The Geographic Distribution of Videolaryngostroboscopy in the United States

Abstract

To study the geographic utilization of videolaryngostroboscopy (VLS) with the hypothesis that office-based voice care is unevenly distributed across the United States. This is a cross-sectional database analysis of Medicare beneficiaries. The Centers for Medicare and Medicaid Services Provider Utilization and Payment Data Physician and Other Supplier Public Use File from 2012 to 2017 was analyzed to evaluate VLS utilization. VLS distribution was assessed by calculating the density of VLS in each of the 306 hospital referral regions (HRRs) nationally. Associations between VLS density and population demographics and health system factors were assessed using Pearson correlation and multivariate regression analyses. In total, 957,648 outpatient VLS were billed to Medicare part B between 2012 and 2017. The annual VLS density per HRR ranged from 0 to 38.2 per 1,000 enrollees. Pearson correlation revealed positive correlations between VLS density and number of Medicare enrollees (r=0.2584, P < 0.001), income (r=0.1913, P=0.0008), education (r=0.2089, P=0.0002), and density of otolaryngologists (r=0.1589, P=0.0053) and medical specialists (r=0.2326, P < 0.0001). A negative Pearson correlation was observed between VLS density and percent male (r=-0.1338, P=0.0192) and Medicare mortality rate (r=-0.1628, P=0.0043). On multivariate regression positive associations between VLS and number of Medicare enrollees (P=0.002) and otolaryngologists (P=0.049), and negative association with Medicare mortality rates (P=0.032) remained significant. The distribution of office-based voice care varies widely across the country, even when analysis by HRR should have homogenized access to specialty care. Greater availability of VLS is seen in HRRs with more Medicare enrollees, greater density of otolaryngologists, and lower mortality rates.