Incorporating predictions of individual patient risk in clinical trials Journal Article
| Authors: | Kattan, M. W.; Vickers, A. J. |
| Article Title: | Incorporating predictions of individual patient risk in clinical trials |
| Abstract: | A risk prediction model is a statistical technique that gives a predicted probability of a certain event for an individual patient. Prediction models outperform the traditional risk classification systems that work by assigning patients into risk groups based on the presence or absence of particular risk factors, such as stage of disease. As such, risk prediction models have a number of important possible uses in clinical trials. For Phase II studies, prediction models can help adjust comparisons with historical control groups for differences in case mix. For Phase III studies, prediction models can ensure that accrued patients are at sufficiently high risk. This improves statistical power and avoids unethical inclusion of low-risk patients. We also propose that prediction models could potentially be used for applying the results of Phase III trials to individual patients. Clinical decisions could be informed by individualized estimates of treatment benefit, rather than by average treatment effects. © 2004 Elsevier Inc. All rights reserved. |
| Keywords: | cancer surgery; cancer risk; patient selection; conference paper; pancreas cancer; research design; cancer staging; recurrence risk; neoplasms; prostate specific antigen; breast cancer; cancer screening; smoking; prediction; risk factor; risk assessment; prostate cancer; sarcoma; probability; models, statistical; stomach cancer; high risk population; kaplan meier method; kidney cancer; decision making; onset age; statistical model; nomogram; sample size; diagnosis-related groups; menstrual cycle; clinical trials, phase iii; power; case mix; humans; prognosis; human; priority journal; sex differentiation; control groups |
| Journal Title: | Urologic Oncology: Seminars and Original Investigations |
| Volume: | 22 |
| Issue: | 4 |
| ISSN: | 1078-1439 |
| Publisher: | Elsevier Inc. |
| Date Published: | 2004-07-01 |
| Start Page: | 348 |
| End Page: | 352 |
| Language: | English |
| DOI: | 10.1016/j.urolonc.2004.04.012 |
| PROVIDER: | scopus |
| PUBMED: | 15283895 |
| DOI/URL: | |
| Notes: | Urol. Oncol. Semin. Orig. Invest. -- Cited By (since 1996):7 -- Export Date: 16 June 2014 -- CODEN: UOSOA -- Source: Scopus |
Altmetric
Citation Impact
BMJ Impact Analytics
Related MSK Work