Miek Smeets1, Séverine Henrard2, Bert Aertgeerts3, Filip Cools4, Stefan Janssens5, Bert Vaes2. 1. Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium. Electronic address: Miek.Smeets@kuleuven.be. 2. Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium; Institute of Health and Society, Université Catholique de Louvain, Brussel, Belgium. 3. Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium; CEBAM, Centre for Evidence Based Medicine, Belgium Cochrane Centre, Belgium. 4. CEBAM, Centre for Evidence Based Medicine, Belgium Cochrane Centre, Belgium; Department of Neonatology, Universitair Ziekenhuis Brussel, Vrije Universiteit Brussel, Brussel, Belgium. 5. Department of Cardiovascular Diseases, Universitair Ziekenhuis Gasthuisberg, Katholieke Universiteit Leuven, Leuven, Belgium.
Abstract
BACKGROUND: Identifying patients with heart failure (HF) in general practice is challenging. Our aim was to provide an overview of methods used to identify patients with HF in general practice and to assess their impact on patient characteristics. METHODS AND RESULTS: A systematic review was conducted using MEDLINE, EMBASE and CENTRAL. Taken together, 105 studies on HF in general practice were included, totalling 196,105 patients. Five main identification methods for HF were distinguished, including 1) echocardiographic assessments, 2) results of echocardiography in general practitioner (GP) charts, 3) GP judgment after chart review, 4) GP judgment of consecutive patients and 5) only chart review. Only 30% of studies used the results of echocardiography. Despite a large heterogeneity between studies the pooled data revealed a predominant phenotype of older women with hypertension rather than ischaemic heart disease. Linear regression analysis showed that the impact of the identification method on patient characteristics was limited. However, study design had a greater impact, with randomized-controlled trials (RCTs) including younger, male patients with ischaemic heart disease and higher HF drug prescription rates at baseline. CONCLUSION: Pooled data of 196,105 patients with HF confirmed a phenotype of older women with hypertension rather than ischaemic heart disease as the predominant HF population in general practice. The lack of a gold standard definition of HF introduced a large heterogeneity in identification methods with remarkably limited impact on patient characteristics. However, RCTs did include patients with a different phenotype, emphasizing the need to promote inclusion of real-world HF patients.
BACKGROUND: Identifying patients with heart failure (HF) in general practice is challenging. Our aim was to provide an overview of methods used to identify patients with HF in general practice and to assess their impact on patient characteristics. METHODS AND RESULTS: A systematic review was conducted using MEDLINE, EMBASE and CENTRAL. Taken together, 105 studies on HF in general practice were included, totalling 196,105 patients. Five main identification methods for HF were distinguished, including 1) echocardiographic assessments, 2) results of echocardiography in general practitioner (GP) charts, 3) GP judgment after chart review, 4) GP judgment of consecutive patients and 5) only chart review. Only 30% of studies used the results of echocardiography. Despite a large heterogeneity between studies the pooled data revealed a predominant phenotype of older women with hypertension rather than ischaemic heart disease. Linear regression analysis showed that the impact of the identification method on patient characteristics was limited. However, study design had a greater impact, with randomized-controlled trials (RCTs) including younger, male patients with ischaemic heart disease and higher HF drug prescription rates at baseline. CONCLUSION: Pooled data of 196,105 patients with HF confirmed a phenotype of older women with hypertension rather than ischaemic heart disease as the predominant HF population in general practice. The lack of a gold standard definition of HF introduced a large heterogeneity in identification methods with remarkably limited impact on patient characteristics. However, RCTs did include patients with a different phenotype, emphasizing the need to promote inclusion of real-world HF patients.
Authors: Miek Smeets; Bert Vaes; Pavlos Mamouris; Marjan Van Den Akker; Gijs Van Pottelbergh; Geert Goderis; Stefan Janssens; Bert Aertgeerts; Séverine Henrard Journal: BMJ Open Date: 2019-01-07 Impact factor: 2.692
Authors: Faye Forsyth; Jonathan Mant; Clare J Taylor; Fd Richard Hobbs; Carolyn A Chew-Graham; Thomas Blakeman; Emma Sowden; Aaron Long; Muhammad Zakir Hossain; Duncan Edwards; Christi Deaton Journal: BJGP Open Date: 2019-11-26