Abstract
This cross-sectional study analyzes Medicare Part B claims to assess the association of changes in hospital charge master markups over time with hospital characteristics.
Hospitals send surprise medical bills to patients who, often unknowingly, receive care from an out-of-network physician; these bills are typically several times the cost of care from an in-network physician and can be financially devastating.1,2 Prior research has shown that hospital charge master prices, specifically the “markups” greater than Medicare reimbursement rates in medical bills for out-of-network care,1 vary widely across hospitals and departments within hospitals. As of September 2019, the US Congress is considering several approaches to protect patients from surprise billing, including limiting the degree of markup.3 This cross-sectional study characterized markups from 2012 through 2016 and examined associations of markup with hospital characteristics by focusing on charges in 2 specialties common in surprise medical billing, emergency medicine and anesthesiology,1 using internal medicine as a reference.
Methods
Using previously reported methodology,1 I analyzed Medicare Part B claims from January 2012 through December 2016 and limited inclusion to hospitals that provided emergency medicine, anesthesiology, and internal medicine services. Markups were quantified using the ratio of the charge relative to the Medicare reimbursed amount. For example, a markup ratio of 4.5 means that for every $100 in Medicare reimbursement, the hospital charged $450 to out-of-network patients, a 350% markup.
To assess the association of changes in markups over time with hospital characteristics, I used a panel regression and adjusted for hospital characteristics from the American Hospital Association directory.1 Adjusted annual growth rates are reported as a percentage of the Medicare reimbursement. P values from the regression model are from the interaction term of the year by the hospital characteristic variables; a significant P value indicates that subgroups of hospitals changed their markups differently over time. This study was deemed exempt from review and approval by the Johns Hopkins Medicine Institutional Review Board because it used publicly available data without identifiable patient information. All reported P values were 2-tailed, and P < .05 was considered to be statistically significant. Analysis was completed using Alteryx 2018 (Alteryx Inc) and Stata, version 14 (StataCorp). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Results
Among the 2042 hospitals included from all 50 states, 373 (18%) were for-profit hospitals, and 402 (20%) served more than 20% uninsured patients. From 2012 to 2016, markups increased substantially for emergency medicine (from a markup ratio of 3.9 to 5.1; adjusted annual growth rate on Medicare reimbursement, 28%) and anesthesiology (from a markup ratio of 6.1 to 7.4; adjusted annual growth rate, 32%) (Figure). By comparison, the internal medicine markup ratio increased from 2.1 to 2.4 (adjusted annual growth rate, 7%).
Figure. Hospitals’ Aggregated Markup Ratios and Adjusted Annual Growth in Charge Master Prices as a Percentage of Medicare Reimbursement, Stratified by Clinical Department, 2012 to 2016.
Adjusted annual growth rates are anesthesiology, 32%; emergency medicine, 28%; and internal medicine, 7%.
For emergency medicine, adjusted annual increases in charges were greatest among for-profit hospitals (46% vs 24%; P < .001), hospitals serving more uninsured patients (36% for hospitals serving ≥20% uninsured patients vs 21% for hospitals serving <10% uninsured patients; P < .001), and hospitals in the Southeast (42%; P < .001) (Table). Similar results were found for anesthesiology in that hospitals with the greatest increases were for-profit and in the Southeast.
Table. Adjusted Annual Percentage Change in Hospitals’ Aggregated Markup as a Percentage of Medicare Reimbursement, Stratified by Hospital Characteristics and Clinical Department, 2012 to 2016.
| Characteristics | Hospitals, No. (%) | Department, % (95% CI) | |||||
|---|---|---|---|---|---|---|---|
| Emergency Medicine | Anesthesiology | Internal Medicine | |||||
| Adjusted Annual Markup Growth Rate | P Valuea | Adjusted Annual Markup Growth Rate | P Valuea | Adjusted Annual Markup Growth Rate | P Valuea | ||
| All hospitals | 2042 (100) | 28 (27-30) | NA | 32 (30-34) | NA | 7 (6-7) | NA |
| Hospital size, No. of beds | |||||||
| <200 | 977 (48) | 32 (30-34) | <.001 | 30 (27-34) | .02 | 8 (7-8) | <.001 |
| 200-399 | 674 (33) | 29 (27-31) | 35 (32-38) | 7 (7-8) | |||
| ≥400 | 391 (19) | 19 (17-22) | 31 (27-35) | 5 (4-5) | |||
| For profit | |||||||
| No | 1669 (82) | 24 (23-26) | <.001 | 30 (28-32) | <.001 | 6 (6-6) | <.001 |
| Yes | 373 (18) | 46 (42-51) | 41 (37-46) | 11 (10-12) | |||
| Academic | |||||||
| No | 1228 (60) | 31 (30-33) | <.001 | 34 (32-37) | .06 | 8 (7-8) | <.001 |
| Yes | 814 (40) | 24 (22-26) | 29 (27-32) | 6 (5-7) | |||
| Urban | |||||||
| No | 1155 (57) | 27 (25-28) | .04 | 30 (28-33) | .03 | 6 (6-7) | .01 |
| Yes | 887 (43) | 30 (28-32) | 35 (32-38) | 8 (7-8) | |||
| Population living in poverty, % | |||||||
| <10 | 521 (26) | 25 (22-27) | .02 | 36 (32-40) | .02 | 8 (7-9) | .001 |
| 10-19 | 827 (41) | 29 (27-31) | 29 (26-32) | 7 (6-8) | |||
| ≥20 | 694 (34) | 30 (27-32) | 33 (29-36) | 6 (5-7) | |||
| Uninsured population, % | |||||||
| <10 | 566 (28) | 21 (19-23) | <.001 | 33 (29-36) | .11 | 6 (6-7) | .16 |
| 10-19 | 1074 (53) | 29 (27-31) | 30 (27-33) | 7 (7-8) | |||
| ≥20 | 402 (20) | 36 (32-40) | 38 (33-42) | 7 (6-8) | |||
| African American population, % | |||||||
| <20 | 1611 (79) | 27 (25-28) | <.001 | 29 (27-31) | <.001 | 7 (7-8) | .007 |
| ≥20 | 431 (21) | 33 (30-37) | 44 (40-49) | 6 (5-7) | |||
| Hispanic population, % | |||||||
| <20 | 1597 (78) | 28 (27-30) | .76 | 33 (31-35) | .13 | 7 (6-7) | .03 |
| ≥20 | 445 (22) | 29 (25-32) | 29 (25-33) | 8 (7-9) | |||
| Region | |||||||
| Northeast | 352 (17) | 26 (23-28) | <.001 | 27 (22-31) | <.001 | 7 (6-8) | .01 |
| Southeast | 772 (38) | 42 (39-45) | 42 (39-46) | 7 (7-8) | |||
| Midwest | 472 (23) | 21 (19-23) | 35 (31-39) | 6 (5-6) | |||
| West | 446 (22) | 15 (13-18) | 17 (14-20) | 7 (6-8) | |||
Abbreviation: NA, not applicable.
Shown are P values from the interaction term of the year by the hospital characteristic variables in a panel regression; a significant P value indicates that subgroups of hospitals changed their markups differently over time.
Discussion
From 2012 to 2016, adjusted annual increases in charge master prices in the 2 common surprise medical bill specialties analyzed in this study—28% in emergency medicine and 32% in anesthesiology—far exceeded economic inflation. These findings are particularly worrisome given that the incidence of emergency department surprise medical bills increased from 32% to 43% between 2010 and 2016, leading to greater patient financial liability.4 Patients can face legal action and wage garnishments if they are unable to pay these excessive prices.5 To date, the Commonwealth Fund has determined that only 13 states have legislation that meets its standard for comprehensive consumer protection to mitigate surprise medical bills,6 so there is growing urgency for prompt national legislation. In this analysis, emergency departments serving a higher percentage of uninsured patients raised their charges more, a potential sign of cost shifting under financial stress.1 Policy makers should consider how legislation limiting the degree of markup3 could give insurers unfair negotiating leverage over these hospitals.
The study has limitations to consider,1 including that claims reflect the billed charge, which some patients may have negotiated down. Second, this study was limited to physician professional fees, which do not account for facility fees charged by the hospitals. Finally, the claims did not differentiate between charges set by hospitals for employed physicians and those that may have been determined independently by contracted physicians.
References
- 1.Xu T, Park A, Bai G, et al. Variation in emergency department vs internal medicine excess charges in the United States. JAMA Intern Med. 2017;177(8):1139-1145. doi: 10.1001/jamainternmed.2017.1598 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Terhune C. His $109K heart attack bill is now down to $332 after NPR told his story. https://www.npr.org/sections/health-shots/2018/08/31/643342598/his-109k-heart-attack-bill-is-now-down-to-332-after-npr-told-his-story. Accessed October 1, 2019.
- 3.Corlette S, Hoadley J, Lucia K Successfully splitting the baby: design considerations for federal balance billing legislation. https://www.healthaffairs.org/do/10.1377/hblog20190708.627390/full/. Accessed October 1, 2019.
- 4.Sun EC, Mello MM, Moshfegh J, Baker LC. Assessment of out-of-network billing for privately insured patients receiving care in in-network hospitals [published online August 12, 2019]. JAMA Intern Med. doi: 10.1001/jamainternmed.2019.3451 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bruhn WE, Rutkow L, Wang P, et al. Prevalence and characteristics of Virginia hospitals suing patients and garnishing wages for unpaid medical bills. JAMA. 2019;322(7):691-692. doi: 10.1001/jama.2019.9144 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hoadley J, Lucia K, Kona M States are taking new steps to protect consumers from balance billing, but federal action is necessary to fill gaps. https://www.commonwealthfund.org/blog/2019/states-are-taking-new-steps-protect-consumers-balance-billing-federal-action-necessary. Accessed October 1, 2019.
