The copyright to this article belongs to Wolters  Kluwer.

From Health Reform WK-EDGE, April 28, 2015

Transparency initiatives slow to bring clarity to decision making

By Michelle L. Oxman, J.D., LL.M.

In the United States, the purchase of health care services is unlike any other decision an individual is likely to make because the consumer has less information than she would about any other major purchase. Specifically, the consumer typically will not know what services are included, who is providing them, or what the price will be. For example, a patient undergoing joint replacement surgery may not know to expect charges for an anesthesiologist, a surgical assistant, or a course of rehabilitation after the operation. If the patient’s insurance plan requires higher cost sharing for the use of out-of-network providers, the cost sharing obligation and the network status of all facilities and professionals are crucial to predicting the patient’s obligation.

The move to assure price transparency is intended to remove the secrecy and give both consumers and health care payers greater predictability. The movement has grown significantly since enactment of the Medicare Prescription Drug, Improvement and Modernization Act (MMA) (P.L. 108-173), which began to bring “pay for performance” to Medicare with voluntary reporting of quality data. Government agencies, notably HHS and CMS, and private payers, including insurers and self-insured employers, have tried several ways to track the cost of care and to tie payments to patient outcomes, safety, and other indicia of quality.

This Strategic Perspective will examine the features of helpful transparency tools and compare the federal government’s Hospital Compare and Physician Compare sites to tools created by private entities.

When consumers have no information other than price, they may assume that higher prices buy higher quality. Yet, it is well established that price has little or no relevance to the quality of health care services.

Obstacles to Transparency

2011 report by the Government Accountability Office (GAO) found that there were many obstacles in the path of consumers who sought information about the price of various health care services, including:

  • Fragmentation of care: The items and services associated with a procedure are provided by different entities, all of whom price their work independently. For example, the patient undergoing a joint replacement will have to pay separate fees not only to the surgeon and anesthesiologist, and physical therapist, but also to the facility, the manufacturer of the implant, and a physical therapist, plus the cost of anesthesia, preoperative testing, and supplies.
  • Lack of knowledge: Practitioners often do not know the cost of drugs or procedures that they order for patients, let alone the charges imposed by other professionals or facilities involved in the episode of care.
  • Reluctance or refusal to share information: Many payers and providers consider the amounts they pay or charge for health care services to be confidential, proprietary information. Some providers actually include “gag” clauses in their contracts with payers, explicitly prohibiting the disclosure of information about prices charged or paid. Some health plans that administer benefits for self-insured employers take the position that the employers are not entitled to use the information about claims they have paid to develop their own tool, with or without the cooperation of another party.
  • Unpredictability: For some procedures, the physician or practitioner does not know what complications may arise that will increase the cost.
  • Variation in consumers’ coverage: The obligation of an insured consumer will depend upon the terms of the consumer’s insurance coverage, including whether the deductible has been met; a provider or practitioner is unlikely to have that information.

Price Information Matters

Consumers and self-insured employers pay for health care services and both may lack control over the extent of their payment obligations. As more employers change their benefit packages to high deductible, “consumer driven” plans, consumers’ cost sharing obligations increase. The financial risk leads them to look for price information.

2013 study by the Robert Wood Johnson Foundation showed that consumers have become both anxious and angry about increasing premiums and deductibles. Some people deferred needed care because of the cost. The investigators noted that for the first time, members of the focus groups knew exactly what they spent on premiums, deductibles, and copayments.

Which price information is helpful depends largely on the role of the person using it. For patients with insurance, the most relevant price information is their out-of-pocket cost, that is, deductible, coinsurance, and any costs not covered by the insurance. In a recent report, the Healthcare Financial Management Association (HFMA) Price Transparency Task Force, tend to agree that this information is most easily available to the health plan, and the health plan should be responsible for providing it. For uninsured patients, however, the HFMA recommends that providers such as hospitals be responsible for communicating price information. For these individuals, complete transparency would include provision of information about available assistance programs.

What cost information is available. Information about cost may be derived from paid claims. Health plans will have their own data, as will some self-insured employers. The Medical Expenditure Panel Survey (MEPS) maintained by the Agency for Healthcare Research and Quality contains data on health care spending and utilization from surveys of patients and families, providers, and employers. There also is a database of payment for and utilization of procedures and services by Medicare beneficiaries in the fee-for-service program. It includes data on the 100 most common diagnosis related groups (DRGs), the 30 most common outpatient services, and all services and procedures performed by physicians or other suppliers on more than ten Medicare beneficiaries. In some states, all payer claims databases (APCDs) combine information on expenditures by health insurers, participating self-insured employers, and government programs. According to Catalyst for Payment Reform, several private companies, including Change Healthcare, use their own proprietary databases.

CMS’ cost information. The payment and utilization data for inpatient hospital, outpatient hospital, or physician services are available on the CMS web site. For inpatient hospital services, the data available cover federal fiscal years (FY) 2011 and 2012. Both the 2011 data and the 2012 data show the number of discharges of Medicare beneficiaries from each hospital for the DRGs described, the “average covered charges,” average total payments, and average Medicare payment. The term “average covered charges” is defined on the page as “the provider’s average charge for services covered by Medicare for all discharges in the DRG.” The data may be sorted by DRG, provider number or name, and average Medicare payment. The page also contains tools for exporting or saving the data.

For outpatient services, the services are described by ambulatory payment classification (APC). The available data is from FYs 2011 and 2012. For each APC at each hospital, the page shows the number of the services billed during the year, “average estimated submitted charges,” and “average total payments.” The average submitted charges are the average amounts billed to Medicare, and the average total payments include the beneficiary’s copayment and any deductible applicable to the service as well as the amount that Medicare paid. There is no further breakdown that would allow a consumer to estimate the patient’s share of the costs.

The information about physicians in the payment and utilization database is available as a tab delimited file, so that additional software is required to perform any kind of analysis. It also is sorted into Excel files alphabetically by the physicians’ last names. The page contains several disclaimers not present on the other pages, noting that the data: (1) is limited to Medicare fee-for-service beneficiaries and may not reflect the scope of the physician’s practice; (2) is not risk-adjusted to account for the severity of patients’ illnesses; and (3) is not intended to inform the user about the quality of services. In order to view data, one must download an entire file. Once the file is downloaded, the user must, once again, be very familiar with data sorts and filters in order to search.

The CMS pages for inpatient hospital, outpatient hospital, and physician and other provider utilization and payments are designed and formatted to convey to the user that the files must be downloaded in order to view any information. Statements about downloading the data appear prominently near the top and middle of each of the pages. The option to search without downloading massive quantities of unnecessary data is not communicated explicitly. Rather, it appears at the bottom of the page as “other related links.”

With respect to the physician information, there are two such links. The first is to the Medicare physician and other supplier look-up tool. From there, one can search for an individual physician or organization by name, city, and state or National Provider Identifier (NPI) number. The tool then retrieves data on services for which the physician billed Medicare during 2012, including the number of Medicare beneficiaries, the average billed amount, average Medicare allowed amount, and the average amount Medicare paid. Each service is designated by Healthcare Common Procedure Coding System (HCPCS®). The information provided is simply raw data concerning the Medicare population that the physician served during that year; no information about the physician’s total patient load or the number of unique Medicare patients is included.

The second “related link,” the Interactive Physician and Other Supplier Dataset, contains similar information. This page contains some help for users who need direction on how to use sorts and filters.

In summary, the information available about the cost of procedures or services, whether from hospitals or physicians and other practitioners, is not in a form that would help patients determine their cost. To the extent that CMS’ purpose is to help Medicare beneficiaries to comparison shop for services or providers, they are not well designed to achieve it. The beneficiary’s health plan could provide more detailed information more easily. In addition, the health plan, not CMS, would, or at least should, have information about which providers are members of the network and are accepting new patients.

What Quality Information Is Available?

Much of the information about quality available through public sources concerns processes or application of best practices, for example, whether a patient presenting with a heart attack is given aspirin, or the number of a physician’s patients who have been screened for diabetes. More specific information on safety, experience with the patient’s condition or planned procedure, and other factors addressing the quality of services, is harder to find.

Nevertheless, the relevance of quality-related information depends greatly on the role of the stakeholder using it. Quality data reported by hospitals such as the frequency of readmissions within 30 days, central line associated bloodstream infections, or catheter-associated urinary tract infections, are important to CMS and providers because they affect payments to the hospital. These indicators of the frequency of hospital-acquired conditions are less helpful to consumers, however, at least when they are presented alone. Patients tend to want information about outcomes, such as how long they should expect to be in post-surgical rehabilitation, and about patients’ experiences with the providers. Whether the nurses communicated clearly, or patients’ questions were answered, tend to be more important to patients than statistics. For physician visits or other outpatient care, some patients and families will consider availability outside of standard business hours, such as evening or Saturday appointments, to be indicators of cost and quality.

The federal government maintains quality information about hospitals and other providers received from mandatory or voluntary reports on specific measures. The results of certification and complaint investigation surveys are publicly available, so that a member of the public may find the deficiencies for which a facility has been cited. CMS uses enrollment and claims data, and, in some cases, Veterans Administration (VA) data to find readmissions within 30 days of discharge, number of Medicare patients treated, and other information. It combines this information with data on adherence to measures considered to be timely and effective that hospitals report to the Centers for Disease Control (CDC) and the CMS Quality Improvement Organization Clinical Data Warehouse. Whether a hospital must report certain data to CMS depends in large part upon how the hospital or unit is paid. Hospitals subject to the standard inpatient prospective payment system (IPPS) report one set of data, while outpatient hospitals, and rehabilitation or psychiatric “distinct part units” as well as stand-alone rehabilitation, long-term care, or psychiatric hospitals. VA hospitals, children’s hospitals, and others are examined separately, as are hospitals with organ transplant facilities.

In addition to CMS, the Agency for Healthcare Research and Quality maintains data and makes software available to providers and researchers to use with their own data. Some professional organizations, such as the American College of Surgeons, collect data that is voluntarily reported to them by members and submit it to CMS.

CMS’ Compare Websites

Federal government efforts to promote transparency have included CMS’ Compare sites, including Hospital CompareNursing Home Compare and Physician Compare. The sites are designed differently and use different sources of data. Even within Hospital Compare, the sources of data vary.

Hospital Compare reports data over several domains:

  • Organizational or structural matters, such as participation in registries and databases devoted to specific conditions, such as heart surgery, stroke, general surgery, or “nursing sensitive care;”
  • Timely and effective care: adherence to best practices in the care of certain conditions, such as heart attack, heart failure, pneumonia, stroke, prevention of blood clots, and care of children with asthma;
  • Frequency of unplanned readmissions or deaths within 30 days of admission to the hospital;
  • Frequency of healthcare-associated infections or other preventable conditions that a patient develops while at the hospital;
  • Frequency of complications from surgery, particularly heart surgery and joint replacements;
  • Patient experiences as documented in the Consumers Assessment of Healthcare Providers and Systems (CAHPS®); and
  • Use of medical imaging.

Most of the information reported is drawn only from Medicare patients who were enrolled in traditional Medicare for the entire year before their admission to the hospital. New Medicare beneficiaries, members of Medicare Advantage or special needs plans, and adults with other sources of payment are excluded. These limits apply to joint replacement surgery. The information on death or unplanned readmission to the hospital within 30 days includes both this group of patients and patients age 65 or older who were admitted to VA hospitals.

There is very little focus on price transparency. In fact, the only price information on Hospital Compare relates to Medicare spending per beneficiary and Medicare payment for heart attack patients. CMS measures the value of care for cardiac patients by comparing the average Medicare expenditures for cardiac patients in relation to the percentage of deaths among those cardiac patients within 30 days of hospitalization. Until April 16, 2015, Hospital Compare did not provide quality ratings, but compared the hospital’s performance on a particular measure to the national and state averages. For example, the national average Medicare payment for heart attack patients is presented with the national average percentage of unplanned readmissions or deaths, and each hospital’s results are reported as greater than, about the same as, or less than the national average. These two facts, taken together, are supposed to indicate the value of the care Medicare patients received at the hospital.

The data in Hospital Compare may not be complete. Hospital Compare did not list all hospitals in the area searched. A search for hospitals in or near Skokie, Illinois, a suburb just north of Chicago, missed Skokie Hospital. Although the hospital is one of four hospitals in the Chicago suburbs that comprise NorthShore University Health System (NorthShore), only two are found in Hospital Compare, and neither is described as part of NorthShore. The search page includes a box to check to search for children’s hospitals. Yet, a search for children’s hospitals in Chicago, Illinois found zero results, notwithstanding the existence of Lurie Children’s Hospital (formerly Children’s Memorial), the Shriner’s Hospital, and the University of Chicago Comer Children’s Hospital. Nothing on the Hospital Compare site explains why hospitals are not listed.

During 2014, CMS added data on links between quality and payment, on hospital readmissions within 30 days, value-based purchasing, and reduction of hospital-acquired conditions. CMS had implemented regulations that required hospitals to report their overall performance and their improvement. Hospitals that score above a certain threshold are rewarded with additional payments, while those at the bottom receive reduced payments. CMS reports information on mortality rates within 30 days of admission for heart attack or heart failure, stroke, pneumonia, chronic obstructive pulmonary disease, complications after total surgical replacement of a hip or knee, and a hospital-wide measure of unplanned readmission for any cause within 30 days of admission. The page on linking quality to payment does not score, rank, or compare hospitals. Instead, the page explains that a score exceeding 1.0 means that a hospital had more of the condition than the average hospital and a hospital with a score below 1.0 was better than average on the same domains. The page instructs the user to go back to the entry to the Hospital Compare to find quality information relevant to choosing a hospital. A click on one of the links to quality measures, such as reduction in readmissions, brings up a list of thousands of hospitals and lists each measure by abbreviations.

Five-star ratings. The Nursing Home Compare site was the first to use the five-star rating system to help consumers digest the information available. It has ratings for health inspections, staffing, quality measures, and an “overall” rating. Nursing homes may score between one star, for much below average, and five stars, much above average. The overall score is supposed to be a composite of the scores on the three domains. The site is consumer-friendly in that it is possible to sort homes by rank or score and the scores are easy to understand. However, as CMS explains, “below average” score on staffing has no effect on the facility’s score.

The Hospital Compare site launched five-star ratings on April 16, 2015. The major difference between the hospitals’ star ratings and those on Nursing Home Compare is that the hospitals’ stars apply only to patient experiences measured in the. The CAHPS is administered by the AHRQ. It is sent to a random sample of Medicare patients after discharge from the hospital, and the results are updated quarterly. Patients are asked about their experiences communicating with their nurses and doctors, whether they received help promptly when they asked for it, and whether they clearly understood what they needed to do upon discharge from the hospital. This patient feedback is likely of interest to prospective patients. Nevertheless, it is subjective, and it may or may not accurately describe the patient’s actual understanding. It is not tied to any information about either price or quality of care.

What good transparency tools provide. Currently, a variety of tools are available from government, health plans, self-insured employers, and private companies. The GAO has updated its research on price and quality transparency to review currently available tools. Other entities, such as Catalyst for Payment Reform (CPR), also have described the features they believe are most helpful to patients and families.

Both the GAO and CPR agree that a good transparency tool uses plain language and clear graphics. It is organized in a way that allows users to select the information they want, to compare providers, and to discern patterns in the data. Long lists of physicians, for example, are overwhelming. The language used to describe or explain a quality measure should be clear and not technical. A good tool provides the information consumers need in a manner that they can understand easily. It should cover a broad range of both services and providers so that consumers will be likely to find information that applies to their situation. Information about cost should be as relevant to the consumer’s situation as possible, considering coverage, deductibles, and copayments. If a user’s coverage either requires or provides incentives for the use of network providers, it is essential that the provider directory be current and that the incentives be clearly explained. Toward that end, in March 2015, Kevin Counihan, Marketplace CEO and Director of the Center for Consumer Information and Insurance Oversight (CCIIO) told the National Summit on Health Care Price, Cost, and Quality Transparency that CMS planned to require health plans offered on the federal Marketplace to update their provider directories at least monthly. Both the GAO Report and the Catalyst for Payment Reform believe that tools should allow for side-by-side comparisons of facilities or providers.

Both the GAO and CPR agree that CMS’ Compare tools are much less helpful than they could be. One possible reason, according to the GAO, is that CMS developed the tools for purposes that are mutually inconsistent. The agency uses the information not only to inform consumers, but also to drive improvements to quality through public exposure and direct effects upon reimbursement. To the extent that the agency encounters resistance from facilities and physicians, it may relax its efforts.

The Physician Compare web site illustrates the effects of these competing objectives. In the five years since the site was launched, the only evaluative information to be posted concerned accountable care organizations and 66 group practices. The site states whether an individual physician participates in a quality reporting or incentive program, but no information about performance is given.

Commercial Transparency Tools

Private companies have entered the market in the last several years to take the publicly available information and organize and format it to be usable by consumers. Companies such as Change Healthcare (CHC) and Castlight create tools for health plans and self-insured employers to make available to their employees and subscribers. In recent years, these firms have refined their products to give patients and families a reasonable approximation of their likely financial responsibility, combined with the quality information in which consumers are most interested. CHC’s tools allow users to search for providers by location and network participation and to predict the costs of procedures.

Tiffany Pack, Vice President of Engagement at Change Healthcare, says that her firm combines the data available through CMS with proprietary information such as patient reviews and claims data to create its tools for consumers. Large self-insured employers find these tools attractive because Pack’s team focuses heavily on educating consumers about how to use their health benefits and what constitutes value in health care services. The company’s Health Care University helps consumers examine how they used their health care benefits and how they could save money on the services they need. Information about doctors and hospitals’ performance is available, but patients or consumers do not rate doctors on Change Healthcare. Rather, the company uses Sternuus, a private vendor, to conduct a structured interview of the subscribers and aggregate the data. Pack says that there is sufficient distance between the providers or practitioners and the consumer opinion that there have been no issues about the quality or accuracy of the feedback.

CHC’s priority is to make the tool as user-friendly as possible. Consumers may earn financial incentives by using the tool, and the incentives play an important part in motivating consumers to use the tool. In addition to search features and the ability to organize the data according to their own priorities, consumers may sign up for alerts via text or email to let them know about opportunities to save on their health care costs.

Refinements needed. CPR Director Suzanne Delbianco and the GAO both find that great strides have been made on the journey to transparency. The next refinements would make the tools more reliable. Perhaps first on the list is the need for consistency in defining episodes of care for specific conditions. When the definitions of episodes of care are inconsistent, patients are likely to be misinformed about the total costs of a procedure. It is necessary not only to assure that the items and services that ordinarily are part of the bundle of services are included, but also to educate consumers about the risk of avoidable complications so that they will ask the necessary questions when complications arise.

Companies: Catalyst for Payment Reform; Change Healthcare

MainStory: StrategicPerspectives NewsFeed HealthCareAcquiredConditionNews HospitalReadmissionNews ProviderPaymentNews PhysicianNews QualityNews InpatientFacilityNews

Health Reform / WK-EDGE