Abstract
Aims
This study aimed to examine the diagnostic pathways and outcomes of patients with heart failure (HF), stratified by left ventricular ejection fraction (EF), and to highlight deficiencies in real‐world HF diagnosis and management.
Methods and results
We conducted a retrospective cohort study in Salford, United Kingdom, utilizing linked primary and secondary care data for HF patients diagnosed between January 2010 and November 2019. We evaluated characteristics, diagnostic patterns, healthcare resource utilization, and outcomes. Patients were categorized according to baseline (the latest measure prior to or within 90 days post‐diagnosis) as having HF with reduced EF (HFrEF), mildly reduced EF (HFmrEF), or preserved EF (HFpEF). The data encompassed a 2 year period before diagnosis and up to 5 years post‐diagnosis. A total of 3227 patients were diagnosed with HF between January 2010 and November 2019. The mean follow‐up time was 2.6 [±1.9 standard deviation (SD)] years. The mean age at diagnosis was 74.8 (±12.7 SD) years, and 1469 (45.5%) were female. HFpEF was the largest cohort (46.6%, n pEF = 1505), HFmrEF constituted 16.1% (n mrEF = 520), and HFrEF 18.5% (n rEF = 596) of the population, while 18.8% (n u = 606) of patients remained unassigned due to insufficient evidence to support categorization. At baseline, measurement of natriuretic peptide (NP; brain NP and N‐terminal pro‐B‐type NP) and echocardiographic report data were available for 592 (18.3%) and 2621 (81.2%) patients, respectively. A total of 2099 (65.0%) of the HF cohort had access to a cardiology‐led outpatient clinic prior to the HF diagnosis, and 602 (18.7%) attended cardiac rehabilitation post‐diagnosis. The 5 year crude survival rate was 37.8% [95% confidence interval (CI) (35.2–40.7%)], 42.3% [95% CI (38.0–47.2%)], and 45.5% [95% CI (41.0–50.4%)] for HFpEF, HFrEF, and HFmrEF, respectively.
Conclusions
Low survival rates were observed across all HF groups, along with suboptimal rates of NP testing and specialist assessments. These findings suggest missed opportunities for timely and accurate HF diagnosis, a pivotal first step in improving outcomes for HF patients. Addressing these gaps in diagnosis and management is urgently needed.
Keywords: Heart failure, Ejection fraction, Echocardiogram, Patient pathways, Missed opportunity, Electronic medical records
Introduction
Over 900 000 people in the United Kingdom (UK) live with heart failure (HF), a complex syndrome resulting from cardiac abnormalities that reduce cardiac output. 1 Approximately one in five individuals will develop HF during their lifetime. 2 For most, HF is a chronic, incurable condition that places a significant burden on patients, families, and healthcare systems. 3 , 4 , 5 Despite diagnostic and treatment advancements, the HF prognosis remains poor, and 50% of HF patients will die within 5 years of diagnosis. 6 Given these factors, early diagnosis and prompt, evidence‐based interventions are critical for improving patient outcomes. Delays in initial diagnosis and treatment negatively affect long‐term survival.
European and American guidelines describe pathways for the initial assessment and diagnosis of HF. 7 , 8 , 9 The National Institute for Health and Care Excellence (NICE) clinical practice guidelines for chronic HF in adults were first released in 2003 and most recently updated in 2018. These UK guidelines recommend that patients with suspected HF undergo a specialist clinical assessment, including echocardiography, to confirm the diagnosis, assess severity, determine the cause, and identify correctable factors. This assessment should occur within 2 weeks for patients with an N‐terminal pro‐B‐type natriuretic peptide (NP) (NT‐proBNP) level exceeding 2000 ng/L or within 6 weeks for those with levels between 400 and 2000 ng/L. 1 However, UK data reveal significant delays, with some patients waiting up to 5 months for an HF diagnosis, with women disproportionately affected by these delays. 10 Geographical variations in resources and clinical practice, combined with clinical inertia, are likely implicated in the delays observed.
Despite being criticized as outdated, the management of HF has traditionally been based on left ventricular ejection fraction (EF) (LVEF). 11 This classification divides HF patients into three groups: HF with reduced EF (HFrEF; EF < 40%), HF with mildly reduced EF (HFmrEF; EF 40–49%), and HF with preserved EF (HFpEF; EF ≥ 50%). 12
For HFrEF patients, NICE recommends an angiotensin‐converting enzyme inhibitor (ACE‐I) or an angiotensin II receptor blocker combined with a neprilysin inhibitor (ARNI), along with a beta‐blocker and a mineralocorticoid receptor antagonist (MRA). Angiotensin II receptor blockers are suggested for those unable to tolerate ACE‐I or ARNI. Diuretics are recommended for symptom relief and fluid retention across all EF categories.
Until recently, treatment options for HFpEF and HFmrEF were limited. In 2020, the European Society of Cardiology (ESC) Guidelines expanded the range of treatments used for HFrEF to include patients with mildly reduced EF. More recently, sodium‐glucose co‐transporter 2 (SGLT2) inhibitors, initially used for diabetes, received ESC and NICE endorsements for HF treatment. This recommendation is based on new data demonstrating their cardiovascular (CV), renal, and metabolic benefits across all EF categories, including those with mildly reduced and preserved EF. 9 , 13
Recent studies have shown that prompt specialist review of HF patients, regardless of their EF, leads to better outcomes. Patients who receive specialist assessment experience fewer all‐cause hospital admissions and lower all‐cause mortality compared with those who do not. 14 , 15 Importantly, in these studies, more than half of these patients had HFpEF, emphasizing that even though there are limited disease‐modifying treatments for HFpEF, the specialist review remains beneficial. Specialist assessment also appears to facilitate the early detection and treatment of associated comorbidities like obesity, renal failure, and pulmonary disease. Additionally, it often leads to referrals for cardiac rehabilitation (CR) services, which can contribute to improved outcomes. 14
However, significant challenges persist in the diagnosis and management of HF, with data indicating suboptimal patient journeys across primary and secondary care, inconsistent adherence to NICE guidelines, and little improvement in UK survival rates over the past two decades. 16 , 17 , 18 , 19 For instance, although the 2018 NICE guidelines reported that uptake of the 2010 guidance ‘appears to be good’, the 2013 ‘Cardiovascular Disease Outcomes Strategy’, also published by NICE, estimated that just 4% of patients with HF receive CR. 20 Recent data from the National Heart Failure Audit reveal that just 12% of HF patients hospitalized in England and Wales in 2020/21 were referred for CR. 21 Furthermore, beyond the National Institute for Cardiovascular Outcomes Research (NICOR) data, which describe management during hospitalization for HF, there is a lack of UK‐level real‐world data on adherence to clinical practice guidelines for HF.
Considering these challenges, we present real‐world data on diagnostic pathways, management, resource utilization, and outcomes of HF patients categorized by EF in a UK urban population. Our aim is to identify areas along the HF patient pathway where potential opportunities for improved patient care exist.
Methods
Data source
We conducted a retrospective observational cohort study to assess the clinical characteristics, healthcare interactions, and mortality of adult patients diagnosed with chronic HF in Salford, Greater Manchester (UK), from January 2010 to November 2019, utilizing an anonymized dataset extracted from the Salford Integrated Record (SIR). 22 , 23 , 24
SIR is a comprehensive repository of linked primary and secondary care electronic medical records (EMRs) for approximately 251 000 people from the population of Salford, including data from 45 general practitioner (GP) surgeries and a large university teaching hospital (Salford Royal). Every time a patient visits a GP surgery, clinic, or hospital within Salford, healthcare data are generated about their condition(s) and treatment and recorded in several different electronic healthcare systems. Through established infrastructure and partnership agreements between primary and secondary care organizations in Salford, these healthcare data are consolidated into a single, easily accessible record. This ensures that medical professionals have instant access to patient information at the point of care. SIR is maintained and used to allow healthcare providers not only to view and share patient information but also to facilitate research and population health management by providing anonymized data extracts to support research studies. 24 Patients who have opted out of sharing their data for research purposes are automatically excluded from the SIR research database and, consequently, from participation in studies that rely on these data, including this study.
The data are largely structured in nature but also provide the potential to utilize information included in semi‐structured data sources [e.g. echocardiographic (ECHO) reports]. The primary care record is coded using Version 2.0 Read Codes and Egton Medical Information Systems (EMIS) codes, while the secondary care record adopted the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD‐10 Version 2010).
Study population
To assess whether treatment was aligned with clinical practice guidelines, patients were classified into one of the three groups (HFrEF, HFmrEF, or HFpEF) based on the LVEF measured during the echocardiogram performed at baseline. For the purpose of this study, ‘baseline’ was defined as the latest measure obtained either prior to or within 90 days after diagnosis, a consistent definition applied throughout this study.
For patients without a numerical EF value, we considered a qualitative description or assessment summary of their left ventricular systolic function (LVSF) as a suitable proxy. Specifically, we defined HFpEF as EF ≥ 50% or qualitatively ‘good/normal’ LVSF, HFmrEF as EF ranging from 40% to 49% or qualitatively ‘mildly/moderately impaired’ LVSF, and HFrEF as LVEF < 40% or qualitatively ‘poor/severely impaired’ LVSF.
Characteristics and comorbidities
The dataset included clinical records containing information on past and current comorbidities, clinical characteristics, blood tests, measurements, and observations. The comorbidities considered in the current analysis included chronic kidney disease (CKD) (defined as CKD stage 1 and greater), atrial fibrillation (AF), implantable cardioverter defibrillator, cardiac resynchronization therapy (with or without a defibrillator), type II diabetes, chronic obstructive pulmonary disease (COPD), asthma, and obesity. Additionally, due to the potential underdiagnosis of cardiac amyloidosis as a cause of HF, especially in cases of HFpEF, we extended the search within the EMR to include indicators or ‘red flags’ for this condition. This involved an exploratory analysis of clinical codes associated with carpal tunnel syndrome, atrioventricular node block, neuropathy, lumbar spinal stenosis, and rupture of the biceps tendon.
Regarding the clinical characteristics of the cohort, we assessed functional status recorded using the New York Heart Association (NYHA) classification and evaluated breathlessness, which was either noted as present or measured using the Medical Research Council (MRC) Breathlessness Scale, ranging from 1 to 5. 25
Measurements and treatments
We conducted an analysis of blood tests and the use of HF‐related medications. Blood tests and measurements of interest included brain NP (BNP) and NT‐proBNP, estimated glomerular filtration rate (eGFR), serum total bilirubin, and creatinine levels. Additionally, we assessed troponin I and T types, as elevated levels can be indicative of cardiac amyloidosis. 26
To determine medication usage, we examined medical records to ascertain the proportion of patients prescribed specific medications related to HF. Medications of interest included ACE‐Is, angiotensin II antagonists, ARNIs, beta‐blockers, aldosterone antagonists, diuretics (loop, potassium‐sparing, thiazide, and combinations), sinus node inhibitors, anticoagulants, cardiac glycosides, nitrates, and hydralazine hydrochloride/isosorbide dinitrate.
Outcome measures
In our health resource utilization (HRU) analysis, we examined secondary care encounters. In particular, all‐cause accident and emergency department (A&E) visits, all‐cause and CV‐related inpatient hospital admissions, cardiology outpatient clinics, HF specialist clinics, and CR encounters for the 2 years leading up to the HF diagnosis and up to 5 years following the diagnosis. The death summary record was utilized to conduct survival analysis.
Statistical approach
Statistical analyses were mainly descriptive in nature. A comparison of various characteristics between HF phenotypes and the reference HFrEF was conducted, and associated P‐values were presented. In the pairwise comparison between groups, non‐parametric statistical tests were used, with the Wilcoxon test for continuous variables, the χ 2 test for categorical variables with all expected cell counts ≥ 5, and Fisher's exact test for categorical variables with any expected cell count < 5.
The clinical endpoint was all‐cause death occurring during the follow‐up period of up to 5 years. Survival analysis according to EF phenotype, both crude and adjusted for age and sex, was carried out using the Kaplan–Meier method. Cox regression was used to examine the associations of covariates with all‐cause mortality; hazard ratios (HRs) and two‐sided 95% confidence intervals (CIs) for means of quantitative variables were provided, as appropriate. The overall statistical significance of Cox models was evaluated using the likelihood ratio test, Wald's test, and log‐rank statistics. A value of P < 0.05 was assumed to indicate a statistically significant result.
All data preparation before analysis was undertaken using Structured Query Language (SQL); statistical analysis was performed using STATA Version 15 and R Version 4.0.5.
Results
Study population
This study included a total of 3227 patients who received a first diagnosis of HF between January 2010 and November 2019. The mean follow‐up time post‐diagnosis was 2.6 [±1.9 standard deviation (SD)] years.
Based on reported EF or LVSF, 1505 (46.6%) HF patients were categorized as having HFpEF, 520 (16.1%) HFmrEF, and 596 (18.5%) HFrEF. An additional 606 (18.8%) patients could not be categorized into any of these defined groups due to insufficient supporting evidence.
Characteristics and comorbidities
The mean age at the time of HF diagnosis was 75 (±13 SD) years, and 45.5% (1469) of the total cohort was female. Compared with patients with HFrEF, those with HFpEF were typically older (median age 78 years for HFpEF vs. 75 years for HFrEF, P < 0.001) and more often female (53.4% for HFpEF vs. 36.9% for HFrEF, P < 0.001). Smoking status also differed significantly among HF phenotypes (P < 0.001), with patients in the HFpEF group being less likely to be a current smoker (15.4% vs. 23.5% for HFrEF) (Table 1 ).
Table 1.
Baseline characteristics of heart failure study population
| Characteristic | Heart failure population | P‐values HFrEF | ||||||
|---|---|---|---|---|---|---|---|---|
| Overall, N = 3227 a | HFrEF, N = 596 a | HFmrEF, N = 520 a | HFpEF, N = 1505 a | Unknown, N = 606 a | vs. HFmrEF b | vs. HFpEF b | vs. unknown b | |
| Sex | >0.9 | <0.001*** | 0.076 | |||||
| Female | 1469 (45.5%) | 220 (36.9%) | 192 (36.9%) | 803 (53.4%) | 254 (41.9%) | |||
| Male | 1758 (54.5%) | 376 (63.1%) | 328 (63.1%) | 702 (46.6%) | 352 (58.1%) | |||
|
Age at diagnosis date (years) Median (IQR) |
77 (67, 84) | 75 (65, 82) | 76 (67, 83) | 78 (70, 85) | 76 (65, 85) | 0.077 | <0.001*** | 0.017* |
| Smoking status | 0.3 | <0.001*** | <0.001*** | |||||
| Current smoker | 586 (18.2%) | 140 (23.5%) | 101 (19.4%) | 231 (15.3%) | 114 (18.8%) | |||
| Ex‐smoker | 1377 (42.7%) | 239 (40.1%) | 233 (44.8%) | 666 (44.3%) | 239 (39.4%) | |||
| Never smoked | 1057 (32.8%) | 190 (31.9%) | 160 (30.8%) | 539 (35.8%) | 168 (27.7%) | |||
| Unknown | 207 (6.4%) | 27 (4.5%) | 26 (5.0%) | 69 (4.6%) | 85 (14.0%) | |||
| BMI (kg/m2) | ||||||||
| Mean (SD) | 28 (7) | 27 (7) | 28 (6) | 29 (7) | 28 (7) | 0.047* | <0.001*** | 0.093 |
| Unknown | 326 (10%) | 53 (8.9%) | 47 (9.0%) | 115 (7.6%) | 111 (18%) | |||
| New York Heart Association (NYHA) classification | 0.2 | <0.001*** | <0.001*** | |||||
| NYHA class I | 36 (1.1%) | 12 (2.0%) | DCS | 12 (0.8%) | <6 | |||
| NYHA class II | 90 (2.8%) | 32 (5.4%) | 16 (3.1%) | 32 (2.1%) | 10 (1.7%) | |||
| NYHA class III | 54 (1.7%) | 17 (2.9%) | DCS | 19 (1.3%) | <6 | |||
| NYHA class IV | <6 | — | <6 | — | — | |||
| Unknown | 3046 (94.4%) | 535 (89.8%) | 483 (92.9%) | 1442 (95.8%) | 586 (96.7%) | |||
| Breathlessness | 0.5 | 0.003** | 0.026* | |||||
| No | 15 (0.5%) | <6 | <6 | <6 | <6 | |||
| Yes | 1280 (39.7%) | 221 (37.1%) | 210 (40.4%) | 668 (44.4%) | 181 (29.9%) | |||
| Unknown | 1932 (59.9%) | 371 (62.2%) | 307 (59.0%) | 833 (55.3%) | 421 (69.5%) | |||
| Medical Research Council (MRC) Breathlessness Scale | 0.2 | 0.006** | 0.3 | |||||
| Grade 1 | 67 (2.1%) | 16 (2.7%) | 16 (3.1%) | 29 (1.9%) | 6 (1.0%) | |||
| Grade 2 | 166 (5.1%) | 31 (5.2%) | 24 (4.6%) | 82 (5.4%) | 29 (4.8%) | |||
| Grade 3 | 190 (5.9%) | 19 (3.2%) | 34 (6.5%) | 109 (7.2%) | 28 (4.6%) | |||
| Grade 4 | 167 (5.2%) | 27 (4.5%) | 22 (4.2%) | 89 (5.9%) | 29 (4.8%) | |||
| Grade 5 | 52 (1.6%) | 8 (1.3%) | 8 (1.5%) | 27 (1.8%) | 9 (1.5%) | |||
| Unknown | 2585 (80.1%) | 495 (83.1%) | 416 (80.0%) | 1169 (77.7%) | 505 (83.3%) | |||
| Comorbidities | ||||||||
| Any comorbidity, (n%) | 2463 (76.3%) | 431 (72.3%) | 391 (75.2%) | 1229 (81.7%) | 412 (68.0%) | 0.3 | <0.001*** | 0.10 |
| Renal failure | 1096 (34.0%) | 182 (30.5%) | 161 (31.0%) | 568 (37.7%) | 185 (30.5%) | 0.9 | 0.002** | >0.9 |
| Atrial fibrillation | 917 (28.4%) | 137 (23.0%) | 149 (28.7%) | 495 (32.9%) | 136 (22.4%) | 0.031* | <0.001*** | 0.8 |
| Type II diabetes | 790 (24.5%) | 140 (23.5%) | 121 (23.3%) | 412 (27.4%) | 117 (19.3%) | >0.9 | 0.068 | 0.077 |
| COPD | 767 (23.8%) | 125 (21.0%) | 111 (21.3%) | 410 (27.2%) | 121 (20.0%) | 0.9 | 0.003** | 0.7 |
| Asthma | 499 (15.5%) | 84 (14.1%) | 77 (14.8%) | 269 (17.9%) | 69 (11.4%) | 0.7 | 0.037* | 0.2 |
| Obesity | 461 (14.3%) | 76 (12.8%) | 64 (12.3%) | 240 (15.9%) | 81 (13.4%) | 0.8 | 0.065 | 0.8 |
| Carpal tunnel syndrome | 157 (4.9%) | 21 (3.5%) | 25 (4.8%) | 88 (5.8%) | 23 (3.8%) | 0.3 | 0.030* | 0.8 |
| Cardiac implantable electronic device (pacemaker or ICD) insertion | 84 (2.6%) | 15 (2.5%) | 12 (2.3%) | 35 (2.3%) | 22 (3.6%) | 0.8 | 0.8 | 0.3 |
| Atrioventricular node block | 20 (0.6%) | <6 | <6 | 9 (0.6%) | <6 | 0.5 | 0.6 | 0.8 |
| Neuropathy | 16 (0.5%) | <6 | <6 | <6 | <6 | 0.5 | 0.4 | >0.9 |
| Lumbar spinal stenosis | 9 (0.3%) | <6 | <6 | <6 | <6 | >0.9 | 0.6 | >0.9 |
| Rupture of bicep tendon | <6 | <6 | — | <6 | — | 0.3 | 0.072 | 0.12 |
BMI, body mass index; COPD, chronic obstructive pulmonary disease; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; ICD, implantable cardioverter defibrillator; IQR, interquartile range; SD, standard deviation.
Data are numbers (%) unless stated otherwise. ‘<6’ = small number suppression if counts ≤ 5. Pairwise P‐values for comparisons of heart failure phenotypes to the HFrEF reference group.
Continuous: mean (SD), median (IQR); categorical: frequency (n%); DCS, double count suppression (the total count does not reveal the suppressed value).
Pearson's χ 2 test; Wilcoxon's rank sum test; Fisher's exact test.
P < 0.05.
P < 0.01.
P < 0.001.
Functional status, as measured by NYHA class, was recorded for 181 patients (5.6%) at baseline. Of these, 90 (2.8%) patients were classified as NYHA class II. Breathlessness was documented as a symptom in 1280 patients (39.7%); HFpEF was 44.4% vs. HFrEF was 37.1% (P = 0.003). The MRC Breathlessness Scale, ranging from 1 to 5, was used to assess symptoms in 642 patients (19.9%), with the most frequent grade assigned being grade 3 (5.9%).
CV and non‐CV comorbidities of interest are summarized in Table 1 . In this study, the most common comorbidities were CKD (34.0%), AF (28.4%), and diabetes (24.5%). The proportion of patients with at least one comorbidity of interest was significantly higher (81.7% vs. 72.3%; P < 0.001) among the HFpEF group compared with the HFrEF. This is driven by a significantly higher proportion of HFpEF patients affected by CKD, AF, COPD, asthma, and carpal tunnel syndrome in comparison with the HFrEF group.
The median body mass index (BMI) was 27.3 kg/m2, with 14.3% classified as obese (BMI > 30 kg/m2). Multimorbidity, defined as the coexistence of two or more chronic diseases, affected 42.2% of the population, more so in HFpEF (49.0%) than in HFrEF (36.6%). A total of 23.7% of the HF population had no reported comorbidities.
Measurements and treatments
The measurement of NP and biomarkers was relatively low, with 18.3% of the HF population receiving BNP or NT‐proBNP at baseline and 9.5% of patients during the follow‐up period (Table 2 ). Troponin I was measured at baseline for 22.4% of the cohort (25.4% post‐HF), and troponin T for 29.3% (19.2% post‐HF). The use of other blood tests was high, with 97.7%, 96.9%, and 98.1% of patients receiving at least one baseline measurement of eGFR, bilirubin, and creatinine, respectively.
Table 2.
Echocardiogram and natriuretic peptide (brain natriuretic peptide or N‐terminal pro‐B‐type natriuretic peptide) testing
| Characteristic | Heart failure population | P‐values HFrEF | ||||||
|---|---|---|---|---|---|---|---|---|
| Overall, N = 3227 a | HFrEF, N = 596 a | HFmrEF, N = 520 a | HFpEF, N = 1505 a | Unknown, N = 606 a | vs. HFmrEF b | vs. HFpEF b | vs. unknown b | |
| Natriuretic peptide (NP) | ||||||||
| Prior to HF diagnosis | 535 (16.6%) | 79 (13.3%) | 76 (14.6%) | 332 (22.1%) | 48 (7.9%) | 0.5 | <0.001*** | 0.003** |
| Baseline | 592 (18.3%) | 88 (14.8%) | 81 (15.6%) | 370 (24.6%) | 53 (8.7%) | 0.7 | <0.001*** | 0.001** |
| Post‐HF diagnosis | 305 (9.5%) | 39 (6.5%) | 37 (7.1%) | 182 (12.1%) | 47 (7.8%) | 0.7 | <0.001*** | 0.4 |
| ECHO | ||||||||
| Baseline | 3011 (93.3%) | 596 (100.0%) | 520 (100.0%) | 1505 (100.0%) | 390 (64.4%) | <0.001*** | ||
| Baseline + 6 months | 3035 (94.1%) | 596 (100.0%) | 520 (100.0%) | 1505 (100.0%) | 414 (68.3%) | <0.001*** | ||
| Baseline + 12 months | 3055 (94.7%) | 596 (100.0%) | 520 (100.0%) | 1505 (100.0%) | 434 (71.6%) | <0.001*** | ||
ECHO, echocardiography; HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.
Data are the number of patients with evidence of at least one measurement/procedure (%), unless stated otherwise. ‘<6’ = small number suppression if counts ≤ 5. Pairwise P‐values for comparisons of HF phenotypes to the HFrEF reference group.
Frequency (n%); DCS, double count suppression (the total count does not reveal the suppressed value).
Fisher's exact test.
P < 0.01.
P < 0.001.
Evidence of baseline echocardiography was available for 3011 (93.3%) patients, which increased to 94.1% and 94.7% at 6 and 12 month post‐diagnosis (Table 2 ).
Temporal changes in baseline ECHO and NP assessments were analysed over 10 calendar years (2010–19) (Figure 1 ). Baseline ECHO assessments consistently exceeded 90% throughout this period. In contrast, the NP measurement was first detected in 2012 and increased between 2012 and 2015, but then stabilized in 2016, remaining slightly above 30% until 2019.
Figure 1.
Baseline echocardiographic (ECHO) and natriuretic peptide (NP; brain natriuretic peptide or N‐terminal pro‐B‐type natriuretic peptide) assessments presented as a proportion of heart failure (HF) patients with the first HF diagnosis within the study period (10 calendar years: 2010–19).
Guideline‐directed medications for HF were frequently prescribed in the study population, with 89.1% of patients receiving at least one prescription of a specified HF medication within ±6 months of diagnosis. The most common class of medications was loop diuretics, prescribed to 2019 (62.6%) patients (66.8% HFpEF and 68.0% HFrEF; P = 0.6). For 3.9% of the study population, a diuretic was the sole HF‐related medication prescribed within 6 months of diagnosis. ACE‐Is were prescribed to 1910 (59.2%) patients (53.3% in HFpEF and 72.8% in HFrEF; P < 0.001); 420 (13.0%) received an angiotensin II receptor antagonist (12.9% in HFpEF and 14.6% in HFrEF; P = 0.3). Beta‐blockers were prescribed to 1927 (59.7%) patients (55.4% in HFpEF and 75.8% in HFrEF; P < 0.001). Aldosterone receptor antagonists (MRAs) were prescribed to 698 (21.6%) of the HF cohort (15.5% in HFpEF and 36.9% in HFrEF; P < 0.001). ARNI was prescribed to fewer than six patients (Table 3 ).
Table 3.
Patients with at least one prescription of medication at heart failure diagnosis ±6 months
| Characteristic | Heart failure population | P‐values HFrEF | ||||||
|---|---|---|---|---|---|---|---|---|
| Overall, N = 3227 a | HFrEF, N = 596 a | HFmrEF, N = 520 a | HFpEF, N = 1505 a | Unknown, N = 606 a | vs. HFmrEF b | vs. HFpEF b | vs. unknown b | |
| Medication class | ||||||||
| Any medication class, (n%) | 2875 (89.1%) | 552 (92.6%) | 472 (90.8%) | 1364 (90.6%) | 487 (80.4%) | 0.3 | 0.15 | <0.001*** |
| Angiotensin‐converting enzyme inhibitor | 1910 (59.2%) | 434 (72.8%) | 352 (67.7%) | 802 (53.3%) | 322 (53.1%) | 0.061 | <0.001*** | <0.001*** |
| Angiotensin II antagonists | 420 (13.0%) | 87 (14.6%) | 69 (13.3%) | 194 (12.9%) | 70 (11.6%) | 0.5 | 0.3 | 0.12 |
| Neprilysin inhibitors/angiotensin II antagonists | <6 | <6 | — | <6 | — | 0.5 | 0.2 | 0.2 |
| Beta‐blockers | 1927 (59.7%) | 452 (75.8%) | 348 (66.9%) | 834 (55.4%) | 293 (48.3%) | <0.001*** | <0.001*** | <0.001*** |
| Alpha–beta‐blocker (carvedilol) | 45 (1.4%) | 17 (2.9%) | 7 (1.3%) | 9 (0.6%) | 12 (2.0%) | 0.084 | <0.001*** | 0.3 |
| Aldosterone antagonists | 698 (21.6%) | 220 (36.9%) | 107 (20.6%) | 233 (15.5%) | 138 (22.8%) | <0.001*** | <0.001*** | <0.001*** |
| Diuretics, loop | 2019 (62.6%) | 405 (68.0%) | 297 (57.1%) | 1005 (66.8%) | 312 (51.5%) | <0.001*** | 0.6 | <0.001*** |
| Diuretics, potassium‐sparing | 26 (0.8%) | <6 | <6 | 19 (1.3%) | <6 | >0.9 | 0.020* | 0.2 |
| Diuretics, thiazide | 377 (11.7%) | 53 (8.9%) | 68 (13.1%) | 203 (13.5%) | 53 (8.7%) | 0.025* | 0.004** | >0.9 |
| Diuretic combinations | 61 (1.9%) | <6 | <6 | 43 (2.9%) | 9 (1.5%) | 0.7 | 0.002** | 0.2 |
| Sinus node IF inhibitors | 177 (5.5%) | 55 (9.2%) | 41 (7.9%) | 48 (3.2%) | 33 (5.4%) | 0.4 | <0.001*** | 0.012* |
| Anticoagulant | 684 (21.2%) | 113 (19.0%) | 124 (23.8%) | 359 (23.9%) | 88 (14.5%) | 0.046* | 0.015* | 0.039* |
| Cardiac glycosides | 394 (12.2%) | 70 (11.7%) | 59 (11.3%) | 206 (13.7%) | 59 (9.7%) | 0.8 | 0.2 | 0.3 |
| Nitrates | 578 (17.9%) | 113 (19.0%) | 119 (22.9%) | 250 (16.6%) | 96 (15.8%) | 0.11 | 0.2 | 0.2 |
| Xanthines | 38 (1.2%) | DCS | <6 | 21 (1.4%) | DCS | 0.042* | >0.9 | >0.9 |
| Hydralazine hydrochloride or isosorbide dinitrate | 7 (0.2%) | <6 | — | <6 | <6 | 0.3 | 0.4 | 0.4 |
HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.
Data are numbers (%) unless stated otherwise. ‘<6’ = small number suppression if counts ≤ 5. Pairwise P‐values for comparisons of heart failure phenotypes to the HFrEF reference group.
Frequency (n%); DCS, double count suppression (the total count does not reveal the suppressed value).
Pearson's χ 2 test; Fisher's exact test.
P < 0.05.
P < 0.01.
P < 0.001.
Healthcare resource utilization
The utilization of secondary care resources was generally high in the study population. Prior to their HF diagnosis, 65.6% of patients had at least one A&E visit for any reason, and this increased to 70.9% after diagnosis. The annualized A&E visit rate post‐diagnosis was, on average, 22.5% higher than the pre‐diagnosis rate. Similarly, 66.6% of the study population experienced at least one inpatient hospital admission for any reason before their HF diagnosis, and this increased to 73.6% post‐diagnosis. The annualized all‐cause admission rate post‐diagnosis was, on average, 28.1% higher than the pre‐diagnosis rate. When considering only CV‐related admissions, 34.2% of patients were admitted before their HF diagnosis, and this increased to 41.2% after the diagnosis, with an average annualized rate increase of 21.0%.
Cardiology‐related outpatient clinics included cardiology (45%), HF (17%), and cardio‐respiratory investigations (38%). In total, 65.1% of the HF cohort had utilized a cardiology‐led outpatient clinic prior to the diagnosis, increasing to 77.1% post‐diagnosis. Specifically, within the first 12 months following HF diagnosis, overall, 72.5% of patients had at least one specialist‐led outpatient visit; for HFpEF, this was 78.7% of patients.
CR services were accessed by 7.7% before HF diagnosis, increasing to 18.7% in the post‐diagnosis follow‐up period.
A primary care referral to a cardiologist was documented for 15.7% of the population before HF diagnosis. However, secondary care outpatient clinic data reported that at least 65.0% of the HF population had accessed a cardiology‐related clinic at some point before the diagnosis. This suggests that referrals to the specialist likely originated from the secondary care setting.
Survival analysis
The crude 5 year survival rate for HFrEF was 42.3% [95% CI (38.0–47.2%)], 45.5% [95% CI (41.0–50.4%)] for HFmrEF, and 37.8% [95% CI (35.2–40.7%)] for HFpEF (Table 4 ). After adjusting for age and sex, the survival rate for HFrEF was 38.0% [95% CI (33.7–42.8%)], 44.2% [95% CI (39.7–49.3%)] for HFmrEF, and 40.1% [95% CI (37.4–43.1%)] for HFpEF (Table 5 ). Among these HF groups, HFmrEF patients had the highest survival, with a median survival estimate of 4.2 years [95% CI (3.4–4.9)]. After adjusting for confounders, median survival decreased for HFrEF. For HFmrEF and the unassigned (unknown) group, there were smaller reductions in median overall survival. In contrast, for HFpEF, survival time increased following age and sex adjustment (Figure 2 ).
Table 4.
Kaplan–Meier life table: summary of survival curves before adjustment for age and sex, estimated for 5 years post‐heart failure diagnosis
| Characteristic | Survival (95% CI) | |||||
|---|---|---|---|---|---|---|
| Median survival (years) | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 | |
| HF phenotype | ||||||
| HFrEF | 3.9 (3.4–4.6) | 76.8% (73.5–80.3%) | 66.3% (62.6–70.3%) | 57.6% (53.6–61.9%) | 49.9% (45.6–54.5%) | 42.3% (38.0–47.2%) |
| HFmrEF | 4.4 (3.7 to ‐) | 79.8% (76.4–83.3%) | 69.0% (65.1–73.1%) | 58.2% (54.0–62.8%) | 52.4% (48.1–57.1%) | 45.5% (41.0–50.4%) |
| HFpEF | 3.1 (2.9–3.5) | 74.3% (72.1–76.6%) | 61.6% (59.2–64.2%) | 51.1% (48.5–53.8%) | 44.7% (42.1–47.5%) | 37.8% (35.2–40.7%) |
| Unknown | 3.8 (3.4–4.4) | 71.8% (68.3–75.4%) | 63.0% (59.2–67.0%) | 56.9% (53.0–61.1%) | 48.1% (44.1–52.5%) | 42.3% (38.2–46.8%) |
CI, confidence interval; HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.
Table 5.
Kaplan–Meier life table: summary of survival curves adjusted for age and sex, estimated for 5 years post‐heart failure diagnosis
| Characteristic | Survival (95% CI) | |||||
|---|---|---|---|---|---|---|
| Median survival (years) | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 | |
| HF phenotype | ||||||
| HFrEF | 3.4 (2.8–4.0) | 73.8% (70.1–77.7%) | 62.0% (58.0–66.4%) | 53.2% (49.0–57.8%) | 45.5% (41.2–50.2%) | 38.0% (33.7–42.8%) |
| HFmrEF | 4.2 (3.4–4.9) | 78.3% (74.6–82.2%) | 67.3% (63.2–71.8%) | 56.5% (52.1–61.3%) | 51.0% (46.5–55.9%) | 44.2% (39.7–49.3%) |
| HFpEF | 3.4 (3.1–3.9) | 75.4% (73.2–77.6%) | 63.2% (60.7–65.7%) | 53.0% (50.4–55.8%) | 46.8% (44.1–49.6%) | 40.1% (37.4–43.1%) |
| Unknown | 3.7 (3.2–4.2) | 71.1% (67.5–74.9%) | 61.8% (58.0–65.9%) | 55.7% (51.7–60.0%) | 46.7% (42.6–51.2%) | 40.7% (36.6–45.2%) |
CI, confidence interval; HF, heart failure; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.
Figure 2.
Kaplan–Meier survival curves for the heart failure groups: solid lines, using marginal adjustment for age and sex; dotted lines, simple risk model with no adjustment for confounders. The correction for confounders reduces median overall survival in the case of heart failure with reduced ejection fraction (HFrEF) and, by smaller fractions, in heart failure with mildly reduced ejection fraction (HFmrEF) and unknown groups. The opposite is observed for heart failure with preserved ejection fraction (HFpEF), where survival time has increased following age and sex adjustment.
The HFmrEF group had significantly lower 5 year all‐cause mortality compared with the HFrEF group [HR 0.82, 95% CI (0.69–0.97); P < 0.017]; specifically, the relative, age‐ and sex‐adjusted risk of death for HFmrEF vs. HFrEF was reduced by 18%. The HFpEF group had numerically lower 5 year all‐cause mortality than the HFrEF group [HR 0.89, 95% CI (0.78–1.02); P < 0.091], but this result was not statistically significant.
Ex‐smoker status was associated with a significant 16% reduction in the relative risk of death compared with current smokers. Non‐smoker status was associated with a 24% reduction in the relative risk of death compared with current smokers.
When comorbidities were included in the Cox model, the presence of COPD was independently related to a 50% higher relative risk of death [HR 1.50, 95% CI (1.34–1.68); P < 0.001], renal failure was associated with a 23% increase in the relative risk of death [HR 1.23, 95% CI (1.11–1.37); P < 0.001], and type II diabetes was associated with an 18% increase in the relative risk of death [HR 1.18, 95% CI (1.05–1.32); P = 0.004] (Figure 3 ).
Figure 3.
The hazard ratio for the risk of death in the cohort of heart failure (HF) patients. CI, confidence interval; COPD, chronic obstructive pulmonary disease; HFmrEF, heart failure with mildly reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction.
Discussion
Despite the high morbidity and mortality that are associated with HF, several findings indicate a significant deviation from the core aspects of clinical practice guidelines for HF diagnosis and management in this UK urban population. The key findings were as follows: (i) the assessment of symptoms was highly variable, with only one‐quarter of patients being assessed using either the MRC dyspnoea scale or the NYHA functional class system; (ii) overall, the use of BNP/NT‐proBNP, which can aid in diagnosing HF, was used in less than one in five cases. While this increased to one in three by 2016, there was no further increase in the frequency of testing thereafter; (iii) although more than three‐quarters of patients were assessed in cardiology outpatient clinics after their HF diagnosis, fewer than one in five patients were referred for CR (or one in six for those with HFpEF), indicating under‐utilization of CR services; (iv) the 5 year survival rates for patients with HFrEF and HFpEF were just 40%, which is lower than what has been reported in previous studies. 19 These findings collectively point to gaps and variations in clinical practice when it comes to HF diagnosis, management, and post‐diagnosis care, and they underscore the need for a more consistent and guideline‐based approach to improve patient outcomes.
It is encouraging that the vast majority of patients in this analysis received an echocardiogram at baseline. This early assessment allowed for the categorization of patients based on their EF. Despite the significant presence of HFpEF in the population (46.6%), there appears to be an unequal provision of services once the diagnosis is established. Specifically, in this analysis, 78.7% of patients with HFpEF received dedicated cardiology follow‐up in the 12 months following their diagnosis. In contrast, a higher proportion of patients with HFrEF and HFmrEF received dedicated cardiology follow‐up: 88.6% and 85.0%, respectively. These findings highlight a potential disparity in the provision of specialized care and follow‐up for patients with HFpEF compared with those with other EF categories.
Consistent with published data, this analysis reveals several important trends related to patients with HFpEF. HFpEF patients in this analysis were older than those with HFrEF, with a high prevalence of comorbidities. HFpEF patients had the highest 5 year crude all‐cause mortality rate among the different HF categories, highlighting the poor prognosis of this HF category. Despite their high mortality rates, HFpEF patients had the lowest rate of annualized cardiology appointments post‐diagnosis, suggesting a potential gap in specialized follow‐up care for HFpEF patients. Despite the fewer evidence‐based treatment options available for people with HFpEF, specialist input aimed at trying to avoid or manage periods of decompensation in patients with HFpEF is important. Such input can also help in managing comorbidities alongside HF, potentially improving outcomes for these patients. However, many HF services are under‐resourced, leading to difficult decisions about prioritizing treatments. In many cases, services may prioritize patients with HFrEF as there are more evidence‐based treatment options available for this group.
Diagnosis of heart failure
Prompt and accurate diagnosis, supported by biomarkers, dedicated cardiac imaging, and specialist input, are mission critical in HF, as delays in diagnosis and treatment are associated with poor outcomes. Although often overlooked, seemingly small deviations from evidence‐based practice, both at initial diagnosis and subsequently, can significantly impact outcomes in both the short and long terms. UK‐level data have reported that 5 year mortality rates for HF are worse than those of some of the most common cancers. 27 In view of these considerations, NICE clinical practice guidelines specifically recommend that patients with signs and symptoms of suspected HF should receive NT‐proBNP testing and be reviewed by a specialist within 2 weeks for urgent cases and within 6 weeks for those with NT‐proBNP levels of 400–2000 pg/mL. This is intended to ensure that patients receive timely initiation of targeted interventions and disease‐modifying treatment, proven to improve patient outcomes soon after their initiation. However, patients in the UK are presently waiting up to 20 weeks for specialist assessment, and this is delaying the initiation of guideline‐directed therapies. 10
In this population of patients with newly diagnosed HF, it was observed that baseline troponin and bilirubin measurements were conducted more frequently than NPs, despite not being recommended in clinical practice guidelines for HF diagnosis. The reasons for this deviation from guidelines are not explored in this retrospective analysis.
Renal profile measurements were conducted in almost all patients at baseline. Combining NP measurements with renal profiles could be a practical approach for strategic testing practices. Additionally, linking this combination to primary care pathway alerting tools when there is an entry of suspected HF in patient records could help in early detection and intervention for HF cases.
Treatment for heart failure
Clinical practice guidelines recommend specific medications as first‐line therapy for HFrEF patients, including ACE‐I/ARNI or angiotensin receptor blocker in combination with a beta‐blocker and an MRA. These medications have been shown to improve outcomes and reduce the risk of hospitalization and mortality in HFrEF patients. However, the results of this analysis suggest that a proportion of patients with HFrEF may not be receiving these recommended therapies as part of their management, despite support from performance‐related payments under the ‘Quality and Outcomes Framework’ incentivization scheme. 28 This lack of adherence to guidelines could potentially impact the quality of care and patient outcomes. Addressing the barriers to guideline adherence, improving awareness among healthcare providers, and optimizing the implementation of quality improvement initiatives may be necessary to ensure that HF patients receive evidence‐based therapies that can improve their prognosis and quality of life.
Although the evidence indicates that CR is associated with improved patient outcomes, HFpEF patients had the lowest rates of referral for this service. Although there are many reasons why this may be the case, including older age and the presence of multiple comorbidities, it is important to consider alternative approaches to address this issue, such as home‐based exercise programmes.
Outcomes
The 5 year all‐cause mortality rate in this urban HF population was considerable. Previous studies report 5 year survival rates for HF in the region of 50%. 29 Compared with previous studies reporting a 50% survival rate, this study found lower age‐ and sex‐adjusted survival rates of 38% for HFrEF, 40% for HFpEF, and 44% for HFmrEF. This is an important observation for many reasons. First, it is well established that certain areas in the north of England have poorer health outcomes than the more affluent suburbs of southern England. Second, lower socio‐economic status is associated with poorer health outcomes. Although the Salford, Greater Manchester, population is broadly representative of an urban UK population in terms of sex and age distribution, the area is relatively deprived, with a larger proportion of people living with disabilities as compared with the rest of the country (as per 2011 census data). The prevalence of chronic conditions closely follows the trends reported by the North West region of England. 30
Finally, it is easy to overlook the importance of those patients for whom an EF cannot be determined. The 5 year adjusted survival rate for these patients was just 41%. While this might reflect difficulties initiating disease‐modifying treatments in the absence of an EF measurement, they represent an important subgroup that requires closer attention to better understand the reasons behind their poor survival rates. While echocardiograms were performed in the majority of patients, one in six cases lacked sufficient evidence in the electronic patient record to categorize the type of HF (EF). This likely mirrors real‐world practices in other geographic areas.
Clinical perspectives
In 2010, the NICE guidelines for the diagnosis and management of HF were released. Around the same time, a legal right was established for National Health Service (NHS) patients with suspected cancer to be seen by a hospital doctor within 2 weeks of a GP referral, known as the Health Service Circular 205 (HSC 205) pathway. Compliance with this 2 week specialist referral pathway is closely monitored, and resources are allocated to support it.
In contrast, waiting times for HF specialist assessments are not nationally monitored, and little is known about the delays in this process. To address this issue, there is a need to collect data to understand the extent of the problem and identify the specific bottlenecks leading to delays in specialist assessment. Once such data are collected, it will be possible to determine where additional resources are required and in what capacity to improve the efficiency and effectiveness of HF specialist assessments.
Limitations
The limitations of this retrospective analysis include those commonly associated with the use of data sourced from EMRs, including potential data misclassification and missing information. Electrocardiogram (ECG) testing was not examined. Despite NYHA class being an established tool for treatment guidance in selected patients, it was infrequently recorded in primary care records, suggesting a disparity in how HF symptoms are documented in primary vs. secondary care. Whether this makes a material difference was not examined in the current study. Medication dosages and compliance were not studied, and the analysis predates the use of certain HF medications. Although of possible interest, levels of NPs and other blood results were not studied in the current examination but could be explored in future analyses. Finally, the current analysis was designed to examine whole‐system pathway management for HF, from initial presentation and diagnosis through follow‐up to 5 years, capturing mortality and hospitalizations that occur in the intervening period. We wished to highlight missed opportunities to improve and standardize how we assess patients across the whole EF spectrum. Although of potential interest, it was not our intention to examine differences in management between GP surgeries and the university teaching hospital.
Conclusions
Despite the recognized limitations of a retrospective study design, this study highlights the need for a renewed focus on HF patient pathways, especially for those with HFpEF. Improving diagnosis, access to specialist care, and comprehensive management are mission critical to improving care and patient outcomes. Further research is suggested to explore these observations in closer detail.
Conflict of interest
Pfizer Ltd paid NWEH for study activities, including data analysis. M.L.E. and E.R.S. are employees of Pfizer Ltd and hold stock and/or stock options. J.B. was employed by Pfizer Ltd at the time of the study's conduct and analysis and holds stock. O.S. was an employee of Pfizer Ltd at the time of the study's design and concept. F.Z.A. has previously received a research grant funded by Medtronic. F.Z.A. has received consultancy fees from AstraZeneca, Medtronic, Pfizer, Pharmacosmos, Servier, and Vifor. Funding was received from Pfizer for the conduct of the study. No funding was received in relation to this manuscript. S.M. is an employee of NWEH, and B.W. was an employee of NWEH at the time of study conduct and manuscript submission. NWEH was a paid consultant to Pfizer in connection with the development of this manuscript, the statistical analysis, and the final quality checks. No honoraria or payments were made for authorship.
Funding
This study was sponsored by Pfizer Ltd.
Acknowledgements
The authors would like to thank Sophie Clarke for her contribution to the protocol development and Vincenzo Leo for his contribution to the interpretation of the data. We are grateful to the SIR Governance Board for reviewing the study proposal and for permission to use the SIR data to conduct the study.
Migas, S. , Ellis, M. L. , Wrona, B. , Rivero Sanz, E. , Brownrigg, J. , Strauss, O. , and Ahmed, F. Z. (2024) Missed opportunities in heart failure diagnosis and management: study of an urban UK population. ESC Heart Failure, 11: 2200–2213. 10.1002/ehf2.14766.
References
- 1. National Institute for Health and Care Excellence . Chronic heart failure in adults: Diagnosis and management. NICE guideline [NG106]. https://www.nice.org.uk/guidance/ng106. Accessed 23 August 2023 [PubMed]
- 2. Lloyd‐Jones DM, Larson MG, Leip EP, Beiser A, D'Agostino RB, Kannel WB, et al. Lifetime risk for developing congestive heart failure. Circulation 2002;106:3068‐3072. doi: 10.1161/01.cir.0000039105.49749.6f [DOI] [PubMed] [Google Scholar]
- 3. Savarese G, Lund LH. Global public health burden of heart failure. Card Fail Rev 2017;3:7‐11. doi: 10.15420/cfr.2016:25:2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Richard Hobbs FD. Clinical burden and health service challenges of chronic heart failure. Br J Gen Pract 2010;60:611‐615. doi: 10.3399/bjgp10X515133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Squire I, Glover J, Corp J, et al. Impact of HF on HRQoL in patients and their caregivers in England: Results from the ASSESS study. Br J Cardiol 2017;24:30‐34. [Google Scholar]
- 6. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Colvin MM, 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. J Am Coll Cardiol 2017;70:776‐803. doi: 10.1016/j.jacc.2017.04.025 [DOI] [PubMed] [Google Scholar]
- 7. Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: A report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2022;145:e895‐e1032. doi: 10.1161/CIR.0000000000001063 [DOI] [PubMed] [Google Scholar]
- 8. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: Developed by the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur Heart J 2021;42:3599‐3726. doi: 10.1093/eurheartj/ehab368 [DOI] [PubMed] [Google Scholar]
- 9. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2023 focused update of the 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: Developed by the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur Heart J 2023;44:3627‐3639. doi: 10.1093/eurheartj/ehad195 [DOI] [PubMed] [Google Scholar]
- 10. Roche Diagnostics Limited . Heart failure: The hidden costs of late diagnosis. 2020. https://hfreport.roche.com/signs‐and‐symptoms. Accessed 23 August 2023
- 11. Konstam MA, Abboud FM. Ejection fraction: Misunderstood and overrated (changing the paradigm in categorizing heart failure). Circulation 2017;135:717‐719. doi: 10.1161/CIRCULATIONAHA.116.025795 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Savarese G, Orsini N, Hage C, Vedin O, Cosentino F, Rosano GMC, et al. Utilizing NT‐proBNP for eligibility and enrichment in trials in HFpEF, HFmrEF, and HFrEF. JACC Heart Fail 2018;6:246‐256. doi: 10.1016/j.jchf.2017.12.014 [DOI] [PubMed] [Google Scholar]
- 13. National Institute for Health and Care Excellence . Heart failure—Chronic: Management. https://cks.nice.org.uk/topics/heart‐failure‐chronic/management/. Accessed 23 August 2023
- 14. Edmonston DL, Wu J, Matsouaka RA, Yancy C, Heidenreich P, Pina IL, et al. Association of post‐discharge speciality outpatient visits with readmissions and mortality in high‐risk heart failure patients. Am Heart J 2019;212:101‐112. doi: 10.1016/j.ahj.2019.03.005 [DOI] [PubMed] [Google Scholar]
- 15. Morton G, Philip L, Gilpin T, Chan PE, Guha K, Kalra PR. Does specialist review for patients with suspected heart failure predict better outcomes? An observational study on the utility of compliance with NICE guidelines. BMJ Open 2018;8:e021856. doi: 10.1136/bmjopen-2018-021856 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bottle A, Kim D, Aylin P, Cowie MR, Majeed A, Hayhoe B. Routes to diagnosis of heart failure: Observational study using linked data in England. Heart 2018;104:600‐605. doi: 10.1136/heartjnl-2017-312183 [DOI] [PubMed] [Google Scholar]
- 17. Koudstaal S, Pujades‐Rodriguez M, Denaxas S, Gho JMIH, Shah AD, Yu N, et al. Prognostic burden of heart failure recorded in primary care, acute hospital admissions, or both: A population‐based linked electronic health record cohort study in 2.1 million people. Eur J Heart Fail 2017;19:1119‐1127. doi: 10.1002/ejhf.709 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Taylor CJ, Ryan R, Nichols L, Gale N, Hobbs FDR, Marshall T. Survival following a diagnosis of heart failure in primary care. Fam Pract 2017;34:161‐168. doi: 10.1093/fampra/cmw145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Taylor CJ, Ordóñez‐Mena JM, Roalfe AK, Lay‐Flurrie S, Jones NR, Marshall T, et al. Trends in survival after a diagnosis of heart failure in the United Kingdom 2000–2017: Population based cohort study. BMJ 2019;364:l223. doi: 10.1136/bmj.l223 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Team DCD . Cardiovascular Disease Outcomes Strategy, Best Practice Guidance. 2013.
- 21. National Cardiac Audit Programme . National Heart Failure Audit (NHFA) 2022 Summary Report. https://www.nicor.org.uk/wp‐content/uploads/2022/06/NHFA‐DOC‐2022‐FINAL.pdf. Accessed 23 August 2023
- 22. Elkhenini HF, Davis KJ, Stein ND, New JP, Delderfield MR, Gibson M, et al. Using an electronic medical record (EMR) to conduct clinical trials: Salford Lung Study feasibility. BMC Med Inform Decis Mak 2015;15:8. doi: 10.1186/s12911-015-0132-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. New JP, Leather D, Bakerly ND, McCrae J, Gibson JM. Putting patients in control of data from electronic health records. BMJ 2018;360:j5554. doi: 10.1136/bmj.j5554 [DOI] [PubMed] [Google Scholar]
- 24. NHS Salford Clinical Commissioning Group [Internet] . Salford Integrated Record: Sharing patient information locally. 2017. https://www.arc‐gm.nihr.ac.uk/media/wpweb/Salford‐Integrated‐Record‐Booklet.pdf. Accessed 23 August 2023
- 25. Stenton C. The MRC breathlessness scale. Occup Med (Lond) 2008;58:226‐227. doi: 10.1093/occmed/kqm162 [DOI] [PubMed] [Google Scholar]
- 26. Vergaro G, Castiglione V, Aimo A, Prontera C, Masotti S, Musetti V, et al. N‐terminal pro‐B‐type natriuretic peptide and high‐sensitivity troponin T hold diagnostic value in cardiac amyloidosis. Eur J Heart Fail 2023;25:335‐346. doi: 10.1002/ejhf.2769 Epub 2023 Jan 11. PMID: 36597836 [DOI] [PubMed] [Google Scholar]
- 27. Mamas MA, Sperrin M, Watson MC, Coutts A, Wilde K, Burton C, et al. Do patients have worse outcomes in heart failure than in cancer? A primary care‐based cohort study with 10‐year follow‐up in Scotland. Eur J Heart Fail 2017;19:1095‐1104. doi: 10.1002/ejhf.822 [DOI] [PubMed] [Google Scholar]
- 28. National Institute for Health and Care Excellence (NICE) . NICE Quality and Outcomes Framework indicator. Heart failure. Quality standard [QS9]. https://www.nice.org.uk/guidance/qs9. Accessed 23 August 2023
- 29. National Institute for Health and Care Excellence . Heart failure—Chronic: What is the prognosis? https://cks.nice.org.uk/topics/heart‐failure‐chronic/background‐information/prognosis/. Accessed 23 August 2023
- 30. Quality outcomes framework. Key indicators of representativeness of the Greater Manchester population to the UK. https://gpcontract.co.uk. Accessed 23 August 2023