Related Experiment Videos
Predicting elderly outpatients' life-sustaining treatment preferences over time: the majority rules
Renate M Houts1, William D Smucker, Jill A Jacobson
1Center for Developmental Science, University of North Carolina at Chapel Hill, USA. rhouts@rti.org
Summary
Predicting patient treatment preferences using modal preferences requires periodic updates. Models based on these preferences accurately predict end-of-life choices, outperforming surrogate predictions.
Area of Science:
- Gerontology
- Medical Ethics
- Health Services Research
Background:
- Understanding patient preferences for life-sustaining treatments is crucial for end-of-life care.
- Longitudinal studies are needed to track changes in these preferences over time.
Purpose of the Study:
- To examine longitudinal changes in modal life-sustaining treatment preferences.
- To assess the accuracy of these preferences as predictors of patient choices.
- To compare prediction models with surrogate decision-makers.
Main Methods:
- Healthy outpatients aged 65+ and their surrogates recorded treatment preferences over two years.
- A statistical prediction model was developed using initial modal preferences and updated later.
- Model accuracy was compared against concurrent surrogate predictions.
Main Results:
- The prediction model showed changes over two years, with some preferences becoming less relevant and new ones emerging.
- Observed preference shifts indicated a trend towards refusing treatments.
- Both initial and updated models were more accurate than surrogates in predicting patient preferences.
- Covariates like age and gender did not improve the model's predictive power over surrogates.
Conclusions:
- Models utilizing modal preferences are valuable tools for patients, surrogates, and physicians in end-of-life decision-making.
- Regularly updating these models is essential for maintaining predictive accuracy.