Estimation of treatment effects in observational stroke care data: comparison of statistical approaches

Marzyeh Amini*, Nikki van Leeuwen, Frank Eijkenaar, Rob van de Graaf, Noor Samuels, Robert van Oostenbrugge, Ido R van den Wijngaard, Pieter Jan van Doormaal, Yvo B W E M Roos, Charles Majoie, Bob Roozenbeek, Diederik Dippel, James Burke, Hester F Lingsma, MR CLEAN Registry Investigators

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

INTRODUCTION: Various statistical approaches can be used to deal with unmeasured confounding when estimating treatment effects in observational studies, each with its own pros and cons. This study aimed to compare treatment effects as estimated by different statistical approaches for two interventions in observational stroke care data.

PATIENTS AND METHODS: We used prospectively collected data from the MR CLEAN registry including all patients (n = 3279) with ischemic stroke who underwent endovascular treatment (EVT) from 2014 to 2017 in 17 Dutch hospitals. Treatment effects of two interventions - i.e., receiving an intravenous thrombolytic (IVT) and undergoing general anesthesia (GA) before EVT - on good functional outcome (modified Rankin Scale ≤2) were estimated. We used three statistical regression-based approaches that vary in assumptions regarding the source of unmeasured confounding: individual-level (two subtypes), ecological, and instrumental variable analyses. In the latter, the preference for using the interventions in each hospital was used as an instrument.

RESULTS: Use of IVT (range 66-87%) and GA (range 0-93%) varied substantially between hospitals. For IVT, the individual-level (OR ~ 1.33) resulted in significant positive effect estimates whereas in instrumental variable analysis no significant treatment effect was found (OR 1.11; 95% CI 0.58-1.56). The ecological analysis indicated no statistically significant different likelihood (β = - 0.002%; P = 0.99) of good functional outcome at hospitals using IVT 1% more frequently. For GA, we found non-significant opposite directions of points estimates the treatment effect in the individual-level (ORs ~ 0.60) versus the instrumental variable approach (OR = 1.04). The ecological analysis also resulted in a non-significant negative association (0.03% lower probability).

DISCUSSION AND CONCLUSION: Both magnitude and direction of the estimated treatment effects for both interventions depend strongly on the statistical approach and thus on the source of (unmeasured) confounding. These issues should be understood concerning the specific characteristics of data, before applying an approach and interpreting the results. Instrumental variable analysis might be considered when unobserved confounding and practice variation is expected in observational multicenter studies.

Original languageEnglish
Article number103
Number of pages12
JournalBMC Medical Research Methodology
Volume22
Issue number1
DOIs
Publication statusPublished - 10 Apr 2022

Keywords

  • Endovascular Procedures/methods
  • Fibrinolytic Agents/therapeutic use
  • Humans
  • Stroke/drug therapy
  • Thrombolytic Therapy
  • Treatment Outcome
  • Instrumental variable
  • Ecological-analysis
  • PROPENSITY SCORE
  • Statistical approaches
  • INSTRUMENTAL VARIABLE METHODS
  • CONSCIOUS SEDATION
  • BINARY DATA
  • THROMBECTOMY
  • General anesthesia
  • ADJUSTMENT
  • GENERAL-ANESTHESIA
  • Acute ischemic stroke
  • ACUTE ISCHEMIC-STROKE
  • ENDOVASCULAR TREATMENT
  • Confounding by indication
  • Unmeasured confounding
  • OUTCOMES
  • Intravenous thrombolysis

Fingerprint

Dive into the research topics of 'Estimation of treatment effects in observational stroke care data: comparison of statistical approaches'. Together they form a unique fingerprint.

Cite this