Using claims-based measures to predict performance status score in patients with lung cancer

Ramzi G. Salloum, Thomas J Smith, Gail A. Jensen, Jennifer Elston Lafata

Research output: Contribution to journalArticle

Abstract

BACKGROUND: Performance status (PS) is a good prognostic factor in lung cancer and is used to assess chemotherapy appropriateness. Researchers studying chemotherapy use are often hindered by the unavailability of PS in automated data sources. To the authors' knowledge, no attempts have been made to estimate PS using claims-based measures. The current study explored the ability to estimate PS using routinely available measures. METHODS: A cohort of insured patients aged ≥50 years who were diagnosed with American Joint Committee on Cancer stage II through IV lung cancer between 2000 and 2007 was identified via a tumor registry (n = 552). PS was abstracted from medical records. Automated medical and pharmaceutical claims from the year preceding diagnosis were linked to tumor registry data. A logistic regression model was fit to estimate good versus poor PS in a random half of the sample. C statistics, sensitivity, specificity, and R2 were used to compare the predictive ability of models that included demographic factors, comorbidity measures, and claims-based utilization variables. Model fit was evaluated in the other half of the sample. RESULTS: PS was available in 80% of medical records. The multivariable regression model predicted good PS with high sensitivity (0.88 or 0.94 depending on how good PS was defined), but moderate specificity (0.45 or 0.32) with a 0.50 prediction cutoff, and good sensitivity (0.64 or 0.83) and specificity (0.69 or 0.55) when the cutoff was 0.70. The goodness-of-fit c statistic was 0.76 or 0.78. CONCLUSIONS: PS can be estimated, with some accuracy, using claims-based measures. Emphasis should be placed on documenting PS in medical records and tumor registries.

Original languageEnglish (US)
Pages (from-to)1038-1048
Number of pages11
JournalCancer
Volume117
Issue number5
DOIs
StatePublished - Mar 1 2011
Externally publishedYes

Fingerprint

Lung Neoplasms
Medical Records
Registries
Neoplasms
Logistic Models
Drug Therapy
Information Storage and Retrieval
Comorbidity
Research Personnel
Demography
Sensitivity and Specificity
Pharmaceutical Preparations

Keywords

  • Automated claims
  • Chemotherapy
  • Lung cancer
  • Medical chart
  • Performance status
  • Tumor registry

ASJC Scopus subject areas

  • Cancer Research
  • Oncology

Cite this

Using claims-based measures to predict performance status score in patients with lung cancer. / Salloum, Ramzi G.; Smith, Thomas J; Jensen, Gail A.; Lafata, Jennifer Elston.

In: Cancer, Vol. 117, No. 5, 01.03.2011, p. 1038-1048.

Research output: Contribution to journalArticle

Salloum, Ramzi G. ; Smith, Thomas J ; Jensen, Gail A. ; Lafata, Jennifer Elston. / Using claims-based measures to predict performance status score in patients with lung cancer. In: Cancer. 2011 ; Vol. 117, No. 5. pp. 1038-1048.
@article{cd8f0cde4095429e8f2cc42256dbf7bf,
title = "Using claims-based measures to predict performance status score in patients with lung cancer",
abstract = "BACKGROUND: Performance status (PS) is a good prognostic factor in lung cancer and is used to assess chemotherapy appropriateness. Researchers studying chemotherapy use are often hindered by the unavailability of PS in automated data sources. To the authors' knowledge, no attempts have been made to estimate PS using claims-based measures. The current study explored the ability to estimate PS using routinely available measures. METHODS: A cohort of insured patients aged ≥50 years who were diagnosed with American Joint Committee on Cancer stage II through IV lung cancer between 2000 and 2007 was identified via a tumor registry (n = 552). PS was abstracted from medical records. Automated medical and pharmaceutical claims from the year preceding diagnosis were linked to tumor registry data. A logistic regression model was fit to estimate good versus poor PS in a random half of the sample. C statistics, sensitivity, specificity, and R2 were used to compare the predictive ability of models that included demographic factors, comorbidity measures, and claims-based utilization variables. Model fit was evaluated in the other half of the sample. RESULTS: PS was available in 80{\%} of medical records. The multivariable regression model predicted good PS with high sensitivity (0.88 or 0.94 depending on how good PS was defined), but moderate specificity (0.45 or 0.32) with a 0.50 prediction cutoff, and good sensitivity (0.64 or 0.83) and specificity (0.69 or 0.55) when the cutoff was 0.70. The goodness-of-fit c statistic was 0.76 or 0.78. CONCLUSIONS: PS can be estimated, with some accuracy, using claims-based measures. Emphasis should be placed on documenting PS in medical records and tumor registries.",
keywords = "Automated claims, Chemotherapy, Lung cancer, Medical chart, Performance status, Tumor registry",
author = "Salloum, {Ramzi G.} and Smith, {Thomas J} and Jensen, {Gail A.} and Lafata, {Jennifer Elston}",
year = "2011",
month = "3",
day = "1",
doi = "10.1002/cncr.25677",
language = "English (US)",
volume = "117",
pages = "1038--1048",
journal = "Cancer",
issn = "0008-543X",
publisher = "John Wiley and Sons Inc.",
number = "5",

}

TY - JOUR

T1 - Using claims-based measures to predict performance status score in patients with lung cancer

AU - Salloum, Ramzi G.

AU - Smith, Thomas J

AU - Jensen, Gail A.

AU - Lafata, Jennifer Elston

PY - 2011/3/1

Y1 - 2011/3/1

N2 - BACKGROUND: Performance status (PS) is a good prognostic factor in lung cancer and is used to assess chemotherapy appropriateness. Researchers studying chemotherapy use are often hindered by the unavailability of PS in automated data sources. To the authors' knowledge, no attempts have been made to estimate PS using claims-based measures. The current study explored the ability to estimate PS using routinely available measures. METHODS: A cohort of insured patients aged ≥50 years who were diagnosed with American Joint Committee on Cancer stage II through IV lung cancer between 2000 and 2007 was identified via a tumor registry (n = 552). PS was abstracted from medical records. Automated medical and pharmaceutical claims from the year preceding diagnosis were linked to tumor registry data. A logistic regression model was fit to estimate good versus poor PS in a random half of the sample. C statistics, sensitivity, specificity, and R2 were used to compare the predictive ability of models that included demographic factors, comorbidity measures, and claims-based utilization variables. Model fit was evaluated in the other half of the sample. RESULTS: PS was available in 80% of medical records. The multivariable regression model predicted good PS with high sensitivity (0.88 or 0.94 depending on how good PS was defined), but moderate specificity (0.45 or 0.32) with a 0.50 prediction cutoff, and good sensitivity (0.64 or 0.83) and specificity (0.69 or 0.55) when the cutoff was 0.70. The goodness-of-fit c statistic was 0.76 or 0.78. CONCLUSIONS: PS can be estimated, with some accuracy, using claims-based measures. Emphasis should be placed on documenting PS in medical records and tumor registries.

AB - BACKGROUND: Performance status (PS) is a good prognostic factor in lung cancer and is used to assess chemotherapy appropriateness. Researchers studying chemotherapy use are often hindered by the unavailability of PS in automated data sources. To the authors' knowledge, no attempts have been made to estimate PS using claims-based measures. The current study explored the ability to estimate PS using routinely available measures. METHODS: A cohort of insured patients aged ≥50 years who were diagnosed with American Joint Committee on Cancer stage II through IV lung cancer between 2000 and 2007 was identified via a tumor registry (n = 552). PS was abstracted from medical records. Automated medical and pharmaceutical claims from the year preceding diagnosis were linked to tumor registry data. A logistic regression model was fit to estimate good versus poor PS in a random half of the sample. C statistics, sensitivity, specificity, and R2 were used to compare the predictive ability of models that included demographic factors, comorbidity measures, and claims-based utilization variables. Model fit was evaluated in the other half of the sample. RESULTS: PS was available in 80% of medical records. The multivariable regression model predicted good PS with high sensitivity (0.88 or 0.94 depending on how good PS was defined), but moderate specificity (0.45 or 0.32) with a 0.50 prediction cutoff, and good sensitivity (0.64 or 0.83) and specificity (0.69 or 0.55) when the cutoff was 0.70. The goodness-of-fit c statistic was 0.76 or 0.78. CONCLUSIONS: PS can be estimated, with some accuracy, using claims-based measures. Emphasis should be placed on documenting PS in medical records and tumor registries.

KW - Automated claims

KW - Chemotherapy

KW - Lung cancer

KW - Medical chart

KW - Performance status

KW - Tumor registry

UR - http://www.scopus.com/inward/record.url?scp=79952035882&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952035882&partnerID=8YFLogxK

U2 - 10.1002/cncr.25677

DO - 10.1002/cncr.25677

M3 - Article

VL - 117

SP - 1038

EP - 1048

JO - Cancer

JF - Cancer

SN - 0008-543X

IS - 5

ER -