Format

Send to

Choose Destination
J Clin Epidemiol. 2016 Oct;78:90-100. doi: 10.1016/j.jclinepi.2016.03.017. Epub 2016 Apr 1.

Explicit inclusion of treatment in prognostic modeling was recommended in observational and randomized settings.

Author information

1
Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands. Electronic address: r.h.h.groenwold@umcutrecht.nl.
2
Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands; Dutch Cochrane Center, University Medical Center Utrecht, PO Box 85500, Utrecht, 3508 GA, The Netherlands.
3
Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands.
4
Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford OX3 7LD, United Kingdom.
5
Research Institute for Primary Care and Health Sciences, Keele University, Staffordshire ST5 5BG, United Kingdom.

Abstract

OBJECTIVES:

To compare different methods to handle treatment when developing a prognostic model that aims to produce accurate probabilities of the outcome of individuals if left untreated.

STUDY DESIGN AND SETTING:

Simulations were performed based on two normally distributed predictors, a binary outcome, and a binary treatment, mimicking a randomized trial or an observational study. Comparison was made between simply ignoring treatment (SIT), restricting the analytical data set to untreated individuals (AUT), inverse probability weighting (IPW), and explicit modeling of treatment (MT). Methods were compared in terms of predictive performance of the model and the proportion of incorrect treatment decisions.

RESULTS:

Omitting a genuine predictor of the outcome from the prognostic model decreased model performance, in both an observational study and a randomized trial. In randomized trials, the proportion of incorrect treatment decisions was smaller when applying AUT or MT, compared to SIT and IPW. In observational studies, MT was superior to all other methods regarding the proportion of incorrect treatment decisions.

CONCLUSION:

If a prognostic model aims to produce correct probabilities of the outcome in the absence of treatment, ignoring treatments that affect that outcome can lead to suboptimal model performance and incorrect treatment decisions. Explicitly, modeling treatment is recommended.

KEYWORDS:

Calibration; Computer simulation; Decision support techniques; Models; Prognosis; Statistical

PMID:
27045189
DOI:
10.1016/j.jclinepi.2016.03.017
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center