oversight

Medical Devices: FDA Has Met Most Performance Goals but Device Reviews Are Taking Longer

Published by the Government Accountability Office on 2012-02-29.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                United States Government Accountability Office

GAO             Report to Congressional Requesters




February 2012
                MEDICAL DEVICES

                FDA Has Met Most
                Performance Goals
                but Device Reviews
                Are Taking Longer




GAO-12-418
                                             February 2012

                                             MEDICAL DEVICES
                                             FDA Has Met Most Performance Goals but Device
                                             Reviews Are Taking Longer
Highlights of GAO-12-418, a report to
congressional requesters




Why GAO Did This Study                       What GAO Found
The Food and Drug Administration             Even though FDA met all medical device performance goals for 510(k)s, the
(FDA) within the Department of Health        elapsed time from submission to final decision has increased substantially in
and Human Services (HHS) is                  recent years. This time to final decision includes the days FDA spends reviewing
responsible for overseeing the safety        a submission as well as the days FDA spends waiting for a device sponsor to
and effectiveness of medical devices         submit additional information in response to a request by the agency. FDA review
sold in the United States. New devices       time excludes this waiting time, and FDA review time alone is used to determine
are generally subject to FDA review via      whether the agency met its performance goals. Each fiscal year since FY 2005
the 510(k) process, which determines if      (the first year that 510(k) performance goals were in place), FDA has reviewed
a device is substantially equivalent to
                                             over 90 percent of 510(k) submissions within 90 days, thus meeting the first of
another legally marketed device, or the
                                             two 510(k) performance goals. FDA also met the second goal for all 3 fiscal
more stringent premarket approval
(PMA) process, which requires
                                             years it was in place by reviewing at least 98 percent of 510(k) submissions
evidence providing reasonable                within 150 days. Although FDA has not yet completed reviewing all of the FY
assurance that the device is safe and        2011 submissions, the agency was exceeding both of these performance goals
effective. The Medical Device User           for those submissions on which it had taken action. Although FDA review time
Fee and Modernization Act of 2002            decreased slightly from FY 2003 through FY 2010, the time that elapsed before
(MDUFMA) authorized FDA to collect           FDA’s final decision increased substantially. Specifically, from FY 2005 through
user fees from the medical device            FY 2010, the average time to final decision for 510(k)s increased 61 percent,
industry to support the process of           from 100 days to 161 days.
reviewing device submissions. FDA
                                             FDA was inconsistent in meeting performance goals for PMA submissions. FDA
also committed to performance goals
that include time frames within which        designates PMAs as either original or expedited; those that FDA considers
FDA is to take action on a proportion of     eligible for expedited review are devices intended to (a) treat or diagnose life-
medical device submissions. MDUFMA           threatening or irreversibly debilitating conditions and (b) address an unmet
was reauthorized in 2007.                    medical need. While FDA met the performance goals for original PMA
                                             submissions for 4 out of 7 years the goals were in place, it met those goals for
Questions have been raised as to             expedited PMA submissions only twice out of 7 years. FDA review time and time
whether FDA is sufficiently meeting the      to final decision for both types of PMAs were highly variable but generally
performance goals and whether                increased in recent years. For example, the average time to final decision for
devices are reaching the market in a
                                             original PMAs increased from 462 days for FY 2003 to 627 days for FY 2008 (the
timely manner. In preparation for
                                             most recent year for which complete data are available).
reauthorization, GAO was asked to
(1) examine trends in FDA’s 510(k)           The three industry groups and four consumer advocacy groups GAO interviewed
review performance from fiscal years         noted a number of issues related to FDA’s review of medical device submissions.
(FY) 2003-2010, (2) examine trends in        The four issues most commonly raised by stakeholders included (1) insufficient
FDA’s PMA review performance from            communication between FDA and stakeholders throughout the review process,
FYs 2003-2010, and (3) describe              (2) a lack of predictability and consistency in reviews, (3) an increase in time to
stakeholder issues with FDA’s review         final decision, and (4) inadequate assurance of the safety and effectiveness of
processes and steps FDA is taking that       approved or cleared devices. FDA is taking steps—including issuing new
may address these issues. To do this
                                             guidance documents, enhancing reviewer training, and developing an electronic
work, GAO examined FDA medical
                                             system for reporting adverse events—that may address many of these issues. It
device review data, reviewed FDA user
fee data, interviewed FDA staff              is important for the agency to monitor the impact of those steps in ensuring that
regarding the medical device review          safe and effective medical devices are reaching the market in a timely manner.
process and FDA data, and                    In commenting on a draft of this report, HHS generally agreed with GAO’s
interviewed three industry groups and        findings and noted that FDA has identified some of the same performance trends
four consumer advocacy groups.               in its annual reports to Congress. HHS also called attention to the activities FDA
View GAO-12-418. For more information,       has undertaken to improve the medical device review process.
contact Marcia Crosse at (202) 512-7114 or
crossem@gao.gov.

                                                                                     United States Government Accountability Office
Contents


Letter                                                                                     1
               Background                                                                  4
               FDA Met All Performance Goals for 510(k)s but the Time to Final
                 Decision Has Increased Substantially in Recent Years                    11
               FDA Was Inconsistent in Meeting Performance Goals for PMAs
                 While FDA Review Time and Time to Final Decision Generally
                 Increased                                                               20
               Stakeholders Noted Issues with the Medical Device Review
                 Process; FDA Is Taking Steps That May Address Many of These
                 Issues                                                                  34
               Concluding Observations                                                   40
               Agency Comments                                                           41

Appendix I     FDA Medical Device Review Performance for Fiscal Years
               (FY) 2000–2011                                                            42



Appendix II    Number of Full-time Equivalent (FTE) FDA Staff Supporting
               Medical Device User Fee Activities, FYs 2003 through 2010                 48



Appendix III   Comments from the Department of Health and Human Services                 50



Appendix IV    GAO Contact and Staff Acknowledgments                                     53



Tables
               Table 1: FDA’s 510(k) Performance Goals, FYs 2003-2011                      7
               Table 2: GAO Definitions of FDA Review Time and Time to Final
                        Decision                                                          8
               Table 3: FDA’s PMA Performance Goals, FYs 2003-2011                       10
               Table 4: FDA Premarket Notification (510(k)) Review
                        Performance, FYs 2000–2011                                       42
               Table 5: FDA Premarket Approval (PMA) Review Performance for
                        Original PMAs, FYs 2000–2011                                     44
               Table 6: FDA Premarket Approval (PMA) Review Performance for
                        Expedited PMAs, FYs 2000–2011                                    46



               Page i                             GAO-12-418 Medical Device Performance Goals
Figures
          Figure 1: Percentage of 510(k)s FDA Reviewed within 90 Days for
                   the Fiscal Year 2000-2010 Cohorts                                 13
          Figure 2: Percentage of 510(k)s FDA Reviewed within 150 Days for
                   the Fiscal Year 2000-2010 Cohorts                                 15
          Figure 3: Average FDA Review Time and Average Time to Final
                   Decision for 510(k)s in the Fiscal Year 2000-2010 Cohorts         16
          Figure 4: Average Number of Review Cycles Per 510(k) for the
                   Fiscal Year 2000-2010 Cohorts                                     17
          Figure 5: Percentage of 510(k) Submissions Receiving FDA First-
                   Cycle Substantially Equivalent Decisions and First-Cycle
                   Additional Information Requests for the Fiscal Year 2000-
                   2010 Cohorts                                                      19
          Figure 6: Percentage of FDA Final Decisions That Devices Were
                   Substantially Equivalent or Not Substantially Equivalent
                   for 510(k) Submissions for the Fiscal Year 2000-2010
                   Cohorts                                                           20
          Figure 7: Percentage of Original PMAs FDA Reviewed within 180
                   Days for the Fiscal Year 2000-2010 Cohorts                        22
          Figure 8: Percentage of Original PMAs FDA Reviewed within 320
                   Days and 295 Days for the Fiscal Year 2000-2010 Cohorts           24
          Figure 9: Percentage of Expedited PMAs FDA Reviewed within 180
                   Days for the Fiscal Year 2000-2010 Cohorts                        26
          Figure 10: Percentage of Expedited PMAs FDA Reviewed within
                   300 Days and 280 Days for the Fiscal Year 2000-2010
                   Cohorts                                                           28
          Figure 11: Average FDA Review Time and Average Time to Final
                   Decision for Original PMAs in the Fiscal Year 2000-2010
                   Cohorts                                                           30
          Figure 12: Average FDA Review Time and Average Time to Final
                   Decision for Expedited PMAs in the Fiscal Year 2000-2010
                   Cohorts                                                           32




          Page ii                             GAO-12-418 Medical Device Performance Goals
Abbreviations

AI                additional information
CBER              Center for Biologics Evaluation and Research
CDRH              Center for Devices and Radiological Health
FDA               Food and Drug Administration
FDAAA             Food and Drug Administration Amendments Act of 2007
FTE               full-time equivalent
FY                fiscal year
HHS               Department of Health and Human Services
IOM               Institute of Medicine
MDUFA             Medical Device User Fee Amendments of 2007
MDUFMA            Medical Device User Fee and Modernization Act of 2002
PMA               premarket approval
SOP               standard operating procedure



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




Page iii                                    GAO-12-418 Medical Device Performance Goals
United States Government Accountability Office
Washington, DC 20548




                                   February 29, 2012

                                   The Honorable Richard Burr
                                   Ranking Member
                                   Subcommittee on Children and Families
                                   Committee on Health, Education, Labor, and Pensions
                                   United States Senate

                                   The Honorable Tom Coburn
                                   Ranking Member
                                   Permanent Subcommittee on Investigations
                                   Committee on Homeland Security and Governmental Affairs
                                   United States Senate

                                   The Food and Drug Administration (FDA) within the Department of Health
                                   and Human Services (HHS) is responsible for overseeing the safety and
                                   effectiveness of medical devices sold in the United States. 1 Congress
                                   passed the Medical Device User Fee and Modernization Act of 2002
                                   (MDUFMA) to provide additional resources for FDA to support the
                                   process of reviewing medical device applications. 2 MDUFMA authorized
                                   FDA to collect user fees from the medical device industry to supplement
                                   its annual appropriation for salaries and expenses for fiscal years
                                   (FY) 2003 through 2007. 3 The medical device user fee program was
                                   reauthorized in 2007 as part of the Food and Drug Administration
                                   Amendments Act (FDAAA); the reauthorization was called the Medical
                                   Device User Fee Amendments of 2007 (MDUFA) and authorizes FDA to
                                   collect user fees for FYs 2008 through 2012. 4 FDA’s authority to collect
                                   user fees for medical devices expires on October 1, 2012, and the


                                   1
                                    Medical devices include instruments, apparatuses, machines, and implants that are
                                   intended for use to diagnose, cure, treat, or prevent disease, or to affect the structure or
                                   any function of the body. See 21 U.S.C. § 321(h). These devices range from simple tools
                                   such as bandages and surgical clamps to complicated devices such as pacemakers.
                                   2
                                    See Pub. L. No. 107-250, § 102(a), 116 Stat. 1588, 1589-1600 (2002) (codified as
                                   amended at 21 U.S.C. §§ 379i and 379j).
                                   3
                                    A user fee is a fee assessed for goods and services provided by the federal government.
                                   FDA collected one type of user fee—application fees—under MDUFMA.
                                   4
                                    FDA collects three types of user fees under MDUFA: application fees, annual
                                   establishment registration fees, and annual fees for periodic reports regarding Class III
                                   devices.




                                   Page 1                                        GAO-12-418 Medical Device Performance Goals
medical device user fee program will need to be reauthorized for FDA to
continue to collect user fees. Medical device user fee amounts have
become a larger proportion of FDA’s funding for medical device review
processes, rising from 10.6 percent of costs in FY 2003—the first year
FDA collected medical device user fees—to 19.5 percent of costs in
FY 2010, the most recent year for which data are available. In FY 2010,
MDUFA user fees collected by FDA—including application,
establishment, and product fees—totaled nearly $67 million, including
over $29 million in application fees. 5 Application fees are collected for a
variety of medical device submission types, including premarket
notifications (510(k)s) and premarket approvals (PMAs). 6

Under each authorization of the medical device user fee program, FDA
committed to performance goals related to the review of medical device
submissions. 7 The performance goals include specific time frames within
which FDA is to take action on submissions. 8 These performance goals,
as well as user fee amounts, are negotiated between FDA and industry
stakeholders and submitted to congressional committees prior to each
reauthorization. Questions have been raised about whether FDA is
sufficiently meeting the user fee performance goals and whether medical
devices are reaching the market in a timely manner. A number of
congressional committees have recently held hearings during which the


5
 For the remainder of this report, we use the term “user fees” to refer to user fees
submitted with device applications such as premarket approvals (PMA), premarket
notifications (510(k)), and various types of PMA supplements.
6
 The PMA review process is more stringent than the 510(k) review process and is
generally used for higher risk devices. As part of a PMA submission, the manufacturer is
required to supply evidence providing reasonable assurance that the device is safe and
effective. Under the 510(k) process, FDA determines whether the device is substantially
equivalent to a legally marketed device. FDA also reviews several other types of medical
device submissions that are outside the scope of our work. PMAs and 510(k)s make up
the majority of device submissions received by FDA. For example, our analysis of PMAs
and 510(k)s included 88.6 percent of all device submissions to FDA in FY 2010.
7
 See Pub. L. No. 110-85, § 201(c), 121 Stat. 823, 842-43 (2007). The performance goals
are identified in letters sent by the Secretary of Health and Human Services to the
Chairman of the Senate Committee on Health, Education, Labor, and Pensions and the
Chairman of the House Committee on Energy and Commerce and are published on FDA’s
website. Each fiscal year, FDA is required to submit a report on its progress in achieving
those goals and future plans for meeting them. See 21 U.S.C. § 379j-1(a).
8
 FDA does not have to approve a PMA or clear a 510(k) for it to be considered acted
upon. There are a number of decisions in addition to an approval or clearance decision
that can end FDA’s review.




Page 2                                       GAO-12-418 Medical Device Performance Goals
medical device industry questioned FDA’s timeliness, while other
stakeholders questioned FDA’s ability to ensure safety and effectiveness.

In preparation for the reauthorization of the medical device user fee
program, you requested that we examine FDA’s medical device review
process. In this report, we (1) examine trends in FDA’s 510(k) medical
device review performance for FYs 2003 through 2010, (2) examine
trends in FDA’s PMA medical device review performance for FYs 2003
through 2010, and (3) describe the issues stakeholders have raised about
the medical device review processes and steps FDA is taking that may
address these issues. We provide additional details on FDA’s medical
device review performance in appendix I. You also asked us to provide
information on the number of full-time equivalent (FTE) staff involved in
the medical device review process; this information is provided in
appendix II.

To determine the trends in FDA’s medical device review performance for
510(k)s and PMAs for FYs 2003 through 2010, we examined data
obtained from FDA on the review process for all 510(k)s and PMAs
submitted to FDA in those years. 9 To provide context for FDA’s
performance prior to enactment of the user fee acts, we also analyzed
review data for all 510(k)s and PMAs submitted for FYs 2000 through
2002. Additionally, we reviewed data on FY 2011 submissions in order to
provide preliminary performance results for that year. 10 Our analyses
focused on the proportion of medical device submissions in each fiscal
year for which FDA met or did not meet the applicable performance
goal(s); the total time from the date of submission to the date a final
decision was made—including both the time FDA spent reviewing a
submission and any time the sponsor took to respond to questions or
requests for additional information from FDA; the FDA review time
(i.e., the time counted toward user fee performance goals, which does not
include any time the sponsor took to respond to any questions from FDA);



9
 We defined a cohort to be complete if fewer than 10 percent of submissions from that
cohort were still under review at the time we received FDA’s data.
10
  The data for FY 2011 are preliminary because the agency had only completed its review
for 61 percent of 510(k) submissions and 51 percent of PMA submissions at the time we
received FDA’s data. As a result, it was too soon to tell what the final results for this cohort
would be. For example, it is possible that some of the reviews taking the most time were
among those not completed when we received FDA’s data.




Page 3                                         GAO-12-418 Medical Device Performance Goals
             and the average number of review cycles prior to approval. 11 We also
             reviewed publicly available FDA user fee data for FY 2003 through 2010
             and interviewed FDA staff regarding the medical device review process
             and the data we received from FDA.

             To describe the issues stakeholders have raised about the device review
             processes and what steps FDA is taking that may address these issues,
             we reviewed congressional testimony and interviewed three industry
             groups and four consumer advocacy groups. 12 All of these groups have
             participated in at least half of the meetings held by FDA to discuss the
             reauthorization of the user fee program. Furthermore, the industry groups
             we interviewed represent a mixture of large and small medical device
             manufacturers and cover a significant portion of the device market. We
             performed content analyses of the interviews to determine the most
             pressing issues based on how often each issue was raised. 13

             We conducted this performance audit from October 2011 through
             February 2012 in accordance with generally accepted government
             auditing standards. Those standards require that we plan and perform the
             audit to obtain sufficient, appropriate evidence to provide a reasonable
             basis for our findings and conclusions based on our audit objectives. We
             believe that the evidence obtained provides a reasonable basis for our
             findings and conclusions based on our audit objectives.


             Medical devices are reviewed primarily by FDA’s Center for Devices and
Background   Radiological Health (CDRH), with a smaller proportion reviewed by the
             Center for Biologics Evaluation and Research (CBER). FDA classifies
             each device type into one of three classes—class I, II, or III—based on
             the level of risk it poses and the controls necessary to reasonably ensure



             11
               The first review cycle begins when a sponsor makes a submission to FDA and ends
             when FDA either makes a decision or contacts the sponsor in writing to request additional
             information. A new review cycle begins when the sponsor sends a response back to FDA.
             12
               When we refer to consumer advocacy groups, we are referring to groups that advocate
             on behalf of consumers and patients.
             13
               It was beyond the scope of our review to describe all issues raised by stakeholders.
             Such issues—including barriers to innovation in the medical device industry or the need
             for increased resources at FDA—have been extensively covered in other reports, forums,
             and hearings.




             Page 4                                      GAO-12-418 Medical Device Performance Goals
its safety and effectiveness. 14 Class I includes devices with the lowest risk
(e.g., tongue depressors, reading glasses, forceps), while class III
includes devices with the highest risk (e.g., breast implants, coronary
stents). Almost all class I devices and some class II devices
(e.g., mercury thermometers, certain adjustable hospital beds) are
exempt from premarket notification requirements. Most class III device
types are required to obtain FDA approval through the PMA process, the
most stringent of FDA’s medical device review processes. 15 The
remaining device types are required to obtain FDA clearance or approval
through either the 510(k) or PMA processes. 16

If eligible, a 510(k) is filed when a manufacturer seeks a determination
that a new device is substantially equivalent to a legally marketed device
known as a predicate device. 17 In order to be deemed substantially
equivalent (i.e., cleared by FDA for marketing), a new device must have
the same technological characteristics and intended use as the predicate
device, or have the same intended use and different technological
characteristics but still be demonstrated to be as safe and effective as the
predicate device without raising new questions of safety and
effectiveness. Most device submissions filed each year are 510(k)s. For




14
  See 21 U.S.C. § 360c(a)(1); 21 C.F.R. pt. 860.
15
  Some class III device types on the market before the enactment of the Medical Device
Amendments of 1976 and those determined to be substantially equivalent to them do not
currently require PMA approval for marketing. FDA has been taking steps to address
these device types; once this process is completed, all class III devices will be required to
go through the PMA review process. For additional information, see GAO, Medical
Devices: FDA’s Premarket Review and Postmarket Safety Efforts, GAO-11-556T
(Washington, D.C.: Apr. 13, 2011).
16
  A small percentage of devices enter the market by other means, such as through the
humanitarian device exemption process that authorizes FDA to exempt certain medical
devices from the premarket review requirement to demonstrate effectiveness in order to
provide an incentive for the development of devices that treat or diagnose rare diseases or
conditions. See 21 U.S.C. § 360j(m); 21 C.F.R. pt. 814, subpt. H.
17
  A legally marketed device to which a new device may be compared for a determination
regarding substantial equivalence is a device that was legally marketed prior to 1976, a
device that has been reclassified from class III to class II or I, or a device that has been
found to be substantially equivalent through the 510(k) process. See 21 C.F.R.
§ 807.92(a)(3).




Page 5                                        GAO-12-418 Medical Device Performance Goals
example, of the more than 13,600 device submissions received by FDA in
FYs 2008 through 2010, 88 percent were 510(k)s. 18

The medical device performance goals were phased in during the period
covered by MDUFMA (the FYs 2003 through 2007 cohorts) and were
updated for MDUFA. 19 Under MDUFA, FDA’s goal is to complete the
review process for 90 percent of the 510(k)s in a cohort within 90 days of
submission (known as the Tier 1 goal) and to complete the review
process for 98 percent of the cohort within 150 days (the Tier 2 goal). 20
(See table 1 for the 510(k) performance goals for the FYs 2003 through
2011 cohorts). FDA may take any of the following actions on a 510(k)
after completing its review:

•    issue an order declaring the device substantially equivalent;

•    issue an order declaring the device not substantially equivalent; or

•    advise the submitter that the 510(k) is not required (i.e., the product is
     not regulated as a device or the device is exempt from premarket
     notification requirements).




18
  See Department of Health and Human Services, Food and Drug Administration,
Quarterly Update on Medical Device Performance Goals (Silver Spring, Md.: July 26,
2011).
19
  A cohort is comprised of all the submissions of a certain type filed in the same fiscal
year. For example, all 510(k)s received by FDA from October 1, 2010, to September 30,
2011, make up the 510(k) review cohort for FY 2011.
20
  Tier 1 and Tier 2 designations refer to the length of time allotted (90 days and 150 days,
respectively) for FDA to complete its review of 510(k) submissions. If FDA completed its
review of a submission in 90 days or less, it met the time frame for both the Tier 1 and
Tier 2 goals. If the review was completed in more than 90 days but not more than
150 days, only the time frame for the Tier 2 goal was met. If the review took longer than
150 days, the time frame for neither goal was met. FDA did not designate 510(k)
performance goals as either Tier 1 or Tier 2 prior to FY 2008. We have aligned the
performance goals in place prior to FY 2008 with the Tier 1 goals for FYs 2008-2011
based on sharing the same 90-day time frame. This placement illustrates the increase in
the goal percentage over time. We defined a 510(k) cohort to be complete if fewer than 10
percent of submissions from that cohort were still under review at the time we received
FDA’s data, which cover reviews by CDRH through October 26, 2011, and reviews by
CBER through December 23, 2011. Using this definition, FY 2011 was the only 510(k)
cohort that was incomplete.




Page 6                                       GAO-12-418 Medical Device Performance Goals
                                          Each of these actions ends the review process for a submission. 21 A
                                          sponsor’s withdrawal of a submission also ends the review process.

Table 1: FDA’s 510(k) Performance Goals, FYs 2003-2011
                                                                                                                                       a
                                              Period covered by MDUFMA                                 Period covered by MDUFA
Fiscal year cohort                 2003            2004          2005         2006    2007           2008        2009       2010           2011
                     b                                                 c         c        c
Tier 1 goal percentage               —                —            75          75       80              90         90          90            90
                     d
Tier 2 goal percentage               —                —              —          —        —              98         98          98            98
                                          Source: GAO analysis of FDA data.

                                          Notes: A review cohort includes all the medical device submissions relating to a particular
                                          performance goal that were submitted in a given fiscal year. For example, all 510(k)s received by
                                          FDA from October 1, 2010, to September 30, 2011, make up the 510(k) review cohort for FY 2011.
                                          There were no 510(k) performance goals prior to MDUFMA. Fiscal years for which there was no
                                          corresponding 510(k) performance goal are denoted with a dash (—).
                                          a
                                           MDUFA performance goals cover the FYs 2008 through 2012 cohorts; we are showing only those
                                          cohorts we examined as part of our analysis.
                                          b
                                              Percentage of 510(k) submissions to be completed by FDA within 90 days of submission.
                                          c
                                           These were not designated as Tier 1 goals prior to FY 2008 because there were no Tier 2 goals for
                                          those cohorts. We have aligned them with the Tier 1 goals for FYs 2008 through 2011 because they
                                          are based on the same 90-day time frame and this placement illustrates the gradual increase in the
                                          goal percentage over time.
                                          d
                                              Percentage of 510(k) submissions to be completed by FDA within 150 days of submission.


                                          Alternatively, FDA may “stop the clock” on a 510(k) review by sending a
                                          letter asking the sponsor to submit additional information (known as an AI
                                          letter). This completes a review cycle but does not end the review
                                          process. The clock will resume (and a new review cycle will begin) when
                                          FDA receives a response from the sponsor. As a result, FDA may meet
                                          its 510(k) performance goals even if the time to final decision (FDA review
                                          time plus time spent waiting for the sponsor to respond to FDA’s requests
                                          for additional information) is longer than the time frame allotted for the
                                          performance goal. For example, a sponsor might have submitted a 510(k)
                                          on March 1, 2009, to start the review process. If FDA sent an AI letter on
                                          April 1, 2009 (after 31 days on the clock), the sponsor provided a
                                          response on June 1, 2009 (after an additional 61 days off the clock), and
                                          FDA issued a final decision on June 11, 2009 (10 more days on the


                                          21
                                            When calculating whether FDA met the performance goal for a 510(k) cohort, FDA and
                                          industry have agreed to include only those submissions receiving a substantially
                                          equivalent or not substantially equivalent decision. For our analysis, we included all
                                          510(k)s that had received a final decision, regardless of the decision received, in order to
                                          provide a broader look at FDA’s review performance.




                                          Page 7                                              GAO-12-418 Medical Device Performance Goals
                                        clock), then the FDA review time counted toward the MDUFA
                                        performance goals would be 41 days (FDA’s on-the-clock time). FDA
                                        would have met both the Tier 1 (90 day) and Tier 2 (150 day) time frames
                                        for that device even though the total number of calendar days (on- and
                                        off-the-clock) from beginning the review to a final decision was 102 days.
                                        (See table 2 for a comparison of FDA review time and time to final
                                        decision.) FDA tracks the time to final decision and reports on it in the
                                        agency’s annual reports to Congress on the medical device user fee
                                        program. 22

Table 2: GAO Definitions of FDA Review Time and Time to Final Decision

Term                                   Definition
FDA review time                        The time used to determine whether FDA met the medical device user fee performance
                                       goals. Determined by counting the on-the-clock time that FDA spends reviewing a
                                       submission during all review cycles and does not include any time that FDA spends
                                       between review cycles waiting for the sponsor to submit additional information.
Time to final decision                 The total elapsed time from the date of submission through the date of FDA’s final
                                       decision. Determined by adding on-the-clock time (FDA review time) for all review cycles
                                       and any off-the-clock time that FDA spends between review cycles waiting for the sponsor
                                       to submit additional information.
                                        Source: GAO.



                                        A PMA is filed when a device is not substantially equivalent to a predicate
                                        device or has been classified as a class III PMA device (when the risks
                                        associated with the device are considerable). The PMA review process is
                                        the most stringent type of medical device review process required by
                                        FDA, and user fees are much higher for PMAs than for 510(k)s. 23 PMAs
                                        are designated as either original or expedited. 24 FDA considers a device
                                        eligible for expedited review if it is intended to (a) treat or diagnose a life-
                                        threatening or irreversibly debilitating disease or condition and


                                        22
                                          In its July 2011 analysis of 510(k) submissions, FDA concluded that reviewers asked for
                                        additional information from sponsors—thus stopping the clock on FDA’s review time while
                                        the total time to reach a final decision continued to elapse—mainly due to problems with
                                        the quality of the submission. See U.S. Department of Health and Human Services, Food
                                        and Drug Administration, Analysis of Premarket Review Times Under the 510(k) Program
                                        (Silver Spring, Md.: July 2011).
                                        23
                                          For example, for FY 2012 the standard PMA application fee is $220,050 while the 510(k)
                                        fee is $4,049.
                                        24
                                          In its annual performance reports, FDA refers to these two types as original PMAs and
                                        expedited original PMAs.




                                        Page 8                                     GAO-12-418 Medical Device Performance Goals
(b) address an unmet medical need. 25 FDA assesses all medical device
submissions to determine which are appropriate for expedited review,
regardless of whether a company has identified its device as a potential
candidate for this program.

To meet the MDUFA goals, FDA must complete its review of 60 percent
of the original PMAs in a cohort within 180 days of submission (Tier 1)
and 90 percent within 295 days (Tier 2). For expedited PMAs, 50 percent
of a cohort must be completed within 180 days (Tier 1) and 90 percent
within 280 days (Tier 2). (See table 3 for the PMA performance goals for
the FYs 2003 through 2011 cohorts.) The various actions FDA may take
during its review of a PMA are the following:

•    approval order;

•    approvable letter;

•    major deficiency letter;

•    not approvable letter; and

•    denial order. 26




25
   See 21 U.S.C. § 360e(d)(5). Unmet medical need is demonstrated by meeting one of
the following criteria: the device represents a breakthrough technology that provides a
clinically meaningful advantage over existing technology; no approved alternative
treatment or means of diagnosis exists; the device offers significant, clinically meaningful
advantages over existing approved alternative treatments; or the availability of the device
is in the best interest of patients.
26
   An approval order informs the applicant that the PMA has been approved. An
approvable letter is sent to inform the applicant that there needs to be resolution of minor
deficiencies or completion of an FDA inspection. A major deficiency letter informs the
applicant that the PMA lacks significant information necessary for the agency to complete
its review and requests the applicant amend the submission to provide the necessary
information. A not approvable letter informs the applicant that the submission cannot be
approved at that time because of significant deficiencies; describes the deficiencies; and,
where practical, describes the measures required to make the submission approvable.
Generally, before FDA issues a not approvable letter, it will first issue a major deficiency
letter to provide the applicant with an opportunity to address its concerns. A denial order
notifies the applicant that the PMA is not approved and informs the applicant of its
deficiencies.




Page 9                                       GAO-12-418 Medical Device Performance Goals
                                          The major deficiency letter is the only one of these actions that does not
                                          end the review process for purposes of determining whether FDA met the
                                          MDUFA performance goal time frame for a given submission. As with the
                                          AI letter in a 510(k) review, FDA can stop the clock during the PMA
                                          review process by sending a major deficiency letter (ending a review
                                          cycle) and resume it later upon receiving a response from the
                                          manufacturer. In contrast, taking one of the other four actions
                                          permanently stops the clock, meaning any further review that occurs is
                                          excluded from the calculation of FDA review time. In addition, the
                                          approval order and denial order are also considered final decisions and
                                          end FDA’s review of a PMA completely. A sponsor’s withdrawal of a
                                          submission also ends the review process.

Table 3: FDA’s PMA Performance Goals, FYs 2003-2011
                                                                                                                                        a
                                                        Period covered by MDUFMA                         Period covered by MDUFA
                                                                               b      b         b
Fiscal year cohort                               2003       2004         2005      2006   2007           2008       2009      2010      2011
                                  c
Original PMA Tier 1 goal percentage                 —           —             —      —        50            60         60         60        60
                                  d
Original PMA Tier 2 goal percentage                 —           —             —      80       90            90         90         90        90
                                      e
Expedited PMA Tier 1 goal percentage                —           —             —      —        —             50         50         50        50
                                      f
Expedited PMA Tier 2 goal percentage                —           —             70     80       90            90         90         90        90
                                          Source: GAO analysis of FDA data.

                                          Notes: A review cohort includes all the medical device submissions relating to a particular
                                          performance goal that were submitted in a given fiscal year. For example, all PMAs received by FDA
                                          from October 1, 2010, to September 30, 2011, make up the PMA review cohort for FY 2011.
                                          There were no performance goals prior to MDUFMA. Fiscal years for which there was no
                                          corresponding PMA performance goal are denoted with a dash (—).
                                          a
                                           MDUFA performance goals cover the FYs 2008 through 2012 cohorts; we are showing only those
                                          cohorts we examined as part of our analysis.
                                          b
                                           PMA performance goals were not designated as Tier 1 or Tier 2 until FY 2008. We have aligned the
                                          performance goals in place prior to FY 2008 with the Tier 1 or Tier 2 goals for FYs 2008 through 2011
                                          based on sharing the same or similar goal time frames. This placement illustrates the increase in the
                                          goal percentage over time.
                                          c
                                              Percentage of original PMAs to be completed by FDA within 180 days of submission.
                                          d
                                           Percentage of original PMAs to be completed by FDA within 320 days (for FYs 2006 through 2007)
                                          or 295 days (FYs 2008 through 2011) of submission.
                                          e
                                              Percentage of expedited PMAs to be completed by FDA within 180 days of submission.
                                          f
                                          Percentage of expedited PMAs to be completed by FDA within 300 days (for FYs 2005 through 2007)
                                          or 280 days (FYs 2008 through 2011) of submission.


                                          FDA’s review of medical device submissions has been discussed in
                                          recent congressional hearings, meetings between FDA and stakeholders
                                          about the medical device user fee program reauthorization, and published
                                          reports. In addition, in August 2010, FDA released reports which
                                          described the results of two internal assessments conducted by FDA of


                                          Page 10                                            GAO-12-418 Medical Device Performance Goals
                      its medical device review programs. 27 In January 2011, FDA released a
                      plan of action that included 25 steps FDA intends to take to address the
                      issues identified in these assessments. 28


                      For FYs 2003 through 2010, FDA met all Tier 1 and Tier 2 performance
FDA Met All           goals for 510(k)s. In addition, FDA review time for 510(k)s decreased
Performance Goals     slightly during this period, but time to final decision increased
                      substantially. The average number of review cycles and FDA’s requests
for 510(k)s but the   for additional information for 510(k) submissions also increased during
Time to Final         this period.
Decision Has
Increased
Substantially in
Recent Years



                      27
                        In September 2009, FDA convened an internal 510(k) working group to conduct a
                      comprehensive assessment of the 510(k) process. This assessment resulted in the
                      publication of a preliminary report in August 2010, which was intended to communicate
                      preliminary findings and recommendations and actions FDA might take to address
                      identified areas of concern. See U.S. Department of Health and Human Services, Food
                      and Drug Administration, 510(k) Working Group: Preliminary Report and
                      Recommendations (Silver Spring, Md.: August 2010). Also in September 2009, FDA
                      convened an internal task force on the utilization of science in regulatory decision making.
                      This task force was responsible for reviewing how FDA uses science in its regulatory
                      decision making for device reviews and making recommendations on how FDA can
                      quickly incorporate new science—including evolving information, novel technologies, and
                      new scientific methods—into its decision making, while maintaining as much predictability
                      as practical. The task force released a preliminary report with its findings and
                      recommendations in August 2010. See U.S. Department of Health and Human Services,
                      Food and Drug Administration, Task Force on the Utilization of Science in Regulatory
                      Decision Making, Preliminary Report and Recommendations (Silver Spring, Md.: August
                      2010).
                      28
                        In January 2011, after reviewing public comments on the August 2010 reports, FDA
                      issued a plan of action for implementing the recommendations in the reports. See U.S.
                      Department of Health and Human Services, Food and Drug Administration, Plan of Action
                      for Implementation of 510(k) and Science Recommendations (Silver Spring, Md.: January
                      2011). FDA began implementing these actions in March 2011 and the majority of the
                      actions had been implemented or were underway at the time of our report. See U.S.
                      Department of Health and Human Services, Food and Drug Administration, CDRH Plan of
                      Action for 510(k) and Science (Silver Spring, Md.: October 2011).




                      Page 11                                      GAO-12-418 Medical Device Performance Goals
FDA Met All Tier 1 and     FDA met all Tier 1 performance goals for the completed 510(k) cohorts
Tier 2 Performance Goals   that had Tier 1 goals in place. 29 The percentage of 510(k)s reviewed
for 510(k)s from 2003      within 90 days (the current Tier 1 goal time frame) exceeded 90 percent
                           for the FYs 2005 through 2010 cohorts (see fig. 1.) Although the 510(k)
through 2010
                           cohort for FY 2011 was still incomplete at the time we received FDA’s
                           data, FDA was exceeding the Tier 1 goal for those submissions on which
                           it had taken action. 30 FDA’s performance varied for 510(k) cohorts prior to
                           the years that the Tier 1 goals were in place but was always below the
                           current 90 percent goal.




                           29
                             When calculating whether FDA met the performance goal for a 510(k) cohort, FDA and
                           industry have agreed to include only those submissions receiving a substantially
                           equivalent or not substantially equivalent decision. For our analysis, we included all
                           510(k)s that had received a final decision, regardless of the decision received, in order to
                           provide a broader look at FDA’s review performance.
                           30
                             Approximately 39 percent of 510(k)s received in FY 2011 were still under review at the
                           time we received FDA’s data, which cover reviews by CDRH through October 26, 2011,
                           and reviews by CBER through December 23, 2011. As a result, it was too soon to tell
                           what the final results for this cohort would be. For example, the percentage of completed
                           510(k)s that met the 90-day performance goal time frame was 97.2 percent. However, the
                           percentage of 510(k)s reviewed within 90 days for the FY 2011 cohort may increase or
                           decrease as those reviews are completed.




                           Page 12                                       GAO-12-418 Medical Device Performance Goals
Figure 1: Percentage of 510(k)s FDA Reviewed within 90 Days for the Fiscal Year
2000-2010 Cohorts




Note: Only 510(k)s that had received a final decision from FDA were included in this analysis. Tier 1
and Tier 2 designations refer to the length of time allotted (90 days and 150 days, respectively) for
FDA to complete its review of 510(k) submissions. If FDA completed its review of a 510(k) submission
in 90 days or less, it met the time frames for both the Tier 1 and Tier 2 goals. If the review was
completed in more than 90 days but not more than 150 days, only the time frame for the Tier 2 goal
was met. If the review took longer than 150 days, FDA did not meet the time frame for either goal.
FDA did not designate 510(k) performance goals as either Tier 1 or Tier 2 prior to FY 2008. We have
aligned the performance goals in place prior to FY 2008 with the Tier 1 goals for FYs 2008 through
2011 based on sharing the same 90-day time frame. This placement illustrates the increase in the
goal percentage over time.




Page 13                                          GAO-12-418 Medical Device Performance Goals
FDA met the Tier 2 goals for all three of the completed cohorts that had
Tier 2 goals in place. Specifically, FDA met the goal of reviewing
98 percent of submissions within 150 days for the FYs 2008, 2009, and
2010 cohorts (see fig. 2.) Additionally, although the 510(k) cohort for
FY 2011 was still incomplete at the time we received FDA’s data, FDA
was exceeding the Tier 2 goal for those submissions on which it had
taken action. 31 FDA’s performance for 510(k) cohorts prior to the years
that the Tier 2 goals were in place was generally below the current
98 percent goal.




31
  Approximately 39 percent of 510(k)s received in FY 2011 were still under review at the
time we received FDA’s data, which cover reviews by CDRH through October 26, 2011,
and reviews by CBER through December 23, 2011. As a result, it was too soon to tell
what the final results for this cohort would be. For example, the percentage of completed
510(k)s that met the 150-day performance goal was 99.8 percent. However, the
percentage of 510(k)s reviewed within 150 days for the FY 2011 cohort may increase or
decrease as those reviews are completed.




Page 14                                     GAO-12-418 Medical Device Performance Goals
                             Figure 2: Percentage of 510(k)s FDA Reviewed within 150 Days for the Fiscal Year
                             2000-2010 Cohorts




                             Note: Only 510(k)s that had received a final decision from FDA were included in this analysis. For
                             purposes of determining whether a goal was met, the percentage is rounded to the nearest whole
                             number. Tier 1 and Tier 2 designations refer to the length of time allotted (90 days and 150 days,
                             respectively) for FDA to complete its review of 510(k) submissions. If FDA completed its review of a
                             submission in 90 days or less, it met the time frames for both the Tier 1 and Tier 2 goals. If the review
                             was completed in more than 90 days but not more than 150 days, only the time frame for the Tier 2
                             goal was met. If the review took longer than 150 days, FDA did not meet the time frame for either
                             goal. FDA did not designate 510(k) performance goals as either Tier 1 or Tier 2 prior to FY 2008.


FDA Review Time for          While the average FDA review time for 510(k) submissions decreased
510(k)s Decreased Slightly   slightly from the FY 2003 cohort to the FY 2010 cohort, the time to final
from 2003 to 2010 but Time   decision increased substantially. Specifically, the average number of days
                             FDA spent on the clock reviewing a 510(k) varied somewhat but overall
to Final Decision            showed a small decrease from 75 days for the FY 2003 cohort to 71 days
Increased Substantially      for the FY 2010 cohort (see fig. 3). However, when we added off-the-
                             clock time (where FDA waited for the sponsor to provide additional
                             information) to FDA’s on-the-clock review time, the resulting time to final
                             decision decreased slightly from the FY 2003 cohort to the FY 2005
                             cohort before increasing 61 percent—from 100 days to 161 days—from



                             Page 15                                            GAO-12-418 Medical Device Performance Goals
the FY 2005 cohort through the FY 2010 cohort. FDA officials told us that
the only alternative to requesting additional information is for FDA to
reject the submission. The officials stated that as a result of affording
sponsors this opportunity to respond, the time to final decision is longer
but the application has the opportunity to be approved.

Figure 3: Average FDA Review Time and Average Time to Final Decision for 510(k)s
in the Fiscal Year 2000-2010 Cohorts




Note: Only 510(k)s that had received a final decision from FDA were included in this analysis.
Average FDA review time refers to the time that FDA spends reviewing a submission and therefore
excludes any time the sponsor may spend responding to FDA requests for additional information.
Average time to final decision includes both the time FDA spends reviewing a submission and the
time the sponsor may spend responding to any requests for additional information.




Page 16                                         GAO-12-418 Medical Device Performance Goals
                           Additionally, although the 510(k) cohort for FY 2011 was still incomplete
                           at the time we received FDA’s data, the average FDA review time and
                           time to final decision were lower in FY 2011 for those submissions on
                           which it had taken action. 32


Number of Review Cycles    The average number of review cycles per 510(k) increased substantially
and Requests for           (39 percent) from FYs 2003 through 2010, rising from 1.47 cycles for the
Additional Information     FY 2003 cohort to 2.04 cycles for the FY 2010 cohort (see fig. 4).
Increased for 510(k)
Submissions from 2003 to   Figure 4: Average Number of Review Cycles Per 510(k) for the Fiscal Year 2000-
                           2010 Cohorts
2010




                           Note: Cycles that were currently in progress at the time we received FDA’s data were included in this
                           analysis.




                           32
                             Approximately 39 percent of 510(k)s received in FY 2011 were still under review at the
                           time we received FDA’s data, which cover reviews by CDRH through October 26, 2011,
                           and reviews by CBER through December 23, 2011. As a result, it was too soon to tell
                           what the final results for this cohort would be. It is possible that some of the reviews taking
                           the most time were among those not completed when we received FDA’s data.




                           Page 17                                           GAO-12-418 Medical Device Performance Goals
In addition, the percentage of 510(k)s receiving a first-cycle decision of
substantially equivalent (i.e., cleared by FDA for marketing) decreased
from 54 percent for the FY 2003 cohort to 20 percent for the FY 2010
cohort, while the percentage receiving first-cycle AI requests exhibited a
corresponding increase. (See fig. 5.) The average number of 510(k)
submissions per year remained generally steady during this period.
Although the 510(k) cohort for FY 2011 was still incomplete at the time we
received FDA’s data, of the first-cycle reviews that had been completed,
the percentage of submissions receiving a first-cycle decision of
substantially equivalent was slightly higher than for the FY 2010 cohort
(21.2 percent in FY 2011 compared with 20.0 percent in FY 2010). 33 In
addition, the percentage receiving a first-cycle AI request was lower
(68.2 percent for FY 2011 compared with 77.0 for FY 2010). 34




33
  Approximately 39 percent of 510(k)s received in FY 2011 were still under review at the
time we received FDA’s data, which cover reviews by CDRH through October 26, 2011,
and reviews by CBER through December 23, 2011. As a result, it was too soon to tell
what the final results for this cohort would be. Therefore, the percentage of 510(k)s in the
FY 2011 cohort receiving a first-cycle substantially equivalent decision may increase or
decrease as those reviews are completed.
34
  Approximately 39 percent of 510(k)s received in FY 2011 were still under review at the
time we received FDA’s data, which cover reviews by CDRH through October 26, 2011,
and reviews by CBER through December 23, 2011. As a result, it was too soon to tell
what the final results for this cohort would be. Therefore, the percentage of 510(k)s in the
FY 2011 cohort receiving a first-cycle AI letter may increase or decrease as those reviews
are completed.




Page 18                                      GAO-12-418 Medical Device Performance Goals
Figure 5: Percentage of 510(k) Submissions Receiving FDA First-Cycle
Substantially Equivalent Decisions and First-Cycle Additional Information Requests
for the Fiscal Year 2000-2010 Cohorts




Notes: Only 510(k)s that had received a first-cycle decision from FDA were included in this analysis.
The percentages for each year do not add to 100 percent because there are other possible actions
classified as first-cycle decisions (e.g., a sponsor’s withdrawal of a submission).
The first review cycle starts when FDA receives a submission and ends when FDA either makes a
decision regarding substantial equivalence or requests additional information from the sponsor, or the
sponsor withdraws the submission. More than one cycle may occur before FDA reaches its final
decision.


The percentage of 510(k)s that received a final decision of substantially
equivalent also decreased in recent years—from a high of 87.9 percent
for the FY 2005 cohort down to 75.1 percent for the FY 2010 cohort. The
percentage of 510(k)s receiving a final decision of not substantially
equivalent increased for each cohort from FYs 2003 through 2010, rising
from just over 2.9 percent to 6.4 percent. (See fig. 6.)




Page 19                                           GAO-12-418 Medical Device Performance Goals
                       Figure 6: Percentage of FDA Final Decisions That Devices Were Substantially
                       Equivalent or Not Substantially Equivalent for 510(k) Submissions for the Fiscal
                       Year 2000-2010 Cohorts




                       Notes: Only 510(k)s that had received a final decision from FDA were included in this analysis.
                       The percentages for each year do not add to 100 percent because there are other possible actions
                       classified as final decisions (e.g., a sponsor’s withdrawal of a submission).


                       For FYs 2003 through 2010, FDA met most of the goals for original PMAs
FDA Was Inconsistent   but fell short on most of the goals for expedited PMAs. In addition, FDA
in Meeting             review time and time to final decision for both types of PMAs generally
                       increased during this period. Finally, the average number of review cycles
Performance Goals      increased for certain PMAs while the percentage of PMAs approved after
for PMAs While FDA     one review cycle generally decreased.
Review Time and
Time to Final
Decision Generally
Increased



                       Page 20                                           GAO-12-418 Medical Device Performance Goals
FDA Met Most Goals for    Since FY 2003, FDA met the original PMA performance goals for four of
Original PMAs but Fell    the seven completed cohorts that had goals in place, but met the goals
Short of Most Goals for   for only two of the seven expedited PMA cohorts with goals. 35
                          Specifically, FDA met its Tier 1 performance goals for original PMAs for
Expedited PMAs from
                          all three of the completed original PMA cohorts that had such goals in
2003-2010                 place, with the percentage increasing from 56.8 percent of the FY 2007
                          cohort to
                          80.0 percent of the FY 2009 cohort completed on time. 36 (See fig. 7.)
                          While the FY 2010 and 2011 cohorts are still incomplete, FDA is
                          exceeding the goals for those submissions on which it has taken action. 37
                          FDA’s performance had declined steadily in the years immediately before
                          implementation of these goals—from 67.1 percent of the FY 2000 cohort
                          to 34.5 percent of the FY 2006 cohort completed within 180 days.




                          35
                            We treated PMA submissions as meeting the time frame for a given performance goal if
                          they were reviewed within the goal time plus any extension to the goal time that may have
                          been made. The only reason the goal time can be extended is if the sponsor submits a
                          major amendment to the submission on its own initiative (i.e., not solicited by FDA).
                          According to FDA, typical situations that might prompt a sponsor to submit an unsolicited
                          major amendment include when the applicant obtains additional test data related to the
                          safety or effectiveness of the device or obtains new clinical data from a previously
                          unreported study.
                          36
                             PMA performance goals were not designated as Tier 1 or Tier 2 until FY 2008. We have
                          aligned the performance goals in place prior to FY 2008 with the Tier 1 or Tier 2 goals for
                          FYs 2008-2011 based on sharing the same or similar goal time frames. This placement
                          illustrates the increase in the goal percentage over time.
                          37
                            For this analysis, the FY 2010 and 2011 original PMA cohorts were still incomplete.
                          Specifically, for 18.5 percent of the FY 2010 original PMA cohort and 48.8 percent of the
                          FY 2011 cohort, a decision that would permanently stop the review clock for purposes of
                          determining whether FDA met its performance goals had not been made at the time we
                          received FDA’s data. As a result, it was too soon to tell what the final results for this cohort
                          would be. It is possible that some of the reviews taking the most time were among those
                          not completed when we received FDA’s data. The percentage of original PMAs reviewed
                          within 180 days for these two cohorts may increase or decrease as those reviews are
                          completed.




                          Page 21                                        GAO-12-418 Medical Device Performance Goals
Figure 7: Percentage of Original PMAs FDA Reviewed within 180 Days for the Fiscal
Year 2000-2010 Cohorts




Notes: Tier 1 and Tier 2 designations refer to the length of time allotted (for the FYs 2008 through
2011 cohorts: 180 days and 295 days, respectively) for FDA to complete its review of original PMA
submissions. FDA did not designate PMA performance goals as either Tier 1 or Tier 2 prior to
FY 2008. Prior to FY 2006 there were no goals for original PMA submissions. For FY 2006 there was
only one goal: 320 days. For FY 2007, the goals for original PMA submissions were 180 days and
320 days. We have aligned the performance goals in place prior to FY 2008 with the Tier 1 and Tier 2
goals for FYs 2008-2011 based on sharing similar time frames. This placement illustrates the
increase in the goal percentage over time. If FDA completed its review of a submission in 180 days or
less, it met the time frames for both the Tier 1 and Tier 2 goals. If the review was completed in more
than 180 days but not more than 320 or 295 days (depending on the cohort), only the time frame for
the Tier 2 goal was met. If the review took longer than 320 or 295 days, FDA did not meet the time
frame for either goal.
We treated PMA submissions as meeting the time frame for a given performance goal if they were
reviewed within the goal time plus any extension to the goal time that may have been made. The only
reason the goal time can be extended is if the sponsor submits a major amendment to the submission
on its own initiative (i.e., unsolicited by FDA).




Page 22                                           GAO-12-418 Medical Device Performance Goals
a
 This analysis includes only those original PMAs for which FDA or the sponsor had made a decision
that would permanently stop the review clock for purposes of determining whether FDA met its
performance goals (i.e., an approval, approvable, not approvable, withdrawal, or denial); this includes
reviews by CBER through September 30, 2011, and reviews by CDRH through December 1, 2011.
Submissions without such a decision are not included in the results for each cohort shown above. We
considered a cohort to be incomplete if more than 10 percent of submissions had not yet received
such a decision. For this analysis, the FY 2010 cohort was still incomplete. Specifically, for
18.5 percent of the FY 2010 original PMA cohort, a decision that would permanently stop the review
clock had not been made at the time we received FDA’s data. As a result, it was too soon to tell what
the final results for this cohort would be. It is possible that some of the reviews taking the most time
were among those not completed when we received FDA’s data. The percentage of original PMAs
reviewed within 180 days for this cohort may increase or decrease as those reviews are completed.


FDA’s performance in meeting the Tier 2 performance goals for original
PMAs fell short of the goal for three of the four completed cohorts during
the years that these goals were in place. FDA met the MDUFMA Tier 2
performance goal (320 days) for the FY 2006 original PMA cohort but not
for the FY 2007 cohort, and did not meet the MDUFA Tier 2 performance
goal (295 days) for either of the completed cohorts (FYs 2008 and 2009)
to which the goal applied (see fig. 8). While the FYs 2010 and 2011
original PMA cohorts are still incomplete, FDA is exceeding the MDUFA
Tier 2 goals for those submissions on which it has taken action. 38 FDA’s
performance varied for original PMA cohorts prior to the years that the
Tier 2 goals were in place but was always below the current goal to have
90 percent reviewed within 295 days.




38
  For this analysis, the FYs 2010 and 2011 original PMA cohorts were still incomplete.
Specifically, for 18.5 percent of the FY 2010 original PMA cohort and 48.8 percent of the
FY 2011 cohort, a decision that would permanently stop the review clock for purposes of
determining whether FDA met its performance goals had not been made at the time we
received FDA’s data. As a result, it was too soon to tell what the final results for this cohort
would be. It is possible that some of the reviews taking the most time were among those
not completed when we received FDA’s data. The percentage of original PMAs reviewed
within 320 and 295 days for these two cohorts may increase or decrease as those reviews
are completed.




Page 23                                            GAO-12-418 Medical Device Performance Goals
Figure 8: Percentage of Original PMAs FDA Reviewed within 320 Days and 295 Days for the Fiscal Year 2000-2010 Cohorts




                                        Notes: Tier 1 and Tier 2 designations refer to the length of time allotted (for the FYs 2008-2011
                                        cohorts: 180 days and 295 days, respectively) for FDA to complete its review of original PMA
                                        submissions. FDA did not designate PMA performance goals as either Tier 1 or Tier 2 prior to
                                        FY 2008. Prior to FY 2006 there were no goals for original PMA submissions. For FY 2006 there was
                                        only one goal: 320 days. For FY 2007, the goals for original PMA submissions were 180 days and
                                        320 days. We have aligned the performance goals in place prior to FY 2008 with the Tier 1 and Tier 2
                                        goals for FYs 2008 through 2011 based on sharing similar time frames. This placement illustrates the
                                        increase in the goal percentage over time. If FDA completed its review of a submission in 180 days or
                                        less, it met the time frames for both the Tier 1 and Tier 2 goals. If the review was completed in more
                                        than 180 days but not more than 320 or 295 days (depending on the cohort), only the time frame for
                                        the Tier 2 goal was met. If the review took longer than 320 or 295 days, FDA did not meet the time
                                        frame for either goal.
                                        We treated PMA submissions as meeting the time frame for a given performance goal if they were
                                        reviewed within the goal time plus any extension to the goal time that may have been made. The only
                                        reason the goal time can be extended is if the sponsor submits a major amendment to the submission
                                        on its own initiative (i.e., unsolicited by FDA).




                                        Page 24                                           GAO-12-418 Medical Device Performance Goals
a
 These analyses include only those original PMAs for which FDA or the sponsor had made a decision
that would permanently stop the review clock for purposes of determining whether FDA met its
performance goals (i.e., an approval, approvable, not approvable, withdrawal, or denial); this includes
reviews by CBER through September 30, 2011, and reviews by CDRH through December 1, 2011.
Submissions without such a decision are not included in the results for each cohort shown above. We
considered a cohort to be incomplete if more than 10 percent of submissions had not yet received
such a decision. For this analysis, the FY 2010 original PMA cohort was still incomplete. Specifically,
for 18.5 percent of the FY 2010 original PMA cohort, a decision that would permanently stop the
review clock had not been made at the time we received FDA’s data. As a result, it was too soon to
tell what the final results for this cohort would be. It is possible that some of the reviews taking the
most time were among those not completed when we received FDA’s data. The percentage of
original PMAs reviewed within 320 and 295 days for this cohort may decrease as those reviews are
completed.


For expedited PMAs, FDA met the Tier 1 and Tier 2 performance goals
for only two of the seven completed cohorts for which the goals were in
effect. FDA met the Tier 1 (180-day) goal for only one of the two
completed cohorts during the years the goal has been in place, meeting
the goal for the FY 2009 cohort but missing it for the FY 2008 cohort (see
fig. 9). FDA’s performance varied for cohorts prior to the years that the
Tier 1 expedited PMA goals were in place but was below the current goal
of 50 percent in all but 1 year.




Page 25                                            GAO-12-418 Medical Device Performance Goals
Figure 9: Percentage of Expedited PMAs FDA Reviewed within 180 Days for the
Fiscal Year 2000-2010 Cohorts




Notes: Tier 1 and Tier 2 designations refer to the length of time allotted (for the FYs 2008 through
2011 cohorts: 180 days and 280 days, respectively) for FDA to complete its review of expedited PMA
submissions. FDA did not designate PMA performance goals as either Tier 1 or Tier 2 prior to
FY 2008. For FYs 2005 through 2007, there was only one goal for expedited PMAs: 300 days. Prior
to FY 2005 there were no goals for expedited PMA submissions. We have aligned the performance
goal in place prior to FY 2008 with the Tier 2 goal for FYs 2008 through 2011 based on sharing a
similar time frame. This placement illustrates the increase in the goal percentage over time. If FDA
completed its review of a submission in 180 days or less, it met the time frames for both the Tier 1
and Tier 2 goals. If the review was completed in more than 180 days but not more than 300 or
280 days (depending on the cohort), only the time frame for the Tier 2 goal was met. If the review
took longer than 300 or 280 days, FDA did not meet the time frame for either goal.
We treated PMA submissions as meeting the time frame for a given performance goal if they were
reviewed within the goal time plus any extension to the goal time that may have been made. The only
reason the goal time can be extended is if the sponsor submits a major amendment to the submission
on its own initiative (i.e., unsolicited by FDA).




Page 26                                          GAO-12-418 Medical Device Performance Goals
a
 This analysis includes only those expedited PMAs for which FDA or the sponsor had made a
decision that would permanently stop the review clock for purposes of determining whether FDA met
its performance goals (i.e., an approval, approvable, not approvable, withdrawal, or denial); this
includes reviews by CBER through September 30, 2011, and reviews by CDRH through December 1,
2011. Submissions without such a decision are not included in the results for each cohort shown
above. We considered a cohort to be incomplete if more than 10 percent of submissions had not yet
received such a decision. For this analysis, the FY 2010 expedited PMA cohort was still incomplete.
Specifically, for 16.7 percent of the FY 2010 expedited PMA cohort, a decision that would
permanently stop the review clock had not been made at the time we received FDA’s data. As a
result, it was too soon to tell what the final results for this cohort would be. It is possible that some of
the reviews taking the most time were among those not completed when we received FDA’s data.
The percentage of expedited PMAs reviewed within 180 days for this cohort may increase or
decrease as those reviews are completed.


FDA’s performance in meeting the Tier 2 performance goals for expedited
PMAs fell short of the goal for four of the five completed cohorts during
the years that these goals were in place. FDA met the MDUFMA Tier 2
performance goal (300 days) for the FY 2005 cohort but not for the
FY 2006 or 2007 cohorts, and did not meet the MDUFA Tier 2
performance goal (280 days) for either of the completed cohorts (FY 2008
and 2009) to which the goal applied (see fig. 10). FDA’s performance
varied for expedited PMA cohorts prior to the years that the Tier 2 goals
were in place but always fell below the current goal to have 90 percent
reviewed within 280 days.




Page 27                                              GAO-12-418 Medical Device Performance Goals
Figure 10: Percentage of Expedited PMAs FDA Reviewed within 300 Days and 280 Days for the Fiscal Year 2000-2010 Cohorts




                                       Notes: Tier 1 and Tier 2 designations refer to the length of time allotted (for the FYs 2008 through
                                       2011 cohorts: 180 days and 280 days, respectively) for FDA to complete its review of expedited PMA
                                       submissions. FDA did not designate PMA performance goals as either Tier 1 or Tier 2 prior to
                                       FY 2008. For FYs 2005 through 2007, there was only one goal for expedited PMAs: 300 days. Prior
                                       to FY 2005 there were no goals for expedited PMA submissions. We have aligned the performance
                                       goal in place prior to FY 2008 with the Tier 2 goal for FYs 2008 through 2011 based on sharing a
                                       similar time frame. This placement illustrates the increase in the goal percentage over time. If FDA
                                       completed its review of a submission in 180 days or less, it met the time frames for both the Tier 1
                                       and Tier 2 goals. If the review was completed in more than 180 days but not more than 300 or
                                       280 days (depending on the cohort), only the time frame for the Tier 2 goal was met. If the review
                                       took longer than 300 or 280 days, FDA did not meet the time frame for either goal.
                                       We treated PMA submissions as meeting the time frame for a given performance goal if they were
                                       reviewed within the goal time plus any extension to the goal time that may have been made. The only
                                       reason the goal time can be extended is if the sponsor submits a major amendment to the submission
                                       on its own initiative (i.e., unsolicited by FDA).




                                       Page 28                                          GAO-12-418 Medical Device Performance Goals
                          a
                           These analyses include only those expedited PMAs for which FDA or the sponsor had made a
                          decision that would permanently stop the review clock for purposes of determining whether FDA met
                          its performance goals (i.e., an approval, approvable, not approvable, withdrawal, or denial); this
                          includes reviews by CBER through September 30, 2011, and reviews by CDRH through December 1,
                          2011. Submissions without such a decision are not included in the results for each cohort shown
                          above. We considered a cohort to be incomplete if more than 10 percent of submissions had not yet
                          received such a decision. For this analysis, the FY 2010 expedited PMA cohort was still incomplete.
                          Specifically, for 16.7 percent of the FY 2010 expedited PMA cohort, a decision that would
                          permanently stop the review clock had not been made at the time we received FDA’s data. As a
                          result, it was too soon to tell what the final results for this cohort would be. It is possible that some of
                          the reviews taking the most time were among those not completed when we received FDA’s data.
                          The percentage of expedited PMAs reviewed within 300 and 280 days for this cohort may decrease
                          as those reviews are completed.




FDA Review Time and       FDA review time for both original and expedited PMAs was highly
Time to Final Decision    variable but generally increased across our analysis period, while time to
Generally Increased for   final decision also increased for original PMAs. Specifically, average FDA
                          review time for original PMAs increased from 211 days in the FY 2003
PMAs from 2003 to 2010    cohort (the first year that user fees were in effect) to 264 days in the
                          FY 2008 cohort, then fell in the FY 2009 cohort to 217 days (see fig. 11).
                          When we added off-the-clock time (during which FDA waited for the
                          sponsor to provide additional information or correct deficiencies in the
                          submission), average time to final decision for the FYs 2003 through 2008
                          cohorts fluctuated from year to year but trended upward from 462 days for
                          the FY 2003 cohort to 627 days for the FY 2008 cohort. 39




                          39
                            The FY 2009 original PMA cohort is complete for purposes of calculating FDA review
                          time but incomplete for the calculation of time to final decision because some submissions
                          in the cohort have received a decision ending a review cycle (e.g., approvable letter) and
                          permanently stopping the review clock for purposes of determining whether FDA met its
                          performance goals, but have not yet received a final decision such as approval or denial
                          that would end the review process.




                          Page 29                                              GAO-12-418 Medical Device Performance Goals
Figure 11: Average FDA Review Time and Average Time to Final Decision for
Original PMAs in the Fiscal Year 2000-2010 Cohorts




Note: FDA review time refers to the time that FDA spends reviewing a submission and therefore
excludes any time the sponsor may spend responding to FDA requests for additional information.
Time to final decision includes both the time FDA spends reviewing a submission and the time the
sponsor may spend responding to any FDA action.
a
 The analysis of time to final decision includes only those original PMAs for which FDA or the sponsor
had made a final decision (i.e., a decision that ends the review process such as an approval, denial,
or withdrawal); this includes reviews by CBER through September 30, 2011, and reviews by CDRH
through December 1, 2011. Submissions without a final decision are not included in the results for
each cohort shown above. We considered a cohort to be incomplete if more than 10 percent of
submissions had not yet received a final decision. For this analysis, the FYs 2009-2010 original PMA
cohorts were still incomplete. Specifically, 22 percent of the FY 2009 original PMA cohort and
46.3 percent of the FY 2010 cohort had not yet received a final decision. As a result, it was too soon
to tell what the final results for this cohort would be. It is possible that some of the reviews taking the
most time were among those not completed when we received FDA’s data. The average time to final
decision for these two cohorts may increase or decrease as those reviews are completed.




Page 30                                             GAO-12-418 Medical Device Performance Goals
b
 The analysis of FDA review time includes only those original PMAs for which FDA or the sponsor
had made a decision that would permanently stop the review clock for purposes of determining
whether FDA met its performance goals (i.e., an approval, approvable, not approvable, withdrawal, or
denial); this includes reviews by CBER through September 30, 2011, and reviews by CDRH through
December 1, 2011. Submissions without such a decision are not included in the results for each
cohort shown above. We considered a cohort to be incomplete if more than 10 percent of
submissions had not yet received such a decision. For this analysis, the FY 2010 original PMA cohort
was still incomplete. Specifically, for 18.5 percent of the FY 2010 original PMA cohort, a decision that
would permanently stop the review clock had not been made at the time we received FDA’s data. As
a result, it was too soon to tell what the final results for this cohort would be. It is possible that some
of the reviews taking the most time were among those not completed when we received FDA’s data.
The average FDA review time for this cohort may increase or decrease as those reviews are
completed.


The results for expedited PMAs fluctuated even more dramatically than
for original PMAs—likely due to the small number of submissions (about
7 per year on average). Average FDA review time for expedited PMAs
generally increased over the period that user fees have been in effect,
from 241 days for the FY 2003 cohort to 356 days for the FY 2008 cohort,
then fell to 245 days for the FY 2009 cohort (see fig. 12). The average
time to final decision for expedited PMAs was highly variable but overall
declined somewhat during this period, from 704 days for the FY 2003
cohort to 545 days for the FY 2009 cohort.




Page 31                                             GAO-12-418 Medical Device Performance Goals
Figure 12: Average FDA Review Time and Average Time to Final Decision for
Expedited PMAs in the Fiscal Year 2000-2010 Cohorts




Note: FDA review time refers to the time that FDA spends reviewing a submission and therefore
excludes any time the sponsor may spend responding to FDA requests for additional information.
Time to final decision includes both the time FDA spends reviewing a submission and the time the
sponsor may spend responding to any FDA action.
a
 The analysis of time to final decision includes only those expedited PMAs for which FDA or the
sponsor had made a final decision (i.e., a decision that ends the review process such as an approval,
denial, or withdrawal); this includes reviews by CBER through September 30, 2011, and reviews by
CDRH through December 1, 2011. Submissions without a final decision are not included in the results
for each cohort shown above. We considered a cohort to be incomplete if more than 10 percent of
submissions had not yet received a final decision. For this analysis, the FY 2010 expedited PMA
cohort was still incomplete. Specifically, 33 percent of the FY 2010 expedited PMA cohort had not yet
received a final decision. As a result, it was too soon to tell what the final results for this cohort would
be. It is possible that some of the reviews taking the most time were among those not completed
when we received FDA’s data. The average time to final decision for this cohort may increase or
decrease as those reviews are completed.




Page 32                                              GAO-12-418 Medical Device Performance Goals
                          b
                           The analysis of FDA review time includes only those expedited PMAs for which FDA or the sponsor
                          had made a decision that would permanently stop the review clock for purposes of determining
                          whether FDA met its performance goals (i.e., an approval, approvable, not approvable, withdrawal, or
                          denial); this includes reviews by CBER through September 30, 2011, and reviews by CDRH through
                          December 1, 2011. Submissions without such a decision are not included in the results for each
                          cohort shown above. We considered a cohort to be incomplete if more than 10 percent of
                          submissions had not yet received such a decision. For this analysis, the FY 2010 expedited PMA
                          cohort was still incomplete. Specifically, for 16.7 percent of the FY 2010 expedited PMA cohort, a
                          decision that would permanently stop the review clock had not been made at the time we received
                          FDA’s data. As a result, it was too soon to tell what the final results for this cohort would be. It is
                          possible that some of the reviews taking the most time were among those not completed when we
                          received FDA’s data. The average FDA review time for this cohort may increase or decrease as those
                          reviews are completed.




The Average Number of     The average number of review cycles per original PMA increased
Review Cycles Increased   27.5 percent from 1.82 in the FY 2003 cohort (the first year that user fees
for Certain PMAs While    were in effect) to 2.32 cycles in the FY 2008 cohort. For expedited PMAs,
                          the average number of review cycles per submission was fairly steady at
the Percentage of PMAs    approximately 2.5 cycles until the FY 2004 cohort, then peaked at 4.0 in
Approved after One        the FY 2006 cohort before decreasing back to 2.5 cycles in the FY 2009
Review Cycle Generally    cohort. We found nearly identical trends when we examined the subsets
Decreased                 of original and expedited PMAs that received a final FDA decision of
                          approval.

                          In addition, the percentage of original PMAs receiving a decision of
                          approval at the end of the first review cycle fluctuated from FYs 2003
                          through 2009 but generally decreased—from 16 percent in the FY 2003
                          cohort to 9.8 percent in the FY 2009 cohort. Similarly, the percentage
                          receiving a first-cycle approvable decision decreased from 12 percent in
                          the FY 2003 cohort to 2.4 percent in the FY 2009 cohort. The percentage
                          of expedited PMAs receiving first-cycle approval fluctuated from year to
                          year, from 0 percent in 5 of the years we examined to a maximum of
                          25 percent in FY 2008.

                          The percentage of original PMAs that ultimately received approval from
                          FDA fluctuated from year to year but exhibited an overall decrease for the
                          completed cohorts from FYs 2003 through 2008. Specifically,
                          74.0 percent of original PMAs in the FY 2003 cohort were ultimately
                          approved, compared to 68.8 percent of the FY 2008 cohort. The
                          percentage of expedited PMAs that were ultimately approved varied
                          significantly from FYs 2003 through 2009, from a low of 0 percent in the
                          FY 2007 cohort to a high of 100 percent in the FY 2006 cohort.




                          Page 33                                           GAO-12-418 Medical Device Performance Goals
                        The industry groups and consumer advocacy groups we interviewed
Stakeholders Noted      noted a number of issues related to FDA’s review of medical device
Issues with the         submissions. The most commonly mentioned issue raised by industry and
                        consumer advocacy stakeholder groups was insufficient communication
Medical Device          between FDA and stakeholders throughout the review process. Industry
Review Process; FDA     stakeholders also noted a lack of predictability and consistency in reviews
Is Taking Steps That    and an increase in time to final decision. Consumer advocacy group
                        stakeholders noted issues related to inadequate assurance of the safety
May Address Many of     and effectiveness of approved or cleared devices. FDA is taking steps
These Issues            that may address many of these issues.


Stakeholders Cite       Most of the three industry and four consumer advocacy group
Insufficient            stakeholders that we interviewed told us that there is insufficient
Communication between   communication between FDA and stakeholders throughout the review
                        process. For example, four stakeholders noted that FDA does not clearly
FDA and Stakeholders    communicate to stakeholders the regulatory standards that it uses to
throughout the Review   evaluate submissions. In particular, industry stakeholders noted problems
Process                 with the regulatory guidance documents issued by FDA. These
                        stakeholders noted that these guidance documents are often unclear, out
                        of date, and not comprehensive. Stakeholders also noted that after
                        sponsors submit their applications to FDA, insufficient communication
                        from FDA prevents sponsors from learning about deficiencies in their
                        submissions early in FDA’s review. According to one of these
                        stakeholders, if FDA communicated these deficiencies earlier in the
                        process, sponsors would be able to correct them and would be less likely
                        to receive a request for additional information. Two consumer advocacy
                        group stakeholders also noted that FDA does not sufficiently seek patient
                        input during reviews. One stakeholder noted that it is important for FDA to
                        incorporate patient perspectives into its reviews of medical devices
                        because patients might weigh the benefits and risks of a certain device
                        differently than FDA reviewers.

                        FDA has taken or plans to take several steps that may address issues
                        with the frequency and quality of its communications with stakeholders,
                        including issuing new guidance documents, improving the guidance
                        development process, and enhancing interactions between FDA and
                        stakeholders during reviews. For example, in December 2011, FDA
                        released draft guidance about the regulatory framework, policies, and
                        practices underlying FDA’s 510(k) review in order to enhance the




                        Page 34                             GAO-12-418 Medical Device Performance Goals
                         transparency of this program. 40 In addition, FDA implemented a tracking
                         system and released a standard operating procedure (SOP) for
                         developing guidance documents for medical device reviews to provide
                         greater clarity, predictability, and efficiency in this process. 41 FDA also
                         created a new staff position to oversee the guidance development
                         process. Additionally, according to an overview of recent FDA actions to
                         improve its device review programs, FDA is currently enhancing its
                         interactive review process for medical device reviews by establishing
                         performance goals for early and substantive interactions between FDA
                         and sponsors during reviews. 42 This overview also notes that FDA is
                         currently working with a coalition of patient advocacy groups on
                         establishing mechanisms for obtaining reliable information on patient
                         perspectives during medical device reviews. 43


Industry Stakeholders    The three industry stakeholders that we interviewed also told us that there
Report a Lack of         is a lack of predictability and consistency in FDA’s reviews of device
Predictability and       submissions. For example, two stakeholders noted that review criteria
                         sometimes change after a sponsor submits an application. In particular,
Consistency in Reviews   one of these stakeholders noted that criteria sometimes change when the
                         FDA reviewer assigned to the submission changes during the review.
                         Additionally, stakeholders noted that there is sometimes inconsistent


                         40
                          See U.S. Department of Health and Human Services, Food and Drug Administration,
                         Draft Guidance for Industry and Food and Drug Administration Staff. The 510(k) Program:
                         Evaluating Substantial Equivalence in Premarket Notifications [510(k)] (Silver Spring, Md.:
                         Dec. 27, 2011).
                         41
                          See U.S. Department of Health and Human Services, Food and Drug Administration,
                         CDRH Guidance Development (Silver Spring, Md.: Aug. 1, 2011).
                         42
                           See U.S. Department of Health and Human Services, Food and Drug Administration,
                         Medical Device Premarket Programs: An Overview of FDA Actions (Silver Spring, Md.:
                         Oct. 19, 2011). Interactive review—which was established following the 2007
                         reauthorization of the medical device user fee program—was created to formalize a
                         process to encourage and facilitate communication between FDA and sponsors during
                         reviews.
                         43
                           Additionally, FDA recently issued draft guidance regarding the factors that FDA
                         considers when making benefit-risk determinations in order to increase the transparency
                         of these determinations. See U.S. Department of Health and Human Services, Food and
                         Drug Administration, Draft Guidance for Industry and Food and Drug Administration Staff.
                         Factors to Consider when Making Benefit-Risk Determinations in Medical Device
                         Premarket Review (Silver Spring, Md.: Aug. 15, 2011). According to FDA, the criteria in
                         this guidance take a patient-centric approach by calling for the consideration of patients’
                         tolerance for risk.




                         Page 35                                      GAO-12-418 Medical Device Performance Goals
application of criteria across review divisions or across individual
reviewers. Stakeholders noted that enhanced training for reviewers and
enhanced supervisory oversight could help resolve inconsistencies in
reviews and increase predictability for sponsors.

In the two internal assessments of its device review programs that FDA
released in August 2010, the agency found that insufficient predictability
in its review programs was a significant problem. 44 FDA has taken steps
that may address issues with the predictability and consistency of its
reviews of device submissions, including issuing new SOPs for reviews
and enhancing training for FDA staff. For example, in June 2011, FDA
issued an SOP to standardize the practice of quickly issuing written
notices to sponsors to inform them about changes in FDA’s regulatory
expectations for medical device submissions. 45 FDA also recently
developed an SOP to assure greater consistency in the review of device
submissions when review staff change during the review. 46 Additionally, in
April 2010, FDA began a reviewer certification program for new FDA




44
  See U.S. Department of Health and Human Services, Food and Drug Administration,
510(k) Working Group: Preliminary Report and Recommendations (Silver Spring, Md.:
August 2010) and U.S. Department of Health and Human Services, Food and Drug
Administration, Task Force on the Utilization of Science in Regulatory Decision Making,
Preliminary Report and Recommendations (Silver Spring, Md.: August 2010). FDA
identified several causes of the issues noted in these assessments, including very high
reviewer and manager turnover at CDRH; insufficient training for staff and industry;
extremely high ratios of front-line supervisors to employees; insufficient oversight by
managers; CDRH’s rapidly growing workload, caused by the increasing complexity of
devices and the number of submissions reviewed; unnecessary and/or inconsistent data
requirements imposed on device sponsors; insufficient guidance for industry; and poor-
quality submissions from industry.
45
  See U.S. Department of Health and Human Services, Food and Drug Administration,
CDRH Standard Operating Procedure for “Notice to Industry” Letters (Silver Spring, Md.:
June 14, 2011). The SOP provides a streamlined, systematic process for communicating
with industry via the guidance process and other means, as appropriate. Notice to Industry
letters are short communications that describe, at a very high level, changes to scientific
data requirements and FDA’s reasons for those changes. Some of these letters may
constitute guidance, while others will not. Because these letters are short and are
overseen by upper management at the Center, they can be developed and released in
roughly 3 weeks. FDA posts these letters on its website and also uses additional methods
for distributing the letters to stakeholders.
46
 See U.S. Department of Health and Human Services, Food and Drug Administration,
SOP: Management of Review Staff Changes During the Review of a Premarket
Submission (Silver Spring, Md.: Dec. 27, 2011).




Page 36                                     GAO-12-418 Medical Device Performance Goals
                           reviewers designed to improve the consistency of reviews. 47 According to
                           the overview of recent FDA actions to improve its device review
                           programs, FDA also plans to implement an experiential learning program
                           for new reviewers to give them a better understanding of how medical
                           devices are designed, manufactured, and used. 48


Industry Stakeholders      The three industry stakeholders we interviewed told us that the time to
Note an Increase in Time   final decision for device submissions has increased in recent years. This
to Final Decision          is consistent with our analysis, which showed that the average time to
                           final decision has increased for completed 510(k) and original PMA
                           cohorts since FY 2003. Additionally, stakeholders noted that FDA has
                           increased the number of requests for additional information, which our
                           analysis also shows. Stakeholders told us they believe the additional
                           information being requested is not always critical for the review of the
                           submission. Additional information requests increase the time to final
                           decision but not necessarily the FDA review time because FDA stops the
                           review clock when it requests additional information from sponsors. Two
                           of the stakeholders stated that reviewers may be requesting additional
                           information more often due to a culture of increased risk aversion at FDA
                           or because they want to stop the review clock in order to meet
                           performance goals.

                           According to FDA, the most significant contributor to the increased
                           number of requests for additional information—and therefore increased
                           time to final decision—is the poor quality of submissions received from
                           sponsors. In July 2011, FDA released an analysis it conducted of review
                           times under the 510(k) program. 49 According to FDA, in over 80 percent
                           of the reviews studied for this analysis, reviewers asked for additional



                           47
                             See U.S. Department of Health and Human Services, Food and Drug Administration,
                           Driving Biomedical Innovation: Initiatives to Improve Products for Patients (Silver Spring,
                           Md.: October 2011). The reviewer certification program includes up to 18 months of
                           training on specific core competencies through online training modules, instructor-led
                           courses, and practical experience.
                           48
                             The experiential learning program will include visits to academic institutions,
                           manufacturers, research organizations, and health care facilities to provide new reviewers
                           with a broader view of the regulatory process for medical devices.
                           49
                             See U.S. Department of Health and Human Services, Food and Drug Administration,
                           Analysis of Premarket Review Times Under the 510(k) Program (Silver Spring, Md.: July
                           2011).




                           Page 37                                      GAO-12-418 Medical Device Performance Goals
                          information from sponsors due to problems with the quality of the
                          submission. 50 FDA officials told us that sending a request for additional
                          information is often the only option for reviewers besides issuing a
                          negative decision to the sponsor. FDA’s analysis also found that
                          8 percent of its requests for additional information during the first review
                          cycle were inappropriate. Requests for additional information were
                          deemed inappropriate if FDA requested additional information or data for
                          a 510(k) that (1) were not justified, (2) were not permissible as a matter of
                          federal law or FDA policy, or (3) were unnecessary to make a substantial
                          equivalence determination. FDA has taken steps that may address issues
                          with the number of inappropriate requests for additional information. For
                          example, the overview of recent FDA actions indicates the agency is
                          developing an SOP for requests for additional information that clarifies
                          when these requests can be made for 510(k)s, the types of requests that
                          can be made, and the management level at which the decision must be
                          made.


Consumer Advocacy         Three of the four consumer advocacy group stakeholders with whom we
Group Stakeholders        spoke stated that FDA is not adequately ensuring the safety and
Suggest That FDA          effectiveness of the devices it approves or clears for marketing. One of
                          these stakeholders told us that FDA prioritizes review speed over safety
Provides Inadequate       and effectiveness. Two stakeholders also noted that the standards FDA
Assurance of the Safety   uses to approve or clear devices are lower than the standards that FDA
and Effectiveness of      uses to approve drugs, particularly for the 510(k) program. Two
Approved or Cleared       stakeholders also expressed concern that devices reviewed under the
Devices                   510(k) program are not always sufficiently similar to their predicates and
                          that devices whose predicates are recalled due to safety concerns do not
                          have to be reassessed to ensure that they are safe. Finally, three
                          stakeholders told us that FDA does not gather enough data on long-term
                          device safety and effectiveness through methods such as postmarket
                          analysis and device tracking.

                          These issues are similar to those raised elsewhere, such as a public
                          meeting to discuss the reauthorization of the medical device user fee
                          program, a congressional hearing, and an Institute of Medicine (IOM)


                          50
                            Quality problems included (1) inadequate device description, (2) discrepancies
                          throughout the submission, (3) problems with the indications for use, (4) failure to follow or
                          otherwise address current guidance documents or recognized standards, (5) lack of
                          performance data, and (6) lack of clinical data.




                          Page 38                                       GAO-12-418 Medical Device Performance Goals
report. For example, during a September 14, 2010, public meeting to
discuss the reauthorization, consumer advocacy groups—including two of
those we interviewed for our report—urged the inclusion of safety and
effectiveness improvements in the reauthorization, including raising
premarket review standards for devices and increasing postmarket
surveillance. 51 Additionally, during an April 13, 2011, congressional
hearing, another consumer advocacy group expressed concerns about
FDA’s 510(k) review process and recalls of high-risk devices that were
cleared through this process. 52 Finally, in July 2011, IOM released a
report summarizing the results of an independent evaluation of the 510(k)
program. FDA had requested that IOM conduct this evaluation to
determine whether the 510(k) program optimally protects patients and
promotes innovation. IOM concluded that clearance of a 510(k) based on
substantial equivalence to a predicate device is not a determination that
the cleared device is safe or effective. 53

FDA has taken or plans to take steps that may address issues with the
safety and effectiveness of approved and cleared devices, including
evaluating the 510(k) program and developing new data systems. For
example, FDA analyzed the safety of 510(k) devices cleared on the basis
of multiple predicates by investigating an apparent association between




51
 See U.S. Department of Health and Human Services, Food and Drug Administration,
Medical Device User Fee Program Public Meeting (Hyattsville, Md.: Sept. 14, 2010).
52
  See A Delicate Balance: FDA and the Reform of the Medical Device Approval Process,
                                                                        th
Hearing Before the Special Committee on Aging, United States Senate, 112 Cong.
(2011) (statement of Diana Zuckerman, President, National Research Center for Women
and Families, Cancer Prevention and Treatment Fund).
53
  See Institute of Medicine of the National Academies, Medical Devices and the Public’s
Health, The FDA 510(k) Clearance Process at 35 Years (Washington, D.C.: July 29,
2011). IOM concluded that the standard for assessing substantial equivalence to a
predicate device for clearance under the 510(k) program generally does not require
evidence of safety or effectiveness of a device. Therefore, IOM concluded that FDA
cannot evaluate the safety and effectiveness of devices as long as the standard for
clearance is substantial equivalence, as directed in statute. IOM also concluded that
available information on postmarket performance of devices does not provide sufficient
information about potential harm or lack of effectiveness to be a useful source of data on
the safety and effectiveness of marketed devices. IOM does not believe, however, that
there is a public-health crisis related to unsafe or ineffective medical devices. See also
GAO, Medical Devices: FDA Should Take Steps to Ensure That High-Risk Device Types
Are Approved through the Most Stringent Premarket Review Process, GAO-09-190
(Washington, D.C.: Jan. 15, 2009).




Page 39                                      GAO-12-418 Medical Device Performance Goals
               these devices and increased reports of adverse events. 54 FDA concluded
               that no clear relationship exists. FDA also conducted a public meeting to
               discuss the recommendations proposed in the IOM report in September
               2011. 55 FDA is also developing a device identification system that will
               allow FDA to better track devices that are distributed to patients, as well
               as an electronic reporting system that will assist with tracking and
               analyzing adverse events in marketed devices. 56


               While FDA has met most of the goals for the time frames within which the
Concluding     agency was to review and take action on 510(k) and PMA device
Observations   submissions, the time that elapses before a final decision has been
               increasing. This is particularly true for 510(k) submissions, which
               comprise the bulk of FDA device reviews. Stakeholders we spoke with
               point to a number of issues that the agency could consider in addressing
               the cause of these time increases. FDA tracks and reports the time to
               final decision in its annual reports to Congress on the medical device user
               fee program, and its own reports reveal the same pattern we found. In its
               July 2011 analysis of 510(k) submissions, FDA concluded that reviewers
               asked for additional information from sponsors—thus stopping the clock
               on FDA’s review time while the total time to reach a final decision
               continued to elapse—mainly due to problems with the quality of the
               submission. FDA is taking steps that may address the increasing time to
               final decision. It is important for the agency to monitor the impact of those
               steps in ensuring that safe and effective medical devices are reaching the
               market in a timely manner.




               54
                 See U.S. Department of Health and Human Services, Food and Drug Administration,
               Medical Device Reporting (MDR) Rate in 510(k) Cleared Devices Using Multiple
               Predicates (Silver Spring, Md.: Oct. 14, 2011). A sponsor may cite more than one
               predicate device in a submission for several reasons. For example, multiple devices, each
               with its own predicate, may be bundled together into one submission. A sponsor may also
               cite multiple predicates when a single device combines the functions of more than one
               device.
               55
                See 76 Fed. Reg. 50,230 (Aug. 12, 2011).
               56
                 FDA held a public workshop on the adoption, implementation, and use of unique device
               identifiers in various health-related electronic data systems in September 2011. See 76
               Fed. Reg. 43,691 (July 21, 2011). According to FDA’s plan of action, FDA is also currently
               developing proposed regulations for the unique device identifier system.




               Page 40                                     GAO-12-418 Medical Device Performance Goals
                  HHS reviewed a draft of this report and provided written comments, which
Agency Comments   are reprinted in appendix III. HHS generally agreed with our findings and
                  noted that FDA has identified some of the same performance trends in its
                  annual reports to Congress. HHS noted that because the total time to
                  final decision includes the time industry incurs in responding to FDA’s
                  concerns, FDA and industry bear shared responsibility for the increase in
                  this time and will need to work together to achieve improvement. HHS
                  also noted that in January 2011, FDA announced 25 specific actions that
                  the agency would take to improve the predictability, consistency, and
                  transparency of its premarket medical device review programs. Since
                  then, HHS stated, FDA has taken or is taking actions designed to create a
                  culture change toward greater transparency, interaction, collaboration,
                  and the appropriate balancing of benefits and risk; ensure predictable and
                  consistent recommendations, decision making, and application of the
                  least burdensome principle; and implement efficient processes and use of
                  resources. HHS also provided technical comments, which we
                  incorporated as appropriate.


                  As agreed with your office, unless you publicly announce the contents of
                  this report earlier, we plan no further distribution until 30 days from the
                  report date. At that time, we will send copies of this report to the
                  Secretary of Health and Human Services, the Commissioner of FDA, and
                  other interested parties. In addition, the report will be available at no
                  charge on the GAO website at http://www.gao.gov.

                  If you or your staff have any questions about this report, please contact
                  me at (202) 512-7114 or crossem@gao.gov. Contact points for our
                  Offices of Congressional Relations and Public Affairs may be found on
                  the last page of this report. GAO staff who made key contributions to this
                  report are listed in appendix IV.




                  Marcia Crosse
                  Director, Health Care




                  Page 41                              GAO-12-418 Medical Device Performance Goals
Appendix I: FDA Medical Device Review
                                              Appendix I: FDA Medical Device Review
                                              Performance for Fiscal Years (FY) 2000–2011



Performance for Fiscal Years (FY) 2000–2011

Table 4: FDA Premarket Notification (510(k)) Review Performance, FYs 2000–2011

                                Prior to implementation
                                 of the Medical Device                                                    Period covered by the
                                      User Fee and                                                       Medical Device User Fee
                                 Modernization Act of                                                     Amendments of 2007
                                    2002 (MDUFMA)               Period covered by MDUFMA                        (MDUFA)
                                                                                                                                        a
Fiscal year cohorts              2000    2001      2002      2003    2004    2005    2006     2007      2008   2009    2010      2011
Total number of 510(k)
submissions                      4,242   4,294    4,365      4,292 3,711 3,713 3,914 3,714             3,901 4,153 3,938         3,878
Number of 510(k)
submissions with a final
FDA decision                     4,242   4,294    4,365      4,292 3,711 3,713 3,914 3,713             3,899 4,148 3,853         2,366
 Number reviewed in
          b
 ≤ 90 days                       3,391   3,279    3,411      3,299 3,121 3,381 3,569 3,364             3,651 3,737 3,530         2,300
 Percentage reviewed in
          b
 ≤ 90 days                        79.9    76.4      78.1      76.9    84.1    91.1    91.2     90.6     93.6    90.1   91.6       97.2
                            c
 Tier 1 goal percentage             —       —           —       —       —       75      75       80       90      90     90        90
                  d                                                                                                          e          e
  Met Tier 1 goal                  n/a     n/a       n/a       n/a     n/a    Yes     Yes      Yes       Yes    Yes    Yes       Yes
 Number reviewed in
           b
 ≤ 150 days                      3,796   3,773    3,863      3,773 3,535 3,623 3,799 3,591             3,835 4,049 3,795         2,361
 Percentage reviewed in
           b
 ≤ 150 days                       89.5    87.9      88.5      87.9    95.3    97.6    97.1     96.7     98.4    97.6   98.5       99.8
                            c
 Tier 2 goal percentage             —       —           —       —       —       —       —        —        98      98     98        98
                  d                                                                                                          e          e
  Met Tier 2 goal                  n/a     n/a       n/a       n/a     n/a     n/a     n/a      n/a      Yes    Yes    Yes       Yes
 Average number of
 review cycles per
            f
 submission                       1.38    1.43      1.39      1.47    1.53    1.61    1.69     1.78     1.83    1.94   2.04       1.64
Of first-cycle decisions,
percentage that were
substantially equivalent
(SE)                              54.1    53.8      55.7      54.0    52.1    47.7    42.2     36.6     33.3    26.2   20.0       21.2
Among 510(k)
submissions with a final
decision, percentage of
final decisions that were:
 Substantially equivalent
 (SE)                             81.4    83.5      83.0      85.6    87.6    87.9    85.7     83.5    81.0    77.3    75.1      87.2
 Not substantially
 equivalent (NSE)                  1.1     1.9       1.9       2.9     3.5     3.7     3.7      3.7     3.8     5.5     6.4       2.4
 Withdrawn                         0.1     0.1       0.2       0.3     0.1     0.2     0.1      0.2     0.1     0.2     0.3       0.1
 Other                            17.4    14.5      14.9      11.3     8.7     8.2    10.4     12.7    15.1    16.9    18.3      10.4
Average FDA review time
(in days) for 510(k)
submissions that were not
reviewed within 150 days          195      201      201        197    195     186     221      228      259     282    226       176




                                              Page 42                                       GAO-12-418 Medical Device Performance Goals
                                         Appendix I: FDA Medical Device Review
                                         Performance for Fiscal Years (FY) 2000–2011




                           Prior to implementation
                            of the Medical Device                                                                    Period covered by the
                                 User Fee and                                                                       Medical Device User Fee
                            Modernization Act of                                                                     Amendments of 2007
                               2002 (MDUFMA)                      Period covered by MDUFMA                                 (MDUFA)
                                                                                                                                                a
Fiscal year cohorts         2000    2001        2002           2003       2004       2005       2006        2007   2008   2009    2010   2011
Average time to final
decision (in days) for
510(k) submissions that
were not reviewed within
150 days                     295      302         300            306        325        374        426        429   457     460    377     269
                                         Source: GAO analysis of Food and Drug Administration (FDA) data.

                                         Note: A review cohort includes all the medical device submissions relating to a particular performance
                                         goal that were submitted in a given fiscal year. For example, all 510(k)s received by FDA from
                                         October 1, 2010, to September 30, 2011, make up the 510(k) review cohort for FY 2011. Cohorts
                                         were considered complete if fewer than 10 percent of submissions were still under review at the time
                                         we received FDA’s data. All cohorts except FY 2011 were complete for the purposes of our analysis.
                                         As a result, it was too soon to tell what the final results for this cohort would be.
                                         a
                                          Approximately 39 percent of 510(k)s received in fiscal year (FY) 2011 were still under review at the
                                         time we received FDA’s data, which cover reviews by CDRH through October 26, 2011, and reviews
                                         by CBER through December 23, 2011. As a result, it was too soon to tell what the final results for this
                                         cohort would be. It is possible that some of the reviews taking the most time were among those not
                                         completed when we received FDA’s data. The percentage of 510(k)s reviewed within 90 days and
                                         within 150 days for the FY 2011 cohort may increase or decrease as those reviews are completed.
                                         The number of 510(k)s reviewed within 90 and 150 days and the average number of review cycles for
                                         the FY 2011 cohort may increase as those reviews are completed but will not decrease.
                                         b
                                          Only 510(k)s that had received a final decision from FDA were used to determine the number and
                                         percentage of 510(k)s reviewed within 90 days and within 150 days.
                                         c
                                          Fiscal years for which there was no corresponding 510(k) performance goal are denoted with a dash
                                         (—).
                                         d
                                          “n/a” denotes not applicable. In these years, there was no corresponding 510(k) performance goal
                                         and therefore no determination of whether the goal was met.
                                         e
                                          These results may change as the remaining 510(k) submissions for the FY 2010 and FY 2011
                                         cohorts receive final decisions.
                                         f
                                         Cycles that were currently in progress at the time we received FDA’s data were included in this
                                         analysis. The average number of review cycles for the FY 2011 cohort may increase as those reviews
                                         are completed but will not decrease.




                                         Page 43                                                       GAO-12-418 Medical Device Performance Goals
                                        Appendix I: FDA Medical Device Review
                                        Performance for Fiscal Years (FY) 2000–2011




Table 5: FDA Premarket Approval (PMA) Review Performance for Original PMAs, FYs 2000–2011

                                  Pre-MDUFMA                                       MDUFMA                             MDUFA
                                                                                                                                  a         a
Fiscal years                     2000   2001      2002          2003        2004    2005    2006   2007     2008    2009   2010       2011
Total number of submissions        73      75         51            50        45      57      55     37       34      41      54        43
Number reviewed in
          b
≤ 180 days                         47      42         27            25        18      18      19     21       21      32      38        20
 Percentage reviewed in
           b
 ≤ 180 days                      67.1   56.8       52.9          50.0       40.0     32.1   34.5   56.8      61.8   80.0    86.4       90.9
                        c
 Tier 1 goal percentage            —       —          —             —         —       —       —      50       60      60      60        60
                    d                                                                                                             e         e
 Met Tier 1 goal                  n/a     n/a        n/a           n/a       n/a      n/a    n/a    n/a      Yes     Yes    Yes       Yes
Number reviewed in
          b
≤ 320 days                         63      67         40            47        43      50      45     33       27      35      44        22
 Percentage reviewed in
           b
 ≤ 320 days                      90.0   90.5       78.4          94.0       95.6     89.3   81.8   89.2      79.4   87.5   100.0      100.0
                        c
 Tier 2 goal percentage            —       —          —             —         —       —       80     90        —      —       —         —
                    d
 Met Tier 2 goal                  n/a     n/a        n/a           n/a       n/a      n/a   Yes     No        n/a    n/a     n/a        n/a
Number reviewed in
          b
≤ 295 days                         58      63         39            43        40      41      34     29       27      35      44        22
 Percentage reviewed in
           b
 ≤ 295 days                      82.9   85.1       76.5          86.0       88.9     73.2   61.8   78.4      79.4   87.5   100.0      100.0
                        c
 Tier 2 goal percentage            —       —          —             —         —       —       —      —        90      90      90        90
                    d                                                                                                             e         e
 Met Tier 2 goal                  n/a     n/a        n/a           n/a       n/a      n/a    n/a    n/a       No      No    Yes       Yes
Average number of review
                     f                                                                                                 g
cycles per submission            1.77   2.05       2.16          1.82       1.98     2.54   2.40   2.51      2.32   2.12    2.04       1.58
Of first-cycle decisions,
percentage that were approval    15.1   17.3         9.8         16.0        8.9      8.8   21.8    8.1       5.9    9.8     7.4        9.3
Among original PMAs with a
final decision, percentage of
final decisions that were:
            g
 Approval                        61.6   69.3       76.5          74.0       79.5     71.4   90.4   70.6      68.8   56.3    82.8       93.3
       g
 Denial                           0.0     0.0        0.0           0.0       0.0      0.0    0.0    0.0       0.0    0.0     0.0        0.0
                g
 Withdrawal                      31.5   28.0       23.5          26.0       20.5     28.6    9.6   29.4      31.3   43.8    17.2        6.7
Average FDA review time
(in days) for original PMAs
that were not reviewed within
          b                                                                                                                       h         h
295 days                          383    468        451           333        368     342     447    422      584     571     —          —
Average time to final decision
(in days) for original PMAs
that were not reviewed within
          g                                                                                                                       h         h
295 days                          752   1006        688           591        775     727     679    645      923     823     —          —
                                        Source: GAO analysis of FDA data.




                                        Page 44                                               GAO-12-418 Medical Device Performance Goals
Appendix I: FDA Medical Device Review
Performance for Fiscal Years (FY) 2000–2011




Notes: A review cohort includes all the medical device submissions relating to a particular
performance goal that were submitted in a given fiscal year. For example, all original PMAs received
by FDA from October 1, 2010, to September 30, 2011, make up the original PMA review cohort for
FY 2011. Cohorts were considered complete if fewer than 10 percent of submissions were still under
review at the time we received FDA’s data.
We treated PMA submissions as meeting the time frame for a given performance goal if they were
reviewed within the goal time plus any extension to the goal time that may have been made. The only
reason the goal time can be extended is if the sponsor submits a major amendment to the submission
on its own initiative (i.e., unsolicited by FDA).
a
 The FYs 2010 and 2011 original PMA cohorts were considered still incomplete. Specifically, for
18.5 percent of the FY 2010 original PMA cohort and 48.8 percent of the FY 2011 cohort, FDA had
not yet made a decision that would permanently stop the review clock for purposes of determining
whether FDA met its performance goals (i.e., an approval, approvable, not approvable, withdrawal, or
denial) at the time we received FDA’s data; this includes reviews by CBER through September 30,
2011, and reviews by CDRH through December 1, 2011. As a result, it was too soon to tell what the
final results for these cohorts would be. It is possible that some of the reviews taking the most time
were among those not completed when we received FDA’s data. The percentage of original PMAs
reviewed within 180 days for the FY 2010 and FY 2011 cohorts may increase or decrease as those
reviews are completed; the number reviewed within 180 days and the number and percentage
reviewed within 320 days and within 295 days may decrease as those reviews are completed.
b
 Only original PMAs that had received a decision permanently stopping the review clock were used to
determine the number and percentage of original PMAs reviewed within 180 days, within 320 days,
and within 295 days.
c
 Fiscal years for which there was no corresponding original PMA performance goal are denoted with
a dash (—).
d
 “n/a” denotes not applicable. In these years, there was no corresponding original PMA performance
goal and therefore no determination of whether the goal was met.
e
 These results may change as the remaining original PMA submissions for the FY 2010 and FY 2011
cohorts receive decisions that permanently stop the review clock for purposes of determining whether
FDA met its performance goals.
f
 Cycles that were currently in progress at the time we received FDA’s data were included in this
analysis. The average number of review cycles for the incomplete cohorts may increase as those
reviews are completed but will not decrease.
g
 This analysis includes only those original PMAs for which FDA or the sponsor had made a final
decision; this includes reviews by CBER through September 30, 2011, and reviews by CDRH through
December 1, 2011. For this analysis, the FYs 2009 through 2011 original PMA cohorts were
considered still incomplete. Specifically, 22 percent of the FY 2009 original PMA cohort, 46.3 percent
of the FY 2010 cohort, and 65.1 percent of the FY 2011 cohort had not yet received a final decision.
As a result, it was too soon to tell what the final results for these cohorts would be. It is possible that
some of the reviews taking the most time were among those not completed when we received FDA’s
data. The percentages of final decisions that were approval, denial, or withdrawal and the average
time to final decision for original PMAs not meeting the 295-day time frame for the FYs 2009 through
2011 cohorts may increase or decrease as those reviews are completed. The average number of
review cycles for the FYs 2009 through 2011 cohorts may increase as those reviews are completed
but will not decrease.
h
 For the FYs 2010 through 2011 cohorts, there were no original PMAs that had received a final
decision that did not meet the 295-day time frame.




Page 45                                             GAO-12-418 Medical Device Performance Goals
                                            Appendix I: FDA Medical Device Review
                                            Performance for Fiscal Years (FY) 2000–2011




Table 6: FDA Premarket Approval (PMA) Review Performance for Expedited PMAs, FYs 2000–2011

                                     Pre-MDUFMA                                        MDUFMA                             MDUFA
                                                                                                                                      a         a
Fiscal years                        2000   2001    2002          2003       2004        2005    2006   2007     2008    2009   2010       2011
Total number of submissions           11     10        12              4         17        6      3       2        4       4       6         7
Number reviewed in
          b
≤ 180 days                            2      3          3              2          8        1      0       0        1       2       2         1
 Percentage reviewed in
           b
 ≤ 180 days                         18.2   30.0     25.0          50.0          47.1    16.7     0.0     0.0     25.0   50.0    40.0       50.0
                            c
 Tier 1 goal percentage               —      —         —             —           —        —       —      —        50      50      50        50
                    d                                                                                                                 e         e
 Met Tier 1 goal                     n/a    n/a       n/a           n/a          n/a     n/a     n/a     n/a      No    Yes      No       Yes
Number reviewed in
          b
≤ 300 days                            8      6         11              3         14        5      2       0        2       3       5         2
 Percentage reviewed in
           b
 ≤ 300 days                         72.7   60.0     91.7          75.0          82.4    83.3    66.7     0.0     50.0   75.0   100.0      100.0
                                c
 ≤ 300 days goal percentage           —      —         —             —           —        70      80     90        —      —       —         —
                        d
 Met ≤ 300 days goal                 n/a    n/a       n/a           n/a          n/a     Yes     No      No       n/a    n/a      n/a       n/a
Number reviewed in
          b
≤ 280 days                            7      6         10              3         14        4      1       0        2       3       5         2
 Percentage reviewed in
           b
 ≤ 280 days                         63.6   60.0     83.3          75.0          82.4    66.7    33.3     0.0     50.0   75.0   100.0      100.0
                            c
 Tier 2 goal percentage               —      —         —             —           —        —       —      —        90      90      90        90
                    d                                                                                                                 e         e
 Met Tier 2 goal                     n/a    n/a       n/a           n/a          n/a     n/a     n/a     n/a      No     No     Yes       Yes
Average number of review
                     f
cycles per submission               2.45   2.60     2.08          2.75          2.29    3.50    4.00    3.00     2.75   2.50    2.00       2.00
Of first-cycle decisions,
percentage that were
approval                             0.0   20.0     16.7            0.0         17.6    16.7     0.0     0.0     25.0    0.0    16.7       14.3
Among expedited PMAs with
a final decision, percentage of
final decisions that were:
            g
 Approval                           81.8   90.0     75.0          75.0          81.3    66.7 100.0       0.0     25.0   75.0    75.0      100.0
        g
 Denial                              0.0    0.0       0.0           0.0          0.0     0.0     0.0     0.0     25.0    0.0      0.0       0.0
                g
 Withdrawal                         18.2   10.0     25.0          25.0          18.8    33.3     0.0   100.0     50.0   25.0    25.0        0.0
Average FDA review time (in
days) for expedited PMAs
that were not reviewed within
          b                                                                                                                           h         h
280 days                             343    425      308           447          483      427     511    344      489     414      —         —
Average time to final decision
(in days) for expedited PMAs
that were not reviewed within
          g                                                                                                                           h         h
280 days                             588    520      334         1125           713      939     792   1111     1017     483      —         —
                                            Source: GAO analysis of FDA data.




                                            Page 46                                                GAO-12-418 Medical Device Performance Goals
Appendix I: FDA Medical Device Review
Performance for Fiscal Years (FY) 2000–2011




Notes: A review cohort includes all the medical device submissions relating to a particular
performance goal that were submitted in a given fiscal year. For example, all expedited PMAs
received by FDA from October 1, 2010, to September 30, 2011, make up the expedited PMA review
cohort for FY 2011. Cohorts were considered complete if fewer than 10 percent of submissions were
still under review at the time we received FDA’s data. All cohorts except FY 2010 and FY 2011 were
complete for the purposes of our analysis. As a result, it was too soon to tell what the final results for
these cohorts would be.
We treated PMA submissions as meeting the time frame for a given performance goal if they were
reviewed within the goal time plus any extension to the goal time that may have been made. The only
reason the goal time can be extended is if the sponsor submits a major amendment to the submission
on its own initiative (i.e., unsolicited by FDA).
a
 The FYs 2010 and 2011 expedited PMA cohorts were considered still incomplete. Specifically,
33 percent of the FY 2010 expedited PMA cohort and 71.4 percent of the FY 2011 cohort had not yet
received a final decision; this includes reviews by CBER through September 30, 2011, and reviews
by CDRH through December 1, 2011. Additionally, for 16.7 percent of the FY 2010 expedited PMA
cohort and 71.4 percent of the FY 2011 cohort, FDA had not yet made a decision that would
permanently stop the review clock for purposes of determining whether FDA met its performance
goals (i.e., an approval, approvable, not approvable, withdrawal, or denial) at the time we received
FDA’s data. As a result, it was too soon to tell what the final results for these cohorts would be. It is
possible that some of the reviews taking the most time were among those not completed when we
received FDA’s data. The percentage of expedited PMAs reviewed within 180 days for the FY 2010
and FY 2011 cohorts may increase or decrease as those reviews are completed; the number
reviewed within 180 days and the number and percentage reviewed within 300 days and within
280 days may decrease as those reviews are completed. The percentages of final decisions that
were approval, denial, or withdrawal and the average time to final decision for the FYs 2010 through
2011 cohorts may increase or decrease as those reviews are completed. The average number of
review cycles for the FYs 2010 through 2011 cohorts may increase as those reviews are completed
but will not decrease.
b
 Only expedited PMAs that had received a decision permanently stopping the review clock were used
to determine the number and percentage of expedited PMAs reviewed within 180, 300, and 280 days.
c
Fiscal years for which there was no corresponding expedited PMA performance goal are denoted
with a dash (—).
d
 “n/a” denotes not applicable. In these years, there was no corresponding expedited PMA
performance goal and therefore no determination of whether the goal was met.
e
 These results may change as the remaining expedited PMA submissions for the FY 2010 and
FY 2011 cohorts receive decisions that permanently stop the review clock for purposes of
determining whether FDA met its performance goals.
f
 Cycles that were currently in progress at the time we received FDA’s data were included in this
analysis. The average number of review cycles for the incomplete cohorts may increase as those
reviews are completed but will not decrease.
g
 Only expedited PMAs that had received a final decision were used to determine the percentages of
final decisions that were approval, denial, or withdrawal, and the average time to final decision for
expedited PMAs not reviewed within 280 days.
h
 For the FYs 2010 through 2011 cohorts, there were no expedited PMAs that had received a final
decision that did not meet the 280-day time frame.




Page 47                                             GAO-12-418 Medical Device Performance Goals
Appendix II: Number of Full-time Equivalent
                                                    Appendix II: Number of Full-time Equivalent
                                                    (FTE) FDA Staff Supporting Medical Device
                                                    User Fee Activities, FYs 2003 through 2010


(FTE) FDA Staff Supporting Medical Device
User Fee Activities, FYs 2003 through 2010

                                                                                                 Number of FTEs in each fiscal year
FDA centers and offices                                                                 2003   2004   2005    2006    2007     2008     2009     2010
Center for Devices and Radiological Health (CDRH)
                                        a
    Office of the Center Director (OCD)                                                   13     12     15      19       34       35       52          63
    Office of Management Operations (OSM/OMO)                                             89     62     64      62       61       50       48          52
                                                b
    Office of Information Technology (OIT)                                                —      —      —       —        —        16       15          17
    Office of Compliance (OC)                                                             40     46     54      55       56       61       60          60
    Office of Device Evaluation (ODE)                                                    305    301    298     287     311      326      322          329
    Office of Science and Engineering Laboratories (OST/OSEL)                             88     94    110      91       89       86       87          94
    Office of Communication Education and Radiation Programs
    (OHIP/OCER)                                                                           42     42     47      35       45       43       35          43
    Office of Surveillance and Biometrics (OSB)                                           86     92    104     106       98     105      117          139
                                            c
    Office of In Vitro Diagnostics (OIVD)                                                 —      49     61      67       65       71       69         105
                                                          d
    Committee Conference Management (CCM)                                                 —      —      —       —        —         1        1          1
    CDRH Total                                                                           662    697    753     721     760      793      805          902
Center for Biologics Evaluation and Research (CBER)
    Center Director’s Office, Office of Management (OM), Office                           13     15     20      23       23       23       24          28
    of Information Management (OIM), and Office of
    Communication, Outreach, and Development (OCOD)
    Office of Blood Research & Review                                                     37     43     49      70       66       66       64          64
    Office of Cellular, Tissue & Gene Therapies                                           2      2       2       2        2        5        6          6
    Office of Vaccines Research & Review                                                  1      0       0       0        1        1        1          4
    Office of Therapeutics Research & Review                                              1      0       0       0        0        0        0          0
    Office of Biostatistics & Epidemiology                                                1      2       2       2        2        2        4          5
    Office of Compliance & Biologics Quality                                              5      6       9       6        8        6        4          7
    CBER Total                                                                            59     68     81     104     101      102      103          114
Office of Regulatory Affairs (ORA)
    ORA Total                                                                             59     60     62      64       64       62       57          69
Office of the Commissioner (OC)
    OC Total                                                                              77     70     77      78       90       74       84          86
                     e
Shared Service (SS)
    SS Total                                                                              —      20     60      53       57       64       60          59
All Centers and Offices Total                                                            857    915   1,034   1,020   1,071   1,095    1,109     1,230
                                                    Source: GAO analysis of FDA data.

                                                    Note: All FTE rounded to the nearest whole number. FTEs for each fiscal year may not add to the
                                                    fiscal year total due to rounding. One FTE represents 40 hours of work per week conducted by a
                                                    federal government employee over the course of 1 year. FTEs do not include contractors and
                                                    therefore provide a partial measure of staffing resources.




                                                    Page 48                                            GAO-12-418 Medical Device Performance Goals
Appendix II: Number of Full-time Equivalent
(FTE) FDA Staff Supporting Medical Device
User Fee Activities, FYs 2003 through 2010




a
 OCD includes Medical Device Fellowship Program employees even though the Fellows were
assigned to work throughout CDRH.
b
    OIT was included in the OMO FTE total prior to FY 2008.
c
OIVD did not exist prior to FY 2004. Also, the Radiology Devices Branch was moved from ODE to
OIVD between FY 2009 and FY 2010.
d
    CCM was included in the OMO FTE total prior to FY 2008.
e
    Shared Service FTE were not separated from the center FTE until FY 2004.




Page 49                                            GAO-12-418 Medical Device Performance Goals
Appendix III: Comments from the
             Appendix III: Comments from the Department
             of Health and Human Services



Department of Health and Human Services




             Page 50                                      GAO-12-418 Medical Device Performance Goals
Appendix III: Comments from the Department
of Health and Human Services




Page 51                                      GAO-12-418 Medical Device Performance Goals
Appendix III: Comments from the Department
of Health and Human Services




Page 52                                      GAO-12-418 Medical Device Performance Goals
Appendix IV: GAO Contact and Staff
                  Appendix IV: GAO Contact and Staff
                  Acknowledgments



Acknowledgments

                  Marcia Crosse, (202) 512-7114 or crossem@gao.gov:
GAO Contact
Staff             In addition to the contact named above, Robert Copeland, Assistant
                  Director; Carolyn Fitzgerald; Cathleen Hamann; Karen Howard; Hannah
Acknowledgments   Marston Minter; Lisa Motley; Aubrey Naffis; Michael Rose; and Rachel
                  Schulman made key contributions to this report.




(290977)
                  Page 53                              GAO-12-418 Medical Device Performance Goals
GAO’s Mission         The Government Accountability Office, the audit, evaluation, and
                      investigative arm of Congress, exists to support Congress in meeting its
                      constitutional responsibilities and to help improve the performance and
                      accountability of the federal government for the American people. GAO
                      examines the use of public funds; evaluates federal programs and
                      policies; and provides analyses, recommendations, and other assistance
                      to help Congress make informed oversight, policy, and funding decisions.
                      GAO’s commitment to good government is reflected in its core values of
                      accountability, integrity, and reliability.

                      The fastest and easiest way to obtain copies of GAO documents at no
Obtaining Copies of   cost is through GAO’s website (www.gao.gov). Each weekday afternoon,
GAO Reports and       GAO posts on its website newly released reports, testimony, and
                      correspondence. To have GAO e-mail you a list of newly posted products,
Testimony             go to www.gao.gov and select “E-mail Updates.”

Order by Phone        The price of each GAO publication reflects GAO’s actual cost of
                      production and distribution and depends on the number of pages in the
                      publication and whether the publication is printed in color or black and
                      white. Pricing and ordering information is posted on GAO’s website,
                      http://www.gao.gov/ordering.htm.
                      Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
                      TDD (202) 512-2537.
                      Orders may be paid for using American Express, Discover Card,
                      MasterCard, Visa, check, or money order. Call for additional information.
                      Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Connect with GAO      Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts.
                      Visit GAO on the web at www.gao.gov.
                      Contact:
To Report Fraud,
Waste, and Abuse in   Website: www.gao.gov/fraudnet/fraudnet.htm
                      E-mail: fraudnet@gao.gov
Federal Programs      Automated answering system: (800) 424-5454 or (202) 512-7470

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

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




                        Please Print on Recycled Paper.