The Prices That Commercial Health Insurers and Medicare Pay for Hospitals’ and Physicians’ Services

Chapter 1Levels of and Trends in Spending and Prices

This chapter describes the growth of spending by commercial health insurers and Medicare’s fee-for-service (FFS) ­program on hospitals’ and physicians’ services in recent years. It illustrates the role of price increases in fueling the growth of spending by commercial insurers and describes how prices are determined for those insurers and for Medicare. This chapter also discusses sources of upward and downward pressure on prices, compares prices for Medicare and commercial insurers, and compares payments to hospitals with hospitals’ costs.

Factors Driving the Growth of Per-Person Health Care Spending

From 2013 to 2018, commercial insurers’ spending per person on inpatient and out­patient hospital care and physicians’ services grew by an average of 3.2 percent a year, the Congressional Budget Office estimates. That spending growth was driven by increases in the prices that commercial insurers paid for such services. Prices rose by an average of 2.7 percent a year—about 1 percentage point faster than average inflation during that period, as measured by the change in the gross domestic product (GDP) price index.1 CBO estimated the increase in the prices paid by commercial insurers per unit of service (visits, procedures, and admissions) from the growth of per-person spending and the quantity of services. That quantity—“utilization” in the figure—was measured as the number of services provided, adjusted for their intensity (the amount of resources or physicians’ effort used to provide them). Utilization grew by less than 0.5 percent per year, on average, during the 2013–2018 period.

Per-person spending grew more slowly for the Medicare FFS program than for commerical insurers—by 1.8 percent a year, on average—from 2013 to 2018. That spending growth also stemmed mostly from price increases. The FFS program’s prices for hospitals’ and physicians’ services, which are updated regularly by statute and regulation, rose by an average of 1.3 percent a year. The quantity and intensity of services provided per person (which CBO estimated from the growth of per-person spending and from price updates) increased by an average of 0.5 percent a year during the 2013–2018 period.

How Medicare’s Prices Are Determined

The prices that the Medicare FFS program pays providers are set administratively through laws and regulations. The program uses various payment schedules for different kinds of services. For hospitals’ services that are covered by the inpatient or outpatient prospective payment systems, the amount that Medicare pays hospitals is predetermined. That amount is based on a flat base-payment rate, which Medicare adjusts to account for geographic differences in input costs, for the intensity of services, and for patients’ health (in the case of inpatient services). Rates are also adjusted for patients who are transferred to another hospital paid under the inpatient prospective payment system or who are discharged to a post-acute care setting (for certain diagnoses).

In addition, Medicare makes hospital-specific adjustments for inpatient services at hospitals that operate residency training programs or that treat a disproportionate share of low-income patients. Medicare also makes “outlier payments”—additional payments for cases that are extraordinarily costly—for inpatient or outpatient services. (Some hospitals, such as certain small hospitals that Medicare considers “critical access hospitals,” are not paid under Medicare’s prospective payment systems.)

For services covered by Medicare’s physician fee schedule, Medicare pays a fixed amount per service, with adjustments for the relative amount of resources (physicians’ effort, practice expenses, and liability insurance) typically used to provide a given service and for geographic differences in the costs of delivering care. Payments are further adjusted if multiple similar services are furnished by the same provider on the same day or for certain characteristics of providers. For example, Medicare reduces prices if a service is provided by a nonphysician professional, such as a nurse practitioner or social worker, or if a physician is serving as an assistant during a surgery.

How Commercial Insurers’ Prices Are Determined

The prices that commercial health insurers pay in-network providers result from negotiations between individual providers or provider groups and insurers. Providers agree to accept lower in-network prices in exchange for having insurers steer patients to them and pay claims promptly. In some instances, providers can negotiate higher prices by threatening to stay out of an insurer’s network. That threat is more credible for providers that have a large market share, provide high-quality services, or provide services that cannot be planned for in advance. Such negotiations—and the contracts and prices that result from them—are often considered trade secrets by both insurers and providers.

Insurers and providers also negotiate about the basis of payments. How hospitals are paid for inpatient care can differ for each insurer and hospital. Methods include a flat rate per stay that varies by patients’ primary diagnosis and other health conditions (and may be set as a percentage of the amount paid by Medicare for that service), a specified rate per day (known as a per diem), or a specific discount from a hospital’s listed charges. Payments for hospitals’ outpatient care are frequently based on such discounts. Hospitals often prefer discounted charges or per diems because a flat rate can leave them at financial risk for high-cost cases or long stays. According to a recent study, hospitals with greater market power were more likely to be paid on the basis of discounted charges.2

Physicians are often paid by commercial insurers using the same structure as the Medicare FFS program, but provider groups negotiate a multiplier that is applied to Medicare’s prices. Some evidence suggests that physician groups with greater market power can negotiate increases to that multiplier or have more ability to deviate from the Medicare FFS payment structure.3

Sources of Upward and Downward Pressure on Prices

The prices that commercial insurers pay for hospitals’ and physicians’ services are much higher, and have grown much more rapidly in recent years, than the Medicare FFS program’s prices. Those outcomes are the net result of competing factors that put upward or downward pressure on prices. For commercial insurers, the upward pressures on prices are stronger than the downward pressures. For Medicare FFS, prices do not rise as fast as commercial insurers’ prices because of factors that restrain their increases.

Compared with Medicare, commercial insurers are subject to fewer constraints on the prices they pay. Private health insurance is mostly provided through employers, and private insurers negotiate prices on employers’ behalf. Employers and insurers mainly try to negotiate lower prices by threatening to exclude providers from their networks. Their ability to do that is often limited, however, for several reasons: Providers’ market power is much greater than employers’ in many markets; enrollees in employment-based plans tend to value having access to broad networks; certain providers may be essential to a network in a given area; and large insurers or employers may have enrollees in many locations with diverse medical needs, making narrow-network plans hard to implement.4 Employers have been slow to adopt other methods to restrain price increases, such as reference pricing or other changes to the design of insurance benefits, because such designs are complex and could put enrollees at financial risk.5

In addition, the incentives for any one insurer or employer to push for lower prices are limited because the benefits of doing so may not accrue directly to them. For example, insurers have less incentive to negotiate lower prices because they can pass providers’ price increases on to employers. A dominant insurer may even limit the discounts that a provider can offer to other insurers, which can raise the prices paid by those other insurers and their premiums.6 Negotiating lower prices may also involve trade-offs for employers. For instance, employers may be reluctant to reduce prices by adopting a plan with a narrower network or by reducing health care benefits if they use those benefits to compete for high-value employees.

Conversely, one factor putting downward pressure on the prices paid by commercial insurers is that employers may have to finance higher provider prices by raising premiums, limiting wage increases, or reducing their plans’ benefits. The strength of that downward pressure is dampened, however, because premium contributions are generally excluded from federal taxes and because the link between higher provider prices and lower wages or narrower benefits is indirect, so most employees do not see it.

For Medicare FFS, price increases have been limited in part because annual updates to prices are set through statute and regulation and have been modest in recent years. Prices for hospitals’ inpatient and outpatient services are updated using a measure of changes in the prices of hospitals’ key inputs, minus the increase in economywide productivity.7 Price updates for physicians’ services are set by law; those prices have risen by less than 0.5 percent a year since 2015. In addition, Medicare offers payment rates to providers on a take-it-or-leave-it basis. Providers that do not want to accept those rates can decline to participate. But because Medicare accounts for a large share of U.S. health care spending, providers may have limited financial ability to opt out of the program.

Another source of downward pressure on Medicare’s prices is that, because Medicare is a federal program, proposals that would raise its spending must be financed through higher taxes or premiums, increases in federal debt, or cuts in Medicare benefits or other spending. The strength of that downward pressure is partly reduced, however, by concerns about providers’ financial stability and Medicare beneficiaries’ access to care.

Average Prices for Hospitals’ Services

CBO reviewed a range of studies published between 2010 and 2020 that compared commercial insurers’ and the Medicare FFS program’s prices for hospitals’ services. On average, the five studies that looked at overall prices for hospitals’ services suggest that the prices paid by commercial insurers were more than twice those paid by Medicare FFS.8 (The combined average price ratio for inpatient and outpatient services reported by those studies ranged from 167 percent of Medicare FFS’s prices to 250 percent.) Those studies also generally reported substantial variation in prices for the same service in different areas, in different hospitals, and within the same hospital.

The ratio of commercial insurers’ prices to Medicare FFS’s prices was generally much higher for outpatient services than for inpatient services. Averaged across the 6 studies on outpatient services and 11 studies on inpatient services that CBO reviewed, commercial insurers’ prices were 240 percent of Medicare FFS’s prices for outpatient services and 182 percent of Medicare FFS’s prices for inpatient services. The reasons for the higher price ratio for outpatient services are not fully understood. Those reasons could include contracting practices: Contracts based on discounted charges are more common for hospitals’ outpatient services than for inpatient services, and that type of contract may encourage higher prices.

In its literature review, CBO included studies that compared commercial insurers’ prices with Medicare FFS’s prices or that reported payment-to-cost ratios both for private sources (including uninsured people) and for Medicare. CBO then calculated a simple average of the studies’ results. Those studies’ estimates may have differed because of differences in the locations, time periods, or types of data used in the studies. For instance, estimates can differ depending on whether a data source includes patients’ total cost-sharing responsibilities or the amounts that were actually collected from patients.

Average Prices for Physicians’ Services

For physicians’ services overall, commercial insurers paid 129 percent of Medicare FFS’s prices, on average, according to studies published between 2010 and 2020. (CBO reviewed seven studies that looked at physicians’ services as a whole; their average price ratio ranged from 118 percent of Medicare FFS’s prices to 163 percent.) Those studies also generally reported substantial variation in prices for the same service among geographic areas, physician practices, and physician specialties.

Relative to Medicare FFS’s prices, the prices paid by commercial insurers for primary care services and office visits were generally lower than the prices paid for specialty services. Among the five studies of primary care services or office visits and five studies of specialty services that CBO reviewed, commercial insurers’ prices were 117 percent of Medicare FFS’s prices for primary care services or office visits, on average, and 144 percent of Medicare FFS’s prices for specialty services.

Higher prices for specialty services could stem from three factors: Fewer doctors provide any one of those services; many doctors who do are employed in larger, multispecialty practices; and more of those practices are affiliated with hospitals. (In this analysis, average prices for specialty services do not include emergency services or physicians’ services provided only in inpatient settings.) Some of the variation in estimates among studies comes from differences in the types of physicians’ services included.

In its literature review, CBO looked only at studies that compared prices paid by commercial insurers and Medicare FFS. CBO then calculated a simple average of the studies’ results. For studies that reported prices separately for specific services or markets (such as employer-­sponsored plans versus individually purchased plans), CBO averaged those prices, weighting them by the frequency of services or by estimates of premiums for different segments of the market from CBO’s baseline budget projections. The studies’ estimates of relative prices for physicians’ services may differ because of differences in data sources, time periods, settings of service (such as hospitals’ outpatient departments versus physicians’ offices), or the specific services or markets included in the studies.

How Payments to Hospitals Compare With Hospitals’ Costs

According to data from the American Hospital Association (AHA), Medicare’s total payments for inpatient and outpatient services are lower than hospitals’ costs to provide those services. Private payers’ payments, by contrast, are much higher than hospitals’ costs. Between 2000 and 2018, private payers’ payments rose from 116 percent to 145 percent of hospitals’ costs, whereas Medicare’s payments fell from 99 percent to 87 percent of hospitals’ costs. (In the AHA data, Medicare includes both the FFS program and Medicare Advantage plans, and private payers consist of all nongovernment payment sources.)

The payment-to-cost ratios in the AHA data reflect hospitals’ total costs for delivering care, including staffs’ salaries and the costs of buildings and equipment. Those total costs vary because of external factors, such as local wages and utilities, and internal factors related to hospitals’ behavior, such as how efficiently they can manage their resources. Although payments by Medicare are lower than hospitals’ costs, on average, research suggests that hospitals have some control over their cost structure and can adjust their expenses on the basis of the financial resources they have available.9 The Medicare Payment Advisory Commission (MedPAC) has identified a subset of hospitals as “efficient,” meaning that they perform well in the areas of cost and quality. Such hospitals had positive Medicare margins from 2010 until about 2015, although they have had slightly negative Medicare margins since 2016.10

Payments for hospitals’ services in the AHA data reflect all payments made for care, not just the prices negotiated by a subset of payers. Private payments, in particular, reflect a wide variety of sources—from uninsured people, who typically pay less than commercial payers, to automobile liability insurance and workers’ compensation, which tend to pay more. Private payments also reflect patients’ unpaid debts, which tends to reduce private payers’ prices in the AHA data relative to the prices reported in other studies. (CBO does not know of any regularly updated source of data about costs or payment-to-cost ratios for physicians’ services.)

---

Chapter 2The Implications of Price Increases

This chapter describes how the high and rising prices that commercial insurers pay for hospitals’ and physicians’ services affect people with private health insurance and the federal budget.

How Price Increases Affect People With Employment-Based Insurance

A rise in providers’ prices would increase health insurers’ spending on claims, all else being equal. Insurers could respond to such spending increases by raising premiums, increasing cost-sharing requirements for patients, reducing the scope of benefits, or making other adjustments. In general, insurers’ greater spending would be passed on to employers that purchase coverage on behalf of their employees.1 Employers’ spending on health insurance represents a large part of their employees’ nonwage compensation, so employers generally take actions to offset increases in health insurance spending in order to maintain their profits.

A recent study found that price increases for hospitals’ services were associated with a rise in employees’ out-of-pocket costs, an increase in the use of high-deductible health plans, and slower wage growth for employees.2 In addition, the Congressional Budget Office analyzed data from the Kaiser Family Foundation’s Employer Health Benefits Survey and information from the Bureau of Labor Statistics and found that between 2010 and 2020, average total premiums rose by 55 percent for a family plan (from $13,770 to $21,342) and by 48 percent for an individual plan (from $5,049 to $7,470), whereas median wages grew by only 32 percent over that period.

How Price Increases Affect Federal Subsidies for Commercial Health Insurance

The federal tax code contains various provisions that subsidize commercial health insurance—both employment-based insurance and nongroup (individually purchased) plans. For example, premiums for people with employment-based insurance are generally excluded from federal income and payroll taxes. In addition, some out-of-pocket medical costs and premium payments are subsidized through the tax deduction for itemized medical expenses. The federal government also provides people who qualify with tax credits to reduce premiums for nongroup health insurance purchased through the marketplaces established under the Affordable Care Act. And the tax code allows some self-employed people who buy nongroup insurance to deduct up to 100 percent of their premiums from their income. Because of such provisions, increases in the prices that plans pay to providers affect federal subsidies for commercial health insurance.

Insurers pass higher costs on to employers through premium increases or changes to plans’ benefits. Since the tax code generally lets employers and employees contribute to the cost of premiums with pretax dollars, higher premiums reduce taxable wages and increase federal subsidies for employment-based insurance. Likewise, when premiums rise for the benchmark plans offered in the federal or state health insurance marketplaces, tax credits for people who buy insurance through those marketplaces rise as well.

---

Chapter 3Variation in Prices Among and Within Geographic Areas

This chapter describes how prices for hospitals’ and physicians’ services vary among different areas, among hospitals or provider groups in the same area, and within hospitals for the same or similar services. Prices can vary for many reasons, but large variation in prices for similar services can be evidence of market inefficiencies.

Prices for Hospitals’ Services: Variation Among States

The average prices that Medicare’s fee-for-service program and commercial employment-based plans paid for hospitals’ inpatient services in 2018—relative to the national average price paid by Medicare FFS—varied widely by state. The statewide average Medicare FFS price ranged from 24 percent lower than the program’s national average price in Arkansas to 55 percent higher than the national average in Massachusetts. Prices for Medicare services differed among states because of the location- and hospital-specific adjustments Medicare makes (see Chapter 1).1

The degree of price variation was much greater for commercial insurers. The average state-level price paid by commercial employment-based plans ranged from 54 percent higher than the national average Medicare FFS price in Arkansas to 294 percent higher than that price in Massachusetts. (The Congressional Budget Office limited this analysis to employment-based plans because price data are not as widely available for other types of commercial plans. Prices for inpatient services are allowed amounts per stay, with stays adjusted to reflect the intensity of the service.)

Prices for Physicians’ Services: Variation Among States

Relative to the national average Medicare FFS price for physicians’ services, the average price paid by Medicare FFS and commercial insurers for a set of 500 common services in 2017 varied among states. (Those 500 services reflect the service codes that appeared most often in aggregate data drawn from employment-based plans in the claims database of the Health Care Cost Institute, or HCCI. They include services such as office visits and knee replacements.) The statewide average Medicare FFS price ranged from 9 percent lower than the program’s national average price in Mississippi to 11 percent higher than the national average in Washington, DC. In this analysis, average Medicare prices differed primarily because of Medicare’s geographic adjustments for input prices.

The magnitude of the price variation was much larger for commercial insurers than for Medicare FFS. The average state-level price paid by commercial insurers ranged from about 10 percent lower than the national average Medicare FFS price in Alabama to 78 percent higher than that price in Wisconsin. Commercial insurers’ average price exceeded the average state Medicare FFS price in every state except Alabama.

Prices for Hospitals’ Services: Variation Within Metropolitan Areas

Commercial insurers’ prices for hospitals’ services vary widely not only by state but also within a given metropolitan statistical area, or MSA (a geographic area that generally consists of one or more cities and the surrounding population). For example, the median price that commercial insurers paid for a vaginal delivery of a baby—a well-defined and fairly standardized service—in San Francisco in 2016 was $13,363. But the range of prices around that median in the San Francisco area was quite broad: The price at the 90th percentile of the distribution of prices ($23,880) was more than twice as high as the price at the 10th percentile ($11,098). In Tucson, Arizona, the median price for a vaginal delivery was much lower, $5,920, and the range of prices was much narrower: The price at the 90th percentile ($6,735) was only 50 percent higher than the price at the 10th percentile ($4,490).

Price variation within an MSA reflects price differences among hospitals in that area as well as price differences within the same hospital. The latter largely reflect differences in the amounts that different commercial insurers pay for the same service at a particular hospital. Price variation within and among hospitals could also reflect differences in patients’ health or in the intensity of a service. However, most studies that examine price variation either control for patients’ health or focus on variation for a relatively standardized service, such as magnetic resonance imaging (MRI) scans.2

Prices for Physicians’ Services: Variation Within Metropolitan Areas

Recent analyses have found that commercial insurers’ prices for a given physicians’ service varied widely among providers in the same metropolitan statistical area. For instance, CBO’s analysis of data about a specific type of colonoscopy (Current Procedural Terminology code 45385) found a roughly twofold variation in commercial insurers’ prices for that service within five MSAs in 2014.3 In the Chicago-Naperville-Elgin area of Illinois, for example, the price for that procedure for the provider at the 90th percentile of the distribution of providers by price ($660) was twice as high as the price for the provider at the 10th percentile ($330).

CBO found similar variation in prices within MSAs for all 20 of the physicians’ services that it examined—including fairly standardized and narrowly defined services, such as echocardiograms and MRIs. For instance, in more than 85 percent of the 218 MSAs included in CBO’s analysis, the price for a cranial MRI for the provider at the 90th percentile was more than 1.5 times the price for the provider at the 10th percentile.

---

Chapter 4Factors Affecting the Prices Paid by Commercial Insurers

This chapter describes possible reasons that commercial insurers’ prices for hospitals’ and physicians’ services are higher and more variable than the prices paid by Medicare’s fee-for-service program. Potential explanations include market concentration for hospitals and physicians, the prices of various inputs for care, the quality of care, providers’ administrative spending, and their possible cost shifting from public to private payers.

Market Power and Commercial Insurers’ Prices

Market power is the ability of firms to raise prices above the level that would exist in a perfectly competitive market. It occurs for many reasons. Hospitals and physicians’ groups may have market power because they have a dominant share of a particular market or because an insurer sees them as essential to its network of providers. Being seen as essential can give providers substantial leverage in price negotiations with insurers. Thus, prices may vary among areas because of differences in the relative market power of providers and insurers in each area.

Differences in market power can also contribute to price variation within a geographic area. Prices may vary among providers because of differences in providers’ and insurers’ relative bargaining power, or they may vary within a hospital or physicians’ group because of differences in insurers’ bargaining power. One study found that, within the same provider group, negotiated prices for physicians’ office visits were 21 percent lower, on average, for insurers with a market share of at least 15 percent than for insurers with a market share of 5 percent or less.1

Market concentration (the extent to which only a few firms provide a large share of the goods or services in a market) can confer market power to firms with larger market shares. Evidence suggests that concentration in the markets for hospitals’ and physicians’ services has been growing and that, in many areas, those markets are now moderately or highly concentrated. The percentage of physicians’ practices that are owned by or affiliated with hospitals has also increased. Such vertical integration can increase concentration in the market for physicians’ services (by consolidating physicians into larger groups) and can increase hospitals’ market power (by steering physicians’ referrals to the acquiring hospital). In addition, evidence suggests that a growing number of health care providers have been acquired by private equity firms, which may leverage providers’ market power to raise prices.

Most of the studies that the Congressional Budget Office reviewed (see Appendix C) suggest that consolidation increases prices in the markets for both hospitals and physicians, as does vertical integration among hospitals and physicians’ groups. There is also a well-documented correlation, across many areas, between concentration in the provider market and prices, suggesting that some of the difference in prices in different areas is attributable to providers’ market power.2 Most studies focus on market share as a measure of providers’ market power, but there is growing evidence that providers can also derive market power from providing specialized services, high-quality services, or services that a patient cannot shop for in advance.3

Concentration in Hospital Markets and Commercial Insurers’ Prices

CBO analyzed data from the Health Care Cost Institute for hospital markets in 42 states and found that the percentage of metropolitan statistical areas with hospital markets that were considered highly or very highly concentrated rose between 2010 and 2017. In 2010, 63 percent of the 124 MSAs in the HCCI data had highly or very highly concentrated hospital markets. By 2017, that share had risen to 70 percent. (The MSAs that HCCI included in that sample were selected on the basis of population, insurance coverage, and number of hospitals.) Over that period, the average Herfindahl-Hirschman index (HHI) for the MSAs in that sample rose from 3,032, already in the “highly concentrated” range, to 3,338. (The HHI is a common measure of market concentration; it indicates the extent to which a market is dominated by one or a few participants.)

Greater market concentration has been linked to less price competition. CBO reviewed 13 studies published since 2010 on the relationship between hospital market concentration and prices for hospitals’ services (see Table C-1 in Appendix C). Of the four studies that tested for correlations, three found a positive association between market concentration and hospitals’ prices. Mergers between hospitals have also increased concentration and typically led to higher prices. Of the five studies that examined hospital mergers within a market, three concluded that prices rose for insurers. (The other two studies showed price increases for some insurers and decreases for others.) There is also some recent evidence that mergers of hospitals in different markets may result in price increases.

CBO focused on the research literature since 2010 because high-quality data on prices have become more readily available since then. But a review of older studies also found a positive association between market concentration and prices for hospitals’ services.4 It also found some evidence that growth in those prices is related to market concentration and that hospital mergers in concentrated markets typically result in much higher prices.

CBO also reviewed studies on the relationship between mergers and efficiency gains for hospitals.5 The evidence about the relationship between mergers and hospitals’ costs is inconsistent; some studies found that acquired hospitals have lower costs after their merger. But lower costs do not necessarily translate into lower prices for hospitals’ services if hospitals do not pass those efficiency gains on to consumers.

Concentration in Physician Markets and Commercial Insurers’ Prices

Market concentration for physicians increased between 2010 and 2016, as measured by the average of the Herfindahl-Hirschman indexes for more than 370 metropolitan statistical areas. During that period, the average HHI rose by nearly 29 percent for primary care physicians and by about 5 percent for four common types of specialist physicians (cardiologists, oncologists or hematologists, radiologists, and orthopedists).6 In addition, the percentage of MSAs whose markets for primary care physicians were considered highly or very highly concentrated (an HHI of more than 2,500) increased from 20 percent in 2010 to 39 percent in 2016. Increases in market concentration were at least partly attributable to a rise in the share of physicians employed by hospitals or hospital systems.

CBO reviewed eight studies on the relationship between market concentration for physicians and prices for physicians’ services; all of the studies found that prices were related to market structure in some way (see Table C-2 in Appendix C). Prices for most services were generally higher in areas with more-concentrated physician markets, or increased more in areas with higher initial levels of market concentration, or rose after mergers or acquisitions of physicians’ practices. In addition, one of the studies found that an increase in the enforceability of noncompete agreements for physicians (which make it harder for physicians to find new employment after leaving a practice) led to greater concentration among physicians’ groups and higher prices.

Employment of Physicians by Health Care Systems

Over the 2010–2016 period, the share of primary care physicians employed by a hospital or health care system rose by 16 percentage points, from 28 percent to 44 percent.7 At the same time, the share of primary care physicians who were in a solo practice or were part of a medical group declined. The trend of hospitals’ employing physicians may stem in part from the fact that Medicare and many commercial insurers have tended to pay higher prices when a service is billed in a hospital’s outpatient department rather than in a physician’s office.8 Physicians may also want to join larger, integrated practices to have more flexible work schedules, lower practice expenses per physician, access to expensive information technology systems, or discounts on drugs they purchase for outpatient care.

Hospitals’ acquisitions of physicians’ practices have been found to result in greater use of electronic health records and care management processes. Such acquisitions can also increase hospitals’ market power in at least two ways. First, they increase the amount of care billed in facilities owned by hospitals and reduce the amount billed in physicians’ offices. Second, they lessen the ability of rival hospitals to refer patients to the acquired physicians’ practices and increase the frequency of referrals within systems.9 Such acquisitions can also increase market concentration among physicians by consolidating them into larger groups.

CBO reviewed seven studies about whether vertical integration of providers relates to prices for their services (see Table C-3 in Appendix C). Four of the studies looked at the relationship between hospital-physician integration and prices for physicians’ services. All four found that prices increased more in areas where vertical integration increased or that prices rose after physicians’ practices were acquired by hospitals. (According to one of the studies, almost half of those price increases were potentially attributable to the services’ being billed as services provided in a facility and thus incurring a higher total fee.)10 Another study looked at integration between physicians of different specialties and found that prices for physicians’ services rose more in areas where that type of integration increased.

Three of the studies in CBO’s review examined the relationship between hospital-physician integration and prices for hospitals’ services. Two found a significant positive association between increases in integration and hospitals’ prices. The other study’s findings were inconclusive. (Although two of those studies used similar data sources, the three differed in their time periods, measures of vertical integration, and methods.)

Acquisition of Health Care Providers by Private Equity Firms

In recent decades, some physicians’ practices and hospitals have been bought by private equity firms (companies that purchase and take over other companies to generate short-term profits for their investors). Several recent studies have documented the volume of such acquisitions. According to one study, only about 2 percent of physicians’ practices were acquired by private equity firms between 2013 and 2016, but such acquisitions more than doubled in frequency during that period.11 Anesthesiologists, emergency medicine specialists, family practitioners, and dermatologists were the most common types of physicians in acquired practices. Another study found that roughly 7 percent of short-term acute care hospitals, responsible for 11 percent of total patient discharges, were acquired by private-equity-backed ventures between 2003 and 2017.12 More than half of those acquisitions (161) occurred in 2006, when Bain Capital purchased Hospital Corporation of America. (HCA became ­publicly traded again in 2011.)

Evidence about the effects on prices when private equity firms acquire medical providers has been mostly anecdotal, until recently. A recent study found that after private equity firms acquired hospitals, the ratio of those hospitals’ charges to costs rose by 7 percent overall, and by 16 percent in emergency rooms, relative to otherwise-similar hospitals not bought by private equity firms.13 (Hospitals’ charges differ from the negotiated prices paid by commercial insurers. But they are often correlated with those prices because, in some contracts, negotiated prices are defined as a percentage of a hospital’s charges.)

For physicians and other providers, another study found that after a private equity firm that employed emergency physicians entered into contracts with hospitals, the percentage of emergency care billed out of network at those hospitals rose by 83 percentage points, and the prices paid to those emergency physicians rose by 114 percent.14 A recent study of dermatologists found that among practices acquired by private equity firms, prices for a few common services rose by 3 percent to 5 percent relative to prices at nonacquired practices, although prices for other services were not significantly affected.15 In addition, a study found that air ambulance companies owned by private equity firms charged 68 percent more in 2017 than other air ambulance companies did.16

The Relationship Between Input Prices and Prices Paid to Providers

Commercial insurers’ prices for hospitals’ and physicians’ services are correlated with various measures of the prices of inputs needed to deliver those services (such as providers’ wages, rent, and malpractice insurance premiums). The strength of those relationships, however, is sensitive to the inclusion of a few areas with particularly high measures of input prices.

Hospitals’ Wages and Commercial Insurers’ Prices

Plotting an index of commercial insurers’ prices for hospitals’ inpatient services in a given area against an index of hospitals’ wages in that area in 2017 shows a positive relation­ship between wages and prices across 105 geographic areas. Of the variation in average prices among those areas, 51 percent is explained by wages. That share falls to 33 percent if the three outlier areas (Anchorage, Alaska; and San Francisco and San Jose, California) are removed.

In this analysis, the geographic areas are a subset of the nation’s core-based statistical areas (CBSAs), which consist of metropolitan statistical areas and other areas with smaller populations, known as micropolitan statistical areas. Commercial insurers’ prices for inpatient services are measured by a commercial price index, which represents the average price for a basket of common services if people in each CBSA used services in the same proportions seen at the national level. That index captures the percentage deviation from the national median price paid by commercial insurers. Hospitals’ wages are measured by Medicare’s hospital wage index. Medicare uses that index to adjust its prices for inpatient services to account for geographic differences in the wages that hospitals face in their local labor markets. Although wages are a large part of hospitals’ input costs, Medicare adjusts its prices for other characteristics of hospitals as well, such as whether they operate medical education programs or serve a disproportionate share of low-income patients.

Physicians’ Input Prices and Commercial Insurers’ Prices

Plotting an index of commercial insurers’ average prices for physicians’ services in each of 42 states in 2017 against an index of average input prices for physicians’ services in those states shows a positive relationship between input prices and the prices that commercial insurers pay for physicians’ services. Differences in input prices explain 27 percent of the variation in commercial insurers’ prices among those states (the states for which the price index that CBO used was available). However, that relationship is sensitive to the inclusion of Alaska, which has high input prices. With Alaska excluded, only 3 percent of the variation in commercial insurers’ prices is explained by differences in input prices.

In this analysis, commercial insurers’ prices for physicians’ services are measured as the state-level average commercial price index for a basket of 500 common physicians’ services in 2017. As in the earlier discussion of how prices for physicians’ services vary by state, the 500 services included in that index reflect the service codes most often seen in aggregate data from HCCI’s database of claims by employment-based plans.

To measure input prices for physicians’ services, CBO used Medicare’s geographic adjustment factor (GAF) as a proxy index for a state’s average input price for such services. That adjustment factor accounts for geographic differences in the prices of three inputs necessary for delivering medical services: physicians’ wages (based on geographic differences in the wages of other professionals, such as architects, pharmacists, and computer scientists), practice expenses (such as wages for administrative and clinical staff and rent), and malpractice insurance premiums. For states with multiple geographic adjustment factors, CBO created a state-level GAF by calculating a population-weighted average of the state’s factors. In addition, the Centers for Medicare & Medicaid Services (CMS) makes a number of legislatively determined adjustments to the parts of the GAF that deal with physicians’ wages and practice expenses, which affect the relationship between that measure and relative input prices.

Quality of Care and Commercial Insurers’ Prices

Analyzing 2016–2018 pricing data from RAND and 2018 quality information from CMS for more than 1,500 hospitals, CBO found a small positive correlation between the average prices that commercial insurers paid for hospitals’ inpatient and outpatient services, relative to Medicare FFS’s prices, and a summary measure of hospitals’ quality. In addition, when hospitals were ranked by commercial insurers’ prices as a percentage of Medicare’s prices, 41 percent of hospitals in the top third of that ranking (those with commercial prices greater than 276 percent of Medicare FFS’s prices) received four or five stars on CMS’s overall quality rating, compared with 32 percent of hospitals in the bottom third of the price ranking (those with commercial prices less than 216 percent of Medicare FFS’s prices). Even among hospitals in the bottom third, however, more than 60 percent received a quality rating of three or more stars.

It is unclear whether hospitals with higher quality can command higher prices from commercial insurers or whether hospitals with more market power, and thus higher prices, can spend more to improve their quality. In addition, correlations between prices and quality scores could both be associated with some other, unmeasured factor—such as a hospital’s size or the health of its patients—that causes the observed correlations.

In the case of physicians’ services, commercial insurers’ prices were not found to be associated with measures of quality for those services, according to a recent research study.17

Providers’ Administrative Spending and Commercial Insurers’ Prices

Many studies have documented the significant costs of providers’ administrative activities, which might contribute to higher prices for their services. Administrative spending can be divided into spending on billing- and insurance-related (BIR) activities (such as processing claims, updating patients’ medical records, and getting prior authorization from insurers) and spending on non-BIR activities (such as general overhead, marketing, and efforts to monitor and improve the quality of care).

In 2018, hospitals spent a total of $101 billion on BIR activities and $148 billion on ­non-BIR activities, while physicians and clinics spent $94 billion on BIR activities and $100 billion on non-BIR activities. A study of BIR spending at a major academic medical center found that providers’ administrative spending per encounter with a patient ranged from $20 for a primary care visit to $215 for inpatient surgery.18

Despite providers’ substantial administrative burden, the effect of that burden on prices for providers’ services is unclear. Providers may pass the costs of their administrative activities on to commercial insurers, but the relationship between administrative spending and prices has not been documented by research studies.

Cost Shifting and Commercial Insurers’ Prices

Some observers assert that hospitals’ costs are inflexible and that hospitals have to negotiate higher prices with commercial insurers to offset smaller payments made by Medicare and Medicaid—a concept known as cost shifting. There is some evidence to support that assertion in specific circumstances.19 But the preponderance of the evidence suggests that hospitals do not engage in cost shifting. Moreover, the idea of cost shifting is inconsistent with economic theory: If hospitals could charge private payers more, it is unclear why they would do so only after payment cuts from public payers.

If hospitals were able to cost shift, then hospitals with larger shares of Medicare and Medicaid patients (for whom prices are relatively low) would be paid relatively high prices by commercial insurers. However, CBO’s analysis of data for more than 1,500 hospitals indicates a weak cross-sectional relationship between commercial insurers’ average prices for a hospital’s inpatient and outpatient services during the 2016–2018 period and the percentage of Medicare and Medicaid patients among the hospital’s discharges. (In this analysis, the average commercial price reflects the sum of commercial insurers’ allowed amounts for inpatient and outpatient services relative to the simulated amount that Medicare would have paid for the same services.) Each 1 percentage-point increase in a hospital’s share of Medicare and Medicaid discharges is associated with only a 0.1 percent increase in commercial insurers’ prices relative to Medicare’s prices. A hospital’s share of discharges paid for by Medicare or Medicaid explains just 0.4 percent of the variation among hospitals in commercial insurers’ prices.

CBO’s analysis is descriptive, however, and does not control for the many other factors that affect hospitals’ costs. In addition, recent evidence suggests that hospitals with larger shares of Medicare patients are more likely to close or be acquired over time, which could increase concentration in their market and thus lead to higher prices for commercial insurers.20

---

Appendix AData Sources for Figures

Chapter 1: Levels of and Trends in Spending and Prices

Average Annual Growth Rates of Spending, Utilization, and Prices for Hospitals’ and Physicians’ Services, 2013 to 2018

Data sources: Growth rates for Medicare’s fee-for-service (FFS) program reflect the Congressional Budget Office’s analysis of information from Medicare Payment Advisory Commission, A Data Book: Health Care Spending and the Medicare Program (July 2020), Section 1; Boards of Trustees of the Medicare Trust Funds, 2020 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds (April 2020), https://tinyurl.com/mb9666c8 (PDF, 2.7 MB); and other, miscellaneous sources.

Growth rates for commercial insurers reflect CBO’s analysis of aggregate data from Health Care Cost Institute, 2017 Health Care Cost and Utilization Report (February 2019), https://tinyurl.com/82jksnn3, and 2018 Health Care Cost and Utilization Report (February 2020), https://tinyurl.com/27v534jv.

CBO estimated total spending for Medicare FFS and for commercial insurers by summing per-person spending on inpatient, outpatient, and physicians’ services but excluding spending on prescription drugs, home health care, and other types of care. In this analysis, commercial insurers’ spending per patient reflects the payments that insurers negotiated with providers and excludes patients’ cost sharing.

Studies’ Estimates of Commercial Insurers’ Prices for Hospitals’ Services as a Percentage of Medicare FFS’s Prices

Data sources: Sources are listed in Appendix B.

Studies’ Estimates of Commercial Insurers’ Prices for Physicians’ Services as a Percentage of Medicare FFS’s Prices

Data sources: Sources are listed in Appendix B.

Payment-to-Cost Ratios for Hospitals, 2000 to 2018

Data source: CBO’s analysis of aggregate national-level data from American Hospital Association, Trendwatch Chartbook 2020, Appendix 4, Table 4.4, https://tinyurl.com/­em4rjw9z (PDF, 3.6 MB).

Chapter 3: Variation in Prices Among and Within Geographic Areas

Average Prices for Hospitals’ Inpatient Services, by State, 2018

Data source: CBO’s analysis of aggregate data from Christopher M. Whaley and others, Nationwide Evaluation of Health Care Paid by Private Plans: Findings From Round 3 of an Employer-Led Transparency Initiative, RR-4394-RWJ (RAND, 2020), https://doi.org/10.7249/RR4394.

Average Prices for Physicians’ Services, by State, 2017

Data source: CBO’s analysis of aggregate data from Bill Johnson and others, “Comparing Commercial and Medicare Professional Service Prices: Public Use File” (Health Care Cost Institute, August 13, 2020), https://tinyurl.com/3xux3hzj.

Commercial Insurers’ Prices for Vaginal Deliveries at Hospitals in Selected MSAs, 2016

Data source: CBO’s analysis of aggregate data from Kevin Kennedy and others, “Past the Price Index: Exploring Actual Prices Paid for Specific Services by Metro Area” (Health Care Cost Institute, April 30, 2019), https://tinyurl.com/ap25xhr8.

Commercial Insurers’ Prices for Colonoscopies Performed by Providers in Selected MSAs, 2014

Data source: CBO’s analysis of data from Daria Pelech, An Analysis of Private-Sector Prices for Physicians’ Services, Working Paper 2018-01 (Congressional Budget Office, January 2018), www.cbo.gov/publication/53441.

Chapter 4: Factors Affecting the Prices Paid by Commercial Insurers

Hospital Market Concentration, 2010 and 2017

Data source: CBO’s analysis of aggregate data from Health Care Cost Institute, “Healthy Marketplace Index” (accessed December 2020), https://tinyurl.com/u86tw3x9.

Average Herfindahl-Hirschman Indexes for Primary Care and Specialist Physicians in Metropolitan Statistical Areas, 2010 to 2016

Data source: CBO’s reproduction of Exhibit 1 in Brent D. Fulton, “Health Care Market Concentration Trends in the United States: Evidence and Policy Responses,” Health Affairs, vol. 36, no. 9 (September 2017), pp. 1530–1538, https://doi.org/10.1377/hlthaff.2017.0556.

Share of Primary Care Physicians, by Ownership of Their Practice, 2010 to 2016

Data source: CBO’s reproduction of Exhibit 4 in Brent D. Fulton, “Health Care Market Concentration Trends in the United States: Evidence and Policy Responses,” Health Affairs, vol. 36, no. 9 (September 2017), pp. 1530–1538, https://doi.org/10.1377/hlthaff.2017.0556.

Number of Physicians’ Practices Acquired Each Year by Private Equity Firms, 2013 to 2016

Data source: CBO, using data from Jane M. Zhu, Lynn M. Hua, and Daniel Polsky, “Private Equity Acquisitions of Physician Medical Groups Across Specialties, 2013–2016,” JAMA, vol. 323, no. 7 (February 18, 2020), pp. 663–665, https://doi.org/10.1001/jama.2019.21844.

Relationship Between Hospitals’ Wages and Commercial Insurers’ Prices for Inpatient Services, 2017

Data sources: CBO’s analysis of aggregate data from Health Care Cost Institute, “Healthy Marketplace Index” (accessed December 2020), https://tinyurl.com/u86tw3x9; and Centers for Medicare & Medicaid Services, “FY 2017 Proposed Rule Tables 2 and 3 (Wage Index Tables) (ZIP),” Table 3 (accessed May 2021), https://tinyurl.com/cukuexa8.

Relationship Between Physicians’ Input Prices and Commercial Insurers’ Prices for Physicians’ Services, 2017

Data sources: CBO’s analysis of aggregate data from Bill Johnson and others, “Comparing Commercial and Medicare Professional Service Prices: Public Use File” (Health Care Cost Institute, August 13, 2020), https://tinyurl.com/3xux3hzj; and Centers for Medicare & Medicaid Services, “CY 2017 PFS Final Rule Addenda (ZIP),” Addendum D (accessed December 2020), https://tinyurl.com/ywftbrad.

Quality Ratings for Hospitals, by Price Group, 2018

Data source: CBO’s analysis of aggregate data from Christopher M. Whaley and others, Nationwide Evaluation of Health Care Paid by Private Plans: Findings From Round 3 of an Employer-Led Transparency Initiative, RR-4394-RWJ (RAND, 2020), https://doi.org/10.7249/RR4394.

Estimates of Providers’ Administrative Spending, 2018

Data source: CBO’s reproduction of Box Figure 1 in David M. Cutler, Reducing Administrative Costs in U.S. Health Care, Policy Proposal 2020-09 (Hamilton Project, Brookings Institution, March 2020), https://tinyurl.com/ap9mcczh.

Relationship Between a Hospital’s Share of Medicare and Medicaid Patients and Commercial Insurers’ Average Price for Its Services Over the 2016–2018 Period

Data sources: CBO’s analysis of aggregate data from Christopher M. Whaley and others, Nationwide Evaluation of Health Care Paid by Private Plans: Findings From Round 3 of an Employer-Led Transparency Initiative, RR-4394-RWJ (RAND, 2020), https://doi.org/10.7249/RR4394; and Chapin White, “RAND Hospital Data: Web-Based Tool,” TL-303 (RAND, 2018), www.rand.org/pubs/tools/TL303.html.

Appendix BSources of Estimated Average Prices for Hospitals’ and Physicians’ Services

This appendix describes the sources that underlie the figures and discussion in Chapter 1 of average prices for hospitals’ and physicians’ services. For this analysis, the Congressional Budget Office reviewed studies that compared commercial insurers’ prices with the Medicare fee-for-service (FFS) program’s prices and studies that reported ratios of payments to providers’ costs for private payers (including people without insurance) and for Medicare. (Studies that looked at data before 2005 were not included because high-quality data on prices were not commonly available at that time.) For each major type of service, CBO calculated a simple average of the studies’ estimates (see Table B-1). The estimates of commercial insurers’ prices as a percentage of Medicare FFS’s prices may differ among the studies because of differences in the locations, data sources, and time periods they examined, as well as differences in their methods for adjusting for the health status of patients.

Appendix CStudies Examining the Relationship Between Market Concentration for Providers and Prices Paid by Commercial Insurers

This appendix describes the sources that underlie the discussion in Chapter 4 of how concentration in the markets for hospitals’ services (see Table C-1) and physicians’ services (see Table C-2) affects prices for those providers’ services. This appendix also describes studies that examine the relationship between prices and vertical integration of hospitals and physicians—that is, when physicians’ practices are owned by or affiliated with hospitals (see Table C-3).

About This Document

This report was prepared at the request of the Chairman of the House Committee on the Budget. In keeping with the Congressional Budget Office’s mandate to provide objective, impartial analysis, the report makes no recommendations.

Michael Cohen, Jared Maeda (formerly of CBO), and Daria Pelech wrote the report with guidance from Lyle Nelson (formerly of CBO) and Chapin White. Yiqun Gloria Chen, Carrie H. Colla, Noelia Duchovny, Caroline Hanson, Tamara Hayford, and Ian McCarthy of CBO offered comments. Christian Henry and Joshua Varcie fact-checked the report.

Comments were also provided by Zack Cooper of the Yale University School of Public Health; Paul Ginsburg of the Price School of Public Policy, University of Southern California; Aditi Sen and Zhera Valencia of the Health Care Cost Institute; and Jeffrey Stensland and Ariel Winter of the Medicare Payment Advisory Commission. The assistance of external reviewers implies no responsibility for the final product; that responsibility rests solely with CBO.

Jeffrey Kling and Robert Sunshine reviewed the report. Christian Howlett edited it, and Casey Labrack created the graphics and prepared the report for publication. The report is available at www.cbo.gov/publication/57422.

CBO seeks feedback to make its work as useful as possible. Please send comments to communications@cbo.gov.

Phillip L. Swagel

Director

January 2022