Predictive modeling in practice: Improving the participant identification process for care management programs using condition-specific cut points

Shannon M E Murphy, Heather K. Castro, Martha Sylvia

Research output: Contribution to journalArticle

Abstract

The objective of this study was to optimize predictive modeling in the participant selection process for care management (CM) programs by determining the ideal cut point selection method. Comparisons included: (a) an evidence-based "optimal" cut point versus an "arbitrary" threshold, and (b) condition-specific cut points versus a uniform screening method. Participants comprised adult Medicaid health plan members enrolled during the entire study period (January 2007-December 2008) who had at least 1 of the chronic conditions targeted by the CM programs (n=6459). Adjusted Clinical Groups Predictive Modeling (ACG-PM) system risk scores in 2007 were used to predict those with the top 5% highest health care expenditures in 2008. Comparisons of model performance (ie, c statistic, sensitivity, specificity, positive predictive value) and identified population size were used to assess differences among 3 cut point selection approaches: (a) single arbitrary cut point, (b) single optimal cut point, and (c) condition-specific optimal cut points. The "optimal" cut points (ie, single and condition-specific) both outperformed the "arbitrary" selection process, yielding higher probabilities of correct prediction and sensitivities. The condition-specific optimal cut point approach also exhibited better performance than applying a single optimal cut point uniformly across the entire population regardless of condition (ie, a higher c statistic, specificity, and positive predictive value, although sensitivity was lower), while identifying a more manageable number of members for CM program outreach. CM programs can optimize targeting algorithms by utilizing evidence-based cut points that incorporate condition-specific variations in risk. By efficiently targeting and intervening with future high-cost members, health care costs can be reduced. (Population Health Management 2011;14:205-210)

Original languageEnglish (US)
Pages (from-to)205-210
Number of pages6
JournalPopulation Health Management
Volume14
Issue number4
DOIs
StatePublished - Aug 1 2011

Fingerprint

Medicaid
Health
Health Expenditures
Population Density
Health Care Costs
Population
Delivery of Health Care
Costs and Cost Analysis
Sensitivity and Specificity

ASJC Scopus subject areas

  • Health Policy
  • Leadership and Management
  • Public Health, Environmental and Occupational Health

Cite this

Predictive modeling in practice : Improving the participant identification process for care management programs using condition-specific cut points. / Murphy, Shannon M E; Castro, Heather K.; Sylvia, Martha.

In: Population Health Management, Vol. 14, No. 4, 01.08.2011, p. 205-210.

Research output: Contribution to journalArticle

@article{18aef9a0139d45368a05cc1a9fae50b0,
title = "Predictive modeling in practice: Improving the participant identification process for care management programs using condition-specific cut points",
abstract = "The objective of this study was to optimize predictive modeling in the participant selection process for care management (CM) programs by determining the ideal cut point selection method. Comparisons included: (a) an evidence-based {"}optimal{"} cut point versus an {"}arbitrary{"} threshold, and (b) condition-specific cut points versus a uniform screening method. Participants comprised adult Medicaid health plan members enrolled during the entire study period (January 2007-December 2008) who had at least 1 of the chronic conditions targeted by the CM programs (n=6459). Adjusted Clinical Groups Predictive Modeling (ACG-PM) system risk scores in 2007 were used to predict those with the top 5{\%} highest health care expenditures in 2008. Comparisons of model performance (ie, c statistic, sensitivity, specificity, positive predictive value) and identified population size were used to assess differences among 3 cut point selection approaches: (a) single arbitrary cut point, (b) single optimal cut point, and (c) condition-specific optimal cut points. The {"}optimal{"} cut points (ie, single and condition-specific) both outperformed the {"}arbitrary{"} selection process, yielding higher probabilities of correct prediction and sensitivities. The condition-specific optimal cut point approach also exhibited better performance than applying a single optimal cut point uniformly across the entire population regardless of condition (ie, a higher c statistic, specificity, and positive predictive value, although sensitivity was lower), while identifying a more manageable number of members for CM program outreach. CM programs can optimize targeting algorithms by utilizing evidence-based cut points that incorporate condition-specific variations in risk. By efficiently targeting and intervening with future high-cost members, health care costs can be reduced. (Population Health Management 2011;14:205-210)",
author = "Murphy, {Shannon M E} and Castro, {Heather K.} and Martha Sylvia",
year = "2011",
month = "8",
day = "1",
doi = "10.1089/pop.2010.0005",
language = "English (US)",
volume = "14",
pages = "205--210",
journal = "Population Health Management",
issn = "1942-7891",
publisher = "Mary Ann Liebert Inc.",
number = "4",

}

TY - JOUR

T1 - Predictive modeling in practice

T2 - Improving the participant identification process for care management programs using condition-specific cut points

AU - Murphy, Shannon M E

AU - Castro, Heather K.

AU - Sylvia, Martha

PY - 2011/8/1

Y1 - 2011/8/1

N2 - The objective of this study was to optimize predictive modeling in the participant selection process for care management (CM) programs by determining the ideal cut point selection method. Comparisons included: (a) an evidence-based "optimal" cut point versus an "arbitrary" threshold, and (b) condition-specific cut points versus a uniform screening method. Participants comprised adult Medicaid health plan members enrolled during the entire study period (January 2007-December 2008) who had at least 1 of the chronic conditions targeted by the CM programs (n=6459). Adjusted Clinical Groups Predictive Modeling (ACG-PM) system risk scores in 2007 were used to predict those with the top 5% highest health care expenditures in 2008. Comparisons of model performance (ie, c statistic, sensitivity, specificity, positive predictive value) and identified population size were used to assess differences among 3 cut point selection approaches: (a) single arbitrary cut point, (b) single optimal cut point, and (c) condition-specific optimal cut points. The "optimal" cut points (ie, single and condition-specific) both outperformed the "arbitrary" selection process, yielding higher probabilities of correct prediction and sensitivities. The condition-specific optimal cut point approach also exhibited better performance than applying a single optimal cut point uniformly across the entire population regardless of condition (ie, a higher c statistic, specificity, and positive predictive value, although sensitivity was lower), while identifying a more manageable number of members for CM program outreach. CM programs can optimize targeting algorithms by utilizing evidence-based cut points that incorporate condition-specific variations in risk. By efficiently targeting and intervening with future high-cost members, health care costs can be reduced. (Population Health Management 2011;14:205-210)

AB - The objective of this study was to optimize predictive modeling in the participant selection process for care management (CM) programs by determining the ideal cut point selection method. Comparisons included: (a) an evidence-based "optimal" cut point versus an "arbitrary" threshold, and (b) condition-specific cut points versus a uniform screening method. Participants comprised adult Medicaid health plan members enrolled during the entire study period (January 2007-December 2008) who had at least 1 of the chronic conditions targeted by the CM programs (n=6459). Adjusted Clinical Groups Predictive Modeling (ACG-PM) system risk scores in 2007 were used to predict those with the top 5% highest health care expenditures in 2008. Comparisons of model performance (ie, c statistic, sensitivity, specificity, positive predictive value) and identified population size were used to assess differences among 3 cut point selection approaches: (a) single arbitrary cut point, (b) single optimal cut point, and (c) condition-specific optimal cut points. The "optimal" cut points (ie, single and condition-specific) both outperformed the "arbitrary" selection process, yielding higher probabilities of correct prediction and sensitivities. The condition-specific optimal cut point approach also exhibited better performance than applying a single optimal cut point uniformly across the entire population regardless of condition (ie, a higher c statistic, specificity, and positive predictive value, although sensitivity was lower), while identifying a more manageable number of members for CM program outreach. CM programs can optimize targeting algorithms by utilizing evidence-based cut points that incorporate condition-specific variations in risk. By efficiently targeting and intervening with future high-cost members, health care costs can be reduced. (Population Health Management 2011;14:205-210)

UR - http://www.scopus.com/inward/record.url?scp=80051774516&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80051774516&partnerID=8YFLogxK

U2 - 10.1089/pop.2010.0005

DO - 10.1089/pop.2010.0005

M3 - Article

C2 - 21241172

AN - SCOPUS:80051774516

VL - 14

SP - 205

EP - 210

JO - Population Health Management

JF - Population Health Management

SN - 1942-7891

IS - 4

ER -