We find that under current regulatory and market conditions, demand for hospital-based emergency services is highly inelastic with respect to price, giving hospitals substantial pricing power over out-of-network emergency services.
Objectives: Empirical evaluation of market power that hospitals gain over health plans through hospitals’ ability to cancel their contracts with plans while keeping large shares of plans’ emergency patients and getting paid for them at above-market rates.
Study Design: Case-study analysis of 5 California hospitals that initially had contracts with most commercial health plans and then cancelled all those contracts at the same time.
Methods: We conducted a before-and-after case-study analysis comparing volume, price, and net revenues for the 5 study hospitals 3 years before and up to 4 years after the cancellation of their commercial contracts. The volume and price trends in study hospitals were compared with data on control hospitals in the same geographic area over the matching study period.
Results: Despite substantially increasing their prices on a noncontracted basis, the 5 study hospitals collectively retained 50% of their commercial health plan volume in first 2 years after the cancellation and 41% of their commercial volume in years 3 and 4, with net commercial revenues increasing as a result. At the same time, the simulated costs of treating the patients from out-of-network hospitals more than doubled for the health plans.
Conclusions: In hospital-payer negotiation, many hospitals have an upper hand: Their threat to retain large portions of their emergency patients and revenues after becoming out of network is credible and it imposes disproportionate costs on the payers, which partially explains the continuing rise in hospital prices.
Am J Manag Care. 2020;26(3):105-110. https://doi.org/10.37765/ajmc.2020.41929
Hospital pricing in the United States is highly complex. For most commercially insured Americans, the prices that they pay hospitals (or prices paid on their behalf) are determined by negotiations between their insurance company and hospitals. Health plans seek to pay the lowest prices possible to include hospitals in their preferred networks. Hospitals can agree to lower prices offered by the plan to stay in network in the hopes of receiving higher volume from the plan. Alternatively, hospitals, in negotiating for higher in-network contract rates, can threaten to cancel their contracts, become out of network for the plan, and then use their much higher billed charges as the price to bill the health plan for health plan patients using the hospital.
Several factors have made the threat of contract cancellation by hospitals more powerful. Most states, including California, adopted regulations (often referred to as “prudent layperson” laws) that require health plans to instruct their members to go to the nearest hospital in the event of a medical emergency, even if the nearest hospital is out of network, and that plans must cover the cost of their emergency care.1,2 At the same time, inpatient admissions through the emergency department (ED) and ED visits have become the largest source of volume at most US hospitals,3 and there has been substantial inflation of hospital billed charges, which are 200% to 400% of contract rates and are generally the prices that most hospitals demand for out-of-network emergency patients.4-7 These factors increase the exposure of a health plan to higher costs if a hospital can go out of network yet can retain a significant share of its pre—contract cancellation emergency volume.
Some have argued that government regulations designed to ensure access to hospital-based emergency care combined with the factors described here have had the unintended consequence of giving hospitals the power to unilaterally raise their prices above competitive levels for a share of medical emergencies in their local markets.8,9 If true, this could undermine price competition in provider markets, weaken health plan bargaining power vis-à-vis providers, and contribute to the trend of rising healthcare prices.
Due to data limitations, it has been difficult for researchers to quantify these dynamics (there has been only 1 published study to date10). The number of contract cancellations is relatively small, even within health plans, resulting in small sample sizes, and the data needed to study the effects are generally confidential and not available to researchers. Our study is feasible because our 5 study hospitals are part of a system whose business model included simultaneous cancellation (and public disclosure) of all commercial contracts at selected system member hospitals along with public reporting of data needed for our analysis.
METHODS AND DATA
Hospital data were accessed from the California Office of Statewide Health Planning and Development (OSHPD), including annual financial data, patient discharge data, and annual utilization data.
Study Sample and Study Period
The study sample includes 5 hospitals located in 3 geographic areas within California that are part of a for-profit, multihospital system (Prime Healthcare). During the study period (2004-2011), the system employed a unique strategy of cancelling all its contracts with commercial health plans and then billing insurers at their billed charge levels. Prime’s unique business model has been widely documented in the media11 as has information on contract cancellation dates for the specific hospitals. We confirmed the pre— and post–contract cancellation dates (all during the 2005-2007 period) for each hospital using OSHPD data and system-membership records.
Methods and Measures
We calculated totals and changes in prices, volume, and total net revenue for commercial payers only (excluding commercial Medicare/Medicare Advantage plans, Medicare, Medicaid, and self-pay patients) before and after contract cancellations. We used a 2-year average in the 2-year period before and 2-year averages for years 1 to 2 and 3 to 4 after contract cancellation for study hospitals and for comparable control hospitals (contract cancellation year is excluded to eliminate partial-year data). Price, volume, and revenue measures are limited to hospital services and do not include physician and other services that may be provided and billed separately during delivery of hospital-based emergency services. Data on control hospitals are included for comparison in order to control for local market forces and other factors that may affect both the study hospitals as well as other hospitals. The sample of control hospitals (n = 51) includes all hospitals located in the same geographic health service area (as defined by OSHPD) for each of the study hospitals and includes all general acute care hospitals (excluding teaching, specialty, and Kaiser hospitals). Results of t tests show that the study hospitals were similar to the control hospitals in terms of volume and case mix in the precancellation period (detailed statistical tests included in eAppendix [available at ajmc.com]).
Volume was calculated using OSHPD methodology that includes actual total inpatient discharges combined with an estimate of equivalent inpatient discharges to include outpatient volume12 and was calculated separately for patients admitted through the ED and for non-ED admissions. Total ED outpatient visits were calculated by OSHPD for third-party payers in each year.
Prices and Total Net Revenue From Third-Party Payers
We used total net revenue from third-party commercially insured patients (including plan and patient payments) to measure hospital net revenue and average price for third-party commercial patients. Price was calculated by dividing total net revenue for all third-party commercial payers by total adjusted inpatient discharges (reflecting both inpatient and outpatient volume). We also calculated a case mix—adjusted price by dividing the price measure by each hospital’s commercial case mix index.
Annual Price and Volume Trends Before and After Cancellation of All Commercial Contracts: Hospital Level
Figure 1 shows prices for the 5 study hospitals for each of the 3 years before they cancelled all their commercial contracts (precancellation period) and 4 years after they cancelled all their commercial contracts (postcancellation period). Prices in 3 of the study hospitals (hospitals 1, 3, and 4) were relatively stable in the 3-year precancellation period whereas prices in 2 of the study hospitals (hospitals 2 and 5) showed price increases during the precancellation period. Prices in all 5 hospitals increased substantially in the postcancellation period compared with the precancellation period.
Figure 2 shows volume for the study hospitals for each of the 3 years before they cancelled all their commercial contracts (precancellation period) and for each of the 4 years after they cancelled all their commercial contracts (postcancellation period). Volume in all 5 study hospitals was relatively stable in the 3-year precancellation period. Volume declined in the postcancellation period compared with the precancellation period in all hospitals, but in hospital 4 the decline was more modest.
Percentage Changes in Price and Volume After Cancellation of All Commercial Contracts: Hospital Level
Table 1 summarizes the percentage changes in average prices and volume for each of the 5 study hospitals for 2-year periods before and after they cancelled all their commercial contracts. In the 2-year period after cancelling their contracts and becoming out of network, all 5 hospitals were able to raise their prices—by 125% to 347% relative to the 2-year precancellation period baseline. Prices in other nearby control hospitals also increased but by far less (by 21%-36%) over the same periods and, as can be seen, the differential price increases in the study hospitals were not driven by increased case mix after cancellation of contracts. T tests indicate statistically significant differences between the 5 study hospitals and the 51 control hospitals for percentage changes in case mix—adjusted net revenue per adjusted discharge for both postcancellation periods, years 1 to 2 and years 3 to 4, compared with the 2-year baseline, precancellation period (detailed statistical tests included in eAppendix). Volume, as expected, trended downward after contract cancellation. Retained volume in the postcancellation period (as a percentage of the precancellation period 2-year average) ranged from a low of 33% to a high of 76% (volume at nearby control hospitals remained relatively steady over the study period).
Comparison of Total Net Revenue Before and After Cancellation of All Commercial Contracts: Hospital Level
Table 2 shows that in the 2 years after contract cancellations, all 5 hospitals increased their total net revenue from commercial health plans and 2 of the 5 hospitals sustained net revenue increases 3 to 4 years after contract cancellations.
Summary of Price and Volume Trends Before and After Cancellation of All Commercial Contracts: System Level
Table 3 shows data aggregated at the system level for the 5 study hospitals. The systemwide average price rose relative to the precancellation period by 331% for years 1 to 2 and 279% for years 3 to 4 following contract cancellation. Overall retained volume (includes inpatient and outpatient) across the system was 50% of the precancellation period volume for years 1 to 2 and 41% for years 3 to 4 following contract cancellation. Retained commercial outpatient ED visit volume was higher than overall volume: 72% of the precancellation period volume for years 1 to 2 and 64% for years 3 to 4 following contract cancellation. Total net revenue collected from third-party commercial payers increased following contract cancellation. In the 2-year precancellation period, the 5 hospitals collected $89.6 million in net revenue per year. Net revenue rose to $146.6 million per year in the first 2-year postcancellation period and $107.0 million in years 3 to 4 following contract cancellation. This indicates that price increases more than offset volume reductions across the system following contract cancellation.
High and rising healthcare costs continue to be one of most pressing social issues in the United States, and a growing body of research shows that rising prices are a main driver of increased healthcare spending in the United States.13
Our data show that all 5 hospitals were able to raise their prices in the 2 years after cancelling their contracts—by 125% to 347%—and although commercial volume declined for all 5 hospitals after they cancelled their contracts, commercial volume did not drop to zero—all the study hospitals retained a share of their precancellation commercial volume. Although they lost most of their nonemergency volume after contract cancellation, all 5 hospitals continued to capture a share of medical emergencies in their local market—despite having no commercial contracts and substantially raising their prices. All 5 hospitals retained a higher proportion of their outpatient ED visit volume compared with overall inpatient volume. And, although not shown here, commercial inpatient admissions through the ED went from about 50% of all inpatient admissions in the precancellation period to more than 90% of all inpatient admissions in the postcancellation periods.
When volume and price effects are combined to calculate total net revenue from commercial payers, before and after contract cancellation, all 5 of the study hospitals were able to increase their total net revenue from commercial payers (5% to 198% after 2 years from precancellation levels; 2 of 5 hospitals sustained net revenue increases from commercial payers through years 3-4 of the postcancellation period).
These results suggest that cancellation of commercial contracts likely increased the hospitals’ net profits because net revenue increased while volume declined and, because costs are correlated with volume, costs likely declined as well. This is important because it makes the hospital’s threat to leave the network economically credible to the health plan. The value of the threat to the health plan will depend on the increase in the total net cost to commercial health plans to cover their members after contract cancellation.
We provide an example based on our results for the 5 case study hospitals. Assuming that all the patients lost post cancellation (n = 8484) sought care elsewhere at the same prices paid to the study hospitals at precancellation levels ($5455 per patient), the total cost to commercial plans would increase substantially, from $89.6 million at precancellation contract rates to $192.9 million (this includes $146.6 million at out-of-network rates paid to the 5 study hospitals plus $46 million paid to other facilities for the displaced patients). In this example, total costs to commercial health plans would increase by $103 million from the precancellation base. This does not include the impact of the disruption to their members caused by having to switch hospitals.
Health plans, in the 2 years following contract cancellation, paid an average of $11,334 per adjusted patient discharge compared with $5455 per discharge precancellation (an increase of 108%) for their entire population, including those still receiving ED care in the out-of-network hospitals and displaced patients going to other hospitals. The implication is that the negotiators for the hospitals, as part of their bargaining, could demand price increases up to 108% of their previously negotiated prices to keep their hospitals in the networks and, if the health plans agree, the health plans would still pay less than they would if they allowed the 5 study hospitals to become noncontracted. One factor that might lower the value to the hospital of the threat of cancellation is the complete loss of non-ED inpatient volume, which may affect the efficiency and effectiveness of services dependent on direct-admit patients (eg, open heart surgery).
These findings illustrate the inflationary potential of this dynamic on contracted hospital prices. Combined, the 5 hospitals in our case study raised their prices an average 231% and retained 50% of their volume after 1 to 2 years (and increased their prices by 179% after 3-4 years and still retained 40% of their volume). It is worth noting that 4 of the hospitals in our sample had competing EDs within 15 minutes of their ED, and the 1 hospital that retained the highest share of precancellation volume (76%) had a more distant competing ED, 21 minutes away.
In addition, the number of US hospitals that are part of multihospital systems has been increasing. About two-thirds of California and US hospitals are part of such systems.14 This is potentially important because we observed variation in both the price and volume effects. Although all 5 hospitals increased their total net revenue in the 2-year period following cancellation, 3 of the 5 saw their net revenue decline from precancellation levels in years 3 to 4 following cancellation, but total net revenue summed across all 5 hospitals was still above precancellation levels. This increased total net revenue result suggests that systems might be able to lower their risk of a negative outcome by spreading the risk of contract cancellation across multiple members, and it could further enhance the credibility of a threat by a system to cancel its contracts during price negotiations.
Our data and empirical findings have several limitations. The 5 study hospitals are from a single for-profit hospital system in a single state (California), and thus the specific empirical findings may not be generalizable to other hospitals or systems. This system has been fined by CMS for unnecessarily admitting Medicare beneficiaries who initially went to EDs at the hospitals but required only outpatient care.15 If these same practices were applied to commercial patients, estimated retained inpatient ED commercial volume post cancellation may be inflated. However, the study hospitals retained a higher percentage of outpatient ED visit volume compared with inpatient admissions, which is likely unaffected by such practices. Net revenue data reported to OSHPD may be underreported, as out-of-network payments are often the subject of lawsuits that are adjudicated over several years and supplemental payments by health plans may not be reflected in the net revenue data reported to OSHPD.
Despite these limitations, our empirical findings are valuable to understand how regulations that ensure access and payment for hospital emergency services (the Emergency Medical Treatment and Labor Act requires hospitals to treat all emergency patients, and prudent layperson laws require plans to pay for emergency services) can affect bargaining dynamics and contribute to overall hospital price increases. Prices paid by commercial health plans to our case study hospitals, once they went out of network, increased substantially. Commercial volume declined but did not fall nearly enough to offset the much higher out-of-network prices.
Finally, the legal and policy landscape related to out-of-network emergency services is evolving. During our study period, hospitals relied on regulations that allowed them to argue quite successfully that their full billed charges were the correct prices for out-of-network emergency services. In 2014, an important court ruling in California broadened the criteria beyond solely billed charges for determining out-of-network prices for emergency services.16 This ruling may, to some extent, blunt the ability of hospitals to successfully demand full billed charges for out-of-network prices for emergency services. Policy makers at both the state and federal levels are also becoming more aware of the issues related to out-of-network billing, resulting in a wide range of legislative proposals which, based on our findings, might include limitations on out-of-network hospital emergency prices.
Our results suggest that under current regulatory and market conditions, demand for hospital-based emergency services is highly inelastic with respect to price, giving hospitals substantial pricing power over emergency patients in their local markets, which can be leveraged to demand higher contract prices from commercial health plans to keep hospitals in their networks. This dynamic results in higher health plan payments, higher insurance premium levels, and higher healthcare spending for everyone. Policy makers should take notice of this dynamic and consider policy actions to restore and strengthen price competition.17,18 In addition, our findings are potentially important to the small but growing literature on hospital—health plan bargaining and patient choice models. These models, to date, have generally ignored the ability of hospitals to retain a large fraction of emergency volume and have assumed that patients in contracted hospitals will be fully absorbed in other nearby hospitals following removal from a network.19,20Author Affiliations: Price School of Public Policy, University of Southern California (GM, KF), Manhattan Beach, CA; Massachusetts Health Policy Commission (KF), Boston, MA.
Source of Funding: None.
Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (GM, KF); acquisition of data (GM, KF); analysis and interpretation of data (GM, KF); drafting of the manuscript (GM, KF); critical revision of the manuscript for important intellectual content (GM, KF); statistical analysis (GM, KF); provision of patients or study materials (GM); obtaining funding (GM); administrative, technical, or logistic support (GM); and supervision (GM).
Address Correspondence to: Glenn Melnick, PhD, Price School of Public Policy, University of Southern California, 2205 Meadows Ave, Manhattan Beach, CA 90266. Email: email@example.com.REFERENCES
1. Hall MA. The impact and enforcement of prudent layperson laws. Ann Emerg Med. 2004;43(5):558-566. doi: 10.1016/S0196064403013416.
2. Williams RM. The prudent layperson definition: will it work for emergency medicine? Ann Emerg Med. 2000;36(3):238-240. doi: 10.1067/mem.2000.109212.
3. Schuur JD, Venkatesh AK. The growing role of emergency departments in hospital admissions. N Engl J Med. 2012;367(5):391-393. doi: 10.1056/NEJMp1204431.
4. Historical impact files for FY 1994 through present. CMS website. cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Historical-Impact-Files-for-FY-1994-through-Present.html. Updated August 1, 2012. Accessed August 1, 2019.
5. Anderson GF. From ‘soak the rich’ to ‘soak the poor’: recent trends in hospital pricing. Health Aff (Millwood). 2007;26(3):780-789. doi: 10.1377/hlthaff.26.3.780.
6. Augustine J. Survey: the ED is the nexus of care. Emergency Physicians Monthly website. epmonthly.com/article/survey-the-ed-is-the-nexus-of-care. Published October 7, 2013. Accessed November 10, 2019.
7. Selden TM, Karaca Z, Keenan P, White C, Kronick R. The growing difference between public and private payment rates for inpatient hospital care. Health Aff (Millwood). 2015;34(12):2147-2150. doi: 10.1377/hlthaff.2015.0706.
8. Murray R. Hospital charges and the need for a maximum price obligation rule for emergency department & out-of-network care. Health Affairs website. healthaffairs.org/do/10.1377/hblog20130516.031255/full. Published May 16, 2013. Accessed August 1, 2019.
9. Melnick GA, Fonkych K, Zwanziger J. The California competitive model: how has it fared, and what’s next? Health Aff (Millwood). 2018;37(9):1417-1424. doi: 10.1377/hlthaff.2018.0418.
10. Shepard M. Hospital network competition and adverse selection: evidence from the Massachusetts health insurance exchange. Scholars at Harvard website. scholar.harvard.edu/files/mshepard/files/mshepard_jmp_hospital_networks_adverse_selection.pdf. Published February 29, 2016. Accessed November 10, 2019.
11. Costello D. Hospital group rejects system and cashes in. Los Angeles Times. July 8, 2007. latimes.com/archives/la-xpm-2007-jul-08-fi-reddy8-story.html. Accessed November 10, 2019.
12. Adjusted admission. OregonLaws.org website. oregonlaws.org/glossary/definition/adjusted_admission. Accessed November 10, 2019.
13. 2017 Health Care Cost and Utilization Report. Health Care Cost Institute website. healthcostinstitute.org/research/hccur. Published February 2019. Accessed August 1, 2019.
14. Fast facts on US hospitals, 2019. American Hospital Association website. aha.org/statistics/fast-facts-us-hospitals. Updated January 2019. Accessed August 1, 2019.
15. Gooch K. Prime Healthcare, CEO to pay $65M Medicare fraud settlement. Becker’s Hospital Review website. beckershospitalreview.com/legal-regulatory-issues/prime-healthcare-ceo-to-pay-65m-medicare-fraud-settlement.html. Published August 6, 2018. Accessed August 1, 2019.
16. Children’s Hospital Central California v Blue Cross of California, 2014 WL 2590823 (Cal App).
17. Murray R, Berenson RA. Hospital rate setting revisited: dumb price fixing or a smart solution to provider pricing power and delivery reform? Urban Institute website. urban.org/sites/default/files/alfresco/publication-pdfs/2000516-Hospital-Rate-Setting-Revisited.pdf. Published November 2015. Accessed November 17, 2016.
18. An analysis of policy options for involuntary out-of-network charges in New Jersey. Avalere Health website. avalere.com/insights/new-avalere-analysis-examines-the-unintended-consequences-of-hold-harmless-rules-in-new-jersey. Published March 25, 2015. Accessed November 1, 2019.
19. Lewis MS, Pflum KE. Diagnosing hospital system bargaining power in managed care networks. Am Econ J Econ Policy. 2015;7(1):243-274. doi: 10.1257/pol.20130009.
20. Town R, Vistnes G. Hospital competition in HMO networks. J Health Econ. 2001;20(5):733-753. doi: 10.1016/S0167-6296(01)00096-0.