With the aging of the US population, the number of Americans with heart failure (HF) is expected to exceed 8 million, with associated costs of $70 billion, by the end of the next decade.1,2 Patients with HF are currently characterized based on left ventricular ejection fraction (LVEF), as HF with reduced ejection fraction (previously referred to as systolic HF) or HF with preserved ejection fraction (previously known as diastolic HF). Although HF with reduced ejection fraction and HF with preserved ejection fraction are the preferred clinical terms, the terms systolic HF and diastolic HF are still used in the International Classification of Diseases, Ninth Revision (ICD-9) and ICD, Tenth Revision (ICD-10).
Medications that have demonstrated unequivocal improvements in patients with systolic HF have not produced similar beneficial effects in patients with diastolic HF.3 In addition, although survival for patients with systolic HF has improved over the past 20 years, little improvement in survival has been seen for patients with diastolic HF.3 Systolic and diastolic HF cannot be differentiated between based on their signs and symptoms.4,5 Although these 2 types of HF tend to occur in different patient populations, considerable overlap exists, with some patients manifesting both types simultaneously.6 Patients with systolic HF respond different from those with diastolic HF to therapeutic interventions, and each HF type displays distinct patterns of ventricular and myocardial cellular remodeling.6
Because of the demographic, therapeutic, and prognostic differences between systolic and diastolic HF, it is important to understand these populations in terms of patient characteristics, treatment patterns, and overall burden of illness in the real-world setting.
The goal of this study was to describe and compare the treatment patterns, healthcare resource utilization (HCRU), costs, and mortality of US patients newly diagnosed with systolic or diastolic HF who have commercial or Medicare Advantage insurance coverage.
This observational retrospective cohort study used claims in the HealthCore Integrated Research Database (HIRD), which is a large administrative claims database containing geographically diverse, longitudinal medical and pharmacy claims from 14 US commercial health plans. The data were used in full compliance with the relevant provisions of the Health Insurance Portability and Accountability Act. The study was conducted under the research provisions of Privacy Rule 45 CFR 164.514(e) and was exempt from Institutional Review Board review.
Patients had ≥1 new diagnoses of HF in an inpatient setting or ≥2 such diagnoses in a physician’s office or emergency department between January 1, 2010, and June 30, 2016. The diagnosis was identified by ICD-9-Clinical Modification (CM) or ICD-10-CM codes; the first observed diagnosis for HF was set as the index date (Appendix Table 1). Patients were required to have ≥1 evaluations of LVEF between 1 year before and 1 year after the index date to provide certainty of the diagnosis, given the pivotal role of LVEF in the classification of HF. All patients were aged ≥18 years and had continuous health plan enrollment (medical and pharmacy coverage) for ≥12 months before the index date.
Patients were excluded from the study if they had any diagnosis of HF during the entire enrolled period before the index date to ensure that only patients with incident HF were included. Patients with unspecified HF (neither diastolic nor systolic HF), combined systolic HF and diastolic HF, or conflicting systolic and diastolic HF diagnoses were excluded.
The systolic HF cohort consisted of patients whose diagnoses during the 12-month postindex period were consistently systolic HF, without a conflicting HF type based on ICD-9 or ICD-10 codes. Similarly, the diastolic HF cohort was composed of patients whose postindex diagnoses were diastolic HF, without a conflicting type of HF. Patients with conflicting diastolic and systolic HF diagnoses and only unspecified diagnoses were excluded (Appendix Table 1).
The patients were followed for at least 1 year from the index date to the end of the study, the end of health plan eligibility, or death, whichever came first. Mortality was assessed among all the patients, with no minimum postindex eligibility requirement imposed. The treatment patterns, HCRU, and costs of care were evaluated among patients with at least 12 months of continuous eligibility after the index date. For second-, third-, and fourth-year outcomes, these variables were estimated among subsets of patients who had a full 2, 3, or 4 years of follow-up. HCRU and costs were presented on a per-patient per-year basis.
The treatment patterns were assessed through pharmacotherapy utilization. The number and types of HF medications used, medication adherence, and medication persistence were reported during the first year after the index date. Medication adherence was measured with the proportion of days covered, defined as the number of days with a drug on hand divided by the number of days in the specified period.7 Medication persistence was measured by the time from the initiation of therapy to the discontinuation of therapy. Medication discontinuation was defined as not refilling an HF medication within 30 days after the depletion of the previous fill’s supply.
The all-cause and HF-related HCRU included inpatient admissions, 30-day readmissions, length of hospital stay, emergency department visits, physician office visits, skilled-nursing facilities, and prescription fills. HCRU was evaluated before the index date and annually thereafter and was reported as frequency and percentage of patients with ≥1 visits and mean (standard deviation [SD]) visits. Healthcare medical encounters were considered HF-related if the claim had an HF-related diagnosis code in any position.
Similarly, healthcare costs before the index date and at annual follow-up periods were calculated directly from the claims. The costs were presented as the sum of the total paid by health plans, the members’ out-of-pocket costs, and coordination of benefits, and were adjusted to 2017 dollars based on the updated Consumer Price Index by the US Bureau of Labor Statistics. The total costs consisted of expenses incurred in inpatient, emergency department, office visits, other outpatient settings, skilled-nursing facilities, and pharmacy costs.
Mortality was derived from the Social Security Death Index (SSDI). Because the completeness of the mortality data in the SSDI has decreased since 2011 as a result of policy changes,8 the study also captured mortality data from administrative claims. Survival time was defined as the time from the index date to the date of death. Patients who were alive at the end of enrollment or at the end of the study period were considered as censored.
All baseline characteristics and study measures were compared between the systolic HF and diastolic HF cohorts, including mean, SD, median, relative frequency, and percentages. The statistical differences between the cohorts were assessed using Wilcoxon rank-sum test for continuous variables and χ2 test for categorical variables. Multivariable logistic regression was used to assess the odds of all-cause hospitalization for the first year after the index date while controlling for baseline demographic and clinical characteristics. The Hosmer-Lemeshow goodness of fit test was used to assess the model fit.
A generalized linear model with gamma distribution and log link function was used to examine the all-cause healthcare costs during the first year after the index date, controlling for the baseline total costs of care, demographics, HF medications, and comorbidity. Mortality was examined using a Cox proportional hazards model. Because patients with systolic HF and diastolic HF presented different characteristics at baseline, we also ran separate models for each of the HF cohorts.
A conventional alpha of .05 with 2-tailed level of significance was used to interpret statistical significance. Statistical analyses were performed using SAS Enterprise Guide 7.1 (SAS Institute; Cary, NC).
A total of 97,377 patients with HF were identified for the mortality and survival analyses. A total of 68,739 patients who had at least 12 months of continuous health plan eligibility after being diagnosed with HF were included in the HCRU and cost of care analyses. Of these 68,739 patients, 46,885 (68.1%) had systolic HF and 21,854 (31.8%) had diastolic HF (Figure 1).
Among the 68,739 patients with HF, 46,602 (67.8%), 30,905 (45%), and 20,000 (29.1%) patients had 2, 3, and 4 years of follow-up, respectively, after being diagnosed with HF.
The patients in the systolic HF cohort were approximately 8 years younger (mean age, 66.1 years vs 74.4 years, respectively; P <.001) and were more likely to be male (61.7% vs 40.1%, respectively; P <.001) than the patients in the diastolic HF cohort (Table 1).
Although most of the patients in both groups were diagnosed by a cardiologist, the proportion of patients identified by a cardiologist was greater in the systolic HF cohort than in the diastolic HF cohort (65.3% vs 52.5%, respectively; P <.001).
In addition, the systolic HF cohort had a lower prevalence of dyslipidemia, hypertension, ischemic stroke, peripheral vascular disorder, diabetes, chronic kidney disease, atrial fibrillation, chronic obstructive pulmonary disorder, anemia, and depression, but a slightly higher prevalence of coronary heart disease and myocardial infarction (all P <.001; Table 1).
Before diagnosis, 59.9% of the study population received 1 or more HF-related medications for an indication other than HF. Patients with systolic HF were less likely to receive any HF-related medications than patients with diastolic HF (57% vs 66.1%, respectively; P <.001; Table 2). During the first year after the index date, only 77.1% of the total patients received treatment with at least 1 HF medication; 21.9% of patients with systolic HF and 25% of patients with diastolic HF received no HF medications. For patients followed for 4 years, the treatment rate further declined to 70.1% (Appendix Figure 1).
Among all the patients, the most frequently prescribed HF medications were angiotensin-converting enzyme (ACE) inhibitors, beta-blockers, and loop diuretics. Patients with systolic HF were more likely to receive ACE inhibitors (46.4% vs 30.3%, respectively; P <.001) and beta-blockers (56.5% vs 44.8%, respectively; P <.001), whereas patients with diastolic HF were more likely to receive loop diuretics (35.5% vs 49.9%, respectively; P <.001; Table 2).
Adherence to HF medications was similar in both cohorts. The proportion of days covered among those who received treatment, which was measured by the number of days with drug on hand during the 365-day period, was 81.9% for patients with systolic HF and 82.2% for those with diastolic HF (P = .364). The average time to treatment discontinuation (medication persistence) was slightly longer among patients with systolic HF than among patients with diastolic HF (278.2 days vs 273.9 days, respectively; P <.001), and although statistically significant, this difference was likely not clinically meaningful.
During the first postindex year, patients with systolic HF had less HCRU than patients with diastolic HF, including hospital admissions (70.2% vs 82.4%, respectively; P <.001; Appendix Table 2), emergency department visits (30.5% vs 39.1%, respectively; P <.001), physician office visits (mean, 14.2 vs 15.5, respectively; P <.001), and skilled-nursing facility use (8.1% vs 17.4%, respectively; P <.001; Appendix Table 2). A smaller proportion of patients with systolic HF was diagnosed in inpatient settings compared with patients with diastolic HF (55.3% vs 67.6%, respectively; P <.001).
All-cause inpatient admissions were high during the first postindex year, at 74% of all patients, but this percentage decreased substantially by 25% to 35% in subsequent years (Figure 2). The higher proportion of hospitalizations during the first year was mainly because 59.2% of patients with HF were first diagnosed in the hospital (Appendix Table 2). In the first year, the all-cause 30-day readmission rate was 26.5%. Although the majority (93.4%) of admissions in the first year were related to HF, only 41.6% of the 30-day readmissions were related to HF (Appendix Table 2).
Compared with patients with diastolic HF, those with systolic HF had relatively fewer hospital admissions (82.4% vs 70.2%, respectively; P <.001) and 30-day readmissions (30.9% vs 24.1%, respectively; P <.001) in the first year. This trend remained consistent through years 2 through 4 (Figure 2). The proportion of all-cause hospitalization remained stable at approximately 25% to 26% in the systolic HF cohort and approximately 35% to 36% in the diastolic HF cohort during years 2 through 4 after diagnosis.
In a multivariable analysis, patients with diastolic HF were significantly more likely to be hospitalized than patients with systolic HF (odds ratio [OR], 1.62; 95% confidence interval [CI], 1.55-1.69; Appendix Figure 2). Female patients and patients older than age 75 years had a higher risk for hospitalization than male patients and patients aged 18 to 44 years. Exposure to HF-related medications before diagnosis significantly reduced the risk for hospitalization during the first year after the index date (OR, 0.67, 0.58, and 0.57 for 1, 2, or ≥3 medication classes, respectively; all P <.0001).
Further stratified analysis showed that HF-related medications prescribed before diagnosis had protective effects in patients with systolic HF (OR, 0.64, 0.55, and 0.55 for 1, 2, and ≥3 medication classes, respectively; all P <.0001) and for patients with diastolic HF (OR, 0.83, 0.73, and 0.68 for 1, 2, and ≥3 medication classes, respectively; all P <.001; Appendix Table 3), with a trend toward a greater protective effect in patients with systolic HF.
Cost of Care
Before a diagnosis of HF, a univariate analysis showed that patients with systolic HF had lower annual all-cause healthcare costs than patients with diastolic HF (mean, $20,455 vs $23,188, respectively; P <.001; Figure 3). Despite having lower costs and fewer hospital admissions before diagnosis, the average costs 1 year after the index date were higher for patients with systolic HF than for patients with diastolic HF ($64,154 vs $59,652, respectively; P <.001; Appendix Table 4, and Figure 3). However, in years 2 through 4, the average total cost for patients with systolic HF was lower than for patients with diastolic HF (Figure 3).
HF-related costs accounted for 62% and 55% of the total healthcare costs in the first year for patients with systolic HF and diastolic HF, respectively (data derived from Figure 3). The HF-related costs decreased markedly from $39,957 for patients with systolic HF and $32,617 for patients with diastolic HF during the first year to approximately $7000 in each group in years 2 through 4 (Figure 3). Compared with patients with diastolic HF, patients with systolic HF had higher total HF-related costs in the first year (P = .0131), but similar costs in the second through fourth years (all P >.15).
The total costs were similar between the groups after controlling for baseline costs and characteristics using a generalized linear model (exponentiated estimate, 1.01; 95% CI, 0.99-1.02; Appendix Figure 3). The number of HF-related medication classes prescribed during the year preceding a diagnosis was associated with lower all-cause costs (exponentiated estimates of 0.92, 0.89, and 0.85 for 1, 2, or ≥3 medication classes, respectively; all P <.0001).
The effects of HF-related medications on the cost of care were more pronounced among patients with systolic HF (exponentiated estimates, 0.88, 0.85, and 0.83 for 1, 2, and ≥3 medication classes, respectively; all P <.0001) than in patients with diastolic HF (exponentiated estimates, 1.01, 0.98, and 0.9; P = .72, P = .21, and P <.0001, respectively; Appendix Table 5).
In the univariate analysis, patients with diastolic HF had a higher mortality rate than patients with systolic HF (19.4% vs 15.2%, respectively; P <.001; data not shown). After controlling for baseline characteristics, the Cox proportional hazards model showed that the adjusted all-cause mortality was slightly lower for patients with diastolic HF than for patients with systolic HF (hazard ratio [HR], 0.96; 95% CI, 0.93-0.99; Appendix Figure 4). The use of HF-related medications for other conditions before being diagnosed with HF was associated with lower all-cause mortality (HR, 0.95, 0.94, and 0.93 for 1, 2, and ≥3 medication classes; P = .03, P .01, P <.001, respectively).
Although additional analysis showed that HF-related medications lowered the all-cause mortality risk among patients with systolic HF (HR, 0.91, 0.89, and 0.89 for 1, 2, or ≥3 medication classes, respectively; all P <.001), these protective effects were not significant among patients with diastolic HF (HR, 1.04, 1.04, and 0.99, respectively; all P >.25; Appendix Table 6).
This real-world observational study provides detailed clinical and economic outcomes associated with systolic HF and diastolic HF. This level of detail across these HF phenotypes is important to assess the resources needed for clinical interventions, because they provide background on the potential cost-savings from successful interventions. This is important information for private and public payers, as well as for health systems and providers who are now increasingly taking financial risk in their contracts with payers. In addition, these data demonstrate gaps in care regarding suboptimal HF medication utilization, as well as the real-world benefits of HF drug use on hospitalizations, costs, and mortality.
The demographic differences in this study between patients with systolic HF and patients with diastolic HF are consistent with previous epidemiologic studies and support the appropriateness of the cohort identification criteria.9-11 Our study population consisted of a higher proportion of patients with systolic HF, unlike previous studies.3,12 The population excluded from analysis, including patients with unspecified and combined HF, were more similar to patients with diastolic HF than those with systolic HF in terms of baseline characteristics and clinical outcomes (data not shown).
Before diagnosis, 57% and 66.1% of the patients with systolic HF and diastolic HF, respectively, had already received HF-related medication for non-HF diagnoses. Because these medications have a role in the management of other cardiovascular conditions, such as hypertension and post–myocardial infarction, they were likely taken for those reasons.
This analysis confirmed a high disease burden for patients (including those with diastolic HF), healthcare providers, and health plans. The proportion of inpatient admissions for patients with systolic HF was consistently lower than for patients with diastolic HF. The systolic HF cohort also had relatively lower HCRU than the diastolic HF cohort, which was partly driven by the patients’ younger age. Our overall findings regarding all-cause hospitalization and readmissions were comparable with the literature.13,14
The slightly higher all-cause and HF-related costs among patients with systolic HF during the first year could reflect cardiac events and comorbidities more frequently associated with systolic HF. Because coronary artery disease is causally linked to systolic HF, the management of coronary artery disease after diagnosis likely drove the cost for patients with systolic HF. However, once the patient was optimally treated (with or without revascularization), the downstream management was less costly, because the treatment was mostly for the management of risk factors.
By contrast, diastolic HF–related costs were less concentrated, because associated comorbidities were not centered on a specific event but instead required long-term consistent management. After controlling for the baseline characteristics, the all-cause costs during the first year after the index were similar between the groups. Our all-cause total cost estimations were higher than those in the literature using Medicare data,15,16 but our inpatient costs were similar to those from a 2017 study.17
As in previous literature,18 our results showed that medication use after an HF diagnosis was suboptimal, even in patients with systolic HF, which has a significant evidence base for recommended therapies. We found that patients with systolic HF were more likely to receive ACE inhibitors and beta-blockers, whereas patients with diastolic HF were more likely to receive loop diuretics.
These observations were consistent with the current literature, reflecting the benefits of medications among patients with systolic HF and the absence of proved therapies that have consistently demonstrated improvements in morbidity and mortality in patients with diastolic HF.18-21 Healthcare providers were left with targeting the symptoms of patients with diastolic HF (with diuretics) rather than long-term outcomes. This pattern is consistent with treatment guidelines for patients with systolic HF, which recommend renin-angiotensin system inhibition and diuretics as first-line therapy, with beta-blockers added after the patient’s condition has been stabilized.22,23
As many as 21.9% of patients with systolic HF and 25% of patients with diastolic HF received no HF medications within the first year after the index date, and the rate of untreated patients increased in subsequent years. This represents a noteworthy gap in care and highlights a significant opportunity for clinical intervention.
The number of HF-related medication classes patients received before HF diagnosis was significantly associated with low hospitalization risk, lower healthcare costs, and reduced mortality risk. This indicated that early intervention and early treatment may have positive effects on the prevention of hospitalization and the reduction of cost for patients with HF. Further stratified analysis showed that HF-related medications had stronger protective effects among patients with systolic HF than patients with diastolic HF, which aligns with the significant evidence-based benefit of recommended therapy for patients with systolic HF.
Existing data comparing patients with systolic HF and patients with diastolic HF were inconsistent with respect to mortality. Patients with diastolic HF were previously considered to have a better prognosis than patients with systolic HF.24 More recent data, however, suggest that mortality rates are similar between the groups.25,26 In addition, the trend analysis showed that the survival of patients with systolic HF had improved over the past 2 decades, but there was no improvement in the survival of patients with diastolic HF.5,27
We observed a higher mortality risk among patients with diastolic HF in the univariate analysis, which was highly driven by the patients’ significantly older age. After adjusting for all baseline demographic and clinical characteristics, the adjusted all-cause mortality of patients with diastolic HF was slightly lower than that for patients with systolic HF.
This study has several limitations. Despite our best efforts, the specific HF group classification using administrative claims alone and the limitation of ICD-9 and ICD-10 coding (using systolic HF and diastolic HF instead of HF with reduced ejection fraction and HF with preserved ejection fraction) could be subject to errors. We also excluded a significant proportion of patients with combined, unspecified, and conflicting diagnoses of HF. The 12-month period without a diagnosis of HF before the index date might not have been long enough to ensure that those individuals had incident HF rather than prevalent HF. However, there was no evidence to indicate that these errors systematically differed across the groups.
All the patients in the study were from US commercially insured and Medicare Advantage populations, which may limit the generalizability of these results to other subpopulations, such as traditional fee-for-service Medicare and Medicaid, and the uninsured. Future research to understand the treatment of HF in traditional fee-for-service Medicare and Medicaid patient populations is needed. Social determinants of health, such as race, education, and socioeconomic status, were not available in our HIRD database at the time our study was conducted. These social, behavioral, and environmental factors could influence a patient’s choice of treatment, adherence to healthcare recommendations, and subsequent health outcomes.
The length of follow-up required to estimate HCRU and cost of care might have excluded patients with severe conditions who died early after diagnosis, which may have potentially presented bias in these categories.
The multivariable analysis only examined the medication effects during the year before an HF diagnosis. Further research with a more sophisticated statistical model is needed to consider the variations of medication exposure over time.
Another limitation was the use of SSDI to ascertain death status. Since 2011, the Social Security Administration no longer requires states to provide death records, and thus a number of deaths captured in SSDI were expunged. Although we supplemented the SSDI data with the mortality data captured in the administrative claims, the mortality rate might have been underestimated. However, there was no evidence to indicate that such potential underestimation was systematically different between the cohorts.
The results of this real-world analysis confirm a high disease burden for patients with HF and highlight a need to improve the management of this condition. Inpatient admissions were high, placing a significant burden on patients, healthcare providers, and health plans. Medication use after a diagnosis of HF was suboptimal, even in patients with systolic HF, which has a significant evidence base for recommended therapies. Approximately 20% of patients with systolic HF and 25% of patients with diastolic HF received no HF medications within the first year after diagnosis, and these high rates of untreated patients underscore a gap in care and provide an opportunity for intervention.
Compared with patients with systolic HF, patients with diastolic HF had a higher risk for hospitalization, but comparable adjusted costs and slightly lower mortality risk. Exposure to HF medications before diagnosis was associated with lower hospitalization risk, lower healthcare costs, and reduced mortality. These protective effects of HF medications were more prominent among patients with systolic HF than those with diastolic HF. These results indicate a critical need for evidence-based therapies in patients with diastolic HF.
The authors thank Zhengzheng Jiang for data programming, and Cheryl Jones for writing and editorial support. All errors remain the responsibility of the authors.
This work was funded by Bayer AG, Germany.
Author Disclosure Statement
Dr Nguyen, Dr Willey, and Mr Tan are employees of HealthCore, which has received research support from pharmaceutical, biotechnology, and medical device companies, including Bayer AG; Dr Evers is an employee of Bayer AG, Germany; Dr Zhang and Dr Power have no conflicts of interest to report.
- Dunlay SM, Roger VL. Understanding the epidemic of heart failure: past, present, and future. Curr Heart Fail Rep. 2014;11:404-415.
- Benjamin EJ, Virani SS, Callaway CW, et al; for the American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics—2018 Update: a report from the American Heart Association. Circulation. 2018;137:e67-e492. Erratum in: Circulation. 2018;137:e493.
- Owan TE, Hodge DO, Herges RM, et al. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med. 2006;355:251-259.
- Shah RV, Desai AS, Givertz MM. The effect of renin-angiotensin system inhibitors on mortality and heart failure hospitalization in patients with heart failure and preserved ejection fraction: a systematic review and meta-analysis. J Card Fail. 2010;16:260-267.
- Pfeffer MA, Claggett B, Assmann SF, et al. Regional variation in patients and outcomes in the Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist (TOPCAT) trial. Circulation. 2015;131:34-42.
- Borlaug BA, Redfield MM. Diastolic and systolic heart failure are distinct phenotypes within the heart failure spectrum. Circulation. 2011;123:2006-2013; discussion 2014.
- Peterson AM, Nau DP, Cramer JA, et al. A checklist for medication compliance and persistence studies using retrospective databases. Value Health. 2007;10:3-12.
- Jones B, Vawdrey DK. Measuring mortality information in clinical data warehouses. AMIA Jt Summits Transl Sci Proc. 2015;2015:450-455.
- Savarese G, Lund LH. Global public health burden of heart failure. Card Fail Rev. 2017;3:7-11.
- Quiroz R, Doros G, Shaw P, et al. Comparison of characteristics and outcomes of patients with heart failure preserved ejection fraction versus reduced left ventricular ejection fraction in an urban cohort. Am J Cardiol. 2014;113:691-696.
- Yancy CW, Lopatin M, Stevenson LW, et al; for the ADHERE Scientific Advisory Committee and investigators. Clinical presentation, management, and in-hospital outcomes of patients admitted with acute decompensated heart failure with preserved systolic function: a report from the Acute Decompensated Heart Failure National Registry (ADHERE) database. J Am Coll Cardiol. 2006;47:76-84. Erratum in: J Am Coll Cardiol. 2006;47:1502.
- Steinberg BA, Zhao X, Heidenreich PA, et al; for the Get With the Guidelines Scientific Advisory Committee and investigators. Trends in patients hospitalized with heart failure and preserved left ventricular ejection fraction: prevalence, therapies, and outcomes. Circulation. 2012;126:65-75.
- Chamberlain AM, Dunlay SM, Gerber Y, et al. Burden and timing of hospitalizations in heart failure: a community study. Mayo Clin Proc. 2017;92:184-192.
- Nichols GA, Reynolds K, Kimes TM, et al. Comparison of risk of re-hospitalization, all-cause mortality, and medical care resource utilization in patients with heart failure and preserved versus reduced ejection fraction. Am J Cardiol. 2015;116:1088-1092.
- Dunlay SM, Shah ND, Shi Q, et al. Lifetime costs of medical care after heart failure diagnosis. Circ Cardiovasc Qual Outcomes. 2011;4:68-75.
- Greiner MA, Hammill BG, Fonarow GC, et al. Predicting costs among Medicare beneficiaries with heart failure. Am J Cardiol. 2012;109:705-711.
- Kilgore M, Patel HK, Kielhorn A, et al. Economic burden of hospitalizations of Medicare beneficiaries with heart failure. Risk Manag Healthc Policy. 2017;10:63-70.
- Oghlakian GO, Sipahi I, Fang JC. Treatment of heart failure with preserved ejection fraction: have we been pursuing the wrong paradigm? Mayo Clin Proc. 2011;86:531-539.
- Davis BR, Kostis JB, Simpson LM, et al; for the ALLHAT Collaborative Research Group. Heart failure with preserved and reduced left ventricular ejection fraction in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Circulation. 2008;118:2259-2267.
- Yusuf S, Pfeffer MA, Swedberg K, et al; for the CHARM investigators and committees. Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-Preserved Trial. Lancet. 2003;362:777-781.
- Massie BM, Carson PE, McMurray JJ, et al; for the I-PRESERVE investigators. Irbesartan in patients with heart failure and preserved ejection fraction. N Engl J Med. 2008;359:2456-2467.
- Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the Diagnosis and Treatment of Acute and Chronic Heart Failure. Rev Esp Cardiol (Engl Ed). 2016;69:1167. doi: 10.1016/j.rec.2016.11.005. Erratum in: Rev Esp Cardiol (Engl Ed). 2017;70:309-310.
- Yancy CW, Jessup M, Bozkurt B, et al. 2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. Circulation. 2017;136:e137-e161.
- Cohn JN, Johnson G. Heart failure with normal ejection fraction. The V-HeFT Study. Veterans Administration Cooperative Study Group. Circulation. 1990;81(2 suppl):III48-III53.
- Lenzen MJ, Scholte op Reimer WJM, Boersma E, et al. Differences between patients with a preserved and a depressed left ventricular function: a report from the EuroHeart Failure Survey. Eur Heart J. 2004;25:1214-1220.
- Bhatia RS, Tu JV, Lee DS, et al. Outcome of heart failure with preserved ejection fraction in a population-based study. N Engl J Med. 2006;355:260-269.
- Miljkovik LV, Spiroska V. Heart failure with preserved ejection fraction – concept, pathophysiology, diagnosis and challenges for treatment. Open Access Maced J Med Sci. 2015;3:521-527.
- Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43:1130-1139.