Developing and testing models to predict mortality in the general population

ABSTRACT

We have previously proposed an approach using information collected from published reports to generate prediction models. The goal of this project was to validate this technique to develop and test various prediction models. A risk indicator (R) is calculated as a linear combination of the hazard ratios for the following predictors: age, male gender, diabetes, albuminuria, and either CKD, CVD or both. We developed a linear and two exponential expressions to predict the probability of the outcome of 2-year mortality and compared to actual outcome in the target dataset from NHANES. The risk indicator demonstrated good performance with area under ROC curve of 0.84. The linear and two exponential expressions generated similar predictions in the lower categories of risk indicator (R ≤ 6). However, in the groups with higher R value, the linear expression tends to predict lower, and the exponential expressions higher, probabilities than the observed outcome. A Combined model which averaged the linear and logistic expressions was shown to approximate the actual outcome data the best. A simple technique (named Woodpecker™) allows derivation functional prediction models and risk stratification tools from reports of clinical outcome studies and their application to new populations by using only summary statistics of the new population.

KEYWORDS:

• Prediction modeling
• decision support
• outcome
• survival
• Woodpecker™
• mortality
• epidemiology
• personalized medicine

List of abbreviations

AUC	=	area under receiver operator characteristics curve
CKD	=	chronic kidney disease
CVD	=	cardio-vascular disease
HR	=	hazard ratio
NHANES	=	National Health and Nutrition Examination Survey
P(R)	=	probability of the outcome
Ȓ	=	average value of risk indicator in population
r	=	mortality ratio in the population
R	=	risk indicator
ROC	=	receiver operator characteristics

Competing interests

The authors declare that they have no competing interests. Prediction modeling technique described in the paper is registered as Woodpecker™ by the Beth Israel Deaconess Medical Center, Boston, MA. AG-R and RB are collecting royalty from their published book and app “Nephrology Pocket” unrelated to this project AG-R is employed by Akebia Therapeutics, Inc (Cambridge, Massachusetts).

Non-financial competing interests

There aren’t any non-financial competing interests (political, personal, religious, ideological, academic, intellectual, commercial or any other) to declare in relation to this manuscript.

Authors’ contributions

AG-R has made substantial contributions to conception and design by developing an idea for prediction modeling technique (Woodpecker™) and has been involved in drafting the manuscript.

RB has made substantial contributions to conception and design of the validation study using NHANES data and has been revising the manuscript critically for important intellectual content.

ND has made substantial contributions to conception and design of the study by developing the approach to combine different prediction models, development of the Woodpecker™ technique, and contributing to validation of the study design.

GS has been involved in analysis and interpretation of data and drafting the manuscript.

PV made contribution to study design, data analysis and drafting the manuscript.

SG has made substantial contributions to development of the Woodpecker™ technique, providing statistical expertise, and has been involved in analysis and interpretation of data (performed the statistical analysis).

All authors have given final approval of the version to be published.

Acknowledgments

All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Availability of data and materials

The original data used in the study are publically available on the NHANES web-site.

Ethics

The study was approved by the Institutional review Board at the Beth Israel Deaconess Medical Center with an exempt status as analysis has to do with de-identified retrospective human data available in public domain.

Additional information

Funding

The study was funded from departmental funds at the Department of Medicine Beth Israel Deaconess Medical Center and did not have any outside sponsor or funding agency.