Statistically Significant Association Does not Imply Improvement in Prediction of Clinical Outcomes
Open Access
- 29 September 2025
- journal article
- Published by American Association for Cancer Research (AACR) in Cancer Prevention Research
- Vol. 18 (12) , 727-733
Abstract
In the current landscape of clinical studies, the concept of statistically significant association is often mixed up with the expectation of improved prediction performance. We discuss the two concepts, association and prediction, and present the epidemiologic principles and statistical constructs that underlie the discrepancy between statistically significant associations and the rationale for their lack of impact on improving prediction in terms of discrimination. This issue is illustrated using an existing breast cancer dataset. The concept of statistically significant association should not be mixed up with the expectation of improved discrimination performance. Although some markers may not markedly improve discrimination, they can still have substantial clinical relevance by identifying critical biological pathways that inform novel treatment or prevention strategies. Development of models for both association and prediction assessments should be directly tied to clinical translation to move adoption forward to advance precision medicine. Prevention Relevance: Development of models for both association and prediction assessments should be directly tied to clinical translation to move adoption forward to advance precision medicine.Keywords
All Related Versions
Funding Information
- National Cancer Institute (R37 CA256810)
- Breast Cancer Research Foundation (BCRF 24-28)
- School of Medicine, Washington University in St. Louis
This publication has 35 references indexed in Scilit:
- Prediction Models and Clinical Outcomes—A Call for PapersJAMA Network Open, 2024
- The risks of cancer in older women with BRCA pathogenic variants: How far have we come?Cancer, 2022
- Concordance measures and time-dependent ROC methodsBiostatistics & Epidemiology, 2021
- Establishment of Best Practices for Evidence for PredictionJAMA Psychiatry, 2020
- 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease: Executive SummaryJACC, 2019
- Clinical Prediction Models2019
- Dynamic predictions with time‐dependent covariates in survival analysis using joint modeling and landmarkingBiometrical Journal, 2017
- Childhood body mass index and adult mammographic density measures that predict breast cancer riskBreast Cancer Research and Treatment, 2016
- Testing for improvement in prediction model performanceStatistics in Medicine, 2013
- Predictive accuracy of risk factors and markers: a simulation study of the effect of novel markers on different performance measures for logistic regression modelsStatistics in Medicine, 2012