An alternative approach identified optimal risk thresholds for treatment indication: an illustration in coronary heart disease

Anoukh van Giessen* (Corresponding Author), G. Ardine de Wit, Karel G.M. Moons, Jannick A.N. Dorresteijn, Hendrik Koffijberg

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

Objectives: Treatment thresholds based on risk predictions can be optimized by considering various health (economic) outcomes and performing marginal analyses, but this is rarely performed. We demonstrate a general approach to identify treatment thresholds optimizing individual health (economic) outcomes, illustrated for statin treatment based on 10-year coronary heart disease (CHD) risk predicted by the Framingham risk score.

Study Design and Setting: Creating a health economic model for a risk-based prevention strategy, risk thresholds can be evaluated on several outcomes of interest. Selecting an appropriate threshold range and decrement size for the thresholds and adapting the health economic model accordingly, outcomes can be calculated for each risk threshold. A stepwise, or marginal, comparison of clinical as well as health economic outcomes, that is, comparing outcomes using a specific threshold to outcomes of the former threshold while gradually lowering the threshold, then takes into account the balance between additional numbers of individuals treated and their outcomes (additional health effects and costs). In our illustration, using a Markov model for CHD, we evaluated risk thresholds by gradually lowering thresholds from 20% to 0%.

Results: This approach can be applied to identify optimal risk thresholds on any outcome, such as to limit complications, maximize health outcomes, or optimize cost-effectiveness. In our illustration, keeping the population-level fraction of statin-induced complications <10% resulted in thresholds of T = 6% (men) and T = 2% (women). Lowering the threshold and comparing quality-adjusted life-years (QALYs) after each 1% decrease, QALYs were gained down to T = 1% (men) and T = 0% (women). Also accounting for costs, net health benefits were favorable down to T = 3% (men) and T = 6% (women).

Conclusion: Using a stepwise risk-based approach to threshold optimization allows for preventive strategies that optimize outcomes. Presenting this comprehensive overview of outcomes will better inform decision makers when defining a treatment threshold.

Original languageEnglish
Pages (from-to)122-131
Number of pages10
JournalJournal of clinical epidemiology
Volume94
DOIs
Publication statusPublished - 1 Feb 2018

Fingerprint

Coronary Disease
Health
Economic Models
Hydroxymethylglutaryl-CoA Reductase Inhibitors
Quality-Adjusted Life Years
Economics
Cost-Benefit Analysis
Insurance Benefits
Therapeutics
Health Care Costs
Costs and Cost Analysis
Population

Keywords

  • Coronary heart disease
  • Cost-effectiveness
  • Marginal analysis
  • Optimization
  • Risk stratification
  • Treatment threshold

Cite this

van Giessen, Anoukh ; de Wit, G. Ardine ; Moons, Karel G.M. ; Dorresteijn, Jannick A.N. ; Koffijberg, Hendrik. / An alternative approach identified optimal risk thresholds for treatment indication : an illustration in coronary heart disease. In: Journal of clinical epidemiology. 2018 ; Vol. 94. pp. 122-131.
@article{28cb4f1e35a54371903d8fe93a849a6b,
title = "An alternative approach identified optimal risk thresholds for treatment indication: an illustration in coronary heart disease",
abstract = "Objectives: Treatment thresholds based on risk predictions can be optimized by considering various health (economic) outcomes and performing marginal analyses, but this is rarely performed. We demonstrate a general approach to identify treatment thresholds optimizing individual health (economic) outcomes, illustrated for statin treatment based on 10-year coronary heart disease (CHD) risk predicted by the Framingham risk score.Study Design and Setting: Creating a health economic model for a risk-based prevention strategy, risk thresholds can be evaluated on several outcomes of interest. Selecting an appropriate threshold range and decrement size for the thresholds and adapting the health economic model accordingly, outcomes can be calculated for each risk threshold. A stepwise, or marginal, comparison of clinical as well as health economic outcomes, that is, comparing outcomes using a specific threshold to outcomes of the former threshold while gradually lowering the threshold, then takes into account the balance between additional numbers of individuals treated and their outcomes (additional health effects and costs). In our illustration, using a Markov model for CHD, we evaluated risk thresholds by gradually lowering thresholds from 20{\%} to 0{\%}.Results: This approach can be applied to identify optimal risk thresholds on any outcome, such as to limit complications, maximize health outcomes, or optimize cost-effectiveness. In our illustration, keeping the population-level fraction of statin-induced complications <10{\%} resulted in thresholds of T = 6{\%} (men) and T = 2{\%} (women). Lowering the threshold and comparing quality-adjusted life-years (QALYs) after each 1{\%} decrease, QALYs were gained down to T = 1{\%} (men) and T = 0{\%} (women). Also accounting for costs, net health benefits were favorable down to T = 3{\%} (men) and T = 6{\%} (women). Conclusion: Using a stepwise risk-based approach to threshold optimization allows for preventive strategies that optimize outcomes. Presenting this comprehensive overview of outcomes will better inform decision makers when defining a treatment threshold.",
keywords = "Coronary heart disease, Cost-effectiveness, Marginal analysis, Optimization, Risk stratification, Treatment threshold",
author = "{van Giessen}, Anoukh and {de Wit}, {G. Ardine} and Moons, {Karel G.M.} and Dorresteijn, {Jannick A.N.} and Hendrik Koffijberg",
year = "2018",
month = "2",
day = "1",
doi = "10.1016/j.jclinepi.2017.09.020",
language = "English",
volume = "94",
pages = "122--131",
journal = "Journal of clinical epidemiology",
issn = "0895-4356",
publisher = "Elsevier",

}

An alternative approach identified optimal risk thresholds for treatment indication : an illustration in coronary heart disease. / van Giessen, Anoukh (Corresponding Author); de Wit, G. Ardine; Moons, Karel G.M.; Dorresteijn, Jannick A.N.; Koffijberg, Hendrik.

In: Journal of clinical epidemiology, Vol. 94, 01.02.2018, p. 122-131.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - An alternative approach identified optimal risk thresholds for treatment indication

T2 - an illustration in coronary heart disease

AU - van Giessen, Anoukh

AU - de Wit, G. Ardine

AU - Moons, Karel G.M.

AU - Dorresteijn, Jannick A.N.

AU - Koffijberg, Hendrik

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Objectives: Treatment thresholds based on risk predictions can be optimized by considering various health (economic) outcomes and performing marginal analyses, but this is rarely performed. We demonstrate a general approach to identify treatment thresholds optimizing individual health (economic) outcomes, illustrated for statin treatment based on 10-year coronary heart disease (CHD) risk predicted by the Framingham risk score.Study Design and Setting: Creating a health economic model for a risk-based prevention strategy, risk thresholds can be evaluated on several outcomes of interest. Selecting an appropriate threshold range and decrement size for the thresholds and adapting the health economic model accordingly, outcomes can be calculated for each risk threshold. A stepwise, or marginal, comparison of clinical as well as health economic outcomes, that is, comparing outcomes using a specific threshold to outcomes of the former threshold while gradually lowering the threshold, then takes into account the balance between additional numbers of individuals treated and their outcomes (additional health effects and costs). In our illustration, using a Markov model for CHD, we evaluated risk thresholds by gradually lowering thresholds from 20% to 0%.Results: This approach can be applied to identify optimal risk thresholds on any outcome, such as to limit complications, maximize health outcomes, or optimize cost-effectiveness. In our illustration, keeping the population-level fraction of statin-induced complications <10% resulted in thresholds of T = 6% (men) and T = 2% (women). Lowering the threshold and comparing quality-adjusted life-years (QALYs) after each 1% decrease, QALYs were gained down to T = 1% (men) and T = 0% (women). Also accounting for costs, net health benefits were favorable down to T = 3% (men) and T = 6% (women). Conclusion: Using a stepwise risk-based approach to threshold optimization allows for preventive strategies that optimize outcomes. Presenting this comprehensive overview of outcomes will better inform decision makers when defining a treatment threshold.

AB - Objectives: Treatment thresholds based on risk predictions can be optimized by considering various health (economic) outcomes and performing marginal analyses, but this is rarely performed. We demonstrate a general approach to identify treatment thresholds optimizing individual health (economic) outcomes, illustrated for statin treatment based on 10-year coronary heart disease (CHD) risk predicted by the Framingham risk score.Study Design and Setting: Creating a health economic model for a risk-based prevention strategy, risk thresholds can be evaluated on several outcomes of interest. Selecting an appropriate threshold range and decrement size for the thresholds and adapting the health economic model accordingly, outcomes can be calculated for each risk threshold. A stepwise, or marginal, comparison of clinical as well as health economic outcomes, that is, comparing outcomes using a specific threshold to outcomes of the former threshold while gradually lowering the threshold, then takes into account the balance between additional numbers of individuals treated and their outcomes (additional health effects and costs). In our illustration, using a Markov model for CHD, we evaluated risk thresholds by gradually lowering thresholds from 20% to 0%.Results: This approach can be applied to identify optimal risk thresholds on any outcome, such as to limit complications, maximize health outcomes, or optimize cost-effectiveness. In our illustration, keeping the population-level fraction of statin-induced complications <10% resulted in thresholds of T = 6% (men) and T = 2% (women). Lowering the threshold and comparing quality-adjusted life-years (QALYs) after each 1% decrease, QALYs were gained down to T = 1% (men) and T = 0% (women). Also accounting for costs, net health benefits were favorable down to T = 3% (men) and T = 6% (women). Conclusion: Using a stepwise risk-based approach to threshold optimization allows for preventive strategies that optimize outcomes. Presenting this comprehensive overview of outcomes will better inform decision makers when defining a treatment threshold.

KW - Coronary heart disease

KW - Cost-effectiveness

KW - Marginal analysis

KW - Optimization

KW - Risk stratification

KW - Treatment threshold

UR - http://www.scopus.com/inward/record.url?scp=85033705978&partnerID=8YFLogxK

U2 - 10.1016/j.jclinepi.2017.09.020

DO - 10.1016/j.jclinepi.2017.09.020

M3 - Article

AN - SCOPUS:85033705978

VL - 94

SP - 122

EP - 131

JO - Journal of clinical epidemiology

JF - Journal of clinical epidemiology

SN - 0895-4356

ER -