This is the accessible text file for GAO report number GAO-14-279 
entitled 'Medicare: Certain Physician Feedback Reporting Practices of 
Private Entities Could Improve CMS's Efforts' which was released on 
March 26, 2014. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

Report to Congressional Committees: 

March 2014: 

Medicare: 

Certain Physician Feedback Reporting Practices of Private Entities 
Could Improve CMS's Efforts: 

GAO-14-279: 

GAO Highlights: 

Highlights of GAO-14-279, a report to congressional committees. 

Why GAO Did This Study: 

Health care payers—-including Medicare-—are increasingly using VBP to 
reward the quality and efficiency instead of just the volume of care 
delivered. Both traditional and newer delivery models use this 
approach to incentivize providers to improve their performance. 
Feedback reports serve to inform providers of their results on various 
measures relative to established targets. The American Taxpayer Relief 
Act of 2012 mandated that GAO compare private entity and Medicare 
performance feedback reporting activities. 

GAO examined (1) how and when private entities report performance data 
to physicians, and what information they report; and (2) how the 
timing and approach CMS uses to report performance data compare to 
that of private entities. GAO contacted nine entities—-health insurers 
and statewide collaboratives-—recognized for their performance 
reporting programs. Focusing on physician feedback, GAO obtained 
information regarding report recipients, data sources used, types of 
performance measures and benchmarks, frequency of reporting, and 
efforts to enhance the utility of performance reports. GAO obtained 
similar information from CMS about its Medicare feedback efforts. 

What GAO Found: 

Private entities GAO reviewed for this study selected a range of 
measures and benchmarks to assess physician group performance, and 
provided feedback reports to physicians more than once a year. Private 
entities almost exclusively focused their feedback efforts on primary 
care physician groups participating in medical homes and accountable 
care organizations, which hold physicians responsible for the quality 
and cost of all services provided. They limited their feedback 
reporting to those with a sufficient number of enrollees to ensure the 
reliability of reported measures. The entities decided on the number 
and type of measures for their reports, and compared each group's 
performance to multiple benchmarks, including peer group averages or 
past performance. All the entities used quality measures, and some 
also used utilization or cost measures. Because of the variety of 
quality measures and benchmarks, feedback report content differed 
across the entities. Some entities noted that in addition to national 
benchmarks, they compared results to state or regional level rates to 
reflect local patterns of care which may be more relevant to their 
physicians. Most health insurers spent from 4 to 6 months to generate 
their performance reports, a period that allowed them to amass claims 
data as well as to make adjustments and perform checks on the measure 
calculations. Commonly, private entities issued interim feedback 
reports, covering a 1-year measurement period, on a rolling monthly, 
quarterly, or semiannual schedule. They told GAO that physicians 
valued frequent feedback in order to make changes that could result in 
better performance at the end of the measurement period. 

Feedback from the Centers for Medicare & Medicaid Services (CMS) 
included quality measures determined by each medical group, along with 
comparison to only one benchmark, and CMS did not provide interim 
reports to physicians. The agency has phased in performance feedback 
in order to meet its mandate to apply value-based payment (VBP) to all 
physicians in Medicare by 2017, a challenge not faced by private 
entities. In September 2013, CMS made feedback reports available to 
6,779 physician groups. While private entities in this study chose the 
measures for their reports, CMS tied the selection of specific quality 
measures to groups' chosen method of submitting performance data. 
Although both CMS and private entities focused their feedback on 
preventive care and management of specific diseases, CMS's reports 
contained more information on costs and outcomes than some entities. 
While private entities employed multiple benchmarks, the agency only 
compared each group's results to the national average rates of all 
physician groups that submitted data on any given measure. CMS's use 
of a single benchmark precludes physicians from viewing their 
performance in fuller context, such as relative to their peers in the 
same geographic areas. CMS's report generation process took 9 months 
to complete, several months longer than health insurers in the study, 
although it included more steps. In contrast to private entity 
reporting, CMS sent its feedback report to physicians once a year, a 
frequency that may limit physicians' opportunity to make improvements 
in advance of their annual payment adjustments. 

The Department of Health and Human Services generally concurred with GAO
's recommendations and asked for additional information pertaining to 
the potential value of using multiple benchmarks to assess Medicare 
physicians' performance. 

What GAO Recommends: 

The Administrator of CMS should consider expanding performance 
benchmarks to include state or regional averages, and disseminating 
feedback reports more frequently than the current annual distribution. 

View [hyperlink, http://www.gao.gov/products/GAO-14-279]. For more 
information, contact James Cosgrove at (202) 512-7114 or 
cosgrovej@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Private Entities Selected a Range of Measures and Benchmarks to Assess 
Physician Group Performance, and Provided Feedback More than Once a 
Year: 

CMS Feedback Included Group-Determined Physician Quality Measures and 
Only One Benchmark; CMS Issued Reports Less Frequently than Private 
Entities: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Private Entity and Medicare Performance Feedback for 
Hospitals: 

Appendix II: Quality Measures Used in Sample Physician Feedback 
Reports Provided by Selected Private Entities: 

Appendix III: Comments from the Department of Health and Human 
Services: 

Appendix IV: GAO Contact and Staff Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Key Features of Hospital Feedback Reporting by Selected 
Private Entities and CMS: 

Table 2: Number of Quality Measures Categorized by Type Found in 
Sample Physician Feedback Reports Provided by Selected Private 
Entities: 

Figures: 

Figure 1: CMS's Phased Approach to Providing Value Modifier (VM) 
Reports to Eligible Professionals (EP): 

Figure 2: Private Entity Examples of Hemoglobin A1C Measures as 
Presented in Four Sample Reports: 

Figure 3: Private Entity's Display of Information on Specialty 
Referral Patterns in a Sample Report: 

Figure 4: Private Entity Examples of Emergency Department (ED) Visit 
Measures as Presented in Two Sample Reports: 

Figure 5: Private Entity Examples of Benchmark Comparisons as 
Presented in Three Sample Reports: 

Figure 6: Illustration of Timelines to Report Production for Health 
Insurers and Statewide Health Care Collaboratives: 

Figure 7: Cost Measures Displayed in CMS's 2013 Quality and Resource 
Use Reports: 

Figure 8: Performance Benchmarks Displayed in CMS's 2013 Quality and 
Resource Use Reports: 

Figure 9: CMS Sample Composite Scores, Quality Tier, and Value 
Modifier: 

Figure 10: Report Generation Timeline for CMS Performance Feedback 
Reports, September 2013: 

Abbreviations: 

ABC: Achievable Benchmark of Care: 

ACO: accountable care organization: 

ACSC: Ambulatory Care Sensitive Condition: 

CAH: critical access hospital: 

CMS: Centers for Medicare & Medicaid Services: 

COPD: chronic obstructive pulmonary disease: 

EHR: electronic health record: 

EP: eligible professional: 

FFS: fee-for-service: 

HHS: Department of Health and Human Services: 

PQRS: Physician Quality Reporting System: 

QRUR: Quality and Resource Use Reports: 

VBP: value-based payment: 

VM: Value Modifier: 

[End of section] 

United States Government Accountability Office: 
GAO:
441 G St. N.W. 
Washington, DC 20548: 

March 26, 2014: 

The Honorable Ron Wyden: 
Chairman: 
The Honorable Orrin G. Hatch: 
Ranking Member: 
Committee on Finance: 
United States Senate: 

The Honorable Fred Upton: 
Chairman: 
The Honorable Henry Waxman: 
Ranking Member: 
Committee on Energy and Commerce: 
House of Representatives: 

The Honorable Dave Camp: 
Chairman: 
The Honorable Sander Levin: 
Ranking Member: 
Committee on Ways and Means: 
House of Representatives: 

Increasingly, health care payers--including Medicare--are rethinking 
the way they reimburse providers in an attempt to shift away from 
paying solely for the volume of care delivered and toward paying them 
for the value of their care. One such approach--known as value-based 
payment (VBP)--links a portion of physician compensation to achieving 
specified levels of performance. VBP can be used as a means to 
improving quality and efficiency in the traditional health care 
delivery environment by encouraging providers to address gaps in 
patient care and consider the likely costs and benefits of care. It 
can also be used with newer care delivery models, such as accountable 
care organizations (ACO) and patient-centered medical homes.[Footnote 
1] Under these arrangements, payers hold teams of providers 
responsible for all of a patient's care. They reward those who 
coordinate services across providers and make cost-effective referral 
decisions, among other practices. 

While physicians and other providers may intend to furnish 
consistently high-quality, efficient care, they may not always know 
how well they do or where practice changes are needed. Therefore, a 
key element of the VBP approach is for payers to develop performance 
feedback reports to indicate specific opportunities for 
improvement.[Footnote 2] Performance feedback entails collecting data 
on measures of quality and cost of care, assessing performance against 
benchmarks, and communicating results to providers. For example, 
periodic feedback reports can make providers aware of the percentage 
of their patients receiving appropriate screening tests, or those with 
potentially avoidable emergency department visits. The expectation is 
that giving detailed, timely feedback to providers will enhance their 
ability to take actions that improve performance.[Footnote 3] 

The American Taxpayer Relief Act of 2012 mandated that GAO compare how 
private entities and the Centers for Medicare & Medicaid Services 
(CMS)--the agency within the Department of Health and Human Services 
(HHS) that administers the Medicare program--conduct performance 
feedback reporting for health care providers.[Footnote 4] To meet this 
requirement, we conducted briefings for congressional staff on our 
preliminary findings in September 2013. This report contains 
information we provided during those briefings, updated with 
additional information, addressing: 

1. how and when private entities--such as health insurers--report 
performance data to physicians, and what information they report; and: 

2. how the timing and approach CMS uses to report performance data to 
physicians compare to that of private entities. 

In addition, appendix I contains information on how private entities 
and Medicare provide performance feedback to hospitals. 

To examine how and when private entities report performance data to 
physicians, and to identify what information they provide, we 
contacted nine private entities that had experience with VBP programs 
or that had innovative features in their performance feedback 
programs. To make our selection, we asked representatives of America's 
Health Insurance Plans,[Footnote 5] Blue Cross Blue Shield 
Association,[Footnote 6] and Network for Regional Healthcare 
Improvement to suggest leading organizations that met those criteria. 
[Footnote 7] We also considered programs profiled in peer-reviewed 
literature, as well as those operating in varying geographic areas 
across the country. We chose six health insurers, and three statewide 
health care collaboratives--organizations comprising providers, 
payers, and employers that focus on quality improvement activities--as 
follows: [Footnote 8] 

* Aetna: 

* Blue Shield of California: 

* Blue Cross and Blue Shield of Florida, Inc. 

* Highmark Blue Cross Blue Shield: 

* Horizon Blue Cross Blue Shield of New Jersey: 

* Iowa Healthcare Collaborative: 

* Maine Health Management Coalition: 

* Oregon Health Care Quality Corporation: 

* WellPoint: 

We interviewed representatives of these private entities regarding the 
feedback report recipients, data sources used, types of performance 
measures and benchmarks, frequency of reporting, and efforts to 
enhance the utility of their performance reports. We also requested 
sample physician feedback reports to learn how the data were 
presented. In some cases, entities had multiple performance reporting 
initiatives. We focused on the reports that were most similar to the 
type of reporting to physicians that CMS provided to medical groups. 

Our findings regarding private entity performance reporting to 
physicians are limited to the entities we interviewed and cannot be 
generalized to other health insurers and health care collaboratives. 
In this report, we describe private entities' feedback programs in 
operation in 2013, although performance reporting continues to evolve 
as organizations adopt newer payment and delivery models. We did not 
evaluate the effectiveness of the feedback in altering physician 
practice patterns. Also, we did not gather information on the payment 
incentives, if any, associated with these entities' reporting 
initiatives, as that issue was beyond our scope.[Footnote 9] Because 
nationwide interest in VBP has been largely aimed at physician care, 
we primarily focused our review on performance reporting to 
physicians, and as noted, present additional information on our 
methodology and findings related to hospital feedback reporting in 
appendix I.[Footnote 10] 

To learn how the timing and approach CMS uses to report performance 
data to physicians compare to that of private entities, we obtained 
CMS documentation similar to that received from the selected entities. 
We analyzed information regarding CMS report recipients, data sources 
used, types of performance measures and benchmarks, frequency of 
reporting, and efforts to enhance the utility of the reports. We also 
examined a prototype of the report CMS provided to physicians in 
September 2013. We spoke with CMS officials about their report 
preparation process and about components of the feedback program that 
differ from those of private entities. 

We conducted this performance audit from April 2013 to March 2014 in 
accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. 

Background: 

Laws enacted since 2006 have directed CMS to collect performance 
information on providers and eventually reward quality and efficiency 
of care rather than reimburse for volume of services alone. 

* The Tax Relief and Health Care Act of 2006 required the 
establishment of the Physician Quality Reporting System (PQRS) to 
encourage physicians to successfully report data needed for certain 
quality measures.[Footnote 11] PQRS applies payment adjustments to 
promote reporting by eligible Medicare professionals (EP)--including 
physicians, nurses, physical therapists, and others.[Footnote 12] In 
2013, EPs could report data to PQRS using claims, electronic health 
records (EHR)[Footnote 13] or a qualified registry,[Footnote 14] or 
opt for CMS to calculate quality measures using administrative claims 
data. Under its group practice reporting option, CMS allows EPs to 
report to PQRS as a group, either through a registry or a web-based 
interface. 

* The Medicare Improvements for Patients and Providers Act of 2008 
established the Physician Feedback Program, under which CMS was 
required, beginning in 2009, to distribute confidential feedback 
reports, known as Quality and Resource Use Reports (QRUR), to show 
physicians their performance on quality and cost measures.[Footnote 15] 

* The Patient Protection and Affordable Care Act required HHS to 
coordinate the Physician Feedback Program with a Value Modifier (VM) 
that will adjust fee-for-service (FFS) physician payments for the 
relative quality and cost of care provided to beneficiaries.[Footnote 
16] In implementing the VM, CMS's Center for Medicare intends to use 
PQRS and cost data from groups of EPs defined at the taxpayer 
identification number level to calculate the VM and then report the 
payment adjustments in the QRURs. As required in the act, CMS plans to 
apply the VM first to select physicians in 2015 and to all physicians 
in 2017. 

As required by law, CMS implemented a performance feedback program for 
Medicare physicians, which serves as the basis for eventual payment 
adjustments. (See figure 1.) 

Figure 1: CMS's Phased Approach to Providing Value Modifier (VM) 
Reports to Eligible Professionals (EP): 

[Refer to PDF for image: illustration] 

Legislative actions: 

2006: 
The Tax Relief and Health Care Act of 2006 required the establishment 
of the Physician Quality Reporting System. 

2008: 
The Medicare Improvements for Patients and Providers Act of 2008 
established the Physician Feedback Program. 

2009: 
The Patient Protection and Affordable Care Act added the VM 
requirement. 

CMS actions: 

2008 and 2009: 
CMS tested certain measures and developed a feedback report prototype 
to be distributed to some physicians. [Feedback report] 

2009 and 2010: 
CMS distributed feedback reports with both quality and resource use 
measures to certain EPs. [Feedback report] 

2011: 
CMS made group-level reports available to 35 groups nationwide. 
[Feedback report] 

December 2012: 
CMS made reports available to groups of 25 or more EPs in nine states 
and others nationwide. [Feedback report] 

September 2013: 
CMS made reports available to 6,779 groups of 25 or more EPs 
nationwide. [Feedback report] 

September 2013: 
Reports include a preview of each group's VM. [Value modifier] 

Fall 2014: 
CMS expects to provide reports to all groups in all states, including 
small groups and solo practitioners. [Feedback report] 

January 1, 2015: 
VM to be applied to physicians in groups of 100 or more EPs. [Value 
modifier] 

January 1, 2016: 
VM to be applied to physicians in groups of 10 or more EPs. [Value 
modifier] 

2017: 
VM to be applied to all physicians. [Value modifier] 

Source: GAO. 

[End of figure] 

In our December 2012 report on physician payment incentives in the VM 
program, we found that CMS had yet to develop a method of reliably 
measuring the performance of physicians in small practices, that CMS 
planned to reward high performers and penalize poor performers using 
absolute performance benchmarks, and that CMS intended to annually 
adjust payments 1 year after the performance measurement period ends. 
We recommended that CMS develop a strategy to reliably measure the 
performance of small physician practices, develop benchmarks that 
reward physicians for improvement as well as for meeting absolute 
performance benchmarks, and make the VM adjustments more timely, to 
better reflect recent physician performance.[Footnote 17] CMS agreed 
with our recommendations, but noted that it was too early to fully 
implement these changes. 

Private Entities Selected a Range of Measures and Benchmarks to Assess 
Physician Group Performance, and Provided Feedback More than Once a 
Year: 

Private entities we reviewed provided feedback mostly to groups of 
primary care physicians practicing within newer delivery models. Each 
entity decided which measures to report and which performance 
benchmarks to use, leading to differences in report content across 
entities. Largely relying on claims data, health insurers spent from 4 
to 6 months to produce the annual reports. To meet the information 
needs of physicians, they all provided feedback throughout the year. 
The entities also generally offered additional report detail and other 
resources to help physicians improve their performance. 

Private Entities Focused Their Feedback Efforts on Groups of Primary 
Care Physicians Practicing within Newer Delivery Models: 

The private entities in our review had discretion in determining the 
number and type of physicians to be included in their performance 
reporting initiatives, and their feedback programs generally included 
only physician groups participating in newer delivery models--medical 
homes and ACOs--with which they contract.[Footnote 18] Within this set 
of providers, the entities used various approaches to further narrow 
the physician groups selected to receive performance feedback. For 
example, one entity told us that only physician groups accredited by a 
national organization focused on quality were eligible for 
participation in its medical home program, which included physician 
feedback reports. 

Private entities' feedback programs were generally directed toward 
primary care physician practices. One entity defined primary care as 
family medicine, internal medicine, geriatrics, and pediatrics; and 
included data on the services furnished by nurse practitioners and 
physician assistants in its medical group reports. The entities 
indicated that they rarely provided reports directly to specialty care 
physician groups. Among those that did, the programs typically focused 
on practice areas considered significant cost drivers--obstetrics/ 
gynecology, cardiology, and orthopedics. 

Entities further limited their physician feedback programs to groups 
participating in medical homes with a sufficient number of attributed 
enrollees to ensure the reliability of the reported measures. In 
medical home models, enrollees are attributed to a physician (or 
physicians) responsible for their care, who is held accountable for 
the quality and cost of care, regardless of by whom or where the 
services are provided. Among those entities we spoke with, the minimum 
enrollment size for feedback reporting varied widely, with most 
requiring a minimum of between 200 and 1,000 attributed enrollees to 
participate in the program. For example, one entity had two levels of 
reporting in its medical home program, differentiated by the number of 
attributed enrollees. In one medical home model, the entity required 
more than 2,000 attributed enrollees for participation and rewarded 
the practices through shared savings. In a second medical home model, 
the entity included practices with fewer than 1,000 attributed 
enrollees, but these practices did not share in any savings. 

According to the entities in our study, small physician practices 
(including solo practitioners) typically received performance reports 
for quality improvement purposes only. Because smaller practices may 
not meet minimum enrollment requirements needed for valid 
measurement,[Footnote 19] private entities generally did not link 
their performance results to payment or use them for other purposes. 
For example, one entity provided feedback to practices of one to three 
primary care physicians upon request, but did not publicly report 
these practices' data on its website. To increase the volume of 
patient data needed for reliable reporting, some entities pooled data 
from several small groups and solo practitioners and issued aggregate 
reports for those small practices. Most of the entities that used this 
method said they applied their discretion in forming these "virtual" 
provider groups; however, another entity commented that allowing small 
practices to voluntarily form such groups for measurement purposes 
would be advantageous. 

Private Entities Decided Which Measures to Report and Compared 
Physician Performance to Various Benchmarks, Leading to Differences in 
Report Content across Entities: 

Because each private entity in our study determined the number and 
types of measures on which it evaluated physician performance, the 
measures used in each feedback program differed. Each entity decided 
on quality measures to include, and many also identified utilization 
or cost measures for inclusion.[Footnote 20] In addition, one entity 
noted that it allowed ACOs to choose 8 to 10 measures from among a set 
of about 18 measures. To assess physicians' quality and 
utilization/cost results, the entities used absolute or relative 
performance benchmarks. 

Quality Measures: 

Private entities generally report on physician quality using many more 
process of care measures than outcomes of care measures.[Footnote 21] 
Entities in our review commonly included indicators of clinical care 
in areas such as diabetes care, cardiovascular health, and prevention 
or screening services for both their adult and pediatric patients. The 
most common measure reported by all entities was breast cancer 
screening, followed by hemoglobin A1C measures,[Footnote 22] a service 
used to monitor diabetes. 

We found wide variation in the number and type of measures in private 
entities' quality measure sets.[Footnote 23] The total number of 
quality measures used in the feedback reports ranged from 14 to 51. 
[Footnote 24] Measures typically fell into one of several measurement 
areas, each with as few as one or as many as 20 individual measures. 
For example, in the quality measurement areas for pulmonary and 
respiratory conditions, one private entity reported on a single 
measure (appropriate use of medications for asthma), while another 
reported three measures (appropriate use of medications for asthma, 
appropriate testing for pharyngitis, and avoidance of antibiotic 
treatment for adults with acute bronchitis). Although primarily 
focused on clinical quality measures, entities also included 
nonclinical measures, such as patient safety and patient satisfaction. 
(See appendix II for more information on the number and types of 
quality measures included in sample reports provided by the entities 
we reviewed.) 

Even when entities appeared to report on similar types of measures in 
common areas, we found considerable variability in each measure's 
definition and specification. For this reason, results shown in 
physician feedback reports may not be comparable across entities. As 
shown in figure 2, the diabetes hemoglobin A1C measure was defined and 
used in different ways in our selected entities' reports. In some 
cases, entities calculated the percentage of enrollees with diabetes 
within a certain age range that received the test. In other cases, the 
entities calculated the percentage of enrollees with diabetes within a 
certain age range that had either good or poor control of the 
condition, as determined from a specified hemoglobin A1C result. In 
addition, some entities defined their diabetic patient population as 
enrollees from 18 to 75 years of age, while another did not indicate 
the age range, and one entity set the age range from 18 to 64 years of 
age. 

Figure 2: Private Entity Examples of Hemoglobin A1C Measures as 
Presented in Four Sample Reports: 

[Refer to PDF for image: 4 report examples] 

Example 1: 

Measure	Grouping: Diabetes; 
Measure	Number: 100024; 
Measure Short Description: Diabetes: Hemoglobin Alc testing; 
Measure Long Description: This measure calculates the percentage of 
members age 18 to 75 with diabetes receiving annual HbA1c testing; 
Eligible Population in Denominator: 589; 
Numerator: 525; 
Rate = Numerator divided by Eligible Population: 89.13%. 

Measure	Grouping: Diabetes; 
Measure	Number: 100244; 
Measure Short Description: Diabetes Hemoglobin Alc poor control (>9.0%; 
Measure Long Description: This measure calculates the percentage of 
members age 18 to 75 with diabetes that demonstrate poor glycemic 
control, based on a HbAlc level greater than 9%; 
Eligible Population in Denominator: 390; 
Numerator: 61; 
Rate = Numerator divided by Eligible Population: 15.64%. 

Measure	Grouping: Diabetes; 
Measure	Number: 100336; 
Measure Short Description: Diabetes: Hemoglobin A1c control (<8.0%); 
Measure Long Description: This measure calculates the percentage of 
members age 18 to 75 with diabetes that demonstrate good glycemic 
control, based on a HbA1c level less than 8%; 
Eligible Population in Denominator: 390; 
Numerator: 287; 
Rate = Numerator divided by Eligible Population: 73.59%. 

Example 2: 

Diabetes Management: 

Alc Testing: The percentage of members 18 to 64 years of age with 
diabetes (type 1 and type 2) who had the following: Hemoglobin Alc 
(HbAlc) testing; 
Total Eligible Members: 591
Members with Care: 548
Rate: 92.72%. 

Example 3: 

Diabetes: HbA1C control: 59%. 

Example 4: 

Measure: Diabetes Care, HbAlc Test (age 18-75); 
Number of Patients: 149; 
Your Quality Scores: 75.8%; 
95% Confidence Interval: (68% - 82%). 
		
Source: Withheld. Images used with permission.

[End of figure] 

Utilization or Cost Measures: 

Some, but not all, private entities in our review included utilization 
or cost measures in their performance reports to physicians. Total 
cost of care per enrollee was the most commonly used measure, but cost 
measures disaggregated by type of service--facility, pharmacy, primary 
care physician, and specialty--were also used. Some entities described 
how they limited their reporting of a total cost of care measure to 
those medical groups with a large number of enrollees. In one case the 
minimum enrollment size was 20,000 enrollees and in another it was 
2,500 enrollees. Officials from one entity also told us that they 
allowed smaller physician practices to combine their data in order to 
meet the required number of enrollees for receiving feedback on cost 
of care. 

In addition to feedback on the total cost of care per enrollee, some 
reports given to groups of primary care physicians contained 
information on the cost of care provided by specialists in the 
entity's network. For example, one entity provided trend data that 
included the number of specialist visits (total and by type) and the 
number of patients with one or more visits for these specialty areas. 
(See figure 3.) For the two specialties with the most enrollee visits 
during the measurement period--orthopedic surgery and dermatology--the 
entity also provided the medical group with data on which specialists 
were seen most frequently and their cost per visit. This information 
was intended to encourage cost-efficient referrals. Another entity 
said it was focused on a program in July 2013 to provide feedback to 
primary care physicians on cardiologists' performance showing where 
care was being delivered most efficiently. By providing such 
information, the entity expected primary care physicians to take cost 
differences into account when making referrals, rather than basing 
referrals solely on historical habits. Disseminating information to 
primary care physicians about the relative cost of specialty care 
providers is a key aspect of medical home and ACO programs. 

Figure 3: Private Entity's Display of Information on Specialty 
Referral Patterns in a Sample Report: 

[Refer to PDF for image: table] 

Total Specialist Visits: 
Prior: 18,585; 
Current: 19,226; 
Trend: 3%. 

Unique Patients: 
Prior: 7,002; 
Current: 7,334; 
Trend: 5%. 

Multiple Visits[A]: 
Prior: 3,941; 
Current: 4,015; 
Trend: 2%. 

Visits by Specialty: 

Orthopedic Surgery; 
Prior: 2,765; 
Current: 3,145; 
Trend: 14%. 

Dermatology; 
Prior: 2,326; 
Current: 2,525; 
Trend: 9%. 
			
Radiology; 
Prior: 2,988; 
Current: 2,400; 
Trend: -20%.

Ophthalmology; 
Prior: 1,669; 
Current: 1,791; 
Trend: 7%. 

Cardiovascular Disease; 
Prior: 1,523; 
Current: 1,551; 
Trend: 7%. 

[A] less than 1 admission in measurement period. 

Source: Withheld. Image used with permission. 

[End of figure] 

The entities were fairly consistent in the number and types of 
utilization measures they selected for feedback reporting. The most 
common utilization measures reported by our private entities were 
physicians' generic drug prescribing rates, followed by emergency 
department visits, inpatient visits, hospital readmissions, and 
specialist visits. One entity provided additional detail under the 
emergency department visits measure to show the number of patients 
that repeatedly seek care at emergency departments. Officials from the 
entity told us that this measure was included to alert physicians of 
potentially avoidable hospital visits so that they can encourage 
patients to use office-based care before seeking care in more costly 
settings. (See examples of this measure as presented by private 
entities in their sample reports in figure 4.) 

Figure 4: Private Entity Examples of Emergency Department (ED) Visit 
Measures as Presented in Two Sample Reports: 

[Refer to PDF for image: 2 report examples] 

Example 1: 

Period: 1/2011-12/2011; 
Member Months: 59,459; 
Average membership: 4,955; 
Actual ED Visits/1,000 Measurement Year: 165.90. 

Example 2: 

Practice: 

Emergency Department Utilization: 
ED Visits/1000: 171.7; 
% ACSE ED Visits: 8.8%; 
Count Frequency ED Users (less than 2): 10; 
% Frequency Users (less than 2): 2%. 

Figure: Emergency Department Vists by Quarter: 

Vist log for time period of 2009 Q3 through 2012 Q2. 

Source: Withheld. Image used with permission. 

Note: In example 2, Ambulatory Care Sensitive Conditions (ACSC) refers 
to conditions where appropriate ambulatory care prevents or reduces 
the need for admission to the hospital. 

[End of figure] 

Benchmarks: 

To evaluate physician performance, the selected private entities 
compared the measures data to different types of benchmarks. Some 
entities compared each physician group's performance results to that 
of a peer group (e.g., others in the entity's network or others in the 
collaborative's state or region); some entities compared physician 
groups' results to a pre-established target; and others gauged 
physician groups' progress relative to their past performance. (See 
figure 5.) Entities generally used two or three such benchmarks in 
their feedback reports. For example, one entity separately displayed 
results for the medical home's commercially insured, Medicare insured, 
and composite patient population. Within each of these population 
groups, it compared the practice's performance to the average for 
nonmedical home practices, as well as to the practice's performance in 
the prior measurement year. The entity also gave narrative detail to 
indicate favorable or unfavorable performance. The most common 
benchmark for the entities in our study was a physician group's 
performance relative to the previous measurement period. However, some 
entities used this benchmark only for utilization/cost measures and 
not for quality measures. 

Figure 5: Private Entity Examples of Benchmark Comparisons as 
Presented in Three Sample Reports: 

[Refer to PDF for image: 3 report examples] 

Example 1: 

Region: Western; 
Specialty: Family Practice; 
Clinical Quality Performance Detail: 

Clinical Category/Quality Measure: Acute Pharyngitis Testing: Throat 
culture or antigen agglutination test for streptococcus 3 days before 
through 3 days after a sole diagnosis of acure pharyngitis is 
identified and after which an antibiotic was dispensed within 3 days; 
Practice Quality %: 91%; 
Specialty Quality %: 78%; 
Practice to Specialty %: 100%; 
Earned Points: 1.00. 

Clinical Category/Quality Measure: Adolescent Well-Care Vistis; 1 or 
more comprehensive well-care visits with a PCP or an OB/FYN; 
Practice Quality %: 52%; 
Specialty Quality %: 59%; 
Practice to Specialty %: 88%; 
Earned Points: 0.00. 

Example 2: 

Adult Quality Measures - Medical Group Results: 

Contqct Name: Sample Q. Sample; 
Group Name: The Sample Group. 

Measure: Breast Cancer Screening (age 40-69); 
Number of Patients: 140; 
Your Quality Score: 64.3%; 
95% Confidence Level: (56%-72%); 
Average Quality Score for State: 70.4%; 
State ABC Benchmark: 87.0%; 
HEDIS 2011 Commercial PPO 90th Percentile: 72.5%. 

Measure: Cervical Cancer Screening (age 21-64); 
Number of Patients: 430; 
Your Quality Score: 69.8%; 
95% Confidence Level: (65%-74%); 
Average Quality Score for State: 71.15; 
State ABC Benchmark: 88.5%; 
HEDIS 2011 Commercial PPO 90th Percentile: 79.0%. 

Measure: Chlamydia Screening (age 16-24); 
Number of Patients: 111; 
Your Quality Score: 50.5%; 
95% Confidence Level: (41%-60%); 
Average Quality Score for State: 41.4%; 
State ABC Benchmark: 72.2%; 
HEDIS 2011 Commercial PPO 90th Percentile: 51.0%. 

Example 3: 

PCHM - ED Visits: 
Practice vs. Non PCMH Comparison. 

ED visits per 1,000 members: 
(based on claims incurred in 5 months and paid in 8 months vs. same 
time period in prior period). 

Commercial: 

Practice: 
Prior: 149; 
Current: 150; 
0%. 

Non PMH: 
Prior: 177; 
Current: 192; 
+8%. 

Medicare: 

Practice: 
Prior: 336; 
Current: 361; 
+8%. 

Non PMH: 
Prior: 297; 
Current: 337; 
+14%. 

Composite: 

Practice: 
Prior: 158; 
Current: 160; 
+1%. 

Non PMH: 
Prior: 184; 
Current: 201; 
+9%. 

Favorable 8% variance when compared to Non PCMH. 

Source: Withheld. Image used with permission.  

Note: In example 2, the state Achievable Benchmark of Care (ABC) 
refers to the health care collaborative's calculated benchmark for the 
state's clinics based on its own criteria. 

[End of figure] 

Largely Relying on Claims Data, Health Insurers Spent 4 to 6 Months to 
Produce Annual Reports and Typically Provided Feedback in the Interim: 

Private entity officials told us they relied on claims as their 
primary data source for performance reporting. However, several 
private entities noted shortcomings in relying solely on claims data--
the billing codes that describe a patient's diagnoses, procedures, and 
medications--for performance reporting.[Footnote 25] Some entities 
supplemented their claims data by obtaining information from EHRs, 
patient satisfaction surveys, or chart extractions.[Footnote 26] 
Entities noted that using EHR data was resource-intensive for both 
providers and payers, because they depended on physician groups to 
submit the information. The entities we spoke to have had limited 
success in using EHR data as a primary data source, although many saw 
it as complementary to claims data. Another entity supplemented its 
claims data with data from registries that compile information from 
administrative data sets, patient medical records, and patient 
surveys, and thus have the capacity to track trends in quality over 
time. 

The health insurers in our review typically spent from 4 to 6 months 
to produce and distribute annual performance reports; in contrast, the 
health care collaboratives spent 9 to 10 months. (See illustrations of 
these timelines in figure 6.) As is common in the health insurance 
industry, payers require a 3-month interval after the performance 
period ends--referred to as the claims run-out--to allow claims for 
the services furnished late in the measurement period to be submitted 
and adjudicated for the report. The claims run-out was followed by 1 
to 3 months to prepare the data, a period that allowed for provider 
attribution, risk-adjustment,[Footnote 27] measure calculation, 
[Footnote 28] and quality assurance.[Footnote 29] One statewide health 
care collaborative stated that the quality assurance process is 
helpful in increasing physician trust because the group is able to 
compare its own data with the collaborative's data before results are 
final. The statewide health care collaboratives we spoke with required 
additional time to collect and aggregate data from multiple health 
insurers, and their final reports were issued at least 9 months after 
the end of the performance period. The time needed for some or all of 
these report production steps varied depending on the entity and the 
types of measures included. 

Figure 6: Illustration of Timelines to Report Production for Health 
Insurers and Statewide Health Care Collaboratives: 

[Refer to PDF for image: 2 timeline illustrations] 

Illustration of health plan timeline: 

Performance period: 12 months; 
4-6 month lag time: 
* Final claims submission: 3 months; 
* Report production: 1-3 months; 
Report issued. 

Illustration of health care collaborative timeline:; 
Performance period: 12 months; 
9-10 month lag time: 
* Final claims submission: 3 months; 
* Data submission: 1 month; 
* Quality assurance: 4 months; 
* Measure calculation: 1 month; 
* Results review: 1 month; 
Report issued. 

Source: GAO analysis of private entity information. 

[End of figure] 

Collaboratives often used all-payer claims databases--centralized data 
collection where each payer submits claims data on that state's health 
care providers--for aggregate reporting to providers. Officials from 
entities told us that all-payer claims databases are helpful because 
they provide physicians with a better picture of their entire patient 
panel, not just results determined by individual payers for limited 
sets of patients. One entity noted that it aggregates its quality data 
with other payers in its commercial market through a statewide 
organization, and no one payer can provide statistically meaningful 
data to a physician group on its own. Officials from one entity with 
all-payer claims database experience told us that the addition of 
Medicare data into these databases would improve the information 
available for measurement and feedback. In addition, one entity 
suggested that a multipayer database could help with feedback to 
physicians in groups of all sizes, including small practices, because 
the higher number of patients would generate sufficient data for 
calculating reliable measures. However, one entity acknowledged that 
using all-payer databases requires more time for merging data from 
different payers in different formats, and another entity noted the 
challenges of customizing reports for each medical groups' patient 
population. 

Private entities told us that physicians valued frequent feedback on 
their performance so that they have time to make practice changes that 
may result in better performance by the end of the measurement period. 
In response, these entities typically provided feedback reports on an 
interim basis throughout the measurement period. Interim reports 
typically covered a 1-year performance period, and were commonly 
issued on a rolling monthly, quarterly, or semiannual schedule. 
Entities also noted that frequent reporting throughout the period 
updated physicians on their performance so that year-end results were 
better expected and understood. Some entities in our study elected to 
issue interim reports that build up to the 12-month performance period 
by continually adding data from month to month. Those that used 
preliminary data that may not account for all final claims in building 
reports told us that such data starts to become useful about 3 to 6 
months into the performance year. They also stated that, although the 
interim reports may be limited by the use of rolling or incomplete 
data, providers generally seek this information for early 
identification of gaps in care. 

Private Entities Generally Offered Access to Additional Report Detail 
and Other Resources to Help Physicians Improve Performance: 

Private entities generally offered additional report detail intended 
to enhance physicians' understanding of the information contained in 
their reports or in response to physician requests for more data. 
Private entity officials told us that, because physicians prefer 
dynamic reports with as much detail as possible, they generally sent 
reports that can be expanded to show individual physician or patient-
level data. Some entities formatted their reports to include summary-
level information on quality and cost measures in labeled sections, 
with supplemental information following the summary data. Other 
entities provided additional reports or supplemental data through a 
web portal that allowed providers to see individual physician or 
patient-level detail. Private entities sent reports in multiple file 
formats, such as in a spreadsheet, some of which allowed report 
recipients to sort their data.[Footnote 30] 

Entities in our study also offered resources designed to assist 
physician groups with actionable steps they can take to improve in the 
next performance period. Most entities told us they offered resources 
to physician groups, such as consultations with quality improvement 
professionals, forums for information-sharing, and documents on best 
practices. For example, one entity's staff worked directly with 
practices to improve their results by distributing improvement 
guidelines for each performance measure included in the feedback 
report. In addition, the entity's officials told us they also convened 
workgroups to review trend information and paid particular attention 
to differences between medical homes and nonmedical homes. 

CMS Feedback Included Group-Determined Physician Quality Measures and 
Only One Benchmark; CMS Issued Reports Less Frequently than Private 
Entities: 

CMS has provided feedback to increasing numbers of physician practices 
each year in order to eventually reach all physicians. Each medical 
group's chosen method of quality data submission determined the 
quality measures included in its report, to which CMS added health 
care costs and certain outcomes measures. CMS's report generation 
process took slightly longer than that of most private entities in our 
study, and the agency did not provide interim performance data during 
the measurement period. CMS feedback reports have included information 
to assist providers in interpreting their performance results. 

CMS Provided Feedback to Physicians in Groups with 25 or More Eligible 
Professionals in 2013: 

Unlike the private entities we contacted, which selected a limited set 
of physicians to receive feedback reports, CMS is mandated to apply 
the VM to all physicians by 2017. Therefore, the agency faces certain 
challenges not faced by private entities as it has expanded its 
feedback program to reach increasing numbers of physicians. In 
preparation for implementation of the VM, CMS provided performance 
reports to nearly 4,000 medical groups in September 2013. In 2014, CMS 
plans to disseminate reports to physicians in practices of all sizes. 

As of September 2013, CMS had not yet determined how to report to 
smaller groups and physicians in solo practices. According to CMS, the 
decision not to present VM information to smaller groups stemmed from 
concerns regarding untested cost metrics and administrative 
complexity. CMS agreed with a 2012 GAO recommendation to develop a 
strategy to reliably measure the performance of solo and small 
physician practices, but has not yet finalized such a strategy. 
[Footnote 31] 

CMS Feedback Contained Varying Quality but Consistent Cost and 
Outcomes Measures and Assessed Performance Against a Single Benchmark: 

Under the CMS approach to performance reporting, the content of 
feedback reports related to quality measures may vary across 
providers. Unlike our selected private entities, the agency has 
allowed physician groups to select the method by which they will 
submit quality-of-care data, which, in turn, determines the measures 
on which they receive feedback. CMS used claims data for a consistent 
set of measures in all of its feedback reports for performance on cost 
and outcomes. 

Quality Measures: 

For the CMS 2013 reports, medical groups submitted data on quality 
measures to CMS via a web interface or through a qualified registry; 
if a group did not select either of these options, the agency 
calculated quality measures based on claims data. Both CMS and private 
entities focused on preventive care and management of specific 
diseases.[Footnote 32] 

* Web interface. Quality measures under this method pertain to care 
coordination, disease management, and preventive services.[Footnote 
33] CMS required groups reporting via the web interface to submit data 
on 17 quality measures--such as hemoglobin A1C levels for control of 
diabetes--for a patient sample of at least 218 beneficiaries.[Footnote 
34] 

* Registries. Some groups submitted data for quality measures via 
qualified registries--independent organizations, typically serving a 
particular medical specialty, that collect and report these data to 
CMS. CMS required groups reporting to a qualified registry to submit 
at least three measures--such as whether cardiac rehabilitation 
patients were referred to a prevention program--for at least 80 
percent of patients.[Footnote 35] 

* Administrative claims. As a default, if a group did not report via 
web interface or qualified registry, CMS calculated quality measures 
using claims data. In September 2013, the majority of groups with 25 
or more EPs--nearly 90 percent--received quality scores based on 
claims data. CMS calculated performance on a set of 17 quality 
indicators, including several composite measures. For example, the 
diabetes composite measure included several different measures of 
diabetes control. 

Regardless of the method a group selected to submit quality-of-care 
data, CMS used claims to calculate three outcomes measures--two 
ambulatory care composite measures and hospital readmission. One 
ambulatory care composite included hospitalization rates for three 
acute conditions: bacterial pneumonia, urinary tract infections, and 
dehydration. Another composite included hospitalization rates for 
three chronic conditions: diabetes, chronic obstructive pulmonary 
disease (COPD), and heart failure. 

Cost Measures: 

CMS included cost measures--several of which differed from the 
measures private entities in our study reported to physicians--in all 
2013 feedback reports (see figure 7). Using claims data, CMS 
calculated an overall measure of the cost of care as the total per 
capita costs for all beneficiaries attributed to each physician 
group.[Footnote 36] In addition, CMS separately reported total per 
capita costs for attributed beneficiaries with any of four chronic 
conditions: diabetes, heart failure, COPD, or coronary artery disease. 
This contrasts with the private entities that typically measured a 
more limited set of measures focused on physicians' generic drug 
prescribing rates and hospital utilization. 

Figure 7: Cost Measures Displayed in CMS's 2013 Quality and Resource 
Use Reports: 

[Refer to PDF for image: table] 

Per Capita Costs for All Attributed Beneficiaries (Domain Score = 
+0.41): 

Cost Categories: All Beneficiaries: 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 1,351; 
Per Capita Costs Before Risk Adjustment: $19,135; 
Per Capita Costs After Risk Adjustment: $10,898; 
Performance of All 1,032 Groups with at Least 100 Eligible	 
Professionals: 
Benchmark Per Capita Costs (Risk-Adjusted): $10,265; 
Average Range: 
Benchmark -1 Standard Deviation: $8,722; 
Benchmark +1 Standard Deviation: $1,808. 

Per Capita Costs for Beneficiaries with Specific Conditions (Domain 
Score = -0.03)							

Cost Categories: Diabetes; 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 373; 
Per Capita Costs Before Risk Adjustment: $25,396; 
Per Capita Costs After Risk Adjustment: $14,732; 
Performance of All 1,032 Groups with at Least 100 Eligible	 
Professionals: 
Benchmark Per Capita Costs (Risk-Adjusted): $14,788; 
Average Range: 
Benchmark -1 Standard Deviation: $12,379; 
Benchmark +1 Standard Deviation: $17,198. 

Cost Categories: COPD: 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 149; 
Per Capita Costs Before Risk Adjustment: $36,685; 
Per Capita Costs After Risk Adjustment: $24,396; 
Performance of All 1,032 Groups with at Least 100 Eligible	 
Professionals: 
Benchmark Per Capita Costs (Risk-Adjusted): $24,153; 
Average Range: 
Benchmark -1 Standard Deviation: $19,840; 
Benchmark +1 Standard Deviation: $28,466. 

Cost Categories: Coronary Artery Disease: 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 418; 
Per Capita Costs Before Risk Adjustment: $26,036; 
Per Capita Costs After Risk Adjustment: $17,750; 
Performance of All 1,032 Groups with at Least 100 Eligible	 
Professionals: 
Benchmark Per Capita Costs (Risk-Adjusted): $17,265; 
Average Range: 
Benchmark -1 Standard Deviation: $14,415; 
Benchmark +1 Standard Deviation: $20,115; 

Cost Categories: Heart Failure: 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 227; 
Per Capita Costs Before Risk Adjustment: $34,857; 
Per Capita Costs After Risk Adjustment: $24,467; 
Performance of All 1,032 Groups with at Least 100 Eligible	 
Professionals: 
Benchmark Per Capita Costs (Risk-Adjusted): $26,013; 
Average Range: 
Benchmark -1 Standard Deviation: $21,237; 
Benchmark +1 Standard Deviation: $30,788. 

Source: CMS.

Note: Each domain score is an average of the standardized scores for 
all measures in the domain with at least 20 cases. Up to six equally 
weighted quality domain scores make up a medical group practice's 
quality composite score. 

[End of figure] 

Benchmarks: 

While private entities typically displayed performance information 
relative to multiple benchmarks for certain quality measures--such as 
peer groups, pre-established targets, or past performance--CMS 
compared each physician group's performance only to the national 
average (see figure 8). National average rates comprised all EPs and 
provider groups that reported on that particular quality measure. For 
quality measures, the agency compared a group's current performance to 
the prior-year national average of all groups. CMS reported 
calculating benchmarks by weighting the performance rate of each 
physician and group of physicians submitting data through any 
reporting mechanism for that specific quality measure, regardless of 
specialty, by the number of beneficiaries used to calculate the 
performance rate. According to CMS, where data for a measure are only 
available through one reporting option, such as a registry, the 
benchmark is the performance of all other groups that used the same 
reporting method to submit data for that quality measure. For cost 
measures, CMS compared a group's current performance to current-year 
national averages.[Footnote 37] 

Figure 8: Performance Benchmarks Displayed in CMS's 2013 Quality and 
Resource Use Reports: 

[Refer to PDF for image: table] 

Chronic Obstructive Pulmonary Disease (COPD): 

Performance Measures: COPD-1 COPD: Bronchodilator Therapy; 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 32; 
Performance Rate: 87.5%; 
Performance of All POPS Participants Reporting the Measure: 
Benchmark Rate: Not Available; 
Average Range: 
Benchmark -1 Standard Deviation: Not Available; 
Benchmark +1 Standard Deviation: Not Available. 

Coronary Artery Disease (CAD): 

Performance Measures: CAD-1 CAD: Antiplatelet Therapy; 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 401; 
Performance Rate: 92.8%; 
Performance of All POPS Participants Reporting the Measure: 
Benchmark Rate: 82.8%; 
Average Range: 
Benchmark -1 Standard Deviation: 76.9%; 
Benchmark +1 Standard Deviation: 88.8%. 

Performance Measures: CAD-2 CAD: Lipid Control; 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 401; 
Performance Rate: 72.1%; 
Performance of All POPS Participants Reporting the Measure: 
Benchmark Rate: 88.8%; 
Average Range: 
Benchmark -1 Standard Deviation: 73.9%; 
Benchmark +1 Standard Deviation: 100.0%. 

Performance Measures: CAD-7: CAD: ACE Inhibitor or ARB Therapy for 
Patients with CAD and Diabetes and/or LVSD; 
Your Medical Group Practice's Performance: 
Number of Eligible Cases: 246; 
Performance Rate: 52.9%; 
Performance of All POPS Participants Reporting the Measure: 
Benchmark Rate: 69.0%; 
Average Range: 
Benchmark -1 Standard Deviation: 53.8%; 
Benchmark +1 Standard Deviation: 84.1%. 
						
Source: CMS.

[End of figure] 

According to CMS, this benchmarking approach should spur cost and 
quality improvement nationwide. However, this method may not always 
produce a national or nationally representative benchmark for quality 
of care, as it only captures providers who reported that measure in 
that year. In addition, a private entity in our review noted that 
national benchmarks do not reflect more local patterns of care. 
Several entities calculate benchmarks at the state or regional level; 
the performance of peers in the same geographic area may differ from 
national averages and have more relevance to the providers. For 
example, absent multiple performance benchmarks--both national and 
local--a physician group might be unaware that while its performance 
exceeds national benchmarks, it is below the average within the state. 
Moreover, as we concluded in a 2012 report,[Footnote 38] without 
consistent measures showing performance over time, a physician group 
cannot gauge whether its performance was improving. 

CMS also provided broader benchmarks, in the form of composite scores. 
CMS's 2013 feedback reports also included a ''first look'' at the VM, 
which did not yet affect the physician group's payment, but showed 
each group how its payments could be affected with the implementation 
of the VM in 2017.[Footnote 39] The agency determined the VM based on 
each group's performance on quality and cost measures relative to 
other Medicare providers. To determine the VM, CMS classified each 
quality measure into one of six domains, following the national 
priorities related to clinical care, patient experience, 
population/community health, patient safety, care coordination, and 
efficiency established in its National Quality Strategy. CMS then 
calculated a quality composite score, weighing each domain equally. 
Similarly, to produce a cost composite score, CMS aggregated the per 
capita costs for beneficiaries with four chronic diseases into one 
composite, and weighted this composite equally with total overall 
costs to produce a cost composite. CMS then combined the quality and 
cost composite scores and compared them to national averages to 
determine an overall quality tier,[Footnote 40] which then determined 
the group's VM.[Footnote 41] (See figure 9.) 

Figure 9: CMS Sample Composite Scores, Quality Tier, and Value 
Modifier: 

[Refer to PDF for image: illustration] 

Illustration depicts the following: 

Your Quality Composite Score: Average; 
Standard deviations from National Mean: -0.3. 

Your Quality Composite Score: Average; 
Standard deviations from National Mean (Negative scores are better): 
0.04. 

Your beneficiaries Average Risk Score: 83rd Percentile. 

Your Quality Tiering Performance: Average Quality, Average Cost. 

Your Value-Based Payment Adjustment Based On Quality Tiering. 

Source: CMS. 

[End of figure] 

Also, unlike the private entities we interviewed, CMS's feedback 
reports did not provide information on a group's performance relative 
to prior years. CMS plans to expand to multiple benchmarks in the 
future--for example, CMS finalized for 2014 adjusting the benchmark 
based on the specialty composition of the groups. In 2012, GAO 
recommended that CMS develop benchmarks that assess the degree of 
performance improvement as well as the extent to which performance 
meets absolute targets.[Footnote 42] However, according to CMS, it is 
premature to establish such benchmarks without more data from medical 
groups, increased participation in PQRS, and an understanding by 
providers that the reported data will be used to adjust payment. 

CMS Spent 9 Months to Produce the 2013 Feedback Reports, and Did Not 
Provide Interim Reports: 

CMS's report generation process took longer than that of most private 
entities in our study because it required more steps. While most 
health insurers generated performance reports in 4 to 6 months, CMS 
issued reports about 9 months after the end of the January to December 
2012 reporting period. 

To produce its 2013 physician feedback reports using administrative 
claims, CMS began with the standard claims run-out period followed by 
intervals for provider attribution, measure calculation, risk-
adjustment, and quality assurance. (See figure 10.) CMS officials said 
they allowed a 3-month run-out interval to account for providers' late-
year claims submissions. After the run-out period, CMS required 5 to 6 
months for a series of additional tasks needed to prepare the data for 
reporting. For groups that submitted data to CMS via the web interface 
or registry options, CMS gave these groups 3 months to submit such 
data after the end of the 12-month performance period. CMS then 
calculated the measures for these options over a period of the next 
several months. Although FFS beneficiaries see multiple physicians, 
CMS attributed each beneficiary to a single medical group through its 
yearly attribution process.[Footnote 43] It used the claims for the 12-
month reporting period to determine which groups provided the 
beneficiary the most primary care and then assigned responsibility for 
performance on quality and cost measures to that group. Following 
attribution, the agency risk-adjusted the cost measures to account for 
differences in beneficiary characteristics[Footnote 44] and 
complexity, and standardized the cost measures by removing all 
geographic payment adjustments. Finally, CMS officials said they 
performed data checks to ensure accuracy before the reports were 
disseminated. 

Figure 10: Report Generation Timeline for CMS Performance Feedback 
Reports, September 2013: 

[Refer to PDF for image: timeline illustration] 

Performance period: 12 months; 
9-10 month lag time: 
* Final claims and quality data submission: 3 months; 
* Provider attribution: less than 1 month; 
* Geographic adjustment removal: 1 month; 
* Measure calculation: 2 months; 
* Risk adjustment and quality assurance: 3 months; 
Report issued. 

Source: GAO analysis of CMS information. 

[End of figure] 

According to health insurers and collaboratives, physicians find that 
frequent feedback enables them to improve their performance more 
quickly;[Footnote 45] however, CMS did not provide physicians interim 
performance feedback.[Footnote 46] However, with only annual feedback 
from CMS, physicians may be missing an opportunity to improve their 
performance on a more frequent basis. Asked if more frequent reporting 
was considered, CMS officials cited concerns about the time it would 
take to generate each set of reports. With each round, the agency 
would need to attribute all beneficiaries to a medical group, risk-
adjust and standardize the cost measures, and compute the benchmarks 
for each measure. In addition, providing interim reports on quality 
data would require certain providers to report more frequently. For 
example, providers who submit via registry would need to finalize 
their data more often than annually. However, experts and CMS 
officials have stated that, with continued adoption of advanced data 
reporting technology, CMS may be able to generate reports more 
frequently. 

CMS Included Explanatory Information in Its Reports and Offered Other 
Resources Online: 

CMS provided general information on its website and through the 
Medicare Learning Network, to assist providers in understanding the 
performance feedback and VM.[Footnote 47] Unlike private entities, CMS 
has not provided tailored guidance or action steps to help providers 
improve their scores. However, CMS resources included steps to access 
reports, a review of methodology, suggested ways to use the data in 
reports, and contact information for technical support. 

A representative acting on behalf of a medical group could access the 
group's QRUR. In addition, CMS's web-based reports allowed providers 
to access further detail on the Medicare beneficiaries attributed to 
the group. For example, physicians could view their patients' 
percentage of total cost by type of service and hospital admission 
data. 

CMS included explanatory information within the reports for providers. 
In addition to comparative performance data, reports made available in 
September 2013 included a description of the attribution methods, the 
number of providers billing in each medical group, information about 
each attributed patient's hospitalizations during the year, and other 
details about the group's performance. In addition, CMS included 
within the QRUR a glossary of terms used in the feedback report. 

Conclusions: 

Payers have been refining their performance reports for physicians, a 
key component of their VBP initiatives. Private entities have 
selectively rolled out their feedback programs, generally applying 
them to relatively large groups of primary care physicians 
participating in medical homes and ACOs. Although they are not uniform 
in their approaches, the entities in our study used their discretion 
to select a limited number of quality and utilization/cost measures, 
calculated them using claims data, and used them to assess performance 
against a variety of benchmarks. In response to physicians' needs, 
their feedback reports tended to be frequent, timely, and dynamic. 

CMS's approach to performance reporting faces some unique challenges. 
First, it is driven by the statutory requirement that, by 2017, 
Medicare pay FFS physicians in groups of all sizes, including 
specialists, using a VM. Second, the agency has had to develop the 
feedback program in the context of pre-existing incentive programs, 
such as PQRS. CMS finalized several key changes to the feedback 
program for future reporting periods, as it expands the application of 
the VM to all physicians. Specifically, CMS continues to modify 
program components such as measures and reporting mechanisms as it 
works to align the reporting and feedback aspects of multiple 
programs. Despite these program modifications, we found that certain 
features of private entities' feedback programs, which are lacking in 
CMS's program, could enhance the usefulness of the reports in 
improving the value of physician care. 

* CMS's use of a single nationwide benchmark to compare performance on 
quality and cost ignores richer benchmarking feedback that could 
benefit physicians. Private entities in our study measured provider 
performance against several benchmarks. CMS's reliance on a national 
average as the sole benchmark precludes providers from gauging their 
performance relative to their peers in the same geographic area. 
Without such contextual information, providers lack the feedback to 
better manage their performance and target improvement efforts. 

* Additionally, CMS disseminates feedback reports only once a year 
(for example, September 2013). This gives physicians little time 
(October through December) to analyze the information and make changes 
in their practices to score better in the next measurement period. The 
private entities we reviewed sent reports more than once a year, and 
reported that greater frequency of reporting enabled more frequent 
improvements. Without interim performance reports, providers may not 
be able to make needed changes to their performance in advance of 
their annual VM payment modifications. 

Our findings also support past GAO recommendations that CMS reward 
physicians for improvement as well as performance against absolute 
benchmarks, and develop a strategy to reliably measure solo and small 
practices, such as by aggregating data. 

Recommendations for Executive Action: 

As CMS implements and refines its physician feedback and VM programs, 
the Administrator of CMS should consider taking the following two 
actions to help ensure physicians can best use the feedback to improve 
their performance: 

* Develop performance benchmarks that compare physicians' performance 
against additional benchmarks such as state or regional averages; and: 

* Disseminate performance reports more frequently than the current 
annual distribution--for example, semiannually. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to HHS for comment. In its written 
response, reproduced in appendix III, the department generally agreed 
with our recommendations, and reiterated our observation that the 
agency faces unique challenges with its mandate to report to Medicare 
FFS providers in groups of all sizes that encompass all specialty care 
areas. 

HHS conditionally agreed with our recommendation that reporting 
physician performance using multiple benchmarks would be beneficial, 
but asked for further information on private entities' practices and 
their potential use for Medicare providers. As we stated in the report, 
private entities generally use two or three different types of 
benchmarks to provide a variety of performance assessments. We found 
alternative benchmarks that could enhance Medicare feedback reporting 
by allowing physicians to track their performance in their own 
historical and geographic context. For example, some entities' reports 
included physician group performance on certain measures relative to 
their past performance, a recommendation we previously made to HHS in 
December 2012. Although it agreed to consider developing benchmarks 
for performance improvement, HHS has yet to do so. A comparison to 
past performance allows a medical group to see how much, if at all, it 
has improved regardless of where it stands relative to its peers. In 
this way, CMS can motivate physicians to continuously improve their 
performance. In addition, some entities in our review compared 
physician performance data to statewide or regional-level benchmarks. 
Because of the number of Medicare physicians, CMS has extensive 
performance data, which could enable more robust localized peer 
benchmarks than any individual health plan could generate. As we 
noted, such benchmarks reflect more local patterns of care that may be 
more relevant to physicians than comparisons to national averages 
alone. 

HHS further asserted that, because the physician feedback program's 
key purpose is to support the national VM program, it is appropriate 
to limit reporting to a single national benchmark. HHS expressed 
concern that displaying other benchmarks could be misleading and 
confusing for the purposes of the VM. However, CMS's reports provide a 
group's VM payment adjustment in a concise, one-page summary, as shown 
in figure 9. We do not believe that additional benchmark data, 
displayed separately, would detract from the information provided on 
the summary page, and could enhance the value of the reports for 
physicians. 

HHS agreed with our second recommendation to disseminate feedback 
reports more frequently than on an annual basis. As seen in the 
private entity practices of using rolling or preliminary data for 
interim reporting, disseminating reports more frequently can assist 
physicians in making improvements to their performance before CMS 
determines their VM payment adjustment. HHS commented that producing 
more frequent reports would first require modifying the PQRS data 
collection schedules. For example, groups of EPs that use the web 
interface and registry options currently are only required to submit 
data to CMS once a year. The registry option will eventually require 
groups to submit data to CMS on a quarterly or semiannual basis, and 
HHS noted that these requirements would have to be synchronized with 
the timing of data submission through the web interface and EHR 
options. 

The agency also provided technical comments, which we incorporated as 
appropriate. 

We are sending copies of this report to the appropriate congressional 
committees and the Administrator of CMS. The report also is available 
at no charge on GAO's website at [hyperlink, http://www.gao.gov]. If 
you or your staffs have any questions regarding this report, please 
contact me at (202) 512-7114 or cosgrovej@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. GAO staff who made major 
contributions to this report are listed in appendix IV. 

Signed by: 

James Cosgrove: 
Director, Health Care: 

[End of section] 

Appendix I: Private Entity and Medicare Performance Feedback for 
Hospitals: 

This appendix contains information on the similarities and differences 
between private entities' and Medicare's performance reporting to 
hospitals. The private entities in our study provided feedback through 
a variety of value-based payment (VBP) initiatives and several 
entities have made accountable care organizations the focus of their 
feedback programs. Payers' efforts to provide feedback to hospitals on 
their performance are centered on rewarding higher-quality and lower-
cost providers of care. 

We followed the same methodology for comparing how private entities 
and the Centers for Medicare & Medicaid Services (CMS) conduct 
performance feedback reporting for hospitals as we did for examining 
physician-focused feedback programs. We interviewed representatives of 
the nine selected private entities about their feedback reporting to 
hospitals, if any, with regard to report recipients, data sources 
used, types of performance measures and benchmarks, frequency of 
reporting, and efforts to enhance the utility of performance reports. 
One statewide health care collaborative in our review was established 
through a partnership between the state medical society and hospital 
association, and only provides feedback reports to hospitals. We 
similarly requested sample feedback reports for hospitals. We 
interviewed CMS officials and obtained CMS documentation on its 
hospital feedback reporting activities, and compared these to private 
entity efforts. We also reviewed a sample CMS hospital feedback report 
from July 2013. 

CMS's hospital VBP efforts over the past decade have evolved to 
provide performance feedback to a range of hospital types, with a 
focus on acute care hospitals. In 2003 the agency began with a quality 
incentive demonstration program designed to see whether financial 
incentives to hospitals were effective at improving the quality of 
inpatient care, and to publicly report that information. Since then, a 
number of laws have required CMS to conduct both feedback reporting 
and VBP programs for hospitals. These included the following: 

* The Medicare Prescription Drug, Improvement, and Modernization Act 
of 2003, which required the establishment of the Hospital Inpatient 
Quality Reporting Program, a pay-for-reporting initiative.[Footnote 
48] The act also required CMS to make downward payment adjustments to 
hospitals that did not successfully report certain quality measures. 
That downward payment adjustment percentage was increased by the 
Deficit Reduction Act of 2005.[Footnote 49] 

* The Patient Protection and Affordable Care Act established 
Medicare's Hospital VBP Program for inpatient care provided in acute 
care hospitals.[Footnote 50] Under this program, CMS withholds a 
percentage of all eligible hospitals' payments and distributes those 
funds to high-performing hospitals. 

In reviewing current feedback reporting practices, we found that 
private entities and CMS report to hospitals on similar performance 
measures and that entities' feedback generally contains publicly 
available data. Table 1 compares features of the hospital feedback 
produced by those private entities in our study that report to 
hospitals through a VBP initiative and CMS's hospital VBP program. 

Table 1: Key Features of Hospital Feedback Reporting by Selected 
Private Entities and CMS: 

Report recipients: 

Private entities in our study reported to a range of hospital types: 
* Private entities reported to acute care general hospitals, including 
critical access hospitals (CAH)[A]; 
* One entity limited the data reported to small hospitals, such as 
CAHs, because of their small volume and limited services, and another 
entity used 5 years of combined data in its annual report to increase 
data reliability for CAHs; 
CMS reported to hospitals covered under statute: 
* CMS provided reports to more than 3,000 acute care general hospitals 
nationwide; 
* CMS did not report to hospitals excluded from its statutory mandate, 
including those excepted from CMS's Hospital Inpatient Quality 
Reporting (IQR) Program,[B] as well as those that do not meet minimum 
volume requirements.[C] 

Data sources: 

Private entities compiled information from Medicare, states, and other 
sources: 
* Private entities generally used publicly available Medicare, state 
hospital association, and other data to report on performance; 
* Medicare data sources included CMS's Hospital Compare website,[D] as 
well as CMS's patient satisfaction surveys; 
* Some entities used other national data sources such as a quality 
indicator tool set from the Agency for Healthcare Research and Quality 
or national registries; 
* One entity used hospital-reported all-payer data, and another entity 
used voluntary hospital-reported data on infections; 
CMS used claims, patient survey, medical record, and other 
administrative data: 
* CMS used data collected through its Hospital IQR Program, which 
included claims and administrative data, its patient satisfaction 
surveys, and other data. 

Measures: 

Private entities reported similar quality measures on processes of 
care, outcomes, patient safety, and patient satisfaction: 
* Private entities generally used similar measures on quality and 
utilization for hospital reporting, including patient safety and 
patient satisfaction; 
* Entities used as few as 15 and as many as 49 total measures in the 
four sample reports we obtained; 
* No entities reported to hospitals on cost measures; 
* One entity stated it focused on readmissions and emergency 
department care because reductions in these areas gave the entity a 
greater ability to control costs; 
CMS reported quality measures on clinical processes of care and 
patient satisfaction: 
* CMS will use three weighted quality domains in fiscal year 2014, 
covering clinical process (45% weighted), patient satisfaction (30% 
weighted), and outcomes; (25% weighted) measures; 
* CMS included 13 clinical process measures, 8 patient satisfaction 
measures, and 3 outcomes measures; (acute myocardial infarction, heart 
failure, and pneumonia 30-day mortality rates); 
* CMS removed measures through annual rulemaking that outlines 
specific criteria, such as measures that have "topped-out"--where 
there is not a significant difference among all hospitals. 

Benchmarks: 

Private entities generally compared hospital indicators to network, 
state, and national averages: 
* Private entities generally compared hospital performance against 
absolute or relative benchmarks at the state and national levels; 
* One entity compared hospitals against network, regional, and 
national levels, as well as against previous measurement period 
performance; 
CMS compared hospitals to achievement against all hospitals' rates and 
to improvement over time: 
* CMS used both an achievement benchmark (a hospital's performance 
against all other hospitals) and an improvement benchmark (a 
hospital's performance from the baseline period to the performance 
period); 
* CMS rewarded the higher of the two benchmark scores and aggregated 
these scores for each quality domain. 

Timeliness and frequency: 

Private entities generally reported to hospitals on an annual basis, 
with some interim reporting: 
* Private entities generally reported to hospitals annually based on a 
9-month or 12-month performance period; 
* Some private entities provided hospitals with interim feedback 
reports; 
CMS generally reported annual data on a quarterly basis: 
* CMS provided annual reports to hospitals; 
* CMS's reporting period varied by measure to ensure sufficient 
reliability; 
* CMS used a performance period of 9 months for value-based payment 
(VBP) in fiscal year 2013 and plans to use a 12-month performance 
period beginning in 2014. 

Enhancing utility: 

Private entities offered resources to assist hospitals in acting on 
performance data: 
* Some private entities offered consultations with quality improvement 
professionals, forums for information-sharing, and mentoring 
opportunities with hospital peers; 
* One entity hosted an annual best practices forum in 2012 where more 
than 400 hospitals shared their quality improvement experiences; 
CMS has offered educational resources on its website and through 
national provider calls: 
* CMS posted information on its hospital reporting and VBP programs on 
its website; 
* CMS held national provider calls to solicit hospitals' feedback. 

Source: GAO analysis of information from CMS and private entities. 

Note: We did not independently verify the information we obtained in 
our interviews with private entities. 

[A] The Balanced Budget Act of 1997 established the CAH designation to 
target small rural hospitals with low patient volumes and short 
patient stays. A number of criteria are used to apply the CAH 
designation, including an average annual length of stay of 96 hours or 
less per patient for acute care services. Pub. L. No. 105-33, § 
4201(c), 111 Stat. 251, 373-374 (1997). 

[B] The Medicare Prescription Drug, Improvement, and Modernization Act 
of 2003 required the establishment of the Hospital IQR Program, a pay-
for-reporting initiative. Pub. L. No. 108-173, § 501(b), 117 Stat. 
2066, 2289 (2003). 

[C] Certain hospitals are excluded from participating in CMS's 
hospital VBP program. These include hospitals that pose immediate 
jeopardy to the health or safety of patients; hospitals that do not 
meet the minimum requirements for cases, measures, or surveys during 
the measurement period; and hospitals that are exempted by the 
Secretary of Health and Human Services (for example, Maryland was 
exempt in fiscal year 2013 by submitting a report to the Secretary 
describing how its similar state program achieves or surpasses CMS's 
hospital VBP program). 

[D] Medicare's Hospital Compare website was started in 2005 through 
its Hospital Quality Initiative in collaboration with other 
stakeholders, such as the American Hospital Association and the 
National Quality Forum. On Hospital Compare, consumers can review 
quality-of-care data for over 4,000 hospitals nationwide. 

[End of table] 

[End of section] 

Appendix II: Quality Measures Used in Sample Physician Feedback 
Reports Provided by Selected Private Entities: 

Table 2 summarizes the number of quality measures included in sample 
physician feedback reports we received from private entities in our 
study. These entities used their discretion to determine which 
measures to include in their reports. We analyzed the measures focused 
on quality of care and categorized them into common areas. 

Table 2: Number of Quality Measures Categorized by Type Found in 
Sample Physician Feedback Reports Provided by Selected Private 
Entities: 

Clinical areas reported: 

Cardiovascular; 
Entity A: 1; 
Entity B: 2; 
Entity C: 7; 
Entity D: 1; 
Entity E: 3; 
Entity F: 1; 
Entity G: 5; 
Entity H: 2. 

Diabetes; 
Entity A: 4; 
Entity B: 3; 
Entity C: 8; 
Entity D: 4; 
Entity E: 9; 
Entity F: 4; 
Entity G: 3; 
Entity H: 4. 

Medication management; 
Entity A: 0; 
Entity B: 0; 
Entity C: 6; 
Entity D: 0; 
Entity E: 0; 
Entity F: 2; 
Entity G: 5; 
Entity H: 0. 

Mental health; 
Entity A: 2; 
Entity B: 0; 
Entity C: 0; 
Entity D: 0; 
Entity E: 0; 
Entity F: 4; 
Entity G: 0; 
Entity H: 0. 

Musculoskeletal conditions; 
Entity A: 1; 
Entity B: 1; 
Entity C: 1; 
Entity D: 2; 
Entity E: 1; 
Entity F: 3; 
Entity G: 0; 
Entity H: 1. 

Pediatric care; 
Entity A: 4; 
Entity B: 0; 
Entity C: 3; 
Entity D: 5; 
Entity E: 4; 
Entity F: 11; 
Entity G: 10; 
Entity H: 17. 

Prevention and screening; 
Entity A: 3; 
Entity B: 5; 
Entity C: 3; 
Entity D: 3; 
Entity E: 4; 
Entity F: 7; 
Entity G: 3; 
Entity H: 3. 

Pulmonary and respiratory conditions; 
Entity A: 3; 
Entity B: 0; 
Entity C: 1; 
Entity D: 2; 
Entity E: 2; 
Entity F: 5; 
Entity G: 3; 
Entity H: 2. 

Nonclinical areas reported: 

Patient safety; 
Entity A: 0; 
Entity B: 0; 
Entity C: 3; 
Entity D: 3; 
Entity E: 0; 
Entity F: 0; 
Entity G: 0; 
Entity H: 0. 

Patient satisfaction; 
Entity A: 0; 
Entity B: 3; 
Entity C: 0; 
Entity D: 0; 
Entity E: 8; 
Entity F: 8; 
Entity G: 0; 
Entity H: 0. 

Other[A]; 
Entity A: 0; 
Entity B: 0; 
Entity C: 0; 
Entity D: 4; 
Entity E: 20; 
Entity F: 0; 
Entity G: 7; 
Entity H: 0. 

Total number of quality measures; 
Entity A: 18; 
Entity B: 14; 
Entity C: 32; 
Entity D: 21; 
Entity E: 51; 
Entity F: 37; 
Entity G: 36; 
Entity H: 29. 

Source: GAO analysis of private entity information: 

Note: For this analysis, we examined the measures included in sample 
physician performance feedback reports obtained from private entities. 
We categorized measures into common areas based on predominant 
patterns. One of the nine entities included in our review only 
provided feedback reports to hospitals, and is not shown here. 

[A] This area included measures related to information technology, 
access to care, participation in quality improvement initiatives or 
external recognition programs, and assessment and care plans for 
urinary incontinence. 

[End of table] 

[End of section] 

Appendix III: Comments from the Department of Health and Human 
Services: 

Department of Health & Human Services: 
Office of The Secretary: 
Assistant Secretary for Legislation: 
Washington, DC 20201: 

March 13, 2014: 

James Cosgrove: 
Director, Health Care: 
U.S. Government Accountability Office: 
441 G Street NW: 
Washington, DC 20548: 

Dear Mr. Cosgrove: 

Attached are comments on the U.S. Government Accountability Office's 
(GAO) draft report entitled, "Medicare: Certain Physician Feedback 
Reporting Practices of Private Entities Could Improve CMS's Efforts" 
(GAO-14-279). 

The Department appreciates the opportunity to review this report prior 
to publication. 

Sincerely, 

Signed by: 

Jim A. Esquea: 
Assistant Secretary for Legislation: 

Attachment: 

General Comments Of The Department Of Health And Human Services (HHS) 
On The Government Accountability Office's (GAO) Draft Report: 
"Medicare: Certain Physician Feedback Reporting Practices Of Private 
Entities Could Improve CMS's Efforts" (GAO-14-279): 

The Department of Health and Human Services (HHS) appreciates the 
opportunity to review and respond to this GAO draft report. In the 
report, GAO examined (1) how and when private health entities report 
performance data to providers, and what information is provided; and 
(2) how the timing and approach CMS uses to report performance data to 
providers compare to that of private health care entities. 

Section 1848(n) of the Social Security Act (the Act) requires CMS to 
provide confidential reports to physicians and, as appropriate, to 
groups of physicians, that measure the resources involved in 
furnishing care to Medicare beneficiaries. Section 1848(n)(1)(A)(iii) 
of the Act also authorizes CMS to include information on the quality 
of care furnished to Medicare beneficiaries by the physician or group. 
CMS has phased in the production and dissemination of these reports. 
On September 16, 2013, CMS made available calendar year (CY) 2012 
reports to 6,779 physician groups nationwide with 25 or more 
physicians and other practitioners. These reports covered 
approximately 400,000 physicians practicing in large medical groups. 
Not only did these reports provide comparative quality of care and 
cost information, but they also previewed how the groups of physicians 
might fare under the value-based payment modifier (VM)
(authorized by section I848(p) of the Act to begin January 1, 2015) 
and how the VM could affect their payments under the Medicare 
Physician Fee Schedule (PFS). 

Additionally, and in response to feedback we received from recipients 
of prior year reports, the 2012 reports contained detailed, 
confidential beneficiary-specific data on each group's attributed 
beneficiaries and their hospitalizations, and the group's associated 
eligible professionals (EP). Complementing the 2012 reports were three 
tables, downloadable only by an authorized and registered group 
representative that provided information on each beneficiary 
attributed to the group and each eligible EP billing under the group's 
Taxpayer Identification Number. 

In a final rule issued in November 2013, CMS described its plans to 
provide feedback reports to all Medicare-enrolled physicians during 
the summer of 2014. We note that these reports will go to hundreds of 
thousands of physicians nationwide — a population that is 
substantially larger than the population to which any one private 
entity will provide reports. 

The Medicare physician feedback program is multi-faceted and broader 
than these private entity programs in terms of scope, magnitude, and 
participation. While there are lessons to be learned from these 
private entities, it is important to note that the subject GAO report 
indicates these entities "almost exclusively focused their feedback 
efforts on primary care physician groups participating in medical 
homes and accountable care organizations." In comparison, Medicare's 
physician feedback program covers all physicians paid under the 
Medicare Physician Fee Schedule and provides feedback reports to 
groups nationwide regardless of specialty or size. 

The GAO recommendations and HHS responses are discussed below. 

Recommendation: 

The GAO recommends that CMS develop performance benchmarks that 
compare physicians' performance against additional benchmarks such as 
state or regional averages. 

HHS Response: 

HHS conditionally concurs with this recommendation pending additional 
information from GAO regarding the types of information being provided 
by private entities and an assessment of whether and to what degree 
that information would be useful to providers serving Medicare 
beneficiaries. To the extent that such additional benchmark 
information could be helpful in this regard, CMS will work with its 
quality improvement organizations to utilize such benchmarks in the 
technical assistance they provide to practitioners, particularly those 
subject to the VM, at the local level. The VM is a national program 
that will affect payment of all physicians under the Medicare PFS. 
Accordingly, we believe that it is appropriate to measure performance 
for all physicians and other EPs that are paid under the Medicare PFS 
and subject to the VM using the same national benchmarks. The key 
purpose of the physician feedback reports is to support the VM and, 
therefore, the reports should focus on the information practitioners 
need in order to understand how and why the VM payment adjustment is 
applied to their payments under the PFS. We believe that including 
additional benchmark information in those reports could be misleading 
and confusing for the purposes of the VM and could also be 
misconstrued to imply that poor performance against national 
benchmarks is acceptable so long as performance is comparable to a 
more local benchmark. 

We also believe that all Medicare FFS beneficiaries are entitled to 
high-quality care, as measured against national benchmarks, regardless 
of where they reside or receive care. 

Recommendation: 

The GAO recommends that CMS disseminate performance reports more 
frequently than the current annual distribution — for example, semi-
annually. 

HHS Response: 

HHS concurs with the recommendation and is examining how to do so in 
future years. We believe that before we begin to provide feedback data 
more frequently, the first step is to complete the phase-in of the 
feedback reporting program to ensure the reports go to all Medicare-
enrolled physicians as required by the statute. As described above, we 
anticipate that the CY 2013 reports (available summer of 2014) will 
accomplish this goal. 

We note that producing the reports more frequently than on an annual 
basis would require CMS to make changes in Physician Quality Reporting 
System (PQRS) data collection, including but not limited to requiring 
physicians to submit data more frequently than annually. Currently, 
physicians reporting under the PQRS through the registry, electronic 
health record (EHR) and web interface reporting methods are required 
to submit data to CMS only once a year. 

It is worth noting, however, that we do require registries that 
participate in PQRS to provide reports more frequently than annually 
to the EPs on whose behalf they report. Specifically, two reports per 
year are required for traditional registry reporting and four per year 
for the new Qualified Clinical Data Registry option implemented for 
2014 PQRS. Therefore, any increase in the frequency of the feedback 
reports may have less impact for physicians reporting through the 
registry methods. Increasing the frequency of feedback reports for 
physicians and other EPs that report using the EHR and web interface 
methods under the PQRS may require a more significant change to 
current PQRS reporting practices. A necessary first step would be to 
increase the frequency of PQRS data collection from EPs and data 
intermediaries such as registries. Once this change is successfully 
implemented, CMS can explore more frequent dissemination of quality 
performance feedback in reports. 

[End of section] 

Appendix IV: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

James Cosgrove, (202) 512-7114 or cosgrovej@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, individuals making key 
contributions to this report include Rosamond Katz, Assistant 
Director; Sandra George; Katherine Perry; and E. Jane Whipple. 

[End of section] 

Related GAO Products: 

Electronic Health Record Programs: Participation Has Increased, but 
Action Needed to Achieve Goals, Including Improved Quality of Care. 
[hyperlink, http://www.gao.gov/products/GAO-14-207]. Washington, D.C.: 
March 6, 2014. 

Clinical Data Registries: HHS Could Improve Medicare Quality and 
Efficiency through Key Requirements and Oversight. [hyperlink, 
http://www.gao.gov/products/GAO-14-75]. Washington, D.C.: December 16, 
2013. 

Medicare Physician Payment: Private-Sector Initiatives Can Help Inform 
CMS Quality and Efficiency Incentive Efforts. [hyperlink, 
http://www.gao.gov/products/GAO-13-160]. Washington, D.C.: December 
26, 2012. 

Medicare Program Integrity: Greater Prepayment Control Efforts Could 
Increase Savings and Better Ensure Proper Payment. [hyperlink, 
http://www.gao.gov/products/GAO-13-102]. Washington, D.C.: November 
13, 2012. 

Medicare Physician Feedback Program: CMS Faces Challenges with 
Methodology and Distribution of Physician Reports. [hyperlink, 
http://www.gao.gov/products/GAO-11-720]. Washington, D.C.: August 12, 
2011. 

Value in Health Care: Key Information for Policymakers to Assess 
Efforts to Improve Quality While Reducing Costs. [hyperlink, 
http://www.gao.gov/products/GAO-11-445]. Washington, D.C.: July 26, 
2011. 

Medicare: Per Capita Method Can Be Used to Profile Physicians and 
Provide Feedback on Resource Use. [hyperlink, 
http://www.gao.gov/products/GAO-09-802]. Washington, D.C.: September 
25, 2009. 

Medicare: Focus on Physician Practice Patterns Can Lead to Greater 
Program Efficiency. [hyperlink, 
http://www.gao.gov/products/GAO-07-307]. Washington, D.C.: April 30, 
2007. 

[End of section] 

Footnotes: 

[1] In an ACO, groups of physicians and hospitals share responsibility 
for providing care to a designated group of patients, and financial 
risk. Payers generally require providers to meet predetermined 
performance targets on quality and cost in order to share in savings 
or receive other incentives. Medical homes allow physicians to earn 
incentives for reducing utilization of high-cost settings, such as 
emergency departments and hospitals, through care coordination 
activities. Medical homes unite physicians of various specialties, 
especially targeting patients with chronic illnesses. 

[2] Private feedback reports are generally designed to identify 
differences between providers' current practices and desired 
performance and may be combined with financial incentives to encourage 
improvement. Other performance reporting makes provider information 
available to the public through recognition programs or websites, thus 
using professional reputation to promote high-quality care. In this 
report, performance feedback reports refer to the private reports sent 
to providers from a payer or other entity. 

[3] How well this approach will work is still unclear. Experts have 
noted that the evidence of VBP's effectiveness is limited. Proponents 
of VBP contend that, with sufficient financial incentive, VBP can 
improve overall quality of care and promote efficient practice 
patterns. However, critics assert that fundamental issues, such as 
developing appropriate performance measures for all provider types, 
have not yet been adequately addressed. 

[4] Pub. L. No. 112-240, § 609(b)(5), 126 Stat. 2313, 2350 (Jan. 2, 
2013). 

[5] America's Health Insurance Plans is an industry trade association 
representing private entities that provide health insurance coverage 
through employers, the individual market, and public programs. 

[6] Blue Cross and Blue Shield Association coordinates legislative, 
regulatory, and political strategy on behalf of a nationwide group of 
37 independent, locally operated insurance companies. 

[7] Network for Regional Healthcare Improvement represents health care 
collaboratives at the city, state, and regional level that provide 
programs for over 110 million Americans. It brings together multiple 
stakeholders--physicians, hospitals, health insurers, and employers--
to address various health care topics, including performance 
measurement and payment and delivery system reform. 

[8] Nearly all of our selected entities reported to physicians on 
their performance. One private entity only reported to hospitals on 
performance, and is included in appendix I. 

[9] We reported on physician payment incentives in 2012. See GAO, 
Medicare Physician Payment: Private-Sector Initiatives Can Help Inform 
CMS Quality and Efficiency Incentive Efforts, [hyperlink, 
http://www.gao.gov/products/GAO-13-160] (Washington, D.C.: Dec. 26, 
2012). 

[10] CMS also includes other types of providers, such as long-term 
care hospitals, under separate VBP programs. These programs are not 
included in our scope because private entities typically do not have 
reporting programs for such provider types. 

[11] Pub. L. No. 109-432, § 101(b), 120 Stat. 2922, 2975 (Dec. 20, 
2006). 

[12] Through 2014, CMS plans to provide an upward payment adjustment 
to physicians who satisfactorily report quality data. In 2015, groups 
that do not satisfactorily report data are to be subject to a downward 
percentage payment adjustment. 

[13] Starting in 2011 and continuing through 2015, eligible Medicare 
physicians who demonstrate meaningful use of certified EHR technology 
consistent with CMS requirements are eligible to receive Medicare EHR 
incentive payments. About 36 percent of EPs received incentive 
payments in 2012. See GAO, Electronic Health Record Programs: 
Participation Has Increased, but Action Needed to Achieve Goals, 
Including Improved Quality of Care, [hyperlink, 
http://www.gao.gov/products/GAO-14-207] (Washington, D.C.: Mar. 6, 
2014). 

[14] Registries qualified for reporting to CMS differ from clinical 
data registries--entities that collect and analyze detailed 
information on the therapies that patients receive and changes in 
their clinical condition over time--in that their sole function is to 
compile and submit data to PQRS. See GAO, Clinical Data Registries: 
HHS Could Improve Medicare Quality and Efficiency through Key 
Requirements and Oversight, [hyperlink, 
http://www.gao.gov/products/GAO-14-75] (Washington, D.C.: Dec. 16, 
2013). 

[15] Pub. L. No. 110-275, § 131(c), 122 Stat. 2494, 2526 (July 15, 
2008). 

[16] Pub. L. No. 111-148, §§ 3003(a), 3007, 124 Stat. 119, 367, 373 
(Mar. 23, 2010). 

[17] [hyperlink, http://www.gao.gov/products/GAO-13-160]. 

[18] Some entities in our study had previous experience in giving 
feedback to providers in these delivery models, while other entities 
were in the process of transitioning their incentive programs from pay 
for performance programs to these newer delivery models. Most entities 
told us they disseminated performance reports to a designated point of 
contact for subsequent distribution to the group's physicians. Many 
entities said they tracked report receipt or website downloads, and 
they stated that report use is high, especially when tied to incentive 
payments. 

[19] Practices with small numbers of attributed enrollees can receive 
skewed results that do not appropriately reflect the practice's 
average performance over time. One entity told us that even one or two 
catastrophic cases in small patient populations could 
disproportionately affect the results and could be misleading. 

[20] Entities told us they generally used nationally endorsed quality 
measures, but have considered adding measures to meet the particular 
needs of their program and physicians. Nationally endorsed measures 
could include measures from the National Quality Forum or the National 
Committee for Quality Assurance's Healthcare Effectiveness Data and 
Information Set. For example, at one entity, a multistakeholder 
Measurement and Reporting Committee supplemented the National Quality 
Forum-endorsed measures with nonendorsed generic drug measures. 

[21] Process measures include clinical activities such as tests, 
screenings, and immunizations. In contrast, outcomes measures require 
data about patient health, such as the percentage of a physician's 
patients at target blood pressure. 

[22] The hemoglobin A1C test is a blood test used to manage type 1 or 
type 2 diabetes by looking at the patient's average blood sugar level 
for a period of time. A high hemoglobin A1C level indicates poor 
control of diabetes. 

[23] In a study of 23 health insurers accounting for 66 percent of 
U.S. commercial enrollment, America's Health Insurance Plans found 
wide variation in measures under common measurement areas. See A. 
Higgins, G. Veselovskiy, and L. McKown, "Provider Performance Measures 
in Private and Public Programs: Achieving Meaningful Alignment With 
Flexibility to Innovate," Health Affairs, vol. 32, no. 8 (2013). 

[24] According to one entity, physicians complained that receiving 
feedback on as many as 29 performance measures did not support 
meaningful patient care. 

[25] Some services a patient receives are not paid for by the entity 
(such as care furnished at a Department of Veterans Affairs facility) 
and therefore the entity's claims data may give an incomplete picture 
of a patient's health. 

[26] Since claims are designed for reimbursement purposes, the data 
are likely to encompass all services and supplies furnished by a 
provider. However, because payment in an FFS setting is not directly 
related to the number or type of conditions for which the physician 
codes, an EHR may have a more complete set of patient diagnoses than 
claims data. In addition, EHRs contain clinical data--lab results or 
vital signs--that can be used to identify conditions that may not have 
been recognized or coded by the physician. 

[27] Data on costs per enrollee are risk-adjusted to reflect the 
enrollee's demographics and health status. This adjustment helps 
ensure that providers are not disadvantaged in the comparisons for 
serving patients with poor health status. 

[28] To calculate the percentage of a defined patient population that 
receives a particular process of care or achieves a particular 
outcome, entities define a denominator population by demographic 
information or diagnosis. For example, to calculate the rate of 
dilated macular examinations, the denominator could be specified as 
all patients aged 50 years and older with a diagnosis of age-related 
macular degeneration. 

[29] The quality assurance process may entail several steps in 
reviewing data for errors and accuracy, including data submission 
audits and internal quality checks. Some entities said that they 
included a physician group review period that allows groups to appeal 
their results using supporting data. 

[30] In response to provider needs, entities were flexible in how they 
distributed reports. Entities told us they primarily distributed 
reports over secure email or through a web portal. One entity sent 
copies by mail to reach small, rural providers. 

[31] See [hyperlink, http://www.gao.gov/products/GAO-13-160]. 

[32] Although specific measures vary by reporting method, CMS places 
all quality measures within an agencywide national quality strategy. 
In 2011, HHS issued a National Quality Strategy establishing six broad 
priority domains: patient and family engagement, patient safety, care 
coordination, population/public health, efficient use of health care 
resources, and clinical process/effectiveness. Experts have 
recommended that CMS align measures among its various Medicare payment 
incentive programs, such as PQRS and the EHR Incentive Programs, and 
CMS has begun to do so. See C. Damberg, Physician Payment Reform: 
Designing a Performance-Based Incentive Program. Testimony before the 
House Energy and Commerce Committee, Subcommittee on Health, on June 
5, 2013. 

[33] CMS's disease management measures relate to coronary artery 
disease, diabetes, heart failure, hypertension, and ischemic vascular 
disease; and the preventive care measures include screenings for 
breast cancer, colorectal cancer, body mass index, tobacco use, high 
blood pressure, and clinical depression; and influenza and 
pneumococcal vaccinations. 

[34] If the pool of eligible assigned beneficiaries is less than 218, 
then providers must report on all assigned beneficiaries. 

[35] Depending on the specialty focus, qualified registries may 
collect and submit data from over 200 CMS-determined measures. 

[36] Total per capita costs include services covered under Medicare 
Part A and services covered under Medicare Part B furnished by all 
providers seen by the beneficiary. 

[37] For cost measures, the peer group for medical groups of 100 or 
more EPs is all groups of 100 or more EPs, while the peer group for 
groups of 25 to 99 EPs is all groups of 25 or more EPs. CMS compares 
cost measures after risk-adjustment and price standardization. 

[38] See [hyperlink, http://www.gao.gov/products/GAO-13-160]. 

[39] The 2015 VM, which will impact FFS payment in 2015 for groups 
with 100 or more EPs is to be based on the groups' performance in 2013. 

[40] CMS determines whether a group's performance was significantly 
different from average performance, with a 95 percent confidence 
interval or one standard deviation above or below the mean. 

[41] Although the feedback report is based on claims from all EPs in 
the medical group, only payments for physician services are affected 
by the VM. 

[42] See [hyperlink, http://www.gao.gov/products/GAO-13-160]. 

[43] Private entities providing reports in a collaborative care 
setting such as a medical home do not have to complete this step, as 
each beneficiary has already been associated with a provider. 

[44] Patient characteristics include age, gender, Medicaid 
eligibility, history of medical conditions, and end-stage renal 
disease. 

[45] In addition to evidence from our selected entities, recent 
research on private feedback reporting found that timeliness of data 
is critical for management of patient care, as well as for tracking 
performance and monitoring progress toward improvement goals. For 
example, based on a review of literature on private reports, an 
examination of selected examples of community quality collaborative 
feedback reports, and a case study of Cincinnati's experience with 
private reporting, one study recommended that private feedback reports 
should be updated at least quarterly. See Dale Shaller and David 
Kanouse, Private "Performance Feedback" Reporting for Physicians: 
Guidance for Community Quality Collaboratives, Publication No. 13-0004 
(Rockville, MD: Agency for Healthcare Research and Quality, Nov. 2012). 

[46] Providers can access quarterly reports from CMS based on claims 
data. However, these interim reports only show the number of measures 
(or measures groups) reported, and the number of those accurately 
reported. They do not show performance rates. 

[47] CMS runs the Medicare Learning Network as an outreach program to 
educate health care professionals on CMS programs, using e-mail, 
calls, web-based courses, and other methods. 

[48] Pub. L. No. 108-173, § 501(b), 117 Stat. 2066, 2289 (Dec. 8, 
2003). 

[49] Pub. L. No. 109-171, § 5001(a), 120 Stat. 4, 28 (Feb. 8, 2006). 

[50] Pub. L. No. 111-148, § 3001(a)(1), 124 Stat. 119, 353 (Mar. 23, 
2010). Other hospital types, such as long-term care hospitals, are 
included under separate VBP programs. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation, and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the 
performance and accountability of the federal government for the 
American people. GAO examines the use of public funds; evaluates 
federal programs and policies; and provides analyses, recommendations, 
and other assistance to help Congress make informed oversight, policy, 
and funding decisions. GAO's commitment to good government is 
reflected in its core values of accountability, integrity, and 
reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select 
"E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO's actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO's 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, DC 20548. 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, DC 20548. 

[End of document]