Evaluating comparative effectiveness with observational data: Endoscopic ultrasound and survival in pancreatic cancer

Abhishek D. Parmar, Kristin M. Sheffield, Yimei Han, Gabriela M. Vargas, Praveen Guturu, Yong Fang Kuo, James S. Goodwin, Taylor S. Riall

Research output: Contribution to journalArticle

9 Scopus citations

Abstract

BACKGROUND A previous observational study reported that endoscopic ultrasound (EUS) is associated with improved survival in older patients with pancreatic cancer. The objective of this study was to reevaluate this association using different statistical methods to control for confounding and selection bias. METHODS Surveillance, Epidemiology, and End Results (SEER)-Medicare linked data (1992-2007) was used to identify patients with locoregional pancreatic cancer. Two-year survival in patients who did and did not receive EUS was compared by using standard Cox proportional hazards models, propensity score methodology, and instrumental variable analysis. RESULTS EUS was associated with improved survival in both unadjusted (hazard ratio [HR] = 0.67, 95% confidence interval [CI] = 0.63-0.72) and standard regression analyses (HR = 0.78, 95% CI = 0.73-0.84) which controlled for age, sex, race, marital status, tumor stage, SEER region, Charlson comorbidity, year of diagnosis, education, preoperative biliary stenting, chemotherapy, radiation, and pancreatic resection. Propensity score adjustment, matching, and stratification did not attenuate this survival benefit. In an instrumental variable analysis, the survival benefit was no longer observed (HR = 1.00, 95% CI = 0.73-1.36). CONCLUSIONS These results demonstrate the need to exercise caution in using administrative data to infer causal mortality benefits with diagnostic and/or treatment interventions in cancer research. Cancer 2013;119:3861-3869.

Original languageEnglish (US)
Pages (from-to)3861-3869
Number of pages9
JournalCancer
Volume119
Issue number21
DOIs
StatePublished - Nov 1 2013

    Fingerprint

Keywords

  • endoscopic ultrasound
  • instrumental variable
  • pancreatic cancer
  • propensity score
  • selection bias

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this