Quality indicators for acute cardiovascular diseases: a scoping review

We conducted a scoping review according to a predefined protocol based on the following five-stage approach developed by the Joanna Briggs Institute (JBI): Stage 1, identifying the research question; Stage 2, identifying relevant studies; Stage 3, study selection; Stage 4, charting the data; and Stage 5, collating, summarizing, and reporting the results [13, 14]. This scoping review followed the preferred reporting items for systematic reviews and meta-analyses extension for scoping reviews statement, and adopted established methodological scoping review frameworks and recommendations [15‐17].

We used the JBI population, concept, and context framework for scoping reviews, to define the inclusion criteria [14]. Our scope was existing literature published after 2000 in which authors built original QIs that focused on acute major cardiovascular diseases (ACS, AHF, and AAD). All published studies that targeted adult (age ≥ 18 years) patients who were hospitalized with acute major cardiovascular diseases diagnosed by physicians were extracted. We reviewed the existing literature that satisfied the following conditions: (1) QIs including any form of quality measures, such as quality metrics and performance measures; (2) QIs assessing any component of Donabedian’s model, namely, any structure (i.e. human resources and hospital equipment), process (i.e., diagnosis and treatment), and outcome measurement (i.e., patient status) [18]; (3) QIs being used in acute care settings; (4) the literature search process (i.e., literature review or systematic review); and (5) a predefined QI creation process (i.e. Delphi or modified Delphi method, or another decision-making process). Studies that assessed or validated the QI sets from existing studies or guidelines were excluded. The context in this review was limited to acute care settings where hospitalization occurred; thus, outpatient and chronic care settings were excluded. There were no restrictions regarding cultural factors, geographic location, race, gender, or particular settings.

Following the initial limited search, a systematic search was performed across MEDLINE and EMBASE on June 24, 2021 (Supplementary methods). We also checked the reference lists of the included studies, including international guidelines for major cardiovascular diseases. Furthermore, we searched the websites of relevant organizations, including the Agency for Healthcare Research and Quality, the National Institute for Health and Care Excellence, and the Australian Commission on Safety and Quality in Health Care. Only the latest article was included in the analysis if updated QI sets had been reported from the same organization. There were no language restrictions. Conference abstracts, systematic reviews of secondary data of existing studies which were not used for the literature search for QI development, and case studies were excluded, following the predefined protocol [13].

Two reviewers (KK and YI) independently conducted the literature search. Any disagreements on study selection were resolved through discussion. The reasons for the exclusion of studies in the full-review process are presented in a flow diagram (Fig. 1).

Two researchers conducted data extraction using a standardized data collection form. First, a summary of each study, including the name of the first author, publication year, study setting (countries of origin), target disease, review and consensus-making process, and the number of QIs, was created. Second, the details of the QIs, including clinical setting, definition of QI, and number of publications cited, were summarized according to the predefined protocol [13]. Any disagreements were resolved by discussion, and if this failed, a resolution was reached through a third researcher (YM). For diseases with a large number of QIs, the most common QIs are listed in the Results section, and all items are presented in the Supporting Information materials. After a discussion, it was decided that similar QIs would be considered as one QI.

Among the 3112 articles screened, 18 articles reporting search and creation processes were included in this scoping review (Fig. 1). Most of the guidelines for each acute cardiovascular disease were not included, as they did not mention how to create QI sets. Table 1 summarizes the characteristics of the included studies. Among the 18 included articles, 12 were on ACS, five on AHF, and two on AAD; one study included QIs for both ACS and AHF [11, 12, 19‐34]. Of all articles, 11 articles were published in North America, and the number of publications differed among different years of publication (Fig. 2). Twelve articles were published after 2010 [11, 12, 19‐25, 30, 33, 34]. Four studies conducted a systematic review [19, 26, 33, 34], and most of the studies used literature reviews that included a search for existing guidelines or statements. Ten studies used the Delphi or modified Delphi method in the QI creation process [11, 19, 22, 24‐26, 28, 29, 32, 34]. Five studies conducted by cardiovascular societies or a government agency used a consensus-making process, based on their original protocols [12, 21, 23, 31]. Furthermore, four studies used expert panel consensus through discussion [21, 27, 31, 33]. We identified 268 QIs in the 18 evaluates articles, including the duplications between them: 191 QIs for ACS, 57 QIs for AHF, and 20 QIs for AAD.

QIs for ACS are listed in Table 2 (QIs mentioned in more than half [≥6] of the articles) and S1 Table (all-QI list). Most of the QIs (85%, n = 45) were process measures, and seven (13%) QIs were structural measures. The most mentioned measures were “time for primary percutaneous coronary intervention (PCI)/timely performed PCI” (n = 9), “beta-blockers prescription for patients with reduced left ventricular (LV) function” (n = 8), and “angiotensin-converting enzyme inhibitor (ACEi) or angiotensin II receptor blocker (ARB) prescription for patients with reduced LV function” (n = 8). Most of the process QIs were mentioned in ≤3 articles, and the process measures varied across studies and clinical settings (upon admission, acute setting, and during hospitalization/at discharge) (Table 3). Recent articles have referred to process measures for patient psychological and social factors, such as patient-reported health status and patient feedback [35]. For outcome measures, mortality or readmission were mentioned in seven articles. Among the 54 QIs, 26 (48%) were mentioned in only one article each. Among the structural indicators, the prehospital electrocardiogram was the most recommended indicator (n = 3).

QIs for AHF are listed in Table 4 (QIs mentioned in more than half [≥3] of the articles) and S2 Table (all-QI list). Most QIs (83%, n = 20) were process measures, and two (8%) were structural measures. The most mentioned measures were “ACEi, ARB, or angiotensin receptor neprilysin inhibitor therapy for patients with heart failure with reduced ejection fraction (HFrEF)” (n = 5) and “beta-blockers prescribed for patients with HFrEF” (n = 4). Outcome measures, such as mortality and readmission, were mentioned in three articles. Among the 24 QIs, 15 (60%) were mentioned in one article each. Among all the process measures (n = 20), 17 (85%) were used during the hospitalization / at discharge, and more than half of QIs were mentioned in only one article (Table 3).

QIs for AAD are listed in Table 5. Two articles mentioned QIs for patients with AAD, and there was little overlap between the QIs listed in these articles [33, 34]. More than half of the QIs (58%, n = 7) were process measures, and four (33%) were structural measures. “Annual operation volume for AAD” was the sole indicator reported in both articles. One article included long-term measures such as follow-up imaging, mortality, and re-intervention. The other article included structural measures such as the designation of the emergency center and the number of surgeons or cardiologists.

This is the first scoping review to systematically review the reporting quality of literature on QIs for acute cardiovascular diseases. This review revealed that few articles conducted a systematic search and provided a search strategy. Most of the articles on ACS and AHF performed a literature review without a search strategy, and performed a review of guidelines and associated literature. Except for one article [26], all of the literature that included systematic reviews was published after 2018. This implies that systematic reviews have been increasingly used to screen candidates for QIs in recent years.

We revealed that most of the studies that used a predefined creation process employed well-established methods for QI creation. In original research conducted by individual researchers, the Delphi or modified Delphi methods were frequently used. Although it is not clear which method is best to use, these systematic methods for decision making were recommended and widely used for QI development in healthcare [36]. In contrast, cardiovascular societies, such as the American Heart Association, and government agencies and the National Institute for Health and Care Excellence, published and used methodology papers to develop QIs. Moreover, these methods were also used for other cardiovascular diseases.

We found that the number of reports varied according to cardiovascular disease. There were more articles on ACS than on AHF and AAD. QIs for ACS have been reported for a longer period, and the list of QIs reported in the guidelines (American Heart Association/American College of Cardiology or European Society of Cardiology) has been updated depending on the care situation [35, 37‐40]. Compared to ACS and AHF, all articles on AAD were published after 2020 [33, 34]. The number of QI sets may differ because the prevalence of ACS and AHF was higher, and QI sets for ACS and AHF have been reported since the 2000s; therefore, there was a large number of relevant studies. Additionally, the literature listed in our study was used for the assessment of the variation of care and showed the association with better outcome in other studies [41‐43]. However, AAD is critical but infrequent, which explains the small number of relevant studies; thus, the validation of QI sets in AAD was limited.

We found that QI measurements for each acute cardiovascular disease varied across articles. Most of the various QI measurements were categorized as process measurements. This finding is consistent with those of previous reviews on non-cardiovascular diseases [44, 45]. The QI measurements in our study were used in a variety of situations (upon admission, acute setting, during hospitalization / at discharge). Moreover, they included various factors, such as patient assessment, medical and surgical treatment, clinical tests, and patient education. In a recent update, novel QI measurements, such as the dual antiplatelet therapy duration and patient Quality of Life, were established as new QIs for ACS. This shows that emphasis is placed on both the acute and chronic care for patients with ACS. In addition, although we summarized outcome measures as a single category according to disease, outcome measures included a wide variety of definitions of the follow-up period and readmission outcomes, including ACS, AHF, and other causes.

Among all QI measurements, the most mentioned were “time for primary PCI/timely performed PCI” for ACS and “ACEi, ARB, or angiotensin receptor neprilysin inhibitor therapy for patients HFrEF” for AHF. These measurements have been reported since 2000, and a consensus was made on their use because most evaluated articles reported the same measurements. Commonly adopted QIs, shown in Table 2 for ACS and in Table 4 for AHF, were mentioned in more than half of the evaluated articles; these QIs are widely recommended in clinical settings.

We also found that there was wide variation in relation to the study setting (year of publication and country), QI construction process, and selection of QIs, among the different articles included in our study. Our scoping review did not focus on exploring the reasons underlying such variations; however, differences in literature searches, treatment strategies, patient backgrounds, and regional characteristics in each clinical setting may play a role in the variation of QIs. In this study, we provided a comprehensive list of QIs for acute cardiovascular diseases and clarified the commonly mentioned QIs for ACS and AHF. These consensus QI lists based on the assessment of the QI development process could be informative in the development of future QIs. Additionally, the use of commonly mentioned QIs may lead to improved outcomes related to the management of cardiovascular diseases.

This study had several limitations. First, we did not evaluate the quality of the QI development methodologies because there are no established tools for such evaluation. However, we assessed a part of the QI development process and whether systematic search with search formulas and established QI creation processes were used. Second, QI sets were created according to the clinical setting, and we did not evaluate the creation process in detail. Finally, this was a scoping review with a synthesis approach, and we did not perform a detailed analysis of each original publication.