The Centre for Statistics in Medicine (CSM) based at NDORMS is committed to improving the quality of healthcare research, from its design and conduct to its analysis and reporting. The group not only provides medical investigators with statistical support but also engages in applied statistical research.
CSM's applied statistical research focuses on prognostic and diagnostic models. CSM Deputy Co-director Gary Collins heads a team that creates and validates these tools for clinical decision-making, and conducts research into the methods behind the models. Diagnostic and prognostic models have the potential to reduce invasive tests for patients, cut measurement costs, and accurately identify people with an undiagnosed disease or an increased risk of developing a disease.
For the first time, diagnostic and prognostic medical research has a dedicated home in an open-access journal. The new BMC Diagnostic & Prognostic Research will provide an online platform for primary studies, systematic reviews and methodology research into these crucial clinical decision-making tools. Professor Collins is one of three founding editors-in-chief.
"Diagnostic and prognostic models are key tools in clinical practice and have a direct impact on the healthcare delivered to patients. This new journal will provide medical researchers in the field with a critical platform based on excellence in scientific research and reporting," says Professor Collins.
Doctors use diagnostic and prognostic models to help them make decisions during their day-to-day work. Diagnostic models use a patient's characteristics now, like their blood pressure and X-ray results, to predict how likely it is that they are suffering from a particular disease right now. Prognostic models use the patient's characteristics now to estimate their risk of a future outcome. We might be interested in their chance of developing a particular disease or in whether a particular symptom will get worse or better.
Diagnostic and prognostic models are key tools in clinical practice and have a direct impact on the healthcare delivered to patients. - Professor Gary Collins
BMC Diagnostic & Prognostic Research will fill key gaps in how research into these models is published today. Diagnostic and prognostic research is often carried out without a study protocol and is frequently opportunistic. The new journal will for the first time encourage publication of protocols for this kind of work, improving both transparency and work quality. A protocol describes the intended study objectives, how it builds on existing research, and any planned or exploratory analyses. It helps to strengthen study design, leading to more convincing, usable research findings.
The journal will also bring together research into the methods behind diagnostic and prognostic methods, which until now has been scattered between methodology, epidemiology and clinical journals.
Professor Collins says: "Regardless of their aims, diagnostic or prognostic models should be designed, developed and evaluated using appropriate methodologies. Research in this area has often been poorly carried out, making it difficult to decide whether to use a particular diagnostic test or prognostic model. The methodological considerations in the research must be handled appropriately, and all results from the research must be reported clearly and in a timely fashion."
The new journal will also publish clinical articles. Articles that attempt to validate existing research findings, such as tests of existing diagnostic tests and prediction models, are particularly encouraged regardless of their findings.
Professor Collins is joined by co-editors-in-chief Karel Moons of UMC Utrecht, the Netherlands, and Nancy Cook of Harvard University, USA. The editorial board spans four continents and has a broad range of expertise.
Diagnostic & Prognostic Research is now accepting submissions, and will accept empirical primary studies, systematic reviews (including meta-analyses), methodology research papers, protocols and commentaries. The journal plans to publish diagnostic and prognostic research addressing the evaluation of medical tests, markers or multivariable prediction models. The journal is committed to the publication of the results of all well-conducted diagnostic and prognostic research, regardless of their outcome.
CSM currently has editors-in-chief for three journals; in addition to Professor Collins, Professor Doug Altman is an editor for Trials and Dr Iveta Simera is an editor for Research integrity and peer review.