Opportunities for diabetes research using the Norwegian Cardiovascular Disease Registry

While Norway has many outstanding health registries that are used as a foundation for surveillance, quality control and epidemiologic research, it was recognized that a knowledge gap existed regarding the prevalence, incidence, and quality of health care delivery for cardiovascular diseases (CVD) which remains the primary cause of death in the country. Given broad agreement for the need for a registry by patient organizations, research environments and health authorities, the Norwegian Parliament decided to establish the Norwegian Cardiovascular Disease Registry (NCVDR) in March 2010 and then adopted new regulations in December 2011 which established the NCVDR. The registry consists of a core registry housed and maintained by the Norwegian Institute of Public Health (NIPH) that is supplemented with information from the medical quality registries housed and maintained by the hospital trusts. Given the burden of CVD among patients with diabetes and the unique challenges of medical care for CVD patients with diabetes, the registries also provides a new opportunity to advance diabetes research in Norway. INTRODUCTION AND DIABETES RESEARCH POTENTIAL OF THE NORWEGIAN CVD


INTRODUCTION AND DIABETES RESEARCH POTENTIAL OF THE NORWEGIAN CVD REGISTRY
While there is evidence of declining mortality for acute myocardial infarction (1), CVD remains the leading cause of mortality in Norway (1,2).Further, there are concerns that the declining trend in CVD mortality may not be sustainable, reflective of subpopulations, or of disease burden.For example, there is evidence of increasing obesity and diabetes prevalence rates (3), and of an increasing incidence of acute myocardial infarction (MI) in women in Tromsø (4).Further, hospitalization rates for CVD do not reflect the same downward trend as that observed for CVD mortality (5), and hospitalizations for acute MI for those over 79 years of age have not been declining (6).
While the burden of diabetes in Norway is not known, an estimated 3-5% of the Norwegian population has diabetes mellitus with prevalence peaking at 11-12% for those 80-89 years of age (3,(7)(8)(9).Diabetes mellitus is associated with a two-to four-fold increased risk of cardiovascular diseases including coronary heart disease, stroke, atrial fibrillation and flutter, and left ventricular hypertrophy (10)(11)(12)(13).Both type 1 and type 2 diabetes mellitus are associated with excess CVD risks (14), and approximately 2/3rds of diabetic patients suffer from cardiovascular disease (15)(16)(17)(18)(19), posing unique challenges in the treatment and management of patients.
A priority research arena is to elucidate best practice for the integrated care of patients with diabetes, particularly for those with cardiovascular complications.Data sources in Norway could provide an opportunity to advance this arena of research in which large population-based registries with comprehensive data on clinical management and outcomes could provide an opportunity to make meaningful contributions to the literature.An example of a research arena requiring further investigation is the extent to which CVD complications among diabetic patients is related to the degree of glycemic control (20).The efficacy of glycemic control interventions has not been observed for all CVD sequelae (21) and there is considerable uncertainty regarding the nature of the glycemic control and CVD dose-response relationship (22)(23)(24).Further, diabetes represents a challenge to health care practitioners and patients alike in that intensive efforts to control hyperglycemia may potentially introduce competing risks.The Action to Control Cardiovascular Risk in Diabetes Study found increased mortality and no significant reduction in cardiovascular disease events in the intensive intervention group (24).Research findings, thus far, have led to recommendations against overly strict glycemic control (25).
Another example of research promoting improved health care delivery for patients with diabetes was the finding of a nearly 4-fold excess risk of CVD observed among diabetic patients with prehypertension (systolic blood pressure of 120-139 mmHg and/or a diastolic blood pressure of 80-89 mmHg) (26) which has led to new recommendations for blood pressure control among diabetic patients.Another important research arena is the exploration of potential differences in drug efficacy in patients with and without diabetes (27)(28)(29).Finally, the extensive variability in the continuum of the diabetic condition needs to be considered in the evaluation of primary and secondary intervention approaches (30).The above mentioned illustrations provide a few examples regarding the need for ongoing clinical epidemiological research for identifying best practices in the treatment of patients with diabetes.

BACKGROUND ON THE NORWEGIAN CARDIOVASCULAR DISEASE REGISTRY (NCVDR) AND LEGAL MANDATE
The advantages of epidemiological research in Norway is well known given our national health care system, registries, and ability to link between data sources based upon the unique identification number of individuals.The formation of the Norwegian Cardiovascular Disease Registry (NCVDR) extends these epidemiological research opportunities to cardiovascular diseases.Given broad agreement for the need for a registry by patient organizations, research environments and health authorities, the Norwegian Parliament decided to establish the Norwegian CVD Registry in March 2010 and formally adopted new regulations on December 16, 2011 which established the Registry with data collection commencing in January 2012 (31,32).In accordance with the statutory authority of the Health Register Law Act §8, the legislation established the NCVDR as a nationwide registry in which personal identification of patients can be collected without the requirement for informed consent from those registered in order to meet the goals of the registry; the legislation expanded the Norwegian Institute of Public Health's (NIPH) disease surveillance mandate.The NIPH has data management responsibility and judicial responsibility for safeguarding patient privacy and confidentiality of patient information for the registry in accordance with applicable laws and regulations (32,33).Further, identifying information on individuals is removed from files used for analyses once a unique number is assigned to each person.This unique number enables researchers to generate health statistics without concern of duplicate enumeration of CVD events and makes it possible to track the same patient over time and from hospital to hospital.
The NCVDR consists of a core registry housed and maintained by the Bergen Division of the NIPH.A collaborative agreement is being finalized between NIPH and St. Olav's Hospital Trust to formalize the inclusion of clinical information collected by the medical quality registries, such as the Norwegian Myocardial Infarct Registry (34) and the Norwegian Stroke Registry (35).The inclusion of data from the medical quality registries will provide details regarding the variable manifestations of cardiovascular disease, risk factors, co-morbidities, treatment and their results.In the future, it is anticipated that the NCVDR will also include information collected by primary care physicians, which would provide additional information for CVD patients with diabetes, such as glycemic control.

OBJECTIVES
There are four primary objectives of the NCVDR: 1. to provide comprehensive data regarding the incidence and prevalence of CVD; 2. to evaluate the prognosis of patients by relevant diagnostic categories and treatment modalities; 3. to provide a means of supporting basic clinical research to enhance the quality of health care delivery; 4. to provide evidence-based knowledge for primary and secondary prevention and health care planning.

HOSPITAL TRUSTS
The NCVDR is based upon the collaborative organizational model, which has been described in the report "Good Health Registries -Better Health" (36), in which the government seeks to register information from a variety of sources for important diseases and medical conditions for improving disease surveillance and ultimately health care delivery and primary prevention strategies.By combining data resources into one nationwide CVD registry, we will have greater oversight and a greater likelihood of reducing the burden of disease in the population.
The core NCVDR will consist of data of cardiovascular disease patients identified starting Jan 1, 2012 from the Norwegian Patient Registry (37) and Cause of Death Registry (38), with supplemental information from the Central Population Registry (39) (Figure 1).CVD patients will be identified using the relevant ICD-10 codes associated with CVD (40) (Table 1).Patients will be included in the registry if they were discharged from hospital or out-patient clinics with one of the listed diagnostic codes (as main or secondary diagnoses).In addition, all the deceased with any CVDrelated diagnose as the underlying or contributing cause of death, will be included in the registry.
Of note for diabetes researchers is that the ICD-10 codes for diabetes found in ICD-10 Chapter IV regarding endocrine, nutritional, and metabolic disorders are not used as qualifying criteria for inclusion in the   1.Many patients with diabetes will have additional CVD diagnoses when being hospitalized, and many CVD patients will have diabetes as a secondary diagnosis.
The NCVDR will also collect information from the medical quality registries for selected conditions and procedures, including stroke, myocardial infarction, and vessel surgery, and, pending approvals, coronary angiography and percutaneous coronary intervention (PCI).

DIABETES RESEARCH OPPORTUNITIES
Given the burden of CVD among diabetic patients and the unique challenges they pose for clinicians, the formation of the NCVDR provides an opportunity to engage in research that will ultimately improve our understanding and clinical care for this high-risk patient population.We foresee the development of research projects comparing the prognosis for CVD patients with and without diabetes.Further, we foresee future data linkages between the NCVDR and other data sources for evaluating specific questions that can not  (44), will provide important details regarding medical care, glycemic control, and comorbidities which, through data linkages with the NCVDR, will represent a significant resource for both type 1 and type 2 diabetes research in Norway.Also, the data provided by other medical quality registries will enable a detailed evaluation of the range of comorbidities, and the prognosis for diabetic patients following vessel surgery or coronary angiography or PCI procedures.These are examples of the synergistic research opportunities that are to emerge once the NCVDR is operational.
A similar model of research in Denmark, for example, allowed for linking patients with nationwide administrative registries of dispensed prescriptions, hospitalizations, and cause of death.The record linkages enabled the evaluation of the efficacy of clopidogrel treatment following myocardial infarction in which the risk reduction for diabetics was lower than that observed for non-diabetic patients (28).Diabetic patients have increased platelet activity and despite clopidogrel treatment have high platelet activity (45,46).These and other research findings speak to the need for identifying appropriate risk reduction strategies for CVD patients with diabetes.

Figure 1 .
Figure 1.Diagram of the Norwegian Cardiovascular Disease Registry (NCVDR) Components.Abbreviations: NPR, National Patient Registry; CDR, Cause of Death Registry; National Reg, National Population Registry.

Table 1 .
Qualifying International Classification of Disease (ICD-10) Codes utilized for identifying patients for inclusion in the Norwegian Cardiovascular Disease Registry.

Table 2 .
Data sources of relevance for diabetes research in Norway.