Effect of Hospital Accreditation Process in Outcomes of Patients with Acute Coronary Syndrome

Mailing Address: Prof. Dr. Gilson Soares Feitosa-Filho Coordenação de Ensino – Hospital Santa Izabel Praça Conselheiro Almeida Couto, 500. Postal Code: 40050-410, Nazaré, Salvador, BA Brazil. E-mail: gilsonfeitosafilho@yahoo.com.br Effect of Hospital Accreditation Process in Outcomes of Patients with Acute Coronary Syndrome Carolina de Deus Leite,1 Thiago Carvalho Pereira,2 Matheus Pamponet Freitas,2 Natália Lima Walsh Tinôco,1 Flávia Guimarães Pereira,1 Roberta Vicente Leite Viana Menezes,1 Milena Quadros Sampaio Andrade,1 Samantha Pereira Rosa Vilas Boas,1 Paulo José Bastos Barbosa,1,2 Gilson Soares Feitosa-Filho1,2 Escola Bahiana de Medicina e Saúde Pública,1 Bahia, BA – Brazil Hospital Santa Izabel da Santa Casa de Misericórdia da Bahia,2 Bahia, BA – Brazil


Introduction
Institutions that offer health services face challenges in improving safety and quality [1][2][3][4] , and it is therefore critical that a global organization be in place for all sectors to work in a systematic way 1,5 . Constant evaluation of this system proves to be useful to ensure its smooth operation 3,6 . Hospital quality certification has then emerged. It is a process of continuing professional education that helps to encourage perfecting through multidisciplinary procedures that improve patient hospitalization and ensure lower rates of in-hospital complications. 3,[6][7][8] Hospital accreditation programs, which are forms of certification in many parts of the world have proven to be a method that assists in the evolution of the quality of health services, besides serving as an external validation of the service [1][2][3]9,10 . These programs analyze many criteria, ranging from hospital infrastructure to teaching and Accreditation in outcomes of patients with ACS Int J Cardiovasc Sci. 2019;32(6):607-614 Original Article research and patient care service 6 . In Brazil, we have the National Accreditation Organization (ONA) that follows the standards established by the Ministry of Health 1 , where the institution is evaluated and receives a classification ranging from level 1 to 3, which represents accreditation with excellence. 3,11 Several studies have analyzed the positive effects of the process and have found that there was a reduction in the length of hospital stay, improved management of preventable outcomes, reduction of hospital mortality, and it helps to create internal protocols. [12][13][14][15] Coronary artery disease (CAD) is the most frequent cause of death in the world, consisting of approximately 13% of all causes of mortality. 16,17 Acute coronary syndrome (ACS) in Brazil represents an important cause of hospitalization and acute myocardial infarction is the second leading cause of death in the country. 18,19 These data demonstrate the importance of adequate management of ACS, especially with regards to the creation of guidelines. 20 The length of stay in both the Intensive Care Unit and at the hospital is an important parameter of quality and better prognosis for the patient, where a decrease of this time is related to reduced hospital costs with the patient and a lower rate of complications, such as lower readmission rates, death or infection (such as mechanical ventilation-related pneumonia, central venous catheter infection or urinary tract infection related to bladder catheter use). 21 In a reference Cardiology hospital in Salvador, Bahia, ACS was considered the main line of care in the ONA 3 accreditation process, held in December 2015.
The primary outcome of this study was the comparison between length of stay at the Coronary Care Unit and hospital stay of patients with ACS before and after ONA 3 accreditation. Secondly, the impact of accreditation on clinical outcomes was analyzed.

Study design and population
This is a prospective observational registry that consecutively included patients diagnosed with Acute Coronary Syndrome hospitalized at the Coronary Care Unit (CCU) of Hospital Santa Izabel (HSI) -Salvador/ BA from February 2015 to August 2016. Santa Casa de Misericórdia da Bahia -HSI is a tertiary philanthropic hospital that underwent ONA 3 accreditation (excellence) in December 2015 and successfully achieved this goal. The study population was divided into two groups: Period 1 (Pre-accreditation: before December 1, 2015) and Period 2 (Post-accreditation: as of December 1, 2015).
All patients with diagnostic confirmation of acute coronary syndrome (including unstable angina, STsegment elevation acute myocardial infarction -STEMI -and non-ST-segment elevation acute myocardial infarction -non-STEMI) met the inclusion criteria, as well as patients receiving drug and/or interventional treatment for acute coronary syndrome at Hospital Santa Izabel. Patients readmitted after elective interventional treatment of previous acute coronary syndrome were excluded.
The data were collected prospectively through a structured electronic medical record of the coronary care unit by a team specifically involved in the collection, completed by the unit's attending physician, always with the possibility of rediscussing some topic with the physician. In summary, the variables include socio-demographic and clinical aspects, in addition to admission, evolution and outcomes.

Statistical analysis
A descriptive analysis of the frequencies of variables was performed using IBM SPSS Statistics 14.0.
Initially, the Kolmogorov-Smirnov test of normality was performed to find out whether the variables had a normal or non-normal distribution pattern and, from this, the statistical tests and the most adequate description of each variable were defined.
The variables were described using measures of central tendency (mean, median) and dispersion (standard deviation, interquartile range) when continuous and by absolute and relative frequencies, for the categorical variables.
Comparisons of clinical outcomes between the preand post-accreditation periods were performed using the chi-square test (X²) or Fisher's exact test when the sample number was < 5. The Mann-Whitney test was used to compare the time of hospitalization at the coronary care unit and hospital admission between the two periods, since non-parametric variables were involved.
In addition, multivariate analysis was performed by multiple linear regression to find out where the hospital accreditation process was an independent predictor of length of hospital stay and length of stay at the CCU, controlling for relevant confounding factors in the The differences observed in all statistical tests (nonparametric and linear regression) were considered significant when the probability (p) of type I error was < 0.05.

Results
A total of 372 patients were included in the study, of which 186 in period 1 (pre-accreditation) and 186 in period 2 (post-accreditation). Of the total, 117 (31.5%) were patients diagnosed with ACS with ST-segment elevation and 255 (68.5%) were diagnosed as non-STsegment elevation ACS.
Regarding the length of hospital stay, in the total period of the study, the median CCU length of stay was 3 (IQR = 2 -4) and the hospital length of stay was 7 days (IQR = 5 -11.75). In period 1, the median was 3 days (IQR = 2 -4) for the CCU length of stay and 8 days for the hospital length of stay (IQR = 5 -12.25). In period 2, the median was 2.5 days for the CCU length of stay (IQR = 2 -4) and 6 days for hospital length of stay (IQR = 4 -11).
By analyzing the median variation between the CCU length of stay and the hospital length of stay between the pre-and post-accreditation periods, it was found that the reduced hospital length of stay in the general sample was statistically significant at p = 0.004. (Table 2) In the analysis of subgroups, the median remained with a tendency of reduction, but only the decrease in hospital length of stay of non-STEMI was relevant, with p = 0.001. (Table 2) Regarding the secondary outcomes, it was found that the type of clinical outcome most commonly presented in the sample was cardiorespiratory arrest (CRA) of any type, evolving to death or not (29 patients -7.8%), followed by death (26 patients -7%). Comparing the two periods, period 1 had a higher number of deaths than period 2 (14 and 12, respectively), but this data did not reach statistical relevance. (Table 3) By cross-comparing the data, it was found that mortality and cardiogenic shock were variables that showed a decrease in the number of cases between the pre-and post-accreditation period, but this data did not reach any statistical significance. Reinfarction, CRA (resulting in death or not) and combined outcomes showed an increase in the absolute number of cases in the comparison between the two periods analyzed, but this difference did not present a significant p-value. (Table 3) By analyzing the clinical outcomes correlated to the types of ACS, it was found that some outcomes increased and others decreased in frequency in the comparison between the pre-and post-accreditation periods, but this change is not statistically relevant. (Table 3) In the multivariate analysis by multiple linear regression, controlling for the variables of age, sex, systemic arterial hypertension, diabetes mellitus, dyslipidemia, previous acute myocardial infarction, previous coronary artery bypass grafting and type of ACS, the post-accreditation period was an independent predictor of reduced time of hospitalization (p = 0.041; B = 2.081; β = 0.105).
By doing the same analysis for the hospitalization time at the coronary care unit, we found that accreditation was not an independent predictor of this change in length of stay (p = 0.834 B = 0.086; β = 0.011).

Discussion
The accreditation process has a positive impact on the standardization of care offered to patients, generating a flow that results in faster and more effective practices, contributing to a better patient prognosis. 22 At Hospital Santa Izabel, where this study was conducted, the hospital accreditation process resulted in better health care processes and had a strong impact on the pursuit of patient safety. With regard to acute coronary syndromes (ACS), its line of care was devised by conducting analyses before the patient arrived at the hospital until their follow-up after discharge. Mortality and bleeding outcomes were established as indicators of the line of care, and these outcomes were adjusted by the GRACE score and the CRUSADE score obtained on admission to the coronary care unit. A set of measures were planned and implemented at the different phases of the line of care, such as taking joint actions with the Municipal Health Department and SAMU (Mobile Emergency Care Service), aiming at improving the time to the implementation of reperfusion in ST-segment elevation acute myocardial infarction. Another relevant aspect was the construction of a therapeutic plan for In this study, regarding the demographic characteristics of the sample, a higher mean age was found, with a difference of about 10 years, compared with the studies of Eagle et al. 15 and Chen et al. 23 which analyzed the impact of the evaluation of hospital services based on the standardization of these services. Moreover, this study has found a lower incidence of ACS in women and a higher prevalence of SAH compared to the results of the two studies mentioned above. 15,22 There was also a higher prevalence of diabetes mellitus and previous angioplasty; approximately the same prevalence of previous acute myocardial infarction; and a lower prevalence of patients with heart failure compared to patients from the study of Chen et al. 22 It can be assumed that, because the population sample of Salvador has more comorbidities than the population analyzed by Eagle et al. 15 and Chen et al. al. 22 , the patients in the sample may be associated with earlier infarction, since a considerable difference was observed between the mean ages of the patients in this study and those of the other authors cited.
The hospitalization times at the Coronary Care Unit and at the hospital were analyzed and a tendency of reduced medians has been found. As for the total sample and that of patients with non-ST-segment elevation ACS, there was a significant reduction in the length of Original Article hospital stay. Patients with ST-segment elevation also had a reduced length of hospital stay, but this data did not present any statistical significance.
In a multicenter study conducted by Sack et al. 24 in 73 hospitals, the quality of care offered and the patient's satisfaction with their hospitalization in accredited and non-accredited hospitals were analyzed. 24 The population considered was more comprehensive, excluding only obstetric patients and pediatric patients. As a result, it was found that the median length of hospital stay in accredited hospitals tended to be lower, but with no statistical relevance. 24 In the sample of this study, the population was more specific, and the reduction in hospital length of stay may be associated with a better systematization of care for these patients, which was confirmed in this study through a multivariate linear regression. Furthermore, it can be assumed that, with improved care, with protocols established and fulfilled, the patient presents more favorable conditions for early discharge. Also, regarding the length of stay, a study conducted by Falstie-Jensen et al. 9 also found that patients hospitalized in accredited hospitals had a shorter hospital stay, 9 which is consistent with this study.
Regarding the outcomes analyzed, in the current study, it can be seen that the tendency related to cardiogenic shock was a reduction in the sample studied and an increase in patients with STEMI. Regarding mortality, there was a decrease in this variable both in the sample as a whole and in patients with non-STsegment elevation, but without statistical significance. In the analysis of subgroups, there was an increase in the number of deaths in patients with ST-segment elevation. Compared to the Eagle 15 study, it can be seen that the patients analyzed in the United States showed an increase in the number of cases of cardiogenic shock and that mortality decreased significantly. 15 One possibility to be raised to increase the number of deaths of patients with STEMI is that there were more admissions of more severe patients in the second period analyzed. According to Greenfield et al. 14 by undergoing a quality assessment such as the hospital accreditation process, the institution tends to receive more patients with more serious disorders 14 , possibly due to the recognition of the effectiveness of the service offered.
Another issue to be emphasized is that the results on mortality obtained in this analysis should be interpreted carefully. In a study conducted by Williams et al. 25 in 2005, a dissociation was found between the variable in-hospital mortality of patients with acute myocardial infarction and the other variables analyzed. These other variables were more associated with the quality of the service offered to the patient, and an improvement was perceived after the hospital evaluation process. The authors pointed out that previous studies reported a lower sensitivity of the clinical outcomes to the detriment of quality parameters with regard to the protocols established at the hospital. 25 In-hospital mortality refers to the management of a specific patient and does not necessarily shows the ineffective outcome of all other procedures of care provided during hospitalization. 25 Due to this, the number of deaths is not considered a good parameter in assessing the impact of the accreditation process.
Regarding the other variables of clinical outcomes, there was an increase in reinfarction and CRA, especially in patients with ST-segment elevation, but without significance. In 2015, Falstie-Jensen et al. 9 studied the relationship between accredited hospitals and acute readmission (up to 30 days), considering all patients admitted to the hospital, and it was found that patients seen in institutions certified as accredited institutions did not present any difference comparing with nonaccredited hospitals. 9 Given that reinfarction can be considered a factor that would lead to acute readmission, this cause may be included in the context of the study.

Limitations
This study has some limitations. Firstly, it includes data from a single hospital, with a relatively small sample. In addition, it is not possible to evaluate the secondary outcomes satisfactorily, since the sample size is not so big. Moreover, it was impossible to have a control group in parallel to the study, since it was conducted in two distinct periods, and the motivation of the team may imply different results. However, note that the motivation of the team is one of the benefits of the accreditation process. Another limitation was that the data of this study were secondary and were derived from medical records, although the researchers made sure they conducted an active search for any information that might be missing or doubtful.

Conclusions
In conclusion, after the ONA 3 accreditation process, there was a reduction in hospital stay. There were no significant differences in the frequency of hospital mortality or combined clinical outcomes, as well as in the length of hospital stay at the CCU.