In recent years, consumers have assumed an increasing share of health care spending through high deductible plans. For example, the average deductible for family coverage in Michigan more than doubled from 2002 to 2012, rising from $810 to $1,877, respectively. Deductibles for individual coverage in Michigan grew by 162 percent over the same period of time, increasing from $375 to $982, respectively.1
In order to control rising health spending and provide more information on the cost of care to consumers, policymakers have increasingly focused on publishing data about payments to providers.
As a result of these trends, the topic of “price transparency” has gained momentum in the United States. For the purpose of this brief, price transparency in health care is defined as “the availability of provider-specific information on the price for a specific health care service or set of services to consumers and other interested parties.”2 This brief provides an overview of initiatives by federal and state governments and private entities; discusses the challenges associated with achieving the current goals of price transparency efforts; and highlights opportunities for moving forward to effectively achieve such goals.
Between 1998 and 2010, the Centers for Medicare and Medicaid Services (CMS) launched five websites to publicly report provider performance information, including Nursing Home Compare, Dialysis Facility Compare, Home Health Compare, Hospital Compare, and Physician Compare. These websites allow the public to access information that is aggregated into measures chosen by CMS and endorsed by The Joint Commission, the National Quality Forum, and the Agency for Healthcare Research and Quality. The data, submitted to CMS by a variety of providers in the Medicare program, include information such as hospital characteristics and inspection information; readmissions, complications, and deaths; management of chronic diseases; and patient surveys. Nursing Home Compare and Dialysis Facility Compare summarize results as a five-star rating system for each included nursing home or dialysis facility.3,4 The other websites display a range of process and outcome-based measures for each individual provider included. Hospital Compare allows consumers to compare hospitals based on over 100 different service-specific measures, while Home Health Compare displays 22 process of care measures and nine outcome of care measures.5,6 Additionally, Hospital Compare allows consumers to see the median Medicare payment rate at each hospital for 43 common inpatient hospital services.7
In addition to the five websites, in 2014, the CMS publicly released a database containing payments made in 2012 to 880,000 providers under the Medicare Part B Fee-For-Service program, including inpatient and outpatient hospitals, physicians, and other suppliers, totaling more than $77 billion and covering more than 6,000 services and procedures. The database allows comparisons by physician, specialty, location, medical service, Medicare payment, and submitted charges.8 The federal government recently announced that it will release Medicare physician-payment records annually.9
Since Maine established the first All-Payer Claims Database (APCD) in 2003, over 30 states, including Michigan, have developed—or are seriously considering developing—an APCD, and 12a have fully operational systems.10 In 2013, the Michigan Legislature passed Public Act 107, which authorized a Michigan Health Care Cost and Quality Advisory Committee to develop “recommendations on the creation of a database on health care costs and health care quality in this state.”11
APCDs typically collect patient-specific information from a broad range of payers including private health insurers, Medicaid, children’s health insurance and state employee health benefit programs, prescription drug plans, dental insurers, self-insured employer plans, and Medicare (if available). Some states—including Colorado, New Hampshire, and Maine—have created public-facing websites that allow consumers to search for health care services by price and review the variation in prices of common services. These sites generally include information on several hundred common procedures and were intended to help consumers shop for medical care. For example, a consumer in Denver interested in a hip replacement could look on CO Medical Price Compare and learn that in 2014, the price for a hip joint replacement in Denver ranged from $21,235 to $36,446.12 The reported price is based on paid claims data, not what each provider charges for the service, but is aggregated across payers so it does not reflect what any one payer would have paid for this service.
Health plans and third-party vendors
In recent years, health plans and independent vendors have developed products that provide consumers with cost and quality information, allowing them to comparison shop for health care services. Health plans and vendors have taken a variety of approaches to provide consumers with cost and quality information, ranging from consumer-facing websites to “high-touch” support in the form of email, text, or phone alerts. Common features of vendor products include:
- In-network physician and hospital information, including credentials and contact information;
- Price information for common routine services at various providers in a consumer’s geographic area, tailored to the consumer’s benefit design and physician network; and
- Basic quality information on hospitals and/or individual physicians, displayed in a variety of ways, including star ratings, check marks, or several separate quality measurements from public and private sources of quality information.13
In addition, certain vendors are also releasing price estimates by condition or episodes of care.14
In February 2015, the Health Care Cost Institute, an independent non-profit organization, partnered with four large health plansb to create Guroo, a free consumer-facing website that displays the average cost of more than 70 common health care services by geographic area.15
Challenges with publishing price data
As noted, policymakers and advocates for APCDs generally cite two goals for these systems:
- To provide more information to enable consumers to be more effective at shopping for care; and
- To help lower health care spending through public disclosure of prices.
While publishing cost information has conceptual appeal, available research suggests that it will not be effective in achieving either of the above goals. States with consumer-facing websites or other reporting tools have not yet studied whether this information has changed consumer decision-making. Anecdotal reports from states suggest that consumers are not yet using available tools to compare costs across providers.16 Further, findings from the Kaiser Family Foundation’s April 2015 Health Tracking Poll suggest that very few consumers use available provider quality and price data (see Figure 1).17
FIGURE 1: Percentage of Consumers Who Saw and Used Provider Quality and Price Data over Past 12 Months
SOURCE: Kaiser Family Foundation April 2015 Health Tracking Poll
While no peer-reviewed studies have assessed consumer use of publicly available cost information, research has been done on consumer use and awareness of publicly available information on quality of care. One study published in the American Journal of Managed Care found that fewer than 30 percent of chronically ill adults are aware of publicly available quality data.18
In a survey of adults with chronic conditions in the Aligning Forces for Quality communities, only 14 percent of respondents reported seeing any information comparing the quality among different doctors in the past 12 months, and 29 percent of those respondents reported using the information to make decisions about doctors.19 Further, Health Affairs researchers conducted focus groups in 2013 and concluded that the majority of consumers were unwilling to consider costs when deciding between comparable clinical options.20
In addition to lack of apparent consumer interest in using cost data to choose practitioners of care, publicly available cost data are too aggregated to provide consumers with useful information for decision making.21 For example, in the CO Medical Price Compare system, when a consumer searches by health care service, the website displays a total price for that service at each provider in his or her geographic area. However, as a result of the complex nature of billing for health care services, the website does not tell the consumer what each provider will actually get paid if he or she receives that service. The exact price paid to each provider for a service is affected by several factors, such as bundled payments, the negotiated rate between the provider and a consumer’s insurance company, and the billing of multiple providers for one episode of care.22
In addition, this information is incomplete for any one consumer. For example, CO Medical Price Compare does not provide a consumer with an understanding of what he or she might owe out-of-pocket for the procedure. The negotiated rate to each provider would differ based on that individual consumer’s insurance plan, and the consumer would be responsible for his or her annual deductible before coverage would begin. Other cost-sharing, like co-payments or co-insurance, could require the consumer to pay part of the negotiated rate out-of-pocket, even after the deductible has been satisfied.
Further, as the name implies, All-Payer Claims Databases collect claims data, which are not real-time and do not allow for the calculation of a complete and accurate price before a consumer receives a service. Additionally, in most cases, such databases provide limited or no quality information to supplement price data in order to help consumers understand value.23 Unless quality and cost data are linked, consumers may inaccurately equate lower price with lower-quality providers, potentially encouraging providers to increase prices.24
In its current form, publicly released price information has also not been shown to be effective in encouraging a more cost effective health care system. The available data have significant flaws that have limited its utility for cost controls. Pure claims-based data do not reflect the mix or complexity of patients for any particular provider. There is considerable evidence that when provider performance data are shared publicly, it may change practitioner behavior such that providers avoid more complex patients so that cost and risk profiles look more favorable.25 In addition, it is difficult to adjust claims data in a public release to account for the variation in costs unique to provider systems such as capital, bad debts and graduate medical education.26 And, finally, claims data inevitably have anomalies that are hard to interpret and explain. For example, when Medicare publicly released payments made to providers in 2014, individual payments for services provided by particular doctors were not distinguished from payments made to doctors who were providing care management services and billing on behalf of groups of doctors. As a result, data were misinterpreted by those not familiar with the details of billing and coding.
Indeed, public reporting of provider pricing data may be counterproductive to cost control goals because public reporting of reimbursement rates may encourage lower-paid providers to demand higher prices. In this way, insurers’ abilities to negotiate volume-based discounts would actually be diluted.27
Providing price information to consumers
In light of the challenges discussed above, health plans are well-positioned to provide price and quality information to their members, and many have already begun to offer such tools. Health plan tools are tailored to the consumer’s benefit design and provider network in order to provide more accurate customer-specific information than publicly available APCDs. Further, while state government would require additional investment to collect and publish the data, health plans already have such capabilities.
Providing price information to providers
According to several third-party transparency vendors, providers who are assuming more financial risk for the cost of care have begun to show interest in accessing price and quality data in order to potentially reduce use of marginally beneficial interventions and to help them refer patients to low-cost, high-quality specialists. While vendors have typically focused on the health plan and employer market, several are taking steps to provide products tailored to providers in 2015.28
Federal and state databases
Some information that is currently collected by federal and state databases is being used to improve health care delivery. Data collected through APCDs and CMS can help with a variety of analyses to improve population health, understand demographic and geographic variations in health care utilization, and enhance performance measurement and benchmarking.29 For example, in Utah, the Department of Health used the state’s APCD data to measure the number of diabetic individuals who received blood sugar level screenings within a measurement year. This allowed the state to create a benchmark in order to understand which cities were top, middle, and low performers. In New Hampshire, the Department of Health and Human Services utilized the state’s APCD to look at COPD prevalence in the commercial and Medicaid markets by age and geographic location.30
Increasingly, health information exchanges and other data aggregation tools are providing information to providers of care to help make decisions at the patient and population health levels.31,32 What distinguishes these approaches from other approaches to APCDs is that the information produced in this way is used by public health leadership, health plans, providers and others to improve health outcomes rather than to publish consumer information about the cost of care. It appears that these population health approaches are likely to grow in the future as several states are currently in the implementation phase of developing an APCD with these types of goals.33