Atrial fibrillation is associated with increased mortality and morbidity, including stroke, transient ischemic attack, myocardial infarction, and heart failure.1–3 Nearly 15% of strokes are attributed to atrial fibrillation, and atrial fibrillation-related strokes are more serious and lead to more disability and deaths.4 Nonvalvular atrial fibrillation (NVAF) is defined as atrial fibrillation, which occurs in the absence of both mechanical prosthetic heart valves and moderate-to-severe mitral stenosis (usually of rheumatic origin).5 It contributes to 77–91% of atrial fibrillation cases.6 In 2010, the estimated number of individuals with atrial fibrillation globally was 33.5 million, whereas the prevalence rate was 596.2 and 373.1 per 100 000 in male and female individuals, respectively.7 The global incidence of atrial fibrillation is reported to be 77.5 and 59.5 per 100 000 person-years for male and female individuals, respectively. Of particular note, atrial fibrillation incidence in the Middle East was 73.4 and 49.9 cases per 100 000 person-years for men and women, respectively.7 There are limited epidemiological data for NVAF patients in African and Middle Eastern countries.8,9
Oral anticoagulants are effective in reducing stroke/systemic embolism among NVAF patients. Vitamin K antagonists (VKAs) are the most commonly used treatments for reducing stroke/systemic embolism but many patients remain untreated because of drug and food interactions, the inconvenience of frequent international normalized ratio tests and increased major bleeding risk. In the last decade, non-VKA oral anticoagulants have been proven to be at least as effective and well tolerated as warfarin with fewer food and drug interactions and without a need for international normalized ratio monitoring.10 The Saudi Ministry of Health, European, United States, and Canadian guidelines advocate the use of non-VKA oral anticoagulants over warfarin for NVAF patients.11–14 However, there are limited available data on atrial fibrillation in Egypt, Lebanon, the Kingdom of Saudi Arabia (KSA), and the United Arab Emirates (UAE), as these regions were underrepresented in the non-VKA oral anticoagulant clinical trials.15–18 The characteristics of patients in this region may be different from patients in the West because of younger populations and a high prevalence of obesity, diabetes, and smoking.19 Due to the significant clinical and economic burden and the change in treatment in recent years, it is important to understand the patient characteristics and treatment patterns among newly diagnosed NVAF patients in African and Middle Eastern countries.
Recent improvements in atrial fibrillation diagnosis and management have prompted the initiation of various national, regional, and international registries. The Saudi Atrial Fibrillation Survey registry conducted in KSA provided mainly epidemiological and clinical data on patients (n = 400) with atrial fibrillation in the country.20 The Gulf Survey of Atrial Fibrillation Events was conducted in 2009–2010 and involved only one country included in the FLOW-AF: UAE (n = 459).21 The GARFIELD-atrial fibrillation (Global Anticoagulant Registry in the FIELD-Atrial Fibrillation) recruited patients from 35 countries over five sequential cohorts since 2008, including 453 patients from Egypt and 389 from UAE, but the country-specific data have not yet been published.22 Lastly, the GLORIA-atrial fibrillation (Global Registry on Long-Term Oral Antithrombotic Treatment in Patients with Atrial Fibrillation) included 597 patients (4% of all patients) from Lebanon, KSA, UAE, and South Africa; although regional data have been published,23 no country analyses have been done so far, possibly because of insufficient patient numbers per country.
Although different registries have included patients from these African and Middle Eastern regions, they are limited by sample size and lack of information on NVAF-related healthcare resource utilization and direct costs associated with the management of NVAF patients. The goal of the FLOW-AF registry will be to evaluate the characteristics, treatment patterns, and clinical and economic outcomes of patients newly diagnosed with NVAF in Egypt, Lebanon, KSA, and UAE. Furthermore, it will provide contemporary data for these countries in a continuously changing field.
Materials and methods
The FLOW-AF registry will be a multicountry, multicenter, prospective observational study aiming to enroll 1446 newly diagnosed NVAF patients at more than 20 sites across Egypt, Lebanon, KSA, and UAE, including a minimum of 427 patients each from Egypt and KSA and 296 patients each from Lebanon and UAE (Table 1). The centers to be included in the study will be those routinely providing clinical care to NVAF patients, and they will be geographically representative of the NVAF patients in these countries. The enrollment period will be 12 months, and patient data will be assessed prospectively for 12 months from their enrollment date.
Data will be collected through a secure online system using electronic Case Report Forms (Supplemental Appendix, http://links.lww.com/JCM/A270) and extracted at the enrollment (baseline), 6-month, and 12-month follow-up points. The same electronic Case Report Forms will be used across all investigator sites in the four countries. This will allow the data to be combined and pooled for the analysis. Data for 3 months before the baseline period will be extracted retrospectively through patients’ medical and pharmacy records. All evaluations and treatment decisions will be performed as per routine clinical practice.
The main aim of this registry will be to generate real-world evidence on patient characteristics and clinical management in newly diagnosed NVAF patients. The study will also evaluate the following patterns and outcomes in the overall population, across countries and by country:
- (1) primary clinical outcomes including stroke/systemic embolism, bleeding complications, and cause-specific/all-cause mortality, which are aligned with the randomized controlled trials;
- (2) secondary outcomes including treatment plan, drug regimen, dosing, frequency, duration of treatment, and reason for interruption and/or termination along with treatment switching at any point during follow-up;
- (3) international normalized ratio values and time in therapeutic range as well as the effect of these on the history of clinical outcomes among patients who received VKA treatment; and
- (4) NVAF-related healthcare resource utilization and direct costs associated with the management of NVAF patients.
To ensure that these four countries’ sites are representative of the clinical management provided to the NVAF patients, the type of practice (community hospital vs. university hospital; private vs. public hospital) and type of healthcare professional(s) involved in the care of atrial fibrillation patients (cardiologist, internist, general physician) will be considered. Approximately 1446 patients will be enrolled from 20 to 27 sites during the recruitment period, including a minimum sample of 427 patients residing in Egypt and KSA each, and of 296 patients residing in both Lebanon and UAE (Table 1).
During the study enrollment period (1 January 2019 to 31 December 2020), eligible patients will be invited by the site investigator to participate in the registry during their routine clinic visits (at the time of a confirmed diagnosis or shortly thereafter) until the target number of patients is enrolled (Table 2). During the study recruitment period, all the patients at the investigation site will be assessed for eligibility based on the criteria listed below:
- (1) Male or female patients aged at least 18 years, newly diagnosed with NVAF (within the enrollment period or ≤90 days before the baseline visit) for whom care has been initiated by the treating physician for the prevention of stroke/systemic embolism (Table 2). Information on the method of NVAF diagnosis including electrocardiogram, implantable devices, Holter monitor, online/smart applications, and others will be collected.
- (2) Patients will be required to sign an informed consent form to participate and enroll in the registry.
- (3) Patients with whom follow-up is unlikely or impossible (e.g. because of invalid residency permits or plans to leave the country) or enrolled in an interventional clinical trial will be excluded.
- (4) Furthermore, patients will be excluded from the registry if prescribed a VKA/non-VKA oral anticoagulant for a condition other than atrial fibrillation or diagnosed with a severe psychiatric illness, a cause for atrial fibrillation that is reversible [defined as noncardiac surgery, postcardiac surgery (atrial fibrillation within 3 months after surgery), hyperthyroidism, pulmonary embolism, pneumonia, or acute myocardial infarction], mechanical heart valves, or valve disease (Table 2).
- (5) Pregnant or breast-feeding women will also be excluded from the study.
A screening log of all eligible patients invited for the study will be maintained at the investigator site to record the characteristics.
Collection of data
Data will be collected through a secure online system using electronic Case Report Forms and will be maintained by the Contract Research Organization. Data collected at baseline include demographics (age, sex, height, weight, BMI, race), medical history (e.g. family history, history of stroke/systemic embolism/transient ischemic attack, myocardial infarction, peripheral arterial disease, history of bleeding, cardiovascular risk factors, baseline comorbidities), risk factors (e.g. smoking), vital signs, and laboratory assessments (e.g. estimated glomerular filtration rate) (Fig. 1).
Follow-up data include:
- (1) clinical events (stroke/systemic embolism, transient ischemic attack, bleeding events, myocardial infarction, all-cause mortality, and cause of death);
- (2) atrial fibrillation treatments [oral anticoagulants/antiplatelet therapy (including acetylsalicylic acid and P2Y12 platelet inhibitors), dosage, and start and end dates];
- (3) laboratory assessments [complete blood count, creatine kinase-muscle/brain, international normalized ratio, Troponin (I or T), electrolytes, thyroid function test, B-type natriuretic peptide, liver function and renal function tests (creatinine clearance or estimated glomerular filtration rate)];
- (4) vital signs (blood pressure and heart rate);
- (5) adverse events (death because of any cause, or nonserious adverse reaction observed in NVAF patients exposed to any therapy throughout the study);
- (6) treatment patterns, including utilization rates, discontinuation, and switching;
- (7) concomitant treatments [lipid-lowering drugs, antihypertensive drugs (diuretics, beta-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers), nonsteroidal anti-inflammatory drugs, oral antidiabetic drugs, insulin, anxiolytics, antidepressants, antifungals, antibiotics, glucocorticoids, synthetic thyroid hormones, oral contraception, hormone replacement therapy, bisphosphonates, calcitonin, and selective estrogen receptor modulators];
- (8) international normalized ratio and time in therapeutic range levels for patients who receive VKAs; and
- (9) healthcare utilization and costs.
Regarding concomitant treatments, the generic name of the drug and the status of usage (including history of use of the drug) will be recorded during the baseline visit, and the start and end dates will be recorded during follow-up visits (Fig. 1).
The following outcomes will be monitored during the 6-month and 12-month follow-up visits:
- (1) Stroke, defined as hemorrhagic or ischemic stroke with time-to-event evaluated during the follow-up period
- (2) Transient ischemic attack
- (3) Systemic embolism
- (4) Bleeding events, with time-to-event evaluated during the follow-up period, will be classified as follows:
- (a) Major bleeding events
- (i) Acute clinically overt bleeding [accompanied by a decrease in hemoglobin of 2 g/dl or more over a 24-h period; a transfusion of at least 2 units of packed red blood cells; or bleeding that occurs in at least one of the critical sites including intracranial, intra-spinal, intraocular (e.g. not conjunctival), pericardial, intra-articular, intramuscular with compartment syndrome, or retroperitoneal]
- (ii) Fatal bleeding
- (b) Clinically relevant nonmajor bleeding events
- (i) Acute or sub-acute clinically overt bleeding that does not satisfy the criteria for major bleeding and that leads to:
- - hospital admission for bleeding
- - physician-guided medical or surgical treatment for bleeding
- - a change in antithrombotic therapy
- (c) Minor bleeding events
- (i) All acute, clinically overt bleeding events not meeting the criteria for either major bleeding or clinically relevant nonmajor bleeding will be classified as minor bleeding.
The following outcomes monitored during the baseline period, and at 6-month and 12-month follow-up visits:
- (1) Healthcare utilization:
- (a) Number and type of imaging exams including electrocardiogram, X-ray, echocardiogram, Holter monitor, stress test, other exams;
- (b) Number of inpatient admissions -- for each admission, the number of days of hospitalization (i.e. length of stay) captured;
- (c) Number of outpatient visits for each type of visit (emergency care visit, outpatient visit to general practitioner, outpatient visit to a specialist, other); and
- (d) Number and type of other healthcare resource use throughout the patients’ treatment journey (including but not limited to laboratory assessments and surgeries).
- (2) Healthcare costs:
- (a) The collection of data on the estimated cost of each unit of healthcare utilization data (specified above) done in a separate piece component of research to be conducted during the study period, using structured interviews at country level.
- (b) The following variables derived: economic burden associated with the management of patients with NVAF -- calculated by multiplying each unit of healthcare utilization by the correspondent medical care cost estimated.
No formal sample size computation was performed. The estimated prevalence of atrial fibrillation among adults aged at least 20 years is 0.95%. A higher precision level, corresponding to a wider 95% confidence interval, was considered for Lebanon and UAE as these countries have lower populations of atrial fibrillation patients but higher population density, which may lead to less heterogeneity in clinical management of the disease. Descriptive analysis will be conducted for all the data collected during baseline and follow-up. Outcomes will be analyzed by country and by groups based on treatment received, demographics, clinical characteristics, medical history, and risk factors. The incidence of each clinical outcome [stroke (hemorrhagic or ischemic), transient ischemic attack, systemic embolism, bleeding events, myocardial infarction, and all-cause mortality] will be reported as the number of events per patient-year. Healthcare utilization will be evaluated as the percentage of patients using each resource and will be reported on a per patient per month basis. The economic burden associated with the management of NVAF patients will be calculated by multiplying each unit of healthcare resource utilization by the corresponding medical care cost estimated and reported on a per patient per month basis.
The cost data will be collected during the study period using structured interviews at the country level. To calculate the total annual cost incurred, cost per patient will be divided by the number of days between diagnosis and follow-up visit at 12 months and multiplied by 365. It will be reported as follows: total healthcare costs, cost of imaging exams, cost of hospitalizations (inpatient admissions), cost of outpatient visits, and cost of treatment.
Comparisons will be made using parametric tests (χ2-test, Student's t test, Poisson regression for resource use, and a gamma model with log link for costs). Nonparametric tests will also be considered depending on the distributions of the observed resource use and costs. Fisher's exact test will be used for categorical variables to analyze small samples. The detailed methodology and statistical analysis of the data collected for this study will be detailed in a statistical analysis plan and finalized prior to the start of the analysis. Outcomes will be analyzed by country and by groups created based on treatment received, demographic and clinical characteristics, medical history, and risk factors.
This study provides an opportunity to evaluate routine clinical practice, including information on healthcare use and related costs among NVAF patients in Egypt, UAE, Lebanon, and KSA. There are limited data available on atrial fibrillation diagnosis and management in these countries.8,9 Firstly, there are multiple regions in the world that had their own registries. For example, ORBIT-AF registry in the USA, GARFIELD UK registry in the UK, AFNET in Germany, and EORP-AF for European Society of Cardiology members.24–27 Secondly, there are global registries like the GARFIELD-AF and the GLORIA-AF that are not specific to a region.22,23 However, they have yet not published any country-specific data, probably because of insufficient sample size. The management of NVAF in Middle Eastern countries is challenging because of the differences in geographical region, practice patterns, and patient demographics, compared with the clinical trials that have been conducted in North America, Europe, and Asia.28 The findings from this study will establish the patterns of care for newly diagnosed NVAF patients in these countries. The study will also evaluate the differences in management and clinical outcomes among patients in different countries and across different care settings and assess their adherence to clinical guidelines and areas of unmet needs.
It will also provide insight on international normalized ratio management among patients who receive VKAs, which has not been investigated previously in other real-world studies in these African and Middle Eastern regions. In addition, the FLOW-AF registry will have the ability to capture the real-world evidence on the reasons for discontinuation and switching, cause of death, and international normalized ratio values. The FLOW-AF registry will provide data on predictors of good time in therapeutic range as well as its relationship to clinical outcomes that will allow providers and healthcare authorities to perform robust and sophisticated analyses on the role of warfarin for the treatment of atrial fibrillation in this region of the world. It will also provide insights into atrial fibrillation management, healthcare utilization, and costs of NVAF patients in these four countries, which is not included in any of the existing registries. As treatment of atrial fibrillation is changing dramatically, there is a need to understand the current treatment patterns.
As an observational study, this has inherent design limitations, such as selection bias and unmeasured confounders. The study results may not be generalizable to patient populations outside these four countries because of possible differences in monitoring procedures and treatment standards across diverse sites and countries. However, the study has been designed to ensure that the study population is representative of NVAF patients in the four countries and that the participating sites are nationally representative of routine NVAF case management. The study permits descriptive analyses; therefore, it cannot formally compare the clinical effectiveness and safety of different medications, as in randomized control trials. Another limitation of using the registry data is the short follow-up time (∼12 months), which may preclude evaluation of the primary clinical outcomes over longer periods of time. Due to its observational nature, the data collected from routine medical practice are prone to bias and confounding. Major potential biases have been identified and taken into consideration in the study design. Due to its observational nature, the data collected from routine medical practice are prone to bias and confounding. Major potential biases have been identified and taken into consideration in the study design. However, unlike clinical trials, the quality and level of care for the patients in the registry are generally not expected to differ from standard patients. Patients in this registry were all enrolled during a clinic visit and did not include those who were admitted to a hospital or emergency department. However, the site selection for the registry will be conducted to include centers that provide clinical management representative of the NVAF management in the four countries. There is also a potential bias in the selection of the sites, as the participating centers may alter practice and quality of care patterns at the study site. However, prior studies have shown that the outcomes presented in registries are similar to those of regular clinical practice.29,30 Other NVAF registries also, have similar selection criteria to those in the current registry.22–24 Additional sensitivity analyses may be conducted to explore the potential impact of bias in the study findings. Finally, the enrolled patients might be using other facilities for atrial fibrillation-related complications and care. Hence, the data might not be wholly representative of the clinical and economic outcomes associated with NVAF management.
FLOW-AF will be the largest registry focusing on the Middle Eastern region. This study will provide valuable information on the uptake of non-VKA oral anticoagulants, treatment patterns, clinical outcomes, and healthcare utilization, and costs among newly diagnosed NVAF patients.
The authors would like to thank Chris Haddlesey of STATinMED Research for his assistance with editing and formatting the manuscript.
Funding: The FLOW-AF registry is funded by Pfizer, Inc.
Conflicts of interest
M.A.S. and M.K. have no conflicts to disclose. W.A.A. is on the Pfizer Advisory board for the study. J.S. is an employee of STATinMED Research, a paid consultant to the study sponsor. M.D.F., J.M., and S.A.K. are employees of Pfizer Inc., the study sponsor. R.D.L. has received research grants from Amgen, Bristol-Myers Squibb, GlaxoSmithKline, Pfizer, and Sanofi-Aventis; and personal fees from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline, Pfizer, and Portola. A.H. had received honoraria from Pfizer Inc. for giving talks and workshops.
1. Marini C, De Santis F, Sacco S, et al. Contribution of atrial fibrillation
to incidence and outcome of ischemic stroke: results from a population-based study. Stroke
2. Andersson T, Magnuson A, Bryngelsson IL, et al. All-cause mortality in 272,186 patients hospitalized with incident atrial fibrillation
1995-2008: a Swedish nationwide long-term case-control study. Eur Heart J
3. Steger C, Pratter A, Martinek-Bregel M, et al. Stroke patients with atrial fibrillation
have a worse prognosis than patients without: data from the Austrian Stroke registry. Eur Heart J
4. Jørgensen HS, Nakayama H, Reith J, Raaschou HO, Olsen TS. Acute stroke with atrial fibrillation
. The Copenhagen Stroke Study. Stroke
5. Heidbuchel H, Verhamme P, Alings M, et al. Updated European Heart Rhythm Association practical guide on the use of nonvitamin K antagonist anticoagulants in patients with nonvalvular atrial fibrillation
: executive summary. Eur Heart J
6. Boriani G, Cimaglia P, Fantecchi E, et al. Nonvalvular atrial fibrillation
: potential clinical implications of the heterogeneous definitions used in trials on new oral anticoagulants. J Cardiovasc Med (Hagerstown)
7. Chugh SS, Havmoeller R, Narayanan K, et al. Worldwide epidemiology of atrial fibrillation
: a Global Burden of Disease 2010 Study. Circulation
8. Johnston KM, Osenenko KM, Qatami L, et al. Healthcare resource utilization and costs in individuals with atrial fibrillation
in United Arab Emirates and Kingdom of Saudi Arabia: a retrospective cohort study. Int J Integr Med
9. El Kadri M, Bazargani N, Farghaly M, et al. Profiling clinical characteristics and treatment patterns among nonvalvular atrial fibrillation
patients: a real-world analysis in Dubai, United Arab Emirates. Open Med J
10. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation
: a meta-analysis of randomised trials. Lancet
11. Hersi A, Al Fagih A, Almusaad A, et al.
Clinical practice guideline on antithrombotic treatment of patients with nonvalvular atrial fibrillation
: The Saudi Center for EBHC Clinical Practice Guideline. The Saudi Arabia Ministry of Health website. Available at: https://www.moh.gov.sa/en/Search/Pages/Default.aspx?k=Clinical%20practice%20guideline%20on%20antithrombotic%20treatment%20of%20patients%20with%20nonvalvular%20atrial%20fibrillation
. (Accessed 15 January 2019)
12. Kirchhof P, Benussi S, Kotecha D, et al. 2016 ESC guidelines for the management of atrial fibrillation
developed in collaboration with EACTS. Eur Heart J
13. Andrade JG, Verma A, Mitchell LB, et al. CCS Atrial Fibrillation
Guidelines Committee. 2018 focused update of the Canadian Cardiovascular Society guidelines for the management of atrial fibrillation
. Can J Cardiol
14. January CT, Wann LS, Calkins H, et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation
. Heart Rhythm
15. Lopes RD, Alexander JH, Al-Khatib SM, et al. ARISTOTLE Investigators. Apixaban for reduction in stroke and other ThromboemboLic events in atrial fibrillation
(ARISTOTLE) trial: design and rationale. Am Heart J
16. Eikelboom JW, O’Donnell M, Yusuf S, et al. Rationale and design of AVERROES: apixaban versus acetylsalicylic acid to prevent stroke in atrial fibrillation
patients who have failed or are unsuitable for vitamin K antagonist treatment. Am Heart J
17. ROCKET AF Study Investigators. Rivaroxaban — once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation
: rationale and design of the ROCKET AF study. Am Heart J
18. Ezekowitz MD, Connolly SJ, Parekh A, et al. Rationale and design of RE-LY: randomized evaluation of long-term anticoagulant therapy, warfarin, compared with dabigatran. Am Heart J
19. Ahmed AM, Hersi A, Mashhoud W, et al. Cardiovascular risk factors burden in Saudi Arabia: the Africa Middle East Cardiovascular Epidemiological (ACE) study. J Saudi Heart Assoc
20. Hersi A, Abdul-Moneim M, Almous’ad A, Al-Samadi F, AlFagih A, Sweidan R. Saudi Atrial Fibrillation
Survey: national, observational, cross-sectional survey evaluating atrial fibrillation
management and the cardiovascular risk profile of patients with atrial fibrillation
21. Zubaid M, Rashed WA, Alsheikh-Ali AA, et al. Gulf Survey of Atrial Fibrillation
Events (Gulf SAFE) Investigators. Gulf Survey of Atrial Fibrillation
Events (Gulf SAFE): design and baseline characteristics of patients with atrial fibrillation
in the Arab Middle East. Circ Cardiovasc Qual Outcomes
22. Kakkar AK, Mueller I, Bassand JP, et al. International longitudinal registry of patients with atrial fibrillation
at risk of stroke: Global Anticoagulant Registry in the FIELD (GARFIELD). Am Heart J
2012; 163:13.e1–19. e1.
23. Mazurek M, Huisman MV, Rothman KJ, et al. GLORIA-AF Investigators. Regional differences in antithrombotic treatment for atrial fibrillation
: insights from the GLORIA-AF phase II registry. Thromb Haemost
24. Piccini JP, Fraulo ES, Ansell JE, et al. Outcomes registry for better informed treatment of atrial fibrillation
: rationale and design of ORBIT-AF. Am Heart J
25. Apenteng PN, Murray ET, Holder R, Hobbs FR, Fitzmaurice DA. UK GARFIELD Investigators and GARFIELD Steering Committee. An international longitudinal registry of patients with atrial fibrillation
at risk of stroke (GARFIELD): the UK protocol. BMC Cardiovasc Disord
26. Nabauer M, Gerth A, Limbourg T, et al. The Registry of the German Competence NETwork on Atrial Fibrillation
: patient characteristics and initial management. EP Europace
27. Lip GY, Laroche C, Dan GA, et al. A prospective survey in European Society of Cardiology member countries of atrial fibrillation
management: baseline results of EURObservational Research Programme Atrial Fibrillation
(EORP-AF) Pilot General Registry. EP Europace
28. Hersi AS, Alhebaishi YS, Hamoui O, et al. Practical perspectives on the use of nonvitamin K antagonist oral anticoagulants for stroke prevention in patients with nonvalvular atrial fibrillation
: a view from the Middle East and North Africa. J Saudi Heart Assoc
29. Heidenreich PA, Fonarow GC. Are registry hospitals different? A comparison of patients admitted to hospitals of a commercial heart failure registry with those from national and community cohorts. Am Heart J
30. Granger CB, Gersh BJ. Clinical trials and registries in cardiovascular disease: competitive or complementary?