Estimating cost-effectiveness from claims and registry data with measured and unmeasured confounders

Elizabeth A. Handorf, Daniel F. Heitjan, Justin E. Bekelman, Nandita Mitra

Research output: Contribution to journalArticle

Abstract

The analysis of observational data to determine the cost-effectiveness of medical treatments is complicated by the need to account for skewness, censoring, and the effects of measured and unmeasured confounders. We quantify cost-effectiveness as the Net Monetary Benefit (NMB), a linear combination of the treatment effects on cost and effectiveness that denominates utility in monetary terms. We propose a parametric estimation approach that describes cost with a Gamma generalized linear model and survival time (the canonical effectiveness variable) with a Weibull accelerated failure time model. To account for correlation between cost and survival, we propose a bootstrap procedure to compute confidence intervals for NMB. To examine sensitivity to unmeasured confounders, we derive simple approximate relationships between naïve parameters, assuming only measured confounders, and the values those parameters would take if there was further adjustment for a single unmeasured confounder with a specified distribution. A simulation study shows that the method returns accurate estimates for treatment effects on cost, survival, and NMB under the assumed model. We apply our method to compare two treatments for Stage II/III bladder cancer, concluding that the NMB is sensitive to hypothesized unmeasured confounders that represent smoking status and personal income.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Cost-effectiveness
Cost-Benefit Analysis
Registries
Costs and Cost Analysis
Costs
Treatment Effects
Bladder Cancer
Accelerated Failure Time Model
Weibull Model
Parametric Estimation
Urinary Bladder Neoplasms
Smoking
Survival Time
Linear Models
Generalized Linear Model
Skewness
Censoring
Confidence Intervals
Bootstrap
Confidence interval

Keywords

  • cost-effectiveness analysis
  • Economic analysis
  • observational studies
  • sensitivity analysis
  • unmeasured confounding

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

Estimating cost-effectiveness from claims and registry data with measured and unmeasured confounders. / Handorf, Elizabeth A.; Heitjan, Daniel F.; Bekelman, Justin E.; Mitra, Nandita.

In: Statistical Methods in Medical Research, 01.01.2018.

Research output: Contribution to journalArticle

@article{04f65c0f5f664931bf4fd75d08d74185,
title = "Estimating cost-effectiveness from claims and registry data with measured and unmeasured confounders",
abstract = "The analysis of observational data to determine the cost-effectiveness of medical treatments is complicated by the need to account for skewness, censoring, and the effects of measured and unmeasured confounders. We quantify cost-effectiveness as the Net Monetary Benefit (NMB), a linear combination of the treatment effects on cost and effectiveness that denominates utility in monetary terms. We propose a parametric estimation approach that describes cost with a Gamma generalized linear model and survival time (the canonical effectiveness variable) with a Weibull accelerated failure time model. To account for correlation between cost and survival, we propose a bootstrap procedure to compute confidence intervals for NMB. To examine sensitivity to unmeasured confounders, we derive simple approximate relationships between na{\"i}ve parameters, assuming only measured confounders, and the values those parameters would take if there was further adjustment for a single unmeasured confounder with a specified distribution. A simulation study shows that the method returns accurate estimates for treatment effects on cost, survival, and NMB under the assumed model. We apply our method to compare two treatments for Stage II/III bladder cancer, concluding that the NMB is sensitive to hypothesized unmeasured confounders that represent smoking status and personal income.",
keywords = "cost-effectiveness analysis, Economic analysis, observational studies, sensitivity analysis, unmeasured confounding",
author = "Handorf, {Elizabeth A.} and Heitjan, {Daniel F.} and Bekelman, {Justin E.} and Nandita Mitra",
year = "2018",
month = "1",
day = "1",
doi = "10.1177/0962280218759137",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - Estimating cost-effectiveness from claims and registry data with measured and unmeasured confounders

AU - Handorf, Elizabeth A.

AU - Heitjan, Daniel F.

AU - Bekelman, Justin E.

AU - Mitra, Nandita

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The analysis of observational data to determine the cost-effectiveness of medical treatments is complicated by the need to account for skewness, censoring, and the effects of measured and unmeasured confounders. We quantify cost-effectiveness as the Net Monetary Benefit (NMB), a linear combination of the treatment effects on cost and effectiveness that denominates utility in monetary terms. We propose a parametric estimation approach that describes cost with a Gamma generalized linear model and survival time (the canonical effectiveness variable) with a Weibull accelerated failure time model. To account for correlation between cost and survival, we propose a bootstrap procedure to compute confidence intervals for NMB. To examine sensitivity to unmeasured confounders, we derive simple approximate relationships between naïve parameters, assuming only measured confounders, and the values those parameters would take if there was further adjustment for a single unmeasured confounder with a specified distribution. A simulation study shows that the method returns accurate estimates for treatment effects on cost, survival, and NMB under the assumed model. We apply our method to compare two treatments for Stage II/III bladder cancer, concluding that the NMB is sensitive to hypothesized unmeasured confounders that represent smoking status and personal income.

AB - The analysis of observational data to determine the cost-effectiveness of medical treatments is complicated by the need to account for skewness, censoring, and the effects of measured and unmeasured confounders. We quantify cost-effectiveness as the Net Monetary Benefit (NMB), a linear combination of the treatment effects on cost and effectiveness that denominates utility in monetary terms. We propose a parametric estimation approach that describes cost with a Gamma generalized linear model and survival time (the canonical effectiveness variable) with a Weibull accelerated failure time model. To account for correlation between cost and survival, we propose a bootstrap procedure to compute confidence intervals for NMB. To examine sensitivity to unmeasured confounders, we derive simple approximate relationships between naïve parameters, assuming only measured confounders, and the values those parameters would take if there was further adjustment for a single unmeasured confounder with a specified distribution. A simulation study shows that the method returns accurate estimates for treatment effects on cost, survival, and NMB under the assumed model. We apply our method to compare two treatments for Stage II/III bladder cancer, concluding that the NMB is sensitive to hypothesized unmeasured confounders that represent smoking status and personal income.

KW - cost-effectiveness analysis

KW - Economic analysis

KW - observational studies

KW - sensitivity analysis

KW - unmeasured confounding

UR - http://www.scopus.com/inward/record.url?scp=85043720901&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043720901&partnerID=8YFLogxK

U2 - 10.1177/0962280218759137

DO - 10.1177/0962280218759137

M3 - Article

C2 - 29468944

AN - SCOPUS:85043720901

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -