Tibor V Varga1, Kristoffer Niss2, Angela C Estampador3, Catherine B Collin2, Pope L Moseley2. 1. Section of Epidemiology, Department of Public Health, University of Copenhagen, Copenhagen, Denmark; Novo Nordisk Foundation Center for Protein Research, Translational Disease Systems Biology Group, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; Department of Clinical Sciences, Genetic and Molecular Epidemiology Unit, Lund University, Skåne University Hospital Malmö, Malmö, Sweden. Electronic address: tibor.varga@sund.ku.dk. 2. Novo Nordisk Foundation Center for Protein Research, Translational Disease Systems Biology Group, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. 3. ustwo AB, Malmö, Sweden.
Abstract
AIMS: Appropriate analysis of big data is fundamental to precision medicine. While statistical analyses often uncover numerous associations, associations themselves do not convey predictive value. Confusion between association and prediction harms clinicians, scientists, and ultimately, the patients. We analyzed published papers in the field of diabetes that refer to "prediction" in their titles. We assessed whether these articles report metrics relevant to prediction. METHODS: A systematic search was undertaken using NCBI PubMed. Articles with the terms "diabetes" and "prediction" were selected. All abstracts of original research articles, within the field of diabetes epidemiology, were searched for metrics pertaining to predictive statistics. Simulated data was generated to visually convey the differences between association and prediction. RESULTS: The search-term yielded 2,182 results. After discarding non-relevant articles, 1,910 abstracts were evaluated. Of these, 39% (n = 745) reported metrics of predictive statistics, while 61% (n = 1,165) did not. The top reported metrics of prediction were ROC AUC, sensitivity and specificity. Using the simulated data, we demonstrated that biomarkers with large effect sizes and low P values can still offer poor discriminative utility. CONCLUSIONS: We demonstrate a landscape of confused reporting within the field of diabetes epidemiology where the term "prediction" is often incorrectly used to refer to association statistics. We propose guidelines for future reporting, and two major routes forward in terms of main analytic procedures and research goals: the explanatory route, which contributes to precision medicine, and the prediction route which contributes to personalized medicine.
AIMS: Appropriate analysis of big data is fundamental to precision medicine. While statistical analyses often uncover numerous associations, associations themselves do not convey predictive value. Confusion between association and prediction harms clinicians, scientists, and ultimately, the patients. We analyzed published papers in the field of diabetes that refer to "prediction" in their titles. We assessed whether these articles report metrics relevant to prediction. METHODS: A systematic search was undertaken using NCBI PubMed. Articles with the terms "diabetes" and "prediction" were selected. All abstracts of original research articles, within the field of diabetes epidemiology, were searched for metrics pertaining to predictive statistics. Simulated data was generated to visually convey the differences between association and prediction. RESULTS: The search-term yielded 2,182 results. After discarding non-relevant articles, 1,910 abstracts were evaluated. Of these, 39% (n = 745) reported metrics of predictive statistics, while 61% (n = 1,165) did not. The top reported metrics of prediction were ROC AUC, sensitivity and specificity. Using the simulated data, we demonstrated that biomarkers with large effect sizes and low P values can still offer poor discriminative utility. CONCLUSIONS: We demonstrate a landscape of confused reporting within the field of diabetes epidemiology where the term "prediction" is often incorrectly used to refer to association statistics. We propose guidelines for future reporting, and two major routes forward in terms of main analytic procedures and research goals: the explanatory route, which contributes to precision medicine, and the prediction route which contributes to personalized medicine.
Authors: Allison L O'Kell; Clive Wasserfall; Joy Guingab-Cagmat; Bobbie-Jo M Webb-Roberston; Mark A Atkinson; Timothy J Garrett Journal: Metabolomics Date: 2021-11-14 Impact factor: 4.290
Authors: Giorgio De Nunzio; Luana Conte; Roberto Lupo; Elsa Vitale; Antonino Calabrò; Maurizio Ercolani; Maicol Carvello; Michele Arigliani; Domenico Maurizio Toraldo; Luigi De Benedetto Journal: Front Med (Lausanne) Date: 2022-05-25
Authors: Tibor V Varga; Jinxi Liu; Ronald B Goldberg; Guannan Chen; Samuel Dagogo-Jack; Carlos Lorenzo; Kieren J Mather; Xavier Pi-Sunyer; Søren Brunak; Marinella Temprosa Journal: BMJ Open Diabetes Res Care Date: 2021-03
Authors: Ashley J W Lim; Lee Jin Lim; Brandon N S Ooi; Ee Tzun Koh; Justina Wei Lynn Tan; Samuel S Chong; Chiea Chuen Khor; Lisa Tucker-Kellogg; Khai Pang Leong; Caroline G Lee Journal: EBioMedicine Date: 2022-01-10 Impact factor: 8.143