Read your latest personalised notifications
No account yet? Start here
Don't miss out
Ok, got it
Registries are vital to providing information about a broad range of patients within a disease population, information that cannot be gained from the highly selected patient groups taking part in clinical trials. However, the quality of the registry output depends on the quality of the input, as Professor Lars Lund (Karolinska Institutet, Stockholm, Sweden), Heart Failure III Registry Chair, and Doctor Aldo Pietro Maggioni (ANMCO Research Centre, Florence, Italy), EURObservational Research Programme (EORP) Scientific Coordinator, discuss here
“There are two main aspects to registry quality. The first aspect is the internal validity of the data, which is a measure of its reliability. In a high-quality registry, the data collected should be correct and meaningful and attempts should be made to minimise the amount of missing data. The second aspect is external validity, which means that the data should be as representative of the general population as possible.” And it is this generalisability of data that Prof. Lund thinks we should focus on. “The message to investigators really is to report as many patients, with as much information, as possible. There is a certain amount of trade-off between internal and external validity; the more patients included, the more missing data likely to be encountered. Also, with a large number of patients, there is a compromise on the amount of data collected to make sure that the process remains feasible for investigators to perform. However, the most important point is that a registry must reflect the breadth of the general population, in terms of age, sex and comorbidities. It is this scope of information that is needed by the registry users, including the policy makers and clinical trial designers,” he explains.
“A registry must reflect the breadth of the general population.”
Dr. Maggioni describes what he sees as elements important in the construction of a high-quality registry. “There should be a good, clearly described scientific question to stimulate the interest of the investigators and the use of clear, simple record forms that are easy to complete and strictly focussed on the scientific aims will help to make sure that the correct information is collected.” Expanding on the point of representativeness, he says, “The geographic distribution and technological capability of the centres included must be representative. For ESC registries, the ESC Atlas of Cardiology is invaluable in helping us to identify the types of centres that are potential candidates for inclusion in a registry in each participating country.” “Finally,” continues Dr. Maggioni, “consecutiveness of data collection is essential if patient selection is to be avoided. For some hospitals, enrolling all patients may not be feasible. In such cases, hospitals can choose to restrict data collection, for example by limiting it to a period of time, such as a week or a month, or by performing it only on a certain number of days in the week.”
Both Prof. Lund and Dr. Maggioni see the main barrier to building high-quality registries as being the lack of incentives for investigators to take part in an activity that can require a considerable amount of time and energy. “We know that being a co-author on a registry publication is an attractive incentive for many investigators and we are actively working towards increasing the prospects of authorship for national and local investigators,” says Prof. Lund. Dr. Maggioni agrees, adding, “We should also focus on creating a culture of investigative research, particularly among the younger generation of doctors, that makes them want to be part of a community that works together to reduce the burden of cardiovascular disease.”
Join Prof. Lund, Dr. Maggioni and other experts at today’s session ‘Heart failure epidemiology and registries’
Monday, 11:00 – 12:30; Agora 1
Our mission: To reduce the burden of cardiovascular disease.
© 2020 European Society of Cardiology. All rights reserved.