Inverse probability weighted least squares regression in the analysis of time-censored cost data: An evaluation of the approach using SEER-medicare

Robert I. Griffiths, Michelle L. Gleeson, Mark D. Danese, Anthony O'Hagan

Research output: Contribution to journalArticle

Abstract

Objectives: To assess the accuracy and precision of inverse probability weighted (IPW) least squares regression analysis for censored cost data. Methods: By using Surveillance, Epidemiology, and End Results-Medicare, we identified 1500 breast cancer patients who died and had complete cost information within the database. Patients were followed for up to 48 months (partitions) after diagnosis, and their actual total cost was calculated in each partition. We then simulated patterns of administrative and dropout censoring and also added censoring to patients receiving chemotherapy to simulate comparing a newer to older intervention. For each censoring simulation, we performed 1000 IPW regression analyses (bootstrap, sampling with replacement), calculated the average value of each coefficient in each partition, and summed the coefficients for each regression parameter to obtain the cumulative values from 1 to 48 months. Results: The cumulative, 48-month, average cost was $67,796 (95% confidence interval [CI] $58,454-$78,291) with no censoring, $66,313 (95% CI $54,975-$80,074) with administrative censoring, and $66,765 (95% CI $54,510-$81, 843) with administrative plus dropout censoring. In multivariate analysis, chemotherapy was associated with increased cost of $25,325 (95% CI $17,549-$32,827) compared with $28,937 (95% CI $20,510-$37,088) with administrative censoring and $29,593 ($20,564-$39,399) with administrative plus dropout censoring. Adding censoring to the chemotherapy group resulted in less accurate IPW estimates. This was ameliorated, however, by applying IPW within treatment groups. Conclusion: IPW is a consistent estimator of population mean costs if the weight is correctly specified. If the censoring distribution depends on some covariates, a model that accommodates this dependency must be correctly specified in IPW to obtain accurate estimates.

Original languageEnglish (US)
Pages (from-to)656-663
Number of pages8
JournalValue in Health
Volume15
Issue number5
DOIs
StatePublished - Jul 2012

Fingerprint

Medicare
Least-Squares Analysis
Regression Analysis
Costs and Cost Analysis
Confidence Intervals
Drug Therapy
Epidemiology
Multivariate Analysis
Databases
Breast Neoplasms
Weights and Measures
Population

Keywords

  • accuracy and precision
  • cost analysis
  • inverse probability weighting
  • observational data

ASJC Scopus subject areas

  • Health Policy
  • Public Health, Environmental and Occupational Health

Cite this

Inverse probability weighted least squares regression in the analysis of time-censored cost data : An evaluation of the approach using SEER-medicare. / Griffiths, Robert I.; Gleeson, Michelle L.; Danese, Mark D.; O'Hagan, Anthony.

In: Value in Health, Vol. 15, No. 5, 07.2012, p. 656-663.

Research output: Contribution to journalArticle

Griffiths, Robert I. ; Gleeson, Michelle L. ; Danese, Mark D. ; O'Hagan, Anthony. / Inverse probability weighted least squares regression in the analysis of time-censored cost data : An evaluation of the approach using SEER-medicare. In: Value in Health. 2012 ; Vol. 15, No. 5. pp. 656-663.
@article{2197e7250437401bafc2ce84b1131862,
title = "Inverse probability weighted least squares regression in the analysis of time-censored cost data: An evaluation of the approach using SEER-medicare",
abstract = "Objectives: To assess the accuracy and precision of inverse probability weighted (IPW) least squares regression analysis for censored cost data. Methods: By using Surveillance, Epidemiology, and End Results-Medicare, we identified 1500 breast cancer patients who died and had complete cost information within the database. Patients were followed for up to 48 months (partitions) after diagnosis, and their actual total cost was calculated in each partition. We then simulated patterns of administrative and dropout censoring and also added censoring to patients receiving chemotherapy to simulate comparing a newer to older intervention. For each censoring simulation, we performed 1000 IPW regression analyses (bootstrap, sampling with replacement), calculated the average value of each coefficient in each partition, and summed the coefficients for each regression parameter to obtain the cumulative values from 1 to 48 months. Results: The cumulative, 48-month, average cost was $67,796 (95{\%} confidence interval [CI] $58,454-$78,291) with no censoring, $66,313 (95{\%} CI $54,975-$80,074) with administrative censoring, and $66,765 (95{\%} CI $54,510-$81, 843) with administrative plus dropout censoring. In multivariate analysis, chemotherapy was associated with increased cost of $25,325 (95{\%} CI $17,549-$32,827) compared with $28,937 (95{\%} CI $20,510-$37,088) with administrative censoring and $29,593 ($20,564-$39,399) with administrative plus dropout censoring. Adding censoring to the chemotherapy group resulted in less accurate IPW estimates. This was ameliorated, however, by applying IPW within treatment groups. Conclusion: IPW is a consistent estimator of population mean costs if the weight is correctly specified. If the censoring distribution depends on some covariates, a model that accommodates this dependency must be correctly specified in IPW to obtain accurate estimates.",
keywords = "accuracy and precision, cost analysis, inverse probability weighting, observational data",
author = "Griffiths, {Robert I.} and Gleeson, {Michelle L.} and Danese, {Mark D.} and Anthony O'Hagan",
year = "2012",
month = "7",
doi = "10.1016/j.jval.2012.03.1388",
language = "English (US)",
volume = "15",
pages = "656--663",
journal = "Value in Health",
issn = "1098-3015",
publisher = "Elsevier Limited",
number = "5",

}

TY - JOUR

T1 - Inverse probability weighted least squares regression in the analysis of time-censored cost data

T2 - An evaluation of the approach using SEER-medicare

AU - Griffiths, Robert I.

AU - Gleeson, Michelle L.

AU - Danese, Mark D.

AU - O'Hagan, Anthony

PY - 2012/7

Y1 - 2012/7

N2 - Objectives: To assess the accuracy and precision of inverse probability weighted (IPW) least squares regression analysis for censored cost data. Methods: By using Surveillance, Epidemiology, and End Results-Medicare, we identified 1500 breast cancer patients who died and had complete cost information within the database. Patients were followed for up to 48 months (partitions) after diagnosis, and their actual total cost was calculated in each partition. We then simulated patterns of administrative and dropout censoring and also added censoring to patients receiving chemotherapy to simulate comparing a newer to older intervention. For each censoring simulation, we performed 1000 IPW regression analyses (bootstrap, sampling with replacement), calculated the average value of each coefficient in each partition, and summed the coefficients for each regression parameter to obtain the cumulative values from 1 to 48 months. Results: The cumulative, 48-month, average cost was $67,796 (95% confidence interval [CI] $58,454-$78,291) with no censoring, $66,313 (95% CI $54,975-$80,074) with administrative censoring, and $66,765 (95% CI $54,510-$81, 843) with administrative plus dropout censoring. In multivariate analysis, chemotherapy was associated with increased cost of $25,325 (95% CI $17,549-$32,827) compared with $28,937 (95% CI $20,510-$37,088) with administrative censoring and $29,593 ($20,564-$39,399) with administrative plus dropout censoring. Adding censoring to the chemotherapy group resulted in less accurate IPW estimates. This was ameliorated, however, by applying IPW within treatment groups. Conclusion: IPW is a consistent estimator of population mean costs if the weight is correctly specified. If the censoring distribution depends on some covariates, a model that accommodates this dependency must be correctly specified in IPW to obtain accurate estimates.

AB - Objectives: To assess the accuracy and precision of inverse probability weighted (IPW) least squares regression analysis for censored cost data. Methods: By using Surveillance, Epidemiology, and End Results-Medicare, we identified 1500 breast cancer patients who died and had complete cost information within the database. Patients were followed for up to 48 months (partitions) after diagnosis, and their actual total cost was calculated in each partition. We then simulated patterns of administrative and dropout censoring and also added censoring to patients receiving chemotherapy to simulate comparing a newer to older intervention. For each censoring simulation, we performed 1000 IPW regression analyses (bootstrap, sampling with replacement), calculated the average value of each coefficient in each partition, and summed the coefficients for each regression parameter to obtain the cumulative values from 1 to 48 months. Results: The cumulative, 48-month, average cost was $67,796 (95% confidence interval [CI] $58,454-$78,291) with no censoring, $66,313 (95% CI $54,975-$80,074) with administrative censoring, and $66,765 (95% CI $54,510-$81, 843) with administrative plus dropout censoring. In multivariate analysis, chemotherapy was associated with increased cost of $25,325 (95% CI $17,549-$32,827) compared with $28,937 (95% CI $20,510-$37,088) with administrative censoring and $29,593 ($20,564-$39,399) with administrative plus dropout censoring. Adding censoring to the chemotherapy group resulted in less accurate IPW estimates. This was ameliorated, however, by applying IPW within treatment groups. Conclusion: IPW is a consistent estimator of population mean costs if the weight is correctly specified. If the censoring distribution depends on some covariates, a model that accommodates this dependency must be correctly specified in IPW to obtain accurate estimates.

KW - accuracy and precision

KW - cost analysis

KW - inverse probability weighting

KW - observational data

UR - http://www.scopus.com/inward/record.url?scp=84864580199&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864580199&partnerID=8YFLogxK

U2 - 10.1016/j.jval.2012.03.1388

DO - 10.1016/j.jval.2012.03.1388

M3 - Article

C2 - 22867774

AN - SCOPUS:84864580199

VL - 15

SP - 656

EP - 663

JO - Value in Health

JF - Value in Health

SN - 1098-3015

IS - 5

ER -