This is the accessible text file for GAO report number GAO-12-418 
entitled 'Medical Devices: FDA Has Met Most Performance Goals but 
Device Reviews Are Taking Longer' which was released on March 29, 2012. 

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

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

February 2012: 

Medical Devices: 

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

GAO-12-418: 

GAO Highlights: 

Highlights of GAO-12-418, a report to congressional requesters. 

Why GAO Did This Study: 

Why GAO Did This Study
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. New 
devices are generally subject to FDA review via the 510(k) process, 
which determines if a device is substantially equivalent to another 
legally marketed device, or the more stringent premarket approval 
(PMA) process, which requires evidence providing reasonable assurance 
that the device is safe and effective. The Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA) authorized FDA to collect user fees 
from the medical device industry to support the process of reviewing 
device submissions. FDA also committed to performance goals that 
include time frames within which FDA is to take action on a proportion 
of medical device submissions. MDUFMA was reauthorized in 2007. 

Questions have been raised as to whether FDA is sufficiently meeting 
the performance goals and whether devices are reaching the market in a 
timely manner. In preparation for reauthorization, GAO was asked to 
(1) examine trends in FDA’s 510(k) review performance from fiscal 
years (FY) 2003-2010, (2) examine trends in FDA’s PMA review 
performance from FYs 2003-2010, and (3) describe stakeholder issues 
with FDA’s review processes and steps FDA is taking that may address 
these issues. To do this work, GAO examined FDA medical device review 
data, reviewed FDA user fee data, interviewed FDA staff regarding the 
medical device review process and FDA data, and interviewed three 
industry groups and four consumer advocacy groups. 

What GAO Found: 

What GAO Found
Even though FDA met all medical device performance goals for 510(k)s, 
the elapsed time from submission to final decision has increased 
substantially in recent years. This time to final decision includes 
the days FDA spends reviewing a submission as well as the days FDA 
spends waiting for a device sponsor to submit additional information 
in response to a request by the agency. FDA review time excludes this 
waiting time, and FDA review time alone is used to determine whether 
the agency met its performance goals. Each fiscal year since FY 2005 
(the first year that 510(k) performance goals were in place), FDA has 
reviewed over 90 percent of 510(k) submissions within 90 days, thus 
meeting the first of two 510(k) performance goals. FDA also met the 
second goal for all 3 fiscal years it was in place by reviewing at 
least 98 percent of 510(k) submissions within 150 days. Although FDA 
has not yet completed reviewing all of the FY 2011 submissions, the 
agency was exceeding both of these performance goals for those 
submissions on which it had taken action. Although FDA review time 
decreased slightly from FY 2003 through FY 2010, the time that elapsed 
before FDA’s final decision increased substantially. Specifically, 
from FY 2005 through FY 2010, the average time to final decision for 
510(k)s increased 61 percent, from 100 days to 161 days. 

FDA was inconsistent in meeting performance goals for PMA submissions. 
FDA designates PMAs as either original or expedited; those that FDA 
considers eligible for expedited review are devices intended to (a) 
treat or diagnose life-threatening or irreversibly debilitating 
conditions and (b) address an unmet 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 expedited PMA 
submissions only twice out of 7 years. FDA review time and time to 
final decision for both types of PMAs were highly variable but 
generally increased in recent years. For example, the average time to 
final decision for original PMAs increased from 462 days for FY 2003 
to 627 days for FY 2008 (the most recent year for which complete data 
are available). 

The three industry groups and four consumer advocacy groups GAO 
interviewed noted a number of issues related to FDA’s review of 
medical device submissions. The four issues most commonly raised by 
stakeholders included (1) insufficient communication between FDA and 
stakeholders throughout the review process, (2) a lack of 
predictability and consistency in reviews, (3) an increase in time to 
final decision, and (4) inadequate assurance of the safety and 
effectiveness of approved or cleared devices. FDA is taking steps-—
including issuing new guidance documents, enhancing reviewer training, 
and developing an electronic system for reporting adverse events—-that 
may address many of these issues. 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. 

In commenting on a draft of this report, HHS generally agreed with 
GAO’s findings and noted that FDA has identified some of the same 
performance trends in its annual reports to Congress. HHS also called 
attention to the activities FDA has undertaken to improve the medical 
device review process. 

View [hyperlink, http://www.gao.gov/products/GAO-12-418]. For more 
information, contact Marcia Crosse at (202) 512-7114 or 
crossem@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

FDA Met All Performance Goals for 510(k)s but the Time to Final 
Decision Has Increased Substantially in Recent Years: 

FDA Was Inconsistent in Meeting Performance Goals for PMAs While FDA 
Review Time and Time to Final Decision Generally Increased: 

Stakeholders Noted Issues with the Medical Device Review Process; FDA 
Is Taking Steps That May Address Many of These Issues: 

Concluding Observations: 

Agency Comments: 

Appendix I: FDA Medical Device Review Performance for Fiscal Years 
(FY) 2000-2011: 

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

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

Appendix IV: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: FDA's 510(k) Performance Goals, FYs 2003-2011: 

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

Table 3: FDA's PMA Performance Goals, FYs 2003-2011: 

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

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

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

Figures: 

Figure 1: Percentage of 510(k)s FDA Reviewed within 90 Days for the 
Fiscal Year 2000-2010 Cohorts: 

Figure 2: Percentage of 510(k)s FDA Reviewed within 150 Days for the 
Fiscal Year 2000-2010 Cohorts: 

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

Figure 4: Average Number of Review Cycles Per 510(k) for the Fiscal 
Year 2000-2010 Cohorts: 

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: 

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: 

Figure 7: Percentage of Original PMAs FDA Reviewed within 180 Days for 
the Fiscal Year 2000-2010 Cohorts: 

Figure 8: Percentage of Original PMAs FDA Reviewed within 320 Days and 
295 Days for the Fiscal Year 2000-2010 Cohorts: 

Figure 9: Percentage of Expedited PMAs FDA Reviewed within 180 Days 
for the Fiscal Year 2000-2010 Cohorts: 

Figure 10: Percentage of Expedited PMAs FDA Reviewed within 300 Days 
and 280 Days for the Fiscal Year 2000-2010 Cohorts: 

Figure 11: Average FDA Review Time and Average Time to Final Decision 
for Original PMAs in the Fiscal Year 2000-2010 Cohorts: 

Figure 12: Average FDA Review Time and Average Time to Final Decision 
for Expedited PMAs in the Fiscal Year 2000-2010 Cohorts: 

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: 

[End of section] 

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.[Footnote 
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.[Footnote 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.[Footnote 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.[Footnote 4] FDA's authority to collect user fees for 
medical devices expires on October 1, 2012, and the 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.[Footnote 5] Application fees are 
collected for a variety of medical device submission types, including 
premarket notifications (510(k)s) and premarket approvals (PMAs). 
[Footnote 6] 

Under each authorization of the medical device user fee program, FDA 
committed to performance goals related to the review of medical device 
submissions.[Footnote 7] The performance goals include specific time 
frames within which FDA is to take action on submissions.[Footnote 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 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.[Footnote 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.[Footnote 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); and the average number of review cycles prior to 
approval.[Footnote 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.[Footnote 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.[Footnote 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. 

Background: 

Medical devices are reviewed primarily by FDA's Center for Devices and 
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 its safety and effectiveness.[Footnote 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.[Footnote 15] The remaining device types are 
required to obtain FDA clearance or approval through either the 510(k) 
or PMA processes.[Footnote 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.[Footnote 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 example, of the 
more than 13,600 device submissions received by FDA in FYs 2008 
through 2010, 88 percent were 510(k)s.[Footnote 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.[Footnote 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).[Footnote 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). 

Each of these actions ends the review process for a submission. 
[Footnote 21] A sponsor's withdrawal of a submission also ends the 
review process. 

Table 1: FDA's 510(k) Performance Goals, FYs 2003-2011: 

Fiscal year cohort: Tier 1 goal percentage[B]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: 75%[C]; 
2006: 75%[C]; 
2007: 80%[C]; 
Period covered by MDUFA[A]: 
2008: 90%; 
2009: 90%; 
2010: 90%; 
2011: 90%. 

Fiscal year cohort: Tier 2 goal percentage[D]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty] 
Period covered by MDUFA[A]: 
2008: 98%; 
2009: 98%; 
2010: 98%; 
2011: 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 [Empty]. 

[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. 

[End of table] 

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 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.[Footnote 
22] 

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

Term: FDA review time; 
Definition: 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. 

Term: Time to final decision; 
Definition: 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. 

[End of table] 

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.[Footnote 23] PMAs are designated as either original or 
expedited.[Footnote 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 (b) address an 
unmet medical need.[Footnote 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.[Footnote 26] 

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: 

Fiscal year cohort: Original PMA Tier 1 goal percentage[C]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005[B]: [Empty]; 
2006[B]: [Empty]; 
2007[B]: 50%; 
Period covered by MDUFA[A]: 
2008: 60%; 
2009: 60%; 
2010: 60%; 
2011: 60%. 

Fiscal year cohort: Original PMA Tier 2 goal percentage[D]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005[B]: [Empty]; 
2006[B]: 80%; 
2007[B]: 90%; 
Period covered by MDUFA[A]: 
2008: 90%; 
2009: 90%; 
2010: 90%; 
2011: 90%. 

Fiscal year cohort: Expedited PMA Tier 1 goal percentage[E]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005[B]: [Empty]; 
2006[B]: [Empty]; 
2007[B]: [Empty]; 
Period covered by MDUFA[A]: 
2008: 50%; 
2009: 50%; 
2010: 50%; 
2011: 50%. 

Fiscal year cohort: Expedited PMA Tier 2 goal percentage[F]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005[B]: 70%; 
2006[B]: 80%; 
2007[B]: 90%; 
Period covered by MDUFA[A]: 
2008: 90%; 
2009: 90%; 
2010: 90%; 
2011: 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 
[Empty]. 

[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. 

[End of table] 

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 its medical device review programs.[Footnote 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.[Footnote 28] 

FDA Met All Performance Goals for 510(k)s but the Time to Final 
Decision Has Increased Substantially in Recent Years: 

For FYs 2003 through 2010, FDA met all Tier 1 and Tier 2 performance 
goals for 510(k)s. In addition, FDA review time for 510(k)s decreased 
slightly during this period, but time to final decision increased 
substantially. The average number of review cycles and FDA's requests 
for additional information for 510(k) submissions also increased 
during this period. 

FDA Met All Tier 1 and Tier 2 Performance Goals for 510(k)s from 2003 
through 2010: 

FDA met all Tier 1 performance goals for the completed 510(k) cohorts 
that had Tier 1 goals in place.[Footnote 29] The percentage of 510(k)s 
reviewed within 90 days (the current Tier 1 goal time frame) exceeded 
90 percent for the FYs 2005 through 2010 cohorts (see figure 1.) 
Although the 510(k) 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.[Footnote 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. 

Figure 1: Percentage of 510(k)s FDA Reviewed within 90 Days for the 
Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: vertical bar graph] 

Fiscal year 2000; 
Complete cohort where goal was not in place (shown for context): 79.9%. 

Fiscal year 2001; 
Complete cohort where goal was not in place (shown for context): 76.4%. 

Fiscal year 2002; 
Complete cohort where goal was not in place (shown for context): 78.1%. 

Fiscal year 2003; 
Complete cohort where goal was not in place (shown for context): 76.9%. 

Fiscal year 2004; 
Complete cohort where goal was not in place (shown for context): 84.1%. 

Fiscal year 2005;
Complete cohort where goal was in place: 91.1%; 
501 (k) Tier 1 Performance goal: 75%. 

Fiscal year 2006; 
Complete cohort where goal was in place: 91.2%; 
501 (k) Tier 1 Performance goal: 75%. 

Fiscal year 2007; 
Complete cohort where goal was in place: 90.6%; 
501 (k) Tier 1 Performance goal: 80%. 

Fiscal year 2008; 
Complete cohort where goal was in place: 93.6%; 
501 (k) Tier 1 Performance goal: 90%. 

Fiscal year 2009; 
Complete cohort where goal was in place: 90.1%; 
501 (k) Tier 1 Performance goal: 90%. 

Fiscal year 2010; 
Complete cohort where goal was in place: 91.6%; 
501 (k) Tier 1 Performance goal: 90%. 

For FY 2005-2006, the Tier 1 performance goal was to have 75 percent 
of 510(k)s reviewed within 90 days. 

For FY 2007, the Tier 1 performance goal was to have 80 percent of 
510(k)s reviewed within 90 days. 

For FY 2008-2010, the Tier 1 performance goal was to have 90 percent 
of 510(k)s reviewed within 90 days. 

Source: GAO analysis of FDA data. 

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. 

[End of figure] 

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 figure 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.[Footnote 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. 

Figure 2: Percentage of 510(k)s FDA Reviewed within 150 Days for the 
Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: vertical bar graph] 

Fiscal year 2000; 
Complete cohort where goal was not in place (shown for context): 89.5%. 

Fiscal year 2001; 
Complete cohort where goal was not in place (shown for context): 87.9%. 

Fiscal year 2002; 
Complete cohort where goal was not in place (shown for context): 88.5%. 

Fiscal year 2003; 
Complete cohort where goal was not in place (shown for context): 87.9%. 

Fiscal year 2004; 
Complete cohort where goal was not in place (shown for context): 95.3%. 

Fiscal year 2005;
Complete cohort where goal was not in place (shown for context): 97.6%. 

Fiscal year 2006; 
Complete cohort where goal was not in place (shown for context): 97.1%. 

Fiscal year 2007; 
Complete cohort where goal was not in place (shown for context): 96.7%. 

Fiscal year 2008; 
Complete cohort where goal was in place: 98.4%; 
501 (k) Tier 2 Performance goal: 98%. 

Fiscal year 2009; 
Complete cohort where goal was in place: 97.6%; 
501 (k) Tier 2 Performance goal: 98%. 

Fiscal year 2010; 
Complete cohort where goal was in place: 98.5%; 
501 (k) Tier 2 Performance goal: 98%. 

For FY 2008-2010, the Tier 2 performance goal was to have 98 percent 
of 510(k)s reviewed within 150 days. 

Source: GAO analysis of FDA data. 

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. 

[End of figure] 

FDA Review Time for 510(k)s Decreased Slightly from 2003 to 2010 but 
Time to Final Decision Increased Substantially: 

While the average FDA review time for 510(k) submissions decreased 
slightly from the FY 2003 cohort to the FY 2010 cohort, the time to 
final decision increased substantially. Specifically, the average 
number of days FDA spent on the clock reviewing a 510(k) varied 
somewhat but overall showed a small decrease from 75 days for the FY 
2003 cohort to 71 days for the FY 2010 cohort (see figure 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 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: 

[Refer to PDF for image: multiple line graph] 

Fiscal year: 2000; 
Average time to final decision: 105.1 days; 
Average FDA review time: 74.1 days. 

Fiscal year: 2001; 
Average time to final decision: 109.3 days; 
Average FDA review time: 79 days. 

Fiscal year: 2002; 
Average time to final decision: 109 days; 
Average FDA review time: 76.9 days. 

Fiscal year: 2003; 
Average time to final decision: 111.2 days; 
Average FDA review time: 75 days. 

Fiscal year: 2004; 
Average time to final decision: 99.2 days; 
Average FDA review time: 62.9 days. 

Fiscal year: 2005; 
Average time to final decision: 100.4 days; 
Average FDA review time: 56.1 days. 

Fiscal year: 2006; 
Average time to final decision: 113.4 days; 
Average FDA review time: 60.5 days. 

Fiscal year: 2007; 
Average time to final decision: 131.8 days; 
Average FDA review time: 65.6 days. 

Fiscal year: 2008; 
Average time to final decision: 137.2 days; 
Average FDA review time: 65.5 days. 

Fiscal year: 2009; 
Average time to final decision: 158 days; 
Average FDA review time: 72.6 days. 

Fiscal year: 2010; 
Average time to final decision: 160.5 days; 
Average FDA review time: 70.8 days. 

Source: GAO analysis of FDA review time. 

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. 

[End of figure] 

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.[Footnote 32] 

Number of Review Cycles and Requests for Additional Information 
Increased for 510(k) Submissions from 2003 to 2010: 

The average number of review cycles per 510(k) increased substantially 
(39 percent) from FYs 2003 through 2010, rising from 1.47 cycles for 
the FY 2003 cohort to 2.04 cycles for the FY 2010 cohort (see figure 
4). 

Figure 4: Average Number of Review Cycles Per 510(k) for the Fiscal 
Year 2000-2010 Cohorts: 

[Refer to PDF for image: line graph] 

Fiscal year: 2000; 
Average number of review cycles per submission: 1.38. 

Fiscal year: 2001; 
Average number of review cycles per submission: 1.43. 

Fiscal year: 2002; 
Average number of review cycles per submission: 1.39. 

Fiscal year: 2003; 
Average number of review cycles per submission: 1.47. 

Fiscal year: 2004; 
Average number of review cycles per submission: 1.53. 

Fiscal year: 2005; 
Average number of review cycles per submission: 1.61. 

Fiscal year: 2006; 
Average number of review cycles per submission: 1.69. 

Fiscal year: 2007; 
Average number of review cycles per submission: 1.78. 

Fiscal year: 2008; 
Average number of review cycles per submission: 1.83. 

Fiscal year: 2009; 
Average number of review cycles per submission: 1.94. 

Fiscal year: 2010; 
Average number of review cycles per submission: 2.04. 

Source: GAO analysis of FDA data. 

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

[End of figure] 

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 figure 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).[Footnote 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).[Footnote 34] 

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: 

[Refer to PDF for image: multiple line graph] 

Fiscal year: 2000; 
Substantially equivalent: 54.1%; 
Additional information requests: 37%. 

Fiscal year: 2001; 
Substantially equivalent: 53.8%; 
Additional information requests: 38.6%. 

Fiscal year: 2002; 
Substantially equivalent: 55.7%; 
Additional information requests: 35.9%. 

Fiscal year: 2003; 
Substantially equivalent: 54%; 
Additional information requests: 40.1%. 

Fiscal year: 2004; 
Substantially equivalent: 52.1%; 
Additional information requests: 43.5%. 

Fiscal year: 2005; 
Substantially equivalent: 47.7%; 
Additional information requests: 49.7%. 

Fiscal year: 2006; 
Substantially equivalent: 42.2%; 
Additional information requests: 55.6%. 

Fiscal year: 2007; 
Substantially equivalent: 36.6%; 
Additional information requests: 61.3%. 

Fiscal year: 2008; 
Substantially equivalent: 33.3%; 
Additional information requests: 64.5%. 

Fiscal year: 2009; 
Substantially equivalent: 26.2%; 
Additional information requests: 71.6%. 

Fiscal year: 2010; 
Substantially equivalent: 20%; 
Additional information requests: 77%. 

Source: GAO analysis of FDA data. 

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. 

[End of figure] 

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 
figure 6.) 

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: 

[Refer to PDF for image: multiple line graph] 

Fiscal year: 2000; 
Substantially equivalent: 81.4%; 
Not substantially equivalent: 1.1%. 

Fiscal year: 2001; 
Substantially equivalent: 83.5%; 
Not substantially equivalent: 1.9%. 

Fiscal year: 2002; 
Substantially equivalent: 83%; 
Not substantially equivalent: 1.9%. 

Fiscal year: 2003; 
Substantially equivalent: 85.6%; 
Not substantially equivalent: 2.9%. 

Fiscal year: 2004; 
Substantially equivalent: 87.6%; 
Not substantially equivalent: 3.5%. 

Fiscal year: 2005; 
Substantially equivalent: 87.9%; 
Not substantially equivalent: 3.7%. 

Fiscal year: 2006; 
Substantially equivalent: 85.7%; 
Not substantially equivalent: 3.7%. 

Fiscal year: 2007; 
Substantially equivalent: 83.5%; 
Not substantially equivalent: 3.7%. 

Fiscal year: 2008; 
Substantially equivalent: 81%; 
Not substantially equivalent: 3.8%. 

Fiscal year: 2009; 
Substantially equivalent: 77.3%; 
Not substantially equivalent: 5.5%. 

Fiscal year: 2010; 
Substantially equivalent: 75.1%; 
Not substantially equivalent: 6.4%. 

Source: GAO analysis of FDA data. 

Notes: Only 510(k)s that had received a final decision from FDA were 
included in this analysis. 

[End of figure] 

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). 

FDA Was Inconsistent in Meeting Performance Goals for PMAs While FDA 
Review Time and Time to Final Decision Generally Increased: 

For FYs 2003 through 2010, FDA met most of the goals for original PMAs 
but fell short on most of the goals for expedited PMAs. In addition, 
FDA review time and time to final decision for both types of PMAs 
generally increased during this period. Finally, the average number of 
review cycles increased for certain PMAs while the percentage of PMAs 
approved after one review cycle generally decreased. 

FDA Met Most Goals for Original PMAs but Fell Short of Most Goals for 
Expedited PMAs from 2003-2010: 

Since FY 2003, FDA met the original PMA performance goals for four of 
the seven completed cohorts that had goals in place, but met the goals 
for only two of the seven expedited PMA cohorts with goals.[Footnote 
35] Specifically, FDA met its Tier 1 performance goals for original 
PMAs for all three of the completed original PMA cohorts that had such 
goals in 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.[Footnote 36] (See figure 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.[Footnote 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. 

Figure 7: Percentage of Original PMAs FDA Reviewed within 180 Days for 
the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: vertical bar graph] 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 67.1%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 56.8%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 52.9%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 50%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 40%. 

Fiscal year: 2005; 
Complete cohort where goal was not in place (shown for context): 32.1%. 

Fiscal year: 2006; 
Complete cohort where goal was not in place (shown for context): 34.5%. 

Fiscal year: 2007; 
Complete cohort where goal was in place: 56.8%; 
Tier 1 performance goal: 50%. 

Fiscal year: 2008; 
Complete cohort where goal was in place: 61.8%; 
Tier 1 performance goal: 60%. 

Fiscal year: 2009; 
Complete cohort where goal was in place: 80%; 
Tier 1 performance goal: 60%. 

Fiscal year: 2010; 
Incomplete cohort where goal was in place: 86.4%; 
Tier 1 performance goal: 60%. 

For FY 2007, the Tier 1 performance goal was to have 50 percent of 
original PMAs reviewed within 180 days. 

For FY 2008-2010, the Tier 1 performance goal was to have 60 percent 
of original PMAs reviewed within 180 days. 

Source: GAO analysis of FDA data. 

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). 

[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. 

[End of figure] 

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 figure 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.[Footnote 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. 

Figure 8: Percentage of Original PMAs FDA Reviewed within 320 Days and 
295 Days for the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: 2 vertical bar graphs] 

Original PMAs reviewed within 320 days (MDUFMA Tier 2 goal): 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 90.0%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 90.5%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 78.4%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 94.0%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 95.6%. 

Fiscal year: 2005; 
Complete cohort where goal was not in place (shown for context): 89.3%. 

Fiscal year: 2006; 
Complete cohort where goal was place: 81.8%; 
Tier 2 performance goal: 80%. 

Fiscal year: 2007; 
Complete cohort where goal was in place: 89.2%; 
Tier 2 performance goal: 90%. 

Fiscal year: 2008; 
Complete cohort where goal was not in place (shown for context): 79.4%. 
Tier 2 performance goal: 60%. 

Fiscal year: 2009; 
Complete cohort where goal was not in place (shown for context): 87.5%. 
Tier 2 performance goal: 60%. 

Fiscal year: 2010; 
Incomplete cohort where goal was not in place (shown for context): 
100%[A]. 

For FY 2006, the Tier 2 performance goal was to have 80 percent of 
original PMAs reviewed within 320 days. 

For FY 2007, the Tier 2 performance goal was to have 90 percent of 
original PMAs reviewed within 320 days. 

[End of first graph] 

Original PMAs reviewed within 290 days (MDUFMA Tier 2 goal): 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 82.9%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 85.1%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 76.5%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 86.0%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 88.9%. 

Fiscal year: 2005; 
Complete cohort where goal was not in place (shown for context): 73.2%. 

Fiscal year: 2006; 
Complete cohort where goal was not in place (shown for context): 61.8%. 

Fiscal year: 2007; 
Complete cohort where goal was not in place (shown for context): 78.4%. 

Fiscal year: 2008; 
Complete cohort where goal was place: 79.4%; 
Tier 2 performance goal: 90%. 

Fiscal year: 2009; 
Complete cohort where goal was place: 87.0; 
Tier 2 performance goal: 90%. 

Fiscal year: 2010; 
Incomplete cohort where goal was in place: 100%[A]; 
Tier 2 performance goal: 90%. 

For FY 2008-2010, the Tier 2 performance goal was to have 90 percent 
of original PMAs reviewed within 295 days. 

Source: GAO analysis of FDA data. 

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). 

[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. 

[End of figure] 

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 figure 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. 

Figure 9: Percentage of Expedited PMAs FDA Reviewed within 180 Days 
for the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: vertical bar graph] 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 18.2%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 30%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 25%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 50%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 47.1%. 

Fiscal year: 2005; 
Complete cohort where goal was not in place (shown for context): 16.7%. 

Fiscal year: 2006; 
Complete cohort where goal was not in place (shown for context): 0; 

Fiscal year: 2007; 
Complete cohort where goal was not in place (shown for context): 0; 

Fiscal year: 2008; 
Complete cohort where goal was place: 25%. 
Tier 1 performance goal: 50%. 

Fiscal year: 2009; 
Complete cohort where goal was place: 50%. 
Tier 1 performance goal: 50%. 

Fiscal year: 2010; 
Incomplete cohort where goal was in place: 40%[A]; 
Tier 1 performance goal: 50%. 

For FY 2008-2010, the Tier 1 performance goal was to have 50 percent 
of original PMAs reviewed within 180 days. 

Source: GAO analysis of FDA data. 

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). 

[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. 

[End of figure] 

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 figure 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. 

Figure 10: Percentage of Expedited PMAs FDA Reviewed within 300 Days 
and 280 Days for the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: 2 vertical bar graphs] 

Expedited PMAs reviewed within 300 days (MDUFMA Tier 2 goal): 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 72.7%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 60.0%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 91.7%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 75.0%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 82.4%. 

Fiscal year: 2005; 
Complete cohort where goal was place: 83.3%; 
Tier 2 performance goal: 70%. 

Fiscal year: 2006; 
Complete cohort where goal was place: 66.7%; 
Tier 2 performance goal: 80%. 

Fiscal year: 2007; 
Complete cohort where goal was not in place (shown for context): 0.0%; 
Tier 2 performance goal: 90%. 

Fiscal year: 2008; 
Complete cohort where goal was not in place (shown for context): 50.0%. 

Fiscal year: 2009; 
Complete cohort where goal was not in place (shown for context): 75.0%. 

Fiscal year: 2010; 
Incomplete cohort where goal was not in place (shown for context): 
100%[A]. 

For FY 2005, the Tier 2 performance goal was to have 70 percent of 
original PMAs reviewed within 300 days. 

For FY 2006, the Tier 2 performance goal was to have 80 percent of 
original PMAs reviewed within 300 days. 

For FY 2007, the Tier 2 performance goal was to have 90 percent of 
original PMAs reviewed within 300 days. 

[End of first graph] 

Expedited PMAs reviewed within 290 days (MDUFMA Tier 2 goal): 

Fiscal year: 2000; 
Complete cohort where goal was not in place (shown for context): 63.6%. 

Fiscal year: 2001; 
Complete cohort where goal was not in place (shown for context): 60.0%. 

Fiscal year: 2002; 
Complete cohort where goal was not in place (shown for context): 83.3%. 

Fiscal year: 2003; 
Complete cohort where goal was not in place (shown for context): 75.0%. 

Fiscal year: 2004; 
Complete cohort where goal was not in place (shown for context): 82.4%. 

Fiscal year: 2005; 
Complete cohort where goal was not in place (shown for context): 66.7%. 

Fiscal year: 2006; 
Complete cohort where goal was not in place (shown for context): 33.3%. 

Fiscal year: 2007; 
Complete cohort where goal was not in place (shown for context): 0.0%. 

Fiscal year: 2008; 
Complete cohort where goal was place: 50.0%; 
Tier 2 performance goal: 90%. 

Fiscal year: 2009; 
Complete cohort where goal was place: 75.0; 
Tier 2 performance goal: 90%. 

Fiscal year: 2010; 
Incomplete cohort where goal was in place: 100%[A]; 
Tier 2 performance goal: 90%. 

For FY 2008-2010, the Tier 2 performance goal was to have 90 percent 
of original PMAs reviewed within 280 days. 

Source: GAO analysis of FDA data. 

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). 

[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. 

[End of figure] 

FDA Review Time and Time to Final Decision Generally Increased for 
PMAs from 2003 to 2010: 

FDA review time for both original and expedited PMAs was highly 
variable but generally increased across our analysis period, while 
time to final decision also increased for original PMAs. Specifically, 
average FDA review time for original PMAs increased from 211 days in 
the FY 2003 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 figure 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.[Footnote 39] 

Figure 11: Average FDA Review Time and Average Time to Final Decision 
for Original PMAs in the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: multiple line and plotted point graph] 

Fiscal year: 2000; 
Average time to final decision: 405 days; 
Average FDA review time: 201 days. 

Fiscal year: 2001; 
Average time to final decision: 593 days; 
Average FDA review time: 223 days. 

Fiscal year: 2002; 
Average time to final decision: 488 days; 
Average FDA review time: 246 days. 

Fiscal year: 2003; 
Average time to final decision: 462 days; 
Average FDA review time: 211 days. 

Fiscal year: 2004; 
Average time to final decision: 395 days; 
Average FDA review time: 225 days. 

Fiscal year: 2005; 
Average time to final decision: 580 days; 
Average FDA review time: 252 days. 

Fiscal year: 2006; 
Average time to final decision: 476 days; 
Average FDA review time: 300 days. 

Fiscal year: 2007; 
Average time to final decision: 564 days; 
Average FDA review time: 232 days. 

Fiscal year: 2008; 
Average time to final decision: 627 days; 
Average FDA review time: 264 days. 

Fiscal year: 2009; 
Average FDA review time: 217 days. 

Three plotted points: [A], [A], [B]. 

Source: GAO analysis of FDA data. 

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. 

[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 figure 
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. 

[End of figure] 

Figure 12: Average FDA Review Time and Average Time to Final Decision 
for Expedited PMAs in the Fiscal Year 2000-2010 Cohorts: 

[Refer to PDF for image: multiple line and plotted point graph] 

Fiscal year: 2000; 
Average time to final decision: 475 days; 
Average FDA review time: 275 days. 

Fiscal year: 2001; 
Average time to final decision: 382 days; 
Average FDA review time: 289 days. 

Fiscal year: 2002; 
Average time to final decision: 459 days; 
Average FDA review time: 224 days. 

Fiscal year: 2003; 
Average time to final decision: 704 days; 
Average FDA review time: 241 days. 

Fiscal year: 2004; 
Average time to final decision: 495 days; 
Average FDA review time: 250 days. 

Fiscal year: 2005; 
Average time to final decision: 1,044 days; 
Average FDA review time: 303 days. 

Fiscal year: 2006; 
Average time to final decision: 652 days; 
Average FDA review time: 417 days. 

Fiscal year: 2007; 
Average time to final decision: 1,111 days; 
Average FDA review time: 344 days. 

Fiscal year: 2008; 
Average time to final decision: 755 days; 
Average FDA review time: 356 days. 

Fiscal year: 2009; 
Average time to final decision: 545 days; 
Average FDA review time: 245 days. 

Two plotted points: [A], [B]. 

Source: GAO analysis of FDA data. 

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. 

[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. 

[End of figure] 

The Average Number of Review Cycles Increased for Certain PMAs While 
the Percentage of PMAs Approved after One Review Cycle Generally 
Decreased: 

The average number of review cycles per original PMA increased 27.5 
percent from 1.82 in the FY 2003 cohort (the first year that user fees 
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 approximately 2.5 cycles until the FY 2004 cohort, then 
peaked at 4.0 in the FY 2006 cohort before decreasing back to 2.5 
cycles in the FY 2009 cohort. We found nearly identical trends when we 
examined the subsets 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. 

Stakeholders Noted Issues with the Medical Device Review Process; FDA 
Is Taking Steps That May Address Many of These Issues: 

The industry groups and consumer advocacy groups we interviewed noted 
a number of issues related to FDA's review of medical device 
submissions. The most commonly mentioned issue raised by industry and 
consumer advocacy stakeholder groups was insufficient communication 
between FDA and stakeholders throughout the review process. Industry 
stakeholders also noted a lack of predictability and consistency in 
reviews and an increase in time to final decision. Consumer advocacy 
group stakeholders noted issues related to inadequate assurance of the 
safety and effectiveness of approved or cleared devices. FDA is taking 
steps that may address many of these issues. 

Stakeholders Cite Insufficient Communication between FDA and 
Stakeholders throughout the Review Process: 

Most of the three industry and four consumer advocacy group 
stakeholders that we interviewed told us that there is insufficient 
communication between FDA and stakeholders throughout the review 
process. For example, four stakeholders noted that FDA does not 
clearly communicate to stakeholders the regulatory standards that it 
uses to evaluate submissions. In particular, industry stakeholders 
noted problems 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 
transparency of this program.[Footnote 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.[Footnote 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. 
[Footnote 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.[Footnote 43] 

Industry Stakeholders Report a Lack of Predictability and Consistency 
in Reviews: 

The three industry stakeholders that we interviewed also told us that 
there is a lack of predictability and consistency in FDA's reviews of 
device submissions. For example, two stakeholders noted that review 
criteria sometimes change after a sponsor submits an application. In 
particular, 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 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.[Footnote 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.[Footnote 45] FDA also recently 
developed an SOP to assure greater consistency in the review of device 
submissions when review staff change during the review.[Footnote 46] 
Additionally, in April 2010, FDA began a reviewer certification 
program for new FDA reviewers designed to improve the consistency of 
reviews.[Footnote 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.[Footnote 48] 

Industry Stakeholders Note an Increase in Time to Final Decision: 

The three industry stakeholders we interviewed told us that the time 
to final decision for device submissions has increased in recent 
years. This 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.[Footnote 49] According to FDA, 
in over 80 percent of the reviews studied for this analysis, reviewers 
asked for additional information from sponsors due to problems with 
the quality of the submission.[Footnote 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 Group Stakeholders Suggest That FDA Provides 
Inadequate Assurance of the Safety and Effectiveness of Approved or 
Cleared Devices: 

Three of the four consumer advocacy group stakeholders with whom we 
spoke stated that FDA is not adequately ensuring the safety and 
effectiveness of the devices it approves or clears for marketing. One 
of these stakeholders told us that FDA prioritizes review speed over 
safety and effectiveness. Two stakeholders also noted that the 
standards FDA uses to approve or clear devices are lower than the 
standards that FDA uses to approve drugs, particularly for the 510(k) 
program. Two stakeholders also expressed concern that devices reviewed 
under the 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) 
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.[Footnote 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.[Footnote 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.[Footnote 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 these devices and increased reports of adverse 
events.[Footnote 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.[Footnote 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.[Footnote 56] 

Concluding Observations: 

While FDA has met most of the goals for the time frames within which 
the agency was to review and take action on 510(k) and PMA device 
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. 

Agency Comments: 

HHS reviewed a draft of this report and provided written comments, 
which 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 [hyperlink, 
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. 

Signed by: 

Marcia Crosse: 
Director, Health Care: 

[End of section] 

Appendix I: FDA Medical Device Review Performance for Fiscal Years 
(FY) 2000-2011: 

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

Fiscal year cohorts: Total number of 510(k) submissions; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 4,242; 
2001: 4,294; 
2002: 4,365; 
Period covered by MDUFMA: 
2003: 4,292; 
2004: 3,711; 
2005: 3,713; 
2006: 3,914; 
2007: 3,714; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 3,901; 
2009: 4,153; 
2010: 3,938; 
2011[A]: 3,878. 

Fiscal year cohorts: Number of 510(k) submissions with a final FDA 
decision; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 4,242; 
2001: 4,294; 
2002: 4,365; 
Period covered by MDUFMA: 
2003: 4,292; 
2004: 3,711; 
2005: 3,713; 
2006: 3,914; 
2007: 3,713; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 3,899; 
2009: 4,148; 
2010: 3,853; 
2011[A]: 2,366. 

Fiscal year cohorts: Number reviewed in less than or equal to 90 
days[B]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 3,391; 
2001: 3,279; 
2002: 3,411; 
Period covered by MDUFMA: 
2003: 3,299; 
2004: 3,121; 
2005: 3,381; 
2006: 3,569; 
2007: 3,364; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 3,651; 
2009: 3,737; 
2010: 3,530; 
2011[A]: 2,300. 

Fiscal year cohorts: Percentage reviewed in less than or equal to 90 
days[B]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 79.9; 
2001: 76.4; 
2002: 78.1; 
Period covered by MDUFMA: 
2003: 76.9; 
2004: 84.1; 
2005: 91.1; 
2006: 91.2; 
2007: 90.6; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 93.6; 
2009: 90.1; 
2010: 91.6; 
2011[A]: 97.2. 

Fiscal year cohorts: Tier 1 goal percentage[C]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: 75; 
2006: 75; 
2007: 80; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 90; 
2009: 90; 
2010: 90; 
2011[A]: 90. 

Fiscal year cohorts: Met Tier 1 goal[D]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: n/a; 
2001: n/a; 
2002: n/a; 
Period covered by MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: Yes; 
2006: Yes; 
2007: Yes; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: Yes; 
2009: Yes; 
2010: Yes[E]; 
2011[A]: Yes[E]. 

Fiscal year cohorts: Number reviewed in less than or equal to 150 
days[B]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 3,796; 
2001: 3,773; 
2002: 3,863; 
Period covered by MDUFMA: 
2003: 3,773; 
2004: 3,535; 
2005: 3,623; 
2006: 3,799; 
2007: 3,591; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 3,835; 
2009: 4,049; 
2010: 3,795; 
2011[A]: 2,361. 

Fiscal year cohorts: Percentage reviewed in less than or equal to 150 
days[B]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 89.5; 
2001: 87.9; 
2002: 88.5; 
Period covered by MDUFMA: 
2003: 87.9; 
2004: 95.3; 
2005: 97.6; 
2006: 97.1; 
2007: 96.7; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 2008: 
98.4; 
97.6; 
98.5; 
99.8. 

Fiscal year cohorts: Tier 2 goal percentage[C]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
Period covered by MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 98; 
2009: 98; 
2010: 98; 
2011[A]: 98. 

Fiscal year cohorts: Met Tier 2 goal[D]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: n/a; 
2001: n/a; 
2002: n/a; 
Period covered by MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: n/a; 
2007: n/a; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: Yes; 
2009: Yes; 
2010: Yes[E]; 
2011[A]: Yes[E]. 

Fiscal year cohorts: Average number of review cycles per submission[F]; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 1.38; 
2001: 1.43; 
2002: 1.39; 
Period covered by MDUFMA: 
2003: 1.47; 
2004: 1.53; 
2005: 1.61; 
2006: 1.69; 
2007: 1.78; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 1.83; 
2009: 1.94; 
2010: 2.04; 
2011[A]: 1.64. 

Fiscal year cohorts: Of first-cycle decisions, percentage that were 
substantially equivalent (SE); 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 54.1; 
2001: 53.8; 
2002: 55.7; 
Period covered by MDUFMA: 
2003: 54.0; 
2004: 52.1; 
2005: 47.7; 
2006: 42.2; 
2007: 36.6; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 33.3; 
2009: 26.2; 
2010: 20.0; 
2011[A]: 21.2. 

Fiscal year cohorts: Among 510(k) submissions with a final decision, 
percentage of final decisions that were: Substantially equivalent (SE); 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 81.4; 
2001: 83.5; 
2002: 83.0; 
Period covered by MDUFMA: 
2003: 85.6; 
2004: 87.6; 
2005: 87.9; 
2006: 85.7; 
2007: 83.5; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 81.0; 
2009: 77.3; 
2010: 75.1; 
2011[A]: 87.2. 

Fiscal year cohorts: Among 510(k) submissions with a final decision, 
percentage of final decisions that were: Not substantially equivalent 
(NSE); 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 1.1; 
2001: 1.9; 
2002: 1.9; 
Period covered by MDUFMA: 
2003: 2.9; 
2004: 3.5; 
2005: 3.7; 
2006: 3.7; 
2007: 3.7; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 3.8; 
2009: 5.5; 
2010: 6.4; 
2011[A]: 2.4. 

Fiscal year cohorts: Among 510(k) submissions with a final decision, 
percentage of final decisions that were: Withdrawn; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 0.1; 
2001: 0.1; 
2002: 0.2; 
Period covered by MDUFMA: 
2003: 0.3; 
2004: 0.1; 
2005: 0.2; 
2006: 0.1; 
2007: 0.2; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 0.1; 
2009: 0.2; 
2010: 0.3; 
2011[A]: 0.1. 

Fiscal year cohorts: Among 510(k) submissions with a final decision, 
percentage of final decisions that were: Other; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 17.4; 
2001: 14.5; 
2002: 14.9; 
Period covered by MDUFMA: 
2003: 11.3; 
2004: 8.7; 
2005: 8.2; 
2006: 10.4; 
2007: 12.7; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 15.1; 
2009: 16.9; 
2010: 18.3; 
2011[A]: 10.4. 

Fiscal year cohorts: Average FDA review time (in days) for 510(k) 
submissions that were not reviewed within 150 days; 
Prior to implementation of the Medical Device User Fee and 
Modernization Act of 2002 (MDUFMA): 
2000: 195; 
2001: 201; 
2002: 201; 
Period covered by MDUFMA: 
2003: 197; 
2004: 195; 
2005: 186; 
2006: 221; 
2007: 228; 
Period covered by the Medical Device User Fee Amendments of 2007 
(MDUFA): 
2008: 259; 
2009: 282; 
2010: 226; 
2011[A]: 176. 

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

[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. 

[End of table] 

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

Fiscal years: 

Total number of submissions; 
Pre-MDUFMA: 
2000: 73; 
2001: 75; 
2002: 51; 
MDUFMA: 
2003: 50; 
2004: 45; 
2005: 57; 
2006: 55; 
2007: 37; 
MDUFA: 
2008: 34; 
2009: 41; 
2010[A]: 54; 
2011[A]: 43. 

Number reviewed in less than or equal to 180 days[B]; 
Pre-MDUFMA: 
2000: 47; 
2001: 42; 
2002: 27; 
MDUFMA: 
2003: 25; 
2004: 18; 
2005: 18; 
2006: 19; 
2007: 21; 
MDUFA: 
2008: 21; 
2009: 32; 
2010[A]: 38; 
2011[A]: 20. 

Percentage reviewed in less than or equal to 180 days[B]; 
Pre-MDUFMA: 
2000: 67.1; 
2001: 56.8; 
2002: 52.9; 
MDUFMA: 
2003: 50.0; 
2004: 40.0; 
2005: 32.1; 
2006: 34.5; 
2007: 56.8; 
MDUFA: 
2008: 61.8; 
2009: 80.0; 
2010[A]: 86.4; 
2011[A]: 90.9. 

Tier 1 goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty];
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: 50; 
MDUFA: 
2008: 60; 
2009: 60; 
2010[A]: 60; 
2011[A]: 60. 

Met Tier 1 goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: n/a; 
2007: n/a; 
MDUFA: 
2008: Yes; 
2009: Yes; 
2010[A]: Yes[E]; 
2011[A]: Yes[E]. 

Number reviewed in less than or equal to 320 days[B]; 
Pre-MDUFMA: 
2000: 63; 
2001: 67; 
2002: 40; 
MDUFMA: 
2003: 47; 
2004: 43; 
2005: 50; 
2006: 45; 
2007: 33; 
MDUFA: 
2008: 27; 
2009: 35; 
2010[A]: 44; 
2011[A]: 22. 

Percentage reviewed in less than or equal to 320 days[B]; 
Pre-MDUFMA: 
2000: 90.0; 
2001: 90.5; 
2002: 78.4; 
MDUFMA: 
2003: 94.0; 
2004: 95.6; 
2005: 89.3; 
2006: 81.8; 
2007: 89.2; 
MDUFA: 
2008: 79.4; 
2009: 87.5; 
2010[A]: 100.0; 
2011[A]: 100.0. 

Tier 2 goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: 80; 
2007: 90; 
2008: [Empty]; 
2009: [Empty]; 
2010[A]: [Empty]; 
2011[A]: [Empty]. 

Met Tier 2 goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: Yes; 
2007: No; 
MDUFA: 
2008: n/a; 
2009: n/a; 
2010[A]: n/a; 
2011[A]: n/a. 

Number reviewed in less than or equal to 295 days[B]; 
Pre-MDUFMA: 
2000: 58; 
2001: 63; 
2002: 39; 
MDUFMA: 
2003: 43; 
2004: 40; 
2005: 41; 
2006: 34; 
2007: 29; 
MDUFA: 
2008: 27; 
2009: 35; 
2010[A]: 44; 
2011[A]: 22. 

Percentage reviewed in less than or equal to 295 days[B]; 
Pre-MDUFMA: 
2000: 82.9; 
2001: 85.1; 
2002: 76.5; 
MDUFMA: 
2003: 86.0; 
2004: 88.9; 
2005: 73.2; 
2006: 61.8; 
2007: 78.4; 
MDUFA: 
2008: 79.4; 
2009: 87.5; 
2010[A]: 100.0; 
2011[A]: 100.0. 

Tier 2 goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
MDUFA: 
2008: 90; 
2009: 90; 
2010[A]: 90; 
2011[A]: 90. 

Met Tier 2 goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: n/a; 
2007: n/a; 
MDUFA: 
2008: No; 
2009: No; 
2010[A]: Yes[E]; 
2011[A]: Yes[E]. 

Average number of review cycles per submission[F]; 
Pre-MDUFMA: 
2000: 1.77; 
2001: 2.05; 
2002: 2.16; 
MDUFMA: 
2003: 1.82; 
2004: 1.98; 
2005: 2.54; 
2006: 2.40; 
2007: 2.51; 
MDUFA: 
2008: 2.32; 
2009: 2.12[G]; 
2010[A]: 2.04; 
2011[A]: 1.58. 

Of first-cycle decisions, percentage that were approval; 
Pre-MDUFMA: 
2000: 15.1; 
2001: 17.3; 
2002: 9.8; 
MDUFMA: 
2003: 16.0; 
2004: 8.9; 
2005: 8.8; 
2006: 21.8; 
2007: 8.1; 
MDUFA: 
2008: 5.9; 
2009: 9.8; 
2010[A]: 7.4; 
2011[A]: 9.3. 

Among original PMAs with a final decision, percentage of final 
decisions that were: Approval[G]; 
Pre-MDUFMA: 
2000: 61.6; 
2001: 69.3; 
2002: 76.5; 
MDUFMA: 
2003: 74.0; 
2004: 79.5; 
2005: 71.4; 
2006: 90.4; 
2007: 70.6; 
MDUFA: 
2008: 68.8; 
2009: 56.3; 
2010[A]: 82.8; 
2011[A]: 93.3. 

Among original PMAs with a final decision, percentage of final 
decisions that were: Denial[G]; 
Pre-MDUFMA: 
2000: 0.0; 
2001: 0.0; 
2002: 0.0; 
MDUFMA: 
2003: 0.0; 
2004: 0.0; 
2005: 0.0; 
2006: 0.0; 
2007: 0.0; 
MDUFA: 
2008: 0.0; 
2009: 0.0; 
2010[A]: 0.0; 
2011[A]: 0.0. 

Among original PMAs with a final decision, percentage of final 
decisions that were: Withdrawal[G]; 
Pre-MDUFMA: 
2000: 31.5; 
2001: 28.0; 
2002: 23.5; 
MDUFMA: 
2003: 26.0; 
2004: 20.5; 
2005: 28.6; 
2006: 9.6; 
2007: 29.4; 
MDUFA: 
2008: 31.3; 
2009: 43.8; 
2010[A]: 17.2; 
2011[A]: 6.7. 

Average FDA review time (in days) for original PMAs that were not 
reviewed within 295 days[B]; 
Pre-MDUFMA: 
2000: 383; 
2001: 468; 
2002: 451; 
MDUFMA: 
2003: 333; 
2004: 368; 
2005: 342; 
2006: 447; 
2007: 422; 
MDUFA: 
2008: 584; 
2009: 571; 
2010[A]: [Empty][H]; 
2011[A]: [Empty][H]. 

Average time to final decision (in days) for original PMAs that were 
not reviewed within 295 days[G]; 
Pre-MDUFMA: 
2000: 752; 
2001: 1006; 
2002: 688; 
MDUFMA: 
2003: 591; 
2004: 775; 
2005: 727; 
2006: 679; 
2007: 645; 
MDUFA: 
2008: 923; 
2009: 823; 
2010[A]: [Empty][H]; 
2011[A]: [Empty][H]. 

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 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 [Empty]. 

[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. 

[End of table] 

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

Fiscal years: 

Total number of submissions; 
Pre-MDUFMA: 
2000: 11; 
2001: 10; 
2002: 12; 
MDUFMA: 
2003: 4; 
2004: 17; 
2005: 6; 
2006: 3; 
2007: 2; 
MDUFA: 
2008: 4; 
2009: 4; 
2010[A]: 6; 
2011[A]: 7. 

Number reviewed in less than or equal to 180 180 days[B]; 
Pre-MDUFMA: 
2000: 2; 
2001: 3; 
2002: 3; 
MDUFMA: 
2003: 2; 
2004: 8; 
2005: 1; 
2006: 0; 
2007: 0; 
MDUFA: 
2008: 1; 
2009: 2; 
2010[A]: 2; 
2011[A]: 1. 

Percentage reviewed in less than or equal to 180 days[B]; 
Pre-MDUFMA: 
2000: 18.2; 
2001: 30.0; 
2002: 25.0; 
MDUFMA: 
2003: 50.0; 
2004: 47.1; 
2005: 16.7; 
2006: 0.0; 
2007: 0.0; 
MDUFA: 
2008: 25.0; 
2009: 50.0; 
2010[A]: 40.0; 
2011[A]: 50.0. 

Tier 1 goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
MDUFA: 
2008: 50; 
2009: 50; 
2010[A]: 50; 
2011[A]: 50. 

Met Tier 1 goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: n/a; 
2007: n/a; 
MDUFA: 
2008: No; 
2009: Yes; 
2010[A]: No[E]; 
2011[A]: Yes[E]. 

Number reviewed in less than or equal to 300 days[B]; 
Pre-MDUFMA: 
2000: 8; 
2001: 6; 
2002: 11; 
MDUFMA: 
2003: 3; 
2004: 14; 
2005: 5; 
2006: 2; 
2007: 0; 
MDUFA: 
2008: 2; 
2009: 3; 
2010[A]: 5; 
2011[A]: 2. 

Percentage reviewed in less than or equal to 300 days[B]; 
Pre-MDUFMA: 
2000: 72.7; 
2001: 60.0; 
2002: 91.7; 
MDUFMA: 
2003: 75.0; 
MDUFMA: 
2004: 82.4; 
2005: 83.3; 
2006: 66.7; 
2007: 0.0; 
MDUFA: 
2008: 50.0; 
2009: 75.0; 
2010[A]: 100.0; 
2011[A]: 100.0. 

less than or equal to 300 days goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: 70; 
2006: 80; 
2007: 90; 
MDUFA: 
2008: [Empty]; 
2009: [Empty]; 
2010[A]: [Empty]; 
2011[A]: [Empty]. 

Met less than or equal to 300 days goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: Yes; 
2006: No; 
2007: No; 
MDUFA: 
2008: n/a; 
2009: n/a; 
2010[A]: n/a; 
2011[A]: n/a. 

Number reviewed in less than or equal to 280 days[B]; 
Pre-MDUFMA: 
2000: 7; 
2001: 6; 
2002: 10; 
MDUFMA: 
2003: 3; 
2004: 14; 
2005: 4; 
2006: 1; 
2007: 0; 
MDUFA: 
2008: 2; 
2009: 3; 
2010[A]: 5; 
2011[A]: 2. 

Percentage reviewed in less than or equal to 280 days[B]; 
Pre-MDUFMA: 
2000: 63.6; 
2001: 60.0; 
2002: 83.3; 
MDUFMA: 
2003: 75.0; 
2004: 82.4; 
2005: 66.7; 
2006: 33.3; 
2007: 0.0; 
MDUFA: 
2008: 50.0; 
2009: 75.0; 
2010[A]: 100.0; 
2011[A]: 100.0. 

Tier 2 goal percentage[C]; 
Pre-MDUFMA: 
2000: [Empty]; 
2001: [Empty]; 
2002: [Empty]; 
MDUFMA: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
MDUFA: 
2008: 90; 
2009: 90; 
2010[A]: 90; 
2011[A]: 90. 

Met Tier 2 goal[D]; 
Pre-MDUFMA: 
2000: n/a; 
2001: n/a; 
2002: n/a; 
MDUFMA: 
2003: n/a; 
2004: n/a; 
2005: n/a; 
2006: n/a; 
2007: n/a; 
MDUFA: 
2008: No; 
2009: No; 
2010[A]: Yes[E]; 
2011[A]: Yes[E]. 

Average number of review cycles per submission[F]; 
Pre-MDUFMA: 
2000: 2.45; 
2001: 2.60; 
2002: 2.08; 
MDUFMA: 
2003: 2.75; 
2004: 2.29; 
2005: 3.50; 
2006: 4.00; 
2007: 3.00; 
MDUFA: 
2008: 2.75; 
2009: 2.50; 
2010[A]: 2.00; 
2011[A]: 2.00. 

Of first-cycle decisions, percentage that were approval; 
Pre-MDUFMA: 
2000: 0.0; 
2001: 20.0; 
2002: 16.7; 
MDUFMA: 
2003: 0.0; 
2004: 17.6; 
2005: 16.7; 
2006: 0.0; 
2007: 0.0; 
MDUFA: 
2008: 25.0; 
2009: 0.0; 
2010[A]: 16.7; 
2011[A]: 14.3. 

Among expedited PMAs with a final decision, percentage of final 
decisions that were: Approval[ G]; 
2000: 81.8; 
2001: 90.0; 
2002: 75.0; 
MDUFMA: 
2003: 75.0; 
2004: 81.3; 
2005: 66.7; 
2006: 100.0; 
2007: 0.0; 
MDUFA: 
2008: 25.0; 
2009: 75.0; 
2010[A]: 75.0; 
2011[A]: 100.0. 

Among expedited PMAs with a final decision, percentage of final 
decisions that were: Denial[G]; 
Pre-MDUFMA: 
2000: 0.0; 
2001: 0.0; 
2002: 0.0; 
MDUFMA: 
2003: 0.0; 
2004: 0.0; 
2005: 0.0; 
2006: 0.0; 
2007: 0.0; 
MDUFA: 
2008: 25.0; 
2009: 0.0; 
2010[A]: 0.0; 
2011[A]: 0.0. 

Among expedited PMAs with a final decision, percentage of final 
decisions that were: Withdrawal[G]; 
Pre-MDUFMA: 
2000: 18.2; 
2001: 10.0; 
2002: 25.0; 
MDUFMA: 
2003: 25.0; 
2004: 18.8; 
2005: 33.3; 
2006: 0.0; 
2007: 100.0; 
MDUFA: 
2008: 50.0; 
2009: 25.0; 
2010[A]: 25.0; 
MDUFA: 
Average FDA review time (in days) for expedited PMAs that were not 
reviewed within 280 days[B]; 
Pre-MDUFMA: 
2000: 343; 
2001: 425; 
2002: 308; 
MDUFMA: 
2003: 447; 
2004: 483; 
2005: 427; 
2006: 511; 
2007: 344; 
MDUFA: 
2008: 489; 
2009: 414; 
2010[A]: [Empty] [H]; 
2011[A]: [Empty][H]. 

Average time to final decision (in days) for expedited PMAs that were 
not reviewed within 280 days[G]; 
Pre-MDUFMA: 
2000: 588; 
2001: 520; 
2002: 334; 
MDUFMA: 
2003: 1125; 
2004: 713; 
2005: 939; 
2006: 792; 
2007: 1111;
MDUFA: 
2008: 1017; 
2009: 483; 
2010[A]: [Empty][H]; 
2011[A]: [Empty][H]. 

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 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 [Empty]. 

[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. 

[End of table] 

[End of section] 

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

Center for Devices and Radiological Health (CDRH): 

FDA centers and offices: Office of the Center Director (OCD)[A]; 
Number of FTEs in each fiscal year: 
2003: 13; 
2004: 12; 
2005: 15; 
2006: 19; 
2007: 34; 
2008: 35; 
2009: 52; 
2010: 63. 

FDA centers and offices: Office of Management Operations (OSM/OMO); 
Number of FTEs in each fiscal year: 
2003: 89; 
2004: 62; 
2005: 64; 
2006: 62; 
2007: 61; 
2008: 50; 
2009: 48; 
2010: 52. 

FDA centers and offices: Office of Information Technology (OIT)[B]; 
Number of FTEs in each fiscal year: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
2008: 16; 
2009: 15; 
2010: 17. 

FDA centers and offices: Office of Compliance (OC); 
Number of FTEs in each fiscal year: 
2003: 40; 
2004: 46; 
2005: 54; 
2006: 55; 
2007: 56; 
2008: 61; 
2009: 60; 
2010: 60. 

FDA centers and offices: Office of Device Evaluation (ODE); 
Number of FTEs in each fiscal year: 
2003: 305; 
2004: 301; 
2005: 298; 
2006: 287; 
2007: 311; 
2008: 326; 
2009: 322; 
2010: 329. 

FDA centers and offices: Office of Science and Engineering 
Laboratories (OST/OSEL); 
Number of FTEs in each fiscal year: 
2003: 88; 
2004: 94; 
2005: 110; 
2006: 91; 
2007: 89; 
2008: 86; 
2009: 87; 
2010: 94. 

FDA centers and offices: Office of Communication Education and 
Radiation Programs (OHIP/OCER); 
Number of FTEs in each fiscal year: 
2003: 42; 
2004: 42; 
2005: 47; 
2006: 35; 
2007: 45; 
2008: 43; 
2009: 35; 
2010: 43. 

FDA centers and offices: Office of Surveillance and Biometrics (OSB); 
Number of FTEs in each fiscal year: 
2003: 86; 
2004: 92; 
2005: 104; 
2006: 106; 
2007: 98; 
2008: 105; 
2009: 117; 
2010: 139. 

FDA centers and offices: Office of In Vitro Diagnostics (OIVD)[C]; 
Number of FTEs in each fiscal year: 
2003: [Empty]; 
2004: 49; 
2005: 61; 
2006: 67; 
2007: 65; 
2008: 71; 
2009: 69; 
2010: 105. 

FDA centers and offices: Committee Conference Management (CCM)[D]; 
Number of FTEs in each fiscal year: 
2003: [Empty]; 
2004: [Empty]; 
2005: [Empty]; 
2006: [Empty]; 
2007: [Empty]; 
2008: 1; 
2009: 1; 
2010: 1. 

FDA centers and offices: CDRH Total; 
Number of FTEs in each fiscal year: 
2003: 662; 
2004: 697; 
2005: 753; 
2006: 721; 
2007: 760; 
2008: 793; 
2009: 805; 
2010: 902. 

Center for Biologics Evaluation and Research (CBER): 

FDA centers and offices: Center Director's Office, Office of 
Management (OM), Office of Information Management (OIM), and Office of 
Communication, Outreach, and Development (OCOD); 
Number of FTEs in each fiscal year: 
2003: 13; 
2004: 15; 
2005: 20; 
2006: 23; 
2007: 23; 
2008: 23; 
2009: 24; 
2010: 28. 

FDA centers and offices: Office of Blood Research & Review; 
Number of FTEs in each fiscal year: 
2003: 37; 
2004: 43; 
2005: 49; 
2006: 70; 
2007: 66; 
2008: 66; 
2009: 64; 
2010: 64. 

FDA centers and offices: Office of Cellular, Tissue & Gene Therapies; 
Number of FTEs in each fiscal year: 
2003: 2; 
2004: 2; 
2005: 2; 
2006: 2; 
2007: 2; 
2008: 5; 
2009: 6; 
2010: 6. 

FDA centers and offices: Office of Vaccines Research & Review; 
Number of FTEs in each fiscal year: 
2003: 1; 
2004: 0; 
2005: 0; 
2006: 0; 
2007: 1; 
2008: 1; 
2009: 1; 
2010: 4. 

FDA centers and offices: Office of Therapeutics Research & Review; 
Number of FTEs in each fiscal year: 
2003: 1; 
2004: 0; 
2005: 0; 
2006: 0; 
2007: 0; 
2008: 0; 
2009: 0; 
2010: 0. 

FDA centers and offices: Office of Biostatistics & Epidemiology; 
Number of FTEs in each fiscal year: 
2003: 1; 
2004: 2; 
2005: 2; 
2006: 2; 
2007: 2; 
2008: 2; 
2009: 4; 
2010: 5. 

FDA centers and offices: Office of Compliance & Biologics Quality; 
Number of FTEs in each fiscal year: 
2003: 5; 
2004: 6; 
2005: 9; 
2006: 6; 
2007: 8; 
2008: 6; 
2009: 4; 
2010: 7. 

FDA centers and offices: CBER Total; 
Number of FTEs in each fiscal year: 
2003: 59; 
2004: 68; 
2005: 81; 
2006: 104; 
2007: 101; 
2008: 102; 
2009: 103; 
2010: 114. 

Office of Regulatory Affairs (ORA): 

FDA centers and offices: ORA Total; 
Number of FTEs in each fiscal year: 
2003: 59; 
2004: 60; 
2005: 62; 
2006: 64; 
2007: 64; 
2008: 62; 
2009: 57; 
2010: 69. 

Office of the Commissioner (OC): 

FDA centers and offices: OC Total; 

Number of FTEs in each fiscal year: 
2003: 77; 
2004: 70; 
2005: 77; 
2006: 78; 
2007: 90; 
2008: 74; 
2009: 84; 
2010: 86. 

Shared Service (SS)[E]: 

FDA centers and offices: SS Total; 
Number of FTEs in each fiscal year: 
2003: [Empty]; 
2004: 20; 
2005: 60; 
2006: 53; 
2007: 57; 
2008: 64; 
2009: 60; 
2010: 59. 

All Centers and Offices Total; 
Number of FTEs in each fiscal year: 
2003: 857; 
2004: 915; 
2005: 1,034; 
2006: 1,020; 
2007: 1,071; 
2008: 1,095; 
2009: 1,109; 
2010: 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. 

[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. 

[End of table] 

[End of section] 

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

Department Of Health And Human Services: 
Office Of The Secretary: 
Assistant Secretary for Legislation: 
Washington, DC 20201: 

February 22 2012: 

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

Dear Ms. Crosse: 

Attached are comments on the U.S. Government Accountability Office's 
(GAO) report entitled, "Medical Devices: FDA Has Met Most Performance 
Goals but Device Reviews Are Taking Longer" (GAO-12-418). 

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

Sincerely, 

Signed by: 

Jim R. Esquea: 
Assistant Secretary for Legislation: 

Attachment: 

[End of letter] 

General Comments Of The Department Of Health And Human Services (HHS) 
On The Government Accountability Office's (GAO) Draft Report Entitled, 
"Medical Devices: FDA Has Met Most Performance Goals But Device 
Reviews Are Taking Longer" (GAO-12-418): 

The Department appreciates the opportunity to review and comment on 
this draft report. As GAO notes, the Food and Drug Administration's 
(FDA) own annual reports to Congress[Footnote 1] regarding medical 
device review times reveal the same patterns that GAO has found in 
this report. FDA has continued to make progress during FY 2011 while 
implementing MDUFA II. Overall, FDA has already met or exceeded, or 
has the potential to meet or exceed, based on preliminary data of 
completed and pending reviews, 17 of 21 Tier 1 performance goals and 
15 of 21 Tier 2 performance goals for the FY 2008 through FY 2011 
performance goal cohorts. 

In addition, GAO noted, and FDA also has identified in it own reports, 
that the total time to a decision (FDA review time plus the time it 
takes a sponsor to provide requested information) has increased, yet 
FDA's performance in meeting its goals is by and large strong. Since 
the total time to a decision includes the time industry incurs in 
responding to FDA's concerns, FDA and industry bear shared 
responsibility for the time increase and will need to work together to 
improve performance on total time to decision. 

FDA recently conducted an analysis of premarket review times under the 
510(k) program and identified quality issues in more than 50 percent 
of 510(k) applications. These quality issues require agency staff to 
prepare and issue additional information letters, resulting in 
additional review cycles. At the same time, FDA has been instituting 
management changes to improve its premarket review programs, which the 
agency expects will help address the increase in total time to 
decision. FDA has conducted a thorough assessment of the 510(k) review 
program and an assessment of how CDRH uses science in regulatory 
decision-making. FDA issued two reports in August 2010 that found that 
FDA needed to take several critical actions to improve the 
predictability, consistency, and transparency of both the 510(k) and 
premarket approval application review programs. 

The agency solicited public comment on the recommendations identified 
in the studies and received a range of perspectives from stakeholders, 
including medical device companies, industry representatives, venture 
capitalists, health care professional organizations, third-party 
payers, patient and consumer advocacy groups, foreign regulatory 
bodies, and others. After considering the public input, in January 
2011, FDA announced 25 specific actions[Footnote 2] that the agency 
would take to improve the predictability, consistency, and 
transparency of its premarket review programs. Subsequently, FDA has 
taken or is taking actions 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. 

These actions exemplify FDA's commitment to increasing the timely 
availability of safe and effective new medical devices to patients and 
healthcare providers. 

Finally, FDA needs adequate, stable funding to manage a device program 
that can approve or clear safe and effective products for patients 
without delay, and only timely reauthorization of a medical device 
user fee program that both FDA and the medical device industry support 
can assure such funding. 

Footnotes: 

[1] [hyperlink, 
http://www.fda.gov/AboutFDA/ReportsManualsForms/Reports/UserFeeReports/P
erformanceReports/MDUFMA/default.htm]. 

[2] [hyperlink, 
http://www.fda.gov/downloads/AboutFDA/CentersOffices/CDRH/CDRHReports/UC
M239450.pdf]. 

[End of section] 

Appendix IV: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Marcia Crosse, (202) 512-7114 or crossem@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Robert Copeland, Assistant 
Director; Carolyn Fitzgerald; Cathleen Hamann; Karen Howard; Hannah 
Marston Minter; Lisa Motley; Aubrey Naffis; Michael Rose; and Rachel 
Schulman made key contributions to this report. 

[End of section] 

Footnotes: 

[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. 

[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. 

[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. 

[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. 

[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, [hyperlink, 
http://www.gao.gov/products/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). 

[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. 

[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. 

[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. 

[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. 

[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). 

[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. 

[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. 

[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. 

[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. 

[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. 

[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. 

[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. 

[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. 

[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). 

[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). 

[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. 

[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, Hearing Before the Special Committee on Aging, 
United States Senate, 112TH 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, [hyperlink, 
http://www.gao.gov/products/GAO-09-190] (Washington, D.C.: Jan. 15, 
2009). 

[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. 

[End of section] 

GAO’s Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

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

Order by Phone: 

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

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

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

Connect with GAO: 

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

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

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

Congressional Relations: 

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

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