-up period (i.e., from 60 to 365 days). A very recent systematic review qualitatively assessed the rates for hospital readmissions, especially at the 30-day follow-up stage [12]. Patients were admitted with primary diagnosis of acute myocardial infarction, pneumonia, heart failure, COPD, diabetes, dysrhythmias and other internal diseases. The review, which includes 37 RCTs and non-RCTs, found that pharmacist-led interventions that include communication with a primary care physician were effective in reducing readmissions. However, the per-

Discussion
The present study reviews prospective studies on the effectiveness of pharmacist-led interventions for heart failure; it is recognized as baseline diagnosis at admission against a range of outcomes and performs a pooled analysis. The most frequent outcome assessed in the studies was the risk of all-cause readmission. This risk was based on a cut-off point of 30 days, where possible; this cut-off point was selected due to the hospital readmissions reduction programs implemented in some countries, including the USA, which penalize hospitals with excess in-patient rehospitalizations within 30 days of index in-patient stays for targeted conditions [49].

Hospitalization Rate and Mortality
The pooled analysis of data from 26 studies, which compared patients receiving a number of pharmacist-led interventions to those receiving usual care, revealed reductions in the rates of all-cause 30-day hospital re-admission and >30-day hospitalization. The sub-group of patients primarily hospitalized for heart failure alone received some benefits, which manifested as a reduced risk of hospitalization more than 30 days after discharge. However, this was not the case among the more heterogeneous group of patients, i.e., those hospitalized for HF and other conditions. These patients displayed a non-significant reduction in hospitalization rate for the 30-day and longer follow-up period (i.e., from 60 to 365 days). A very recent systematic review qualitatively assessed the rates for hospital readmissions, especially at the 30-day follow-up stage [12]. Patients were admitted with primary diagnosis of acute myocardial infarction, pneumonia, heart failure, COPD, diabetes, dysrhythmias and other internal diseases. The review, which includes 37 RCTs and non-RCTs, found that pharmacist-led interventions that include communication with a primary care physician were effective in reducing readmissions. However, the performed analyses did not concern the HF-patient sub-group.
A meta-analysis by Van Spall et al. (2017) compared the effectiveness of transitional care services in decreasing all-cause death and all-cause readmission rates following hospitalization for heart failure based on 53 RCT studies [13]. In contrast to the current study, some interventions were performed by various health care professionals. The authors classified services into the following categories: education alone, telemonitoring, telephone support (without remote telemonitoring), nurse home visits (clinical assessment and education) and a combination of nurse home visits with structured telephone support and follow-up visits at a hospital. A final category-pharmacist interventions-concerned visits by pharmacists for education, medication reconciliation and optimization after discharge according to data reported in four trials. Among the services that significantly decreased all-cause re-hospitalization and mortality compared with usual care, nurse home visits were most effective. Telephone, telemonitoring, pharmacist, and education interventions did not significantly improve clinical outcomes.
Despite the diverse and multi-component structure of interventions performed by pharmacists in the trials included in the current analysis, it was possible to distinguish two sub-groups of patients. The first sub-group received services that were focused on in-patient medication reviews and/or reconciliation at discharge, while the second group was mainly prepared for the transition from hospital to home through counselling and education. Some authors indicated that the latter program can be more readily adapted to a variety of "real life" settings suitable for patients with HF [45]. The present study demonstrates the efficacy of such activities; rates for hospitalization were reduced significantly. In general, clinical pharmacists reconciled pre-admission and at-discharge medications with the patient and reported any inconsistencies to the medical team prior to hospital discharge. They also could contact physicians regarding drug therapy if rationalization of therapy or simplification of dosage regimens was appropriate. Educational interventions varied in individual protocols and could be performed prior to and after discharge. They were based on the tailored HF management programs, supported by written or audiovisual materials and adherence aids for patients. The emphasis was placed on simple language, adapted to the social and cultural level of the patient [33]. In a majority of studies, the benefits of patient education were reported for longer follow-ups of typically three-to-six months, which was probably due to a whole range of positive factors that came about as a result of the multi-component interventions. It could be hypothesized that intervention patients tended to seek help more frequently at the casualty department, perhaps indicating a better understanding of the need for early medical intervention in management their heart failure when symptoms were deteriorating. This view was supported by phenomena where the nature of intervention additionally involved home-based pharmacist visits or contacts with community pharmacists who were familiarized with the discharge medication list.
The mortality rates were assessed in n = 17 studies. For most sub-groups of patients and follow-up periods, the interventions performed by pharmacists did not reduce the rates of death from any cause. The obtained results are also similar to those reported in reviews including populations of patients hospitalized for a wide spectrum of chronic disorders [50].

Medication Adherence
Undoubtedly, medication adherence among HF patients is a continual challenge. Few strategies for reducing DRPs, including medication non-adherence, in patients hospitalized for heart failure have been presented in the existing literature. In a very recent systematic review, Hernández-Prats et al. (2022) summarized the available evidence resulting from interventions led by pharmacists aimed at reducing inappropriate medication prescription in patients with heart failure [11]. The authors conclude that evaluating the suitability of treatment to specific HF guidelines can be crucial. Other papers concerned chronically ill subjects without addressing their particular disorders. For example, Kelly et al. (2021) assessed the relationship between pharmacist medication counseling, medication adherence, 30-day hospital readmission rates and mortality [50]. The authors included 62 RCTs; in most studies, the participants were older patients with various chronic diseases and polypharmacy. Pharmacist medication counseling was associated with a statistically significant 30% increase in relative risk for medication adherence and a 24-30% risk reduction in 30-day hospital readmission rates or emergency department visits.
In the present study, due to different scales and follow-ups the pooled analysis of dichotomous data was performed for only five studies, where all patients were hospitalized for heart failure. In general, medication adherence was not improved by interventions being performed within multidisciplinary teams compared to usual care; however, these findings need to be verified by a higher number of studies.

Quality of Life and Health Literacy
Better medication adherence should be accompanied by increased health literacy; such improved knowledge was reported among HF patients in some trials. This could concern items such as knowledge of weight control, self-adjustment of diuretics, signs and symptoms of heart failure and side-effects of drugs [31,46]. This disorder and its symptoms can greatly affect the ability of patients to perform normal daily activities. Some studies focused on the potential role of the pharmacist in improving quality of life among patients hospitalized for heart failure. In general, the results indicated that benefits were obtained in both the Intervention and Usual Care sub-groups. Various measures of quality of life were used, including SF-36 scores, Minnesota Living with Heart Failure (MLHF), health visual analogue scale (VAS) and EUROQOL, EQ-5D and Assessment of Quality of Life instrument (AQoL). Only in a study by Varma et al. (1999) did patients receiving pharmacist-led interventions tend to score lower (improved QoL) on the MLHF questionnaire throughout the study; the only significant difference was at 9 months (p = 0.04) [46]. However, as quality of life (QoL) was reported by means of a variety of scales and follow-ups, more detailed comparative analyses could not be performed. In isolated studies, QoL or health literacy assessments were accompanied by an assessment of patient satisfaction score [33,37].

Practical Implications
In trials that were included in the systematic review, interprofessional teams consisted of clinical pharmacists and other health care providers. As discussed previously, clinical pharmacists can successfully contribute to such teams, offering effective and safe care [51]. This can include optimization of medical therapy or dealing with drug interactions or medicine side-effects, as employed by the health care systems in the USA, the UK, Canada or New Zealand [52]. Indeed, most of the clinical studies were conducted in the USA (20/34) and the UK (4/34). Moreover, in other European countries, such as Norway and Sweden, 5/34 trials showed that the need for clinical pharmacy services and agreements to perform these services in hospitals has increased substantially in recent years. Nevertheless, the structure of the multicomponent service and the degree of cooperation between clinical pharmacists and other health care professionals may represent important limitations in the present paper. Another point is the extent of transmission of patient information from the in-patient setting to the community pharmacy. This point might be key to reducing fragmentation in health care and empowering patients with knowledge about their disease state and therapy; this approach provides them with a sense of self-efficacy, which further encourages positive lifestyle changes and medication adherence [53].

Strengths and Limitations
Although the present study was not restricted to RCTs, the comparative analyses did not reveal any significant differences between RCT and non-RCT studies according to rate of hospitalization. Little publication bias was observed across the studies. As mentioned earlier, most of the reviewed studies were conducted in the USA and UK. This phenomenon seems to indicate the substantial role of clinical pharmacy services in the USA and UK. Subgroup analyses were performed where possible and several variables/factors that could impact the final outcome were identified. This included the participation of patients with heart failure in primary admission diagnosis. If possible, the results were (re)calculated according to intention-to-treat analysis. The main outcomes reported in the individual studies-no matter if they were included into meta-analysis or not-are summarized in Table 2.
Despite its strengths, the review has some limitations. The first limitation is the high risk of bias, as shown in the case of 44.2% of trials. The trials also demonstrated high heterogeneity with regard to the intervention type and its multicomponent structure (e.g., in-patient medication review and/or conciliation, at discharge, education and counselling, post-discharge visits at home, contacts with community pharmacists, etc.), follow-up periods and diverse measures for outcomes. For example, due to differences in the scales and measures (dichotomous or continuous data, lower/higher score-related increases in an individual parameter, etc.), it was not possible to perform analyses in relation to QoL. In addition, due to incomplete reporting, it was possible to perform neither comparative analyses regarding the inter alia relationship between medication adherence and hospitalization rate in HF patients, nor assess the impact of clinical characteristics on final outcomes. Indeed, regressive analyses were only performed for mortality and age. For the same reason, the specific-cause hospitalization rates, i.e., cardiac-or DRP-related rates, were not assessed. Another point is that patient-oriented activities were performed by specific HF pharmacists in none of 34 studies. In most trials, clinical pharmacists took care of participants with specialists in geriatrics, cardiology, internal medicine and/or specialist heart failure nurses. Further trials are needed to evaluate the advantages of HF pharmacist-driven interventions in reducing the hospitalizations related to heart failure. As mentioned above, a key "interpretative" limitation of the present study can be the degree of cooperation between clinical pharmacists and other health care professionals, as well as the extent of transmission of patient information from the in-patient setting to the community pharmacy in "real life".

Conclusions
In conclusion, given that patients with heart failure often have complex treatment regimens and multiple comorbid conditions, while HF accounts for 2-5% of hospital admissions, the obtained results point to the need for greater involvement of skilled clinical and community pharmacists in disease management. Only the multicomponent interventions that includes both in-patient medication reviews and reconciliation at discharge and post-discharge follow-up with the contribution of community pharmacies, as well as patient education and counseling at each of these stages, could produce some benefit. These benefits could be associated with decreased hospitalization rates, including 30-day re-admission rates.