Bayesian hierarchical network autocorrelation models for estimating direct and indirect effects of peer hospitals on outcomes of hospitalized patients
2024

Estimating Effects of Peer Hospitals on Patient Outcomes

Sample size: 1306 publication Evidence: moderate

Author Information

Author(s): Chen Guanqing, O’Malley A. James

Primary Institution: Geisel School of Medicine at Dartmouth

Hypothesis

How do peer hospitals influence the outcomes of hospitalized patients through direct and indirect effects?

Conclusion

The study found that peer hospitals' adoption of robotic surgery may indirectly associate with shorter hospital stays for patients, although the results were not statistically significant.

Supporting Evidence

  • The study utilized a cohort of Medicare beneficiaries to analyze hospital peer effects.
  • Bayesian hierarchical models were developed to assess direct and indirect peer effects.
  • The findings suggest that peer hospitals' adoption of robotic surgery may influence patient outcomes.

Takeaway

This study looks at how hospitals can affect each other's patients, especially when using new technologies like robotic surgery.

Methodology

The study developed two hierarchical network autocorrelation models using Bayesian methods to analyze the effects of peer hospitals on patient outcomes.

Potential Biases

The study acknowledges potential biases related to homophily among hospitals.

Limitations

The findings were inconclusive due to the high density of the network, which limited the information available about the peer effects.

Participant Demographics

The cohort included 45 hospitals and 1,306 patients, primarily Medicare beneficiaries undergoing prostatectomy.

Statistical Information

Confidence Interval

ρ: -0.525 (-1.481, 0.804); α: 1.355 (-2.539, 4.168)

Digital Object Identifier (DOI)

10.1007/s41109-024-00627-1

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication