Prevalence and Risk Factors of Readmission Among Patients with Cardiovascular Diseases in Africa: A Systematic Review

Abstract

Background

Cardiovascular diseases (CVDs) are a leading cause of morbidity and mortality globally, with a disproportionately high burden in Africa. Readmissions among patients present an additional challenge. This systematic review aimed to map evidence on the prevalence and risk factors of readmissions among patients with CVDs in Africa.

Method

A systematic review was conducted following the PRISMA guidelines. Comprehensive searches were carried out in PubMed Central, Scopus, JSTOR, CINAHL and other databases. Studies published in English from January 2000 that reported on the prevalence and risk factors of readmission among CVD patients in Africa were included. The Joanna Briggs Institute critical appraisal tools were used to assess the quality of included studies. Descriptive and thematic analysis were used to synthesise the findings.

Results

A total of 37 records were included in the final synthesis. Heart failure was the most extensively studied CVD. The prevalence of readmission among patients with CVDs ranged from 1.53% to 67.7%. Older age (> 65 years) was a key predictor. Clinical factors included hypertension, arrhythmias, advanced disease stages including LVEF < 40%, NYHA Class II-IV, NHISS > 7, anemia, infections and chronic comorbidities. Facility-related factors comprised prolonged hospitalisation and discharge without essential medications (beta-blockers for heart failure, and aspirin, clopidogrel, statins for myocardial infarction). Poor adherence to treatment regimen emerged as self-management determinant.

Conclusion

The findings suggest that readmissions among patients with CVDs in Africa remain a substantial burden. The review highlights the need for evidence-based interventions focusing on standardised discharge protocols, enhanced patient education, better screening of patients and management of comorbidities. Future research should address geographical gaps and focus on a wider range of CVDs to provide a more comprehensive overview.

Supplementary Information

The online version contains supplementary material available at 10.1007/s44197-026-00542-x.

Introduction

Cardiovascular diseases (CVDs) remain the leading cause of morbidity and mortality globally with significant health and economic implications. It has been estimated that more than a billion people worldwide are affected by CVDs [1]. According to the World Health Organisation (WHO), CVDs contribute the largest to global deaths, accounting for about 17.9 million deaths in 2019 with more than three-quarters occurring in low- and middle-income countries (LMICs) [2]. In addition to the above, CVDs account for a significant proportion of economic burden for people in LMICs [3].

In Africa, the burden of CVDs is disproportionately high compared with global estimates [1, 4, 5]. A report by the WHO indicated that the prevalence of CVDs in Africa increased by 50% over three decades [6]. Furthermore, it has been reported that around 80% of CVD deaths occur in LMIC, with Africa accounting for the highest proportion, and death rate disproportionately high in Sub-Saharan Africa (SSA) [1]. In line with this, CVDs have been reported as the most frequent causes of non-communicable disease (NCD) deaths in SSA, responsible for approximately 13% of all deaths and 37% of all NCD-related deaths [5]. Additionally, in SSA, CVDs account for about $9 billion in national and regional economic loss, representing about 7% of gross domestic product [3]. These indicate a disproportionately high and concerning burden of CVDs in the African region.

Hospital readmissions among patients with CVD represent a major challenge for both patients and health systems [7]. Readmissions are associated with increased mortality, prolonged morbidity, reduced quality of life, and substantial healthcare costs [7]. In many settings, a significant proportion of readmissions following cardiovascular hospitalisation are considered potentially preventable and may reflect gaps in discharge planning, medication adherence, continuity of care, and management of comorbidities [8–12]. This means understanding the patterns and determinants of readmissions is therefore essential for improving patient outcomes and strengthening health system performance [11–13].

In addition, readmissions among patients are additional challenges to the disease burden of CVDs. For instance, in the United States of America (USA), CVDs are among the leading causes of readmissions [7]. Similarly, studies in Africa have reported that CVDs lead the reasons for hospitalization [8–11]. Beyond index admissions, the burden of readmissions among patients with CVD remains substantial. In a cohort of 11,000 patients with heart failure (HF) in USA, 56% to 70% of hospitalisations were attributed to HF-related readmissions [12]. Comparable evidence from United Arab Emirates reported a 42.1% readmission rate among patients with HF [13]. National data from the United States indicate a 30-day readmission rate of 11.9% among stroke survivors [14], while a cardiology cohort in South Australia, including patients admitted for arrhythmias and elective cardiac procedures, demonstrated a 20% 30-day all-cause unplanned readmission rate [15]. These findings highlight the high prevalence of readmissions and implications for patients and health systems. Evidence from global studies indicates that readmission is influenced by multiple factors, including chronic comorbidities, advanced CVD severity, poor adherence to treatment, suboptimal discharge planning, financial barriers and inadequate family or social support [7, 13, 15–17]. Further compounding the issue, it has been indicated that majority of the readmissions among patients with CVDs are preventable [13, 15]. Additionally, the burden of readmissions in CVDs has been linked to poor quality of life among patients, high costs of care and poor outcomes including increased mortality [7, 16, 18]. It is therefore prudent to explore the nature of readmissions among patients with CVD in Africa to understand the current state of the problem and intervene strategically.

A number of reviews have explored the prevalence of readmission among patients with different types of CVDs, revealing both the scale of the problem and its complexity. For instance, Labrosciano et al. [15] assessed the readmission among Australian patients following a hospitalisation for CVDs, highlighting substantial post-discharge morbidity and identifying clinical and system-level predictors such as comorbidities, length of stay, and prior emergency department attendance. Additionally, Foroutan et al. [17] synthesised global evidence on HF, reporting marked international variation in 30-day readmission rates and highlighting the influence of health-system factors such as discharge clarity and access to affordable medications. Another review also mapped clinical prediction models for unplanned readmissions in heart diseases globally indicating limited utilisation of readmission risk stratification tools [16]. Furthermore, Mbalinda et al. [19] explored stroke recurrence among patients in SSA and identified hypertension, chronic alcohol use, diabetes mellitus and poor treatment adherence as key contributors. These reviews bring to light the critical problem of readmissions among patients with CVD, providing insights into occurrence and risk factors. However, many of these studies focus on non-African populations [15–17, 20].

Despite growing research on CVD outcomes in Africa, the evidence on hospital readmissions remains fragmented across different settings, cardiovascular conditions, and follow-up periods. Existing studies report varying readmission rates and identify diverse predictors, making it difficult to draw consolidated conclusions about the burden and determinants of readmissions across the continent. In addition, considering the heightening epidemiological shift in Africa towards NCD, alongside challenges such as insufficient CVD-related policies, poor adherence among patients and poorly resourced facilities and healthcare staff, it is imperative to understand the patterns of readmissions among patients with CVDs. This systematic review aimed to synthesise evidence on prevalence and determinants of hospital readmissions among patients with CVDs in Africa. This review contributes to literature by enhancing availability of evidence needed for the management of CVD and prevention of poor outcomes in Africa. The findings of this review will guide policymakers and healthcare professionals in developing evidence-based interventions that support the achievement of Sustainable Development Goal 3, specifically target 3.4, which aims to reduce premature mortality from NCD [21].

Methods

This review followed the guidelines proposed by Aromataris & Pearson [22] for conducting systematic reviews. The suggested steps include clearly defining the objective and question, provision of a detailed inclusion and exclusion criteria, providing a comprehensive search strategy, conducting a quality appraisal of the included studies, analysing the data extracted, presenting and synthesising findings, and providing a transparent reporting of the approach undertaken [22]. Furthermore, The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines proposed by Page et al. [23] guided the reporting of this review. The protocol for this review was registered in Open Science Frame work with DOI: 10.17605/OSF.IO/AWBVK.

Inclusion and Exclusion

The research questions for this review were developed using the Population, Concept and Context (PCC) framework. In this review, CVDs refer to conditions affecting the heart and blood vessels, including but not limited to HF, coronary artery disease, hypertension-related complications and stroke. Readmission was defined as an unplanned hospital admission occurring after discharge following an index hospitalisation for a CVD, irrespective of whether the cause of readmission was cardiovascular or non-cardiovascular. The timeframe for readmission assessment in the included studies ranged from 30 days to two years. The risk factors encompass any cause of readmission that may result directly from the CVD or from related complications. The inclusion and exclusion criteria was utilised in identifying studies that will address the review questions. The eligibility criteria are presented in Table 1 below.

Table 1.

Eligibility Criteria

Inclusion	Exclusion
1. Studies that reported on the prevalence or risk factors associated with readmission among patients with CVD in Africa.	1. Letters to editors, conferences papers, reviews and commentaries
2. Studies conducted among patients with CVDs.	2. Studies among patients admitted as referrals from other facilities
3. Peer-reviewed journal articles and grey literature, including Masters Theses or Dissertations
4. Papers published in English language from 1st January, 2000 to 31st July, 2025.

Search Strategy

The search for relevant literatures was conducted in four main databases which are PubMed Central, Scopus, JSTOR and CINAHL (see Table 2). Additional hand search was conducted in Dimensions AI, Google scholar, ProQuest, Google and institutional repositories. The search was conducted from 1st July, 2025 to 31st July, 2025. A search strategy that involves the combination of key terms and Medical Subject Headings (MeSH) terms was developed in consultation with a librarian. Boolean operators (“AND”, “OR”) and truncation symbols were applied to combine and expand search terms. The search strategy was first used for search in PubMed Central and later modified for search in the other main databases. Additional hand search was conducted in Google scholar, ProQuest, Google and institutional repositories. Furthermore, the reference list of all the included studies was manually searched to identify other eligible records and a librarian was also consulted for articles. All these were done to ensure maximum coverage of a broad spectrum of literature. The keywords that were used for the search included prevalence, rate, percentage, readmission, rehospitalisation, unplanned admission, risk factors, predictors, factors, determinants, patients, inpatients, cardiovascular diseases, heart diseases, heart failure, myocardial infarction, acute coronary syndrome, hypertension, stroke, Africa.

Table 2.

Search Strategy in Databases

Database	Search strategy
PubMed Central	((((((((((“epidemiologies“[All Fields] OR “epidemiology“[MeSH Subheading] OR “epidemiology“[All Fields] OR “epidemiology“[MeSH Terms] OR “epidemiology s“[All Fields] OR (“epidemiology“[MeSH Subheading] OR “epidemiology“[All Fields] OR “prevalence“[All Fields] OR “prevalence“[MeSH Terms] OR “prevalance“[All Fields] OR “prevalences“[All Fields] OR “prevalence s“[All Fields] OR “prevalent“[All Fields] OR “prevalently“[All Fields] OR “prevalents“[All Fields]) OR “rate“[All Fields]) OR (“percentage“[All Fields] OR “percentages“[All Fields]) AND (“readmission“[All Fields] OR “readmissions“[All Fields])) OR (“rehospitalisation“[All Fields] OR “rehospitalisations“[All Fields] OR “rehospitalised“[All Fields]) OR (“patient readmission“[MeSH Terms] OR (“patient“[All Fields] AND “readmission“[All Fields]) OR “patient readmission“[All Fields] OR “rehospitalization“[All Fields] OR “rehospitalizations“[All Fields] OR “rehospitalized“[All Fields]) OR (“unplanned“[All Fields] AND (“admission“[All Fields] OR “admissions“[All Fields]))) AND (“risk factors“[MeSH Terms] OR (“risk“[All Fields] AND “factors“[All Fields]) OR “risk factors“[All Fields])) OR (“predictor“[All Fields] OR “predictors“[All Fields]) OR (“factor“[All Fields] OR “factor s“[All Fields] OR “factors“[All Fields]) OR (“analysis“[MeSH Subheading] OR “analysis“[All Fields] OR “determination“[All Fields] OR “determinant“[All Fields] OR “determinants“[All Fields] OR “determinate“[All Fields] OR “determinated“[All Fields] OR “determinates“[All Fields] OR “determinating“[All Fields] OR “determinations“[All Fields] OR “determine“[All Fields] OR “determined“[All Fields] OR “determines“[All Fields] OR “determining“[All Fields])) AND (“patient s“[All Fields] OR “patients“[MeSH Terms] OR “patients“[All Fields] OR “patient“[All Fields] OR “patients s“[All Fields])) OR (“inpatient s“[All Fields] OR “inpatients“[MeSH Terms] OR “inpatients“[All Fields] OR “inpatient“[All Fields])) AND (“cardiovascular diseases“[MeSH Terms] OR (“cardiovascular“[All Fields] AND “diseases“[All Fields]) OR “cardiovascular diseases“[All Fields])) OR (“heart diseases“[MeSH Terms] OR (“heart“[All Fields] AND “diseases“[All Fields]) OR “heart diseases“[All Fields]) OR (“heart failure“[MeSH Terms] OR (“heart“[All Fields] AND “failure“[All Fields]) OR “heart failure“[All Fields]) OR (“myocardial infarction“[MeSH Terms] OR (“myocardial“[All Fields] AND “infarction“[All Fields]) OR “myocardial infarction“[All Fields]) OR (“acute coronary syndrome“[MeSH Terms] OR (“acute“[All Fields] AND “coronary“[All Fields] AND “syndrome“[All Fields]) OR “acute coronary syndrome“[All Fields]) OR (“hypertense“[All Fields] OR “hypertension“[MeSH Terms] OR “hypertension“[All Fields] OR “hypertension s“[All Fields] OR “hypertensions“[All Fields] OR “hypertensive“[All Fields] OR “hypertensive s“[All Fields] OR “hypertensives“[All Fields]) OR (“stroke“[MeSH Terms] OR “stroke“[All Fields] OR “strokes“[All Fields] OR “stroke s“[All Fields])) AND (“africa“[MeSH Terms] OR “africa“[All Fields] OR “africa s“[All Fields] OR “africas“[All Fields])) OR (“ghana“[MeSH Terms] OR “ghana“[All Fields] OR “ghana s“[All Fields]) OR (“nigeria“[MeSH Terms] OR “nigeria“[All Fields] OR “nigeria s“[All Fields]) OR (“kenya“[MeSH Terms] OR “kenya“[All Fields] OR “kenya s“[All Fields]) OR (“tanzania“[MeSH Terms] OR “tanzania“[All Fields] OR “tanzania s“[All Fields]) OR (“egypt“[MeSH Terms] OR “egypt“[All Fields] OR “egypt s“[All Fields]) OR (“morocco“[MeSH Terms] OR “morocco“[All Fields]) OR (“cameroon“[MeSH Terms] OR “cameroon“[All Fields] OR “cameroons“[All Fields] OR “cameroon s“[All Fields]) OR (“democratic republic of the congo“[MeSH Terms] OR (“democratic“[All Fields] AND “republic“[All Fields] AND “congo“[All Fields]) OR “democratic republic ? ? congo“[All Fields]) OR (“south africa“[MeSH Terms] OR (“south“[All Fields] AND “africa“[All Fields]) OR “south africa“[All Fields]) OR (“zambia“[MeSH Terms] OR “zambia“[All Fields] OR “zambia s“[All Fields]) AND 2000/1/1:2025/7/31[pdat]
Scopus	TITLE-ABS-KEY (epidemiology OR prevalence OR rate OR percentage) AND TITLE-ABS-KEY (readmission OR rehospitalisation OR (“unplanned” AND admission)) AND TITLE-ABS-KEY (“risk factors” OR (“risk” AND “factors”) OR predictors OR factors OR determinants) AND TITLE-ABS-KEY (patients OR inpatients) AND TITLE-ABS-KEY (“cardiovascular diseases” OR (“cardiovascular” AND diseases) OR “heart diseases” OR (“heart” AND failure) OR “heart failure” OR (“myocardial” AND infarction) OR “myocardial infarction” OR (“acute” AND coronary AND syndrome) OR “acute coronary syndrome” OR hypertension OR stroke) AND TITLE-ABS-KEY (africa) AND PUBYEAR > 1999 AND PUBYEAR < 2026
JSTOR	(epidemiology OR prevalence OR rate OR percentage) AND (readmission OR rehospitalisation OR (“unplanned” AND admission)) AND (“risk factors” OR (“risk” AND “factors”) OR predictors OR factors OR determinants) AND (patients OR inpatients) AND (“cardiovascular diseases” OR (“cardiovascular” AND diseases) OR “heart diseases” OR (“heart” AND failure) OR “heart failure” OR (“myocardial” AND infarction) OR “myocardial infarction” OR (“acute” AND coronary AND syndrome) OR “acute coronary syndrome” OR hypertension OR stroke) AND africa
CINAHL	TX (epidemiology OR prevalence OR rate OR percentage) AND TX (readmission OR rehospitalisation OR (“unplanned” AND admission)) AND TX (“risk factors” OR (“risk” AND “factors”) OR predictors OR factors OR determinants) AND TX (patients OR inpatients) AND TX (“cardiovascular diseases” OR (“cardiovascular” AND diseases) OR “heart diseases” OR (“heart” AND failure) OR “heart failure” OR (“myocardial” AND infarction) OR “myocardial infarction” OR (“acute” AND coronary AND syndrome) OR “acute coronary syndrome” OR hypertension OR stroke) AND TX (africa)

Selection of Relevant Studies

A criterion for eligibility was established prior to the screening process to help guide making decisions pertaining to what studies are relevant or not to this review. Screening for eligible records was conducted in three phases. Firstly, all records that were retrieved from all the databases were imported into Mendeley Reference Manager for removal of duplicates. During the second phase, four graduate students from the School of Nursing and Midwifery of University of Cape Coast were trained and supervised to screen the remaining records using the titles and abstract. The purpose of this screening was to select full-text records closely aligned with the review objectives for further assessment. Records that were excluded included studies that did not report on readmissions or risk factors of readmissions, were conducted outside Africa, or reported readmissions on other NCDs aside CVDs. The reference list of the full-text records were further checked to identify additional relevant records. Finally, the identified full-text records were independently screened by two groups of researchers (FAN, AS, HKK and EO) against the eligibility criteria to ascertain their inclusion. Through discussions facilitated by two independent reviewers (MA and NIEE), all disagreements that arose at the study selection stage were resolved. The search results and screening process were summarised using the PRISMA flow diagram.

Data Extraction

An extraction form was developed using Microsoft Word. The extraction form was piloted using five randomly selected papers for reliability and credibility of the form and the extracted data. The form contained key information including name of the authors and year of publication, country, purpose of the study, study design, study population, sample size, period of readmission assessment, readmission rate and risk factors of readmission. Two groups of authors (E.O., M.D.) and (F.A.N., A.G.) extracted the data independently. Discrepancies that arose from the data extraction were resolved by N.I.E.E and M.A during meetings of all authors.

Critical Quality Assessment

The Joanna Briggs Institute (JBI) [24] critical appraisal tools for cohort studies and cross-sectional designs were used in the appraisal of eligible records. The scoring procedures adopted for JBI cohort study were 1–4 (high risk), 5–7 (moderate risk) and 8–11 (low risk). The scoring criteria adopted for JBI analytical cross-sectional study were 1–3 (high risk), 4–5 (moderate risk) and 6–8 (low risk) [25]. The use of these tools was very important in ensuring the reliability, quality and credibility of the studies were selected for the systematic review. Additionally, they provided insights into a holistic understanding of the quality and risk of bias of the evidence gathered. It is worth noting that the aim of this appraisal was not to exclude studies, but to provide an evaluation of the quality of the papers included in the review. Through this process, we sought to openly relay the robustness of the existing literature to readers enabling them to estimate the reliability and trustworthiness of the synthesised evidence. In line with this, the use of these appraisal tools ensured that the review process was methodologically sound, enhancing the credibility and validity of the findings presented. The results of the quality appraisal were used to inform the interpretation of the findings and to highlight methodological limitations within the included studies.

Data Analysis and Synthesis

The extraction form was reviewed repeatedly to ensure familiarity with the data extracted and also to ensure thorough understanding of the extracted data. We aggregated and synthesised information on general study characteristics including geographical setting, methodological characteristics (study design, population characteristics, sample size), types of CVDs, timepoints of readmissions, readmission rates and the risk factors associated with readmission. This enabled a comprehensive description of the evidence base and facilitated comparison across studies conducted in diverse African contexts. Microsoft Excel was used to analyse the characteristics of the included studies due to its suitability for managing structured data. The countries where the studies were conducted were represented on a map whiles the extracted data relevant to answering the research questions were analysed using thematic analysis. The identified risk factors of readmission were further synthesised into sub-themes to reflect patient-related factors, healthcare system-related factors, and disease-specific factors.

The synthesis focused on identifying patterns, consistencies, and variations across the included studies. Prevalence estimates were compared across different timeframes and study settings, while reported predictors of readmission were grouped and analysed thematically based on their conceptual similarities. Interpretations were grounded directly in the data extracted from the included studies to ensure that conclusions were derived from the available evidence. Due to substantial methodological and clinical heterogeneity across studies such as differences in follow-up periods, CVD categories, study designs, and outcome measurement, a meta-analysis was not considered appropriate. Therefore, findings were synthesised using a structured narrative approach.

Consultation

A librarian was consulted to support formulating the search terms in PubMed Central, modification of search terms to suit other databases and the removal of duplicates.

Results

Search Results

This systematic review followed the PRISMA guideline for literature search and study selection. Relevant records were retrieved through a comprehensive search in PubMed Central, Scopus, JSTOR and CINAHL. Additional hand search was conducted in Dimensions AI, Google scholar, ProQuest, Google and institutional repositories. A total of 8,553 records were identified from database searches. Out of these, 4,476 were retrieved from PubMed Central, 1,148 from Scopus, 2,033 from JSTOR, 896 from CINAHL and 68 from additional hand search in Google Scholar (32), Dimensions AI (20) and ProQuest (16). Mendeley Reference Manager was used to remove 73 duplicates leaving 8,548 records for further screening. After thorough screening of abstracts and titles, a total of 8,506 records were excluded leaving 42 full-text eligible records. Six additional records were retrieved from the reference lists of the full-text eligible records leaving 48 records for final screening. During the full-text eligibility assessment, 11 out of the 48 full text eligible records were excluded from the study. The 11 records were excluded for reasons including lack of report on the variables of interest (n = 9), and publication not being in English (n = 2). A total of 37 records met the inclusion criteria and were included in this systematic review. Figure 1 presents a PRISMA flow diagram of our search strategy.

Fig. 1.

PRISMA flow Diagram of Search Strategy

Characteristics of Included Studies

Included records [26–62] have been presented in Supplementary File 1. The country with the highest number of studies was Nigeria (n = 12). This was followed by Ethiopia with eight studies and Tanzania with seven studies. Cameroon contributed six studies, while South Africa and Kenya were each represented by five studies. Uganda and Senegal each had four studies. Mozambique and Sudan had three studies each. Egypt, Côte d’Ivoire, and Guinea were represented by two studies each. Ghana, Morocco, Mali, Togo, and Zimbabwe each contributed one study. It is also worth noting that five of the included studies were multinational, which utilised patients from facilities located in different African countries [28, 37, 45, 46, 62]. This has been presented in Fig. 2 below.

Fig. 2.

Map Showing Where the Included Studies were Conducted

Two study designs were used by included records. Cohort study design was the most frequently used methodology, with 28 studies representing 75% of included records. The cross-sectional study design appeared in nine studies. The sample size ranged from 59 to 1,291 and majority of the studies reported on patients 18 years or older. The included studies reported on six main types of CVD. HF was the most extensively studied type of CVD, reported in 27 studies. Acute coronary syndrome (ACS) and myocardial infarction (MI) followed with four and three studies respectively. All the others had one study each reporting on the variables of interest. This has been illustrated in Fig. 3.

Fig. 3.

Types of CVDs Reported in Included Studies

Quality Appraisal Results of Included Studies

A summary of the quality appraisal has been presented in Supplementary File 2. Of the 37 included studies, 35 were assessed as low risk of bias and two were rated as moderate risk. The 35 low-risk studies comprised 27 cohort studies [26–47, 51, 57, 58, 60, 62] and eight cross-sectional studies [48–50, 52, 55, 56, 59, 61]. The studies rated moderate risk included one cohort study [54] and one cross-sectional study [53].

The unmet JBI items contributing to lower quality assessments were absence of reported strategies to address incomplete follow-up, failure to identify and control confounding variables, unclear methods for measurement of exposure or condition and use of inappropriate analytic procedures.

Failure to identify or adequately control confounding variables was noted in 11 studies [27, 30, 33, 37, 47, 51–53, 55, 56, 59]. These studies largely relied on univariable analyses without adjusting for confounders such as age, sex socioeconomic status and comorbidities, although many of these studies measured them. This methodological weakness compromises validity, as the observed associations between predictors and readmission outcomes may be overestimated or underestimated. Lack of reported strategies for addressing incomplete follow-up was identified in 12 records [26, 27, 33, 36, 37, 43, 44, 47, 51, 54, 57, 58]. These studies did not report the extent, reasons or handling of loss to follow-up. The absence of clear follow-up management introduces potential distortion of outcome estimates and weakening the robustness of longitudinal inferences.

Issues related to measurement of exposure or condition were identified in two studies [48, 54]. These limitations stemmed from insufficiently described or non-standardised measurement of the various CVD. This raises significant concerns about the accuracy of diagnosis of CVD, which is very essential in appropriate management. Moreover, such measurement inaccuracies may dilute or exaggerate associations, thereby reducing the reliability of the findings. Additionally, two studies [53, 55] did not utilise appropriate analytical procedures aligned with their study objectives. These studies relied on inadequate statistical analyses, while attempting to identify the predictors of readmission. This was also done in the availability of the measurement of confounding factors [55]. The study by Ogbemudia & Asekhame [53] relied largely on descriptive statistics to classify and quantify the burden of preventable readmissions. For an analytical study aiming to understand why these factors lead to readmission, multivariate analysis is essential to account for potential confounders.

Review Findings

Prevalence of Readmission

Table 3 presents the prevalence of readmission among patients with CVDs. In general, readmission ranged from 1.53% to 67.7% for all patients with any type of CVD. The findings revealed variations in the prevalence of readmission at different time points. The most frequently used time points for measuring readmission were at one month and 12 months, each with eight studies. Readmission at 12 months was measured in five studies. The time points of three months, two months, and 24 months were reported in a few studies. Readmission rate within 30 days ranged 1.5% to 37.1%. The lowest recorded 30-day burdens were found in North and West Africa [40, 60]. Conversely, the highest 30-day rates were concentrated in East Africa, particularly in Tanzania [55], where prevalence reached 37.1% for HF patients. The readmission rate within 6 months ranged from 12% to 49.6% and that of 1 year peaked at 50.8%, indicating an increasing prevalence as the timepoint of assessment increases.

Table 3.

Prevalence of Readmission Among Patients with CVDs

Citation	Design	Sample size	Type of CVD	Period to readmission	Prevalence
Ogah et al. [40]	Cohort study	262	Heart Failure	30 days	1.53%
Elbarbary et al. [60]	Cohort study	400	Acute Coronary Syndrome (Acs)	30 days	5.25% (Patients with HF) 1.5% (Patients with reinfarction)
Okello et al. [42]	Cohort study	274	Heart Failure	30 days	7.3%
Ahadzi et al. [28]	Cohort study	201	Heart Failure	30 days	8.0%
Prattipati et al. [44]	Cohort study	267	Heart Failure	30 days	14.1%
Goli et al. [35]	Cohort study	85	Myocardial Infarction	30 days	20%
Ayenew et al. [31]	Cohort study	572	Heart Failure	30 days	26.4%
Kajembula et al. [48]	Cross-sectional study	134	Heart Failure	30 days	28.4%
Maro et al. [55]	Cross-sectional study	97	Heart Failure	30 days	37.1%
Sani et al. [62]	Cohort study	209	Heart Failure	60 days	7.2%
Sani et al. [45]	Cohort study	1006	Heart Failure	60 days	7.36%
Sliwa et al. [46]	Cohort study	1005	Heart Failure	60 days	7.4%
Ahadzi et al. [28]	Cohort study	201	Heart Failure	3 months	13.9%
Maro et al. [55]	Cross-sectional study	97	Heart Failure	3 months	62.9%
Shavadia et al. [47]	Cohort study	162	Atrial Fibrillation and Atrial Flutter	6 months	12%
Ogah et al. [40]	Cohort study	262	Heart Failure	6 months	12.2%
Ogah et al. [39]	Cohort study	285	Heart Failure	6 months	13.9%
Ogunmodede et al. [41]	Cohort study	148	Heart Failure	6 months	16.2%
Adebayo et al. [27]	Cohort study	160	Heart Failure	6 months	23.8%
Ali et al. [30]	Cohort study	100	Heart Failure	6 months	31%
Iheji et al. [36]	Cohort study	160	Heart Failure	6 months	35.6%
Pallangyo et al. [43]	Cohort study	419	Heart Failure	6 months	49.6%
Varwani et al. [51]	Cohort study	230	Acute Coronary Syndrome	1 year	6.6%
Alamaw et al. [29]	Cohort study	429	Myocardial Infarction	1 year	12.35%
Nkoke et al. [38]	Cohort study	254	Stroke	1 year	15.2%
Schamroth et al. [58]	Cohort study	548	Acute Coronary Syndrome	1 year	20.3%
Uys et al. [57]	Cohort study	174	Acute Coronary Syndrome	1 year	27%
Okello et al. [42]	Cohort study	274	Heart Failure	1 year	31.4%
Hertz et al. [34]	Cohort study	59	Myocardial Infarction	1 year	50.8%
Bamba-Kamagaté et al. [33]	Cohort study	396	Heart Failure	2 years	49.5%
Manwere et al. [52]	Cross-sectional study	65	Heart Failure	2 years	67.7%
Kingué et al. [37]	Cohort study	1291	Rheumatic Heart Disease	Unspecified	13.6%
Sana et al. [59]	Cross-sectional study	437	Heart Failure	Unspecified	28.83%
Owona et al. [56]	Cross-sectional study	160	Heart Failure	Unspecified	30.6%
Gedfew et al. [61]	Cross-sectional study	663	Heart Failure	Unspecified	35.7%
Sadiq et al. [49]	Cross-sectional study	353	Heart Failure	Unspecified	38.5%
Tukeni et al. [50]	Cross-sectional study	69	Heart Failure	Unspecified	56.5%

CVD Cardiovascular disease, HF Heart Failure

The overall readmission rates among cohort studies were comparatively lower (1.5%–50.8%) compared to cross-sectional studies (28.4%–67.7%). A sub-analysis of the prevalence rates revealed the prevalence rates were higher for studies that examined the prevalence after 30 days than studies that assessed readmission within one month. Furthermore, more than half of the studies reported prevalence of readmissions of 20% or higher. Among these, 16 studies were conducted among patients with HF [27, 30–33, 36, 42, 43, 48–50, 52, 55, 56, 59, 61] and four among patients with ACS and MI [34, 35, 57, 58]. Additionally, the highest readmission rate (67.7%) was recorded among patients with HF [52]. Countries in the Western and Southern parts of SSA reported higher rates of readmission compared to countries in the Eastern part of SSA.

Risk Factors of Readmission Among Patients with CVDs

This review identified a number of risk factors associated with readmission among patients with CVDs, presented in Table 4. These determinants have been categorised into socio-demographic factors, patient’s clinical factors, facility-related factors and self-management practices. Pertaining to socio-demographic factors, being of age > 65 years [31, 32, 54, 60], coming from southern regions of SSA [45, 46] and being a rural resident [31, 32] were mostly reported as the risk factors for readmission. The methodological quality of the studies identifying age, residency, and geographic region as risk factors was predominantly low risk.

Table 4.

Risk Factors of Readmission Among Patients with CVDs

Themes	Specific risk factors	CVD Category	Authors
Socio-demographic factors	Being a woman	Acute Coronary Syndrome	[60]
	Being of old age (> 65 years)	Heart Failure Acute Coronary Syndrome	[31, 32, 54, 60]
	Being in a rural residence	Heart Failure	[31, 32]
	Being from Southern part of SSA	Heart Failure	[45, 46]
	Being Nigerian	Heart Failure	[28]
Clinical factors	Being hypertensive	Heart Failure	[50, 53, 55]
	Having right abdominal pain	Heart Failure	[26]
	Having lower systolic blood pressure	Heart Failure	[26]
	Having arrhythmia	Heart Failure Rheumatic heart disease Acute Coronary Syndrome	[37, 50, 53, 55, 60]
	Presence of mitral regurgitation	Heart Failure	[40]
	Severe valvular lesions	Rheumatic heart disease	[37]
	Having pulmonary thromboembolism	Heart Failure	[53]
	Having Cor Pulmonale	Heart Failure	[46]
	Right or left chamber dilation	Rheumatic heart disease	[37]
	Having advanced stage of disease:
	LVEF (< 40%)	Heart Failure Myocardial Infarction Acute Coronary Syndrome	[29, 54, 60]
	NYHA Class II to IV	Heart Failure	[28, 36, 55]
	Discharge NIHSS > 7	Stroke	[38]
	Having a sub-acute HF on admission	Heart Failure	[28]
	Admission GCS of less than 14	Stroke	[38]
	Being hyperthermic	Stroke	[38]
	Being anaemic	Heart Failure	[31, 36, 50, 53]
	Having chronic comorbidities (asthma, CKD, COPD, HIV/AIDS, malignancy)	Heart Failure Myocardial Infarction Acute Coronary Syndrome	[29, 31, 46, 53, 60, 62]
	Having infections (pneumonia, UTI, infective endocarditis)	Heart Failure	[50, 53]
	Having low serum sodium (< 130 mmol/L)	Heart Failure	[40]
	Having eGFR < 60 mls/min/ 1.73m2	Heart Failure	[36, 41]
	Having elevated BUN	Heart Failure	[45, 46]
	Having higher heart rate on admission	Heart Failure	[28, 54]
	Having higher respiratory rate	Heart Failure	[61]
	Longer duration of medication intake	Heart Failure	[61]
	Having higher admission temperature (> 37.5 °C)	Stroke	[38]
	Having pleural effusion	Heart Failure	[49]
	Having baseline rales of 2/3	Heart Failure	[45, 46]
	Having a syncope	Myocardial Infarction	[29]
	Having a BMI < 19 Kg/m2	Heart Failure	[40]
Facility-related factors	Longer hospital stays (≥ 7 days)	Stroke Heart Failure	[32, 38]
	Being discharged without beta-blockers, statins, clopidogrel or aspirin	Heart Failure Myocardial Infarction	[29, 32]
	Being discharged with a diuretic	Heart Failure	[32]
Self-management factors	Inadequate adherence to treatment regimen	Heart Failure	[43, 49, 50, 53, 55]
	Poor self-care practices	Heart Failure	[52]
	Non-adherence to clinic visits	Heart Failure	[48]

BMI Body Mass Index, CKD Chronic Kidney Disease, LVEF Left Ventricular Ejection Fraction, GCS Glasgow Coma Scale, NYHA New York Heart Association, UTI Urinary Tract Infection, NIHSS National Institutes of Health Stroke Scale, BUN Blood Urea Nitrogen, COPD Chronic Obstructive Pulmonary Disease, eGFR estimated Glomerular Filtration Rate

Clinical status factors accounted for the largest proportion of the determinants of readmission. Factors such as being hypertensive [50, 53, 55], having arrhythmias [37, 53, 55], mitral regurgitation and pulmonary thromboembolism were reported by included records. It was also further indicated that having advanced stages of a CVD was a predictor of rehospitalisation. For patients with HF, studies reported left ventricular ejection fraction (LVEF) < 40% [29, 54, 60] and New York Heart Association (NYHA) Class II to IV as predictors of readmission [28, 36, 55, 60]. Similarly, a National Institutes of Health Stroke Scale (NIHSS) score of more than seven was found to be a predictor of readmission among stroke patients [38], likewise severe valvular lesions among patients with rheumatic heart disease [37]. Furthermore, clinical markers such as anemia [31, 36, 50, 53], infections [50, 53], worsening biochemical indicators and chronic comorbidities [29, 31, 45] emerged as predictors of readmission among patients. The evidence supporting clinical status factors as determinants of readmission was generally of high quality. The majority of studies employed objective clinical measures and appropriate analyses to ascertain the predictability of readmission in patients with advanced CVD stage, comorbidities and poor biochemical indicators. A small number of studies were rated moderate risk of bias. This was mainly due to inadequate control of confounding and inadequate reporting of follow-up.

Additionally, various facility-related factors were identified as determinants of readmission. These included factors such as having prolonged hospital stays [31, 38]. Moreover, discharging a patient with HF without a beta-blocker [31] was a predictor for readmission and for patients with MI, this encompassed being discharged without aspirin, clopidogrel or statins [29]. Furthermore, patients’ self-care practices also emerged as determinants of readmission. This included poor adherence to treatment regimen [43, 49, 50, 53, 55].

Discussion

Summary of Findings

This systematic review synthesised evidence from 37 studies examining the prevalence and determinants of hospital readmissions among patients with CVD in Africa. Overall, the findings indicate that readmissions remain a substantial burden across the continent, with prevalence estimates varying widely across settings and timeframes. Readmission rates ranged 1.53% to 67.7% for all CVD. The risk factors were thematised into socio-demographic, clinical, facility-related, and self-management factors. Older age and rural residence were consistent socio-demographic predictors. On the other hand, hypertension, arrhythmias, mitral regurgitation, anemia, poor estimated glomerular filtration rate (eGFR), and advanced disease staging stood out as major clinical contributors. Furthermore, facility-level determinants included prolonged hospital stays, lack of essential discharge medications, and suboptimal discharge planning. Self-management barriers such as medication non-adherence, poor clinic follow-up, and inadequate self-care practices were also found to be predictors of readmission. Collectively, these findings indicate that CVD readmissions in Africa are influenced by an interplay of patient, clinical, health system and behavioral factors.

Prevalence of Readmission Among Patients with CVDs

This review found out different timepoint assessments of readmission among patients with CVDs ranging from one month to two years, thus providing enough time to determine readmission rate. This is in line with global studies that have reported similar timepoints of readmission among patients with CVDs [15–17, 20]. Overall, readmission rates were highly variable, ranging from 1.53% to 67.7% across different patient populations and CVD categories. This variability may be explained by the methodological heterogeneity across studies, including differences in population and settings, study design, follow-up duration and outcome measurement. For instance, it was revealed in this review that cohort studies reported relatively lower readmission rates compared to cross-sectional designs, likely due to follow-up and reduced reliance on patient recall. In addition, clinical heterogeneity, such as differences in CVD type and severity, may have further influenced the observed readmission rates. Nevertheless, the observation that more than half of the included studies reported readmission rates exceeding 20% suggests that rehospitalisation remains a persistent challenge in the management of CVD across the region.

The 30 days readmission rate among participants in this review ranged from 1.5% to 37.1%. These findings align with the findings of a review by Labrosciano et al. [15] which found 6.3 to 27% readmission rate within 30-days for patients with CVDs. The higher upper limit of the 30-day readmission rate (37.1%) observed in the present review likely reflects differences in study design, patient characteristics, and healthcare system capacity. However, this high readmission rate was reported by a cross-sectional study [55]. In estimating readmission outcomes, cohort study designs are methodologically more appropriate, as they allow adequate follow-up for events to occur and apply analytical approaches to account for incomplete follow-up and reduce bias. Nevertheless, these findings suggest that 30-day readmission rates among patients with CVDs in Africa poses a significant challenge, as it has been reported among patients in other settings.

Furthermore, studies that followed patients for a longer reported higher prevalence rates. This depicts a link between the length of the follow-up period and the measured prevalence of readmission, with longer tracking times revealing more intricate outcomes among patients. The highest readmission rate (67.7%) was recorded among patients with HF [52]. However, this estimate was derived from a cross-sectional study involving 65 participants, in which readmission history over a two-year period was assessed using self-reported recall subjecting reports to higher propensity of recall bias and misclassification. That notwithstanding, this adds further evidence to the high burden of HF in Africa. Studies have indicated that HF is highly prevalent in Africa affecting millions, has high case fatalities and poor outcomes and is the leading cause of admission in specialised cardiology centers [4, 63]. Despite HF contributing the largest to the high readmission rates, studies reported concerning readmission rates for patients with other CVDs. For example, one study reported 50.8% prevalence of readmission among a cohort of 59 patients with MI [34]. This indicates that other CVDs aside from HF also contribute significantly to the burden of hospital readmissions in Africa. The predominance of HF in the included studies reflects its well-documented status as one of the leading causes of hospitalisation and rehospitalisation among patients with CVD in Africa.

The studies reported varying readmission rates across different types of CVDs in different geographical regions. However, these studies were concentrated in the Western, Eastern and some part so Southern parts of SSA and a few from North Africa. This concentration buttresses the reported burden of CVDs in these regions of Africa. For instance, a review by Minja et al. [63] revealed that the Western and Eastern regions of SSA have shown a steeper rise in CVD mortality in recent years. Perhaps, rapid epidemiological transition, increasing prevalence of CVD risk factors, constrained health system capacity and limited access to structured post-discharge care could explain the high burden of CVD-related readmissions in these regions. Additionally, five of the included studies were multinational using multicenter registries with large cohorts. This indicates enhanced efforts of researchers and institutions in Africa to collaborate across national borders and generate large-scale, multi-country data on CVDs.

More than three-quarters of the studies were conducted using cohort design spread across both prospective and retrospective approaches. In line with this, patients were followed up at varying timepoints from 30 days up to two years. This provides a strong longitudinal perspective of tracking patient outcomes and identifying relationships between risk factors and readmission events. Pertaining to the type of CVD, HF was reported in 73% of the included studies. This depicts the predominant focus of CVD readmission research on HF. This adds up to the evidence that HF is a primary driver of hospital readmissions and a major concern for healthcare systems in Africa. For instance, a review by Yuyun et al. [5] found that HF is not only highly prevalent in SSA but also presents earlier in life, often with poorer outcomes and is driven by different primary etiologies compared to developed countries.

Risk Factors of Readmission Among Patients with CVD

This review identified various risk factors that increase patients’ likelihood of readmission. The risk factors were organised thematically into socio-demographic, clinical status, facility-related and self-management practices factors.

Socio-demographic Factors

The findings of this review revealed that older age, specifically being 65 years or older, is a significant risk factor for unplanned readmission in HF patients. One study reported a more than threefold increase in the hazard of readmission compared to younger adults [31]. This predictive factor is attributable to a number factors including physiological deterioration, a higher burden of comorbidities, reduced organ function and increased susceptibility to psychological stress among the aged. Furthermore, patients from rural settings were found to have a significantly higher risk of unplanned readmission. This disparity may be linked to lower health literacy levels, a slower adoption of and adherence to healthy behaviors, and a lack of access to healthcare facilities and resources which are contextual problems severely impacting healthcare in Africa. For instance, rural–urban access disparities such as uneven facility distribution, transport and workforce shortages have been highlighted as barriers to timely care and long-term disease management [64].

Clinical Factors

Clinical factors represented the most consistently reported predictors of readmission across the included studies. These included underlying conditions such as hypertension, arrhythmias, mitral regurgitation and pulmonary thromboembolism. These underlying cardiac related conditions have been reported as major risk factors for poor outcomes of a number of CVD. For example, hypertension has been indicated as the most prevalent risk factor for HF and inadequate control has been linked to worsening HF [65]. Mitral regurgitations also have significant impact on the prognosis of HF. A review by Pagnesi et al. [66] found that patients with moderate to severe mitral regurgitation tend to have larger left ventricular dimensions, including increased LV end-diastolic and end-systolic diameters, as well as larger left atrium diameter. Other clinical factors that were reported included having advanced stages of CVDs. This was indicated among studies that reported reduced LVEF, poor NYHA staging and poor NIHSS scores among stroke patients. These suggest poor control and treatment of the underlying CVD, thus, increasing patients’ susceptibility to readmissions.

A number of studies also reported anemia as a significant predictor of readmission among patients with HF. Anaemia may worsen cardiac function and contribute to clinical deterioration requiring rehospitalization. Other risk factors reported included infections and poor biochemical markers including lower eGFR. These findings highlight the importance of disease severity and comorbidity burden as key drivers of rehospitalisation among patients with CVD. However, the quality appraisal revealed that some studies did not assess CVDs using standardised diagnostic and staging criteria. This could stem from challenges commonly encountered in resource-constrained healthcare setting such as restricted access to diagnostic tools, limited specialist availability and disparities in adherence to clinical guidelines. Consequently, these methodological shortcomings and associated findings reflect broader health system limitations in the comprehensive screening, diagnosis, and management of CVDs in some African hospitals.

Facility-related Factors

This review also found some facility-related factors that are linked to readmission among patients with CVDs. Patients who had longer hospital stays, or were discharged without some medications were more likely to be readmitted. Longer hospital stays may indicate the severity of the conditions patients may present with. Conversely, it may also be influenced by delays in diagnostic investigations, limited availability of specialist services, medication stock-outs, and challenges in arranging timely discharge and follow-up care within many African healthcare facilities. It is worth noting that patients who spend longer times at the hospital are susceptible to nosocomial infections [67]. Perhaps, this may explain their return to the hospital for treatment and poor outcomes after discharge. This review found out that some patients were not being discharged with some essential medications that have been enumerated in standard treatment. This likely increases their susceptibility to readmissions. For instance, patients with MI who were not put on aspirin or clopidogrel had higher readmission risks [29]. Similarly, patients with HF who were not put on beta-blockers had higher likelihoods of readmission [32]. These risk factors suggest shortfalls in standard treatment and suboptimal discharge planning for patients with CVDs in some facilities in Africa. Also, findings suggest that aspects of healthcare delivery, particularly discharge planning and medication provision, play an important role in preventing avoidable readmissions.

Self-management Factors

Self-management practices emerged as an important determinant of readmission across several studies. In particular, medication non-adherence and poor follow-up attendance were repeatedly associated with increased likelihood of rehospitalisation. This indicates the importance of adequate client involvement in care. Similar findings linking medication non-adherence to readmissions among patients with CVDs have been reported in other reviews [68, 69]. For patients with chronic conditions such as CVDs, long term commitment to adherence is essential to ensure effective disease management. Perhaps, an explanation to this may be that patients who do not take their medications become susceptible to worsening disease progression and poor outcomes. Furthermore, irregular follow-up affects monitoring, dose titration, early detection of deterioration, and reinforcement of self-care. Patients who miss scheduled appointments may be more likely to experience preventable relapses requiring readmission. Nevertheless, these findings may be attributed to contextual factors such as patients not understanding discharge instructions, financial constraints that impede clinic visits and restocking of medications and low awareness of the importance of follow-up. This raises critical questions about the adequacy of follow-up by healthcare workers, the quality of discharge education, patient’s adherence to discharge instructions and the adequacy of support systems for patients. These findings highlighted the essence of strengthening patient education and post-discharge support systems in cardiovascular care.

Implication for Policy, Practice and Research

It was identified in this review that a number of patient- and facility-related factors encompassing discharge planning and standard treatment predicted readmissions. It is therefore imperative that healthcare policies Africa prioritise the implementation of standardised, evidence-based discharge protocols for CVDs. This will help address inconsistencies in treatment and follow-up care. Ministries of Health in Africa and healthcare facilities need to allocate resources to establish cardiac rehabilitation programs and strengthen community-based care models to improve long-term management and reduce readmissions. This includes making available standard equipment that healthcare workers will need in diagnosing patients. Furthermore, there is a critical need for policies that address the rural-urban health disparity, including equitable healthcare workforce distribution to rural areas. Investments are also needed in transportation systems or telemedicine solutions to improve healthcare access for remote populations. Health education policies should mandate clear, culturally appropriate discharge instructions and patient education to improve health literacy and self-management capabilities.

The review findings also call for improvement in clinical practice. To begin with, the appraisal of studies revealed some disparities in the standard criteria for diagnosing patients with CVDs. In practice, clinicians must perform thorough investigation and assessment of patients and also utilise standard diagnostic procedures in diagnosing, staging and treating CVDs. Aside from that, healthcare workers need to enhance comprehensive discharge planning that includes early initiation, comprehensive patient education and scheduling definite follow-up appointments. This review revealed a plethora of clinical factors that predicted readmission among patients with CVDs. In view of that, patients should be comprehensively screened when on admission and during reviews for these risk factors such as anemia, poor eGFR, high BUN and other markers in order to promptly address abnormalities. Furthermore, practice should adopt a multidisciplinary approach to manage comorbidities and ensure comprehensive treatment. To add up, nurses who are usually at the forefront of discharge planning should ensure adequate discharge preparedness of patients before they leave facilities. In line with that, healthcare workers must ascertain readmission risk assessment scores for patients on admission in order to optimise care. Additionally, healthcare facilities need to audit and improve the quality of discharge processes to minimise factors like prolonged hospital stays and ensure consistent provision of standard treatment to patients. Moreover, there is the need for tools and interventions that can help remind patients of review dates and discharge instructions in order to enhance compliance.

With regards to research, future studies must address the significant geographical gaps in evidence, particularly in Central and Northern Africa, to contribute to evidence that encompass CVD-related readmissions. Additionally, future research should focus on other types of CVDs aside HF to provide a more comprehensive overview of CVD-related readmission in Africa. The quality appraisal revealed that cohort study designs dominated the included records. However, some did not report adequately on follow up and loss to follow up. Studies in the future should provide adequate follow up details on patients. Furthermore, there is the need for studies that will investigate the effectiveness of specific interventions. These include standardised discharge protocols, readmission risk assessment tools, patient education programs and follow-up strategies to help reduce readmission rates. Further studies should also explore discharge readiness prior to patients leaving hospital, quality of discharge teachings given to patients, prevalence and determinants of non-adherence and poor self-care and experiences of patients. This will help inform the development of more effective and tailored support programs for patients.

Limitations in this Review

This systematic review has several key limitations. First, the results revealed that studies are geographically skewed, with studies concentrated in Western and Eastern Africa. This limits the representativeness for the entire continent. In addition, the review is dominated by research on HF. This provides less insight into readmissions for other CVDs and skewing the synthesis towards HF. Furthermore, the synthesis of data is hampered by significant heterogeneity in how readmissions were measured, particularly the wide variation in follow-up timeframes across studies. Moreover, this review is based on cohort, cross-sectional and one descriptive study. This limits the findings to associations between variables but not causation. Additionally, this review included all studies irrespective of their scores on quality appraisal. Although this broad inclusion improves the comprehensiveness of the review, findings from studies with moderate or high risk of bias may influence the overall reliability of the synthesis. In view of that, findings of this study should be interpreted with caution. The omission of non-English articles might have potentially resulted in excluding relevant studies.

Another limitation relates to the inability to conduct a meta-analysis of readmission prevalence. Although a relatively large number of studies were included, the studies exhibited substantial methodological and clinical heterogeneity. These differences included wide variation in follow-up periods used to define readmission (ranging from 30 days to two years), differences in study design (cohort and cross-sectional), variability in the types and severity of CVD examined, and inconsistencies in outcome measurement and reporting across studies. Due to these variations, pooling prevalence estimates using meta-analysis could have produced misleading summary estimates. Hence, the findings were synthesised using descriptive and thematic approaches to allow a more appropriate interpretation of the available evidence.

Another limitation relates to the analytical approaches used in the included studies when identifying predictors of readmission. Although several studies employed multivariable models and reported adjusted associations, a number of studies relied on univariable analyses or did not adequately account for potential confounding variables. This variability in analytical methods limited the ability to consistently compare the strength of associations across studies and to synthesise effect size estimates such as odds ratios or hazard ratios. While the methodological quality appraisal using the JBI tools helped identify studies that addressed or failed to address confounding, the inconsistency in reporting across primary studies should be considered when interpreting the reported predictors of readmission. Despite these limitations, a comprehensive search strategy was conducted across multiple databases, enhancing the breadth of evidence captured.

Conclusion

This review demonstrates that hospital readmissions among patients with CVD in Africa remain a substantial and persistent challenge. The synthesis highlights a range of socio-demographic, clinical, facility-related, and self-management factors that increase the risk of readmission. It was revealed that many of these risk factors are modifiable through improved access, guideline-based care and patient support. Nevertheless, evidence remains unevenly distributed, with gaps in Central and Northern Africa and limited focus on non-HF conditions. It is therefore essential to address these knowledge gaps, as well as strengthen discharge planning, medication adherence, follow-up care and patient self-management. These will cumulatively contribute to reducing the burden of avoidable readmissions and improve long-term CVD outcomes across the continent.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CVDs

Cardiovascular Diseases

NCD

Non-communicable Diseases

WHO

World Health Organisation

LMICs

Low- and middle-income Countries

SSA

Sub-Saharan Africa

HF

Heart Failure

ACS

Acute Coronary Syndrome

MI

Myocardial infarction

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

JBI

Joanna Briggs Institute

LVEF

Left Ventricular Ejection Fraction

NYHA

New York Heart Association

NIHSS

National Institutes of Health Stroke Scale

eGFR

Estimated Glomerular Filtration Rate

Authors’ Contributions

FAN, MA and NIEE conceptualised the study. Data collection, analysis, and drafting of the initial manuscript were conducted by FAN, EO, AG and HKK. EO, MD, FAN and AG extracted the data. Database search, screening of articles and removal of duplicates were conducted by FAN, EO HKK and AG. FAN, MD, EO conducted data synthesis and interpretation was done by FAN, MD and EO. Writing of final manuscript was done by FAN, MAA & MA. MA and NIEE supervised the overall review process. All authors read and approved the final manuscript.

Funding

This review received no funding.

Data Availability

All data generated or analysed during this study are included in this article and its Supplementary File 1 and 2.

Declarations

Ethical Approval and Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Competing Interests

The authors declare no competing interests.