Cardiovascular

Provider networks for pulmonary hypertension in Massachusetts: implications for improving referrals to expert care.

TL;DR

Few providers caring for patients with pulmonary hypertension were connected to PH experts, with significant gaps driven by provider specialty, geographic location, and limited PH panel volume, suggesting multifaceted strategies are needed to improve referral rates.

Key Findings

Few providers shared patients with pulmonary hypertension experts across all constructed networks.

  • Only 31% of providers in the All-PH network were connected to PH experts
  • 35% of providers in the PAH network were connected to PH experts
  • 19% of providers in the CTEPH network were connected to PH experts
  • Connection to a PH expert was defined as sharing at least 1 patient with an expert

Primary care providers had significantly decreased odds of being connected to PH experts compared to nonexpert pulmonologists.

  • Primary care providers had 59% decreased odds of PH expert connection compared to nonexpert pulmonologists
  • Adjusted odds ratio was 0.41 (95% confidence interval, 0.32–0.51)
  • This finding was observed within the All-PH network multivariable regression model
  • Findings were similar among the PAH and CTEPH networks

A large cohort of patients with pulmonary hypertension and their providers was identified in Massachusetts over a five-year period.

  • 8,970 patients with PH were identified from the Massachusetts All-Payer Claims Database in 2014–2018
  • 4,766 providers were identified as caring for these patients
  • 1,768 patients (19.7%) had PAH risk factors
  • 2,164 patients (24.1%) had CTEPH risk factors
  • Patient-level data were linked with provider-level data to construct care networks

Providers practicing outside the greater Boston area were less likely to be connected to a PH expert.

  • Geographic practice location was a significant provider-level factor associated with connection to PH experts
  • This association was observed in the All-PH network multivariable regression model
  • Findings were similar among the PAH and CTEPH networks
  • This was identified as a potential geographic barrier to expert care

Providers with smaller PH panel volumes were less likely to be connected to a PH expert.

  • PH panel volume was a significant provider-level factor associated with connection to PH experts in multivariable regression
  • This suggests limited provider experience with PH may contribute to gaps in expert connections
  • The finding was consistent across the All-PH, PAH, and CTEPH networks

Social network analysis was used to characterize provider-level care networks and identify factors associated with expert connections for pulmonary hypertension patients.

  • Three separate provider networks were constructed: All-PH, PAH, and CTEPH networks
  • The Massachusetts All-Payer Claims Database was the primary data source, linked with provider-level data
  • The provider-level outcome was connection to PH experts, defined as sharing at least 1 patient with an expert
  • Multivariable regression models were conducted within each network to assess provider specialty, practice location, and PH panel volume

Despite guideline recommendations, significant gaps in referral to expert PH care exist even among providers caring for high-risk patients.

  • Guidelines clearly recommend referral to expert care for patients with PAH and CTEPH
  • Gaps in expert connections were found even in the PAH and CTEPH subnetworks, representing higher-risk patient populations
  • The authors attributed these gaps in part to limited provider experience, geographic barriers, and disconnected providers and care networks
  • The authors concluded that multifaceted strategies may be needed to improve referral rates

Have a question about this study?

Citation

Gillmeyer K, Rinne S, Klings E, Elwy A, Wiener R. (2026). Provider networks for pulmonary hypertension in Massachusetts: implications for improving referrals to expert care.. Annals of the American Thoracic Society. https://doi.org/10.1093/annalsats/aaoaf039