Adherence to a Federal Hospital Price Transparency Rule and Associated Financial and Marketplace Factors

Abstract

This study evaluates US hospitals’ disclosure of standard service charges as mandated by a federal price transparency rule and hospital characteristics among acute care hospitals.

The federal Hospital Price Transparency Final Rule aims to increase health care price transparency and facilitate patient price shopping online. Hospitals are required to disclose 5 types of standard charges for all services in an accessible file and provide a consumer-friendly display for at least 300 shoppable services.¹ We evaluated adherence 6 to 9 months after the final rule effective date (January 1, 2021) across all US hospitals and its association with market- and hospital-level characteristics across acute care hospitals.

Methods

We collected data on hospital characteristics and adherence to the final rule between July 1 and September 30, 2021, for all US hospitals that were registered with the Centers for Medicare & Medicaid Services and with an identifiable website.² We collected data on whether each hospital had posted all 5 required price types (gross charges, discounted prices, payer-specific negotiated prices, and minimum and maximum negotiated prices) in a machine-readable file, and a separate accessible display or price estimator for at least 300 shoppable items. Final rule adherence required that both conditions be met. Characteristics of all hospitals were compared between nonadherent and adherent facilities by calculating standardized differences, with values greater than 0.1 considered significant. Our measure of inpatient hospital market concentration, the Herfindahl-Hirschman Index (HHI), was collected for 185 of 929 core-based statistical areas using 2019 data.³ The HHI categories include unconcentrated, moderately concentrated, and highly or very highly concentrated, with greater concentration denoting fewer hospitals accounting for a larger share of admissions within a geographic region. Examination of characteristics associated with final rule adherence was restricted to acute care hospitals because different hospital types vary in characteristics and services provided. The missing indicator method was used for hospitals without HHI data. Information on hospital revenues based on 2020 Medicare Cost Reports and number of patient-days for acute care hospitals was obtained from the American Hospital Directory.⁴ Logistic regression analysis included total revenue quartiles, revenue per patient-day quartiles, HHI categories, urbanicity, hospital size, emergency services, and hospital ownership as independent variables. Statistical significance was defined as a 95% CI that excluded 1. Analyses were conducted with SPSS version 23 (SPSS Inc). See the eMethods in the Supplement for additional details.

Results

Across 5239 total hospitals, 729 (13.9%) had an adherent machine-readable file but no shoppable display, 1542 (29.4%) had an adherent shoppable display but no machine-readable file, and 300 (5.7%) had both. There were 2668 hospitals (50.9%) without an adherent machine-readable file or a shoppable display. There was a significant difference in the proportion of adherent vs nonadherent facilities that were in unconcentrated and highly or very highly concentrated markets (Table 1).

Table 1. Characteristics of All Hospitals by Adherence to the Final Rule.

Hospital characteristics	No. (%)		Standardized difference
	Nonadherent (n = 4939)	Adherent (n = 300)
Urban vs rurala
Rural	2188 (44.3)	120 (40)	0.09
Urban	2751 (55.7)	180 (60)	0.09
Hospital size (No. of beds)a
Small (<100)	2550 (51.6)	163 (54.3)	0.05
Medium (100-300)	1456 (29.5)	82 (27.3)	0.05
Large (>300)	933 (18.9)	55 (18.3)	0.02
Emergency services capablea
No	776 (15.7)	58 (19.3)	0.09
Yes	4163 (84.3)	242 (81.7)	0.07
Hospital typea
Acute care	3053 (61.8)	170 (56.7)	0.10
Psychiatric	533 (10.8)	41 (13.7)	0.09
Critical access	1268 (25.7)	80 (26.7)	0.02
Children’s	85 (1.7)	9 (3.0)	0.09
Hospital ownershipa
Local government	1094 (22.2)	67 (22.3)	0.00
Federal government	55 (1.1)	4 (1.3)	0.02
Private nonprofit	2456 (49.7)	160 (53.3)	0.07
Religious	311 (6.3)	15 (5.0)	0.06
Private for-profit	1023 (20.7)	54 (18.0)	0.07
Herfindahl-Hirschman Index scoreb
Unconcentrated	642 (13.0)	59 (19.7)	0.18
Moderately concentrated	748 (15.1)	47 (15.7)	0.02
Highly or very highly concentrated	1087 (22.0)	51 (17.0)	0.13
Unclassified	2462 (49.8)	143 (47.7)	0.04

^a Hospital characteristics as defined by Centers for Medicare & Medicaid Services.

^b Market categories were based on definitions by the Federal Trade Commission. Markets with a Herfindahl-Hirschman Index score less than 1500 points were categorized as unconcentrated; those with a score between 1501 and 2500 points as moderately concentrated; and those with a score greater than 2501 points as highly or very highly concentrated.

There were 2783 of 3223 acute care hospitals (86%) with available revenue data. Total gross revenue had no significant association with final rule adherence (Table 2). In contrast, being in the first quartile (lowest) of revenue per patient-day was associated with greater rates of adherence than was being in other quartiles. Compared with being in unconcentrated markets, being in a moderately concentrated one (odds ratio, 0.58; 95% CI, 0.35-0.96) and highly or very highly concentrated one (odds ratio, 0.33; 95% CI, 0.19-0.56) was associated with worse adherence. Urban vs rural location was associated with better adherence to the final rule (odds ratio, 1.86; 95% CI, 1.08-3.17). Hospital size, emergency service capabilities, and hospital ownership were not associated with adherence.

Table 2. Factors Associated With Adherence to the Final Rule Among Acute Care Hospitals in a Multivariable Logistic Regression Analysis.

Hospital characteristics	No. of acute care hospitals	Odds ratio (95% CI)
Total revenue, quartilea
First (lowest)	696	1 [Reference]
Second	696	0.66 (0.38-1.14)
Third	696	0.91 (0.49-1.70)
Fourth (highest)	695	0.81 (0.42-1.96)
Total revenue per patient-day, quartileb
First (lowest)	696	1 [Reference]
Second	696	0.54 (0.34-0.86)
Third	696	0.55 (0.34-0.89)
Fourth (highest)	695	0.53 (0.32-0.90)
Herfindahl-Hirschman Index scorec
Unconcentrated	473	1 [Reference]
Moderately concentrated	527	0.58 (0.35-0.96)
Highly or very highly concentrated	720	0.33 (0.19-0.56)
Unclassified	1063	0.65 (0.40-1.05)
Urban vs rural
Rural	811	1 [Reference]
Urban	1972	1.86 (1.08-3.17)
Hospital size
Small	813	1 [Reference]
Medium	1135	0.63 (0.30-1.33)
Large	835	0.59 (0.23-1.55)
Emergency services capable
No	203	1 [Reference]
Yes	2580	1.58 (0.64-3.89)
Hospital ownership
Government	424	1 [Reference]
Private nonprofit	1497	1.52 (0.91-2.54)
Religious	227	0.98 (0.44-2.19)
Private for-profit	635	0.63 (0.32-1.27)

^a Total revenue was calculated as the sum of gross patient and nonpatient revenue from the hospital’s 2020 Medicare Cost Report.

^b Total revenue per patient-day was calculated by dividing total revenue by the number of total patient-days taken from the hospital’s 2020 Medicare Cost Report.

^c Market categories were based on definitions by the Federal Trade Commission. Markets with a Herfindahl-Hirschman Index score less than 1500 points were categorized as unconcentrated; those with a score between 1501 and 2500 points as moderately concentrated; and those with a score greater than 2501 points as highly or very highly concentrated.

Discussion

Adherence to the final rule price transparency mandate 6 to 9 months after its effective date was low. Acute care hospitals with lesser revenue per patient-day, within unconcentrated health care markets, and in urban areas were more likely to be transparent. Greater scrutiny of hospitals without these characteristics may be needed to ensure hospital price transparency. Because multiple factors affect revenue per patient-day, including patient acuity, operational expenses, and provision of specialty care, refining which financial determinants are associated with adherence is needed. Longer-term trends in hospital adherence and whether changes in penalties beginning in 2022 may lead to greater adherence remain to be elucidated.

Study limitations include that final rule adherence may have been underestimated, given that data abstraction was conducted during 3 months, and some hospitals may have disclosed standard charges during that time. Also, financial and HHI data for all hospitals were unavailable, as was revenue and marketplace concentration in non–acute care hospitals and in all geographic regions.

Section Editors: Jody W. Zylke, MD, Deputy Editor; Kristin Walter, MD, Associate Editor.

Supplement.

eMethods