Survival analyses in cardiovascular research, part II: statistical methods in challenging situations

Adequacy of Dialysis and Incidence of Atrial Fibrillation in Patients Undergoing Hemodialysis

BACKGROUND

Atrial fibrillation (AF) can lead to stroke, heart failure, and mortality and has a greater prevalence in dialysis patients than in the general population. Several studies have suggested that uremic toxins may contribute to the development of AF. However, the association between dialysis adequacy and incident AF has not been well established.

METHODS

In this retrospective nationwide cohort study, we analyzed data from the Korean National Periodic Hemodialysis Quality Assessment from 2013 to 2015 of patients who received outpatient maintenance hemodialysis 3× a week. The main exposure was single pooled Kt/V (spKt/V), which is the dialysis adequacy index, and the primary outcome was the development of AF. For the primary analysis, patients were categorized into quartiles according to baseline spKt/V. The lowest quartile, representing the lowest adequacy, was used as the reference group. Fine-Gray subdistribution hazard models were used, treating all-cause mortality as a competing risk.

RESULTS

Of 25 173 patients, the mean age was 60 (51-69) years, and 14 772 (58.7%) were men. During a median follow-up of 5.7 years, incident AF occurred in a total of 3883 (15.4%) patients. Participants with a higher spKt/V tended to have lower AF incidence. In survival analysis, a graded association was observed between the risk of incident AF and spKt/V quartiles: subdistribution hazard ratios and 95% CIs for the second, third, and the highest quartile compared with the lowest quartile were 0.90 (95% CI, 0.82-0.98), 0.84 (95% CI, 0.77-0.93), and 0.79 (95% CI, 0.72-0.88), respectively.

CONCLUSIONS

This nationwide cohort study showed that a higher spKt/V is associated with a reduced risk of incident AF. These findings suggests that reducing uremic toxin burden through enhanced dialysis clearance may be associated with a lower risk of AF development in patients undergoing maintenance hemodialysis.

Association of Use and Dose of Lipid-Lowering Therapy Post Acute Myocardial Infarction With 5-Year Survival in Older Adults

BACKGROUND

Older people are underrepresented in randomized trials. The association between lipid-lowering therapy (LLT) and its intensity after acute myocardial infarction and long-term mortality in this population deserves to be assessed.

METHODS

The FAST-MI (French Registry of Acute ST-Elevation or Non-ST-Elevation Myocardial Infarction) program consists of nationwide French surveys including all patients admitted for acute myocardial infarction ≤48 hours from onset over a 1- to 2-month period in 2005, 2010, and 2015, with long-term follow-up. Numerous data were collected and a centralized 10-year follow-up was organized. The present analysis focused on the association between prescription of LLT (atorvastatin ≥40 mg or equivalent, or any combination of statin and ezetimibe) and 5-year mortality in patients aged ≥80 years discharged alive. Cox multivariable analysis and propensity score matching were used to adjust for baseline differences.

RESULTS

Among the 2258 patients aged ≥80 years (mean age, 85±4 years; 51% women; 39% ST-segment elevation myocardial infarction; 58% with percutaneous coronary intervention), 415 were discharged without LLT (18%), 866 with conventional doses (38%), and 977 with high-dose LLT (43%). Five-year survival was 36%, 47.5%, and 58%, respectively. Compared with patients without LLT, high-dose LLT was significantly associated with lower 5-year mortality (adjusted hazard ratio, 0.78 [95% CI, 0.66-0.92]), whereas conventional-intensity LLT was not (adjusted hazard ratio, 0.93 [95% CI, 0.80-1.09]). In propensity score-matched cohorts (n=278 receiving high-intensity LLT and n=278 receiving no statins), 5-year survival was 52% with high-intensity LLT at discharge and 42% without statins (hazard ratio, 0.78 [95% CI, 0.62-0.98]).

CONCLUSIONS

In these observational cohorts, high-intensity LLT at discharge after acute myocardial infarction was associated with reduced all-cause mortality at 5 years in an older adult population. These results suggest that high-intensity LLT should not be denied to patients on the basis of old age.

REGISTRATION

URL: https://www.clinicaltrials.gov; Unique identifiers: NCT00673036, NCT01237418, and NCT02566200.

The different risk of new-onset, chronic, worsening, and advanced heart failure: A systematic review and meta-regression analysis

AIMS

Heart failure (HF) is a chronic and progressive syndrome associated with a poor prognosis. While it may seem intuitive that the risk of adverse outcomes varies across the different stages of HF, an overview of these risks is lacking. This study aims to determine the risk of all-cause mortality and HF hospitalizations associated with new-onset HF, chronic HF (CHF), worsening HF (WHF), and advanced HF.

METHODS AND RESULTS

We performed a systematic review of observational studies from 2012 to 2022 using five different databases. The primary outcomes were 30-day and 1-year all-cause mortality, as well as 1-year HF hospitalization. Studies were pooled using random effects meta-analysis, and mixed-effects meta-regression was used to compare the different HF groups. Among the 15 759 studies screened, 66 were included representing 862 046 HF patients. Pooled 30-day mortality rates did not reveal a significant distinction between hospital-admitted patients, with rates of 10.13% for new-onset HF and 8.11% for WHF (p = 0.10). However, the 1-year mortality risk differed and increased stepwise from CHF to advanced HF, with a rate of 8.47% (95% confidence interval [CI] 7.24-9.89) for CHF, 21.15% (95% CI 17.78-24.95) for new-onset HF, 26.84% (95% CI 23.74-30.19) for WHF, and 29.74% (95% CI 24.15-36.10) for advanced HF. Readmission rates for HF at 1 year followed a similar trend.

CONCLUSIONS

Our meta-analysis of observational studies confirms the different risk for adverse outcomes across the distinct HF stages. Moreover, it emphasizes the negative prognostic value of WHF as the first progressive stage from CHF towards advanced HF.

Sacubitril/valsartan compared to ramipril in high-risk post-myocardial infarction patients stratified according to use of mineralocorticoid receptor antagonists: Insight from the PARADISE MI trial

AIM

It is unknown whether safety and clinical endpoints by use of sacubitril/valsartan (an angiotensin receptor-neprilysin inhibitor [ARNI]) are affected by mineralocorticoid receptor antagonists (MRA) in high-risk myocardial infarction (MI) patients. The aim of this study was to examine whether MRA modifies safety and clinical endpoints by use of sacubitril/valsartan in patients with a MI and left ventricular systolic dysfunction (LVSD) and/or pulmonary congestion.

METHODS AND RESULTS

Patients (n = 5661) included in the PARADISE MI trial (Prospective ARNI vs. ACE Inhibitor Trial to Determine Superiority in Reducing Heart Failure Events After MI) were stratified according to MRA. Primary outcomes in this substudy were worsening heart failure or cardiovascular death. Safety was defined as symptomatic hypotension, hyperkalaemia >5.5 mmol/L, or permanent drug discontinuation. A total of 2338 patients (41%) were treated with MRA. Safety of ARNI compared to ramipril was not altered significantly by ± MRA, and both groups had similar increase in symptomatic hypotension with ARNI. In patients taking MRA, the risk of hyperkalaemia or permanent drug discontinuation was not significantly altered by ARNI (p > 0.05 for all comparisons). The effect of ARNI compared with ramipril was similar in those who were and were not taking MRA (hazard ratio [HR]MRA 0.96, 95% confidence interval [CI] 0.77-1.19 and HRMRA- 0.87, 95% CI 0.71-1.05, for the primary endpoint; p = 0.51 for interaction [Clinical Endpoint Committee adjudicated]); similar findings were observed if investigator-reported endpoints were evaluated (p = 0.61 for interaction).

CONCLUSIONS

Use of a MRA did not modify safety or clinical endpoints related to initiation of ARNI compared to ramipril in the post-MI setting in patients with LVSD and/or congestion.

Daily Physical Activity Measured by Wearable Smartwatch for Patients Undergoing Transcatheter Aortic Valve Replacement: Insights From the SMART TAVR Study

BACKGROUND

This study aims to evaluate limited data about daily physical activity patterns, influential factors, and their association with 1-year mortality or rehospitalization after transcatheter aortic valve replacement (TAVR) through smartwatches.

METHODS

Consecutive severe aortic stenosis patients undergoing elective transfemoral TAVR in a Chinese tertiary hospital were enrolled from July 2021 to May 2022 and received a Huawei smartwatch at least 1 day before TAVR. The primary outcome was a composite of all-cause mortality or hospital readmission within 1 year. Linear mixed-effects models were applied to determine influential factors of daily step counts, and Cox proportional hazard regression models were to estimate the association between baseline step counts within 1 month since discharge and composite outcome from months 2 to 12. The dose-response association was assessed using restricted cubic spline curves.

RESULTS

A total of 222 participants and 59 469 valid monitoring person-day records were included (mean age, 72.7 years; 61% women). Step counts increased rapidly within the first 2 months (P<0.001), followed by a slower increase for those without composite outcomes (P=0.029) and a gradual decrease for those who developed composite outcomes (P<0.001). In multivariate linear mixed models, a 1-m increase in baseline 6-minute walk test and a 1-month delay after discharge were associated with 4 (95% CI, 1-7) and 170 (95% CI, 145-194) additional step counts, respectively. In restricted cubic spline analysis, the hazard ratio declined progressively until ≈5000 steps per day, after which they leveled. Below 5000 steps, the adjusted hazard ratio of composite outcome associated with each 1000-step count increase was 0.67 (0.50-0.89; P=0.007). However, above 5000 steps, step counts were not significantly associated with the composite outcome (P=0.645), with a hazard ratio of 1.12 (0.70-1.79).

CONCLUSIONS

Daily step counts rapidly increased within the first 2 months post-TAVR. Increased physical activity was associated with a lower risk of 1-year mortality or rehospitalization after TAVR for patients with daily step counts below 5000.

REGISTRATION

URL: https://www.clinicaltrials.gov; Unique identifier: NCT04454177.

Intensive Versus Traditional Cardiac Rehabilitation: Mortality and Cardiovascular Outcomes in a 2016-2020 Retrospective Medicare Cohort

BACKGROUND

Traditional cardiac rehabilitation (CR) improves cardiovascular outcomes and reduces mortality, but less is known about the relative benefit of intensive CR (ICR) which incorporates greater lifestyle education through 72 sessions (versus 36 in CR). Our objective was to determine whether ICR is associated with a mortality and cardiovascular benefit compared with CR.

METHODS

Retrospective cohort study of Medicare Fee-For-Service beneficiaries in a 100% sample, claims data set. Qualifying events were captured from May 1, 2016 to December 31, 2019 and ICR/CR utilization captured from May 1, 2016 to December 31, 2020. Among patients attending at least 1 day of either CR or ICR, Cox proportional hazards models using a 1 to 5 propensity score match were used to compare utilization and the association of ICR versus CR participation with (1) all-cause mortality and (2) cardiovascular-related hospitalizations or nonfatal cardiac events. Dose-response was assessed by the number of days attended.

RESULTS

From 2016 to 2019, 1 277 358 unique patients met at least one qualifying indication for ICR/CR from 2016 to 2019. Of these, 262 579 (20.6%) and 4452 (0.4%) attended at least one session of CR or ICR, respectively (mean [SD] age, 73.2 [7.8] years; 32.3% female). In the matched sample, including 26 659 total patients (median, 2.4-year follow-up), ICR was associated with 12% lower all-cause mortality (multivariable adjusted hazard ratio, 0.88 [95% CI, 0.78-0.99]; P=0.036) compared with CR but no significant difference for cardiovascular-related hospitalization or nonfatal cardiac events. The mortality benefit was seen for both ICR and CR per day strata, with each modality demonstrating a clear dose-response benefit.

CONCLUSIONS

ICR is associated with lower mortality than traditional CR among Medicare beneficiaries but no difference in cardiovascular-related hospitalization or nonfatal cardiac events. Moreover, ICR and CR demonstrate a dose-response relationship for mortality. Additional studies are needed to confirm these observations and to better understand the mechanisms by which ICR may lead to a reduction in mortality.

Relationship Between Community-Level Distress and Cardiac Rehabilitation Participation, Facility Access, and Clinical Outcomes After Inpatient Coronary Revascularization

BACKGROUND

Although disparities in cardiac rehabilitation (CR) participation are well documented, the role of community-level distress is poorly understood. This study evaluated the relationship between community-level distress and CR participation, access to CR facilities, and clinical outcomes.

METHODS

A retrospective cohort study was conducted on a 100% sample of Medicare beneficiaries undergoing inpatient coronary revascularization between July 2016 and December 2018. Community-level distress was defined using the Distressed Community Index quintile at the beneficiary zip code level, with the first and fifth quintiles representing prosperous and distressed communities, respectively. Outpatient claims were used to identify any CR use within 1 year of discharge. Beneficiary and CR facility zip codes were used to describe access to CR facilities. Adjusted logistic regression models evaluated the association between Distressed Community Index quintiles, CR use, and clinical outcomes, including one-year mortality, all-cause hospitalization, and acute myocardial infarction hospitalization.

RESULTS

A total of 414 730 beneficiaries were identified, with 96 929 (23.4%) located in the first and 67 900 (16.4%) in the fifth quintiles, respectively. Any CR use was lower for beneficiaries in distressed compared with prosperous communities (26.0% versus 46.1%, P<0.001), which was significant after multivariable adjustment (odds ratio, 0.41 [95% CI, 0.40-0.42]). A total of 98 458 (23.7%) beneficiaries had a CR facility within their zip code, which increased from 16.3% in prosperous communities to 26.6% in distressed communities. Any CR use was associated with absolute reductions in mortality (-6.8% [95% CI, -7.0% to -6.7%]), all-cause hospitalization (-5.9% [95% CI, -6.3% to -5.6%]), and acute myocardial infarction hospitalization (-1.3% [95% CI, -1.5% to -1.1%]), which were similar across each Distressed Community Index quintiles.

CONCLUSIONS

Although community-level distress was associated with lower CR participation, the clinical benefits were universally received. Addressing barriers to CR in distressed communities should be considered a significant priority to improve survival after coronary revascularization and reduce disparities.

Association Between Electrocardiographic Age and Cardiovascular Events in Community Settings: The Framingham Heart Study

BACKGROUND

Deep neural networks have been used to estimate age from ECGs, the electrocardiographic age (ECG-age), which predicts adverse outcomes. However, this prediction ability has been restricted to clinical settings or relatively short periods. We hypothesized that ECG-age is associated with death and cardiovascular outcomes in the long-standing community-based FHS (Framingham Heart Study).

METHODS

We tested the association of ECG-age with chronological age in the FHS cohorts in ECGs from 1986 to 2021. We calculated the gap between chronological and ECG-age (Δage) and classified individuals as having normal, accelerated, or decelerated aging, if Δage was within, higher, or lower than the mean absolute error of the model, respectively. We assessed the associations of Δage, accelerated and decelerated aging with death or cardiovascular outcomes (atrial fibrillation, myocardial infarction, and heart failure) using Cox proportional hazards models adjusted for age, sex, and clinical factors.

RESULTS

The study population included 9877 FHS participants (mean age, 55±13 years; 54.9% women) with 34 948 ECGs. ECG-age was correlated to chronological age (r=0.81; mean absolute error, 9±7 years). After 17±8 years of follow-up, every 10-year increase of Δage was associated with 18% increase in all-cause mortality (hazard ratio [HR], 1.18 [95% CI, 1.12-1.23]), 23% increase in atrial fibrillation risk (HR, 1.23 [95% CI, 1.17-1.29]), 14% increase in myocardial infarction risk (HR, 1.14 [95% CI, 1.05-1.23]), and 40% increase in heart failure risk (HR, 1.40 [95% CI, 1.30-1.52]), in multivariable models. In addition, accelerated aging was associated with a 28% increase in all-cause mortality (HR, 1.28 [95% CI, 1.14-1.45]), whereas decelerated aging was associated with a 16% decrease (HR, 0.84 [95% CI, 0.74-0.95]).

CONCLUSIONS

ECG-age was highly correlated with chronological age in FHS. The difference between ECG-age and chronological age was associated with death, myocardial infarction, atrial fibrillation, and heart failure. Given the wide availability and low cost of ECG, ECG-age could be a scalable biomarker of cardiovascular risk.

Implications of Social Disadvantage Score in Cardiovascular Outcomes and Risk Assessment: Findings From the Multi-Ethnic Study of Atherosclerosis

BACKGROUND

Social determinants of health contribute to disparate cardiovascular outcomes, yet they have not been operationalized into the current paradigm of cardiovascular risk assessment.

METHODS

Data from the Multi-Ethnic Study of Atherosclerosis, which includes participants from 6 US field centers, were used to create an index of baseline Social Disadvantage Score (SDS) to explore its association with incident atherosclerotic cardiovascular disease (ASCVD) and all-cause mortality and impact on ASCVD risk prediction. SDS, which ranges from 0 to 4, was calculated by tallying the following social factors: (1) household income less than the federal poverty level; (2) educational attainment less than a high school diploma; (3) single-living status; and (4) experience of lifetime discrimination. Cox models were used to examine the association between SDS and each outcome with adjustment for traditional cardiovascular risk factors. Changes in the discrimination and reclassification of ASCVD risk by incorporating SDS into the pooled cohort equations were examined.

RESULTS

A total of 6434 participants (mean age, 61.9±10.2 years; female 52.8%; non-white 60.9%) had available SDS 1733 (26.9%) with SDS 0; 2614 (40.6%) with SDS 1; 1515 (23.5%) with SDS 2; and 572 (8.9%) with SDS ≥3. In total, 775 incident ASCVD events and 1573 deaths were observed over a median follow-up of 17.0 years. Increasing SDS was significantly associated with incident ASCVD and all-cause mortality after adjusting for traditional risk factors (ASCVD: per unit increase in SDS hazard ratio, 1.15 [95% CI, 1.07-1.24]; mortality: per unit increase in SDS hazard ratio, 1.13 [95% CI, 1.08-1.19]). Adding SDS to pooled cohort equations components in a Cox model for 10-year ASCVD risk prediction did not significantly improve discrimination (P=0.208) or reclassification (P=0.112).

CONCLUSIONS

Although SDS is independently associated with incident ASCVD and all-cause mortality, it does not improve 10-year ASCVD risk prediction beyond pooled cohort equations.

Patient-Reported Versus Physician-Assessed Health Status in Heart Failure With Reduced and Preserved Ejection Fraction From ASIAN-HF Registry

BACKGROUND

We aimed to assess if discordance between patient-reported Kansas City Cardiomyopathy Questionnaire (KCCQ)-overall summary (os) score and physician-assessed New York Heart Association (NYHA) class is common among patients with heart failure (HF) with reduced or preserved ejection fraction, and determine its association with outcomes.

METHODS

A total of 4818 patients with HF were classified according to KCCQ-os score (range 0-100, dichotomized by median value 71.9 into high [good] versus low [bad]) and NYHA class (I/II [good] or III/IV [bad]) as concordant good (low NYHA class, high KCCQ-os score), concordant bad (high NYHA class, low KCCQ-os score), discordant worse NYHA class (high NYHA class, high KCCQ-os score), and discordant worse KCCQ-os score (low NYHA class, low-KCCQ-os score). The composite of HF hospitalization or death at 1 year was compared across groups.

RESULTS

There were 2070 (43.0%) concordant good, 1099 (22.8%) concordant bad, 331 (6.9%) discordant worse NYHA class, and 1318 (27.4%) discordant worse KCCQ-os score patients. Compared with concordant good, adverse outcomes were the highest in concordant bad (HR, 2.7 [95% CI, 2.2-3.5]) followed by discordant worse KCCQ-os score (HR, 1.8 [95% CI, 1.4-2.2]) and discordant worse NYHA class (HR, 1.5 [95% CI, 1.0-2.3]); with no modification by HF phenotype (preserved versus reduced ejection fraction, P_interaction=0.52). At 6 months, 1403 (48%) experienced clinically significant improvement in KCCQ-os score (≥5 points increase over 6 months). Patients with improved KCCQ-os at 6 months (HR, 0.65 [95% CI, 0.47-0.92]) had better outcomes and the association was not modified by HF phenotype (P_interaction=0.40).

CONCLUSIONS

One-third of patients with HF had discordance between patient-reported and clinician-assessed health status, largely attributable to worse patient-reported outcomes. Such discordance, particularly in those with discordantly worse KCCQ, should alert physicians to an increased risk of HF hospitalization and death, and prompt further assessment for potential drivers of worse patient-reported outcomes relative to physicians' assessment.

Comparative Effectiveness and Safety of Direct Oral Anticoagulants Versus Warfarin Among Adults With Cancer and Atrial Fibrillation

BACKGROUND

While clinical guidelines recommend direct-acting oral anticoagulants (DOAC) over warfarin to treat isolated nonvalvular atrial fibrillation, guidelines are silent regarding nonvalvular atrial fibrillation treatment among individuals with cancer, reflecting the paucity of evidence in this setting. We quantified relative risk of ischemic stroke or systemic embolism and major bleeding (primary outcomes), and all-cause and cardiovascular death (secondary outcomes) among older individuals with cancer and nonvalvular atrial fibrillation comparing DOACs and warfarin.

METHODS

This retrospective cohort study used Surveillance, Epidemiology, and End Results cancer registry and linked US Medicare data from 2010 through 2016, and included individuals diagnosed with cancer and nonvalvular atrial fibrillation who newly initiated DOAC or warfarin. We used inverse probability of treatment weighting to control confounding. We used competing risk regression for primary outcomes and cardiovascular death, and Cox proportional hazard regression for all-cause death.

RESULTS

Among 7675 individuals included in the cohort, 4244 (55.3%) received DOACs and 3431 (44.7%) warfarin. In the inverse probability of treatment weighting analysis, there was no statistically significant difference among DOAC and warfarin users in the risk of ischemic stroke or systemic embolism (1.24 versus 1.19 events per 100 person-years, adjusted hazard ratio 1.41 [95% CI, 0.92-2.14]), major bleeding (3.08 versus 4.49 events per 100 person-years, adjusted hazard ratio 0.90 [95% CI, 0.70-1.17]), and cardiovascular death (1.88 versus 3.14 per 100 person-years, adjusted hazard ratio 0.82 [95% CI, 0.59-0.1.13]). DOAC users had significantly lower risk of all-cause death (7.09 versus 13.3 per 100 person-years, adjusted hazard ratio 0.81 [95% CI, 0.69-0.94]) compared to warfarin users.

CONCLUSIONS

Older adults with cancer and atrial fibrillation exposed to DOACs had similar risks of stroke and systemic embolism and major bleeding as those exposed to warfarin. Relative to warfarin, DOAC use was associated with a similar risk of cardiovascular death and a lower risk of all-cause death.

Short-term/long-term prognosis with or without beta-blockers in patients without heart failure and with preserved ejection fraction after acute myocardial infarction: a multicenter retrospective cohort study

Background

The role of beta-blockers in acute myocardial infarction patients without heart failure and with preserved left ventricular ejection fraction (LVEF ≥ 50%) is unknown. Our study aimed to retrospectively analyze the associations of beta-blockers on such patients.

Methods

This is a multicenter, retrospective study. After screening 5,332 acute myocardial infarction patients, a total of 2519 patients without heart failure and with LVEF ≥ 50% were included. The patients were divided into two groups: the prescribed (n = 2049) and unprescribed (n = 470) beta-blockers group. The propensity score inverse probability treatment weighting was used to control confounding factors. We analyzed the associations between beta-blockers and outcomes in the short-term (1-year) and long-term (median, 3.61 years).

Results

The primary outcome was all-cause mortality. The secondary outcomes were all-cause rehospitalization, cardiac death, recurrent myocardial infarction, new-onset heart failure rehospitalization. This study shows no statistically significant association between discharged with beta-blockers and all-cause mortality, either in the short-term [IPTW Adjusted, HR 1.02; 95%CI 0.43–2.40; P = 0.966] or long-term [IPTW Adjusted, HR 1.17; 95%CI 0.70–1.94; P = 0.547]. Discharged with beta-blockers was significantly associated with a reduced risk of short-term recurrent myocardial infarction [IPTW Adjusted, HR 0.44; 95%CI 0.20–0.97; P = 0.043], but there was no long-term relationship [IPTW Adjusted, HR 1.11; 95%CI 0.61–2.03; P = 0.735]. Other outcomes, such as new-onset heart failure rehospitalization and all-cause rehospitalization, were not observed with meaningful differences in either the short- or long-term. The results of sensitivity analysis were consistent with this.

Conclusions

Beta-blockers might be associated with a reduced risk of recurrent myocardial infarction in patients without heart failure and with preserved left ventricular ejection fraction after acute myocardial infarction, in the short term. Beta-blockers might not be related to all-cause mortality in those patients, either in the short-term or long-term.

Clinical trial registration Influence of Beta-blockers on Prognosis in Patients with Acute Myocardial Infarction Complicated with Normal Ejection Fraction, NCT04485988, Registered on 24/07/2020. Retrospectively registered.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12872-022-02631-8.

Evidence-Based Process Performance Measures and Clinical Outcomes in Patients With Incident Heart Failure With Reduced Ejection Fraction: A Danish Nationwide Cohort Study

BACKGROUND

Data on the association between quality of heart failure (HF) care and outcomes among patients with incident HF are sparse. We examined the association between process performance measures and clinical outcomes in patients with incident HF with reduced ejection fraction.

METHODS

Patients with incident HF with reduced ejection fraction (n=10 966) between January 2008 and October 2015 were identified from the Danish HF Registry. Data from public registries were linked. Multivariable regression analyses were used to assess the association between 6 guideline-recommended HF care processes (New York Heart Association assessment, use of angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers, beta-blockers, and mineralocorticoid receptor antagonists, exercise training, and patient education) and all-cause and HF readmission, all-cause and HF hospital days, and mortality within 3 to 12 months after HF diagnosis. The associations were analyzed according to the percentages of all relevant performance measures fulfilled for the individual patient (0%-50% [reference group], >50%-75%, and >75%-100%) and for the individual performance measures.

RESULTS

Fulfilling >75% to 100% of the performance measures (n=5341 [48.7%]) was associated with lower risk of all-cause readmission (adjusted hazard ratio, 0.78 [95% CI, 0.68-0.89]) and HF readmission (adjusted hazard ratio, 0.71 [95% CI, 0.54-0.92]), lower use of all-cause hospital days (adjusted mean ratio, 0.73 [95% CI, 0.70-0.76]) and HF hospital days (adjusted mean ratio, 0.79 [95% CI, 0.70-0.89]), and lower mortality (adjusted hazard ratio, 0.42 [95% CI, 0.32-0.53]). A dose-response relationship was observed between fulfilling more performance measures and mortality (adjusted hazard ratio, 0.62 [95% CI, 0.49-0.77] fulfilling >50%-75% of the measures). Fulfilling individual performance measures, except mineralocorticoid receptor antagonist therapy, was associated with lower adjusted all-cause readmission, lower adjusted use of all-cause and HF hospital days, and lower adjusted mortality.

CONCLUSIONS

Fulfilling more process performance measures was associated with better clinical outcomes in patients with incident HF with reduced ejection fraction.