Correlational science discovers associations between patient characteristics, symptoms and biomarkers. Correlational science using data from cross-sectional studies is the most frequently applied study design in palliative care research. The purpose of this review is to address the importance and potential pitfalls in correlational science.
Associations observed in correlational science studies can be the basis for generating hypotheses that can be tested in experimental studies and are the basic data needed to develop classification systems that can predict patient outcomes. Major pitfalls in correlational science are that associations do not equate with causality and that statistical significance does not necessarily equal a correlation that is of clinical interest. Researchers should be aware of the end-points that are clinically relevant, that end-points should be defined before the start of the analyses, and that studies with several end-points should account for multiplicity.
Correlational science in palliative care research can identify related clinical factors and biomarkers. Interpretation of identified associations should be done with careful consideration of the limitations underlying correlational analyses.
aDepartment of Anesthesiology and Acute Medicine, St. Olavs University Hospital
bDepartment of Circulation and Medical Imaging, Medical Faculty, Norwegian University of Science and Technology
cEuropean Palliative Care Research Centre (PRC), Faculty of Medicine, Norwegian University of Science and Technology
dCancer Clinic, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway
Correspondence to Pål Klepstad, MD, PhD, Department Of Anesthesiology and Acute Medicine, St. Olavs University Hospital, 7006 Trondheim, Norway. Tel: +47 72575709; e-mail: firstname.lastname@example.org