Factors affecting hospital readmission rates following an acute coronary syndrome: A systematic review

Abstract Aim To synthesise quantitative evidence on factors that impact hospital readmission rates following ACS with comorbidities. Design Systematic review and narrative synthesis. Data sources A search of eight electronic databases, including Embase, Medline, PsycINFO, Web of Science, CINAHL, Cochrane Library, Scopus and the Joanna Briggs Institute (JBI). Review methods The search strategy included keywords and MeSH terms to identify English language studies published between 2001 and 2020. The quality of included studies was assessed by two independent reviewers, using Joanna Briggs Institute (JBI) critical appraisal tools. Results Twenty‐four articles were included in the review. All cause 30‐day readmission rate was most frequently reported and ranged from 4.2% to 81%. Reported factors that were associated with readmission varied across studies from socio‐demographic, behavioural factors, comorbidity factors and cardiac factors. Findings from some of the studies were limited by data source, study designs and small sample size. Conclusion Strategies that integrate comprehensive discharge planning and individualised care planning to enhance behavioural support are related to a reduction in readmission rates. It is recommended that nurses are supported to influence discharge planning and lead the development of nurse‐led interventions to ensure discharge planning is both coordinated and person‐centred.

to the health care system (Hess et al., 2016;Litovchik et al., 2019;McManus et al., 2016). People with ACS are at high risk for readmission and the rates of readmission following ACS are increasing (Belitardo & Ayoub, 2015). A recent study in Brazil reported 42.6% of people who experienced ACS were readmitted between 30 and 180 days post-discharge (Oliveira et al., 2019). Similarly, the US study of 3536 people post-acute MI, reported a 24.5% readmission rate within one year (Dreyer et al., 2015). A study in Australia and New Zealand indicated that 25% of patients were readmitted or died within 30 days following an acute MI (Labrosciano et al., 2017).
However, a Canadian study of 3411 people who experienced ACS reported the highest readmission rate at 61.7% within one year following discharge (Southern et al., 2014). Readmission to hospital is influenced by a number of factors indicating the occurrence of complications after discharge, which can impact on disease severity (Oliveira et al., 2019), and this reinforces the importance of identifying factors such as the presence of comorbidities, clinical condition and patients' individual features both during and after hospitalisation to reduce complications that may affect readmission outcomes (Oliveira et al., 2019;Walraven et al., 2012). Readmission to hospital among ACS patients is related to increased medical costs reduced quality of life and poorer health outcomes (Budiman et al., 2016). Identifying factors that affect readmission has the potential to improve quality of life, discharge follow-up, care coordination and reduce avoidable healthcare expenditure (Dreyer et al., 2015;Oliveira et al., 2019;Southern et al., 2014). Readmission is a multifactorial phenomenon and influenced by underlying comorbidities, patient factors, healthcare system and organisational factors (Rocca et al., 2020). To date, a study has assessed and evaluated readmission related to patients with MI (Labrosciano et al., 2017); however, readmission rates post-ACS are not clear (Kwok et al., 2017;Nguyen et al., 2018) and further evidence on factors influencing readmission associated with ACS is minimal (Oliveira et al., 2019). Moreover, prior research on readmission rates following an ACS plus comorbidity has not been systematically reviewed and factors affecting hospital readmission are currently lacking. Therefore, improved knowledge of factors affecting readmissions is important to develop targeted strategies to reduce readmissions rates. The aim of this review was to identify factors that are associated with hospital readmission within 30 days following ACS.

| AIM
To synthesise quantitative evidence on factors that impact hospital readmission rates following ACS with comorbidities.

| Methods
A systematic review was conducted in accordance with the Joanna Briggs Institute (JBI) methodology for systematic reviews of quantitative evidence (Moher et al., 2010). This review follows the PRISMA (File S1) (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement reporting (Moher et al., 2010) and was registered with PROSPERO (*registration number will be inserted here post-review).

| Study selection and inclusion criteria
All search results were imported into EndNote X9 with the aim of removing duplicates records. Two independent reviewers screened the titles and abstracts of the full-text articles against the inclusion criteria set out in the protocol. All peer-reviewed experimental and observational study designs were considered for inclusion. Only studies classified as primary research, published in English since 2001 in a peer-reviewed journal were considered. This time frame was considered as preliminary searches indicated an upsurge in the discourse on treatment adherence potentially influencing 30-day readmission rates.
This review considered populations diagnosed with acute coronary syndrome or ischaemic-related cardiovascular conditions and identified comorbidity on initial admission with a subsequent presentation to hospital for treatment within 30 days. The diagnosis of ACS includes the suspicion or confirmation of acute myocardial ischaemia or infarction. Non-ST-elevation myocardial infarction (NSTEMI), ST-elevation MI (STEMI) and unstable angina are the three traditional classifications of ACS, which were considered for this review. This review focused on adults aged 18 years and over. Studies were excluded from the review if they included participants under 18 years old, had non-ischaemic heart failure or had received open heart surgery. Animal studies and grey literature were excluded. This review assessed the readmission rate of patients occurring within 30 days of receiving care for an acute coronary syndrome episode in an acute care setting. The outcomes of interest for this review were the factors that influenced 30-day readmission in patients that had an initial hospitalisation for acute coronary syndrome or ischaemic-related cardiovascular condition. These factors could include, but were not limited to, education, treatment plan implementation and adherence, communication, quality of life, general health outcomes, physical well-being, emotional wellbeing, comorbidities, stress or psychosocial well-being as assessed through data retrieved from medical records/database review, patient questionnaires or patient interviews.

| Quality appraisal and data extraction
Methodological quality and risk of bias within included studies were independently appraised by two reviewers. JBI Critical appraisal tools were utilised to assess the quality of studies (Manual, 2014). Any disagreements that arose between reviewers were discussed and resolved by all authors. A Grading of Recommendations Assessment, Development and Evaluation (GRADE) was completed to evaluate the certainty of the evidence by assessing consistency of results between studies, directness and precision of findings, study limitations and probability of publication bias. The overall quality of evidence was categorised as high, moderate, low or very low. Two researchers completed the GRADE assessments independently. Any discrepancies between reviewers were discussed by the research team.
Data were extracted from included studies using the JBI standardised data extraction tools by study design (Piper, 2019). Data extracted included country, study design, type of patient (referring to reason for initial admission), sample size, sample demographics (age and sex), intervention details, data source or measurement and comorbidity or outcome of interest (referring to the factors that could impact on readmission rates), as well as key findings. Queries on data were followed up with the corresponding authors of each study. One author was contacted to gain clarification on the data, as conflicting sample characteristics were reported; however, no return correspondence was received.

| Data synthesis
Key findings have been narratively synthesised to demonstrate key barriers and facilitators for 30 days readmission rates. Metaanalysis was not completed due to heterogeneity in data collection time points, source of data, study design and collection methods. Therefore, a narrative synthesis was conducted according to the Synthesis Without Meta-analysis (SWiM) guidelines (Campbell et al., 2020). The review was approved by the Human Research Ethics Committee of University (2021-02514).

| Study inclusion
A PRISMA flow diagram outlining the selection of eligible studies is presented in Appendix 2. Initially, 13515 articles were retrieved from databases and manual searching. After the removal of duplicates, 9132 papers remained. These titles were screened by two independent reviewers. A total of 818 abstracts were reviewed for suitability. Full texts of 321 articles were assessed for eligibility by two independent reviewers, with 297 excluded for not meeting inclusion criteria. Reasons for exclusion included the primary diagnosis on admission (n = 104), patients who underwent a coronary artery bypass grafting procedure during the admission (n = 38), readmission outside of the 30-day period (n = 62), comorbidities not reported (n = 80), grey literature (n = 7), reviews (n = 4), non-adult sample (n = 1) and paper not published in English (n = 1). An overall GRADE quality rating of low was assigned to findings in relation to factors that influence 30 days readmission rates, mostly due to the nature of the study design being observational.

| Factors affecting hospital readmission
Studies focused on identifying factors that affected readmission within 30 days among ACS patients and across the review studies, a variety factors related to higher hospital readmission rates within a 30-day period. The factors have been presented in four categories: socio-demographic factors, behavioural factors, comorbidities and cardiac disease.
Age was strongly associated with readmission within 30 days, with increased age associated with increased rates of readmission (Dharmarajan et al., 2013;Dodson et al., 2019;Hess et al., 2016;Khera et al., 2017;McManus et al., 2016;Przybysz-Zdunek et al., 2012;Tripathi et al., 2019). However, three studies (Dreyer et al., 2015;Khera et al., 2017;Southern et al., 2014) reported that older age was associated with a lower risk of readmission. One study (Li et al., 2019) demonstrated that there was no association between age and readmission. Two studies suggested that lower household income was associated with high readmission rates (Cheung et al., 2018;Khera et al., 2017). Two studies (Li et al., 2019;Tripathi et al., 2017) did not report an association between household income and rates of readmission. The association between race and readmission was examined in two studies that reported non-white patients (Hess et al., 2016) and Hispanic patients (Rodriguez et al., 2011) were more likely to be readmitted within 30 days.

| Behavioural factors
Four studies (Cheung et al., 2018;Dodson et al., 2019;Hess et al., 2016;Mahmoud & Elgendy, 2018) identified an association between smoking and readmission. Cigarette smoking was significantly associated with a higher likelihood of readmission, irrespective of whether the patient was a current smoker (Cheung et al., 2018; Mahmoud & Elgendy, 2018). One study indicated an association between the status, current smoker and readmission within 30 days (Dodson et al., 2019). However, another study identified that cigarette smoking was linked with lower rates of unplanned readmission within 30 days (Hess et al., 2016).

| Cardiac disease factors
Of the 24 studies reviewed, five studies did not report cardiac factors as the principal diagnosis for hospital readmission (Borzecki et al., 2016;Gasbarro et al., 2015;Khera et al., 2017;Nuti et al., 2016;Zabawa et al., 2018). However, a recurrent ischaemic event, including re-infarction and unstable angina, was consistently reported as the leading cause of readmission in 12 studies

| DISCUSS ION
To the best of our knowledge, this is the first review to assess the prevalence of readmission within 30 days among people with ACS and comorbidity. This review was based on 24 studies that met the inclusion criteria. The findings suggest that hospital readmission within 30 days varied among persons who had experienced ACS and many factors contributed to an increased risk of readmission. The most reported factors associated with an increased risk of readmission were gender, age, ethnicity, smoking status, recurrent ischaemic events, arrhythmia, heart failure, renal disease, anaemia, diabetes, hypertension and pulmonary complications. Of interest, comorbidities were often reported as related to readmission, however, cardiac disease factors were most often reported as the principal diagnosis for hospital readmission. Further clarity on the reason for readmission, for example, recurrent MI, acute kidney injury, exacerbation of heart failure or pneumonia versus the factors related to readmission, for example history of renal disease, existing lung disease and diabetes is needed within the literature to make this distinction clearer.
Hospital readmission in people with ACS is increasingly rec- Reducing readmission during this intense period of recovery and adjustment requires hospital intervention that shares characteristics of patient-centred care involving the interdisciplinary team and better discharge planning practices.
This review also indicates that readmission varied by gender.
Females had a higher likelihood of readmission within 30 days compared to males. This may be because they have home/family roles, including providing care for family members, undertake more domestic work and additional caregiving responsibilities.
The influence of these factors could add to their vulnerability for readmission and can act as a potential explanation for the gender differences in readmission. This finding links with previous research in which females had worse physical and mental health outcomes after MI (Cenko et al., 2018;Shih et al., 2019). Thirty days post-index hospitalisation, females were more likely to be readmitted than males (38.3% versus 29.6%; p < .001) (Wasfy et al., 2013). This disparity has been related to a number of potential factors including actions taken or omitted during the initial hospital stay, a consequence of incomplete treatment or coordination of post-discharge care (Kwok et al., 2018) and may include the roles and responsibilities of women in providing care for children and family that impacts on recovery and makes women more susceptible to readmission (Rose et al., 1996). Further studies need to be conducted to determine why women are more likely to be readmitted than men (Dreyer et al., 2015). Our findings also suggested racial disparities in readmission rates. Hispanic and nonwhite groups have a higher risk of readmission, compared with non-Hispanic and white patients in Spain and USA, respectively (Correa-de-Araujo et al., 2006;Joynt et al., 2011). One study suggested that higher readmission rates among Hispanic patients may be related to lower quality care, for example Hispanics were less likely to be offered an assessment of left ventricular function, (Correa-de-Araujo et al., 2006;Joynt et al., 2011). This finding is echoed in a further study (Joynt et al., 2011) that highlighted disparities in readmission rates between black and white patients in the USA and related to race itself as well as to the site where care was provided.
Prominent non-cardiac factors that influenced readmission rates were pulmonary causes or complications, hypertension, diabetes and renal disease, followed by cardiac factors, including recurrent ischaemic event, PCI, arrhythmia and heart failure. Having the diagnosis of ACS and the presence of comorbidities places people at high risk for hospital readmission (Condon & Mcdonnell, 2012;Gandjour et al., 2012). Reported comorbid conditions varied by studies, some studies considered the number of chronic conditions or the Charlson comorbidity score to assess comorbidity. A similar study also indicated that patients with acute myocardial infarction (AMI) and diabetes were twice as likely to be readmitted than those without diabetes (Tavani et al., 2002). Those who experienced an AMI and were diagnosed with renal disease are high risk of readmission within 30 days (Dunlay et al., 2012). Indeed, previous studies indicated that 8% of people were readmitted within 30 days after PCI (Yost et al., 2013). However, a recent review suggested that there is still ongoing debate on whether PCI can decrease 30-day readmission rates (Wang et al., 2019). Studies indicated that 58% of recurrent coronary readmissions were due to unstable angina (Yudi et al., 2019). This review summarised the factors that influenced readmission within 30 days following ACS. It highlights the need for early follow-up with attention to the predictors of readmission, and interventions to reduce readmissions within 30 days for this population.

| S TRENG TH S AND LIMITATI ON S
The strength of this review was the implementation of a thorough search strategy across eight databases. The JBI manual provided a comprehensive guide to conducting this systematic review. However, this review has limitations. First, due to heterogeneity among included studies, a meta-analysis was not possible; therefore, a narrative review of the existing evidence was provided. Secondly, the inclusion of English only studies could result in relevant studies being missed. Thirdly, heterogeneity among sampling made some comparisons challenging. Apart from three studies, readmission rates based on living conditions and marital status was not reported. Further investigation is needed to examine if these factors play a role in readmission rates, and why marital status. Fourthly, most of reviewed studies were retrospective and prospective designs, and limited RCT and experimental studies have been conducted to demonstrate the cause and effect of relationships between factors. Lastly, studies have used information from databases, administrative data and clinical data making it difficult to generalise some of the findings. Further research is needed to examine readmission following medical index admissions, and to develop and evaluate the effectiveness of nurseled interventions to reduce readmission among this population. Such interventions need to be integrated with cardiology services to target conditions with the highest rates of readmission in order to improve patient's outcome and reduce preventable readmissions.

| CON CLUS I ON AND IMPLIC ATI ON FOR PR AC TI CE
This review identified comorbidities and cardiac factors that were associated with readmission and there is clear and convincing evidence that hospital readmission is prevalent in patients with ACS.
Recurrent ischaemic events, heart failure and arrythmias were the principal cardiac factors related to readmission. Meanwhile, common comorbidities such as kidney disease, hypertension, diabetes and pulmonary complications were significant factors related to readmission. However, little evidence indicates the degree to which readmission may be preventable. Some factors such as smoking, with diabetes, anaemia and hypertension are modifiable factors and can be influenced by nurses. Therefore, efforts among nurses to reduce readmission could include comprehensive discharge planning, individualised care planning, behavioural support and home visits.
Interventions that address patient need, especially unmet needs have the potential to improve health outcomes and reduce readmission rates. To effectively reduce hospital readmissions in individuals with ACS, nurses can identify individuals at high risk for readmission based on the risk factors highlighted in this review. Addressing the issues related to disparities is challenging; however, the evidence has highlighted differences in the investigations undertaken, care provided and attention to discharge planning. The findings in this review highlight the key areas related to readmission and those areas that are modifiable to inform and shape the development of targeted strategies to reduce readmission rates among people post-ACS.

| RELE VAN CE OF CLINI C AL PR AC TI CE
• This review highlights that healthcare providers, particularly nurses need to provide comprehensive and individualised patientcentred interventions to reduce readmission by identifying factors related to readmission and unmet patient needs.

A PPE N D I X 2 S TU DY S E LEC TI O N A N D PR I S M A FLOW D I AG R A M A PPE N D I X 3 M E TH O D O LO G I C A L QUA LIT Y O F S TU D I E S JBI critical appraisal checklist for cohort studies
Abbreviation: Y, yes; N, no; U, unclear; NA; not applicable.

Questions:
1. Were the two groups similar and recruited from the same population?
2. Were the exposures measured similarly to assign people to both exposed and unexposed groups?
3. Was the exposure measured in a valid and reliable way?
5. Were strategies to deal with confounding factors stated?
6. Were the groups/participants free of the outcome at the start of the study (orat the moment of exposure)?
7. Were the outcomes measured in a valid and reliable way?
8. Was the follow-up time reported and sufficient to be long enough for outcomes to occur?
9. Was follow-up complete, and if not, were the reasons to loss to follow-up described and explored?
10. Were strategies to address incomplete follow-up utilised?

Questions:
1. Is it clear in the study what is the 'cause' and what is the 'effect' (i.e. there is no confusion about which variable comes first)?
2. Were the participants included in any comparisons similar?
3. Were the participants included in any comparisons receiving similar treatment/care, other than the exposure or intervention of interest?
4. Was there a control group?
5. Were there multiple measurements of the outcome both pre and post the intervention/exposure? 6. Was follow-up complete and if not, were differences between groups in terms of their follow-up adequately described and analysed?
7. Were the outcomes of participants included in any comparisons measured in the same way?
8. Were outcomes measured in a reliable way?
9. Was appropriate statistical analysis used?

JBI Critical Appraisal Checklist for Randomised Controlled Trials studies
Abbreviation: Y, yes; N, no; U, unclear; NA; not applicable.