[Senate Hearing 114-557]
[From the U.S. Government Publishing Office]
S. Hrg. 114-557
HEALTHCARE.GOV: A REVIEW OF OPERATIONS AND ENROLLMENT
=======================================================================
HEARING
before the
COMMITTEE ON FINANCE
UNITED STATES SENATE
ONE HUNDRED FOURTEENTH CONGRESS
SECOND SESSION
__________
MARCH 17, 2016
__________
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Printed for the use of the Committee on Finance
______
U.S. GOVERNMENT PUBLISHING OFFICE
24-057-PDF WASHINGTON : 2017
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Publishing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800;
DC area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001
COMMITTEE ON FINANCE
ORRIN G. HATCH, Utah, Chairman
CHUCK GRASSLEY, Iowa RON WYDEN, Oregon
MIKE CRAPO, Idaho CHARLES E. SCHUMER, New York
PAT ROBERTS, Kansas DEBBIE STABENOW, Michigan
MICHAEL B. ENZI, Wyoming MARIA CANTWELL, Washington
JOHN CORNYN, Texas BILL NELSON, Florida
JOHN THUNE, South Dakota ROBERT MENENDEZ, New Jersey
RICHARD BURR, North Carolina THOMAS R. CARPER, Delaware
JOHNNY ISAKSON, Georgia BENJAMIN L. CARDIN, Maryland
ROB PORTMAN, Ohio SHERROD BROWN, Ohio
PATRICK J. TOOMEY, Pennsylvania MICHAEL F. BENNET, Colorado
DANIEL COATS, Indiana ROBERT P. CASEY, Jr., Pennsylvania
DEAN HELLER, Nevada MARK R. WARNER, Virginia
TIM SCOTT, South Carolina
Chris Campbell, Staff Director
Joshua Sheinkman, Democratic Staff Director
(ii)
C O N T E N T S
----------
OPENING STATEMENTS
Page
Hatch, Hon. Orrin G., a U.S. Senator from Utah, chairman,
Committee on Finance........................................... 1
Wyden, Hon. Ron, a U.S. Senator from Oregon...................... 3
WITNESSES
Bliss, Erin, Assistant Inspector General for Evaluation and
Inspections, Office of Inspector General, Department of Health
and Human Services, Washington, DC............................. 6
Bagdoyan, Seto J., Director of Audits, Forensic Audits and
Investigative Service, Government Accountability Office,
Washington, DC................................................. 8
ALPHABETICAL LISTING AND APPENDIX MATERIAL
Bagdoyan, Seto J.:
Testimony.................................................... 8
Prepared statement........................................... 21
Responses to questions from committee members................ 27
Bliss, Erin:
Testimony.................................................... 6
Prepared statement........................................... 31
Responses to questions from committee members................ 37
Hatch, Hon. Orrin G.:
Opening statement............................................ 1
Prepared statement........................................... 43
Wyden, Hon. Ron:
Opening statement............................................ 3
Prepared statement........................................... 45
(iii)
HEALTHCARE.GOV: A REVIEW OF OPERATIONS AND ENROLLMENT
----------
THURSDAY, MARCH 17, 2016
U.S. Senate,
Committee on Finance,
Washington, DC.
The hearing was convened, pursuant to notice, at 10:03
a.m., in room SD-215, Dirksen Senate Office Building, Hon.
Orrin G. Hatch (chairman of the committee) presiding.
Present: Senators Burr, Isakson, Coats, Heller, Scott,
Wyden, Stabenow, Brown, and Bennet.
Also present: Republican Staff: Chris Campbell, Staff
Director; Christopher Armstrong, Deputy Chief Oversight
Counsel; Kimberly Brandt, Chief Health-care Investigative
Counsel; and Jill Wright, Detailee. Democratic Staff: Joshua
Sheinkman, Staff Director; David Berick, Chief Investigator;
Elizabeth Jurinka, Chief Health Advisor; and Juan Machado,
Professional Staff Member.
OPENING STATEMENT OF HON. ORRIN G. HATCH, A U.S. SENATOR FROM
UTAH, CHAIRMAN, COMMITTEE ON FINANCE
The Chairman. The committee will come to order. It has been
a little bit disruptive here this morning. We have a lot on our
plate. It is a pleasure to welcome everybody here this morning.
Today, we will be talking with representatives from the
Office of Inspector General for the Department of Health and
Human Services and from the Government Accountability Office
about their ongoing oversight work with respect to
HealthCare.gov and enrollment in the Federal health insurance
marketplace. I want to thank both entities for their hard work
on these issues and acknowledge the contributions both have
made to help this committee perform more accurate and timely
oversight.
Now, it is no secret that I have never been a fan of the
so-called Affordable Care Act, and, as we approach the sixth
anniversary of this law and look closely into how it is working
and being implemented, the evidence overwhelmingly shows that I
and the many others who opposed the law from the beginning have
been right all along.
The facts speak for themselves. Since Obamacare was signed
into law, HHS/OIG and GAO have cumulatively released at least
six dozen reports detailing various operation and
implementation issues, demonstrating the numerous areas where
the law is falling short. These reports are specific and
focused on key operational failures, like enrollment controls
or system issues, some of which we will hear more about today.
Let us keep in mind that GAO and HHS/OIG are not partisan
entities. They are independent watchdogs tasked with the
responsibility of objectively and dispassionately assessing
what is and what is not working in various Federal programs,
including those created or amended by the Affordable Care Act.
And there is no better record showing how this happened than
the reports we received from these offices.
Today, we are going to specifically discuss operations
issues related to HealthCare.gov and enrollment problems at the
Federal insurance marketplace, otherwise known as the Federal
exchange.
Let us start with the HealthCare.gov launch. As a result of
numerous problems and shortcuts taken with the initial
development and deployment of HealthCare.gov and its supporting
systems, consumers encountered widespread performance issues
when trying to create accounts and enroll in health plans.
After numerous inquiries and reports, we now know what
ultimately caused these performance issues. For example, there
was inadequate capacity planning. The Centers for Medicare and
Medicaid Services, CMS, cut corners and did not plan for
adequate capacity to maintain HealthCare.gov and its supporting
systems.
There were also problems with the software that were
entirely avoidable. CMS and its contractors identified errors
in the software coding for the website, but did not adequately
correct them prior to the launch, and we saw a lack of
functionality as CMS did not adequately prepare the necessary
systems and functions of the website and its supporting systems
prior to the initial launch.
CMS also failed to apply recognized best practices for
system development, which contributed to the problem.
Admittedly, since the initial launch, CMS has taken steps to
address these problems, including increasing capacity,
requiring additional software quality reviews, and awarding a
new contract to complete development and improve the
functionality of key systems. However, many of the problems
have still not been entirely resolved and continue to cause
frustration, especially for consumers trying to obtain health
insurance.
I wish we could boil down all of Obamacare's problems to
the functions of a single website. Indeed, if this was just an
IT problem, all of our jobs would be a lot easier. However, the
problems with Obamacare, and the Federal insurance marketplace
in particular, go much deeper, and many of them remain
unaddressed.
We know, for example, that the enrollment controls for the
Federal marketplace have been inadequate. During undercover
testing by GAO, the Federal marketplace approved insurance
coverage with taxpayer-funded subsidies for 11 out of 12
fictitious phone or online applicants. In 2014, the GAO
applicants--which, once again, were fake, made-up people--
obtained a total of about $30,000 in annual advanced premium
tax credits, plus eligibility for lower insurance costs at the
time of the service. These fictitious enrollees maintained
subsidized coverage throughout the year even though GAO sent
either clearly fabricated documents or no documents at all to
resolve the application inconsistencies.
While the subsidies, including those granted to GAO's
fictitious applicants, are paid to health-care insurers, they
nevertheless represent a benefit to consumers and a cost to the
government.
Now, GAO did find that CMS relies on a contractor charged
with document processing to basically uncover and report
possible instances of fraud. Yet, GAO also found that the
agency does not require that the contractor have any fraud
detection capability.
According to GAO, CMS has not performed a single
comprehensive fraud risk assessment--the recommended best
practice--of the Obamacare enrollment and eligibility process.
Until such assessment is completed, CMS is unlikely to know
whether existing control activities are suitably designed and
implemented to reduce inherent fraud risk to an acceptable
level. In other words, CMS is not even sure if CMS's fraud
prevention systems are designed correctly or if they are
effective.
Lastly, while it is not the focus of the reports that will
be covered by the testimony today, another matter we have been
tracking closely and where the GAO is issuing a report today is
CMS's oversight of the health care CO-OPs. We had a hearing on
this topic in late January, where we examined a number of
financial and oversight-related explanations for the abject
failure of the CO-OP programs.
Today's GAO report describes CMS's efforts to deal with
financial or operations issues at the CO-OPs, including the use
of an escalation plan for CO-OPs with serious problems that may
require corrective actions or enhanced oversight.
As of November 2015, 18 CO-OPs had enough problems that
they had to submit to a CMS escalation plan, including nine
that have discontinued operation. And just last week, we heard
that yet another CO-OP, this time the one in Maine, is on the
verge of financial insolvency, despite the fact that it had
been on a CMS-
mandated escalation plan.
In other words, CMS's efforts to address all the problems
faced by CO-OPs appear to have failed, just like virtually
every other element of this program. The failure of CMS to
adequately implement the CO-OP program is well-documented here
on the Finance Committee and elsewhere. As with so many other
parts of Obamacare, the high-minded rhetoric surrounding this
program has fallen short of reality.
With nearly half of the CO-OPs now closed, the failed
experiment has wasted taxpayer dollars and forced patients and
families to scramble for new insurance. With so many CO-OPs now
in financial jeopardy, I believe that CMS should work with and
not against States to safeguard taxpayer dollars.
So as always, we have a lot to discuss, and I look forward
to hearing more from the officials we have testifying here
today.
So with that, I will turn to Senator Wyden for his opening
remarks.
[The prepared statement of Chairman Hatch appears in the
appendix.]
OPENING STATEMENT OF HON. RON WYDEN,
A U.S. SENATOR FROM OREGON
Senator Wyden. Thank you, Mr. Chairman.
Mr. Chairman and colleagues, it is old news that the
initial rollout of HealthCare.gov 3 years ago was botched. It
is new news that the Inspector General of the Health and Human
Services Department recently said, and I want to quote here--
this is a quote, colleagues, from the Inspector General: ``CMS
recovered the HealthCare.gov website for high consumer use
within 2 months and adopted more effective organizational
practices.'' That is what the Inspector General said, that the
Department recovered the website for high consumer use within 2
months. That quote comes from one of two reports looking back
at 2013 and 2014 that the Finance Committee will be presented
with today.
I think we ought to start by recognizing that the story
here is well-documented. After the launch went badly, some of
the best minds in technology and a new contractor were brought
in. They scrambled to overhaul the system, and the exchange was
soon up and running, and the Centers for Medicare and Medicaid
Services are now following up on each of the Inspector
General's recommendations, which the Inspector General notes in
its report.
In the most recent enrollment period, nearly 10 million
Americans used HealthCare.gov to sign up for a plan or reenroll
automatically. In my home State, which had its own problems,
close to 150,000 people have used the website to sign up for a
plan as of January 31st. That is up by more than 30 percent
compared to last year.
The committee will also hear an update from the Government
Accountability Office on what has been called the secret
shopper investigation. The Government Accountability Office
first brought this study before the committee in July of last
year. I am going to repeat what I said back then.
On this side of the aisle, we do not take a back seat to
anybody in fighting fraud and protecting taxpayer dollars--$1
ripped off is $1 too many. But let us recognize that what was
true last summer remains true today.
This GAO investigation has not uncovered one single shred
of real-world fraud in the insurance marketplace. It was built
on fictitious characters with specially created identities, not
real consumers and not real fraudsters. It is true that the
Government Accountability Office found that there are sometimes
differences between the information on somebody's insurance
application and their tax forms and citizenship records. But
when it comes to these inconsistencies in people's data, this
investigation cannot differentiate between fraud and a typo.
Meanwhile, Health and Human Services does not look the
other way when it finds the red flags. In 2014, the year of
GAO's investigation, the Centers for Medicare and Medicaid
Services closed more than 100,000 insurance policies because
documents did not match or were not provided. Tax credits were
adjusted for nearly 100,000 households. In 2015, Health and
Human Services closed more policies and adjusted more tax
credits.
If you come at this from the left, you might say that is
too harsh. If you come at it from the right, you might take a
different view. But there is no basis whatsoever for the
argument that Health and Human Services ignores problems in
people's records or leaves the door open to fraud. It seems to
me, rather than rehashing old news, we ought to be looking at
the facts and talking in a bipartisan way about how to move
forward together.
Because of the Affordable Care Act, the number of Americans
without health insurance is at or near its lowest point in half
a century. For the 160 million people who get their insurance
from their employer, colleagues, premiums climbed 4 percent
last year. Let me repeat that. For 160 million people who get
their insurance from their employer, premiums climbed only 4
percent. Working-age Americans in Oregon and nationwide with
preexisting conditions--80 million people or more--can no
longer be denied insurance.
So, instead of battling out what happened 3 years ago, we
ought to be pulling on the same end of the rope and solving
some problems. For example, Democrats and Republicans ought to
be working together to look at ways in which we can provide
even more competition and bring costs down for consumers, and a
lot of you in this room have worked with me on that issue for
some time.
Second, there are going to be spectacular new cures in the
future, and there are real questions as to whether our health-
care system is going to be able to afford them. Here, Senator
Grassley has worked very closely with me to put together a
bipartisan case study, which looked at one blockbuster drug
involving hepatitis C. Solving the cost of these blockbuster
drugs is going to take a lot of hard work. It, again, can only
be done on a bipartisan basis.
Finally, I want to express my appreciation to colleagues on
both sides of the aisle, because I think we are on the cusp of
being able to make real progress on a huge opportunity for
older people in our country, and that is protecting the
Medicare guarantee, this very sacred guarantee we have for
seniors, while updating the program to look at the great new
challenge, which is chronic illness.
I want to thank Senator Bennet, who was out in front on
this issue for some time. He is not here, but Senator Isakson
and Senator Warner were champions as well. I want to express my
appreciation to the chairman for the progress that we are
making.
I have to make some comments with respect to something we
did not know about until about an hour ago, and that is this
matter of the CO-OPs.
What we have said is that we want to work in a bipartisan
way to improve a variety of sections of the Affordable Care
Act. Now, this new material on the CO-OPs, which neither I nor
anyone on this side knew anything about, was available
something like an hour ago. I intend to look at it with an eye
to what can be done on a bipartisan basis going forward.
But my work, and I think the work of colleagues here,
always ought to come back to this idea of making health-care
policy more accessible and more affordable. And for now--and I
certainly have not seen this report--I am not going to be
participating in any celebration of people suffering, because
the CO-OPs were tied up in a congressionally induced economic
straightjacket.
Thank you, Mr. Chairman.
[The prepared statement of Senator Wyden appears in the
appendix.]
The Chairman. Thank you, Senator.
I am going to introduce today's witnesses. Our first
witness is Ms. Erin Bliss, the Assistant Inspector General for
Evaluation and Inspections in the Office of Inspector General,
or OIG, at HHS.
Ms. Bliss has served in many roles at OIG since her career
began. I think your career began in 2000, if I have it
correctly. She started as an analyst for the Office of
Evaluation and Inspections and later went on to serve as a
senior advisor, where she provided management advice and expert
analysis to the Inspector General and other senior executives
on programmatic priorities and internal policies and
operations.
Afterwards, she worked from 2009 to 2014 as Director of
External Affairs at OIG and was responsible for overseeing and
implementing OIG's communication strategies and relationship
management with the administration, Congress, media, the
health-care industry and providers, and the public.
Ms. Bliss received her bachelor's degree in government from
the University of Notre Dame before receiving her master's
degree in public policy from the University of Chicago.
Our second witness is Mr. Seto Bagdoyan, the Director for
Audit Services in GAO's Forensics, Audits, and Investigative
Service Mission Team. During his GAO career, Mr. Bagdoyan has
served in a variety of positions, including as Legislative
Advisor in the Office of Congressional Relations and as
Assistant Director for Homeland Security in Justice.
Mr. Bagdoyan has also served on congressional details with
the Senate Finance Committee and the House Committee on
Homeland Security. We are glad to see you back here again.
Mr. Bagdoyan has also held a number of senior positions in
consultancies in the private sector, including most recently
focusing on political risk in homeland security.
Mr. Bagdoyan received his bachelor's degree in
international relations and economics from Claremont McKenna
College and an MBA in strategy from Pepperdine University.
I want to thank you both for coming. We will hear the
witness testimonies in the order that they were introduced.
Ms. Bliss, please proceed with your 5-minute statement.
STATEMENT OF ERIN BLISS, ASSISTANT INSPECTOR GENERAL FOR
EVALUATION AND INSPECTIONS, OFFICE OF INSPECTOR GENERAL,
DEPARTMENT OF HEALTH AND HUMAN SERVICES, WASHINGTON, DC
Ms. Bliss. Thank you. Good morning, Chairman Hatch and
other distinguished members of the committee. Thank you for the
opportunity to testify today about the Office of Inspector
General's case study which examines the management of
HealthCare.gov.
This is the website consumers use to apply for health
insurance through the Federal marketplace. As is well known, on
October 1, 2013, the HealthCare.gov website failed almost
immediately upon launch. Yet, within 2 months, CMS had
substantially improved the site's performance.
How did such a high priority project start so poorly, and
how did CMS turn the website around? Our case study provides
insights into these questions and lessons learned to help
HealthCare.gov and other Federal projects work better.
We believe that our assessment of the intersection of
technology, policy, and management can benefit a broad range of
Federal projects and programs. Our report chronicles the
breakdown and turnaround of HealthCare.gov over a 5-year
period. This morning I will summarize the highlights.
From the outset, the HealthCare.gov project faced a high
risk of failure. It was technically complex, with a fixed
deadline and many uncertainties. Still, HHS and CMS made many
missteps in its implementation. Most critical was the absence
of clear leadership and overall project responsibility, which
had ripple effects.
Policy decisions were delayed, affecting the technical
decisions. Policy and technical staff were in silos and not
well-coordinated, and contract management was disjointed.
Changes to the project were not well-documented and progress
not adequately monitored.
This culminated in CMS not fully communicating or acting
upon many warnings of problems before the launch. CMS failed to
fully grasp the poor status of the build. One reason was that
no one had a full view into all of the problems and how they
fit together.
Red flags raised to leadership did not always flow to staff
working on the build, and staff did not always alert leadership
to problems on the front lines. CMS was unduly optimistic.
Last-minute attempts to correct problems were rushed and
insufficient. In the 2 months before the launch, CMS added
twice the staff to the project and cut many planned website
functions. And just 72 hours ahead, CMS asked its contractor to
double its computing capacity.
Even with these efforts, the HealthCare.gov website
experienced major problems within hours of its launch. The
website received five times the number of expected users, but
the problems went beyond capacity. The website entry tool
worked poorly, and software coding defects caused malfunctions.
CMS and its contractors did not have coordinated tools to
diagnose these problems. However, CMS pivoted quickly to make
corrections to the website. They brought in additional staff
and expertise from across government and the private sector.
One key was creating a badgeless culture, where Federal
employees and contractors worked together as a team. CMS
designated clear leadership, integrated policy and technical
staff, and developed redundant systems to avoid future website
problems.
CMS also took a more realistic approach to building website
functions. It practiced what officials called ruthless
prioritization, which focused on effectively developing the
most critical functions, like reenrollment, and delaying other
features. They measured progress and monitored problems to
respond more quickly and effectively. These factors contributed
to an improved website and important organizational changes.
Looking ahead, CMS continues to face challenges in
improving HealthCare.gov and managing the Federal marketplace.
This includes addressing more than 30 recommendations from
OIG's other Federal marketplace reports. We will continue to
monitor CMS's actions in response to our recommendations and
its overall management of this and other programs.
Thank you again for inviting OIG to speak with the
committee today, and I will be happy to answer your questions.
[The prepared statement of Ms. Bliss appears in the
appendix.]
The Chairman. Thank you so much.
Mr. Bagdoyan, we will turn to you.
STATEMENT OF SETO J. BAGDOYAN, DIRECTOR OF AUDITS, FORENSIC
AUDITS AND INVESTIGATIVE SERVICE, GOVERNMENT ACCOUNTABILITY
OFFICE, WASHINGTON, DC
Mr. Bagdoyan. Good morning, Chairman Hatch, Ranking Member
Wyden, and members of the committee. I am pleased to be here
today to discuss results from our February 2016 report on
enrollment and verification controls for ACA health-care
coverage obtained through the Federal marketplace during the
2014 open enrollment period.
Our results are based on extensive forensic analyses of
relevant data from CMS and other agencies, such as SSA, IRS,
and DHS--involving originally the entire 2014 applicant and
enrollee universe--and are independent of the undercover work
we performed for that period.
A central feature of ACA's enrollment controls is the
Federal data services hub, which is the primary vehicle for CMS
to initially check information provided by applicants against
various Federal data sources. In addition, the ACA established
a process to resolve inconsistencies, i.e., instances where
applicant information does not match that from marketplace
sources.
In terms of context for our work, coverage offered through
the Federal marketplace is a significant expenditure for the
Federal Government. Current levels of coverage involve millions
of enrollees, of whom about 85 percent receive subsidies. CBO
has estimated Fed subsidy costs at about $880 billion for
fiscal years 2016 through 2025.
I would note that while subsidies are paid to insurers and
not directly to enrollees, they nevertheless represent a
financial benefit to them. As I have stressed before, a program
of this scope and scale remains inherently at risk for errors,
including improper payment and fraud. Accordingly, it is
essential that effective enrollment controls are in place to
help narrow the window of opportunity for such risk and
safeguard the government's investment.
Against this backdrop, I will now discuss our two principal
analytical results.
First, we found that CMS does not track or analyze
aggregate outcomes of data hub inquiries or the extent to which
a queried agency delivers information responsive to a request
or whether an agency reports that that information was not
available.
In this regard, for example, we found that SSA could not
match 4.3 million queries related to names, dates of birth, or
Social Security numbers, and 8.2 million queries related to
citizenship claims. IRS could not match queries involving about
31 million people related to income and family size, and,
within this, 1.3 million people had ID theft issues. Finally,
DHS could not match 510,000 queries related to citizenship and
immigration status.
Accordingly, CMS foregoes opportunities for gaining
valuable insights about significant program integrity issues,
including vulnerabilities to potential fraud, as well as
information useful for enhancing overall program management.
Second, we found that CMS did not have an effective process
for resolving inconsistencies for applicants using the Federal
marketplace. For example, we found that about 431,000
applications, with about $1.7 billion in associated subsidies,
still had about 679,000 inconsistencies unresolved as of April
2015. That is 4 months after the close of the 2014 coverage
year.
Within these, CMS did not resolve Social Security number
inconsistencies for about 35,000 applications, with about $154
million in associated subsidies, or incarceration
inconsistencies for about 22,000 applications, with about $68
million in associated subsidies.
By leaving inconsistencies unresolved, CMS risks granting
eligibility to and making subsidy payments on behalf of
individuals who are ineligible to enroll in qualified health
plans. One important example emphasizes this point. According
to IRS, accurate data are vital for income tax compliance and
the reconciliation of advanced premium tax credits through
filing tax returns, which is a key backend control under ACA.
In closing, our work to date collectively shows that CMS
has assumed a generally passive approach to managing fraud
risks in ACA, weakening the program's integrity. Accordingly,
we continue to underscore that CMS needs to make ACA program
integrity a priority and implement effective controls to help
reduce improper payment and fraud risks and preclude them from
being embedded early in the program's life cycle.
In this regard, we made eight recommendations to CMS in our
February report, which are intended to help mitigate the
vulnerabilities and risks we identified. While the agency
agreed with the recommendations, it is incumbent on CMS to
implement them in a timely fashion and achieve and sustain
measurable results.
Mr. Chairman, this concludes my statement. I look forward
to the committee's questions, and I appreciate the indulgence
for an extra 30 seconds.
[The prepared statement of Mr. Bagdoyan appears in the
appendix.]
The Chairman. I am happy to give you that extra time.
Ms. Bliss, previous reports at the Office of Inspector
General criticized HealthCare.gov and the marketplace,
describing important problems with internal controls, such as
inadequate procedures for checking the eligibility of
enrollees.
How does the case study differ from previous reports of the
Office of Inspector General on the same topic?
Ms. Bliss. Thank you for your question, Mr. Chairman. The
case study is one of a dozen reports that OIG has issued on the
Federal marketplaces. Most of those were more-targeted audits
or evaluations examining aspects of eligibility controls,
payment accuracy, contracting, and security of information.
The case study took a different approach and cast a wide
lens at CMS's management of the project in its entirety, from
multiple perspectives and over a long period of time, in order
to glean lessons learned about what went wrong and what went
right in an effort to help improve both this HealthCare.gov
project and other Federal projects moving forward.
The Chairman. Thank you.
Mr. Bagdoyan, your report pointed out the key role played
by the, quote, ``data services hub,'' which is the electronic
clearinghouse for checking applicant information against
Federal databases.
Now, you said that CMS needs to make better use of this
important enrollment control process. Would you explain that a
little bit?
Mr. Bagdoyan. I would be happy to do that, Mr. Chairman.
Basically, the data hub is a key cog, if you will, in the
overall control environment for ACA. It is up-front. It
processes a lot of queries for information. A lot of those
queries--all of those queries, in fact, are not captured for
future analysis.
We believe that such capture and analysis would provide CMS
with a lot of insight into potential indicators of improper
payments, as well as fraud. So a comprehensive control system
would theoretically enable that sort of analysis for the long
term, and we do actually have a recommendation to that effect
to CMS.
The Chairman. Thank you. We have been long told by CMS,
``Do not worry. Even if there are issues with awarding
subsidies, everything eventually gets fixed when people file
their income taxes.''
The GAO found practices that undermine tax compliance. Am I
right about that?
Mr. Bagdoyan. Yes. We identified a number of
inconsistencies. Out of the 431,000, I believe we had about
35,000 that involved tax or SSN inconsistencies. And according
to IRS, when we discussed this at length, they told us that
this was not only important for tax compliance purposes, but
also for the tax reconciliation process to reconcile the
advanced premium tax credits at the end.
This is the third main back-end control, if you will, in
the overall setup. So without that information that is accurate
and reliable, IRS pointed out that their job is made much more
difficult, not only to do the tax return processing, but also
to reconcile the subsidies.
So it is a long-term problem if it is not addressed.
The Chairman. All right. Ms. Bliss, what are the most
important lessons learned from HealthCare.gov for the
administration, and do you think that the lessons learned from
your case study apply to other large programs and projects,
whether being planned by the Department of Health and Human
Services or other government agencies?
Ms. Bliss. Thank you. We certainly do. The intersection
between policy, technology, and management is not only
essential for HealthCare.gov, but we believe these lessons will
apply to other Federal projects and Federal programs.
We gleaned 10 lessons learned, and I will highlight what I
believe to be the three most significant.
First is establishing clear leadership. We found that the
lack of clear leadership in overall responsibility and clear
lines of delegation had ripple effects, caused a number of
cascading problems across the project, and made problem
resolution more difficult.
We also found that a disconnect between those working on
the policy and making decisions and those working on the
technical aspects of the project created problems on both
sides. And delays in policy decision-making compressed an
already tight time frame for achieving the technical build
successfully.
So, better integration across lines of business, policy,
and technical, as well as across government and contractors
through this badgeless culture, are some of the keys we saw to
correction and success.
And then, finally, taking a posture of continuous learning
is essential, which means being flexible and adaptable,
especially with a startup-type project like HealthCare.gov was.
We found that CMS got stuck on an unwinnable path, and it was
too late before they realized it and tried to make changes.
So keeping that continuous learning posture, being
innovative and flexible, and constantly monitoring for problems
to adjust plans where needed are all important.
Thank you.
The Chairman. Thank you.
Senator Stabenow, we will turn to you.
Senator Stabenow. Thank you very much, Mr. Chairman. And
welcome and thank you to both of you.
Ms. Bliss, I am wondering--to start, just a ``yes'' or
``no'' question. Based on your case study, do you think that
the HealthCare.gov website should be taken down and a
completely new website be built?
Ms. Bliss. No.
Senator Stabenow. Thank you. Like many of my colleagues, we
were very frustrated about what happened in the past, and
clearly you have laid out the problems with the launch, and I
think everyone agrees that there were serious problems with the
launch of HealthCare.gov and it created a lot of difficulties,
and certainly for people in Michigan, to get coverage in 2013.
But that was 3 years ago, and we are now in year 3 of the
Affordable Care Act marketplace operations. So when we look at
the report, the report is really looking backwards, and we can
agree there were problems.
The question is moving forward and how do we ignore the
fact that over 20 million people have received health-care
coverage because of the Affordable Care Act, literally saving
people's lives? That is not just a rhetorical statement. I have
talked to people who were able to get surgery or were able to
get care for their children that they have never been able to
receive before and save lives, and I think that is a good part
of things when we talk about the numbers, the real-life
experiences of people.
The un-insurance rate is the lowest it has ever been, and
Medicaid expansion has resulted in literally millions of our
most vulnerable families receiving the care that they deserve.
So, given the fact that the ACA is the law of the land and
it is our responsibility to make it better, I first want to say
that I hope that all of us will work on how to make it better,
and that is why we appreciate your recommendations as we look
forward, not just in the case of this particular website and
process, but others as well.
But the question is, how do we make it better? So we want
to make sure that we have quality access to health care for
every American, whether it is Medicare, Medicaid, the
Children's Health Program, and so on and so on.
So with that in mind, Mr. Bagdoyan, let me ask about any
other recommendations from a GAO standpoint that you have not
already spoken of today on how we can make these better,
because, frankly, I want the over 20 million people who have
health insurance today who did not have it before to have the
peace of mind going to bed at night of knowing they are going
to be able to take their children to a doctor if they get sick.
I want to keep that. And I am hopeful we can even get as close
to zero as possible in terms of the number of people in our
country who do not have access to health care.
So I am interested in your recommendations on how we go
forward to work together to make this system work better.
Mr. Bagdoyan. Sure. Thank you for your question, Senator
Stabenow.
As you mentioned, we operate under the premise that this is
the law on the books, and my charge is to help make it work as
intended. With that in mind, our report makes eight specific
recommendations. We try not to be too prescriptive to allow CMS
some latitude to explore various options.
However, the key recommendation, I believe, the big-picture
recommendation, is for CMS to conduct a comprehensive risk
assessment of the entire program, sort of top to bottom, and
identify the control vulnerabilities and the risks for improper
payments and fraud.
In that regard, GAO issued, in July of 2015, its framework
for managing fraud risk in Federal programs. So that is a
comprehensive leading practice compilation from the private and
public sectors that would provide the agency with quite a solid
roadmap to perform that risk assessment.
So everything should flow from that assessment in terms of
the types of actions, policy changes, control improvements, and
so forth.
Senator Stabenow. And are you working with CMS? What is
their reaction on this? Are they objecting to that?
Mr. Bagdoyan. No. I think I should give CMS credit that
they accepted all eight recommendations, including this one.
But as they say, the proof is in the pudding. They need to
execute, do so successfully, and then achieve results and
sustain them over the long term. This is not a one-and-done
proposition by any means.
Senator Stabenow. Sure. So just to be clear, you have made
the recommendations. They have accepted all eight
recommendations, and they are in the process of doing them.
Mr. Bagdoyan. That is correct. We had informal discussions,
as well as the formal letter responding to our recommendations.
Senator Stabenow. Thank you, Mr. Chairman.
The Chairman. Senator Coats?
Senator Coats. Thank you, Mr. Chairman. And I want to thank
our two witnesses.
Lord knows where we would be if we did not have GAO and
Inspectors General. The alarming malfeasance and incompetence
of the rollout of this plan is just stunning. And here we are,
we cannot just simply brush it off and say, well, this was a
bad start, but everything is going great now.
On the cost to the taxpayer, probably we will never know.
But thank goodness that we have your organizations providing us
information and spurring on a seemingly bureaucratic nightmare
that exists within the Federal Government in terms of handling
these kinds of programs.
Anybody in the private sector who had done this would have
been bankrupt; investors would have lost all their money. It is
just stunning to continue to observe what it takes to get these
agencies to--I think they are well-intended, they are just
overwhelmed in terms of the complexity of getting this done.
I go the floor of the Senate every week and talk about a
waste of the week, and, Mr. Bagdoyan, I have referenced your
name, not as part of the problem, but as part of the solution.
And the information that you have provided here for me
continues to stun people when they hear about some of the
incompetency.
I was particularly interested, because I think it speaks to
a bigger problem, in your, what was called the secret shopper
investigation, where you deliberately made applications as a
test--you made applications for compliance with the Affordable
Care Act and receiving subsidies. And 11 of the 12--I think my
numbers are right--everything you submitted was fraudulent, but
11 of the 12 were accepted. And even after it was revealed that
they were accepted, follow-up phone calls, pretending to be
that person who was given notice that they were not eligible,
were accepted.
That percentage is pretty high, and if you multiply that
out, it just really makes you wonder if this whole thing was
not gamed or at least so intent on providing numbers to make it
look successful that we really were not getting the
information, the verification, that we needed.
Then there was the question with CMS at one point releasing
a statement, ``Well, we are not in the verification business.''
I think basically what you just said was that they are now
taking a different stand on that.
But I wonder if you could respond to where are we now in
terms of verification capacity so that we do not have this
fraudulent and wasteful situation moving on. I am happy to have
either one of you or both of you address that. But it just
seems easy: an evaluation of Social Security numbers to
determine their validity would make it fairly easy to make a
determination as to whether they qualify or whether they do not
qualify.
But where is CMS in terms of putting that process in place,
and what is the success to date of that process?
Mr. Bagdoyan. Sure. If I may, Ms. Bliss, take first crack
on that.
First, I appreciate the plug on the floor, Senator.
Senator Coats. Sure. [Laughter.]
Keep sending us stuff; I will keep going to the floor.
Mr. Bagdoyan. So in terms of where CMS is with the
controls, what we call the control environment, which is a
series of controls designed to verify information, identify
potential indicators of fraud, and so forth, as our undercover
work indicated, both for 2014 and 2015, where we were equally
successful, there is a semblance of controls in place.
Senator Coats. A semblance?
Mr. Bagdoyan. A semblance of controls in place, some basic
things in place, like identity-proofing the document
reconciliation process to clear inconsistencies, for example.
But in each case, we were able to work around those reasonably
easily and obtain coverage both for 2014 and 2015. So the
vulnerabilities are still in place.
Now, with the recommendations we made in this report,
actually, in late February, we made eight recommendations. As I
explained to Senator Stabenow, the big one is to perform a
comprehensive risk assessment.
Now, that is going to take time. It is going to take time
for CMS to absorb the results and then craft, hopefully,
appropriate solutions for the future. So this is a long-term
proposition. It is not going to be an easy fix.
Senator Coats. Well, I think this speaks to the point that
we got a bad start and everything is going great right now.
Everything is not going great right now. As you said, this is
going to take a long-term effort to try to put these
verification procedures in place and to be able to say that we
are successfully avoiding fraud and waste and an inefficiency
and taxpayer cost level that is just absolutely astounding.
So, with due respect to my colleagues, to tout this as
something that has happened in the past but is corrected now
and we are sailing into the bright future, I think we have a
lot of work to do.
Thanks, Mr. Chairman.
The Chairman. Senator Wyden?
Senator Wyden. Thank you, Mr. Chairman.
I want to again say that the initial rollout was botched,
and I appreciated the Inspector General making it clear that a
couple months in, there was serious progress.
So you all reported that after the first open enrollment,
the agency demonstrated a strong sense of urgency to take
action, accepted new work processes, and they, quote,
``improved the HealthCare.gov website substantially within 2
months.''
I think it would be helpful, Ms. Bliss, if you could tell
us two things. What were the operational and strategic changes
that were made after that first open enrollment, and do you
feel they are better-equipped to deal with the challenge now?
Ms. Bliss. Thank you, Ranking Member Wyden, for that
question. As we discussed in the case study, some of the key
strategic and operational changes that were made as part of the
correction were to, one, establish more clear leadership and
designate roles and responsibilities, and they did it in a way
that really brought together staff and contractors across all
of the important business lines that were affected and needed
to be involved in the correction. That includes the policy
people, the technical, the communications, and the contractors
all coming together.
With the influx of experts from across government and the
private sector, there was the potential that it could have
become more chaotic, but, in fact, we saw that the reverse was
true. It was well-organized. Folks were working together in a
badgeless culture as a team. There was better communication,
there was better measurement and monitoring of problems, and
there was progress in order to apply solutions more quickly and
effectively.
Senator Wyden. So, in effect, after the first few months,
which everybody has acknowledged were not ideal, your
characterization was that essentially it was well-organized.
Ms. Bliss. It was much better organized----
Senator Wyden. I was using your word----
Ms. Bliss [continuing]. And they continue to make progress.
Senator Wyden. All right. Good. Mr. Bagdoyan, first, I am
probably the biggest user of GAO products here in the Congress.
I so admire the professionalism of the agency, and I think you
heard me say I do not take a back seat to anybody when it comes
to cracking down on actual real-world fraud.
My question to you is, is it not correct that when you
testified before the committee last year, you stated that the
secret shopper investigation failed to uncover a single real-
world example of fraud?
Mr. Bagdoyan. Yes, that is what I said, Senator Wyden, and
I would also couch that very carefully for you and the
committee.
The intent of that investigation was not to uncover fraud
but to flag control vulnerabilities, as well as identify
indicators of potential fraud, which I think we did quite
successfully.
So I just want to make clear my charge is not to find
fraud. Fraud is determined through a separate criminal
proceeding in courts to definitively determine that. So my job,
again, is to look for vulnerabilities in controls, as well as
identify indicators of potential fraud or improper payments.
Senator Wyden. So let us go then from last year when there
was not one single real-world example of fraud to where we are
now. Is it correct to say that the entire investigation failed
to identify any actual fraud?
Mr. Bagdoyan. Well, again, I would refer you to my answer.
That was not our intent. So if I am not looking for fraud, I am
not going to find it. What I am looking for is vulnerabilities
in controls and indicators of potential fraud, such as the
inconsistencies with the Social Security numbers, as well as,
in the case of the IRS, 1.3 million people having potential ID
theft issues, which is a significant red flag.
Senator Wyden. I think that, as is always the case, you all
are right to talk about various issues that ought to be part of
the debate. That is not what is going on here. What people are
saying is, this is fraud, fraud, fraud, fraud, and I appreciate
your taking us through this in, I think, a better-balanced
view.
Ms. Bliss, at HHS, you all do audits, OIG does audits. Have
you uncovered, in connection with this, any confirmed cases of
fraud?
Ms. Bliss. No, we have not had any cases that have resulted
in criminal convictions or civil settlements to date. We do
have a few investigations that are ongoing, and I cannot
predict what those outcomes will be.
Senator Wyden. Look, I do not know how many times I have
said in this committee that when there are big, important
issues--and certainly the Affordable Care Act is right at the
top--we need to work in a bipartisan fashion, and there is not
a program anywhere in government where you cannot find
opportunities to work together and be bipartisan.
I ticked off a number of them. The chairman and I are
working together on what I think is the future of the Medicare
program, chronic care; Senator Grassley and I are finishing
what I think is a blockbuster study looking at hepatitis C. And
it raises the question of, when we have cures, will people be
able to afford them?
What I think is important is that, to do bipartisan work,
we have to move away from, first, the past, because everybody
has acknowledged that the first few months were botched. I do
not know how many times you can say it. But you all said--and I
read your comments--after the first few months, you said they
had made substantial improvements. I think I can come back to
it and perhaps read it one more time.
``The Centers for Medicare and Medicaid Services recovered
the health care government website for high consumer use within
2 months.'' Now, that is the new news. That is just a few weeks
old. That is new news, and I want people to hear that, and I
want people to hear that there were no actual real-world cases
of fraud uncovered.
Now, one final question, if I might, for you, Ms. Bliss. Do
you disagree with the statement that I made with respect to the
accomplishments of the Affordable Care Act? That is not your
formal role as Inspector General, but does anything strike you
as being inaccurate there with respect to the uninsured rate or
anything of that nature?
Ms. Bliss. As an independent oversight agency, we do not
take positions on whether particular programs should exist, but
we look to make sure they are operating correctly.
Senator Wyden. That is not the question. The question was
about the facts, and what I think, again, is, this is a hard
fact that is not in dispute, that the uninsured rate is now at
or near the lowest level recorded across 5 decades of data,
with about 20 million previously uninsured Americans gaining
coverage since the Act's provisions went into effect.
So I will keep the record open so that if you or your
agency has any information suggesting that is wrong, I would
surely like to know about it.
Ms. Bliss. Thank you. I do not have any information
suggesting that that is wrong.
Senator Wyden. Wonderful. Mr. Chairman, thank you.
The Chairman. Senator Scott?
Senator Scott. Ms. Bliss, do you have any information
suggesting that those numbers are right?
Ms. Bliss. I cannot validate those numbers. I do not have
any reason to believe they are not.
Senator Scott. But you have no indication either way,
actually.
Ms. Bliss. I have no basis, no.
Senator Scott. If I tell you that the number is 30 million,
you have no reason to believe that it is not 30 million.
Ms. Bliss. I do not have a basis for validating that
number. Our case study--I am sorry.
Senator Scott. Thank you. Mr. Bagdoyan, our ranking member
asked you several questions about fraud, and I certainly
understand and appreciate why so many Americans look at this
process and become disenchanted.
Your objective was never to figure out how much fraud was
in the system. Your objective, it appeared to me, was to show
us how fraud could happen.
Mr. Bagdoyan. Yes, essentially, Senator, you are correct.
The big picture we are looking at is for any vulnerabilities in
the controls that are in place and also for any indicators of
potential fraud that pop up. For example, our ability to
circumvent the controls we encountered during our undercover
work--we did that for 2014, and we repeated that experience in
2015, in which case we were successful 17 out of 18 attempts.
Now, I would have to caution that, of course, further to
the point that Senator Coats made earlier, that is not a
projectable number. So we have to be very careful that that
does not represent the actual universe; that is just a data set
that we use to continue our work in this area.
Senator Scott. Thank you very much. No one is going to
mistake me for a fan of Obamacare or the ACA, without any
question. For a number of reasons, I am not a fan of the
website nor the actual policy itself, the legislation.
I think of the Independent Payment Advisory Board, what
some have referred to as a death panel, and the ability to
ration care into the future. This is one of the classic
examples of why so few Americans have the same appreciation of
the ACA that others have talked about.
I think the fact that we are talking about taxing
Americans, whether it is their income or their profits, an
additional 3.8-percent tax, raising somewhere over $120
billion, is another reason why so few Americans have the same
positive theme that we have heard from some of our friends on
the other side.
Think about the whole notion of how the health-care law is
going to regulate the posting of calories at pizza parlors,
grocery stores, all over the place, and, by default, increase
the price of these groceries, these pizzas and other non-food
items, reducing the number of employees' hours, talking about
the impact on middle-income America, so many Americans losing
perhaps up to 25 percent of their income because of the ACA.
We can see why so many Americans have found frustration
with where we are with the ACA, that it is not old news to
them.
It is not old news, actually, when you think about the fact
that so many Americans are facing higher premiums. We have
heard so many different numbers this morning. We know that at
least some States have seen an increase of more than 25 percent
in their health-care costs. Two States have seen those numbers
go over 35 percent. Those are real dollars for struggling
Americans who cannot afford the cost of health insurance.
Not only are the premiums higher, the deductibles are
higher, the out-of-pocket expenses are higher. The only thing
that is actually lower are the doctors to choose from and the
hospitals to go to.
We have seen a catastrophic occurrence under this health-
care law. And even at one of the most recent Democratic town
halls, a young lady, supportive of President Obama, who
supports the health-care law, said that her premiums had
doubled, tripled. Her concerns were strong, clear.
Here is one real case example that, Ms. Bliss, I hope is no
longer happening. A young man named Tom Dougall from Elgin, SC,
who created an account on HealthCare.gov, was called shortly
thereafter by a man named Mr. Justin Hadley from North Carolina
who had done the exact same thing: gone online to
HealthCare.gov and created an account. But what he found
populating his account was information from Mr. Dougall.
He called HHS and could not get any assistance. Finally,
they called our office, and, during one of the hearings, we
were able to get that situation solved, or at least the
beginning of that situation solved.
Can you guarantee me that that situation is no longer
occurring anywhere within HealthCare.gov?
Ms. Bliss. I cannot guarantee that. We have overseen and
conducted reviews of the controls to ensure that the website
and other parts of the program for identity verification, no
eligibility verification, are working properly. But we have
raised concerns about some flaws or weaknesses in those
controls, similar to GAO, and I cannot make that guarantee. But
we are certainly working hard to identify where there is a
vulnerability of that happening and make recommendations on how
to improve it.
Senator Scott. My last statement, since I am out of time so
quickly here today, is, it appears that as we have celebrated
the success of improving the system in the first couple
months--I will note it was a new $1-trillion program--one of
the recommendations is for clear leadership. Earth-shattering.
Thank you.
The Chairman. Senator Isakson?
Senator Isakson. Thank you, Mr. Chairman.
I apologize for missing your testimony, and I apologize for
being late. But I do have one question based on a letter that I
have sent previously to CMS, and I want to ask this question.
Mr. Bagdoyan, do you agree that increasing the utilization
of existing, tested data sources is one easy way that CMS could
reach the mutual goal of expanding program integrity and
management and better assess fraud risk?
Mr. Bagdoyan. Yes. That is, in fact, one of our
recommendations to CMS: to consider doing that on an active
basis, both to capture the data and then analyze the data for
whatever indicators that they may throw off and act upon those.
Yes.
Senator Isakson. Then do you have any idea when CMS is
going to move forward to actually take advantage of that and do
it?
Mr. Bagdoyan. Well, as I stated before in response to
several Senators' questions, CMS has accepted those
recommendations. They are on record in writing as having done
so. And as I said in my opening statement, it is now incumbent
on the agency to take action on a timely basis. But as I said,
it will take time to work through this. It is not an easy fix.
It is not a short-term fix. It is not a one-and-done fix.
Senator Isakson. Well, I apologize for being late, because
obviously you covered it in your opening statement. But there
is readily available data and companies that are already under
contract to CMS that are available to provide information that
could greatly enhance the integrity of the program and uproot
fraud a lot easier, and I appreciate your testimony to that
effect.
Mr. Bagdoyan. Yes. The data are available, definitely.
Senator Isakson. Thank you. Thank you, Mr. Chairman.
The Chairman. Thank you. I want to thank our witnesses for
appearing here today. The work that each of you does is very
important, as far as we are concerned, you and your
organizations. It is vitally important to this committee, and
we are thankful for the quality product that both the HHS/OIG
and GAO produce to assist us in our policymaking and oversight
efforts.
I also want to thank my colleagues for their participation
in this important hearing. I think the hearing has been
insightful. It has been enlightening. Unfortunately, I think
this hearing further revealed that we are only now getting to
the water level of the Obamacare iceberg, it seems to me.
As premiums continue to skyrocket and insurance options
become more and more limited, an increasing number of Americans
are being hung out to dry. Over the past year, we had a
reasonable amount of consensus on several of the unworkable and
failed provisions of Obamacare, but for some reason, many still
have their heads stuck in the sand hoping that things will
finally start working out at some point.
Now, I implore my Democratic colleagues to work with me and
my Republican friends to repeal and replace the so-called
Affordable Care Act before it is altogether too late. Insurance
premiums and health-care costs continue to rise, and little is
being done to stem the tide.
It is high time to put partisan politicking and bickering
aside and find workable bipartisan solutions. There is more we
can do. There is more we, it seems to me, have to do. Honestly,
I earnestly believe that we can do it. The American people
deserve better than what they have right now and, more
importantly, than what they are about to have in the next few
years.
So I encourage each of my colleagues to meet with me and
find workable solutions, and I encourage both of you to keep
doing the jobs that you are doing. They are very important to
this committee and I think to our country at-large.
I would ask that any written questions for the record be
submitted by Thursday, March 31st of this year.
With that, this hearing will be adjourned. Thank you for
being here.
[Whereupon, at 11:10 a.m., the hearing was concluded.]
A P P E N D I X
Additional Material Submitted for the Record
----------
Prepared Statement of Seto J. Bagdoyan, Director of Audits, Forensic
Audits and Investigative Service, Government Accountability Office
Chairman Hatch, Ranking Member Wyden, and members of the committee:
I am pleased to be here today to discuss enrollment and
verification controls for health-care coverage that individuals obtain
through the Federal health-insurance exchange under the Patient
Protection and Affordable Care Act (PPACA). The act expands the
availability of subsidized health-care coverage, and it provides for
the establishment of health-insurance exchanges, or marketplaces, to
help consumers in comparing and selecting among insurance plans offered
by participating private issuers of health-care coverage. Under PPACA,
States may elect to operate their own health-care marketplaces, or may
rely on the federally facilitated marketplace, or Health Insurance
Marketplace (Marketplace).\1\ The Centers for Medicare and Medicaid
Services (CMS), a unit of the Department of Health and Human Services
(HHS), maintains the Federal Marketplace.
---------------------------------------------------------------------------
\1\ Specifically, the act required, by January 1, 2014, the
establishment of health-insurance marketplaces in all States. In States
not electing to operate their own marketplaces, the Federal Government
was required to operate a marketplace.
To be eligible to enroll in a qualified health plan offered through
a marketplace, an individual must be a U.S. citizen or national, or
otherwise lawfully present in the United States; reside in the
marketplace service area; and not be incarcerated (unless incarcerated
while awaiting disposition of charges). Marketplaces, in turn, are
required by law to verify application information to determine
eligibility for enrollment and, if applicable, determine eligibility
---------------------------------------------------------------------------
for income-based subsidies the act provides.
PPACA provides subsidies to those eligible to purchase private
health-insurance plans who meet certain income and other requirements.
According to the Congressional Budget Office, the estimated cost of
subsidies and related spending under the act is $880 billion for fiscal
years 2016-2025. While subsidies under the act are not paid directly to
enrollees, participants nevertheless benefit through reduced monthly
premiums or lower costs due at time of service, such as copayments.\2\
Because subsidy costs hinge on eligibility for coverage, enrollment
controls that help ensure only qualified applicants are approved for
coverage with subsidies are a key factor in determining Federal
expenditures under the act.
---------------------------------------------------------------------------
\2\ Enrollees can pay lower monthly premiums by virtue of a tax
credit the act provides.
A central feature of the enrollment controls is the Federal ``data
services hub'' (data hub), which, among other things, provides a
vehicle to check applicant-provided information against a variety of
data sources.\3\ Verification steps include validating an applicant's
Social Security number, if one is provided; \4\ verifying citizenship,
status as a national, or lawful presence by comparison with Social
Security Administration (SSA) or Department of Homeland Security (DHS)
records; and verifying household income and family size by comparison
against tax-return data from the Internal Revenue Service (IRS), as
well as data on Social Security benefits from SSA.
---------------------------------------------------------------------------
\3\ In particular, PPACA requires that consumer-submitted
information be verified, and that determinations of eligibility be
made, through either an electronic verification system or another
method approved by HHS. To implement this verification process, CMS
developed the data hub, which acts as a portal for exchanging
information between the Federal Marketplace, State-based marketplaces,
and Medicaid agencies, among other entities, and CMS's external
partners, including other Federal agencies. The Marketplace uses the
data hub in an attempt to verify that applicant information necessary
to support an eligibility determination is consistent with external
data sources.
\4\ A marketplace must require an applicant who has a Social
Security number to provide the number. 42 U.S.C. Sec. 18081(b)(2) and
45 CFR Sec. 155.310(a)(3)(i). However, having a Social Security number
is not a condition of eligibility.
If the eligibility information applicants provide to the Federal
Marketplace cannot be verified through the external sources, such as
SSA, IRS, and DHS, an ``inconsistency'' will result. In particular, an
inconsistency can arise when the data hub query process yields no
information; or when information is available through the data hub, but
it does not match information the applicant has provided.\5\
---------------------------------------------------------------------------
\5\ When an inconsistency is generated, the Marketplace is to
proceed with determining other elements of eligibility using the
attestations of the applicant, and ensure that subsidies are provided
on behalf of the applicant, if he or she is qualified to receive them,
while the inconsistency is being resolved. As part of this resolution
process, the applicant is generally required to submit documentation to
substantiate eligibility for the program. In the case of the Federal
Marketplace, CMS uses a document-processing contractor, which reviews
documentation applicants submit, by mail or online upload, to resolve
inconsistencies. Inconsistencies are discussed more fully later in this
testimony.
My testimony today is based on a report we issued on February 23,
2016, that examined eligibility and enrollment controls, and fraud
risk, of the Federal Marketplace.\6\ It addresses:
---------------------------------------------------------------------------
\6\ GAO, Patient Protection and Affordable Care Act: CMS Should Act
to Strengthen Enrollment Controls and Manage Fraud Risk, GAO-16-29
(Washington, DC: February 23, 2016). In addition, we have presented two
other related testimonies prior to issuance of the report. See GAO,
Patient Protection and Affordable Care Act: Preliminary Results of
Undercover Testing of Enrollment Controls for Health Care Coverage and
Consumer Subsidies Provided Under the Act, GAO-14-705T (Washington, DC:
July 23, 2014); and GAO, Patient Protection and Affordable Care Act:
Observations on 18 Undercover Tests of Enrollment Controls for Health-
Care Coverage and Consumer Subsidies Provided under the Act, GAO-15-
702T (Washington, DC: July 16, 2015).
1. The extent to which applicant information is verified through
---------------------------------------------------------------------------
the data hub--the primary means for verifying eligibility; and
2. The extent to which the Federal Marketplace resolved
inconsistencies that resulted from the data hub verification
process.\7\
---------------------------------------------------------------------------
\7\ In addition to findings presented in this testimony--and as
described in detail in our July 2015 testimony, with additional
information provided in our February 2016 report--our work also
identified vulnerability to fraud, when we obtained, through covert
vulnerability testing, Federal Marketplace approval of subsidized
coverage for 11 of 12 fictitious applicants for 2014, with coverage
continuing into 2015. We concluded CMS has assumed a passive approach
to identifying and preventing fraud, and that adopting a more
strategic, risk-based approach could help identify fraud
vulnerabilities before they could be exploited in the enrollment
process. We recommended that HHS direct CMS to conduct a fraud risk
assessment, consistent with best practices provided in GAO's framework
for managing fraud risks in Federal programs, of the potential for
fraud in the process of applying for qualified health plans through the
Federal Marketplace. HHS concurred with our recommendation and said it
plans to conduct such an assessment. See the framework at GAO, A
Framework for Managing Fraud Risks in Federal Programs, GAO-15-593SP
(Washington, DC: July 2015).
In our report, to examine outcomes of the data hub applicant
verification process, we obtained summary data from key Federal
agencies involved in the process--SSA, IRS, and DHS--on the nature and
extent of their responses to electronic inquiries made through the data
hub, for the 2014 and 2015 coverage years.\8\ We also interviewed
agency officials and reviewed statutes, regulations, and other policy
and related information. In addition, we obtained applicant data on
inconsistencies, subsidies awarded, and submission of required
verification documentation, from CMS data systems for coverage year
2014. To determine the reliability of data we used, we interviewed CMS
officials and others responsible for their respective data, reviewed
relevant documentation, and performed electronic testing to determine
the validity of specific data elements we used to perform our work.
Based on this reliability examination, we concluded that the data we
used were sufficiently reliable for our purposes.
---------------------------------------------------------------------------
\8\ In this testimony, we use ``outcomes'' to mean results obtained
from inquiries made through the data hub, and not any ultimate
determination made whether an applicant inconsistency exists.
Additional details on our scope and methodology can be found in our
report. We conducted our performance audit from January 2014 to
February 2016 in accordance with generally accepted government auditing
standards.
cms's approach to applicant verification information needs improvement
HHS officials described the data hub process to us as being part of
an innovative, multilayered approach to verifying applicant information
efficiently and without undue burden on individuals and families.
Through secure electronic connections, the data hub provides real-time
responses to eligibility queries, HHS told us.
In our February 2016 report, however, we found that although the
data hub plays a key role in the eligibility and enrollment process,
CMS officials said the agency does not track the extent to which the
Federal agencies deliver responsive information to a request, or,
alternatively, whether they report that information was not available.
Additionally, CMS officials said they do not analyze data provided in
response to data hub inquiries. This is because, they said, by design,
the data hub does not store individual transactional data that could be
collectively analyzed over time. For policy reasons, the officials
said, the agency did not want the data hub to become a data repository
itself, and in particular, a repository of sensitive personal data.\9\
The CMS officials also said the agency is barred legally from
maintaining IRS taxpayer information in the data hub.
---------------------------------------------------------------------------
\9\ In particular, according to CMS officials, the data hub does
not read and store the content of requests received. It only validates
message structure and determines routing information to send the
request to the correct destination. The data hub next returns the
response it receives to the requester. The data hub stores data such as
transaction identifier for each request. By CMS requirements, the data
hub cannot store privacy data, the officials said.
Asked about analysis of data hub responses, CMS officials told us
when we conducted work for our February 2016 report that the key
performance measures for the data hub are the extent to which the
system is available for queries, and the extent to which transmissions
of queries and responses are successfully accomplished; that is, that
an inquiry is made and a corresponding reply received, without regard
to content.\10\
---------------------------------------------------------------------------
\10\ According to CMS officials, the data hub only captures a code
for type of reply that is generated when agencies respond to the
inquiries, and those codes are not associated with any other applicant-
identifying information or information that may have been provided in
response to the query. There are no additional data kept on what
information might have been transmitted in the source agency's
response, such as income or family size. Likewise, the data hub does
not track whether information provided through the data hub matches
information originally provided by the applicant, the officials said.
Further, the Federal agencies responding to data hub queries
generally told us they do not analyze outcomes of data hub inquiries.
Instead, SSA, IRS, and DHS officials said they focus on responding to
inquiries received. Our review also found that SSA, IRS, and DHS had
limited information on the nature and extent of the inquiries made by
the data hub. According to the three agencies, available statistics
reflect data hub inquiries in general, and cannot be broken out by
program, such as a qualified health plan or Medicaid. In addition,
according to agency officials, an unknown number of data hub applicant
inquiries were duplicates, which we could not eliminate from our
examination.\11\ Instead, agency officials told us, they generally
process inquiries sequentially as they are received from the data hub.
Thus, we found that while the agencies can provide some information on
data hub queries, they cannot provide comprehensive information
specifically on number of inquiries and individuals represented by
those queries.
---------------------------------------------------------------------------
\11\ The agencies could not comprehensively identify the number of
duplicates.
We further found, based on our examination of available statistics
from SSA, IRS, and DHS, that while the agencies could successfully
provide applicant verification information in a large percentage of
cases, they did not have data in their records to verify information
for millions of data hub inquiries over the course of PPACA's first two
enrollment cycles, for 2014 and 2015 coverage.\12\
---------------------------------------------------------------------------
\12\ For example, SSA accomplished a match on name, Social Security
number, and date of birth in about 95 percent of cases for PPACA's
first enrollment cycle, for 2014 coverage. However, for about 4.4
million inquiries--or about 5 percent of the total--the applicant
information did not match SSA records. In addition, after completion of
the name, Social Security number, and date of birth match, when SSA
attempted to verify additional information, the agency could not
confirm citizenship in about 8.2 million inquiries where individuals
claimed they were citizens.
We concluded that by not assessing the extent to which data hub--
provided data matches applicant-provided information, CMS foregoes
analysis of the extent to which responding agencies successfully
deliver applicant verification information in response to data hub
requests. In doing so, CMS foregoes information that could suggest
potential program issues or potential vulnerabilities to fraud, as well
as information that might be useful for enhancing program
management.\13\ We recommended that HHS direct CMS to conduct a
comprehensive feasibility study on actions CMS can take to monitor and
analyze, both quantitatively and qualitatively, the extent to which
data hub queries provide requested or relevant applicant verification
information, for the purpose of improving the data-matching process and
reducing the number of applicant inconsistencies; and for those actions
identified as feasible, create a written plan and schedule for
implementing them. HHS said it concurred with our recommendation and is
reviewing options for such a study.
---------------------------------------------------------------------------
\13\ By analyzing the outcomes of data hub inquiries, and in
particular, clarifying the nature and extent of inconsistencies arising
from this process, CMS could, for example, assess whether other sources
of data, such as the National Directory of New Hires, could be useful
for more current applicant information on income. Similarly, CMS could
analyze the information to examine whether other sources of citizenship
information, such as the Department of State's passport data, could be
used to aid in verifying applicant citizenship. There may also be
correlations observed between various types of applicants and types of
information available from trusted data sources.
---------------------------------------------------------------------------
the federal marketplace did not resolve about one-third of applicant
inconsistencies for coverage year 2014, involving $1.7 billion in
associated subsidies
For qualifying applicants, the act provides two forms of subsidies
for consumers enrolling in individual health plans, both of which are
paid directly to insurers on consumers' behalf. One is a Federal income
tax credit, which enrollees may elect to receive in advance of filing
tax returns, and which reduces a consumer's monthly premium payment.
This is known as the advance premium tax credit (APTC).\14\ The other,
known as cost-sharing reduction (CSR), is a discount that lowers the
amount consumers pay for out-of-pocket charges such as deductibles,
coinsurance, and copayments.
---------------------------------------------------------------------------
\14\ When applicants apply for coverage, they report family size
and the amount of projected income. Based, in part, on that
information, the Marketplace will calculate the maximum allowable
amount of APTC. An applicant can then decide if he or she wants all,
some, or none of the estimated credit paid in advance, in the form of
payment to the applicant's insurer that reduces the applicant's monthly
premium payment. If an applicant chooses to have all or some of his or
her credit paid in advance, the applicant is required to ``reconcile''
on his or her Federal tax return the amount of advance payments the
government sent to the applicant's insurer on the applicant's behalf
with the tax credit for which the applicant qualifies based on actual
reported income and family size.
In our report, for applicants who obtained subsidies but had
application inconsistencies, we identified about 1.1 million
applications with a total of about 2 million inconsistencies.\15\ These
applications had combined APTC and CSR subsidies of about $4.4 billion
associated with them for coverage year 2014. We found, based on our
analysis of CMS data, that the agency resolved about 58 percent of the
total inconsistencies, meaning the inconsistencies were settled by
consumer action, such as document submission, or removed due to events
such as life change, application deletion, or consumer cancellation.
Meanwhile, our analysis found that about 34 percent of inconsistencies,
with about $1.7 billion in associated subsidies, remained open, as of
April 2015--that is, inconsistencies still open several months
following the close of the 2014 coverage year.\16\
---------------------------------------------------------------------------
\15\ In particular, we obtained data from CMS on applicant
inconsistencies generated for the Federal Marketplace and the value of
APTC and CSR subsidies associated with them, for the 2014 coverage
year. Specifically, to observe the number of inconsistencies created
and subsequently resolved, we examined applications that were awarded
subsidies and that were created and submitted during the 2014 open-
enrollment period plus a special enrollment period extension that
followed. The open-enrollment period ran from October 1, 2013, to March
31, 2014, and the extension was through April 19, 2014.
We excluded from our analysis applications modified after
submission, because CMS officials told us that inconsistencies can be
generated or resolved based on consumer actions, such as updating of
application information. We selected the unmodified applications that
had received subsidies as presenting the simplest case for examining
inconsistency generation and subsequent resolution.
Our selection criteria meant excluding 17 percent of the total
number of applications with subsidies and inconsistencies because they
had been modified. A single application may reflect more than one
person, each of whom might have different inconsistencies in different
stages of resolution. The CMS data provided the APTC and CSR amounts at
the application level. Consequently, the results of our analysis are
not mutually exclusive by type of inconsistency, and applications and
their associated subsidy amounts may be represented in multiple
categories.
\16\ The remainder were terminations or adjustments based on
failure to submit documentation to resolve inconsistencies. By
comparison with the inconsistency results in our analysis, HHS reported
that more than 8.84 million people selected or were automatically
reenrolled in 2015 plans through the Federal Marketplace as of the end
of the second open enrollment period on February 15, 2015.
We also found, based on our analysis of the 2014 data, that CMS did
not terminate or adjust subsides for any applications with
incarceration or Social Security number inconsistencies, plus other
inconsistencies.\17\ Further, CMS officials told us that they currently
do not plan to take any actions on individuals with unresolved Social
Security number or incarceration inconsistencies.
---------------------------------------------------------------------------
\17\ These other inconsistencies relate to American Indian status,
and presence of qualifying
employer-sponsored coverage or other minimum essential coverage.
Social Security number inconsistencies. Under CMS regulations, the
Marketplace must validate all Social Security numbers provided by
submitting them to SSA along with other identifying information. If the
Marketplace is unable to validate the Social Security number, it must
follow the standard process for resolving all types of
inconsistencies.\18\ In our analysis, we identified about 35,000
applications that had an unresolved Social Security number
inconsistency, which were associated with about $154 million in
combined subsidies.
---------------------------------------------------------------------------
\18\ 45 CFR Sec. 155.315(b).
We reported that CMS officials told us they did not take action to
terminate coverage or adjust subsidies during 2014 based on Social
Security number inconsistencies. They said this was because such
inconsistencies are generally related to other inconsistencies, such as
citizenship or immigration status, and that document submissions for
citizenship or immigration status may also resolve Social Security
number inconsistencies. Overall, CMS officials told us they do not
consider missing or invalid Social Security number information to be a
stand-alone inconsistency that must be resolved, and do not take
---------------------------------------------------------------------------
adverse action in such cases.
However, CMS regulations state that ``to the extent that the
[Marketplace] is unable to validate an individual's Social Security
number through the Social Security Administration,'' the Marketplace
must follow its standard inconsistency procedures.\19\ Further, when
promulgating this regulation, CMS explained that transmitting Social
Security numbers to SSA for validation ``is separate from the [PPACA]
provision regarding citizenship verification, and only serves to ensure
that SSNs [Social Security numbers] provided to the [Marketplace] can
be used for subsequent transactions, including for verification of
family size and household income with IRS.'' \20\
---------------------------------------------------------------------------
\19\ 45 CFR Sec. 155.315(b).
\20\ 77 Fed. Reg. 18310, 18355 (March 27, 2012).
In addition to unresolved Social Security number inconsistencies
generally, our analysis also found in particular more than 2,000
applications with Social Security number inconsistencies that had no
corresponding citizenship or immigration inconsistencies. We also
identified nearly 5,500 applications with Social Security number
inconsistencies that had no corresponding income inconsistency. These
applications had total subsidies of about $10 million and $31 million
associated with them, respectively. They indicate that Social Security
number inconsistencies can stand alone, unrelated to other
---------------------------------------------------------------------------
inconsistencies.
Social Security number inconsistencies also affect tax compliance.
Missing or invalid Social Security numbers can affect IRS verification
that taxpayers have properly filed APTC information on their tax
returns, as well as impair IRS outreach to taxpayers who have received
the APTC subsidy.\21\
---------------------------------------------------------------------------
\21\ See GAO-16-29 for a full discussion.
We recommended that HHS direct CMS to identify and implement
procedures to resolve Social Security number inconsistencies where the
Marketplace is unable to verify Social Security numbers or applicants
do not provide them. HHS concurred with our recommendation, but did not
---------------------------------------------------------------------------
provide details on how it would seek to implement it.
Incarceration inconsistencies. In our inconsistency analysis that
we reported on in February 2016, we identified about 22,000
applications having an unresolved incarceration inconsistency, which
were associated with about $68 million in combined subsidies. CMS
officials, however, told us they did not terminate eligibility for
incarceration inconsistencies, because the agency determined in fall
2014 that SSA's Prisoner Update Processing System (PUPS) was unreliable
for use by the Marketplace.\22\ As a result, CMS officials told us the
agency elected to rely on applicant attestations on incarceration
status.\23\
---------------------------------------------------------------------------
\22\ The PUPS system contains information on incarcerated
individuals in all 50 State corrections departments, the Federal Bureau
of Prisons, and local and other facilities. According to SSA, it is the
only national database with records of Federal, State, and local
incarcerations. SSA uses PUPS to identify individuals who may no longer
be eligible for SSA benefits due to incarceration. In addition to SSA,
other Federal programs, such as Medicare, use PUPS data.
\23\ In the absence of an approved data source, the Marketplace may
accept applicant attestation on incarceration status without further
verification, unless the attestation is not reasonably compatible with
other information in its records. 45 CFR Sec. 155.315(e).
PPACA provides that incarcerated individuals are not eligible to
enroll in a qualified health plan through a marketplace, with the
exception of those incarcerated pending disposition of charges. CMS
currently uses PUPS to generate incarceration inconsistencies when
there are indications an applicant may be incarcerated. As part of the
inconsistency resolution process, the Marketplace notifies applicants
to send documentation to resolve the inconsistency. To do so, consumers
---------------------------------------------------------------------------
can submit documentation such as release papers, CMS officials told us.
Under CMS's approach to incarceration inconsistencies, agency
officials told us, the Marketplace continues to make an initial
verification attempt using the PUPS data. If a consumer maintains he or
she is not incarcerated, CMS will rely on that representation and not
take adverse action, regardless of what PUPS indicates, officials told
us. According to HHS officials, based on the data reliability issue,
the Marketplace no longer requires applicants to submit documentation
on incarceration status.
In its 2013 computer-matching agreement with CMS, SSA acknowledged
that PUPS is not as accurate as other SSA data and contains information
that SSA may not have independently verified. Thus, the agreement
states that CMS will independently verify information it receives from
PUPS and will provide individuals an opportunity to contest an
incarceration inconsistency before any adverse action in an eligibility
determination. Overall, according to SSA officials, PUPS information
can be used to identify individuals who require additional follow-up to
determine eligibility.
We reported that our review of documentation CMS provided for its
decision to take no adverse action on incarceration inconsistencies
showed it did not contain key information supporting the agency's
decision to not use PUPS data. Specifically, the documentation did not
provide specific details on why, or to what extent, people were
misidentified as incarcerated; why CMS also judged inmate release
information to be unreliable; any criteria or assessment employed to
conclude that the PUPS data were not sufficiently current or accurate;
or the potential cost associated with not verifying incarceration
status.
We concluded that without clearly identifying such elements as
analysis, scope, and costs of significant decisions, CMS is at greater
risk of providing benefits to ineligible applicants, and also may
undermine confidence in the applicant verification process and
compromise overall program integrity. We further concluded that by not
using PUPS data as a lead for further investigation, and by relying on
applicant attestation in the alternative, CMS may be granting
eligibility to, and making subsidy payments on behalf of, individuals
who are ineligible to enroll in qualified health plans.
We recommended that HHS direct CMS to reevaluate use of PUPS
incarceration data and make a determination to either (1) use the PUPS
data, among other things, as an indicator of further research required
in individual cases, and to develop an effective process to clear
incarceration inconsistencies or terminate coverage; or (2) if no
suitable process can be identified to verify incarceration status,
accept applicant attestation on status in all cases, unless the
attestation is not reasonably compatible with other information that
may indicate incarceration, and forego the inconsistency process. HHS
concurred with our recommendation, but did not provide details on how
it would seek to implement it.
We also recommended that HHS direct CMS to fully document prior to
implementation, and have readily available for inspection thereafter,
any significant decision on qualified health plan enrollment and
eligibility matters, with such documentation to include details such as
policy objectives, supporting analysis, scope, and expected costs and
effects. HHS concurred with our recommendation, and said it was
committed to documenting significant decisions.\24\
---------------------------------------------------------------------------
\24\ In all, our February 2016 report contained eight
recommendations to HHS, and the agency concurred with all of them. See
GAO-16-29 for the complete list of recommendations, as well as HHS
agency comments and our evaluation of them.
Chairman Hatch, Ranking Member Wyden, and members of the committee,
this completes my prepared statement. I would be pleased to respond to
---------------------------------------------------------------------------
any questions that you may have at this time.
______
Questions Submitted for the Record to Seto J. Bagdoyan
Questions Submitted by Hon. Chuck Grassley
I would like the record to show that I was unable to attend this
hearing in person. As Chairman of the Senate Judiciary Committee, I was
overseeing the debate about the Supreme Court vacancy. The two hearings
were at the same time.
It was pretty disturbing to hear the report of your undercover
testing of the Federal Marketplace last year.
Eleven of 12--that's over 90%--of fictitious GAO phone or online
applicants were able to obtain a total of $30,000 in Obamacare
subsidies and were able to maintain coverage throughout 2014. People
signed up and received subsidies and coverage in some cases without any
documentation.
And as shocking as that is, today's report is even worse. You
reveal vulnerabilities at the data hub of HealthCare.gov that could
cost taxpayers billions of dollars.
Question. Your report found that CMS does not track the responses
to inquiries made of Federal agencies at the data hub. If CMS does not
track or analyze this information, how can eligibility for Obamacare
and subsidies be verified, as required by law?
Answer. In accordance with our audit objectives, our work focused
on macro-
analysis of data hub query outcomes, and did not address the process of
making eligibility determinations for individual applications.
Thus, in our February 2016 report, we found that although the data
hub plays a key role in the eligibility and enrollment process, Centers
for Medicare and Medicaid Services (CMS) officials said the agency does
not track the extent to which the Federal agencies deliver responsive
information to a request, or, alternatively, whether they report that
information was not available. Additionally, CMS officials said they do
not analyze data provided in response to data hub inquiries.\1\
---------------------------------------------------------------------------
\1\ See GAO, Patient Protection and Affordable Care Act: CMS Should
Act to Strengthen Enrollment Controls and Manage Fraud Risk, GAO-16-29
(Washington, DC: February 23, 2016). A central feature of enrollment
controls under the Patient Protection and Affordable Care Act (PPACA)
is the Federal ``data services hub'' (data hub), which, among other
things, provides a vehicle to check applicant-provided information
against a variety of data sources. In particular, the act requires that
consumer-submitted information be verified, and that determinations of
eligibility be made, through either an electronic verification system
or another method approved by the Department of Health and Human
Services (HHS). To implement this verification process, CMS developed
the data hub, which acts as a portal for exchanging information between
the Federal Health Insurance Marketplace, State-based marketplaces, and
Medicaid agencies, among other entities, and CMS's external partners,
including other federal agencies.
By analyzing the outcomes of data hub inquiries and, in particular,
clarifying the nature and extent of inconsistencies arising from this
process, CMS could, for example, assess whether other sources of data
could be useful to provide more current information on applicant
income.\2\ There may also be correlations observed between various
types of applicants and types of information available from data
sources.
---------------------------------------------------------------------------
\2\ An ``inconsistency'' arises when an applicant's information
does not match information from marketplace data sources.
We concluded that by not assessing the extent to which data
provided through the data hub matches applicant-provided information,
CMS foregoes analysis of the extent to which responding agencies
successfully deliver applicant verification information in response to
data hub requests. Without such an analysis, CMS foregoes information
that could suggest potential program issues or potential
vulnerabilities to fraud, as well as information that might be useful
---------------------------------------------------------------------------
for enhancing program management.
Question. Monitoring activities are an important part of fraud
prevention. What does CMS need to do in order to implement effective
monitoring activities?
Answer. In our February 2016 report, we recommended that CMS should
conduct a comprehensive feasibility study on actions it can take to
monitor and analyze, both quantitatively and qualitatively, the extent
to which data hub queries provide requested or relevant applicant
verification information, for the purpose of improving the data-
matching process and reducing the number of applicant inconsistencies;
and for those actions identified as feasible, create a written plan and
schedule for implementing them. The Department of Health and Human
Services (HHS), CMS's parent agency, concurred with our recommendation.
Question. Rather than track the amounts of the subsidies, CMS
compiles the number of individuals or households affected when a
subsidy is terminated for incomplete information. These changes to
these subsidies have a cost to taxpayers. Federal internal control
standards state that managers need financial information to make
operating decisions, among other activities. How can CMS improve in
this area in order to be a better steward of the taxpayers' dollars?
Answer. In our February 2016 report, we recommended that CMS track
the value of advance premium tax credit and cost-sharing reduction
subsidies that are terminated or adjusted for failure to resolve
application inconsistencies, and use this information to inform
assessments of program risk and performance.\3\ HHS concurred with our
recommendation.
---------------------------------------------------------------------------
\3\ For qualifying applicants, PPACA provides two possible forms of
subsidies for consumers enrolling in individual health plans, both of
which are paid directly to insurers on consumers' behalf. One is a
Federal income tax credit, which enrollees may elect to receive in
advance, and which reduces a consumer's monthly premium payment. This
subsidy is known as the advance premium tax credit. The other, known as
cost-sharing reduction, is a discount that lowers the amount consumers
pay for out-of-pocket charges for deductibles, coinsurance, and
copayments.
Question. In your testimony, you explained that GAO had used a
``secret shopper'' technique to explore possible vulnerabilities to
---------------------------------------------------------------------------
fraud within HealthCare.gov.
Is this technique a well-accepted tool for government and private
auditors to use?
Answer. We have not examined the prevalence of covert testing in
other settings, but GAO has conducted undercover work, including the
use of fictitious applicants, to test controls in other Federal
programs, in addition to health-care coverage provided under the
Patient Protection and Affordable Care Act. GAO has received requests
for undercover work from chairs or ranking members of a number of
congressional committees and subcommittees.
Question. How do you interpret your findings of this particular
``secret shopper'' activity?
Answer. Through covert vulnerability testing, we obtained Federal
Health Insurance Marketplace (Marketplace) approval of subsidized
coverage for 11 of 12 fictitious applicants for 2014, with coverage
continuing into 2015. These results, while not generalizable,
nevertheless illustrate that the Marketplace enrollment process is
vulnerable to fraud. Further, for a second round of testing--for 2015
coverage, and which also included two State marketplaces in addition to
the Federal Marketplace--we obtained subsidized marketplace coverage
for fictitious applicants in 10 of 10 instances.\4\
---------------------------------------------------------------------------
\4\ See GAO, Patient Protection and Affordable Care Act:
Preliminary Results of Undercover Testing of the Federal Marketplace
and Selected State Marketplaces for Coverage Year 2015, GAO-16-159T
(Washington, DC: October 23, 2015). In addition to our 10 applications
for subsidized private health plans, we also made eight additional
fictitious applications for Medicaid coverage, in order to test the
ability to apply for that program through the marketplaces. In these
tests, we were approved for subsidized health-care coverage for seven
of the eight applications. For three of the eight applications, we were
approved for Medicaid, as originally sought. For four of the eight
applications, we did not obtain Medicaid approval, but instead were
subsequently approved for subsidized qualified health-plan coverage.
Thus, for the second round of testing overall, we obtained coverage for
17 of 18 applicants.
Question. In your professional opinion, are the findings of this
---------------------------------------------------------------------------
``secret shopper'' activity concerning?
Answer. As noted above, our results illustrate that the Marketplace
enrollment process is vulnerable to fraud. A program of this scope and
scale is inherently at risk for errors, including improper payments and
fraud. Accordingly, it is essential that effective enrollment controls
are in place to help narrow the window of opportunity for such risk and
safeguard the government's investment. Based on our testing and related
work, we concluded that CMS has assumed a passive approach to
identifying and preventing eligibility and enrollment fraud. In
February 2016, we recommended that CMS conduct a fraud risk assessment,
consistent with best practices provided in GAO's framework for managing
fraud risks in Federal programs, of the potential for fraud in the
process of applying for qualified health plans through the Federal
Marketplace.\5\ HHS concurred with our recommendation.
---------------------------------------------------------------------------
\5\ See GAO, A Framework for Managing Fraud Risks in Federal
Programs, GAO-15-593SP (Washington, DC: July 2015).
______
Questions Submitted by Hon. Rob Portman
Question. I am extremely concerned that a large number of enrollees
in Ohio and across the country may be receiving coverage when they are
not legally eligible to do so, because they have already been offered
coverage through their employers. Many are unaware that they will owe
enormous tax bills at the end of the year, all because HHS lacks the
ability to verify whether an applicant has actually been offered
coverage and is providing subsidies when they should not.
That is why I introduced S. 1996, the Commonsense Reporting Act.
This would allow employers to prospectively report to the IRS before
the open enrollment period opens whether they are offering coverage to
their employees. The government would definitively have this
information well before enrollment begins--today, they only receive
this information after employees may already be enrolled in coverage
and receiving subsidies, meaning the IRS will have to claw back the
money.
Do you believe it would help make subsidy approval more accurate if
IRS and HHS had this type of information prior to open enrollment
beginning?
Why has HHS been unable to help tax payers avoid these repayment
penalties?
Do you believe it would lower the instances of employees who are
offered coverage from being hit with surprise tax bills because they
were not eligible for the subsidies they received?
Answer. We did not report on the availability of employer-sponsored
insurance at time of enrollment, and thus we cannot offer any
observations.
Question. Your report also identifies that CMS is not tracking the
extent to which agencies respond to individual's inaccuracies. Can you
explain how this information could potentially be used by CMS, if they
were to track it?
Answer. Please see the answer to Senator Grassley's first question.
As discussed there, by not tracking outcomes of data hub queries, CMS
foregoes information that could suggest potential program issues or
potential vulnerabilities to fraud, as well as information that might
be useful for enhancing program management.
Question. Wouldn't a better system provide timely updated
information about the availability of employer-sponsored insurance at
the time of enrollment?
Answer. We did not report on the availability of employer-sponsored
insurance at time of enrollment, and thus we cannot offer any
observations.
______
Questions Submitted by Hon. Dean Heller
Question. As of March 2016, there were 88,145 Nevadans on the
Exchange, but only 76,821 of these enrollees have selected a plan.
How can CMS count consumers as ``enrollees'' if they have not
``enrolled'' in a plan?
What barriers exist that would have prevented nearly 12,000
Nevadans from selecting a plan?
Answer. According to the HHS Office of the Assistant Secretary for
Planning and Evaluation (ASPE), Nevada activity for the 2016 open
enrollment period (November 1, 2015 to February 1, 2016), as of March
2016, was as follows:
------------------------------------------------------------------------
STAGE OF APPLICATION NUMBER
------------------------------------------------------------------------
A. Total completed applications 93,255
B. Total individuals applying for coverage in completed 134,454
applications
C. Total individuals eligible to enroll in a marketplace 107,525
plan
D. Number of individuals eligible to enroll in a marketplace 89,716
plan with financial assistance
E. Number of individuals who selected a marketplace plan 88,145
F. Number of individuals with 2016 marketplace plan 76,821
selections with advance premium tax credit
------------------------------------------------------------------------
Notes: For items A-E, for complete details, including explanatory notes
on the figures shown, see Department of Health and Human Services,
Office of the Assistant Secretary for Planning and Evaluation,
``Addendum to the Health Insurance Marketplaces 2016 Open Enrollment
Period: Final Enrollment Report,'' ASPE Issue Brief (March 11, 2016).
For item F, for complete details and explanatory note, see Department
of Health and Human Services, Office of the Assistant Secretary for
Planning and Evaluation, ``Health Insurance Marketplaces 2016 Open
Enrollment Period: Final Enrollment Report,'' ASPE Issue Brief (March
11, 2016).
______
Question Submitted by Hon. Michael F. Bennet
Question. The investigation concludes that no cases of real world
fraud were found. Did GAO take measures to identify cases of actual
fraud in the scope of the overall investigation?
Answer. As we said in our March 17, 2016 testimony, the purpose of
our work was to seek to identify eligibility and enrollment control
vulnerabilities, and not to attempt to identify actual cases of fraud.
As noted above, our covert vulnerability testing demonstrated that the
Marketplace enrollment process is vulnerable to fraud, as 11 of 12
applicants in our tests obtained coverage through fraudulent means.
______
Questions Submitted by Hon. Sherrod Brown
Question. CMS is responsible for working in collaboration with
public and private entities--including other Federal agencies, State
Medicaid agencies, private contractors, health insurance issuers, and
not-for-profit organizations--to manage the Affordable Care Act (ACA)
marketplace.
As part of its responsibilities, CMS must ensure accurate
eligibility determinations, process enrollments, facilitate Medicaid
enrollment for those who qualify, and communicate timely and accurate
information to issuers and consumers.
CMS also provides support functions for the State marketplaces and
administers Federal financial assistance and premium stabilization
programs related to the marketplaces.
The GAO report discusses some inconsistencies that have come up as
CMS has balanced these priorities during a period of time where
Republicans were doing everything they could to sabotage the law's
implementation. But the GAO report fails to acknowledge two things:
These inconsistencies are not indicative of any
widespread fraud by real-world enrollees, and
The significant work CMS has undertaken to resolve
inconsistencies by either eliminating coverage or adjusting an
individual's advance premium tax credit.
Isn't it true that, in order to correct these inconsistencies,
during the first nine months of 2015, CMS ended the enrollments of
approximately 471,000 individuals because they failed to properly
verify their identities?
Isn't it also true that CMS has adjusted the tax credits of
approximately 1,153,000 households whose incomes could not be properly
verified?
Answer. We cannot comment on these figures, as CMS provided us with
statistics covering a different period, which we included in our
February 2016 report. Specifically, according to CMS officials, from
April through June of 2015, enrollment in coverage through the Federal
Marketplace was terminated for about 306,000 consumers with citizenship
or immigration status data-matching issues who failed to produce
sufficient documentation. In addition, according to the officials,
about 735,000 households with income inconsistencies had their advance
premium tax credit or cost-sharing reduction subsidies adjusted for
coverage year 2015.
In February 2016, we also reported on results of a GAO analysis of
application inconsistencies and subsequent resolutions for the 2014
coverage year. We found that for a group of about 1.1 million
applications with a total of about 2 million inconsistencies, about 34
percent of the inconsistencies, with about $1.7 billion in associated
subsidies, remained open as of April 2015--that is, still open several
months following the close of the 2014 coverage year.
Otherwise, as noted above, the objectives of our work were to test
enrollment controls and identify vulnerabilities, if any, and not to
determine the extent of fraud perpetrated by actual enrollees.
Question. Has the GAO elsewhere acknowledged the work CMS did to
correct these inconsistencies?
Answer. Yes, our February 2016 report (p. 34) reflects CMS's
actions on terminations and adjustments. Also in that report (beginning
at pp. 17 and 45), as noted above, we presented results of an analysis
of application inconsistencies, including by type of inconsistency
(such as income or citizenship/immigration status) and resolution.
Terminations of policies and adjustments of subsidies were among
resolutions we reported.
______
Prepared Statement of Erin Bliss, Assistant Inspector General for
Evaluation and Inspections, Office of Inspector General, Department of
Health and Human Services
healthcare.gov: case study of cms management of the federal marketplace
Good morning, Chairman Hatch, Ranking Member Wyden, and other
distinguished Members of the Committee. I am Erin Bliss, Assistant
Inspector General for Evaluation and Inspections in the Office of
Inspector General (OIG), U.S. Department of Health and Human Services
(HHS or the Department). Thank you for the opportunity to testify about
OIG's case study reviewing the management of the Federal Marketplace
website HealthCare.gov by the Centers for Medicare and Medicaid
Services (CMS).
OIG's mission is to protect the integrity of HHS programs and the
health and welfare of the people they serve. We advance our mission
through a nationwide network of audits, evaluations, investigations,
enforcement actions, and compliance efforts. OIG has identified
oversight and operation of the Health Insurance Marketplaces as a Top
Management Challenge for HHS.
The case study is an important component of our marketplace
oversight strategy. It primarily examines implementation of
HealthCare.gov, the consumer-facing website for the Federal
Marketplace, by CMS from passage of the Patient Protection and
Affordable Care Act (ACA) in 2010 through the second open enrollment
period in 2015. As required by the ACA, HealthCare.gov is the Federal
website that facilitates purchase of private health insurance for
consumers who reside in States that did not establish health insurance
marketplaces. At its highly publicized launch on October 1, 2013, and
for some time after, HealthCare.gov users experienced substantial
website outages and technical malfunctions. After corrective action by
CMS and contractors following the launch, CMS ended the first open
enrollment period with 5.4 million individuals having selected a plan
through the Federal Marketplace.
oig's strategy for oversight of the marketplaces
OIG has completed and planned a significant body of audits and
evaluations regarding the Federal Marketplace and other ACA provisions
of high interest and concern to the Department, Congress, and other
stakeholders. OIG's marketplace oversight strategy focuses on four
areas that we have determined to be most critical: payment,
eligibility, management and administration, and security.
My testimony focuses on the OIG report ``HealthCare.gov: Case Study
of CMS Management of the Federal Marketplace'' (OEI-06-14-00350)
released on February 23, 2016. The case study report evaluates CMS's
implementation and management of HealthCare.gov. Consistent with the
OIG's statutory purpose to promote economy, efficiency, and
effectiveness in the administration of Departmental programs, the
rollout of HealthCare.gov presented a unique opportunity to assess
CMS's management and operations. The implementation of HealthCare.gov
provides lessons that will be increasingly important as the success of
Government programs becomes more dependent on the effective
intersection of policy, technology, and management. The case study
enabled OIG to draw conclusions about factors that contributed to the
website's breakdown and subsequent improvement, and lessons learned to
promote effective Government operations moving forward.
In summary, our case study report provides three takeaways about
the development and implementation of HealthCare.gov, presented in
chronological order over a 5-year period from passage of the ACA
through the Marketplace's second open enrollment period:
Development and Launch: The poor launch of the website was
caused by many avoidable organizational missteps, in addition
to problems with website technology;
Correction Through Second Open Enrollment Period: After the
breakdown, CMS improved processes and worked with contractors
and others to fix the website, and this approach led to broader
organizational changes focused on leadership, decisionmaking,
and communication; and
Call for Continued Progress: Challenges remain in managing the
Federal Marketplace and improving operations and services
provided by HealthCare.gov, including issues identified in
related OIG reports. CMS must continue applying lessons learned
from HealthCare.gov to complete this work and address new
challenges as they arise.
background on the federal marketplace and healthcare.gov
The ACA was signed into law on March 23, 2010, and amended on March
30, 2010.\1\ The ACA required the establishment of a health insurance
exchange (marketplace) in each State that would be operational on or
before January 1, 2014.\2\ For States that elected not to establish
their own marketplaces, the Federal Government was required to operate
a marketplace on behalf of the State.\3\
---------------------------------------------------------------------------
\1\ Pub. L. No. 111-148 (March 23, 2010), as amended by the Health
Care and Education Reconciliation Act of 2010, Pub. L. No. 111-152
(March 30, 2010), collectively referred to as the Affordable Care Act
(ACA).
\2\ Ibid. Sec. 1311(a), (b).
\3\ Ibid. Sec. 1321(c).
The marketplaces provide those seeking health insurance a single
point of access to view qualified health plan (health plan) \4\
options, determine eligibility for coverage, and purchase insurance
coverage. Individuals also use the marketplaces to determine
eligibility for insurance affordability programs (e.g., Medicaid,
premium tax credits, and cost-sharing reductions) that lower insurance
premiums and costs of care.\5\ At the beginning of the third open
enrollment period, November 1, 2015, the Federal Government operated a
marketplace (the Federal Marketplace) for 38 States, including 7 State-
partnership marketplaces for which HHS and the State share
responsibilities for core functions and 4 federally supported State
marketplaces in which States perform most marketplace functions.\6\
Thirteen States (including the District of Columbia) operated their own
State marketplaces.\7\
---------------------------------------------------------------------------
\4\ Private health insurance plans certified as meeting certain
standards and covering a core set of benefits including doctor visits,
preventive care, hospitalization, and prescriptions.
\5\ ACA Sec. Sec. 1401, 1402.
\6\ The Henry J. Kaiser Family Foundation, State Decisions on
Health Insurance Marketplaces and the Medicaid Expansion, December 17,
2015. Accessed at http://kff.org/health-reform/state-indicator/state-
decisions-for-creating-health-insurance-exchanges-and-expanding-
medicaid/ on January 6, 2016. CMS, Hawaii: For 2016 insurance coverage,
use HealthCare.gov to apply and enroll. Accessed at https://
www.healthcare.gov/hawaii-2016/ on January 6, 2016.
\7\ Ibid.
CMS has had responsibility for managing the marketplace programs
since January 2011.\8\ To implement the ACA provisions related to the
marketplaces, CMS has worked in collaboration with public and private
entities, including other Federal agencies as required by the ACA,\9\
State Medicaid agencies, private contractors, health insurance issuers
(issuers), and not-for-profit organizations. As it continues to operate
the Federal Marketplace, CMS must ensure accurate eligibility
determinations, process enrollments, facilitate Medicaid enrollment for
those who qualify, and communicate timely and accurate information to
issuers and consumers. CMS also provides support functions for the
State marketplaces and administers Federal financial assistance and
premium stabilization programs related to the marketplaces.
---------------------------------------------------------------------------
\8\ 76 Fed. Reg. 4703 (Jan. 26, 2011).
\9\ ACA Sec. Sec. 1411, 1412.
HealthCare.gov is the public website for the Federal Marketplace
through which individuals can browse health insurance plans, enroll in
plans, and apply for Federal financial assistance to help cover their
premiums and other costs. This is the consumer-facing, or ``front
end,'' portion of the marketplace. The ``back end'' systems of the
Federal Marketplace perform functions such linking consumers'
information from HealthCare.gov to multiple supporting systems that
---------------------------------------------------------------------------
facilitate the enrollment process and payment to issuers.
Key components of HealthCare.gov and the Federal Marketplace
include an identity management system to enable consumers to create
accounts and verify their identities; the Data Services Hub, which
routes information requests from the marketplaces to other Federal
agencies and back, such as the Internal Revenue Service (IRS); and the
Federally-facilitated Marketplace (FFM) that comprises the core of the
overall system. The FFM includes three main subcomponents to facilitate
various aspects of acquiring health insurance: eligibility and
enrollment determinations, plan management, and financial management.
oig's case study approach
The objective of the case study was to gain insight into CMS
implementation and management of the Federal Marketplace, focusing
primarily on HealthCare.gov. The case study identifies organizational
factors that contributed to the website's poor launch and subsequent
improvement, and lessons for employing core management principles in
navigating program implementation and change. These organizational
factors and the lessons learned identify principles that can contribute
not only to improving the Marketplace, but also contribute to improving
the economy, efficiency, and effectiveness of the Department's other
programs and operations.
Our review examined the 5-year period from March 2010 to February
2015, providing a chronology of events and identifying factors that
contributed to the website's breakdown at launch, its recovery
following corrective action, and implementation of HealthCare.gov
through the second open enrollment period. In conducting this review,
we interviewed current and former HHS and CMS officials, staff, and
contractors involved with the development and management of the
website. We also reviewed thousands of HHS and CMS documents, including
management reports, internal correspondence, and website development
contracts.
oig findings from preparation and development of healthcare.gov
(march 2010-september 2013)
The development of HealthCare.gov faced a high risk of failure,
given the technical complexity required; the fixed deadline; and a high
degree of uncertainty about mission, scope, and funding. Still, we
found that HHS and CMS made many missteps throughout development and
implementation. Most critical was the absence of clear leadership,
which caused delays in decisionmaking, lack of clarity in project
tasks, and the inability of CMS to recognize the magnitude of problems
as the project deteriorated.
The HealthCare.gov project encountered problems at the beginning of
development that set the stage for the poor launch. Implementing the
Federal Marketplace required substantial policy development and
decisionmaking to inform technical planning and implementation of the
website. This included not only writing regulations to govern the
marketplaces, but also establishing partnerships with other entities
involved in implementation, such as other departments, States, and
issuers. This policy work was made more difficult and protracted by a
lack of certainty regarding the mission, scope, and funding for the
Federal Marketplace and website and by varying internal and external
expectations for the marketplaces. Delays caused by the lack of
certainty used valuable time and made an already compressed time frame
more difficult.
Additionally, the project's poor transition to CMS after 10 months
in the HHS Office of the Secretary resulted in problems that lasted
long after the move. Initial work in the HHS Office of the Secretary
made significant strides in establishing the policy framework, but did
not focus attention on planning for the project's longer-term technical
and operational needs. CMS had to reconfigure roles and timelines,
determine how it would leverage its resources, and begin work behind
schedule. Further, while CMS's infrastructure and experience provided
greater resources for the project, it led to the Federal Marketplace
operating within a large bureaucratic structure that separated
contract, policy, and technical staff, further diffusing the project
team and making implementation more complex. Our review found that CMS
leadership failed to foster effective collaboration and communication,
particularly between CMS policy and technical staff and with
contractors.
Lack of clear project leadership led to project diffusion and poor
coordination. From the beginning and well into the project, CMS did not
assign clear project leadership, which was particularly problematic for
the policy and technological work needed to set up HealthCare.gov. For
example, CMS continued to make changes to the project's business
requirements that then changed technical aspects of the website build,
in large part because mid-level staff and managers did not have clear
direction or the authority to make decisions. Effective leadership
would have enabled a comprehensive view across the project to better
identify problems and determine priorities. Instead, lack of a single
lead entity inhibited progress assessments and changing course as
needed.
IT contracting for the FFM encountered significant problems. CMS
mismanagement of the key HealthCare.gov contract continued throughout
the website build. CMS did not employ an acquisition strategy to
develop contracts and solicit contractors, a tool used to precisely
assess project needs and make a systematic assessment of the
contractors' ability to meet those needs. Further, due to CMS's
contracting process and uncertainty about funding and specifications,
CMS received a limited number of bids for the contract. CMS hired CGI
Federal to build the core of the overall FFM system, as well as the
online application for consumers. CMS oversight of the contract was
disjointed and spread across different divisions with little
coordination. CMS made frequent changes to contract specifications, and
did not effectively communicate these changes or adequately assess how
they would affect staffing and schedules.
Despite many warnings of substantial problems, CMS moved forward
without serious discussion of delaying the launch. Throughout the
course of building HealthCare.gov, staff at HHS and CMS, as well as
outside entities, identified problems with the program and warned that
these problems warranted action. In all, CMS received 18 ``documented
warnings'' of concerns regarding HealthCare.gov between July 2011 and
July 2013. These documented warnings contained substantial detail about
the project's shortcomings and were formally submitted to CMS senior
leadership or project managers at CMS. However, these reports were not
shared broadly due to diffuse leadership and poor communication. As a
result, no one person in CMS had a comprehensive view of the poor
progress and, given the problems were complex, information became
unwieldy and difficult to prioritize. Without a single comprehensive
view, CMS leadership and staff took little action to respond to
warnings, remained overly optimistic about the launch, and developed
few contingency plans. As the project degraded further and problems
became more well-known, CMS officials and staff became desensitized to
bad news about progress.
In early 2013, CMS attempted to take corrective action, but these
efforts were largely unsuccessful because they were not fully and
diligently executed. For example, after criticism that there was no
clear leadership, CMS assigned its newly appointed Chief Operating
Officer in early 2013 to head the Federal Marketplace program, but the
assignment was not formally announced, the position was not supported
by clear responsibilities, and the designee had an already large
responsibility as CMS Chief Operating Officer. As another example, a
CMS advisor recommended that the project hire a technical systems
integrator to coordinate operations, and CMS and contractors discussed
this need at several points in the project. However, in correspondence
and congressional testimony, it was clear CMS technical leadership
perceived that CMS itself was already serving as the systems
integrator.\10\ CGI Federal managers reported that the lack of a true
systems integrator created extra work that was outside the scope of
their contract.
---------------------------------------------------------------------------
\10\ U.S. House of Representatives, House Energy and Commerce
Committee, PPACA Implementation Failures: Answers from HHS, October 30,
2013.
Due to the poor contract management and ensuing delays, the final
months of development and implementation for HealthCare.gov were
chaotic. CMS continued to make changes to business requirements and
technical specifications well into 2013, delaying development to a
point where it was not feasible to complete and test the website as
initially planned. Critical tasks went uncompleted, including testing
website functionality and security and ensuring adequate capacity for
users. CMS continued with the same plans for a full launch. Changing
the project's path would have required a leader or team to conduct a
comprehensive assessment of status, and to either possess the authority
to alter tasks and processes or to fully communicate that assessment to
leaders with authority. Instead, CMS and contractors continued with the
initial strategy and goals, falling further behind schedule, with
largely the same diffuse leadership structure, staffing, and project
---------------------------------------------------------------------------
plan.
By the time CMS took more drastic action to change the project's
path in August and September of 2013, it was too late to adequately
affect change, given the substantial need for progress and improved
execution. CMS cut functions that were at one time considered critical
to a successful launch, such as the Spanish language and SHOP websites,
to divert resources to the main build. This occurred in the last few
weeks before launch, when developers and testers reported they were
months behind schedule. The rush to launch affected all aspects of the
build, including moving forward with only an interim authorization to
operate and requesting double computing capacity late in September. CMS
sought to deliver a version of HealthCare.gov that had only the minimum
necessary functions to operate, but did so without a comprehensive and
thoughtful strategy.
oig findings from launch, correction and turnaround of healthcare.gov
(october 2013-february 2015)
HealthCare.gov launched at midnight on October 1, 2013, and
experienced substantial problems within hours. The website received
five times the number of expected users, but the problems involved more
than capacity. The website entry tool was overwhelmed, and software
code defects caused malfunctions. Fixing the website required
substantial corrections to the software code and to further increase
capacity. Compounding problems further, some responsible staff were
furloughed when the Government shut down on October 1, 2013.
CMS began corrective action, reorganizing the work to focus on key
priorities and to improve execution. CMS and contractors quickly
brought in new staff and expertise following the launch, developing an
all-hands environment wherein fixing problems with HealthCare.gov was
the key agency mission. Most of the additional staffing came to the
project within 3 weeks, including technological and project management
experts from CMS, contractors, and the private sector. By late October,
CMS and contractors began to move command center operations,
establishing what would become the formal HealthCare.gov command
center--the Exchange Operations Center (XOC). The structure at the XOC
was based on active coordination between technical and policy staff, a
key component missing during the website preparation and development.
It also employed comprehensive website monitoring tools to identify
problems and assess performance. The widespread attention to the launch
and the number of parties involved could have created bureaucratic
paralysis, but those working on the repairs directed their attention to
immediate action and improved the HealthCare.gov website substantially
in 2 months.
Before the launch, artificial distinctions and divisions among
staff contributed to poor collaboration, lack of communication,
disjointed management, and slow progress. Following the launch, first
with the technological team and then more broadly, CMS promoted a
culture that was ``badgeless'' and ``titleless,'' working as a single
team regardless of employer and job title.
According to CMS, this change in culture fostered a greater sense
of mission and teamwork that further improved daily operations.
CMS initiated organizational change, such as a deeper integration
between policy and technological tasks. The Federal Marketplace and
HealthCare.gov needed expertise and coordination across CMS divisions
and many contractors. CMS integrated the various functions within the
project, which improved daily work. This integration allowed CMS to
identify and address problems more quickly, make informed decisions,
and provide clearer direction to those involved in the website
development and operations. CMS also assigned clear project and
technical leadership, hiring a technical systems integrator, and
restructuring its divisions to allow for greater visibility and
oversight of technical staff and contractors by senior leadership.
This greater sense of ``operational awareness'' also prompted CMS
to plan for and mitigate potential problems by considering
contingencies, building redundant systems, and increasing capacity.
CMS's lack of contingency plans before the launch meant that CMS had
few options when the functionality and computing capacity of
HealthCare.gov encountered problems. Essential to success was
identifying possible problems and developing systems and strategies
specific to each concern.
By the end of the first open enrollment period, CMS had a stable
website that functioned well at high capacity, but some planned
components had yet to be completed. CMS immediately began preparation
for the second open enrollment period to begin seven months later. CMS
practiced what officials called ``ruthless prioritization'' of tasks to
focus on the most urgent needs and functionality. This strategy served
to align goals with available resources, guide daily work and
accountability, and temper unrealistic expectations about results.
According to CMS, officials developed a list of technological needs,
then debated and cut about half of the items requested. Cuts included
key elements of the Federal Marketplace system, such as completion of
the automated financial management system.
This process for strategic and organized prioritization marked a
significant improvement over the rushed reprioritization efforts that
occurred prior to launch. Project documentation indicated that in 2013,
CMS and contractors were frantic to establish basic website
functionality. They pushed forward faulty and untested functionality
and hoped to fix it after the launch. Project documentation indicated
that in 2014, CMS maintained a more disciplined project schedule,
meeting deadlines with a goal to implement only technology that had
what project documentation referred to as optimal functionality, or
``perfect execution.'' When this standard could not be met in time, CMS
identified problems more quickly to allow time to employ contingency
plans. CMS stated that this higher standard led to improved practices
overall, such as targeting earlier deadlines for delivery and imposing
stricter testing standards. For example, the new HealthCare.gov
consumer application, App 2.0, was tested through a ``soft launch''
prior to open enrollment. This approach meant that CMS did not always
deliver according to schedule, but was able to test the application's
functionality prior to use in the second open enrollment.
CMS documentation indicated the technical aspects of the website
and supporting systems performed well during the second open enrollment
period, with no system outages and few consumer reports of problems
applying for coverage or selecting plans. CMS further solidified
project leadership, worked to better align project goals with
resources, and renewed its focus on contract management, particularly
emphasizing the agency-contractor relationship. As of February 1, 2016,
CMS reported that over 9.6 million consumers had selected a health
insurance plan through the Federal Marketplace or had their coverage
automatically renewed.\11\
---------------------------------------------------------------------------
\11\ CMS, Health Insurance Marketplace Open Enrollment Snapshot--
Week 13, February 4, 2016. Accessed at https://www.cms.gov/Newsroom/
MediaReleaseDatabase/Fact-sheets/2016-Fact-sheets-items/2016-02-04.html
on February 8, 2016.
---------------------------------------------------------------------------
call for continued progress
CMS continues to face challenges in implementing the Federal
Marketplace, and in improving operations and services provided through
HealthCare.gov. As CMS moves forward, challenges include improving the
website and systems as planned, such as completing the automated
financial management system and improving consumer tools to select
plans. CMS must also continue to address areas OIG has identified in
past reports as problematic or needing improvement, including contract
oversight, the accuracy of payments and eligibility determinations, and
information security controls.
CMS concurred with OIG's call for continued progress, stating that
it will continue to employ the lessons identified in the case study and
that, since OIG's review, it has implemented several initiatives to
further improve its management. The 10 lessons articulated in the
report comprise core management principles that could apply to other
organizations. They include assigning clear project leadership,
encouraging staff to identify and communicate problems, and better
integrating policy and technological work. OIG will continue to monitor
CMS's actions to address specific recommendations from our work, as
well as its overall management of this program.
In addition to the lessons learned from the case study, OIG has
also completed 12 audits and evaluations of the Federal Marketplace,
which combined make over 30 recommendations to CMS. We continue to
monitor CMS's progress toward implementing these recommendations. OIG
has also published numerous other reports related to State marketplaces
and other significant programs created by the ACA. All of our ACA-
related work is available at: http://oig.hhs.gov/reports-and-
publications/aca/.
OIG has ongoing and planned work in several areas related to
marketplaces, including examining the accuracy of financial assistance
payments for individual enrollees for the Federal Marketplace, analysis
of CMS's oversight of the State marketplaces, and a review of the
funding that established the Federal Marketplace. We are also currently
developing work related to the premium stabilization programs. In
addition, OIG has established relationships with its law enforcement
partners to investigate fraud and closely monitor activities and
concerns related to the marketplaces.
conclusion
The Department, and the health care system generally, are in the
process of implementing major changes to health care delivery. Most of
those changes will depend on the successful implementation of
information technology, but success will require more than just
ensuring that the right code is written or that the right technology is
purchased. As our case study demonstrates, whether these changes will
result in more effective, efficient, and economic health care and human
service programs will depend on the interaction of technology,
management, and policy.
OIG believes the lessons learned identified in the case study may
be beneficial to the Department beyond the operation and management of
the Federal Marketplace. Assessing Departmental management will
continue to be a vital component of OIG's oversight of Department
programs going forward. Many programs or projects that OIG oversees
will not require the same level of coordination or resources required
of the Federal Marketplace; however, the principles identified in the
Case Study can help foster the effectiveness and efficiency of
Departmental and program management.
The growing intersection of programs and technology requires OIG to
grow its own capabilities to provide effective oversight. OIG is
building necessary expertise in data analytics, information technology,
and forensic accounting. Increasing our proficiencies and resources in
these disciplines will allow OIG to provide meaningful analysis to
inform decision-makers and program managers.
Thank you again for inviting OIG to speak with the committee today
to share the results of the case study reviewing CMS management of
HealthCare.gov. I would be happy to submit the case study report for
the record, and I would be happy to answer any questions the committee
may have.
______
Questions Submitted for the Record to Erin Bliss
Question Submitted by Hon. Chuck Grassley
I would like the record to show that I was unable to attend this
hearing in person. As Chairman of the Senate Judiciary Committee, I was
overseeing the debate about the Supreme Court vacancy. The two hearings
were at the same time.
Question. Ms. Bliss, in your recent case study report, you
summarized many of the mistakes at HHS and CMS that resulted in the
disastrous launch of HealthCare.gov.
Some of the problems identified by OIG included a lack of
leadership, a failure to act on warnings and address problems, and
corrections that were too little too late.
In the final countdown to the launch of HealthCare.gov, it seems
CMS was more concerned about launching anything rather than following
the advice of their own advisors and waiting and then releasing a
product that would be useful to Americans.
CMS continues to have problems with implementing and operating the
Federal Marketplace.
Ms. Bliss, what are the 10 lessons CMS should have learned?
Answer. The Centers for Medicare and Medicaid Services (CMS)
continue to face challenges in implementing the Federal Marketplace and
in improving services provided through HealthCare.gov. The HHS OIG
called on CMS to continue progress in applying lessons learned from
HealthCare.gov to avoid future problems and to maintain improvement
across the agency. These 10 lessons comprise core management principles
that address both specific project challenges and organizational
structure, and could apply to other organizations.
1. Assign clear project leadership for cohesion across tasks and
a comprehensive view of progress.
2. Align project and organizational strategies with the resources
and expertise available.
3. Identify and address factors of organizational culture that
may affect project success.
4. Seek to simplify processes, particularly for projects with a
high risk of failure.
5. Integrate policy and technological work to promote operational
awareness.
6. Promote acceptance of bad news and encourage staff to identify
and communicate problems.
7. Design clear strategies for disciplined execution and
continually measure progress.
8. Ensure effectiveness of information technology (IT) contracts
by promoting innovation, integration, and rigorous oversight.
9. Develop contingency plans that are quickly actionable, such as
redundant and scalable systems.
10. Promote continuous learning to allow for flexibility and
changing course quickly when needed.
CMS concurred with OIG's call for continued progress and stated
that it will continue to employ the lessons. Since OIG's review, CMS
has implemented several initiatives to further improve its management.
______
Questions Submitted by Hon. Michael B. Enzi
Question. In the hearing, some of the operational and strategic
changes which were made after the first enrollment period were briefly
discussed. Those changes were described as ``clearer leadership amongst
staff and contractors,'' and ``key organizational changes.''
Please list and describe the specific changes made to make the
leadership amongst HHS staff and contractors more clear.
Answer. Assigning and supporting clear project leadership was
perhaps the most critical change CMS made to improve HealthCare.gov
project management. Prior to the launch, HealthCare.gov lacked clear
project leadership to give direction and unity of purpose,
responsiveness in execution, and a comprehensive view of progress. CMS
and its contractors took a number of steps to improve leadership,
including the following:
1. Tasked top agency management to assume daily leadership of the
HealthCare.gov project, first the CMS Administrator and subsequently
the Principal Deputy Administrator. This served to provide project
visibility across agency functions and clear authority to implement
changes. Following the website recovery and before the start of the
second open enrollment period, CMS hired a Marketplace CEO to serve in
that leadership role. CMS senior leaders also alternated as the
website's Executive on Call, serving 24-hour shifts to make decisions
more rapidly.
2. Improved the coordination and transparency of work shared among
CMS staff and multiple contractors by hiring a systems integrator to
organize technological tasks and report progress and problems to
leadership. CMS continues to employ a contractor to serve as systems
integrator, having established this position within weeks of the launch
in late October 2013.
3. Promoted coordination of CMS divisions working on various
aspects of the Federal and State-based marketplaces by forming the
Strategic Planning and Management Council, composed of five workgroups:
appeals; eligibility and enrollment; plan oversight; security and
privacy; and workforce planning.
4. Shifted from following known processes to continually assessing
outcomes and progress. For example, CMS identified a number of problems
in the lead-up to the second open enrollment through ``deep dives,''
assigning staff to scrutinize the performance of a specific area or
function (such as eligibility) and bring to CMS leadership their
assessments of weaknesses.
5. Focused on blending the policy and technical components of CMS
with a greater sense of what CMS leaders called ``the physics of
operations'' or ``operational awareness.'' Policymakers acquired a
better understanding of the effort required to effectuate policy
decisions, both in terms of time and resources, so that those
considerations could better inform decision making and avoid a
longstanding agency bifurcation between operations and policy.
6. Instructed staff coordinating technological work and
supervising contracts about rules and provisions related to contracted
work; also instructed contractors to work more collaboratively with
each other and the CMS divisions. This served to avoid staff concerns
about violating Government contracting rules, promoting closer
collaboration and a sense that CMS and contractors were a single team
working toward the same goals.
Question. Please list and describe the key organizational changes
mentioned by Ms. Erin Bliss.
Answer. CMS also made other key organizational changes as it
improved the HealthCare.gov website. This required significant and
focused effort to measure website performance, correct problems with
website capacity and functions, and establish a new project structure.
To implement these practices, CMS and its contractors took the
following steps:
1. Began correction of website problems immediately following
launch, changing the project management strategy. CMS and contractors
brought in new staff and expertise following the launch, developing an
all-hands environment wherein fixing problems with HealthCare.gov was
the key agency mission. These changes allowed CMS to make quick
progress in identifying the source of problems and developing a
strategy going forward.
2. Adopted a ``badgeless'' culture that encouraged full
collaboration by CMS staff and contractors regardless of employer
status and job title, fostering innovation, problem solving, and
communication among teams. The enhanced team of CMS staff, contractors,
and technological experts correcting problems with HealthCare.gov
included people at all levels of CMS and contracted entities with
varied experience on the project. First with the technological team and
then more broadly, CMS promoted a culture wherein all team members
could speak out about problems and develop creative solutions.
3. Integrated all functions into the organizational structure to
align with project needs, enhancing CMS and contractor accountability
and collaboration. The Federal Marketplace needed expertise and
personnel across CMS, including policy, technical, and communications
staff, as well as many contractors. Key to the correction, CMS
integrated the various functions both operationally and technically,
improving daily work and promoting the larger project mission. This
integration allowed CMS to identify and address problems more quickly,
make informed decisions, and provide clearer direction to those
involved in the website development and operations.
4. Planned for problems, establishing redundant (backup) systems
in the event of further breakdowns and restructuring the key
development contract to ensure better performance. CMS began to plan
for and mitigate potential problems by considering contingencies,
building redundant systems, and increasing capacity. Given limited
resources, CMS leadership had to analyze past problems with
HealthCare.gov and carefully consider how and to what extent it would
develop new systems and strategies, such as enhancing training for call
center staff. Key to success was identifying all possible problems and
developing systems and strategies specific to the concern.
5. Adopted a policy of ``ruthless prioritization'' to reduce
planned website functionality, focusing resources on the highest
priorities. Because the time frame and resources available to prepare
for the second open enrollment period were fixed, CMS focused on
reducing scope to meet deadlines. The day after first open enrollment
closed, CMS leadership met to prioritize tasks to focus on the most
urgent needs and functionality. These decisions and resulting changes
were then locked down and measured for progress and results. Ruthless
prioritization served to align goals with the resources available,
guide daily work and accountability, and temper unrealistic
expectations about results.
6. Prioritized quality over on-time delivery, employing extensive
testing to identify and fix problems and delaying new website
functionality if unready. CMS adopted a project management approach of
going live with website functionality only when it could ensure what
one CMS official called ``perfect execution.'' This policy of requiring
optimal functioning before delivery led to improved practices overall,
such as targeting earlier deadlines for delivery and imposing stricter
testing standards.
7. Simplified systems and processes to enable closer monitoring of
progress, increased transparency and accountability, and clearer
prioritization. CMS simplified both technical aspects of the build and
the organizational structure of the agency itself by closely monitoring
progress and results with daily reports and close communication with
contractors. Reduced complexity in tasks and organizational structure
made it easier for CMS to identify those responsible for carrying out
tasks and to track progress toward goals.
8. Adopted continuous learning for policy and technological tasks,
balancing project plans with system and team capacity and changing
course as needed to improve operations. In preparation for the second
open enrollment period, much about the HealthCare.gov project was still
unfolding. For example, CMS did not know how much website capacity
consumers would require, and it was still developing and testing new
and improved functionality in the final weeks before open enrollment.
Given that the design and proportion of the project was evolving, it
was critical to CMS's success that the organization continuously learn
as the project progressed. As the HealthCare.gov project matured, CMS's
knowledge and experience became more concrete and its planning more
effective, but the project continued to require adaptation.
Question. In the hearing, Ms. Bliss was asked whether or not the
office of Inspector General had uncovered, in connection with open
enrollment, any confirmed cases of fraud. Ms. Bliss responded that
there are no fraudulent cases which have resulted in criminal
convictions or civil settlements. However, she did mention that there
are some ongoing investigations.
How many ``ongoing investigations'' remain since the first open
enrollment period?
Answer. At this time, OIG has two ongoing investigations
specifically looking at enrollment fraud in the Federally Facilitated
or State-Based Marketplaces. The conduct under investigation may
involve more than one open enrollment period.
Question. There have been concerns about the ``back end'' of the
HealthCare.gov website, specifically as it concerns subsidy payments to
insurance companies.
Please describe the process in which insurers receive payments of
subsidies for premiums and out-of-pocket expenses.
Answer. To date, OIG has performed two audits that specifically
examine the ``back end'' of the HealthCare.gov website.\1\ These
reports examine the interim manual payment process that CMS utilized
during the first 2 years of Marketplace operations. In both reports,
OIG found deficiencies that limited the effectiveness of the interim
manual payment process. Please see these reports for detailed
explanations of the interim process.
---------------------------------------------------------------------------
\1\ HHS OIG, CMS's Internal Controls Did Not Effectively Ensure the
Accuracy of Aggregate Financial Assistance Payments Made to Qualified
Health Plan Issuers Under the Affordable Care Act (A-02-14-02006) (June
2015); and HHS OIG, CMS Could Not Effectively Ensure That Advance
Premium Tax Credit Payments Made Under the Affordable Care Act Were
Only for Enrollees Who Paid Their Premiums (A-02-14-02025) (December
2015). Available at http://www.oig.hhs.gov/reports-and-publications/
aca/.
In response to these reports, CMS explained that it expected all
issuers on the Federal Marketplace would be using an automated policy-
based system in 2016. In December 2015, CMS issued guidance related to
issuer implementation of the automated policy-based system.\2\
According to this policy, issuers are expected to transition to this
system in early 2016. CMS has provided training regarding the new
system, which OIG attended to gain a better understanding as CMS
continues to work with issuers to implement automated policy-based
payments.
---------------------------------------------------------------------------
\2\ HHS CMS, ``Policy-Based Payments Bulletin--INFORMATION''
(December 2015). Available at https://www.cms.gov/CCIIO/Resources/
Regulations-and-Guidance/Downloads/Policy-based
-Payment-Guidance.pdf.
We also plan to conduct work on CMS's automated policy-based
payments system at the Federal Marketplace potentially by looking at
the accuracy of the determination of financial assistance payments and
the use of enrollment and payment data. The CMS Center for Consumer
Information and Insurance Oversight (CCIIO) would be the best source of
---------------------------------------------------------------------------
information about specific details of this process.
Question. What steps have been taken to ensure that those
individuals who have signed up for a plan on the HealthCare.gov website
end up in the right plan?
Answer. To date, OIG has not conducted work specifically assessing
internal controls at HealthCare.gov that would ensure individuals are
enrolled in the Qualified Health Plan they select.
Based on our related work on the interim manual payment process,
controls for confirming enrollment may be included in the initial and
confirmation ``834'' transactions.\3\ OIG's audits did not assess the
834 transactions with respect to this specific issue, and the automated
policy-based payment system may utilize different processes. Due to
those considerations, CMS-CCIIO would be the best source of information
about specific internal controls the Federal Marketplace has in place
to ensure individuals are enrolled in the plans they select.
---------------------------------------------------------------------------
\3\ For more information about the 834 transaction process, see
OIG's report CMS's Internal Controls Did Not Effectively Ensure the
Accuracy of Aggregate Financial Assistance Payments Made to Qualified
Health Plan Issuers Under the Affordable Care Act (A-02-14-02006) (June
2015), pp. 6-7.
Question. Please describe the process of adding new information,
---------------------------------------------------------------------------
like the birth of a child or marriage to an existing plan.
OIG has not assessed the process for adding information to an
existing plan. The HealthCare.gov website provides instructions for
consumers to report information about life changes through their
Marketplace accounts online or by phone. CMS-CCIIO would be the best
source of further information about this process.
______
Questions Submitted by Hon. Michael F. Bennet
Question. As HHS and CMS concur with and move toward implementing
OIG recommendations, how can you continue to be a resource to the
agency based on your findings?
Answer. OIG believes that our case study's assessment of the
intersection of technology, policy, and management, and the lessons
that we identified can benefit a broad range of Federal projects and
programs. These lessons comprise core management principles that
address both specific project challenges and organizational structure
and can be applied broadly.
We will continue to monitor CMS's actions in response to our
recommendations and its overall management of the Federal Marketplace
and other programs. OIG has ongoing and planned work in several areas
related to Marketplaces, including examining the accuracy of financial
assistance payments for individual enrollees for the Federal
Marketplace, analysis of CMS's oversight of the State Marketplaces, and
a review of the funding that established the Federal Marketplace. We
are also developing work related to the premium stabilization programs.
In addition, OIG has established relationships with its law enforcement
partners to investigate fraud related to the Federal Marketplace and
make appropriate referrals to partner agencies. OIG will also draw upon
these lessons to examine HHS's and CMS's implementation and management
of other programs.
Question. The OIG Case Study report published in February 2016
highlights changes that CMS underwent including realignment of project
goals and resources as well as enhancing coordination between CMS staff
and contractors. What are the main operational changes that CMS should
focus on as they work toward optimizing the Federal Marketplace?
Answer. In addition to the lessons learned from the case study, OIG
has also completed 12 audits and evaluations of the Federal
Marketplace, which combined make over 30 recommendations to CMS.\4\ We
continue to monitor CMS's progress toward implementing these
recommendations. OIG has also published numerous other reports related
to State Marketplaces and other significant programs created by the
ACA.\5\
---------------------------------------------------------------------------
\4\ For a list of significant unimplemented recommendations related
to OIG's work on both the Federal and State-based marketplaces, see HHS
OIG, Compendium of Unimplemented Recommendations (April 2016), pp. 34-
40. Available at http://oig.hhs.gov/reports-and-publications/
compendium/index.asp.
\5\ All of OIG's ACA-related work is available at http://
oig.hhs.gov/reports-and-publications/aca/.
With regard to the Federal Marketplace, in particular, successful
implementation of the automated policy-based payment system is one of
the main operational challenges facing CMS in 2016. Additionally, CMS
must continue to refine the eligibility determination systems for the
Federal Marketplace to improve the accuracy and timeliness of those
determinations, and continue to focus on resolving inconsistencies in
---------------------------------------------------------------------------
eligibility determinations.
In general, as CMS improves the consumer-facing enrollment process
for HealthCare.gov, CMS must continue to assess and improve the Federal
Marketplace systems that operate behind the scenes, particularly the
eligibility, administrative, and financial management functions. CMS
must ensure that all pathways for enrollment operate with integrity and
that consumers' personal information is secure. Vigilant monitoring and
testing and rapid mitigation of identified vulnerabilities are
essential. Attention must be paid to sound operation of financial
assistance and the risk corridor, reinsurance, and risk-adjustment
programs. CMS must ensure that consumers and issuers receive accurate
Marketplace information, including information relevant for tax
purposes, such as Form 1095A tax forms. Furthermore, Marketplaces must
continue to protect personally identifiable information and strengthen
security controls.
______
Questions Submitted by Hon. Sherrod Brown
Question. We know that HealthCare.gov had a rocky start. But--as
was acknowledged in the HHS OIG report--after just 2 months (during
which a wholly preventable government shutdown occurred, furloughing
relevant members of staff), the Centers for Medicare and Medicaid
Services (CMS) managed to ``substantially'' improve the website.
By the end of the first open enrollment period, CMS had a website
that functioned at high capacity. The agency ended the first open
enrollment period with 5.4 million individuals having selected a plan
through the Federal Marketplace.
Ms. Bliss, what corrective actions were taken by CMS and
contractors following the website's launch to address the lack of
communication and leadership challenges that existed during the first
enrollment period?
Answer. Assigning and supporting clear project leadership was
perhaps the most critical change CMS made to improve HealthCare.gov
project management. Prior to the launch, HealthCare.gov lacked clear
project leadership to give direction and unity of purpose,
responsiveness in execution, and a comprehensive view of progress. To
improve project leadership, top CMS leadership assumed daily management
of the HealthCare.gov project, first the CMS Administrator and later
the Principal Deputy Administrator. This served to provide a view
across agency functions and clear authority to implement changes.
Following the website recovery and before the start of the second open
enrollment period, CMS hired a Marketplace CEO to serve in that role
and coordinate project activities across the agency. CMS senior leaders
also alternated as the website's Executive on Call, serving 24-hour
shifts to make decisions more rapidly.
CMS also worked to improve the coordination and transparency of
work shared among CMS staff and multiple contractors. These efforts
included hiring a systems integrator to organize technological tasks
and report progress and problems to leadership and merging the work of
policy and technical staff more closely to allow for a better sense of
the work needed to complete project goals and integrate CMS staff and
contractors into a single team. CMS also promoted coordination of CMS
divisions working on various aspects of the Federal and State-based
marketplaces by forming the Strategic Planning and Management Council,
comprising subject area workgroups such as health plan oversight and
workforce planning. This improved coordination across divisions allowed
CMS to shift from separated, engrained processes to continually
assessing outcomes and progress.
CMS made other key organizational changes as it improved the
HealthCare.gov website. Following the poor launch, CMS and contractors
pivoted quickly to corrective action, reorganizing the work to focus on
key priorities and improve execution. These efforts benefited from CMS
adopting a ``badgeless'' culture that encouraged full collaboration by
CMS staff and contractors regardless of employer status and job title,
fostering innovation, problem solving, and communication among teams.
The agency also planned for problems, establishing redundant (backup)
systems in the event of further breakdowns and restructuring its key
development contract to ensure better performance.
Question. What has CMS done to address these challenges and
continue to improve operations through the second and third open
enrollments? How do you anticipate CMS will build upon these
improvements for the upcoming open enrollment period?
Answer. In preparation for the second open enrollment period, CMS
restructured project tasks to set more realistic goals and meet core
objectives. For example, CMS managers met immediately following the
first open enrollment to ``ruthlessly prioritize'' remaining tasks and
focus resources on the highest priorities. This served to align goals
with the resources available, guide daily work and accountability, and
temper unrealistic expectations about results. CMS also prioritized
quality over on-time delivery, employing extensive testing to identify
and fix problems and delaying new website functionality if unready for
perfect execution. This contrasted with the launch of HealthCare.gov
wherein CMS delivered what it knew was faulty functionality, planning
to improve the website later.
To manage these efforts, CMS simplified systems and processes to
enable closer monitoring of progress, increased transparency and
accountability, and clearer prioritization. Reduced complexity in tasks
and organizational structure made it easier for CMS to identify those
responsible for carrying out tasks and to track progress toward goals.
At the same time, the broader CMS organization adopted a strategy of
continuous learning for policy and technological tasks, balancing
project plans with system and team capacity and changing course as
necessary to improve operations. Given that the design and proportion
of the HealthCare.gov project was evolving, it was critical to CMS's
success that the organization continuously learned as the project
progressed. CMS continued these strategies, experiencing few technical
problems and no system outages during the second open enrollment
period.
Still, the agency faces ongoing challenges in implementing the
Federal Marketplace and in improving operations and services provided
through HealthCare.gov. OIG called on CMS to continue progress in
applying lessons learned from HealthCare.gov to avoid future problems
and to maintain improvement across the agency. As CMS moves forward,
challenges include improving the website and systems as planned, such
as completing the automated financial management system and improving
consumer tools to select plans. Also, given CMS's large organization
and complex mission, prior management problems could resurface and new
problems could emerge. CMS placed intense organizational focus on the
Federal Marketplace during the recovery of the website. This level of
focus will, by necessity, change in the face of new challenges and
priorities within CMS, and inevitably officials and staff with key
expertise and deep knowledge of the Federal Marketplace will leave CMS
or the project. Such changes in priorities and resources reinforce the
need for CMS to fully embed core management principles in its daily
work.
In its comments in response to the Case Study, CMS concurred with
OIG's call for continued progress in applying the lessons that CMS
learned from the HealthCare.gov recovery in its management of the
Federal Marketplace and CMS's broader organization. CMS stated that
since the OIG review it has implemented several initiatives to improve
its management, striving to incorporate principles aligned with this
report's lessons learned in its culture, operations, and daily work.
These principles include a focus on leadership and accountability,
continuous reevaluation of priorities and how the project could be more
efficient, program measurement, and a flexible and evolving IT strategy
aligned with policy requirements. CMS also indicated a commitment to
overcoming challenges and deliver results in a transparent manner. OIG
will continue to monitor CMS operation and management of the Federal
and State-based Marketplaces and HealthCare.gov, focusing on oversight
of critical aspects such as the integrity of enrollment processes and
payment accuracy.
______
Prepared Statement of Hon. Orrin G. Hatch,
a U.S. Senator From Utah
WASHINGTON--Senate Finance Committee Chairman Orrin Hatch (R-Utah)
today delivered the following opening statement at a hearing examining
deficiencies with HealthCare.gov and the current state of the Patient
Protection and Affordable Care Act (PPACA) and how it is impacting
patients and taxpayers:
Good morning. It is a pleasure to welcome everyone to today's
hearing.
Today, we'll be talking with representatives from the Office of
Inspector General for the Department of Health and Human Services (HHS-
OIG) and from the Government Accountability Office (GAO) about their
ongoing oversight work with respect to HealthCare.gov and enrollment in
the Federal health insurance marketplace. I want to thank both entities
for their hard work on these issues and acknowledge the contributions
both have made to help this committee perform more accurate and timely
oversight.
Now, it is no secret that I have never been a fan of the so called
Affordable Care Act. And, as we approach the sixth anniversary of this
law and look closely into how it's working and being implemented, the
evidence overwhelmingly shows that I--and the many others who opposed
the law from the beginning--have been right all along.
The facts speak for themselves.
Since Obamacare was signed into law, HHS-OIG and GAO have
cumulatively released at least six dozen reports detailing various
operation and implementation issues demonstrating the numerous areas
where the law is falling short. These reports are specific and focused
on key operational failures like enrollment controls or system issues,
some of which we'll hear more about today.
Let's keep in mind that GAO and HHS-OIG are not partisan entities.
They are independent watchdogs, tasked with the responsibility of
objectively and dispassionately assessing what is and what is not
working in various Federal programs, including those created or amended
by the Affordable Care Act. And, there is no better record showing how
and why Obamacare is not working than the reports we've received from
these offices.
Today, we are going to specifically discuss operations issues
related to HealthCare.gov and enrollment problems at the Federal
insurance marketplace, otherwise known as the Federal exchange.
Let's start with the HealthCare.gov launch.
As a result of numerous problems and shortcuts taken with the
initial development and deployment of HealthCare.gov and its supporting
systems, consumers encountered widespread performance issues when
trying to create accounts and enroll in health plans. After numerous
inquiries and reports, we now know what ultimately caused these
performance issues.
For example, there was inadequate capacity planning. The Centers
for Medicare and Medicaid Services (CMS) cut corners and did not plan
for adequate capacity to maintain HealthCare.gov and its supporting
systems.
There were also problems with the software that were entirely
avoidable. CMS and its contractors identified errors in the software
coding for the website, but did not adequately correct them prior to
launch.
We saw a lack of functionality as CMS did not adequately prepare
the necessary systems and functions of the website and its supporting
systems prior to the initial launch.
CMS also failed to apply recognized best practices for system
development, which contributed to the problems.
Admittedly, since the initial launch, CMS has taken steps to
address these problems, including increasing capacity, requiring
additional software quality reviews, and awarding a new contract to
complete development and improve the functionality of key systems.
However, many of the problems have still not been entirely resolved
and continue to cause frustration for consumers trying to obtain health
insurance.
I wish we could boil down all of Obamacare's problems to the
functions of a single website. Indeed, if this was just an IT problem,
all of our jobs would be a lot easier. However, the problems with
Obamacare--and the Federal insurance marketplace in particular--go much
deeper and many of them remain unaddressed.
We know, for example, that the enrollment controls for the Federal
marketplace have been inadequate.
During undercover testing by GAO, the Federal marketplace approved
insurance coverage with taxpayer-funded subsidies for eleven out of
twelve fictitious phone or online applicants. In 2014, the GAO
applicants--which, once again, were fake, made-up people--obtained a
total of about $30,000 in annual advance premium tax credits, plus
eligibility for lower insurance costs at the time of service. These
fictitious enrollees maintained subsidized coverage throughout the
year, even though GAO sent either clearly fabricated documents or no
documents at all to resolve application inconsistencies.
While the subsidies, including those granted to GAO's fictitious
applicants, are paid to health-care insurers, they nevertheless
represent a benefit to consumers and a cost to the government. Now, GAO
did find that CMS relies on a contractor charged with document
processing to uncover and report possible instances of fraud. Yet, GAO
also found that the agency does not require that the contractor has any
fraud detection capabilities.
And, according to GAO, CMS has not performed a single comprehensive
fraud risk assessment--a recommended best practice--of the Obamacare
enrollment and eligibility process. Until such an assessment is
completed, CMS is unlikely to know whether existing control activities
are suitably designed and implemented to reduce inherent fraud risk to
an acceptable level.
In other words, CMS isn't even sure if CMS's fraud prevention
systems are designed correctly or if they're effective.
Lastly, while it is not the focus of the reports that will be
covered by the testimony today, another matter we've been tracking
closely, and where the GAO is issuing a report today, is CMS's
oversight of the health care CO-OPs. We had a hearing on this topic in
late January where we examined a number of financial and oversight
related explanations for the abject failure of the CO-OP program.
Today's GAO report describes CMS's efforts to deal with financial
or operations issues at the CO-OPs, including the use of an escalation
plan for CO-OPs with serious problems that may require corrective
actions or enhanced oversight.
As of November 2015, 18 CO-OPs had enough problems that they had to
submit to a CMS escalation plan, including nine that have discontinued
operations. And, just this week, we heard that yet another CO-OP, this
time the one in Maine, is on the verge of financial insolvency, despite
the fact that it had been on a CMS-mandated escalation plan.
In other words, CMS's efforts to address all the problems faced by
CO-OPs appear to have failed, just like virtually every other element
of this program.
The failure of CMS to adequately implement the CO-OP program is
well-
documented here on the Finance Committee and elsewhere. As with so many
other parts of Obamacare, the high-minded rhetoric surrounding this
program has fallen short of reality.
With nearly half of the CO-OPs now closed, the failed experiment
has wasted taxpayer dollars and forced patients and families to
scramble for new insurance. With so many CO-OPs now in financial
jeopardy, I believe that CMS should work with, not against States, to
safeguard taxpayer dollars.
So, as always, we have a lot to discuss. And I look forward to
hearing more from the officials we have testifying here today.
______
Prepared Statement of Hon. Ron Wyden,
a U.S. Senator From Oregon
It is old news that the rollout of HealthCare.gov 3 years ago was
botched. It is new news that the Inspector General of HHS recently
said, quote, ``CMS recovered the HealthCare.gov website for high
consumer use within 2 months, and adopted more effective organizational
practices. . . .''
That quote comes from one of two reports looking back at 2013 and
2014 that the Finance Committee will be presented with today. Let's
recognize that the story here is well-documented. After the launch went
badly, some of the best minds in tech and a new contractor were brought
in, they scrambled to overhaul the system, and the exchange was soon up
and running. And CMS is following up on every one of the Inspector
General's recommendations, which the IG notes in its report.
In the most recent enrollment period, nearly 10 million Americans
used HealthCare.gov to sign up for a plan or re-enroll automatically.
And in my home State, which has had its own problems, close to
150,000 people have used the website to sign up for a plan as of
January 31st. That's up by more than 30 percent compared to last year.
The committee will also hear an update today from the Government
Accountability Office on what's called a ``secret shopper''
investigation. GAO first brought this study before the committee in
July last year. And I'll repeat now what I said back then: On this side
of the aisle, we don't take a back seat to anybody in fighting fraud
and protecting taxpayer dollars. One dollar ripped off is one dollar
too many. But let's recognize that what was true last summer remains
true today. This GAO investigation has not uncovered one single shred
of real-world fraud in the insurance marketplaces. It was built on
fictitious characters with specially-created identities, not real
consumers or real fraudsters.
It's true the GAO found that there are sometimes differences
between the information on somebody's insurance application and their
tax forms and citizenship records. But when it comes to these
inconsistencies in people's data, this investigation can't
differentiate between fraud and a typo.
And meanwhile, HHS does not look the other way when it finds these
red flags. In 2014, the year of GAO's investigation, CMS closed more
than 100,000 insurance policies because documents didn't match or
weren't provided. Tax credits were adjusted for nearly 100,000
households. In 2015, HHS closed more policies and adjusted more tax
credits. If you come at this from the left, you might say that's too
harsh. If you come at it from the right, you might take a different
view. But there is no basis whatsoever for the argument that HHS
ignores problems in people's records or leaves the door open to fraud.
So in closing, rather than rehashing old news, I'd prefer to look
at the facts. Because of the ACA, the number of Americans without
health insurance is at or near its lowest point in half a century. For
the 160 million people who get their insurance from their employer,
premiums climbed only 4 percent last year. Working-age Americans in
Oregon and nationwide with preexisting conditions--80 million people or
more--can no longer be denied insurance.
Now, instead of battling it out over what happened 3 years ago,
let's start pulling on the same end of the rope and solve some real
problems. For example, Democrats and Republicans ought to be working
together to stoke more competition in the insurance marketplace and
bring costs down for consumers.
Next, there are going to be spectacular, new cures available in the
future, and there are real questions as to whether our health care
system will be able to afford them. Senator Grassley and I put together
a bipartisan case study that looked into one such drug, which treats
Hepatitis C. Solving this issue of blockbuster drug costs is going to
take a lot of hard work on a bipartisan basis.
Congress also has a duty to take Medicare's historic guarantee and
reinforce it for a new generation of Americans. In my view, it starts
with revolutionizing the way Medicare handles caring for seniors with
chronic conditions like diabetes, cancer and Alzheimer's. Members on
both sides of this committee, thanks to steadfast work by Senators
Warner and Isakson, and Chairman Hatch's leadership, have built a
bipartisan game plan for chronic care. And it's my hope that the
committee is able to continue its progress on that front.
Those are the kind of health care challenges I believe this
committee should be focused on tackling. With that, I want to thank our
witnesses for being here today.
[all]