This is the accessible text file for GAO report number GAO-09-104 
entitled 'Medicare: Callers Can Access 1-800-MEDICARE Services, but 
Responsibility within CMS for Limited English Proficiency Plan Unclear' 
which was released on January 28, 2009.

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. 

Report to the Chairman, Subcommittee on Health, Committee on Ways and 
Means, House of Representatives: 

United States Government Accountability Office: 
GAO: 

December 2008: 

Medicare: 

Callers Can Access 1-800-MEDICARE Services, but Responsibility within 
CMS for Limited English Proficiency Plan Unclear: 

GAO-09-104: 

GAO Highlights: 

Highlights of GAO-09-104, a report to the Chairman, Subcommittee on 
Health, Committee on Ways and Means, House of Representatives. 

Why GAO Did This Study: 

The Centers for Medicare & Medicaid Services (CMS) is responsible for 
providing beneficiaries timely and accurate information about Medicare. 
Receiving nearly 30 million calls in 2007, 1-800-MEDICARE, operated by 
a contractor, is the most common way members of the public get program 
information. The help line provides services both to English-speaking 
and limited English proficiency (LEP) callers. In this report, GAO 
describes (1) the extent to which access performance standards and 
targets have been met by the current contractor, (2) the efforts by CMS 
to provide LEP callers access to help line services and wait times 
experienced by these callers, and (3) CMS’s oversight of callers’ 
access to 1-800-MEDICARE and the information’s accuracy. To conduct 
this work, GAO reviewed documents and analyzed help line data through 
July 2008. In addition, GAO interviewed agency staff, industry experts, 
and officials at four federal agencies with high call volume contact 
centers. 

What GAO Found: 

The 1-800-MEDICARE contractor met most standards and some targets for 
the required telephone performance metrics and indicators CMS designed 
to ensure callers’ access—from July 2007 through July 2008. The 1-800-
MEDICARE contractor’s performance met the standard for each of the 
three access-related metrics—the average amount of time callers wait to 
reach customer service representatives (CSR), the percent of unhandled 
calls, such as abandoned calls, and the percent of calls transferred 
among CSRs—in 10 of 13 months analyzed. Because of waivers granted by 
CMS, the contractor was considered by the agency to have met the 
standards in 12 of 13 months. During that time, the contractor met the 
target for only one of three access-related indicators—the percent of 
CSRs answering calls. Other indicators were the average amount of time 
needed to respond to callers’ inquiries and the accuracy of CSR call 
volume forecasting. 

CMS’s efforts to provide LEP callers with access have led to shorter 
average wait times for Spanish-speaking callers, but are not consistent 
with all elements of the HHS LEP Plan. CMS requires its help line 
contractor to provide services to Spanish-speaking callers by employing 
bilingual CSRs and to provide interpretation services for other LEP 
callers, which the contractor does by using telephone interpreters. In 
20 of the 32 months reviewed, Spanish-speaking callers waited less 
time, on average, to reach a CSR than English-speaking callers. CMS 
officials with primary responsibility for 1-800-MEDICARE said they were 
not aware of the LEP Plan when awarding the current contract, and CMS 
has not identified an office responsible for acting as a point of 
contact for management of the LEP Plan. Without a responsible office or 
official, an internal control for federal agencies, CMS staff lack a 
source of guidance to assist them in taking steps consistent with the 
LEP Plan when considering the needs of people with LEP. However, CMS 
has taken steps consistent with some elements of the agency’s adopted 
LEP Plan, such as the element related to oral language assistance, but 
not others, such as the element identifying the need for complaint 
mechanisms for language issues. 

To oversee 1-800-MEDICARE callers’ access to services and accurate 
information, CMS uses all six commonly used contact center management 
practices. Based on GAO’s review of the literature and interviews with 
federal agencies and industry experts, these management practices are: 
(1) clearly defining performance metrics, (2) performing accurate 
capacity planning, (3) conducting customer satisfaction surveys, (4) 
ensuring information for CSRs to reference is accurate, (5) evaluating 
CSRs’ interaction with callers, and (6) validating contact center 
performance reports. These practices are addressed in the current 1-800-
MEDICARE contract and reflected in CMS’s ongoing contract oversight. 

What GAO Recommends: 

To ensure CMS offices, including those that oversee the operation of 1-
800-MEDICARE, are aware of, and take steps consistent with, the 
Department of Health and Human Services (HHS) LEP Plan when considering 
the needs of people with LEP, CMS should designate an official or 
office with responsibility for managing the LEP Plan. In commenting on 
a draft of this report, CMS generally concurred with our 
recommendation. 

To view the full product, including the scope and methodology, click on 
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-09-104]. For more 
information, contact Kathleen King at (202) 512-7114 or kingk@gao.gov. 

[End of section] 

Contents: 

Letter: 

Results in Brief: 

Background: 

The 1-800-MEDICARE Contractor Met Most Standards and Some Targets 
Designed to Ensure Caller Access, but Average Wait Times under the 
Current Contract Varied by Type and Complexity of Inquiry: 

Efforts to Provide LEP Callers Access to 1-800-MEDICARE Have Led to 
Shorter Wait Times for Spanish-speaking Callers, on Average, but Are 
Not Consistent with All LEP Plan Elements: 

CMS Uses Common Management Practices to Oversee Callers' Access to 
Services and to Accurate Information from 1-800-MEDICARE: 

Conclusions: 

Recommendation for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Elements of the Department of Health and Human Services' 
Limited English Proficiency Plan: 

Appendix II: Average Caller Wait Time by Type and Complexity of Call 
and Language of Caller, September 2007 through July 2008: 

Appendix III: Comments from the Centers for Medicare & Medicaid 
Services: 

Appendix IV: GAO Contact and Staff Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Required Monthly Access-Related Performance Metrics and 
Standards, July 2007 through July 2008: 

Table 2: 1-800-MEDICARE Contractor's Unhandled CSR Call Performance, 
July 2007 through July 2008: 

Table 3: Selected Access-Related Performance Indicators, July 2007 
through July 2008: 

Table 4: Six Common Management Practices Used to Oversee Callers' 
Access to Services and to Accurate Information: 

Figures: 

Figure 1: Total 1-800-MEDICARE Call Volume, July 2005 through July 
2008: 

Figure 2: How Callers Navigate 1-800-MEDICARE's Automated IVR System: 

Figure 3: How Callers Navigate 1-800-MEDICARE's CSR-Level Interaction: 

Figure 4: 1-800-MEDICARE Contractor's Average Wait Time, July 2007 
through July 2008: 

Figure 5: 1-800-MEDICARE Contractor's CSR Transfer Rate, July 2007 
through July 2008: 

Figure 6: Monthly Average Wait Time for Callers with General Medicare 
Inquiries and Callers with Claims Inquiries, September 2007 through 
July 2008: 

Figure 7: Monthly Average Wait Time for Tier 1 and Tier 2 CSRs, 
September 2007 through July 2008: 

Figure 8: Monthly Average Wait Time, December 2005 through July 2008: 

Figure 9: Monthly Average Wait Time for English-and Spanish-speaking 
Callers, December 2005 through July 2008: 

Figure 10: Monthly Average Wait Times for General Medicare and Claims 
Inquiries by Tier 1 CSRs: 

Figure 11: Monthly Average Wait Times for General Medicare and Claims 
Inquiries by Tier 2 CSRs: 

Figure 12: Monthly Average Wait Time for Tier 1 and 2 CSRs Responding 
to Part A Inquiries: 

Figure 13: Monthly Average Wait Time for Tier 1 and Tier 2 CSRs 
Responding to Part B Inquiries: 

Figure 14: Monthly Average Wait Times for Tier 1 and Tier 2 CSRs 
Responding to DME Inquiries: 

Figure 15: Monthly Average Wait Times for Tier 1 CSRs Responding to 
English-and Spanish-speaking Callers' General Medicare Inquiries: 

Figure 16: Monthly Average Wait Times for Tier 2 CSRs Responding to 
English-and Spanish-speaking Callers' General Medicare Inquiries: 

Figure 17: Monthly Average Wait Times for Tier 1 CSRs Responding to 
English-and Spanish-speaking Callers' Claims Inquiries: 

Figure 18: Monthly Average Wait Times for Tier 2 CSRs Responding to 
English-and Spanish-speaking Callers' Claims Inquiries: 

Abbreviations: 

BBA: Balanced Budget Act of 1997: 

CMS: Centers for Medicare & Medicaid Services: 

CSR: customer service representative: 

DME: durable medical equipment: 

FFS: fee-for-service: 

HHS: Department of Health and Human Services: 

IVR: Interactive Voice Response System: 

LEP: limited English proficiency: 

MMA: Medicare Prescription Drug, Improvement, and Modernization Act of 
2003: 

NDW: National Data Warehouse: 

OBIS: Office of Beneficiary Information Services: 

OCR: Office for Civil Rights: 

OEOCR: Office of Equal Opportunity and Civil Rights: 

TQC: Training, Quality, and Content contractor: 

[End of section] 

United States Government Accountability Office:
Washington, DC 20548: 

December 29, 2008: 

The Honorable Pete Stark: 
Chairman: 
Subcommittee on Health Committee on Ways and Means: 
House of Representatives: 

Dear Mr. Chairman: 

The nearly 45 million elderly and disabled Medicare beneficiaries face 
a complex set of coverage choices. For example, they must choose 
whether to participate in the original Medicare fee-for-service program 
or obtain coverage from private health plans that contract with 
Medicare, and whether to obtain Medicare coverage for outpatient 
prescription drugs.[Footnote 1] These choices require them to obtain 
information about the comparative benefits, costs, and quality of 
available options. Additionally, once enrolled and receiving services, 
Medicare beneficiaries may seek additional information about their 
coverage, such as the status of payments for services or whether their 
health plan benefits have changed. 

The Department of Health and Human Services' (HHS) Centers for Medicare 
& Medicaid Services (CMS), the federal agency that administers 
Medicare, is responsible for providing clear, accurate, and timely 
information about this complex program. In March 1999, CMS implemented 
a nationwide toll-free telephone help line--1-800-MEDICARE--that 
beneficiaries, their families, and other members of the public can call 
to ask questions about Medicare. 1-800-MEDICARE, which is operated by a 
contractor, is now the most common way beneficiaries and other 
interested members of the public get information from CMS about 
Medicare, with the help line receiving more than 30 million calls in 
2007.[Footnote 2] The help line operates 24 hours a day, 7 days a week, 
with calls first answered by an automated system and then routed to a 
customer service representative (CSR), if requested, for more specific 
inquiries about Medicare benefits. 

The 1-800-MEDICARE help line provides assistance both to English-and 
non-English-speaking callers--in 2007, approximately 1 million callers 
of the more than 30 million total callers were assisted in a language 
other than English. Among non-English-speaking callers, Spanish was the 
most requested language, representing about 98 percent of all non- 
English-speaking callers to 1-800-MEDICARE in 2007. HHS, like other 
federal agencies, is required by Executive Order 13166 to examine the 
services it provides and develop and implement a system by which people 
with limited English proficiency (LEP) can meaningfully access 
federally conducted programs and activities, such as the 1-800-MEDICARE 
help line. The executive order also requires federal agencies, 
including HHS, to prepare a plan identifying the steps they will take 
to ensure meaningful access by people with LEP to such activities. 
[Footnote 3] 

Since the launch of the help line in 1999, changes to the Medicare 
program have required CMS to provide more information through 1-800- 
MEDICARE. For example, the Medicare Prescription Drug, Improvement, and 
Modernization Act of 2003 (MMA) created a voluntary outpatient 
prescription drug program for Medicare beneficiaries and required 
information on this new program to be made available through 1-800- 
MEDICARE.[Footnote 4] This and other changes to the Medicare program 
have resulted in increases in the number of calls received by 1-800- 
MEDICARE. Additionally, while in the past two contractors operated 1- 
800-MEDICARE, in October 2006 CMS opted to begin using a single 
contractor in an effort to be more efficient, according to CMS 
officials. This single contractor assumed responsibility for 1-800- 
MEDICARE operations as of October 30, 2006.[Footnote 5] CMS monitors 
the contractor's performance using metrics and indicators with 
corresponding performance standards and targets.[Footnote 6] 

Given the complexity of Medicare benefits, it is important that callers 
contacting 1-800-MEDICARE be able to access timely and accurate 
information as they make health care decisions. You asked us to provide 
information on the extent to which 1-800-MEDICARE provides timely 
services to callers, including LEP callers, and how CMS oversees the 
current 1-800-MEDICARE contractor. In this report, we describe (1) the 
extent to which performance standards and targets designed by CMS to 
ensure caller access have been met by the current 1-800-MEDICARE 
contractor and the extent to which the time a caller waits to reach a 
CSR has varied, (2) the efforts by CMS to provide LEP callers access to 
1-800-MEDICARE services and the wait times experienced by these 
callers, and (3) CMS's oversight of callers' access to 1-800-MEDICARE 
and the accuracy of the information provided. 

To describe the extent to which performance standards and targets 
designed by CMS to ensure caller access have been met by the current 1- 
800-MEDICARE contractor and the extent of efforts by CMS to provide LEP 
callers access to 1-800-MEDICARE, we reviewed documents related to 1- 
800-MEDICARE access requirements, including both the prior and current 
1-800-MEDICARE contract and other related contracts.[Footnote 7] We 
also interviewed staff from the CMS office with primary responsibility 
for 1-800-MEDICARE. We used these documents and interviews to identify 
the metrics and indicators and their corresponding standards and 
targets--criteria used to measure the 1-800-MEDICARE contractor's 
performance--that we considered to be most closely associated with 
caller access. In conducting our work, we focused on telephone services 
provided by 1-800-MEDICARE and did not review requirements for 
information provided through other communication channels, such as e- 
mail. Using data from CMS's National Data Warehouse (NDW)[Footnote 8] 
on calls received by 1-800-MEDICARE, we compared the selected 
performance metrics and indicators to the contractor's actual 
performance. We also used these data to examine changes in the amount 
of time callers waited to reach a CSR. To assess the reliability of the 
NDW data used, we reviewed CMS documents describing how the data are 
collected and CMS's efforts to ensure data reliability. We also 
interviewed CMS officials responsible for collecting and analyzing the 
data and conducted a series of tests on the data files CMS provided. We 
determined the data were sufficiently reliable for the purposes of this 
report and reviewed data for three primary periods:[Footnote 9] 

* December 2005 through July 2008. We identified changes during this 
time period in the amount of time callers, including LEP callers, 
waited to reach a CSR.[Footnote 10] This time period encompassed both 
the prior and current 1-800-MEDICARE contracts. 

* July 2007 through July 2008. We analyzed data to compare the metrics 
and indicators implemented in July 2007 of the current 1-800-MEDICARE 
contract with the actual performance of the 1-800-MEDICARE 
contractor.[Footnote 11] 

* September 2007 through July 2008. We identified differences in the 
amount of time callers waited to speak to CSRs under the current 1-800- 
MEDICARE contract, depending on the skill level of staff responding to 
the calls, the language in which assistance was provided, and the type 
of calls--general inquiries about the Medicare program or calls about 
payments for services or benefits, referred to as claims.[Footnote 12] 

To provide additional information on efforts by CMS to provide LEP 
callers access to 1-800-MEDICARE services, we reviewed Executive Order 
13166 and the LEP Plan developed by HHS. We also interviewed staff from 
the HHS Office for Civil Rights and the CMS Office of Equal Opportunity 
and Civil Rights.[Footnote 13] Additionally, we reviewed GAO work 
related to internal control standards when considering implementation 
of the LEP Plan. While we reviewed certain 1-800-MEDICARE activities to 
identify steps consistent with the LEP Plan, we did not review all CMS 
and 1-800-MEDICARE activities for consistency with the LEP Plan. 

To describe how CMS oversees callers' access to, and the accuracy of 
information provided by, 1-800-MEDICARE, we identified six common 
oversight practices used by contact centers to ensure both caller 
access and information accuracy.[Footnote 14] To identify these 
practices, we reviewed prior GAO work and federal publications related 
to contact centers[Footnote 15] and conducted interviews with four 
federal agencies that receive inquiries about their services by 
telephone and with contact center industry experts.[Footnote 16] We 
selected federal agency contact centers to interview based on our 
review of prior related GAO work, selecting those that had annual call 
volumes greater than 5 million calls and other characteristics similar 
to those of 1-800-MEDICARE, such as a comparable population, extended 
hours, periods of particularly high call volume, and service provision 
in multiple languages. Using our review of publications and the 
interviews with our selected federal contact centers, we identified the 
six common oversight practices, which we then confirmed in interviews 
with industry experts and an industry publication.[Footnote 17] To 
determine whether CMS used these common oversight practices, we 
reviewed documents related to CMS's oversight of the 1-800-MEDICARE 
contractor, including meeting minutes, policies and procedures, and 
contractor evaluations, and interviewed CMS officials. We also 
conducted a site visit to one of the five 1-800-MEDICARE contact center 
sites to gain a better understanding of how the centers function. 
[Footnote 18] In addition, we reviewed a contract awarded by CMS in 
December 2007 to a separate contractor--referred to as the Training, 
Quality, and Content (TQC) contractor--responsible for developing CSR 
training materials and for conducting quality assurance efforts for 1-
800-MEDICARE. We interviewed officials both from CMS and the current 1-
800-MEDICARE contractor to determine steps the TQC contractor is taking 
to assume responsibility for these activities. As the TQC contract was 
not fully implemented at the time of our review, we could not evaluate 
its effectiveness. 

Our work did not assess particular issues encountered by callers with 
visual, hearing, or speech impairments or those with low literacy. In 
addition, we did not evaluate the accuracy of information callers 
receive from 1-800-MEDICARE CSRs, the accuracy of information CSRs use 
to answer beneficiary questions, or the appropriateness of CSR 
training. We also did not evaluate the effectiveness of the individual 
management practices CMS uses to oversee callers' access to, and 
accuracy of information provided by, 1-800-MEDICARE and did not analyze 
the award process for, or management of, the contract. 

We conducted this performance audit from December 2007 through November 
2008 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. 

Results in Brief: 

The 1-800-MEDICARE contractor met most standards and some targets for 
telephone access-related performance measures--metrics and indicators 
designed to ensure all callers' access to services--from July 2007 
through July 2008. In 10 of the 13 months analyzed, the 1-800-MEDICARE 
contractor's performance met the standards for each of three required 
access performance metrics--the average amount of time a caller waits 
to reach a CSR; the percent of calls that are unhandled, such as 
abandoned calls; and the percent of calls transferred among CSRs. 
Because of waivers granted by CMS, the contractor was considered by the 
agency to have met the standards in 12 of 13 months. For example, the 
current contractor met the performance metric standard of maintaining a 
monthly average wait time for all callers of between 5 minutes and 8 
minutes, 30 seconds in 12 of 13 months analyzed and, because of a 
waiver granted by CMS, was considered by the agency to have met the 
relevant standard for average wait time in all 13 months. However, the 
contractor met targets for only one of three monthly access-related 
indicators during that same 13-month period--the percent of CSRs 
answering calls. Other indicators were the average amount of time 
needed to respond to callers' inquiries and the accuracy of CSR call 
volume forecasting. The 1-800-MEDICARE contract does not include 
metrics or indicators specifically related to the type of call, but the 
agency expected the contractor to manage wait times to ensure that 
callers with different types of inquiries had similar wait times. 
However, during the current contract period, callers experienced 
differing wait times depending on the type and complexity of the 
inquiry. For example, callers with more complex inquiries generally 
experienced longer wait times to reach a CSR. The amount of time 
callers wait to speak with a CSR has increased since December 2005, but 
the performance standards related to caller wait time also have varied 
over time. 

CMS's efforts to provide LEP callers with access to 1-800-MEDICARE have 
led to shorter average wait times for Spanish-speaking callers, but are 
not consistent with all elements of the LEP Plan. To provide LEP 
callers with access to 1-800-MEDICARE, CMS requires its help line 
contractor to provide services to callers in both English and Spanish 
by employing bilingual CSRs and to provide interpretation services for 
callers speaking other languages, which the contractor does through a 
telephone interpretation service. In two-thirds of the months we 
reviewed--December 2005 through July 2008--Spanish-speaking callers 
waited, on average, less time to reach a CSR than their English- 
speaking counterparts. However, Spanish-speaking callers have recently 
experienced average wait times that were slightly longer than those of 
English-speaking callers. CMS officials responsible for 1-800-MEDICARE 
were not aware of the LEP Plan--developed by HHS for its agencies and 
programs as required by Executive Order 13166 and adopted by CMS to 
improve language access--and CMS was not able to identify an office 
responsible for management of this Plan. Despite this limited 
awareness, steps CMS has taken to provide access to 1-800-MEDICARE for 
LEP callers are consistent with some, but not all, of the LEP Plan 
elements. For example, by requiring its contractor to provide services 
in multiple languages, CMS has taken a step consistent with the Plan's 
element related to oral language assistance. However, CMS has not 
identified a specific office or official with whom LEP callers could 
register complaints regarding language assistance services, despite a 
Plan element addressing this issue. Without a responsible office or 
official--a key internal control for federal agencies--to manage the 
LEP Plan, CMS staff lack a source of guidance that could assist them in 
taking steps consistent with the Plan when considering the needs of 
people with LEP. 

CMS uses all six of the commonly used management practices for contact 
centers we identified to oversee 1-800-MEDICARE callers' access to 
services and to accurate information. These practices are: (1) clearly 
defined performance metrics relating to access, (2) customer 
satisfaction surveys, (3) assurance that information referenced by CSRs 
to answer callers' inquiries is accurate, (4) evaluation of CSRs' 
interaction with callers, (5) capacity planning to meet caller demand 
for services, and (6) validation of contact center performance reports. 
These practices are addressed in the current 1-800-MEDICARE contract 
and reflected in CMS's ongoing oversight of the help line. For example, 
CMS used information collected by customer satisfaction surveys to 
identify callers' dissatisfaction with the automated system and took 
steps to improve the system. In addition, CMS identified periods when 
the 1-800-MEDICARE contractor did not accurately forecast call volume, 
which is used to determine whether systems and staffing will be 
sufficient to meet the caller demand for services, and the agency 
continues to monitor the contractor's performance in this area. 

To ensure CMS offices, including those that oversee the operation of 
the 1-800-MEDICARE help line, are aware of, and take steps consistent 
with, the HHS LEP Plan when considering the needs of people with LEP, 
we recommend that CMS designate an official or office with 
responsibility for managing the LEP Plan. In commenting on a draft of 
this report, CMS generally concurred with our recommendation. 

Background: 

CMS administers 1-800-MEDICARE through a single contractor that 
operates the telephone help line 24 hours a day, 7 days a week. CMS 
initially designed 1-800-MEDICARE to assist beneficiaries in obtaining 
information about Medicare programs, including Medicare's managed care 
program, and has since expanded the telephone line to handle increased 
call volume and additional inquiries, including inquiries about 
Medicare's prescription drug benefit. Calls are answered by an 
automated system and, if requested, callers may be routed to a CSR for 
additional assistance. The contractor trains CSRs to respond to two 
types of inquiries--(1) general inquiries about the Medicare program 
(such as general inquiries about prescription drug coverage or 
beneficiary address changes) and (2) specific inquiries about some 
Medicare Part A and Part B claims--for both English-and Spanish- 
speaking callers. CMS uses performance metrics and indicators to 
oversee and measure the 1-800-MEDICARE contractor's performance. To 
meet the needs of people with LEP, HHS developed an LEP Plan that 
identifies steps for its agencies, including CMS, to take to improve 
access for people with LEP to agency programs and activities, including 
1-800-MEDICARE. 

Contact Center Operations: 

CMS administers 1-800-MEDICARE to answer callers' inquiries 24 hours a 
day, 7 days a week about Medicare eligibility, enrollment, and 
benefits. CMS initially developed 1-800-MEDICARE to assist 
beneficiaries and other members of the public in obtaining information 
about Medicare programs, including Medicare's managed care program, as 
required by the Balanced Budget Act of 1997 (BBA).[Footnote 19] In 
2003, the MMA required CMS to make information about the Medicare 
prescription drug benefit available through 1-800-MEDICARE and gave CMS 
increased flexibility in the administration of Medicare.[Footnote 20] 
CMS officials reported that they used this flexibility to transfer 
responsibility for information about Medicare-related Part A and Part B 
(including DME) claims calls, which previously had been handled by 
separate Part A and Part B contractors with their own help line 
numbers, to 1-800-MEDICARE.[Footnote 21] CMS completed the transition 
of Part A and Part B claims calls to 1-800-MEDICARE in September 2007. 

Each of these initiatives contributed to an increase in the overall 
volume of calls received by 1-800-MEDICARE. During the period 
encompassed in our review, overall call volume for 1-800-MEDICARE 
increased approximately 22 percent, from about 1.6 million calls in 
July 2005 to more than 2.0 million calls in July 2008. (See figure 1.) 
A large increase in volume occurred close to the beginning of the 
Medicare prescription drug benefit in late 2005 and early 2006. Since 
that time, monthly call volume has ranged from about 2 million 
inquiries to slightly more than 3 million. 

Figure 1: Total 1-800-MEDICARE Call Volume, July 2005 through July 
2008: 

[Refer to PDF for image] 

This figure is a line graph depicting the following data: 

Month and year: Jul-05; 
Total call volume: 1,645,885. 

Month and year: Aug-05; 
Total call volume: 1,941,761. 

Month and year: Sep-05; 
Total call volume: 1,877,062. 

Month and year: Oct-05 (peak activity period); 
Total call volume: 3,012,118. 

Month and year: Nov-05 (peak activity period) (Medicare Part D 
enrollment begins); 
Total call volume: 4,519,292. 

Month and year: Dec-05 (peak activity period); 
Total call volume: 4,546,120. 

Month and year: Jan-06 (peak activity period); 
Total call volume: 5,794,455. 

Month and year: Feb-06; 
Total call volume: 3,266,135. 

Month and year: Mar-06; 
Total call volume: 3,426,478. 

Month and year: Apr-06; 
Total call volume: 3,450,312. 

Month and year: May-06; 
Total call volume: 4,730,113. 

Month and year: Jun-06; 
Total call volume: 2,720,194. 

Month and year: Jul-06; 
Total call volume: 2,442,976. 

Month and year: Aug-06; 
Total call volume: 2,416,495. 

Month and year: Sep-06; 
Total call volume: 1,972,670. 

Month and year: Oct-06 (peak activity period) (CMS selects single 
contractor to administer 1-800-MEDICARE); 
Total call volume: 2,375,042. 

Month and year: Nov-06 (peak activity period); 
Total call volume: 2,533,031. 

Month and year: Dec-06 (peak activity period); 
Total call volume: 2,352,702. 

Month and year: Jan-07 (peak activity period); 
Total call volume: 3,046,708. 

Month and year: Feb-07; 
Total call volume: 2,448,259. 

Month and year: Mar-07; 
Total call volume: 2,662,862. 

Month and year: Apr-07; 
Total call volume: 2,400,565. 

Month and year: May-07 (Single contractor assumes all operations for 
1-800-MEDICARE); 
Total call volume: 2,287,219. 

Month and year: Jun-07; 
Total call volume: 2,154,346. 

Month and year: Jul-07; 
Total call volume: 2,200,938. 

Month and year: Aug-07; 
Total call volume: 2,258,247. 

Month and year: Sep-07 (Transfer of Part A/Part B claims calls to 1-800-
MEDICARE completed); 
Total call volume: 2,324,874. 

Month and year: Oct-07 (peak activity period); 
Total call volume: 2,780,090. 

Month and year: Nov-07 (peak activity period); 
Total call volume: 2,808,805. 

Month and year: Dec-07 (peak activity period); 
Total call volume: 2,643,230. 

Month and year: Jan-08 (peak activity period); 
Total call volume: 2,856,120. 

Month and year: Feb-08; 
Total call volume: 2,222,779. 

Month and year: Mar-08; 
Total call volume: 2,254,077. 

Month and year: Apr-08; 
Total call volume: 2,155,090. 

Month and year: May-08; 
Total call volume: 2,015,819. 

Month and year: Jun-08; 
Total call volume: 2,042,624. 

Month and year: Jul-08; 
Total call volume: 2,001,851. 

Source: CMS monthly data. 

[End of figure] 

Callers to 1-800-MEDICARE receive information from an automated 
interactive voice response (IVR) system, a CSR, or a combination of 
both. Calls to the line are initially answered by the IVR, which 
responds to voice and electronic prompts by the caller.[Footnote 22] 
Callers also may use the IVR to obtain assistance in either English or 
Spanish. Those who do not indicate a language preference in the IVR and 
wish to speak to a CSR are automatically routed to a CSR for assistance 
in English. If the IVR cannot address the needs of the caller or if the 
caller requests to speak to a person, the call is routed to a CSR for 
assistance. (See figure 2.) 

Figure 2: How Callers Navigate 1-800-MEDICARE's Automated IVR System: 

[Refer to PDF for image] 

This figure is an illustration of how callers navigate 1-800-MEDICARE's 
automated IVR system, as follows: 

Caller contacts 1-800-MEDICARE: 

IVR-level interaction: 
Caller's inquiry is answered by IVR[A]: 
* Yes: End process with answer; 
* No: Back to IVR for new answer; 
* No: Hang up without answer; 
* No: Request CSR for answer (process continues in figure 3 at CSR-
level interaction). 

Source: GAO analysis of CMS documents (data); Art Explosion (graphics). 

[A] Callers can request assistance in English or Spanish. If callers 
request assistance in Spanish, they are assisted by a Spanish IVR that 
is consistent with the IVR available to English-speaking callers. If a 
caller chooses neither English nor Spanish--for example, a caller who 
speaks neither language--the IVR routes that caller directly to a CSR 
for assistance in English. 

[End of figure] 

CSR Assistance: 

1-800-MEDICARE classifies incoming calls to CSRs into two types of 
requests: 

* Inquiries about general Medicare issues, such as general inquiries 
about prescription drug coverage or beneficiary address changes. 

* Specific inquiries about claims for Medicare Parts A and B. To 
provide assistance with these claims inquiries, 1-800-MEDICARE CSRs are 
able to access certain Part A and Part B claims data.[Footnote 23] 

Newly hired CSRs initially receive training to provide services related 
to general Medicare inquiries and, as they become more proficient, 
receive additional training to respond to claims inquiries. The 1-800- 
MEDICARE contractor hires additional, temporary CSRs to handle 
anticipated increases during 1-800-MEDICARE's annual peak period from 
October through mid-January; these CSRs are trained to handle general 
Medicare inquiries.[Footnote 24] 

The IVR routes callers with general Medicare or specific claims 
inquiries to CSRs according to complexity. The approximately 2,600 
CSRs[Footnote 25] responding to these calls are grouped into two skill 
levels, or "tiers," based on their training and experience. (See figure 
3.) 

* Tier 1 CSRs, who made up the majority of CSRs as of January 1, 2008, 
receive training to respond to simple inquiries about Medicare and 
claims. Tier 1 CSRs handle the majority of 1-800-MEDICARE inquiries. 
Between September 2007 and July 2008, Tier 1 CSRs handled over 15.1 
million calls--95 percent of the approximately 15.9 million calls 
handled by CSRs. Approximately 11.8 million of these Tier 1-handled 
calls were general Medicare inquiries while approximately 3.3 million 
were claims inquiries. 

* Tier 2 CSRs receive training to assist callers with more complex 
general Medicare and claims inquiries. For example, callers wishing to 
enroll in a Medicare Advantage plan would speak to a Tier 2 CSR. These 
CSRs also can respond to calls at the Tier 1 level if needed. Between 
September 2007 and July 2008, Tier 2 CSRs handled about 800,000 calls-
-5 percent of all calls handled by CSRs. Approximately 523,000 of these 
calls were general Medicare inquiries, while approximately 273,000 were 
claims inquiries. 

Particularly complex inquiries, about 1 percent of all calls received, 
may be referred by Tier 1 or Tier 2 CSRs to a reference center staffed 
by CSRs with additional training above the level of the Tier 2 CSRs. 
[Footnote 26] 

Figure 3: How Callers Navigate 1-800-MEDICARE's CSR-Level Interaction: 

[Refer to PDF for image] 

This figure is an illustration of how callers navigate 1-800-MEDICARE's 
CSR-Level Interaction, as follows: 

Process continued from IVR-level interaction in figure 2: 

CSR-Level Interaction: 

Caller’s inquiry is answered by Tier 1 CSR? (Simple questions): 
Yes: End process with answer; 
No: Call transferred to Tier 2 CSR. 

Caller’s inquiry is answered by Tier 2 CSR? (Complex questions): 
Yes: End process with answer; 
No: Call referred to Reference Center[A]. 

Source: GAO analysis of CMS documents (data); Art Explosion (graphics). 

[A] Callers are not transferred to the reference center. Rather, Tier 1 
and Tier 2 CSRs provide information on callers' inquiries to Reference 
Center staff, who will research the issue and contact the caller. 

[End of figure] 

To provide responses to callers, CSRs in Tier 1 and Tier 2 use defined 
"scripts"--standard language explaining elements of the Medicare 
program--that they access using desktop computer software. CSRs listen 
to a caller's inquiry and then enter related keywords into this system 
to generate a list of suggested scripts that could be used to answer 
the inquiry. The CSRs then select the script they consider best suited 
to answer the inquiry and read either excerpts or the entire script to 
the caller.[Footnote 27] CSRs may also consult other information 
sources if appropriate. For example, CSRs sometimes use tools available 
on the Medicare Web site to help beneficiaries select a prescription 
drug plan.[Footnote 28] CMS requires CSRs to read scripts in an effort 
to ensure that all callers receive consistent information from 1-800- 
MEDICARE--regardless of the expertise of the CSR--and that information 
being shared is easily understood by the caller.[Footnote 29] 

1-800-MEDICARE Contract and Related Contracts: 

In October 2006, CMS awarded a single performance-based contract 
[Footnote 30] for the operation of its 1-800-MEDICARE contact centers 
valued at approximately $496 million over the life of the contract. 
This contract consisted of an initial 7-month transition period, during 
which the current 1-800-MEDICARE contractor assumed full responsibility 
for 1-800-MEDICARE help line operations, and 2 additional years, 
referred to as "option years."[Footnote 31] The contractor is 
reimbursed for its costs and receives a fixed base fee. In addition, 
the contractor can earn an "award fee"--a percent of the total contract 
amount--based on its performance in meeting specific metric standards 
and indicator targets that CMS identified as being particularly 
important to providing services to callers and controlling contract 
costs. The current metrics and their standards were put in place as of 
July 2007.[Footnote 32] Using the standards and targets, CMS evaluates 
the contractor's performance three times a year. For each evaluation 
period, CMS can decide to award all, part, or none of the available 
award fee amount depending on the contractor's performance. 

The current performance metrics quantify and measure the caller 
experience with 1-800-MEDICARE. Several metrics are specifically 
related to telephone access, such as the average amount of time callers 
wait after indicating that they want to speak to a CSR until their 
calls are answered by the CSR.[Footnote 33] The 1-800-MEDICARE 
contractor must meet the specific standards for these performance 
metrics in order to be awarded up to 15 percent of the award fee. Each 
evaluation period is 4 months long. If the contractor does not meet a 
performance metric standard during any month in an evaluation period, 
that month's portion of the award fee is withheld at the end of that 
evaluation period. 

CMS also has designed several performance indicators that it considers 
when measuring the performance of the 1-800-MEDICARE contractor. 
Several of these indicators are specifically related to telephone 
access, such as the average amount of time a CSR talks to a caller. 
These performance indicators do not have fixed standards against which 
the contractor's performance is measured; rather, CMS has established 
performance targets and compares the contractor's actual performance 
against these targets, using a scale ranging from substandard to 
superior, to determine the contractor's award fee. According to CMS 
officials, the purpose of the targets is to provide a baseline for 
planning and management of contact center activities. The performance 
indicators, when combined with other contract elements, account for up 
to 45 percent of the award fee. CMS also considers the contractor's 
performance related to program management and communication, contract 
compliance, and fiscal responsibility when determining the award fee. 
These elements, when combined, account for up to 40 percent of the 
award fee.[Footnote 34] 

In addition to the main 1-800-MEDICARE contract, CMS has awarded four 
additional contracts to individual companies for other activities 
directly related to the support of 1-800-MEDICARE: 

* A contract with a telephone company to manage the phone lines used by 
1-800-MEDICARE. This company also ensures that calls to CSRs are routed 
to the next available CSR with the skill set to assist the caller. 

* A contract to support and maintain the desktop application used by 
CSRs, including software CSRs use to access scripts. 

* A contract to manage the NDW, a central data repository that 
captures, aggregates, and integrates data on 1-800-MEDICARE from 
multiple sources. 

* A contract--referred to as the TQC contract--to conduct activities 
that include training, development of the scripts used by CSRs, and 
quality assurance, such as evaluation of CSR calls.[Footnote 35] 

Access to Services for Persons with LEP: 

As required by Executive Order 13166, HHS has developed a plan that 
identifies the necessary steps for the department and its agencies 
intended to ensure access to timely, quality language assistance 
services by eligible LEP persons to its programs and activities, such 
as 1-800-MEDICARE. The HHS LEP Plan, issued in December 2000, 
identifies seven elements designed to help each HHS agency, program, 
and activity to meet the department's goal of providing "access to 
timely, quality language assistance services to [LEP] persons." For 
example, the Plan includes elements related to oral language assistance 
services and efforts to assess accessibility and the quality of 
language assistance activities. (See appendix I for a list of the seven 
LEP Plan elements.) The Plan reflects HHS's overall goals for improving 
language access for individuals and includes strategies for improving 
technical assistance for language access services. HHS officials said 
that the Plan provides a "road map" for addressing HHS's goals, while 
allowing individual operating divisions and agencies, including CMS, 
some flexibility in implementing the Plan's elements. 

The 1-800-MEDICARE Contractor Met Most Standards and Some Targets 
Designed to Ensure Caller Access, but Average Wait Times under the 
Current Contract Varied by Type and Complexity of Inquiry: 

The current 1-800-MEDICARE contractor met most standards and some 
targets for telephone access-related performance metrics and 
indicators--measures designed to ensure all callers' access to 
services--from July 2007 through July 2008. In 10 of 13 months we 
analyzed, the current 1-800-MEDICARE contractor's performance met the 
standards for each of the required access-related performance metrics. 
Because of waivers granted by CMS, the agency considered the contractor 
to have met the relevant standards in 12 of the 13 months. While 
generally meeting the required standards for the three access-related 
performance metrics, the 1-800-MEDICARE contractor consistently met the 
target for only one of the three access-related performance indicators 
we analyzed. However, the amount of time callers waited to access 
services has varied depending on the type and complexity of callers' 
inquiries. The amount of time callers waited to speak with a CSR has 
increased since December 2005, but the performance standards related to 
caller wait time also have varied over time. 

Overall, the Current 1-800-MEDICARE Contractor Met CMS Monthly Access- 
Related Performance Standards: 

In 10 of 13 months we analyzed (July 2007 through July 2008), the 
current 1-800-MEDICARE contractor's performance met the standards for 
all three telephone access-related performance metrics required under 
its contract with CMS.[Footnote 36] Because of waivers granted by CMS, 
the contractor was considered by the agency to have met the relevant 
standards in 12 of the 13 months. Implemented in July 2007, the 
performance metrics and their associated standards were designed to 
measure, on a monthly basis, the 1-800-MEDICARE contractor's ability to 
ensure callers can access services and to determine the contractor's 
award fee during each evaluation period, which occur three times a 
year.[Footnote 37] The metrics, described below in table 1, are: (1) 
the average wait time (also referred to by CMS as the average speed of 
answer), (2) the percentage of unhandled CSR calls, and (3) the 
percentage of transfers. 

Table 1: Required Monthly Access-Related Performance Metrics and 
Standards, July 2007 through July 2008: 

Performance metric: Average wait time; 
CMS Definition of Performance Metric: The average amount of time 
callers wait until their calls are answered by a CSR[A]; 
Performance Standard: July 2007 through September 2007, between 7 and 8 
minutes, 30 seconds; October 2007 through July 2008, between 5 and 8 
minutes, 30 seconds[B]; 
Contractor Performance: Met standard in 12 of 13 months; received 
waiver for one month when standard not met[C]. 

Performance metric: Unhandled CSR calls[D]; 
CMS Definition of Performance Metric: The percentage of callers 
indicating they want to speak with a CSR but who are unable to because 
the call is abandoned, disconnected, or delayed because of high call 
volumes; 
Performance Standard: Percentage of unhandled calls standard varies 
with the average wait time--for example, if the average wait time 
standard is between 7 and 8 minutes the unhandled call standard is 40 
percent or less; if the average wait time standard is between 3 to 4 
minutes, the related unhandled CSR call standard would be 20 
percent[E]; 
Contractor Performance: Met standard in 11 of 13 months; received 
waiver for one month when standard not met[F]. 

Performance metric: Transfers; 
CMS Definition of Performance Metric: The percentage of calls 
transferred from one CSR to another CSR; 
Performance Standard: 20 percent or less; 
Contractor Performance: Met standard in 13 of 13 months. 

Source: GAO analysis of CMS data and 1-800-MEDICARE contract. 

Note: Prior to June 2007, CMS had contract metrics related to 
performance but, according to CMS officials, the contractor was 
primarily measured on activities related to the transition period, 
rather than on those metrics. CMS began using performance metrics in 
June 2007, with the current metrics implemented in July 2007. 

[A] The average wait time performance metric averages the time callers 
wait before CSRs answer their calls on a monthly basis. Wait times for 
individual calls may be lower or higher than the monthly average. 

[B] Effective August 2008, CMS revised its average wait time standard. 
The new standard requires the contractor to have an average wait time 
between 1 minute and 5 minutes. 

[C] Because of this waiver, CMS did not consider the contractor's 
performance on this metric for that month when determining overall 
performance or in determining an award fee. 

[D] Data on unhandled calls are collected by the NDW. 

[E] The current 1-800-MEDICARE contract includes several possible 
unhandled call time standards, all dependent on the average wait time 
standard. CMS officials said that the agency included a range of 
unhandled CSR call standards in the contract to facilitate future 
changes to the metrics. 

[F] Because of this waiver, CMS did not consider the contractor's 
performance on this metric for that month when determining overall 
performance or in determining an award fee. 

[End of table] 

Average wait time. The 1-800-MEDICARE contractor met the overall 
monthly average wait time performance standard--between 5 minutes and 8 
minutes, 30 seconds each month--in 12 of the 13 months (from July 2007 
through July 2008) we analyzed and, because of a waiver granted by CMS, 
was considered by the agency to have met the relevant standard in all 
13 months. (See figure 4.) CMS granted the waiver for one month's wait 
time standard for multiple reasons, including to account for the 
interruption of normal 1-800-MEDICARE contact center operations because 
of flooding in the Midwest. During the 13-month period we studied, the 
average wait time ranged from a low of less than 6 minutes in July 2008 
to a high of more than 8 minutes in September 2007.[Footnote 38] For 
the 5 months (October 2007 to February 2008) encompassing the most 
recent annual coordinated election period and a portion of the open 
enrollment period[Footnote 39]--when call volume to 1-800-MEDICARE 
typically increases--the contractor's monthly average wait time for all 
calls was between 6 minutes, 30 seconds and 7 minutes. 

Figure 4: 1-800-MEDICARE Contractor's Average Wait Time, July 2007 
through July 2008: 

[Refer to PDF for image] 

This figure is a line graph depicting the following data: 

Wait time standard maximum: 8.5 minutes; 
Wait time standard minimum: 7 minutes until October 2007; 5 minutes 
thereafter. 

Date: July 2007; 
Average wait time: 7.65 minutes. 

Date: August 2007;	
Average wait time: 7.38 minutes. 

Date: September 2007;	
Average wait time: 8.02 minutes. 

Date: October 2007;	
Average wait time: 6.58 minutes. 

Date: November 2007;	
Average wait time: 6.5 minutes. 

Date: December 2007;	
Average wait time: 6.5 minutes. 

Date: January 2008;	
Average wait time: 6.85 minutes. 

Date: February 2008;	
Average wait time: 6.95 minutes. 

Date: March 2008;	
Average wait time: 7.93 minutes. 

Date: April 2008;	
Average wait time: 7.88 minutes. 

Date: May 2008;	
Average wait time: 7.8 minutes. 

Date: June 2008;	
Average wait time: 8.72 minutes. 

Date: July 2008; 
Average wait time: 5.93 minutes. 

Source: CMS monthly data. 

[End of figure] 

CMS's monthly average wait time standard during the time period we 
analyzed allowed longer wait times than were typical at two of the four 
federal agencies we interviewed that use this metric for their contact 
centers.[Footnote 40] These two agencies used average wait time as a 
performance metric, with performance standards ranging from 4 minutes, 
30 seconds to 5 minutes, 30 seconds.[Footnote 41] CMS officials said 
they recognized that the 1-800-MEDICARE average wait time performance 
standard was long, but said that they selected it based on what could 
be reasonably achieved within the current budgeted amount available for 
the 1-800-MEDICARE contract. CMS officials also said that they believed 
that the budget would not allow for a standard that was more comparable 
to standards used in the contact center industry. However, CMS 
officials reported that in August 2008, they began requiring the 
contractor to meet an average wait time standard of between 1 minute 
and 5 minutes. CMS officials said that in order to better serve 
callers, they and the 1-800-MEDICARE contractor implemented processes 
and improved technologies that have increased the efficiency of the 
help line. For example, CMS and the contractor implemented new staffing 
initiatives to ensure that 1-800-MEDICARE has the necessary CSRs 
available to answer calls. 

Unhandled CSR call rate. The 1-800-MEDICARE contractor met its 
unhandled CSR call rate standard, keeping the number of unhandled calls 
within the required standards, in 11 of the 13 months we analyzed and, 
because of a waiver granted by CMS, was considered by the agency to 
have met the standard in 12 of 13 months. (See table 2 below.) The 
unhandled CSR call metric provides CMS with data on the extent to which 
callers abandon their calls while waiting for a CSR to answer the call, 
[Footnote 42] and CMS officials said that they expect the number of 
abandoned calls to decrease with shorter average wait times.[Footnote 
43] CMS granted the waiver for one month's unhandled CSR call rate 
standard because of changes during that month in its expectations of 
the contractor's performance. Though the performance metric standard 
was not changed, CMS encouraged the contractor to keep its average wait 
time below 8 minutes. To do this, the contractor increased the number 
of callbacks--instances in which callers are called back within 48 
hours rather than remaining on hold. Because CMS includes callbacks in 
the unhandled CSR call metric, in this month the contractor exceeded 
the unhandled CSR call standard. An expert in federal call centers 
indicated that the unhandled call rate standard used by CMS was high 
compared to some other federal agencies, but that these performance 
standards are usually based on what level of service an agency can 
afford to purchase with its contact center budget. 

Table 2: 1-800-MEDICARE Contractor's Unhandled CSR Call Performance, 
July 2007 through July 2008: 

Month: July 2007; 
Unhandled CSR call rate: 40.61%; 
Average wait time: 7:39; 
Unhandled CSR call rate performance standard, based on average wait 
time: 40% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
No. 

Month: August 2007; 
Unhandled CSR call rate: 38.57%; 
Average wait time: 7:23; 
Unhandled CSR call rate performance standard, based on average wait 
time: 40% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: September 2007; 
Unhandled CSR call rate: 38.45%; 
Average wait time: 8:01; 
Unhandled CSR call rate performance standard, based on average wait 
time: 45% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: October 2007; 
Unhandled CSR call rate: 35.00%; 
Average wait time: 6:35; 
Unhandled CSR call rate performance standard, based on average wait 
time: 35% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: November 2007; 
Unhandled CSR call rate: 37.61%; 
Average wait time: 6:30; 
Unhandled CSR call rate performance standard, based on average wait 
time: 35% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Waiver[A]. 

Month: December 2007; 
Unhandled CSR call rate: 26.82%; 
Average wait time: 6:31; 
Unhandled CSR call rate performance standard, based on average wait 
time: 35% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: January 2008; 
Unhandled CSR call rate: 29.73%; 
Average wait time: 6:51; 
Unhandled CSR call rate performance standard, based on average wait 
time: 35% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: February 2008; 
Unhandled CSR call rate: 25.81%; 
Average wait time: 6:57; 
Unhandled CSR call rate performance standard, based on average wait 
time: 35% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: March 2008; 
Unhandled CSR call rate: 29.41%; 
Average wait time: 7:56; 
Unhandled CSR call rate performance standard, based on average wait 
time: 40% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: April 2008; 
Unhandled CSR call rate: 27.38%; 
Average wait time: 7:53; 
Unhandled CSR call rate performance standard, based on average wait 
time: 40% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: May 2008; 
Unhandled CSR call rate: 26.08%; 
Average wait time: 7:48; 
Unhandled CSR call rate performance standard, based on average wait 
time: 40% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: June 2008; 
Unhandled CSR call rate: 28.48%; 
Average wait time: 8:43; 
Unhandled CSR call rate performance standard, based on average wait 
time: 45% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Month: July 2008; 
Unhandled CSR call rate: 20.39%; 
Average wait time: 5:56; 
Unhandled CSR call rate performance standard, based on average wait 
time: 30% or less; 
Did contractor meet the unhandled CSR call rate performance standard?: 
Yes. 

Source: GAO analysis of CMS 1-800-MEDICARE contract and evaluation 
data. 

[A] Because of the waiver granted for this month, CMS did not consider 
the contractor's performance on this metric for that month when 
determining overall performance or in determining an award fee. 

[End of table] 

Transfer rate. Transfers from one CSR to another occurred for no more 
than 20 percent of calls in each of the 13 months of data we analyzed, 
meeting the contractually required performance standard. While CMS 
officials said that they expect some transfers to occur normally, they 
also said a monthly transfer rate that exceeds 20 percent may imply 
that CSRs are inappropriately transferring callers. CSRs transferred 
nearly 13 percent of calls in July 2007, but the transfer rate 
increased to just under 20 percent by October 2007. Since October 2007, 
the transfer rate has generally decreased, with the lowest percentage 
of transferred calls of the 10 months we studied occurring in June 
2008. (See figure 5.) 

Figure 5: 1-800-MEDICARE Contractor's CSR Transfer Rate, July 2007 
through July 2008: 

[Refer to PDF for image] 

This figure is a line graph depicting the following data: 

Transfer Rate maximum: 20%. 

Date: Jul-07; 
Transfer rate: 12.75%. 

Date: Aug-07; 
Transfer rate: 14.25%. 

Date: Sep-07; 
Transfer rate: 12.51%. 

Date: Oct-07; 
Transfer rate: 19.76%. 

Date: Nov-07; 
Transfer rate: 18.52%. 

Date: Dec-07; 
Transfer rate: 17.61%. 

Date: Jan-08; 
Transfer rate: 16.62%. 

Date: Feb-08; 
Transfer rate: 17.38%. 

Date: Mar-08; 
Transfer rate: 17.63%. 

Date: Apr-08; 
Transfer rate: 16.31%. 

Date: May-08; 
Transfer rate: 13.78%. 

Date: Jun-08; 
Transfer rate: 12.74%. 

Date: Jul-08; 
Transfer rate: 13.26%. 

Source: CMS monthly data. 

[End of figure] 

CSRs can transfer calls to other CSRs for various reasons, including 
transferring a call from a Tier 1 CSR to a Tier 2 CSR if the complexity 
of a caller's inquiry requires a greater degree of training. According 
to CMS officials, most transfers occur between Tier 1 and Tier 2 CSRs, 
although transfers can also occur between two Tier 1 CSRs. For example, 
a caller may have a general Medicare inquiry answered initially by a 
Tier 1 CSR, but then raise a claims-related inquiry that may be better 
answered by another Tier 1 CSR with additional training. 

The Current 1-800-Medicare Contractor Consistently Met Performance 
Targets for One of Three Monthly Access-Related Performance Indicators: 

In addition to the three required performance metrics, CMS also 
established several performance indicators and associated targets for 
the 1-800-MEDICARE contractor to meet in order to receive a portion of 
its award fee. Three of the indicators are particularly related to 
callers' telephone access: (1) the average amount of time CSRs spend 
assisting callers, referred to as average handle time;[Footnote 44] (2) 
agent occupancy--the percentage of CSRs answering calls; and (3) 
forecasting of call volume going to CSRs.[Footnote 45] (See table 3 for 
more information on these selected performance indicators.) 

Table 3: Selected Access-Related Performance Indicators, July 2007 
through July 2008: 

Performance indicator name: Average handle time; 
CMS definition of performance indicator: The average amount of time 
CSRs take to respond to callers' inquiries, including talk time, hold 
time, and after-call work time; 
Performance indicator target: 
* July 2007--9.5 minutes; 
* August 2007--9 minutes; 
* September 2007 and beyond--8 minutes; 
Contractor performance: Met target in 0 of 13 months. 

Performance indicator name: Agent occupancy; 
CMS definition of performance indicator: The percentage of CSRs 
answering calls (versus those not available to answer calls, such as 
CSRs in training); 
Performance indicator target: 80 percent; 
Contractor performance: Met target in 13 of 13 months. 

Performance indicator name: CSR call volume forecast; 
CMS definition of performance indicator: The percentage difference 
between actual CSR call volume and forecasted CSR call volume; 
Performance indicator target: < 10 percent variance; 
Contractor performance: Met target in 4 of 13 months. 

Source: GAO analysis of CMS data and 1-800-MEDICARE contract. 

Notes: Prior to June 2007, CMS had metrics related to performance but 
according to CMS officials, the contractor was primarily measured on 
activities related to the transition period, rather than on those 
metrics. CMS began using performance metrics in June 2007, with the 
current metrics being implemented in July 2007. 

[End of table] 

CMS, in the current 1-800-MEDICARE contract, also specifies access- 
related performance indicators and associated targets related to the 
availability of 1-800-MEDICARE's network and information systems. CMS 
officials told us, as of August 2008, that they were working with the 1-
800-MEDICARE contractor to ensure the contractor understands the 
required data reporting for each of these performance indicators. 

The 1-800-MEDICARE contractor met its performance target in each of the 
13 months between July 2007 through July 2008 for one performance 
indicator--agent occupancy--but had varying experience in meeting its 
targets for the average handle time and CSR call volume forecasting. 

* For agent occupancy, the 1-800-MEDICARE contractor met its 
performance target--to have 80 percent or more of its CSRs answering 
calls--in each of the 13 months of data we analyzed. During these 13 
months, the 1-800-MEDICARE contractor's agent occupancy ranged from 83 
percent to 89 percent. 

* For average handle time, the 1-800-MEDICARE contractor did not meet 
its performance target--most recently set at 8 minutes--in any of the 
13 months analyzed. Average handle time peaked in December 2007 at 11 
minutes, 18 seconds and then decreased each month, reaching an average 
handle time of 8 minutes, 59 seconds in July 2008. CMS officials said 
that, because there are many factors that could affect call length, 
they consider average handle time in conjunction with other metrics and 
indicators to determine if the contractor is managing CSRs' time 
effectively. 

* The 1-800-MEDICARE contractor met its CSR call volume forecasting 
target--having a variance of less than 10 percent between its 
forecasted CSR call volume and actual CSR call volume--in 4 of the 13 
months of data we analyzed. 

Callers' Average Wait Times during the Current Contract Period Varied 
by Type and Complexity of Inquiry: 

During the current contract period, callers with claims inquiries or 
complex inquiries who needed assistance from a Tier 2 CSR generally 
experienced longer wait times than other callers. Although the current 
1-800-MEDICARE contract does not have an average wait time metric or 
indicator specifically related to the type of call, CMS's evaluation 
reports have noted that the agency expected the contractor to manage 
wait times by type of call to ensure that a caller with a claims 
inquiry does not wait longer than a caller with a general Medicare 
inquiry. 

Wait times varied by type of call. Callers waited, on average, less 
time to have general Medicare inquiries answered than to have calls 
about claims inquiries answered in all but 1 of the 11 months we 
analyzed under the current contract. During that month--December 2007-
-the difference between wait times for these two types of calls was 1 
second. (See figure 6.) During the time period we reviewed--ending July 
2008--callers with general Medicare inquiries waited between 5 minutes 
and 8 minutes, 30 seconds on average--a time that matches the 
performance standard for all calls during that period--in 10 of the 11 
months we analyzed. In contrast, claims callers were within that range 
in only 5 of the 11 months with the longest waits in the first two 
months of this period--almost 18 minutes in September 2007 and just 
over 11 minutes in October 2007. In September 2007, about 48 percent of 
callers disconnected before a CSR could answer their calls. Beginning 
in November 2007 and continuing through July 2008, average wait times 
for callers with claims inquiries were more closely aligned with that 
of callers with general Medicare inquiries. (For more information on 
differences in average wait times by type of claims call, see appendix 
II.) 

Figure 6: Monthly Average Wait Time for Callers with General Medicare 
Inquiries and Callers with Claims Inquiries, September 2007 through 
July 2008: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
General Medicare inquiries: 5.15 minutes; 
Claims inquiries: 17.75 minutes. 

Date: 10/2007; 
General Medicare inquiries: 5.16 minutes; 
Claims inquiries: 11.1 minutes. 

Date: 11/2007; 
General Medicare inquiries: 6.37 minutes; 
Claims inquiries: 7.2 minutes. 

Date: 12/2007; 
General Medicare inquiries: 6.58 minutes; 
Claims inquiries: 6.58 minutes. 

Date: 01/2008; 
General Medicare inquiries: 6.75 minutes; 
Claims inquiries: 7.38 minutes. 

Date: 02/2008; 
General Medicare inquiries: 6.42 minutes; 
Claims inquiries: 7.27 minutes. 

Date: 03/2008; 
General Medicare inquiries: 7.73 minutes; 
Claims inquiries: 8.83 minutes. 

Date: 04/2008; 
General Medicare inquiries: 7.72 minutes; 
Claims inquiries: 8.68 minutes. 

Date: 05/2008; 
General Medicare inquiries: 7.6 minutes; 
Claims inquiries: 8.82 minutes. 

Date: 06/2008; 
General Medicare inquiries: 8.7 minutes; 
Claims inquiries: 8.83 minutes. 

Date: 07/2008; 
General Medicare inquiries: 5.82 minutes; 
Claims inquiries: 6.3 minutes. 

Source: CMS monthly data. 

[End of figure] 

Wait times varied by complexity of calls. Between September 2007 and 
July 2008, callers generally waited longer, on average, to speak with a 
Tier 2 CSR--a CSR with additional training to respond to more complex 
calls--than to speak with a Tier 1 CSR. (See figure 7.) The average 
wait time for a Tier 1 CSR was between 5 minutes and 8 minutes, 30 
seconds--a time that matched the performance standard for all calls 
during that period--in 10 of 11 months we reviewed. In contrast, 
callers waited more than 8 minutes, 30 seconds to speak with a Tier 2 
CSR in 5 of 11 months we reviewed. During the first two months of this 
period, the average wait time for callers needing to speak with a Tier 
2 CSR was much longer--more than 12 minutes in September 2007 and 9 
minutes in October 2007. Beginning in November 2007 through July 2008, 
average wait times for callers needing to speak to a Tier 2 CSR were 
more closely aligned with that of callers needing to speak to a Tier 1 
CSR. (For more information on differences in average wait times by 
complexity of inquiry, see appendix II.) 

Figure 7: Monthly Average Wait Time for Tier 1 and Tier 2 CSRs, 
September 2007 through July 2008: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Tier 1 CSRs: 7.52 minutes; 
Tier 2 CSRs: 12.15 minutes. 

Date: 10/2007; 
Tier 1 CSRs: 6.5 minutes; 
Tier 2 CSRs: 9.38 minutes. 

Date: 11/2007; 
Tier 1 CSRs: 6.6 minutes; 
Tier 2 CSRs: 6.16 minutes. 

Date: 12/2007; 
Tier 1 CSRs: 6.6 minutes; 
Tier 2 CSRs: 6.43 minutes. 

Date: 01/2008; 
Tier 1 CSRs: 6.83 minutes; 
Tier 2 CSRs: 7.62 minutes. 

Date: 02/2008; 
Tier 1 CSRs: 6.57 minutes; 
Tier 2 CSRs: 7.38 minutes. 

Date: 03/2008; 
Tier 1 CSRs: 7.95 minutes; 
Tier 2 CSRs: 8.8 minutes. 

Date: 04/2008; 
Tier 1 CSRs: 7.9 minutes; 
Tier 2 CSRs: 8.55 minutes. 

Date: 05/2008; 
Tier 1 CSRs: 7.83 minutes; 
Tier 2 CSRs: 8.48 minutes. 

Date: 06/2008; 
Tier 1 CSRs: 8.72 minutes; 
Tier 2 CSRs: 9.02 minutes. 

Date: 07/2008; 
Tier 1 CSRs: 5.92 minutes; 
Tier 2 CSRs: 6.42 minutes. 

Source: CMS monthly data. 

[End of figure] 

Since Late 2005, Callers' Average Wait Times Have Increased: 

CMS's performance standards for caller wait times have varied since 
December 2005, and, since that time, callers have waited longer, on 
average, to have their calls answered by a CSR. (See figure 8.) The 
varying performance standards have allowed for a very short caller wait 
time, such as requiring 80 percent of calls to be answered in 30 
seconds, while at other times allowing for a longer wait time, such as 
requiring that 80 percent of calls be answered in 10 minutes. During 
the current contract period, CMS's performance standards have 
consistently allowed for a longer average caller wait time. From 
December 2005 to October 2006, when two contractors operated 1-800- 
MEDICARE, the average wait time in 8 of 11 months was below 5 minutes-
-the minimum average wait time standard under the current contract 
period through July 2008.[Footnote 46] During the transition period of 
the current 1-800-MEDICARE contract, October 30, 2006 through May 31, 
2007, the average wait time ranged from a low of 2 minutes, 6 seconds 
to a high of 10 minutes, 22 seconds. From the end of the transition 
period through July 2008, the monthly average wait times have been at 
or above 6 minutes, 30 seconds in all but one month, though they have 
not varied as widely as in prior periods we reviewed. 

Figure 8: Monthly Average Wait Time, December 2005 through July 2008: 

[Refer to PDF for image] 

This figure is a line graph depicting the following data: 

Date: Dec-05; 
Average wait time: 0.75 minutes. 

Date: Jan-06; 
Average wait time: 2.82 minutes. 

Date: Feb-06
Average wait time: 0.85 minutes. 

Date: Mar-06; 
Average wait time: 0.42 minutes. 

Date: Apr-06; 
Average wait time: 0.82 minutes. 

Date: May-06; 
Average wait time: 3.15 minutes. 

Date: Jun-06; 
Average wait time: 4.4 minutes. 

Date: July-06; 
Average wait time: 9.0 minutes. 

Date: Aug-06; 
Average wait time: 6.47 minutes. 

Date: Sept-06; 
Average wait time: 4.43 minutes. 

Date: Oct-06; 
Average wait time: 5.95 minutes. 

Start of current contract, October 30, 2006: 

Date: Nov-06; 
Average wait time: 5.03 minutes. 

Date: Dec-06; 
Average wait time: 2.1 minutes. 

Date: Jan-07; 
Average wait time: 5.27 minutes. 

Date: Feb-07; 
Average wait time: 8.63 minutes. 

Date: Mar-07; 
Average wait time: 10.32 minutes. 

Date: Apr-07; 
Average wait time: 10.37 minutes. 

Date: May-07; 
Average wait time: 7.47 minutes. 

End of current contractor's transition period, May 31, 2007. 

Date: June-07; 
Average wait time: 7.45 minutes. 

Date: July-07; 
Average wait time: 7.65 minutes. 

Date: Aug-07; 
Average wait time: 7.38 minutes. 

Date: Sept-07; 
Average wait time: 7.72 minutes. 

Date: Oct-07; 
Average wait time: 6.62 minutes. 

Date: Nov-07; 
Average wait time: 6.5 minutes. 

Date: Dec-07; 
Average wait time: 6.5 minutes. 

Date: Jan-08; 
Average wait time: 6.85 minutes. 

Date: Feb-08; 
Average wait time: 6.57 minutes. 

Date: Mar-08; 
Average wait time: 7.98 minutes. 

Date: Apr-08; 
Average wait time: 7.92 minutes. 

Date: May-08; 
Average wait time: 7.85 minutes. 

Date: June-08; 
Average wait time: 8.72 minutes. 

Date: July-08; 
Average wait time: 5.93 minutes. 

Source: CMS monthly data. 

[End of figure] 

Efforts to Provide LEP Callers Access to 1-800-MEDICARE Have Led to 
Shorter Wait Times for Spanish-speaking Callers, on Average, but Are 
Not Consistent with All LEP Plan Elements: 

CMS has made efforts to provide LEP callers with access to services 
through 1-800-MEDICARE by requiring the contractor to provide services 
in either English or Spanish and to provide interpretation services for 
callers speaking other languages. To meet these requirements, the 
contractor uses CSRs who are bilingual in English and Spanish and a 
telephone interpretation service to assist callers who speak other 
languages. Spanish-speaking callers waited less time on average to 
reach a CSR than their English-speaking counterparts in almost two- 
thirds of the months we reviewed--from December 2005 through July 
2008.[Footnote 47] Officials from the Office of Beneficiary Services 
(OBIS)--the CMS office with primary responsibility for 1-800-MEDICARE-
-said they were not aware of the HHS LEP Plan when awarding and 
assigning the current 1-800-MEDICARE contract and CMS has not 
identified an office responsible for acting as a point of contact for 
its management of the LEP Plan. Nonetheless, steps CMS has taken to 
provide services to LEP callers are consistent with some elements of 
the HHS LEP Plan adopted by the agency without modification, such as 
the element related to oral language assistance, but not others, such 
as the Plan's element for assessing quality and accessibility, which 
identifies the need for complaint mechanisms for language issues. 

1-800-MEDICARE Provides LEP Callers with Access to Services and Spanish-
speaking Callers Frequently Experience Shorter Average Wait Times to 
Reach a CSR than English-speaking Callers: 

As required by CMS, the 1-800-MEDICARE contractor provides callers with 
service in either English or Spanish and provides interpretation 
services for callers speaking other languages. The 1-800-MEDICARE 
contractor uses CSRs who are bilingual in English and Spanish to 
provide services to Spanish-speaking callers. As of January 1, 2008, 
slightly more than 7 percent of all CSRs at 1-800-MEDICARE were 
bilingual. Bilingual CSRs complete the same training required of CSRs 
who are not bilingual and must successfully handle test calls in both 
English and Spanish prior to answering 1-800-MEDICARE calls. Like CSRs 
who speak only English, bilingual CSRs use scripts, translated into 
Spanish, to provide assistance to callers. 

To meet CMS's requirement that the contractor ensure "real-time phone 
translations" for callers who speak neither English nor Spanish, the 1- 
800 contractor subcontracts with a "language line"--a telephone 
interpretation service. Through this language line, English-speaking 
CSRs have access to interpretation services in more than 150 languages. 
According to the 1-800-MEDICARE contractor, the language line provider 
uses internal certification and assessment to assure the quality of its 
interpreters.[Footnote 48] The number of languages for which CSRs have 
requested assistance has increased over time, from 40 languages in 2005 
to 73 as of March 2008. The language line also is used to provide 
interpretation services for Spanish-speaking callers if the wait for a 
bilingual CSR exceeds 20 minutes. In these cases, Spanish-speaking 
callers are transferred to non-Spanish-speaking CSRs who then use the 
language line to assist these callers. According to CMS officials, the 
cost of language line interpreters and the additional length of time 
required to connect and handle the call using the language line 
increases the costs of these calls. Calls needing Spanish 
interpretation accounted for the majority of calls to the language line 
during the time period we reviewed, growing from 60 percent of all 
calls to the line in 2005 to 80 percent of all calls in 2007. However, 
while the use of the language line for Spanish interpretation has 
increased, in most months we reviewed more than 90 percent of all 
Spanish-speaking callers to 1-800-MEDICARE were assisted by bilingual 1-
800-MEDICARE CSRs. 

Spanish-speaking callers who speak with bilingual CSRs frequently 
experienced shorter average monthly wait times to reach a CSR than 
their English-speaking counterparts. (See figure 9.) Spanish-speaking 
callers waited less time, on average, in slightly more than 60 percent 
of the months we reviewed (20 of 32), which encompassed parts of both 
the prior and current contract. 

Figure 9: Monthly Average Wait Time for English-and Spanish-speaking 
Callers, December 2005 through July 2008: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: Dec-05; 
Average wait time, English-speaking callers: 0.72 minutes; 
Average wait time, Spanish-speaking callers: 0.87 minutes. 

Date: Jan-06; 
Average wait time, English-speaking callers: 2.78 minutes; 
Average wait time, Spanish-speaking callers: 3.33 minutes. 

Date: Feb-06
Average wait time, English-speaking callers: 0.83 minutes; 
Average wait time, Spanish-speaking callers: 1.05 minutes. 

Date: Mar-06; 
Average wait time, English-speaking callers: 0.42 minutes; 
Average wait time, Spanish-speaking callers: 0.33 minutes. 

Date: Apr-06; 
Average wait time, English-speaking callers: 0.82 minutes; 
Average wait time, Spanish-speaking callers: 0.93 minutes. 

Date: May-06; 
Average wait time, English-speaking callers: 3.03 minutes; 
Average wait time, Spanish-speaking callers: 5.62 minutes. 

Date: Jun-06; 
Average wait time, English-speaking callers: 4.55 minutes; 
Average wait time, Spanish-speaking callers: 2.1 minutes. 

Date: July-06; 
Average wait time, English-speaking callers: 9.43 minutes; 
Average wait time, Spanish-speaking callers: 2.6 minutes. 

Date: Aug-06; 
Average wait time, English-speaking callers: 6.73 minutes; 
Average wait time, Spanish-speaking callers: 1.87 minutes. 

Date: Sept-06; 
Average wait time, English-speaking callers: 4.47 minutes; 
Average wait time, Spanish-speaking callers: 3.6 minutes. 

Date: Oct-06; 
Average wait time, English-speaking callers: 6.05 minutes; 
Average wait time, Spanish-speaking callers: 3.63 minutes. 

Start of current contract, October 30, 2006: 

Date: Nov-06; 
Average wait time, English-speaking callers: 5.22 minutes; 
Average wait time, Spanish-speaking callers: 1.83 minutes. 

Date: Dec-06; 
Average wait time, English-speaking callers: 2.15 minutes; 
Average wait time, Spanish-speaking callers: 0.77 minutes. 

Date: Jan-07; 
Average wait time, English-speaking callers: 5.42 minutes; 
Average wait time, Spanish-speaking callers: 2.07 minutes. 

Date: Feb-07; 
Average wait time, English-speaking callers: 8.87 minutes; 
Average wait time, Spanish-speaking callers: 3.93 minutes. 

Date: Mar-07; 
Average wait time, English-speaking callers: 10.57 minutes; 
Average wait time, Spanish-speaking callers: 5.65 minutes. 

Date: Apr-07; 
Average wait time, English-speaking callers: 10.57 minutes; 
Average wait time, Spanish-speaking callers: 6.65 minutes. 

Date: May-07; 
Average wait time, English-speaking callers: 7.63 minutes; 
Average wait time, Spanish-speaking callers: 4.05 minutes. 

End of current contractor's transition period, May 31, 2007. 

Date: June-07; 
Average wait time, English-speaking callers: 7.55 minutes; 
Average wait time, Spanish-speaking callers: 4.92 minutes. 

Date: July-07; 
Average wait time, English-speaking callers: 7.78 minutes; 
Average wait time, Spanish-speaking callers: 5.16 minutes. 

Date: Aug-07; 
Average wait time, English-speaking callers: 7.52 minutes; 
Average wait time, Spanish-speaking callers: 4.43 minutes. 

Date: Sept-07; 
Average wait time, English-speaking callers: 7.82 minutes; 
Average wait time, Spanish-speaking callers: 5.68 minutes. 

Date: Oct-07; 
Average wait time, English-speaking callers: 6.68 minutes; 
Average wait time, Spanish-speaking callers: 5.42 minutes. 

Date: Nov-07; 
Average wait time, English-speaking callers: 6.45 minutes; 
Average wait time, Spanish-speaking callers: 7.87 minutes. 

Date: Dec-07; 
Average wait time, English-speaking callers: 6.5 minutes; 
Average wait time, Spanish-speaking callers: 7.13 minutes. 

Date: Jan-08; 
Average wait time, English-speaking callers: 6.85 minutes; 
Average wait time, Spanish-speaking callers: 7.13 minutes. 

Date: Feb-08; 
Average wait time, English-speaking callers: 6.57 minutes; 
Average wait time, Spanish-speaking callers: 7.02 minutes. 

Date: Mar-08; 
Average wait time, English-speaking callers: 7.95 minutes; 
Average wait time, Spanish-speaking callers: 8.7 minutes. 

Date: Apr-08; 
Average wait time, English-speaking callers: 7.9 minutes; 
Average wait time, Spanish-speaking callers: 8.33 minutes. 

Date: May-08; 
Average wait time, English-speaking callers: 7.83 minutes; 
Average wait time, Spanish-speaking callers: 8.32 minutes. 

Date: June-08; 
Average wait time, English-speaking callers: 8.73 minutes; 
Average wait time, Spanish-speaking callers: 8.7 minutes. 

Date: July-08; 
Average wait time, English-speaking callers: 5.93 minutes; 
Average wait time, Spanish-speaking callers: 5.8 minutes. 

Source: CMS monthly data. 

[End of figure] 

While the current 1-800-MEDICARE contract does not have metrics 
specifically focused on call language, the average wait time 
experienced by Spanish-speaking callers under the current contract, 
beginning at the end of October 2006, was consistent with CMS's 
performance standard for overall average caller wait times as of July 
2008 in almost two-thirds of the months we reviewed.[Footnote 49] 
However, Spanish-speaking callers have recently waited slightly longer, 
on average, than their English-speaking counterparts. In addition, 
while Spanish-speaking callers with simple claims inquiries have 
experienced shorter average wait times than English-speaking callers, 
they have waited, on average, longer for all general Medicare inquiries 
and more complex claims inquiries. (For more information on differences 
in average wait times by caller language, see appendix II.) 

CMS Did Not Consider the LEP Plan When Determining Service Provision, 
Although Steps Taken Are Consistent with Some Plan Elements: 

Officials from OBIS--the CMS office with primary responsibility for 1- 
800-MEDICARE--said they were not aware of the LEP Plan when awarding 
and assigning the current contract for operation of the help line. 
[Footnote 50] Rather, they relied primarily on industry best practices 
when determining how to require the 1-800-MEDICARE contractor to 
provide LEP services.[Footnote 51] In addition, while officials from 
HHS's OCR and CMS's OEOCR said CMS chose to adopt the HHS LEP Plan as 
issued--although HHS allowed its agencies flexibility to modify it-- 
officials from multiple CMS offices were unable to identify an office 
or official responsible for acting as the central point of contact 
responsible for the agency's management of the Plan.[Footnote 52] A key 
factor in meeting standards for internal control in federal agencies is 
defining and assigning key areas of authority and responsibility--such 
as a point of contact for an agency-wide plan--and communicating that 
information throughout the organization.[Footnote 53] Without an office 
or official responsible for management of the Plan, staff lack a source 
of guidance that could assist them in taking steps consistent with the 
Plan to provide services to people with LEP. 

Although CMS officials did not consider the LEP Plan when determining 
how services for LEP callers to 1-800-MEDICARE would be provided, steps 
they have taken to implement language services are consistent with 
some--but not all--elements of the Plan.[Footnote 54] In particular, by 
requiring its contractor to provide services to all LEP callers, CMS 
has taken a step consistent with the LEP Plan element stating that 
"each agency, program, and activity… will arrange for the provision of 
oral language assistance in response to the needs of LEP customers, 
both in face-to-face and telephone encounters."[Footnote 55] In 
addition, the LEP Plan states that agencies, activities, and programs 
will implement mechanisms to assess the LEP status and needs of current 
and potential customers. To do this, CMS officials said they and the 
contractor regularly review the frequency with which LEP callers 
contact the 1-800 line to determine appropriate bilingual CSR staffing 
levels. In addition, CMS officials noted that this information could be 
used to determine whether CSRs who speak languages other than English 
and Spanish should be made available to callers. 

However, CMS has not taken steps consistent with other Plan elements 
related to 1-800-MEDICARE. For example, CMS officials did not identify, 
and did not require the contractor to identify, a specific official or 
office to allow LEP callers to register concerns or complaints 
regarding language assistance services provided by 1-800-MEDICARE, as 
indicated by the LEP Plan element for assessing accessibility and 
quality.[Footnote 56] In addition, while CMS considers the number and 
languages of LEP callers currently using 1-800-MEDICARE, consistent 
with the LEP Plan's element related to the assessment of needs and 
capacity, neither the number nor proportion of Medicare-eligible or 
Medicare-enrolled LEP beneficiaries is specifically considered when 
planning how 1-800-MEDICARE language services will be provided. The LEP 
Plan also directs agencies to develop policies and procedures for each 
Plan element, to designate staff responsible for implementing these 
policies and procedures, and to provide staff with training on them. 
However, given that CMS staff within OBIS were not aware of the 
existence of the LEP Plan until our review, no uniform CMS policies and 
procedures had been used to implement any elements of the LEP Plan for 
1-800-MEDICARE, nor had training specific to the LEP Plan occurred. 
[Footnote 57] 

CMS Uses Common Management Practices to Oversee Callers' Access to 
Services and to Accurate Information from 1-800-MEDICARE: 

CMS uses all six of the management practices we identified as commonly 
used by contact centers to oversee 1-800-MEDICARE callers' access to 
services and to accurate information.[Footnote 58] (See table 4 for a 
description of these six common oversight practices.) These practices 
are addressed in the current 1-800-MEDICARE contract and reflected in 
CMS's ongoing oversight of the help line. For example, CMS awarded some 
of the available award fee to the 1-800-MEDICARE contractor for meeting 
certain performance metrics and used CSR feedback to improve the 
oversight of information CSRs provide to beneficiaries. In addition, 
CMS used customer satisfaction surveys to identify areas for 
improvement, including the IVR, and worked with the contractor to 
improve call volume forecasting in an effort to ensure appropriate 
staffing levels to meet callers' needs. 

Table 4: Six Common Management Practices Used to Oversee Callers' 
Access to Services and to Accurate Information: 

Management practice: Clearly defined performance metrics relating to 
access; 
Description of management practices: Selects quantifiable, clearly 
defined units to measure aspects of performance that affect a caller's 
access to information or services[A]. 

Management practice: Customer satisfaction surveys; 
Description of management practices: Surveys callers' satisfaction with 
the service received from the contact center, including, to a limited 
extent, callers' opinions about whether the information received was 
accurate. 

Management practice: Assurance that information referenced by CSRs to 
answer callers' questions is accurate; 
Description of management practices: Takes steps to ensure the 
information used by CSRs to respond to callers' inquiries is accurate, 
including scripts and organizational internet resources. 

Management practice: Evaluation of CSRs' interaction with callers; 
Description of management practices: Reviews information provided 
during calls to evaluate how well CSRs handled the inquiry; the 
evaluation typically uses a score sheet that allows the reviewer to 
rate the CSRs in multiple areas, such as courtesy, accuracy of 
information provided, timeliness, and completeness. 

Management practice: Capacity planning to meet caller demand for 
services; 
Description of management practices: Uses quantitative or judgmental 
methods to predict the number of calls that will come into a contact 
center--or workload--in future time periods. A contact center's 
workload determines the resources, including staff, information 
technology, and corresponding funding, needed to meet the demand for 
services. 

Management practice: Validation of contact center performance reports; 
Description of management practices: Validates the data provided in 
contractor-prepared reports to ensure accuracy; these reports, which 
could be provided daily, weekly, monthly, or quarterly, include 
operational information such as the contact center's workload volumes, 
handling times, and results of the contractor's contact monitoring. 

Source: GAO analysis based on prior GAO reports, literature on industry 
practices, discussions with federal agencies and industry experts. 

[A] The current 1-800-MEDICARE contract uses the terms "metrics" and 
"indicators" with corresponding standards and targets to refer to 
quantifiable, clearly defined units used to measure callers' experience 
with 1-800-Medicare. 

[End of table] 

Clearly defining performance metrics and indicators. The performance 
metrics and indicators that CMS officials said they designed to 
encourage improved performance and correct identified problems are 
clearly defined in the 1-800-MEDICARE contract.[Footnote 59] The 
current 1-800-MEDICARE contract identifies each metric and indicator, 
provides a definition for these measures, and sets a standard or target 
for each measure. CMS uses these measures to evaluate contractor 
performance three times a year and to provide an award fee if the 
contractor performs at or above set levels. Because the contractor did 
not meet all standards and targets for any evaluation period as of May 
2008, CMS has awarded only part of the possible award fee to date, in 
accordance with the award guidance in the 1-800-MEDICARE contract. 
[Footnote 60] Award fees are based on a scoring system, in which a 
percentage of the award fee is distributed depending on CMS's rating of 
the contractor's performance, ranging from the lowest rating, 
substandard, to the highest rating, superior.[Footnote 61] CMS rated 
the contractor as good or very good from the start of the current 
contract through May 2008. 

CMS also uses the performance metrics and indicators on a continuous 
basis to monitor and evaluate callers' ability to access information 
from 1-800-MEDICARE. For example, CMS officials said they receive 
updates from the contractor on the average wait times--a measure of 
caller access--through daily e-mails. If callers' average wait time 
exceeds 5 minutes, CMS officials receive e-mail notification every half 
hour. During that time, CMS officials said the contractor also notifies 
them of actions it is taking to improve caller wait times. 

Customer satisfaction surveys. CMS officials said they use the results 
of customer satisfaction surveys to gain insight into callers' 
experiences with 1-800-MEDICARE services and to identify opportunities 
for improving services. In June 2008, the TQC contractor assumed 
responsibility for conducting a customer satisfaction survey of 20 
percent of randomly selected 1-800-MEDICARE callers.[Footnote 62] The 
TQC contractor will provide CMS with the results from this survey on an 
ongoing basis; however, initial results were not available as of May 
2008. Prior to June 2008, CMS required the 1-800-MEDICARE contractor to 
call back randomly selected beneficiaries who contacted 1-800-MEDICARE 
and administer a customer satisfaction survey developed by CMS. CMS 
officials said they used the information from the surveys to make 
changes to 1-800-MEDICARE. For example, based on survey participant 
concerns about the IVR, which is used by 18 to 20 percent of callers to 
resolve their inquiries, CMS officials indicated they made changes to 
the IVR prompts to make them easier to follow. CMS officials said that 
they anticipate future callers to 1-800-MEDICARE will be more willing 
to use technologies such as the IVR to obtain information. However, as 
of November 2008, only limited testing and focus groups had been 
conducted to confirm this anticipated trend.[Footnote 63] 

Ensuring accurate information. To ensure that beneficiaries receive 
consistent and accurate program information, CSRs are required to use 
scripts approved by the agency and CMS, and contractor officials 
reported taking steps designed to make correct scripts easier for CSRs 
to identify and to make scripts easier to understand.[Footnote 64] In 
April 2008, the TQC contractor began developing content for 1-800- 
MEDICARE scripts, which previously had been developed by the 1-800- 
MEDICARE contractor and reviewed by CMS. The TQC contractor employs 
staff with expertise in Medicare who develop and revise scripts as 
needed. CMS reviews and approves all scripts before making them 
available to CSRs. According to CMS officials, in August 2008 the TQC 
contractor began quarterly script reviews for accuracy as well as 
legislative, program, or policy changes. In addition to reviews 
conducted by the TQC contractor, CMS officials said that they have used 
the results of customer satisfaction surveys and call evaluations to 
assess how well scripts meet the needs of callers and how easy scripts 
are for CSRs to find. CMS can also access a computer application that 
captures CSR feedback on scripts, including how easy scripts are to 
understand, an issue identified in previous GAO work. 

Evaluating CSR interaction with callers. To assess whether callers 
receive consistent and accurate information from 1-800-MEDICARE, CMS 
requires both the 1-800-MEDICARE contractor and the TQC contractor to 
evaluate CSRs' interactions with callers through call monitoring. The 1-
800-MEDICARE contractor listens to four calls a month for each CSR, 
evaluating the CSRs' performance on customer service skills identified 
by CMS, including tone, using scripts appropriately, and completeness 
of information provided. CMS officials said they designed the 1-800- 
MEDICARE contractor's call evaluation process for contractor 
supervisors to coach CSRs on their performance and to help improve 
CSRs' ability to use software to find appropriate scripts. The 1-800- 
MEDICARE contractor has reported on the required evaluations monthly. 
Using these evaluation reports, CMS officials said, they worked with 
the 1-800-MEDICARE contractor to identify and correct trends or issues 
with scripts, software, or CSR training. For example, CMS used the call 
evaluation process to determine the reasons why callers may need to be 
transferred to a more experienced CSR or place another call to 1-800- 
MEDICARE. This analysis was part of the "First Call Resolution 
Initiative," an effort to increase the number of callers who have their 
inquiries resolved with one phone call rather than many.[Footnote 65] 
However, in its evaluation of the 1-800-MEDICARE contractor for the 
period ending May 2008, CMS reported that it observed callers receiving 
poor service from CSRs who received perfect call evaluation scores from 
the 1-800-MEDICARE contractor for those calls. CMS noted the 1-800- 
MEDICARE contractor needed to improve consistency between actual CSR 
call evaluation scores and the quality of service callers receive. 

In addition, since April 1, 2008, CMS has required the TQC contractor 
each month to monitor and evaluate 600 randomly selected calls in 
English and 225 randomly selected calls in Spanish. This call sample is 
designed to allow CMS to generalize trends that emerge from this sample 
to the call volume of 1-800-MEDICARE as a whole. CMS officials said 
they anticipate analyzing any trends identified from this call 
monitoring to note areas for improvement to the 1-800-MEDICARE help 
line. CMS officials also said that they will use the results of the TQC 
scores as part of the 1-800-MEDICARE contractor's regular performance 
evaluation and resulting award payment beginning in October 2008. 
[Footnote 66] In addition, CMS officials reported monitoring calls 
themselves and meeting weekly with the 1-800-MEDICARE and TQC 
contractors to listen to and rate recorded calls as a group. CMS 
officials said that these meetings help to ensure that the contractors 
understand the standard of service CMS expects callers to receive. CMS 
officials also planned to perform similar evaluations on calls for 
which customer satisfaction survey information is available--a practice 
identified by an industry expert as a method to improve contact center 
service. 

Capacity planning. CMS officials said they work with the 1-800-MEDICARE 
contractor to create short-and long-term call volume forecasts and to 
determine whether systems and staffing can handle call volume, taking 
into account Medicare's annual coordinated election and open enrollment 
periods when inquiries peak. CMS requires the 1-800-MEDICARE contractor 
to produce call volume forecasts that are accurate within 10 percent of 
actual call volumes for the forecasted period. CMS oversight of the 
contractor's forecasting efforts has identified significant differences 
between the long-term forecasts and actual monthly volume of calls 
going to CSRs--forecasting up to 35 percent more calls than were 
actually received by CSRs for the performance period ending January 
2008 and causing projected staffing costs for this period to be 
overstated.[Footnote 67] In its evaluation of the performance period 
ending January 2008, CMS notified the contractor that performance in 
this area needed to improve and stated that it wanted the contractor to 
identify methods of ensuring consistent and accurate forecasts. 
However, CMS noted that when the forecasted call volume was not 
realized, the 1-800-MEDICARE contractor adjusted its staffing so that 
only the number of CSRs needed to meet the performance standard related 
to average wait times were available. In its evaluation for the period 
ending May 2008, CMS noted a significant improvement in long-term 
forecasting and indicated that the 1-800-MEDICARE contractor had a 
better understanding of events that affected call volume throughout the 
year. 

To ensure that 1-800-MEDICARE systems, such as phone lines and desktop 
software, are available to handle forecasted call volumes, CMS requires 
the 1-800-MEDICARE contractor to notify the agency of systems outage 
incidents that affect callers' access. In the evaluation period ending 
January 2008, CMS identified inconsistencies in the outage reporting 
process, which caused key information to be omitted from reports about 
systems outage incidents. In its evaluation for the period ending May 
2008, CMS noted some improvement in reporting systems outage incidents, 
but indicated that the 1-800-MEDICARE contractor needed to improve the 
consistency of its reporting practices in the future. CMS officials 
said they are working with the contractor to address this issue and 
reported finalizing a process for reporting systems outages in early 
November 2008. 

Validation of contractor reports. CMS officials said they validate 
contractor reports--many of which are used to determine contract award 
fees--by analyzing data captured by 1-800-MEDICARE computer systems and 
having regular meetings with their contractors. To ensure the integrity 
of data collected by 1-800-MEDICARE systems, CMS collects and stores 
these data in the NDW, which is managed by a separate contractor. CMS 
officials compare data from the NDW on the actual call center 
performance to reports submitted by the 1-800-MEDICARE contractor, such 
as the long-term call volume forecasts. Using this method of validating 
reports, CMS determined that the 1-800-MEDICARE contractor's 
forecasting reports were inaccurate and required the contractor to 
improve in this area, which the contractor did over the next evaluation 
period. In addition to validating contractor reports through the NDW, 
CMS officials said they used weekly and monthly status reports, 
meetings with contractors, and visits to 1-800-MEDICARE contractor 
sites to monitor contractor performance between periodic evaluations. 

Conclusions: 

To date, the current 1-800-MEDICARE contractor has met most of CMS's 
performance standards and some of the performance targets designed to 
ensure callers' access to services from the help line. In addition, by 
employing all six commonly used management practices to oversee 1-800- 
MEDICARE callers' access to services and accurate information, CMS 
gains valuable data to assess the contractor's performance and identify 
areas for improvement. In particular, the new TQC contract provides CMS 
with an opportunity to continue to improve both access to, and accuracy 
of, information. 

However, while callers with LEP can access services through 1-800- 
MEDICARE, CMS has not taken steps to ensure that officials throughout 
the agency, including within OBIS, are fully aware of the LEP Plan, 
which HHS designed to be a "road map" for providing appropriate 
services to this population. By not identifying an official point of 
contact responsible for management of the Plan, CMS is lacking a key 
internal control measure--a clearly defined area of responsibility that 
has been communicated agencywide. While CMS has taken steps to ensure 
access for LEP callers to 1-800-MEDICARE, a clearly identified office 
or official responsible for the Plan could provide guidance in areas 
where steps consistent with the LEP Plan have not been taken and could 
work to ensure consistent use of the Plan across the agency. 

Recommendation for Executive Action: 

To ensure CMS offices, including those that oversee the operation of 
the 1-800-MEDICARE help line are aware of, and take steps consistent 
with, the HHS LEP Plan when considering the needs of people with LEP, 
CMS should designate an official or office with responsibility for 
managing the LEP Plan. 

Agency Comments and Our Evaluation: 

We provided CMS with a draft of this report for its review and comment. 
The agency provided written comments, which we addressed as appropriate 
and which have been reprinted in appendix III. The current 1-800- 
MEDICARE contractor stated that the report was factually accurate and 
provided oral technical comments, which we incorporated as appropriate. 
The Social Security Administration and the departments of Defense, 
Education, and Treasury told us they had no comments on the draft 
report. 

In responding to our draft, CMS stated that it has taken steps to 
implement our recommendation. More specifically, CMS has identified an 
official responsible for the development of an LEP Plan for the agency 
that, when finalized, is intended to define responsibility for ensuring 
consistent and reliable LEP tracking and reporting. CMS also noted that 
the report identified issues related to accurate forecasts and wait 
times experienced by callers for general Medicare and claims calls. The 
agency reiterated that these issues were affected by the transition, 
completed in 2007, of claims calls previously answered by FFS 
contractors to 1-800-MEDICARE. CMS also stated that its intent is to 
try to fully answer callers' questions rather than focusing exclusively 
on reducing average handle time--the average amount of time CSRs take 
to respond to callers' inquiries. Additionally, CMS stated that while 
it has not specifically evaluated the number or proportion of LEP 
beneficiaries who are Medicare eligible and enrolled, it continues to 
review demand and consider other opportunities to best serve LEP 
callers. Finally, CMS noted that some information regarding the 1-800- 
MEDICARE contractor's operations and contract with the agency are 
considered proprietary. 

As agreed with your office, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies to the 
Secretary of Health and Human Services and others. The report also will 
be available at no charge on the GAO Web site at [hyperlink, 
http://www.gao.gov]. 

If you or your staff have any questions about this report, please 
contact me at (202) 512-7114 or at kingk@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. 

Sincerely yours, 

Signed by: 

Kathleen M. King: 
Director, Health Care: 

[End of section] 

Appendix I: Elements of the Department of Health and Human Services' 
Limited English Proficiency Plan: 

Table: 

Element: Assessment: needs and capacity; 
Element description: "Each agency, program, and activity of the HHS 
[Department of Health and Human Services] will have in place mechanisms 
to assess, on a regular and consistent basis, the LEP [limited English 
proficiency] status and language assistance needs of current and 
potential customers, as well as mechanisms to assess the agency's 
capacity to meet these needs according to the elements of this [HHS 
LEP] plan." 

Element: Oral language assistance services; 
Element description: "Each agency, program, and activity of HHS will 
arrange for the provision of oral language assistance in response to 
the needs of LEP customers, in both face-to-face and by telephone 
encounters." 

Element: Written translations; 
Element description: "Each agency, program, and activity of HHS will 
provide vital documents in languages other than English where a 
significant number or percentage of the customers served or eligible to 
be served has [LEP]. These written materials may include paper and 
electronic documents such as publications, notices, correspondence, web 
sites and signs." 

Element: Policies and procedures; 
Element description: "Each agency, program, and activity of HHS will 
have in place specific written policies and procedures related to each 
of the plan elements and designated staff who will be responsible for 
implementing activities related to these policies." 

Element: Notification of the availability of free language services; 
Element description: "Each agency, program, and activity of HHS will 
proactively inform LEP customers of the availability of free language 
assistance services through both oral and written notice, in his or her 
primary language." 

Element: Staff training; 
Element description: "Each agency, program, and activity of HHS will 
train front-line and managerial staff on the policies and procedures of 
its language assistance activities." 

Element: Assessing accessibility and quality; 
Element description: "Each agency, program, and activity of HHS will 
institute procedures to assess the accessibility and quality of 
language assistance activities for LEP customers." 

Source: HHS LEP Plan. 

[End of table] 

[End of section] 

Appendix II: Average Caller Wait Time by Type and Complexity of Call 
and Language of Caller, September 2007 through July 2008: 

This appendix provides more detailed information on the average wait 
times experienced by callers depending on the type of inquiry--general 
Medicare and claims--and complexity of their call or the language in 
which they need assistance for the period September 2007 through July 
2008 of the current contract.[Footnote 68] 

Wait Times by Type and Complexity of Calls: 

Figures 10 and 11 provide information on differences in average wait 
times experienced by callers depending on their type of inquiry-- 
general Medicare and claims--and the complexity of their inquiry. While 
simple inquiries may be resolved by Tier 1 CSRs, inquiries of greater 
complexity may require the assistance of a Tier 2 CSR. Figures 12 
through 14 compare the wait times for callers with simple inquiries to 
callers with complex inquiries by type of claims inquiry--Medicare Part 
A, Part B, and, within Part B, durable medical equipment (DME). 
[Footnote 69] 

Figure 10: Monthly Average Wait Times for General Medicare and Claims 
Inquiries by Tier 1 CSRs: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, Tier 1, General Medicare: 4.9 minutes; 
Average wait time, Tier 1, Claims: 18.25 minutes. 

Date: 10/2007; 
Average wait time, Tier 1, General Medicare: 4.98 minutes; 
Average wait time, Tier 1, Claims: 11.12 minutes. 

Date: 11/2007; 
Average wait time, Tier 1, General Medicare: 6.37 minutes; 
Average wait time, Tier 1, Claims: 7.32 minutes. 

Date: 12/2007; 
Average wait time, Tier 1, General Medicare: 6.57 minutes; 
Average wait time, Tier 1, Claims: 6.7 minutes. 

Date: 01/2008; 
Average wait time, Tier 1, General Medicare: 6.72 minutes; 
Average wait time, Tier 1, Claims: 7.33 minutes. 

Date: 02/2008; 
Average wait time, Tier 1, General Medicare: 6.38 minutes; 
Average wait time, Tier 1, Claims: 7.2 minutes. 

Date: 03/2008; 
Average wait time, Tier 1, General Medicare: 7.68 minutes; 
Average wait time, Tier 1, Claims: 8.78 minutes. 

Date: 04/2008; 
Average wait time, Tier 1, General Medicare: 7.7 minutes; 
Average wait time, Tier 1, Claims: 8.63 minutes. 

Date: 05/2008; 
Average wait time, Tier 1, General Medicare: 7.58 minutes; 
Average wait time, Tier 1, Claims: 8.82 minutes. 

Date: 06/2008; 
Average wait time, Tier 1, General Medicare: 8.68 minutes; 
Average wait time, Tier 1, Claims: 8.83 minutes. 

Date: 07/2008; 
Average wait time, Tier 1, General Medicare: 5.8 minutes; 
Average wait time, Tier 1, Claims: 6.3 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 11: Monthly Average Wait Times for General Medicare and Claims 
Inquiries by Tier 2 CSRs: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, Tier 2, General Medicare: 10.95 minutes; 
Average wait time, Tier 2, Claims: 13.81 minutes. 

Date: 10/2007; 
Average wait time, Tier 2, General Medicare: 8.87 minutes; 
Average wait time, Tier 2, Claims: 10.28 minutes. 

Date: 11/2007; 
Average wait time, Tier 2, General Medicare: 6.4 minutes; 
Average wait time, Tier 2, Claims: 5.67 minutes. 

Date: 12/2007; 
Average wait time, Tier 2, General Medicare: 6.85 minutes; 
Average wait time, Tier 2, Claims: 5.15 minutes. 

Date: 01/2008; 
Average wait time, Tier 2, General Medicare: 7.43 minutes; 
Average wait time, Tier 2, Claims: 8.05 minutes. 

Date: 02/2008; 
Average wait time, Tier 2, General Medicare: 7.02 minutes; 
Average wait time, Tier 2, Claims: 8.2 minutes. 

Date: 03/2008; 
Average wait time, Tier 2, General Medicare: 8.45 minutes; 
Average wait time, Tier 2, Claims: 9.6 minutes. 

Date: 04/2008; 
Average wait time, Tier 2, General Medicare: 8.1 minutes; 
Average wait time, Tier 2, Claims: 9.23 minutes. 

Date: 05/2008; 
Average wait time, Tier 2, General Medicare: 8.08 minutes; 
Average wait time, Tier 2, Claims: 8.92 minutes. 

Date: 06/2008; 
Average wait time, Tier 2, General Medicare: 9.28 minutes; 
Average wait time, Tier 2, Claims: 8.67 minutes. 

Date: 07/2008; 
Average wait time, Tier 2, General Medicare: 6.46 minutes; 
Average wait time, Tier 2, Claims: 6.33 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 12: Monthly Average Wait Time for Tier 1 and 2 CSRs Responding 
to Part A Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, Tier 1, Part A: 18.8 minutes; 
Average wait time, Tier 2, Part A: 14.95 minutes. 

Date: 10/2007; 
Average wait time, Tier 1, Part A: 11.45 minutes; 
Average wait time, Tier 2, Part A: 10.5 minutes. 

Date: 11/2007; 
Average wait time, Tier 1, Part A: 7.52 minutes; 
Average wait time, Tier 2, Part A: 4.98 minutes. 

Date: 12/2007; 
Average wait time, Tier 1, Part A: 6.58 minutes; 
Average wait time, Tier 2, Part A: 2.5 minutes. 

Date: 01/2008; 
Average wait time, Tier 1, Part A: 7.42 minutes; 
Average wait time, Tier 2, Part A: 8.13 minutes. 

Date: 02/2008; 
Average wait time, Tier 1, Part A: 7.2 minutes; 
Average wait time, Tier 2, Part A: 8.08 minutes. 

Date: 03/2008; 
Average wait time, Tier 1, Part A: 8.73 minutes; 
Average wait time, Tier 2, Part A: 9.52 minutes. 

Date: 04/2008; 
Average wait time, Tier 1, Part A: 8.62 minutes; 
Average wait time, Tier 2, Part A: 9.03 minutes. 

Date: 05/2008; 
Average wait time, Tier 1, Part A: 8.72 minutes; 
Average wait time, Tier 2, Part A: 8.62 minutes. 

Date: 06/2008; 
Average wait time, Tier 1, Part A: 9.6 minutes; 
Average wait time, Tier 2, Part A: 10.27 minutes. 

Date: 07/2008; 
Average wait time, Tier 1, Part A: 6.83 minutes; 
Average wait time, Tier 2, Part A: 7.75 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 13: Monthly Average Wait Time for Tier 1 and Tier 2 CSRs 
Responding to Part B Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, Tier 1, Part B: 19.63 minutes; 
Average wait time, Tier 2, Part B: 13.23 minutes. 

Date: 10/2007; 
Average wait time, Tier 1, Part B: 12.05 minutes; 
Average wait time, Tier 2, Part B: 10.33 minutes. 

Date: 11/2007; 
Average wait time, Tier 1, Part B: 7.45 minutes; 
Average wait time, Tier 2, Part B: 6.08 minutes. 

Date: 12/2007; 
Average wait time, Tier 1, Part B: 6.85 minutes; 
Average wait time, Tier 2, Part B: 5.77 minutes. 

Date: 01/2008; 
Average wait time, Tier 1, Part B: 7.27 minutes; 
Average wait time, Tier 2, Part B: 7.9 minutes. 

Date: 02/2008; 
Average wait time, Tier 1, Part B: 7.13 minutes; 
Average wait time, Tier 2, Part B: 7.92 minutes. 

Date: 03/2008; 
Average wait time, Tier 1, Part B: 8.73 minutes; 
Average wait time, Tier 2, Part B: 9.45 minutes. 

Date: 04/2008; 
Average wait time, Tier 1, Part B: 8.58 minutes; 
Average wait time, Tier 2, Part B: 9.12 minutes. 

Date: 05/2008; 
Average wait time, Tier 1, Part B: 8.82 minutes; 
Average wait time, Tier 2, Part B: 8.85 minutes. 

Date: 06/2008; 
Average wait time, Tier 1, Part B: 9.55 minutes; 
Average wait time, Tier 2, Part B: 9.97 minutes. 

Date: 07/2008; 
Average wait time, Tier 1, Part B: 6.68 minutes; 
Average wait time, Tier 2, Part B: 7.23 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 14: Monthly Average Wait Times for Tier1 and Tier 2 CSRs 
Responding to DME Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, Tier 1, DME: 13.6 minutes; 
Average wait time, Tier 2, DME: 15.62 minutes. 

Date: 10/2007; 
Average wait time, Tier 1, DME: 7.63 minutes; 
Average wait time, Tier 2, DME: 10.0 minutes. 

Date: 11/2007; 
Average wait time, Tier 1, DME: 6.58 minutes; 
Average wait time, Tier 2, DME: 5.08 minutes. 

Date: 12/2007; 
Average wait time, Tier 1, DME: 6.23 minutes; 
Average wait time, Tier 2, DME: 5.42 minutes. 

Date: 01/2008; 
Average wait time, Tier 1, DME: 7.48 minutes; 
Average wait time, Tier 2, DME: 8.35 minutes. 

Date: 02/2008; 
Average wait time, Tier 1, DME: 7.43 minutes; 
Average wait time, Tier 2, DME: 8.98 minutes. 

Date: 03/2008; 
Average wait time, Tier 1, DME: 8.98 minutes; 
Average wait time, Tier 2, DME: 10.0 minutes. 

Date: 04/2008; 
Average wait time, Tier 1, DME: 8.9 minutes; 
Average wait time, Tier 2, DME: 9.72 minutes. 

Date: 05/2008; 
Average wait time, Tier 1, DME: 8.87 minutes; 
Average wait time, Tier 2, DME: 9.28 minutes. 

Date: 06/2008; 
Average wait time, Tier 1, DME: 6.83 minutes; 
Average wait time, Tier 2, DME: 5.43 minutes. 

Date: 07/2008; 
Average wait time, Tier 1, DME: 4.88 minutes; 
Average wait time, Tier 2, DME: 3.13 minutes. 

Source: CMS monthly data. 

[End of figure] 

Wait Times by Language of Caller: 

Figures 15 through 18 provide information on differences in average 
wait times experienced by callers depending on the language in which 
they are assisted. The information provided compares the wait times of 
English-and Spanish-speaking callers for each type of inquiry, by 
complexity level. 

Figure 15: Monthly Average Wait Times for Tier 1 CSRs Responding to 
English-and Spanish-speaking Callers' General Medicare Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, English-speaking callers: 4.88 minutes; 
Average wait time, Spanish-speaking callers: 5.03 minutes. 

Date: 10/2007; 
Average wait time, English-speaking callers: 4.95 minutes; 
Average wait time, Spanish-speaking callers: 5.62 minutes. 

Date: 11/2007; 
Average wait time, English-speaking callers: 6.23 minutes; 
Average wait time, Spanish-speaking callers: 8.22 minutes. 

Date: 12/2007; 
Average wait time, English-speaking callers: 6.53 minutes; 
Average wait time, Spanish-speaking callers: 7.28 minutes. 

Date: 01/2008; 
Average wait time, English-speaking callers: 6.7 minutes; 
Average wait time, Spanish-speaking callers: 6.2 minutes. 

Date: 02/2008; 
Average wait time, English-speaking callers: 6.37 minutes; 
Average wait time, Spanish-speaking callers: 6.68 minutes. 

Date: 03/2008; 
Average wait time, English-speaking callers: 7.65 minutes; 
Average wait time, Spanish-speaking callers: 8.4 minutes. 

Date: 04/2008; 
Average wait time, English-speaking callers: 7.68 minutes; 
Average wait time, Spanish-speaking callers: 8.1 minutes. 

Date: 05/2008; 
Average wait time, English-speaking callers: 7.57 minutes; 
Average wait time, Spanish-speaking callers: 8.12 minutes. 

Date: 06/2008; 
Average wait time, English-speaking callers: 8.67 minutes; 
Average wait time, Spanish-speaking callers: 9.05 minutes. 

Date: 07/2008; 
Average wait time, English-speaking callers: 5.8 minutes; 
Average wait time, Spanish-speaking callers: 5.95 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 16: Monthly Average Wait Times for Tier 2 CSRs Responding to 
English-and Spanish-speaking Callers' General Medicare Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, English-speaking callers: 13.57 minutes; 
Average wait time, Spanish-speaking callers: 14.3 minutes. 

Date: 10/2007; 
Average wait time, English-speaking callers: 7.62 minutes; 
Average wait time, Spanish-speaking callers: 8.0 minutes. 

Date: 11/2007; 
Average wait time, English-speaking callers: 6.55 minutes; 
Average wait time, Spanish-speaking callers: 7.65 minutes. 

Date: 12/2007; 
Average wait time, English-speaking callers: 6.23 minutes; 
Average wait time, Spanish-speaking callers: 6.37 minutes. 

Date: 01/2008; 
Average wait time, English-speaking callers: 7.42 minutes; 
Average wait time, Spanish-speaking callers: 9.05 minutes. 

Date: 02/2008; 
Average wait time, English-speaking callers: 7.33 minutes; 
Average wait time, Spanish-speaking callers: 9.23 minutes. 

Date: 03/2008; 
Average wait time, English-speaking callers: 8.93 minutes; 
Average wait time, Spanish-speaking callers: 10.13 minutes. 

Date: 04/2008; 
Average wait time, English-speaking callers: 8.87 minutes; 
Average wait time, Spanish-speaking callers: 9.8 minutes. 

Date: 05/2008; 
Average wait time, English-speaking callers: 8.85 minutes; 
Average wait time, Spanish-speaking callers: 9.25 minutes. 

Date: 06/2008; 
Average wait time, English-speaking callers: 7.05 minutes; 
Average wait time, Spanish-speaking callers: 4.52 minutes. 

Date: 07/2008; 
Average wait time, English-speaking callers: 5.03 minutes; 
Average wait time, Spanish-speaking callers: 2.88 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 17: Monthly Average Wait Times for Tier 1 CSRs Responding to 
English-and Spanish-speaking Callers' Claims Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, English-speaking callers: 18.53 minutes; 
Average wait time, Spanish-speaking callers: 9.8 minutes. 

Date: 10/2007; 
Average wait time, English-speaking callers: 11.35 minutes; 
Average wait time, Spanish-speaking callers: 4.77 minutes. 

Date: 11/2007; 
Average wait time, English-speaking callers: 7.37 minutes; 
Average wait time, Spanish-speaking callers: 5.45 minutes. 

Date: 12/2007; 
Average wait time, English-speaking callers: 6.7 minutes; 
Average wait time, Spanish-speaking callers: 6.65 minutes. 

Date: 01/2008; 
Average wait time, English-speaking callers: 7.3 minutes; 
Average wait time, Spanish-speaking callers: 8.4 minutes. 

Date: 02/2008; 
Average wait time, English-speaking callers: 7.15 minutes; 
Average wait time, Spanish-speaking callers: 8.63 minutes. 

Date: 03/2008; 
Average wait time, English-speaking callers: 8.75 minutes; 
Average wait time, Spanish-speaking callers: 10 minutes. 

Date: 04/2008; 
Average wait time, English-speaking callers: 8.62 minutes; 
Average wait time, Spanish-speaking callers: 9.47 minutes. 

Date: 05/2008; 
Average wait time, English-speaking callers: 8.8 minutes; 
Average wait time, Spanish-speaking callers: 9.15 minutes. 

Date: 06/2008; 
Average wait time, English-speaking callers: 8.92 minutes; 
Average wait time, Spanish-speaking callers: 7.12 minutes. 

Date: 07/2008; 
Average wait time, English-speaking callers: 6.35 minutes; 
Average wait time, Spanish-speaking callers: 5 minutes. 

Source: CMS monthly data. 

[End of figure] 

Figure 18: Monthly Average Wait Times for Tier 2 CSRs Responding to 
English-and Spanish-speaking Callers' Claims Inquiries: 

[Refer to PDF for image] 

This figure is a multiple line graph depicting the following data: 

Date: 09/2007; 
Average wait time, English-speaking callers: 13.83 minutes; 
Average wait time, Spanish-speaking callers: 13.12 minutes. 

Date: 10/2007; 
Average wait time, English-speaking callers: 10.26 minutes; 
Average wait time, Spanish-speaking callers: 10.8 minutes. 

Date: 11/2007; 
Average wait time, English-speaking callers: 5.53 minutes; 
Average wait time, Spanish-speaking callers: 9.88 minutes. 

Date: 12/2007; 
Average wait time, English-speaking callers: 5.1 minutes; 
Average wait time, Spanish-speaking callers: 6.72 minutes. 

Date: 01/2008; 
Average wait time, English-speaking callers: 7.98 minutes; 
Average wait time, Spanish-speaking callers: 9.6 minutes. 

Date: 02/2008; 
Average wait time, English-speaking callers: 8.1 minutes; 
Average wait time, Spanish-speaking callers: 10.9 minutes. 

Date: 03/2008; 
Average wait time, English-speaking callers: 9.5 minutes; 
Average wait time, Spanish-speaking callers: 11.7 minutes. 

Date: 04/2008; 
Average wait time, English-speaking callers: 9.17 minutes; 
Average wait time, Spanish-speaking callers: 11.15 minutes. 

Date: 05/2008; 
Average wait time, English-speaking callers: 8.87 minutes; 
Average wait time, Spanish-speaking callers: 9.9 minutes. 

Date: 06/2008; 
Average wait time, English-speaking callers: 8.55 minutes; 
Average wait time, Spanish-speaking callers: 10.53 minutes. 

Date: 07/2008; 
Average wait time, English-speaking callers: 6.25 minutes; 
Average wait time, Spanish-speaking callers: 7.8 minutes. 

Source: CMS monthly data. 

[End of figure] 

[End of section] 

Appendix III: Comments from the Centers for Medicare & Medicaid 
Services: 

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

December 9, 2008: 

Kathleen M. King: 
Director, Health Care: 
U.S. Government Accountability Office: 
441 G Street N.W. 
Washington, DC 20548: 

Dear Ms. King: 

Enclosed are comments on the U.S. Government Accountability Office's 
(GAO) report entitled: "Medicare: Callers Can Access 1-800-MEDICARE 
Services, but Responsibility Within CMS for Limited English Proficiency 
Plan Unclear" (GAO 09-104). 

The Department appreciates the opportunity to review this report before 
its publication. 

Sincerely, 

Signed by: 

Vincent J. Ventimiglia, Jr. 
Assistant Secretary or Legislation: 

Attachment: 

Department Of Health To Human Services:	
Centers for Medicare & Medicaid Services: 
200 Independence Avenue SW: 
Washington, DC 20201: 

Date: December 9, 2008: 

To: Vincent J. Ventimiglia, Jr. 
Assistant Secretary for Legislation: 
Department of Health and Human Services: 

From: [Signed by] Kerry Weems: 
Acting Administrator: 

Subject: Government Accountability Office (GAO) Draft Correspondence: 
Medicare: Callers Can Access 1-800-MEDICARE Services, but 
Responsibility Within CMS for Limited English Proficiency Plan Unclear" 
(GAO 09-104): 

Thank you for the opportunity to review and comment on the GAO Draft 
correspondence, "MEDICARE: Callers Can Access 1-800-MEDICARE Services, 
but Responsibility within CMS for Limited English Proficiency Plan 
Unclear" (GAO-09-104). We are very pleased to see the results confirm 
our belief that we provide a quality service to callers, that our 
contract oversight is effective and that we service all callers 
appropriately, regardless of the caller's disability or language 
preference. In this draft correspondence, the GAO described (I) the 
extent to which performance standards and targets designed by the 
Centers for Medicare & Medicaid (CMS) to ensure caller access have been 
met by the current I-800-MEDICARE contractor and the extent to which 
the time a caller waits to reach a customer service representative(CSR) 
has changed; (2) efforts by CMS to provide limited English proficiency 
(LEP) callers access to 1-800-MEDICARE services and the wait times 
experienced by these callers, and (3) CMS oversight of callers' access 
to 1-800-MEDICARE. and the accuracy of the information provided. 

We appreciate the effort GAO put into reviewing our processes, 
procedures and the large amounts of data associated with operating the 
I-800-MEDICARE helpline. It is our goal to provide timely and accurate 
information to a population comprised of the aged, the disabled, their 
families and caregivers. We appreciate the professionalism and courtesy 
extended to CMS by the auditors who analyzed an enormous amount of 
contractual and operational data to reach their finding. We understand 
that this was a detailed audit requiring extensive research and 
coordination between CMS and GAO and we believe that both entities 
worked collaboratively and transparently as a team to ensure that all 
opportunities for improvement were identified. 

Below is our response to the GAO recommendation as well as additional 
comments. 

GAO Recommendation: 

To ensure CMS offices, including those that oversee the operation of 
the 1-800-MEDICARE help line, are aware of, and take steps consistent 
with, the Department of Health and Human Services' (HHS) LEP Plan when 
considering the needs of people with LEP, CMS should designate an 
official or office with responsibility for managing the LEP Plan. 

CMS Response: 

As to the specific recommendation for appointing an official 
responsible for managing LEP programs, CMS has identified Tasha 
Richburg in the Office of Equal Opportunity and Civil Rights (OEOCR) as 
the officer responsible for the development of the CMS LEP Plan. OEOCR 
will coordinate with the HHS Office for Civil Rights (OCR) to develop 
the CMS LEP Plan. The draft CMS LEP Plan will be vetted throughout the 
Agency before the final draft is submitted to OCR for approval. Once 
approved, the CMS LEP plan will serve as the internal control measure 
that will clearly define the area of responsibility used to ensure 
consistent and reliable LEP tracking and reporting. 

Other Comments: 

1) Forecasting Targets - We are pleased that GAO recognized the 
considerable improvement already made in this area and we are confident 
that as we gain more experience with our changed environment, we will 
continue to improve our forecasting model. We would like to note that 
during the months associated with this report, there were major changes 
occurring to our program and process. We were transitioning the fee-for-
service (FFS) contractors' claims calls into the Beneficiary Call 
Center (BCC) as well as transitioning from the 1-800-MEDICARE contract 
to the BCC contract. With all of the changes occurring, it took a 
longer time than we anticipated to fully understand the effects on 
forecasting call volumes. 

2) Average Speed of Answer (ASA) differences by line of business - GAO 
noted a significant difference between the ASA of the claims lines of 
business and General Medicare line of business for September and 
October 2007. It is important to note that the transition of FFS 
contractors' claims lines of business into the BCC was completed in 
September 2007, a full year ahead of schedule. We are pleased that GAO 
recognized a marked service improvement only I month after the end of 
transition of the claims volumes into the BCC. It is also important and 
most notable that this marked improvement was accomplished during our 
annual enrollment period, the highest call volume months of the year. 

3) Average Handle Time (AHT) -AHT is a performance indicator in the 
contract. The AHT for a call is used by the contractor to project the 
number of CSRs needed to handle the anticipated call volume. CMS makes 
an effort to be as efficient as possible with calls and did target many 
areas for improvement including a reduction in AHT. However, CMS' 
guidance to the contractor and the CSRs was and continues to be that 
they shall fully answer questions and not just work to reduce AHT. 
Contrary to some industry practices, we have made the business decision 
that fully answering a caller's questions is paramount and while we 
obviously need to manage AHT it is not a metric we attempt to drive 
down aggressively from an operational standpoint. 

4) We are pleased that the GAO acknowledged that we do service all 
callers to the best of our ability regardless of their language 
preference. While it is true that we have not specifically evaluated 
the number or proportion of Medicare-eligible or Medicare-enrolled LEP 
beneficiaries when planning how language services will be provided, we 
have gathered data on an ongoing basis related to language use for 
planning purposes and the result is that the next largest number of LEP 
calls received at the BCC (after English and Spanish) is Russian with a 
rate of 0.1 percent. While it would not be financially or operationally 
prudent to add another standard language to our CSR base to address a 
call volume of less than one tenth of one percent, we will continue to 
evaluate incoming call patterns and look for opportunities to improve 
the service we provide the beneficiaries. In addition, we are proud of 
our ability to service over 70 languages via our language line and 
believe that we go well beyond the basic requirements in our efforts to 
fully service LEP beneficiaries. 

5) There are several references in the GAO report that contain Vangent 
proprietary information that should be removed from the report before 
publication. These include: 

* Specific references to the exact award fee amounts. The Vangent 
evaluations and award fee scores are considered proprietary. 

* The exact amount of fee. This is considered proprietary, as it is 
part of the bid strategy when the contract is competed. 

* Detailed number of CSRs as of January 1 and specifically the number 
of tier 2 CSRs. This is considered proprietary as competitors can use 
these detailed data to reverse engineer the Vangent rates. 

Once again, we appreciate the efforts of the GAO and the 
professionalism exhibited by the staff responsible for this study. We 
are committed to improving our service wherever possible and we will 
continue to work in partnership and keep you apprised as we implement 
the Report's recommendation. 

[End of section] 

Appendix IV: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Kathleen M. King, (202) 512-7114 or kingk@gao.gov: 

Acknowledgments: 

In addition to the person named above, Karen Doran, Assistant Director; 
Jennie F. Apter; Hernan Bozzolo; Eleanor M. Cambridge; Emily R. Gamble 
Gardiner; Barbara A. Hills; Martha R.W. Kelly; Ba Lin; Lisa S. Rogers; 
and Hemi Tewarson made key contributions to this report. 

[End of section] 

Related GAO Products: 

Medicare-Related Reports: 

Medicare: Communications to Beneficiaries on the Prescription Drug 
Benefit Could Be Improved. [hyperlink, 
http://www.gao.gov/products/GAO-06-654]. Washington, D.C.: May 3, 2006. 

Medicare: Accuracy of Responses from the 1-800-MEDICARE Help Line 
Should Be Improved. [hyperlink, 
http://www.gao.gov/products/GAO-05-130]. Washington, D.C.: December 8, 
2004. 

Medicare: Call Centers Need to Improve Responses to Policy-Oriented 
Questions from Providers. [hyperlink,
http://www.gao.gov/products/GAO-04-669]. Washington, D.C.: July 16, 
2004. 

Federal Contact Center Reports: 

Tax Administration: 2007 Filing Season Continues Trend of Improvement, 
but Opportunities to Reduce Costs and Increase Tax Compliance Should be 
Evaluated. [hyperlink, http://www.gao.gov/products/GAO-08-38]. 
Washington, D.C.: November 15, 2007. 

Tax Administration: Most Filing Season Services Continue to Improve, 
but Opportunities Exist for Additional Savings. [hyperlink, 
http://www.gao.gov/products/GAO-07-27]. Washington, D.C.: November 15, 
2006. 

Federal Contact Centers: Mechanisms for Sharing Metrics and Oversight 
Practices along with Improved Data Needed. [hyperlink, 
http://www.gao.gov/products/GAO-06-270]. Washington, D.C.: February 8, 
2006. 

Tax Administration: IRS Improved Some Filing Season Services, but Long- 
term Goals Would Help Manage Strategic Trade-offs. [hyperlink, 
http://www.gao.gov/products/GAO-06-51]. Washington, D.C.: November 14, 
2005. 

Social Security Administration: Additional Actions Needed in Ongoing 
Efforts to Improve 800-Number Service. [hyperlink, 
http://www.gao.gov/products/GAO-05-735]. Washington, D.C.: August 8, 
2005. 

Immigration Services: Better Contracting Practices Needed at Call 
Centers. [hyperlink, http://www.gao.gov/products/GAO-05-526]. 
Washington, D.C.: June 30, 2005. 

Tax Administration: IRS Improved Performance in the 2004 Filing Season, 
But Better Data on the Quality of Some Services Are Needed. [hyperlink, 
http://www.gao.gov/products/GAO-05-67]. Washington, D.C.: November 15, 
2004. 

Tax Administration: IRS Needs to Further Refine Its Tax Filing Season 
Performance Measures. [hyperlink, 
http://www.gao.gov/products/GAO-03-143]. Washington, D.C.: November 22, 
2002. 

IRS Telephone Assistance: Limited Progress and Missed Opportunities to 
Analyze Performance in the 2001 Filing Season. [hyperlink, 
http://www.gao.gov/products/GAO-02-212]. Washington, D.C.: December 7, 
2001. 

Customer Service: Human Capital Management at Selected Public and 
Private Call Centers. [hyperlink, 
http://www.gao.gov/products/GAO/GGD-00-161]. Washington, D.C.: August 
8, 2000. 

[End of section] 

Footnotes: 

[1] The original Medicare fee-for-service (FFS) program has two parts: 
Part A, which covers hospital and other inpatient services, and Part B, 
which is optional insurance and covers certain physician, outpatient 
hospital, and other services for beneficiaries who pay monthly 
premiums. Part B also includes durable medical equipment (DME), such as 
wheelchairs. As an alternative to FFS, Medicare beneficiaries may 
obtain coverage from private health plans that participate in the 
Medicare Advantage program--known as Medicare Part C. Medicare 
beneficiaries receiving coverage under Medicare FFS or Medicare Part C 
generally also may obtain coverage for outpatient prescription drugs 
under Medicare Part D. Medicare Advantage plans also may include drug 
coverage for beneficiaries. 

[2] CMS also receives inquiries through the mail, e-mail, and its 
Medicare Web site. 

[3] Executive Order 13166, Improving Access to Services for Persons 
with Limited English Proficiency, 65 Fed. Reg. 50121-22 (Aug. 16, 
2000). HHS issued an LEP Plan in December 2000 that applied to each of 
its agencies and programs, including CMS. The executive order further 
states that these LEP plans are to be consistent with the standards set 
forth in the Department of Justice guidance, which defines four factors 
for agencies to consider when forming their agency plans. These factors 
are: (1) the number or proportion of LEP persons in the eligible 
service population; (2) the frequency with which LEP individuals come 
in contact with the program; (3) the importance of the services 
provided by the program; and (4) the resources available. 

[4] Pub. L. No. 108-173, § 101, 117 Stat. 2066, 2071-2152 (2003) 
(codified, as amended, at 42 U.S.C. §§ 1395w-101-1395w-152). 

[5] A transition period for phasing out operations of the second 
contractor occurred from October 30, 2006 through May 31, 2007. 

[6] The standards are quantitative measures used to assess the 
contractor's performance on a pass/fail basis. The targets are 
guidelines used to assess the contractor's performance on a scale of 
substandard to superior. 

[7] In this report, we define the current 1-800-MEDICARE contract as 
including the original task order awarded to a single contractor and 
the subsequent modifications to this task order. 

[8] The NDW is a repository for data related to 1-800-MEDICARE, 
including data on the number of calls received by 1-800-MEDICARE and 
the amount of time a caller waits to speak with a CSR. 

[9] July 2008 data reflect the most current data available at the time 
of our review. 

[10] Data related to the amount of time callers initially waited to 
speak with CSRs were not available prior to December 2005. 

[11] Prior to July 2007, CMS used measures primarily focused on 
evaluating the 1-800-MEDICARE contractor's performance during the 
transition of operations from two contractors to the single contractor. 

[12] Data identifying differences in the contractor's performance on 
these three elements were not available prior to September 2007. 

[13] Officials with the HHS Office for Civil Rights (OCR) and CMS 
Office of Equal Opportunity and Civil Rights (OEOCR) said their primary 
LEP focus was enforcement of requirements under Title VI of the Civil 
Rights Act of 1964, Pub. L. No. 88-352, § 601, 78 Stat. 241, 252-253 
(1964) (codified, as amended, at 42 U.S.C. §§ 2000d-2000d-7). Executive 
Order 13166 also requires federal agencies to prepare Title VI guidance 
for their recipients of federal financial assistance and HHS has issued 
this Title VI guidance. However, because Title VI requirements and 
related guidance do not apply to programs conducted by federal 
agencies, such as 1-800-MEDICARE, they were not included in the scope 
of our work. In addition to this enforcement role, HHS OCR is 
responsible for promoting compliance with and enforcement of other 
federal civil rights statutes that prohibit discrimination on the basis 
of race, color, national origin, age, disability, and, in some cases, 
sex and religion. CMS OEOCR also conducts other activities, including 
managing the agency's equal employment opportunity discrimination 
complaints and providing relevant information to CMS programs on civil 
rights issues. 

[14] A contact center refers to a customer service center that handles 
telephone and written inquiries. 

[15] See, for example, GAO, Federal Contact Centers: Mechanisms for 
Sharing Metrics and Oversight Practices along with Improved Data 
Needed, [hyperlink, http://www.gao.gov/products/GAO-06-270] 
(Washington, D.C.: Feb. 8, 2006). For more of the reports we reviewed 
to identify common practices, see the Federal Contact Center Reports 
section of the "Related GAO Products" page at the end of this report. 

[16] We interviewed officials from the Department of Defense's TRICARE 
Management Activity, the Department of Education's Office of Federal 
Student Aid, the Department of the Treasury's Internal Revenue Service, 
and the Social Security Administration. 

[17] Brad Cleveland. Call Center Management on Fast Forward. Maryland: 
ICMI Press, 2006. 

[18] A sixth contact center site is available to accommodate callers 
during periods of increased call volume. 

[19] The BBA established Medicare+Choice--now referred to as Medicare 
Advantage or Medicare Part C--and required that the Secretary of HHS 
maintain a toll-free number for inquiries on the new program. Pub. L. 
No. 105-33, § 4001, 111 Stat. 251, 275-327 (1997) (codified, as 
amended, at 42 U.S.C. §§ 1395w-21, et seq.) 

[20] Pub. L. No. 108-173, §§ 101, 911, 117 Stat. 2066, 2071-2152, 2378- 
2386 (2003) (codified, as amended, at 42 U.S.C. §§ 1395w-101-1395w-152, 
1395kk-1). The MMA allowed the Secretary of HHS to expand certain 
responsibilities of Medicare contractors, including with respect to 
Medicare Parts A and B claims and beneficiary and provider 
communications. These Medicare contractors are required to maintain a 
toll-free line where beneficiaries, providers, and suppliers may obtain 
information regarding billing, coding, claims, coverage, and other 
relevant information relating to Parts A and B. 

[21] CSRs refer callers with specific inquiries about amounts owed for 
their deductibles or co-payments related to Medicare's Part C or Part D 
to the plan in which the beneficiary is enrolled because CSRs do not 
have access to this information. 

[22] Prior work by the HHS Office of the Inspector General reported on 
concerns raised by callers about difficulties in using the 1-800- 
MEDICARE IVR. See Department of Health and Human Services, Office of 
Inspector General, 1-800-MEDICARE: Caller Satisfaction And Experiences, 
OEI-07-06-00530 (Washington, D.C.: September 2007). 

[23] CSRs do not have access to information about amounts owed for 
callers' deductibles or copayments for Medicare's Part C or Part D 
plans. 

[24] Events that affect volume during this period include the annual 
mailing of the Medicare & You handbook and annual beneficiary 
enrollment in Medicare Advantage plans and prescription drug plans. 

[25] The 2,600 CSRs were employed as of January 1, 2008, by the 1-800- 
MEDICARE contractor. 

[26] As of January 1, 2008, CSR staffing also included Tier 2 Senior 
Representative CSRs who assist Tier 1 and Tier 2 CSRs. Additional CSRs 
also are dedicated to responding to written and e-mail correspondence. 
These CSRs are also trained and available to respond to calls during 
periods of high call volume. 

[27] CSRs use scripts to the extent possible when answering claims 
inquiries, though some claims information (e.g., specific claims 
amounts) may vary by caller. 

[28] The Medicare Web site is [hyperlink, http://www.medicare.gov]. 

[29] Prior GAO work has raised concerns about the ability of both CSRs 
and callers to understand scripts for commonly asked questions. See 
GAO, Medicare: Accuracy of Responses from the 1-800-MEDICARE Help Line 
Should Be Improved, [hyperlink, http://www.gao.gov/products/GAO-05-130] 
(Washington, D.C.: Dec. 8, 2004). 

[30] Performance-based contracts specify the desired outcomes and allow 
the contractors to determine how best to achieve those outcomes, rather 
than prescribe the methods to be used. Performance-based contracts can 
encourage contractors to be innovative and to find cost-effective ways 
of delivering services. See [hyperlink, 
http://www.gao.gov/products/GAO-06-270]. 

[31] CMS has exercised its option to extend this contract for two 
additional years. Accordingly, the current 1-800-MEDICARE contract 
extends from October 30, 2006, through May 31, 2009. 

[32] Prior to July 2007, CMS used measures primarily focused on 
evaluating the 1-800-MEDICARE contractor's performance during the 
transition of operations from two contractors to the single contractor. 

[33] CMS has categorized the performance metrics into two "bundles," 
one related to telephone access and another related to other 
communication channels, such as responses to e-mail and written 
inquiries. 

[34] The element related to program management and communication 
accounts for 20 percent of the award fee while the elements related to 
contract compliance and fiscal responsibility account for 10 percent 
each. We considered the program management and communication indicators 
related to access and accuracy as part of our review of CMS oversight 
of 1-800-MEDICARE; however, the performance indicators for contract 
compliance and fiscal responsibility were not related to access or 
oversight of access and accuracy and therefore were considered outside 
the scope of our review. 

[35] CMS awarded the TQC contract in December 2007. According to CMS 
officials, elements of the TQC contract were continuing to be 
implemented during the period encompassed in this review. 

[36] We analyzed metrics for incoming calls only. CMS has also 
developed performance metrics related to services other than incoming 
calls, including web chats and responding to e-mails. 

[37] The standard for one metric, average wait time, was modified in 
August 2008. All data in this report are through July 2008. 

[38] This excludes June 2008, the month in which the contractor 
received a waiver for the average wait time metric and standard. 

[39] Medicare beneficiaries generally may change or disenroll from 
either Medicare Advantage plans or prescription drug plans during the 
annual coordinated election period, which begins on November 15 and 
ends on December 31 of each year. Beneficiaries may also change or 
disenroll from most Medicare Advantage plans once during the open 
enrollment period, which begins on January 1 and ends on March 31 of 
each year. 

[40] An industry expert we interviewed also noted that CMS's average 
wait time performance standard was relatively long, although 
performance standards are usually based on the level of service an 
agency can afford to purchase with its contact center budget. 

[41] The other two federal agencies we interviewed used a performance 
metric defined as the percentage of calls answered within a designated 
timeframe, with a standard of answering 80 percent or more calls within 
20 seconds. 

[42] The calculation of this metric includes callbacks, an option in 
which basic caller information is taken and the caller is contacted at 
a later time. CMS may exercise this option when call volumes cause 
average wait times to reach a threshold of 20 minutes. Two types of 
callbacks can occur to help reduce average wait times: (1) CMS can 
program the IVR to provide callers with the option to receive a 
callback from a CSR within 48 hours or (2) the 1-800-MEDICARE 
contractor's CSRs can offer to call back callers within 48 hours. 

[43] As wait times to reach a CSR increase, callers may be more likely 
to abandon calls. 

[44] This measure also includes any time a CSR spends to complete work 
related to a call after the call is completed. 

[45] CMS's current 1-800-MEDICARE contract specifies about 20 
performance indicators and associated targets; we reviewed only 
indicators related to telephone access. 

[46] Effective August 2008, CMS revised its average wait time standard 
to a new minimum of 1 minute. 

[47] Because callers who speak neither English nor Spanish are assisted 
by English-speaking CSRs using a telephone interpretation service, data 
for this population are included with data for English-speaking 
callers. 

[48] The language line certifies their interpreters and is affiliated 
with multiple professional interpretation and translation associations, 
such as the American Translators Association. 

[49] In just over three-quarters of the months between the end of 
October 2006 and July 2008 where the average wait time was outside the 
performance standard range, the wait times were less than 5 minutes. 

[50] Services have been available in Spanish since 1-800-MEDICARE began 
in 1999, pre-dating the HHS LEP Plan and the current contract. Services 
for callers who speak neither English nor Spanish were added in 2003. 

[51] CMS officials provided the 1-800-MEDICARE contractor with a 
limited number of LEP information resources from HHS and other federal 
agencies related to LEP populations, such as a language planning and 
self-assessment tool developed by the Department of Justice. However, 
they noted that were not aware of the extent of information about 
services for LEP callers available from other HHS and CMS offices. 

[52] Officials from both HHS's OCR and CMS's OEOCR stated that their 
offices do not focus on LEP issues within federally conducted programs, 
such as 1-800-MEDICARE. 

[53] Standards for internal control in the federal government state 
that agency management is responsible for establishing and maintaining 
a control environment that sets a positive attitude toward internal 
control and conscientious management, including an organizational 
structure with clearly defined areas of authority and responsibility. 
See GAO, Standards for Internal Control in the Federal Government, 
[hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] 
(Washington, D.C.: November 1999) and GAO, Internal Control Management 
and Evaluation Tool, GAO-01-1008G (Washington, D.C.: August 2001). 

[54] Because OBIS officials said that they were not aware of the HHS 
LEP Plan, we did not conduct an exhaustive review of all 1-800-MEDICARE 
activities that may relate to elements of the Plan. 

[55] Because 1-800-MEDICARE does not provide services in-person, the 
plan element regarding face-to-face oral language assistance is not 
applicable. 

[56] While no office or official specific to language access complaints 
has been identified, callers can register general complaints about 
services provided by 1-800-MEDICARE through the line. 

[57] While no CMS policies and procedures exist related to the Plan, 
the 1-800-MEDICARE and TQC contractors have developed standard 
operating procedures--as required by their contracts with CMS--which 
include procedures related to LEP callers. 

[58] CMS also has mechanisms to oversee contract elements not related 
to access and accuracy, such as security issues and financial 
oversight. The scope of our review did not include an examination of 
these mechanisms. 

[59] We analyzed metrics and indicators for incoming calls only. CMS 
has also developed performance metrics related to services other than 
incoming calls, including CSR responses to Web chat and to e-mails. 

[60] The first evaluation period encompassed the entire 7-month 
contract transition period and, as such, was longer than the standard 4-
month period and did not use the current contract performance metrics 
and indicators. It ran from October 30, 2006 to May 31, 2007. 

[61] If the contractor's performance is rated as substandard or 
satisfactory, the contractor receives no award fee. If its performance 
is rated as good, the contractor can receive between 0 and 50 percent 
of the award fee. If the contractor's performance is rated as very good 
it can receive between 51 and 90 percent of the award fee and if its 
performance is rated as superior it can receive between 91 and 100 
percent of the award fee. 

[62] The TQC customer satisfaction survey begins after a caller 
completes his or her inquiry and allows the caller to provide feedback 
about the IVR system, their interaction with the CSR, or both, 
depending on the services the caller accessed. 

[63] The HHS Office of the Inspector General recommended that CMS 
reevaluate their investment in the IVR due to low rates of customer 
inquiries resolved with the technology. See OEI-07-06-00530. 

[64] In prior work related to contact centers, GAO has noted that the 
information CSRs use to assist callers should be electronically 
searchable so that CSRs can expeditiously provide correct and complete 
responses. In addition, GAO has noted that 1-800-MEDICARE CSRs and 
callers have had difficulty understanding scripts, and recommended that 
scripts be revised as needed and be pretested to ensure that CSRs can 
effectively use them to accurately answer callers' inquiries. See GAO, 
Medicare: Call Centers Need to Improve Responses to Policy-Oriented 
Questions from Providers, [hyperlink, 
http://www.gao.gov/products/GAO-04-669] (Washington, D.C.: July 16, 
2004) and [hyperlink, http://www.gao.gov/products/GAO-05-130]. 

[65] This initiative is a part of CMS's response to a recommendation 
that CMS should ensure callers' inquiries are fully answered during 
their first call to 1-800-MEDICARE, made by the HHS Office of the 
Inspector General. See OEI-07-06-00530. 

[66] The evaluation period starting October 1, 2008 will also coincide 
with the anticipated increased call volume associated with Medicare's 
annual coordinated election and open enrollment periods. 

[67] An industry expert indicated that call volume forecasts should not 
differ from actual call volume more than 25 percent to be worthwhile in 
producing the forecasts. 

[68] All tables within this appendix reflect the average wait time 
standard of between 5 minutes and 8 minutes, 30 seconds required by 
CMS's contract with the 1-800-MEDICARE contractor as of July 2008. 
Effective August 2008, CMS lowered its average wait time standard to 
between 1 minute and 5 minutes. 

[69] Medicare guidance defines DME as equipment that serves a medical 
purpose, can withstand repeated use, is generally not useful in the 
absence of an illness or injury, and is appropriate for use in the 
home. DME includes items such as wheelchairs, hospital beds, and 
walkers. CMS tracks information on calls related to DME separately from 
Part B calls. 

[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 Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, 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 Web site, 
[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. 

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

Contact: 

Web site: [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: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 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, D.C. 20548: