Cardiovascular disease (CVD) imposes significant health and economic burdens in Brazil, and the country has one of the highest mortality rates for CVD globally, comparable to China and Eastern Europe.¹ To address this issue, the Best Practice in Cardiology (BPC) program was implemented in Brazil. Collaboratively led by Hospital do Coração (HCor), the Brazilian Society of Cardiology (SBC), the American Heart Association (AHA), and the Ministry of Health, the BPC program aims to improve CVD care by adapting quality improvement programs from Get with the Guidelines (GWTG)™.^2-6

Notably, the BPC program in Brazil is the first instance of simultaneously initiating three quality programs adapted from the AHA outside the USA. Specifically, the program focuses on enhancing care for acute coronary syndrome (ACS), heart failure (HF), and atrial fibrillation (AF). Its objectives include (1) describing the characteristics, in-hospital treatment, and outcomes of patients admitted to public hospitals in Brazil, (2) evaluating the effectiveness of quality improvement programs in enhancing care quality and outcomes, and (3) exploring and optimizing quality improvement strategies within the Brazilian healthcare system.

This article presents the outcomes and findings of the BPC program, aiming to identify additional opportunities for quality improvement and provide guidance for the development of effective strategies and tools to improve CVD outcomes in Brazil.

From March 2016 to November 2022, a total of 12,167 patients with a diagnosis of ACS, HF, or AF were enrolled at 19 institutions located in different regions of Brazil, mostly the northeast and southeast regions. Demographic and clinical data of the patient population, which included 2,503 AF patients, 3,574 HF patients, and 6,090 ACS patients, are presented in Table 2. The median age of the patient population was 62.5 years, and 61.1% were men. There was a high prevalence of comorbidities, including hypertension (68.7%), diabetes mellitus (32%), and dyslipidemia (24.1%) in the patient population.

Table 2

	AF (n=2,503)	HF (n=3,574)	ACS(n=6090)	Total (n=12,167)
Age; median [IQR]	66 [57.3 - 74.2]	61 [51 - 70.8]	61.7 [54.3-69.1]	62.5 [53.8 - 71]
Gender; male (%)	52.5	58.4	66.3	61.1
Race; White (%)	49.7	31.5	38.4	38.7
Race; Brown (%)	36.9	52.2	47.7	46.7
Race; Black (%)	12.0	15.4	12.8	13.4
Race; Asian (%)	0.9	0.8	0.7	0.8
Hypertension (%)	71.5	70.6	82.5	68.7
Diabetes mellitus (%)	24.1	35.1	41.5	32.0
Dyslipidemia (%)	26.3	21.8	30.3	24.1
Previous stroke (%)	14.3	9.8	6.6	8.3
Coronary artery disease (%)	10.5	14	18	13.5
Peripheral artery disease (%)	4.9	3.8	2.5	3.1
Myocardial Infarction (%)	11.7	18.7	20.7	16.2
Permanent Atrial Fibrillation / Atrial flutter (%)	81.9	25.4	3.2	24.8
Rheumatic valvular disease (%)	6.6	6	0.3	3.1
Valvular disease (%)	13.5	15.9	1.1	8.0
Dialysis (%)	0.8	1.8	1.5	1.3
Kidney failure (Cr>2.0)(%)	5.4	14.6	3.6	6.6
Previous angioplasty (%)	5.5	10.9	11.4	10
Previous CABG (%)	4.2	5.7	3.4	4.2
Valvular prosthesis (%)	9.4	9.1	0.6	4.9
Current smoking (%)	5.2	7.6	25.5	15.2
BMI, median [IQR]	25.3 [21.9 - 28.9]	25.7 [22.2 - 29.6]	26.3 [23.7 - 29.3]	25.4 [22.3 - 28.9]
LVEF, median [IQR]	61 [46 - 67]	36 [26 - 53]	54 [42 - 62]	51 [36 - 63]

CABG: coronary artery bypass grafting; BMI: body mass index; LVEF: left ventricular ejection fraction; IQR: interquartile range; AF: atrial fibrillation; HF: heart failure; ACS: acute coronary syndrome.

Figure 1 presents the results for the compound performance measures over 22 quarters of follow-up. To account for the variation in the length of time that participating centers were involved in data collection, an analysis was conducted over a 20-month period during which the highest number of patients was present (Table 3).

Figure 1

Table 3

Atrial fibrillation
Quarter	Mean Value	Difference from baseline	p value
Baseline	52.92 (42.53; 63.3)	-	-
Quarter 2	59.24 (50.18; 68.31)	6.33 (2.93; 9.72)	< 0.001
Quarter 3	63.83 (55.33; 72.32)	10.91 (4.93; 16.89)	< 0.001
Quarter 4	66.66 (58.28; 75.04)	13.74 (5.69; 21.79)	< 0.001
Quarter 5	67.74 (58.86; 76.62)	14.83 (4.76; 24.89)	0.004
Quarter 6	67.08 (56.55; 77.61)	14.16 (1.59; 26.74)	0.027
Heart Failure
Quarter	Mean Value	Difference from baseline	p value
Baseline	79.94 (74.78; 85.1)	-	-
Quarter 2	83.02 (78.43; 87.61)	3.08 (1.09; 5.06)	0.002
Quarter 3	85.24 (80.8; 89.69)	5.3 (2.1; 8.5)	0.001
Quarter 4	86.61 (82.34; 90.88)	6.67 (2.88; 10.46)	< 0.001
Quarter 5	87.12 (83.01; 91.24)	7.18 (3.08; 11.29)	< 0.001
Quarter 6	86.78 (82.15; 91.41)	6.84 (2.02; 11.66)	0.005
Acute Coronary Syndrome
Quarter	Mean Value	Difference from baseline	p value
Baseline	87.72 (83.89; 91.54)	-	-
Quarter 2	87.31 (83.69; 90.94)	-0.4 (-2.72; 1.92)	0.734
Quarter 3	88.16 (84.64; 91.69)	0.45 (-2.58; 3.47)	0.773
Quarter 4	89.19 (85.44; 92.94)	1.48 (-2.27; 5.22)	0.44
Quarter 5	89.34 (84.97; 93.71)	1.62 (-3.08; 6.32)	0.499
Quarter 6	87,53 (82,36; 92,71)	-0,18 (-5,98; 5,61)	0,951

Table 4 presents individual performance measures for HF – angiotensin receptor blocker/angiotensin-converting enzyme inhibitor (ARB/ACEI) use at discharge for patients with left ventricular systolic dysfunction (LVSD) and aldosterone antagonist prescription – which had significant improvement from baseline; for AF – prothrombin time/international normalized ratio (PT/INR) planned follow-up, HAS-BLED score, and statin at discharge), which also had significant improvement; and for ACS - aspirin within 24 hours of admission, beta-blockers and statin at discharge had significant improvement.

Table 4

Performance measures for AF*	Baseline (n=364)	at 6 continuos quarters (n=181)	p
ARB/ACEI at discharge	83.7	80.6	0.973
CHADS-VASC2 score	47.4	43.1	0.461
β-Blockers at discharge	81	91.8	0.101
Anticoagulation	78.4	82.8	0.318
PT/INR Planned Follow-Up	87.1	97.7	0.002
HAS-BLED score	23.6	38.1	0.001
Statin at discharge	64.8	82.1	0.006
Composite score	65.8 ± 36.2	73 ± 31.2	0.024
Performance Measures for HF**	Baseline (n=326)	at 6 continuous quarters (n=224)	p
ARB/ACEI at discharge for patients with LVSD	72.9	87.2	0.008
β-Blockers at discharge for patients with LVSD	86.8	91.9	0.263
LVEF assessment	90.8	94.9	0.128
Postdischarge appointment	82.5	88.8	0.077
Aldosterone Antagosnist	64.2	82.8	0.004
Composite score	81 ± 23.6	89.9 ± 19.3	<0.001
Performance Measures for ACS***	Baseline (n=485)	at 6 continuous quarters (n=394)	p
Aspirin within 24 h of admission	90.2	94.7	0.035
Aspirin at discharge	94.4	96.2	0.281
β-Blockers at discharge	82.1	88.6	0.013
ARB/ACEI at discharge for patients with LVEF < 40%	79.6	83.8	0.546
Statin at discharge	87.3	95.2	<0,001
Blood pressure control at discharge	95.4	96.5	0.682
Adult smoking cessation advice/counseling	92	75.3	0.004
Door-to-needle time	55.6	33.3	1
Door-to-balloon time	64.4	68.3%	0.621
Composite score	88 ± 19.1	91.2 ± 14.9	<0.001

*Proportion of non-valvular AF/Flutter patients with a documented CHADS2-VASc risk score assessment; Proportion of patient with a documented HAS-BLED risk score assessment; Proportion of HF patients with LVEF < 40% or AF patients with LVEF ≤ 40% or ACS patients with LVEF < 45% with an ACEI/ARB prescribed at discharge; Proportion of ACS patients with a beta blocker prescribed at discharge; Proportion of AF patients at high risk for thromboembolism according to the CHADS2_VASc taking anticoagulants; Proportion of AF patients with CAD, stroke/TIA, PVD or diabetes who were prescribed a statin at discharge; Proportion of ACS patients without contraindications with statin prescribed for LDL control at discharge; Proportion of AF patients discharged on Warfarin who had an INR follow up planned prior to hospital discharge. **Proportion of HF patients with a documented LV dysfunction either in the medical records or other reports accessible in hospital charts in the 12 months before admission or during hospitalization or with a scheduled evaluation planned to be performed after discharge; Proportion of HF patients with LVEF < 40% or AF patients with LVEF ≤ 40% or ACS patients with LVEF < 45% with an ACEI/ARB prescribed at discharge; Proportion of ACS patients with a beta blocker prescribed at discharge; Proportion of HF patients with LVEF ≤ 35% taking aldosterone inhibitors; Proportion of HF patients for whom a follow-up appointment was scheduled and documented. ***Proportion of ACS patients receiving aspirin within 24 hours of hospital arrival; Proportion of ACS patients with aspirin prescribed at discharge; Proportion of ACS patients with a beta blocker prescribed at discharge; Proportion of HF patients with LVEF < 40% or AF patients with LVEF ≤ 40% or ACS patients with LVEF < 45% with an ACEI/ARB prescribed at discharge; Proportion of ACS patients without contraindications with statin prescribed for LDL control at discharge; Proportion of ACS patients under medication for blood pressure control; Proportion of ACS patients, who are active smoker within the past 12 months, who receive smoking cessation advice during hospitalization or at discharge; Proportion of STEAMI patients submitted to thrombolysis within 30 min; Proportion of STEAMI patients submitted to primary angioplasty within 90 min from hospital arrival.

Table 5 presents in-hospital mortality, mortality at 180 days, and new hospitalizations.

Table 5

	Atrial fibrillation	Heart failure	Acute coronary syndrome
Hospital mortality
n/N (% [CI95%])	9/135 (6.67%)*	120/1305 (9.2%)	44/1959 (2.25%)
Mortality
Rate per 100 patient-year (95%CI)	28/518 (5.4 [3.4 - 8.4])	146/440 (32.9 [27.0 - 40.1])	70/764 (9.1 [6.8 - 12.1])
Hospitalization
Rate per 100 patient-year (95%CI)	34/508 (6.5 [4.3 - 9.8])	189/389 (42.6 [35.8 - 50.7])	167/715 (21.7 [18.1 - 26.1])
Hospitalization or death
Rate per 100 patient-year (95%CI)	57/508 (10.9 [8.0 - 15.0])	312/389 (70.4 [61.5 - 80.5])	222/715 (28.9 [24.6 - 33.9])

Confidence interval for Rate estimated by Poisson regression with time to event as offset.

Brazil has one of the largest publicly financed patient care systems regarding population coverage, although the quality of assistance provided by the public system is frequently challenged. Cardiovascular in-hospital mortality in Brazil is still high, and well-designed, robust quality improvement programs are desirable and needed. Programs like GWTG have shown to improve healthcare value by identifying critical gaps, promoting quality improvement interventions, measuring the rate and degree of change, and identifying potential for new quality measures based on evolving scientific results.^6-9

Since its beginning, the strategy of the BPC program to 1) generate new knowledge, 2) identify opportunities for improvement, 3) prioritize actions and 4) implement improvements based on evidence has been an asset to the participating institutions. Once identified from the analysis of the indicators, the proposed interventions² are centrally coordinated by the project management group. They include checklists and reminders, webinars, automatic and real-time reports through an electronic database, educational materials, quarterly meetings for audit, feedback and recognition, and training of hospitals in quality improvement methodologies for the implementation of rapid improvement cycles using tools promoted by the Institute for Healthcare Improvement to enable hospitals to develop action plans to achieve the desired improvement.

As a real-world registry study, the BPC program provides comprehensive performance and quality information. During this period, we observed an overall sustained improvement from quarter to quarter in evidence-based care for AF, ACS, and HF.

Selected university hospitals for the BPC program had previous experience in the management of patients with HF. This explains their better baseline performance, with higher adherence to performance measures compared to the BREATHE Registry.¹⁰ Although this represents a favorable scenario for the university programs, there was room for quality improvement as seen in ARB/ACEI at discharge for patients with LVSD and aldosterone antagonist prescription. Interestingly, ARB/ACEI at discharge was remarkably low during the period compared to the GWTG program.^6,7 For the aldosterone receptor blockers prescription, a life-saving medication, there was a marked adherence in its use.

Compared with the PEACE 5r-HF (China Patient-centered evaluative Assessment of Cardiac Events Retrospective Study of Heart Failure) in China¹¹ the BPC program patients were younger, more likely to be female and had higher rates of high blood pressure and diabetes.

At discharge, prescription rates of ACEIs and ARBs were much lower in China (51.5% for either ACEIs or ARBs) compared to the BPC program (87.2%). Among eligible candidates with HF with reduced ejection fraction, prescription rates of beta-blockers were 46.2% in China and 91.9% in Brazil and aldosterone receptor antagonists were 64.2% at discharge in China compared to 82.8% in Brazil.

The evaluation of left ventricular ejection fraction (LVEF) is a fundamental measure of quality for treating patients with HF,¹² and it is noticeable that the BPC program institutions obtain indices similar to those of American and European centers. However, note that the bias of the selected centers may have affected this result.^6,13

The composite baseline score for AF in the BPC program was quite similar to the recent report of the Chinese experience.¹⁴ Hypertension and diabetes were more frequent in our series. As expected, we had fewer cases of newly diagnosed AF as outpatients were also included in the analysis.

There was marked improvement in HAS-BLED score determination and statin prescription. According to guideline recommendations,^15-17high adherence to warfarin therapy would be expected using PT/INR during follow-up for treatment control.

Regarding ACS, our data differed from the Chinese study.¹⁰ Our patients were younger, and with a greater proportion of female. Hypertension, diabetes, dyslipidemia, and previous myocardial infarction were more prevalent in our series. Our results also differ from a previous national registry in Brazil. In our BPC program population, we had more patients with myocardial infarction rather than unstable angina.¹⁸

The prescription rate of aspirin at discharge was 96.2%, which is comparable to different countries like the United Kingdom (98.1%) and Sweden (94.6%).

Also, beta-blockers at discharge from the hospital was 88.6%, also comparable with the United Kingdom (95.6%) and Sweden (88.7%).^19-21

Comparing the BPC results with a published series of GWTG for ACS,⁶ we had similar performance rates, over 90% adherence to the performance measures. One exception was ARB/ACEI rates at discharge for patients with LVEF < 40% in the BPC.

Compared to the CCC-ACS project from China,²² ACS data from the BPC program revealed better performance, in terms of prescription of aspirin, ACEI/ARB, beta-blockers, and statins.

For ACS, we found an increase in the prescription ratesof aspirin within 24 h of admission, beta-blockers and statins at discharge.

The BPC program is a quality improvement program in Brazil in which real-time data, derived from cardiology guidelines, were implemented with an overall sustained improvement in AF, HF, and ACS management.