This is the accessible text file for GAO report number GAO-15-202 
entitled 'Drug Shortages: Better Management of the Quota Process for 
Controlled Substances Needed; Coordination between DEA and FDA Should 
Be Improved' which was released on March 4, 2015. 

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 2015: 

Drug Shortages: 

Better Management of the Quota Process for Controlled Substances 
Needed; Coordination between DEA and FDA Should Be Improved: 

GAO-15-202: 

GAO Highlights: 

Highlights of GAO-15-202, a report to congressional requesters. 

Why GAO Did This Study: 

In the last decade, shortages of prescription drugs containing 
controlled substances, such as narcotics and stimulants, have 
increased nationwide, preventing providers and patients from accessing 
essential medications for treatment. Controlled substances are 
regulated by DEA because of the potential for abuse and addiction. To 
prevent diversion of controlled substances, DEA sets quotas that limit 
the amount of certain substances that are available in the United 
States. 

GAO was asked to examine shortages of drugs containing controlled 
substances. This report examines (1) the trends in such shortages, (2) 
the effect on patients and providers, (3) DEA's administration of the 
quota process, and (4) coordination between DEA and FDA to prevent and 
mitigate shortages. GAO analyzed data from 2001 through 2013 from the 
University of Utah Drug Information Service, which is generally 
regarded as the most comprehensive source of drug shortage data, and 
from 2011 and 2012 from YERS/QMS, which is the official record for the 
quota process. GAO interviewed officials from DEA, FDA, organizations 
representing patients and providers, and drug manufacturers. GAO 
reviewed relevant statutes, regulations, and documents. 

What GAO Found: 

Shortages of prescription drugs containing controlled substances have 
increased sharply in recent years; of the 168 shortages reported from 
January 2001 through June 2013, nearly 70 percent began after 2007. 
Such shortages lasted for nearly a year, on average. Additionally, 
many shortages involved generic pain relievers and drugs where there 
was only one manufacturer. 

The Food and Drug Administration (FDA), an agency within the 
Department of Health and Human Services (HHS), and organizations 
representing patients and providers report that during shortages of 
drugs containing controlled substances, patients may receive less 
effective care, experience medication errors, or not receive treatment 
at all. They said providers are also affected as they spend time and 
resources mitigating the effects of shortages, rather than providing 
care. 

The Drug Enforcement Administration (DEA), an agency within the 
Department of Justice (DOJ), has not effectively administered the 
quota process that limits the amount of controlled substances 
available for use in the United States. Each year, manufacturers apply 
to DEA for quota needed to make their drugs. DEA, however, has not 
responded to them within the time frames required by its regulations 
for any year from 2001 through 2014. DEA officials attributed this 
lack of compliance to inadequate staffing. Manufacturers who reported 
quota-related shortages cited late quota decisions as causing or 
exacerbating shortages of their drugs. Additionally, DEA's weak 
internal controls jeopardize the agency's ability to effectively 
manage the quota process. For instance, agency officials said that DEA 
does not conduct quality checks to ensure the accuracy of the data in 
its Year-End Reporting and Quota Management System (YERS/QMS). GAO 
estimates that 44 percent of YERS/QMS records in 2011 and 10 percent 
in 2012 had errors. DEA officials said that 2011 was the first year 
manufacturers applied for quota electronically and they expected data 
from 2012 and beyond to be more accurate. DEA also lacks critical 
management information because it does not have performance measures 
related to setting quotas, nor does it monitor data to assess its 
performance. Moreover, DEA does not have reasonable assurance that the 
quotas it sets are in accordance with its requirements and cannot 
ensure continuity of its operations, as it does not have protocols, 
policies, training materials, or other documentation to manage the 
quota process. 

Despite statutory provisions requiring DEA and FDA to coordinate 
certain efforts to address shortages of drugs containing controlled 
substances, the agencies have not established a sufficiently 
collaborative relationship. For example, DEA and FDA disagree about 
what constitutes a shortage. DEA officials also said that they do not 
believe FDA appropriately validates or investigates the shortage 
information it posts on its website and that posting this information 
encourages manufacturers to falsely report shortages to obtain 
additional quota. However, FDA reports that it takes steps to 
investigate and confirm the shortages on its website. Given such 
barriers to coordination, DEA and FDA cannot effectively act to 
prevent or alleviate shortages. Although DEA and FDA have a memorandum 
of understanding (MOU) in place, it has not been revised since the 
1970s and they have been working for more than two years to update it. 
Officials from both agencies said an updated MOU could facilitate 
information sharing and help prevent and mitigate future shortages of 
drugs containing controlled substances. 

What GAO Recommends: 

DEA should take five actions to improve its management of the quota 
process; DEA and FDA should quickly update their MOU and agree on 
steps each will take regarding drug shortages. HHS agreed with the 
applicable recommendations. DEA neither agreed nor disagreed, but 
raised multiple objections to this report. 

View [hyperlink, http://www.gao.gov/products/GAO-15-202]. For more 
information, contact Marcia Crosse at (202) 512-7114 or 
crossem@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Shortages of Drugs Containing Controlled Substances Rose Sharply After 
2007; Many Involved Generic Pain Relievers and Drugs Manufactured by 
Few Companies: 

Shortages of Drugs Containing Controlled Substances Can Compromise 
Patient Care and Create Challenges for Providers: 

DEA Has Not Effectively Administered the Quota Process Due to Missed 
Time Frames and Weak Internal Controls: 

DEA and FDA Have Not Established a Sufficiently Collaborative 
Relationship, Which Could Hinder Their Abilities to Effectively 
Coordinate and Address Future Shortages: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: List of Provider and Patient Organizations Interviewed: 

Appendix II: Classes of Schedule II Controlled Substances with Medical 
Use: 

Appendix III: Scope and Methodology - Analysis of the Year-End 
Reporting and Quota Management System Data: 

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

Appendix V: Comments from the Drug Enforcement Administration: 

Appendix VI: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: Summary of the Duration of Shortages of Drugs Containing 
Controlled Substances, January 2001 through June 2013: 

Table 2: Summary of Critical Shortages of Drugs Containing Controlled 
Substances by Therapeutic Class, January 2001 through June 2013: 

Table 3: The Drug Enforcement Administration's (DEA) Estimated 
Response Times for Supplemental Applications in Year-End Reporting and 
Quota Management System (YERS/QMS) Source Documents, 2011 and 2012: 

Table 4: Estimated Percent of Errors in the Drug Enforcement 
Administration's (DEA) Year-End Reporting and Quota Management System 
(YERS/QMS): 

Table 5: Records Sampled from the Drug Enforcement Administration's 
Year-End Reporting and Quota Management System (YERS/QMS) for Schedule 
II Substances with Medical Use: 

Figures: 

Figure 1: Timeline for Manufacturers to Apply for Quotas and for the 
Drug Enforcement Administration (DEA) to Propose or Establish Quotas: 

Figure 2: New Shortages of Drugs Containing Controlled Substances and 
All Drugs Reported January 2001 through June 2013, by Year: 

Figure 3: Number of Active Shortages of Drugs Containing Controlled 
Substances, January 2001 through June 2013, by Year: 

Figure 4: Distribution of Critical Shortages of Drugs Containing 
Controlled Substances by Therapeutic Class and Product Type, January 
2001 through June 2013: 

Abbreviations: 

ADHD: attention deficit hyperactivity disorder: 

API: active pharmaceutical ingredient: 

APQ: aggregate production quota: 

ARCOS: Automation of Reports and Consolidated Orders System: 

CSA: Controlled Substances Act: 

DEA: Drug Enforcement Administration: 

DOJ: Department of Justice: 

DSS: Drug Shortage Staff: 

EMS: emergency medical services: 

FDA: Food and Drug Administration: 

FDASIA: Food and Drug Administration Safety and Innovation Act: 

GPRA: Government Performance and Results Act of 1993: 

HHS: Department of Health and Human Services: 

MOU: memorandum of understanding: 

UUDIS: University of Utah Drug Information Service: 

YERS/QMS: Year-End Reporting and Quota Management System: 

[End of section] 

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

February 2, 2015: 

The Honorable Charles E. Grassley: 
Chairman: 
Committee on the Judiciary: 
United States Senate: 

The Honorable Sheldon Whitehouse: 
Ranking Member: 
Subcommittee on Crime and Terrorism: 
Committee on the Judiciary: 
United States Senate: 

In the last decade, shortages of prescription drugs have increased 
nationwide, preventing providers and patients from accessing 
medications that are essential for treatment. Some shortages involve 
drugs that contain controlled substances, such as narcotics, 
stimulants, and sedatives, which play an important role in health 
care.[Footnote 1] For example, anesthesia, which is used to sedate 
patients, is routinely administered prior to surgical procedures and 
most patients receive some kind of drug containing a controlled 
substance during inpatient hospital stays. Emergency medical service 
(EMS) providers also frequently rely on a variety of drugs containing 
controlled substances to treat patients experiencing heart attacks, 
seizures, traumatic injuries, and other medical crises. For example, 
data from the National EMS Information System, a nationwide repository 
of information collected by EMS organizations, show that fentanyl and 
morphine--both narcotics that are used to manage pain--were 2 of the 
10 most frequently administered medications by such organizations in 
2012. In addition, some drugs containing controlled substances are 
available by prescription for outpatient use as pain relievers or 
sleep aids, as well as to treat individuals with attention deficit 
hyperactivity disorder (ADHD) and anxiety. During shortages, 
providers--including hospitals, physicians, and pharmacists--may have 
to use alternative medications, if alternatives are available at all. 
The Food and Drug Administration (FDA), an agency within the 
Department of Health and Human Services (HHS), which oversees the 
safety and effectiveness of drugs marketed in the United States, 
considers drug shortages to be a serious public health concern and 
works to prevent, alleviate, and resolve them.[Footnote 2] 

Despite their medical value, prescription drugs containing these 
substances have the potential for abuse and psychological and physical 
dependence. They are therefore required to be regulated by the Drug 
Enforcement Administration (DEA), an agency within the Department of 
Justice (DOJ), in accordance with the Controlled Substances Act (CSA). 
The Centers for Disease Control and Prevention has declared that the 
United States is in the midst of an epidemic of prescription drug 
overdose deaths. In 2011, more than 22,000 Americans died from drug 
overdoses attributable to prescription drugs, and most of those deaths-
-almost 17,000--were due to prescription pain relievers containing 
controlled substances. To prevent diversion of controlled substances, 
DEA maintains a closed system of distribution, which includes limiting 
the amount of certain controlled substances that are available. To do 
so, DEA establishes quotas for the maximum amount of each basic class 
of controlled substance--such as amphetamine or morphine--that can be 
produced each year in the United States, known as aggregate production 
quotas (APQ). It also establishes quotas for individual manufacturers, 
who must apply to DEA to obtain quota for specific classes of 
controlled substances. In setting quotas, DEA is required to provide 
for the estimated medical, scientific, research, and industrial needs 
of the United States.[Footnote 3] 

You raised questions about shortages of drugs containing controlled 
substances and asked us to examine the effect of such shortages on 
patients. You also asked us to review DEA's administration of the 
quota process to understand its potential effects on shortages of 
drugs containing controlled substances, particularly those used by EMS 
providers and to treat ADHD. This report (1) identifies the trends in 
shortages of drugs containing controlled substances that occurred from 
January 2001 through June 2013, (2) describes the reported effects of 
shortages of drugs containing controlled substances on patients and 
providers, (3) examines DEA's administration of the quota process for 
controlled substances, and (4) examines DEA and FDA coordination 
activities to prevent and mitigate shortages of drugs containing 
controlled substances. 

To identify the trends in shortages of drugs containing controlled 
substances from January 2001 through June 2013, we analyzed University 
of Utah Drug Information Service (UUDIS) data during this time 
period.[Footnote 4] UUDIS data is generally regarded as the most 
comprehensive and reliable information on drug shortages for the time 
period we reviewed and are what we used in preparing our 2011 and 2014 
reports on drug shortages.[Footnote 5] For this report we used an 
extract of the UUDIS dataset used to prepare our 2011 and 2014 
reports, which covered shortages from 2001--the first year available 
from UUDIS--through June 2013. We also examined the characteristics of 
drugs containing controlled substances that were in shortage at some 
point from January 2001 through June 2013 and that UUDIS identified as 
critical shortages because alternative medications were unavailable, 
the shortages affected multiple manufacturers, or the shortages were 
widely reported. These critical shortages were a subset of the total 
number of shortages reported during this time period.[Footnote 6] For 
this subset of shortages, we obtained drug shortage bulletins created 
by UUDIS, which contain national drug codes associated with each 
shortage.[Footnote 7] Using these national drug codes, we analyzed Red 
Book data to determine the product types, routes of administration, 
and therapeutic classes.[Footnote 8] We reviewed all UUDIS data for 
reasonableness, outliers, and consistency, and based on our review we 
determined that the data were sufficiently reliable for our purposes. 
Additionally, to give context to and inform these analyses we 
interviewed officials at the FDA and UUDIS. For this objective as well 
as our remaining objectives, the scope of our work was limited to 
prescription drugs, as opposed to over-the-counter or behind-the-
counter medications. 

To describe the reported effects of shortages of drugs containing 
controlled substances on patients and providers, we interviewed 
officials from provider and patient organizations, such as the 
American Society of Anesthesiologists and Children and Adults with 
Attention Deficit/Hyperactivity Disorder. We selected provider and 
patient organizations that represent populations we determined were 
likely to be affected by shortages of drugs containing controlled 
substances. The information that we obtained from these organizations 
is not generalizable to all provider and patient organizations, but 
provided us with valuable insights. For a complete list of patient and 
provider organizations we interviewed, see appendix I. We also 
interviewed officials from FDA, the Office of the Assistant Secretary 
for Preparedness and Response within HHS, and the National Highway 
Traffic Safety Administration's Office of Emergency Medical Services. 
We reviewed relevant documentation from these organizations and 
agencies, including the results of surveys they had conducted of their 
constituencies. 

To examine DEA's administration of the quota process for controlled 
substances, we reviewed the CSA and DEA's regulations and other 
documentation regarding the agency's process for establishing quotas 
for certain classes of controlled substances with medical use. 
[Footnote 9] See appendix II for the specific substance classes 
subject to quotas that we considered to have medical use. We also 
interviewed DEA officials about the quota process, including factors 
that affect the agency's timeliness and responsiveness to 
manufacturers' quota applications. In addition, we reviewed DEA's 
administration of the quota process in light of relevant federal 
standards for internal controls and our prior work related to the 
establishment of agency performance measures.[Footnote 10] To examine 
the timeliness of DEA's quota decisions for certain classes of 
controlled substances, we reviewed Federal Register notices for APQs 
for years 2001 through 2014. From our review of Federal Register 
notices, we were also able to determine when DEA established annual 
quotas for individual manufacturers from 2001 through 2014, as these 
types of quota are established after the APQ has been established in 
the Federal Register. We also analyzed data from the system DEA uses 
to track and record quota applications and decisions--the Year-End 
Reporting and Quota Management System (YERS/QMS). We analyzed data for 
certain controlled substances with medical use from 2011 and 2012, the 
most recent years for which data were available when we began our 
work. We analyzed the source documents for a stratified random sample 
of 440 YERS/QMS records from 2011 and 2012 (15 percent of the total 
records during these years for certain controlled substances with 
medical use). We used these source documents to verify YERS/QMS data, 
each of which includes a manufacturer's quota application for a 
particular substance and the corresponding decision letter from DEA 
informing the manufacturer of its quota authorization. We compared the 
information in these documents with corresponding YERS/QMS records and 
found many instances of discrepancies between them. We therefore 
determined the data from YERS/QMS to be unreliable for the purposes of 
our review, and so we instead are reporting on our analysis of the 
information contained in the source documents for our sample of 
YERS/QMS records to determine DEA's timeliness in responding to 
manufacturers' quota applications. See appendix III for additional 
information on the sample design and methodology we used to determine 
the reliability of the data and to report on the source documents. We 
then compared the timing of DEA's quota decisions, both for the APQs 
and manufacturers' quotas, to required deadlines in the CSA and DEA's 
regulations. In addition, we interviewed manufacturers about their 
experiences with DEA's quota process, including any effect the quota 
process may have had on shortages of their products. We also reviewed 
documentation provided by manufacturers about their quota applications 
and DEA's responses. Specifically, we spoke with 13 dosage form 
manufacturers, including 7 that had the greatest number of drugs 
containing controlled substances subject to quota in shortage from 
January 2010 through June 2013.[Footnote 11] We also spoke with 2 bulk 
manufacturers that supplied active pharmaceutical ingredients (API) 
for some of the drugs containing controlled substances that were in 
shortage.[Footnote 12] We identified these manufacturers through our 
analysis of FDA data on the manufacturer-reported causes of shortages 
of drugs containing controlled substances from January 2010 through 
June 2013, the most recent years for which data were available when we 
began our analysis. We reviewed FDA's data for reasonableness, 
outliers, and consistency, and based on our review, determined that 
the data were sufficiently reliable for our purposes. Lastly, we 
interviewed DEA and FDA officials about the causes of shortages of 
drugs containing controlled substances. 

To examine DEA and FDA coordination activities to prevent and mitigate 
shortages of drugs containing controlled substances, we reviewed the 
relevant provisions of the Food and Drug Administration Safety and 
Innovation Act (FDASIA) as well as other documentation governing how 
the two agencies are to coordinate.[Footnote 13] We interviewed DEA 
and FDA officials about their coordination activities. In addition, we 
reviewed our past work related to interagency collaboration and the 
necessary elements for a collaborative working relationship.[Footnote 
14] 

We conducted this performance audit from October 2012 through February 
2015 in accordance with generally accepted government auditing 
standards.[Footnote 15] 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 the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. 

Background: 

DEA's Oversight of Controlled Substances: 

Within DEA, the Office of Diversion Control is responsible for 
preventing, detecting, and investigating the diversion of controlled 
substances. The mission of this Office is dually focused on preventing 
diversion while ensuring an adequate and uninterrupted supply of these 
substances for legitimate needs. The CSA requires DEA to maintain a 
closed system of distribution of controlled substances in the United 
States from the point of import and manufacture through dispensing to 
patients or disposal. DEA funds diversion control activities, 
including the quota process, through fees that it sets for businesses, 
individuals, or entities that are required to register with DEA to 
import, export, manufacture, distribute, dispense, or conduct research 
on controlled substances.[Footnote 16] The CSA requires that DEA set 
registration fees at a level that ensures DEA recovers the full costs 
of operating its Diversion Control Program.[Footnote 17] DEA adjusts 
registration fees through the notice and comment rulemaking process 
and last adjusted registration fees in March 2012.[Footnote 18] 

The CSA places each controlled substance in one of five schedules 
based on whether the substance has a currently accepted medical use in 
treatment in the United States, its relative potential for abuse, and 
the degree of dependence the drug or other substance may 
cause.[Footnote 19] Schedule I controlled substances, such as heroin 
and LSD, have a high potential for abuse and no currently accepted 
medical use in treatment in the United States.[Footnote 20] Schedule 
II controlled substances have a high potential for abuse, which may 
lead to severe psychological or physical dependence, but also have a 
currently accepted medical use; for instance, amphetamine is a 
schedule II controlled substance and is used to treat ADHD. As part of 
the closed system of distribution to prevent the diversion of 
controlled substances, the CSA requires DEA to establish quotas for 
each basic class of schedule I and II controlled substances.[Footnote 
21] Quotas are not established for schedule III, IV, and V controlled 
substances, which have a currently accepted medical use, a lower 
potential for abuse, and a lower physical and psychological dependence 
relative to schedule I and II controlled substances.[Footnote 22] 

DEA's Process for Establishing Quotas for Certain Controlled 
Substances: 

Each year the Office of Diversion Control establishes three types of 
quotas for schedule II substances: APQs, bulk manufacturing quotas, 
and procurement quotas.[Footnote 23] Within the Office of Diversion 
Control, the Quota Unit is responsible for calculating and proposing 
quotas that are then established by the agency.[Footnote 24] 

Aggregate production quotas, which are established for each basic 
class of schedule I and II substances--such as amphetamine or 
morphine--specify the maximum amount of each controlled substance that 
can be manufactured in the United States in a given year to provide 
for the estimated medical, scientific, research, and industrial needs 
of the United States, lawful export requirements, and the 
establishment and maintenance of reserve stocks. APQs limit the amount 
of bulk materials that may be manufactured or synthesized in the 
United States, and subsequently available for use in the manufacture 
of drug products containing schedule I and II controlled substances. 
In establishing APQs for each basic class of schedule I and II 
controlled substances, DEA considers information from many sources, 
including: 

* manufacturers' production history and anticipated needs; 

* estimates from IMS Health on retail consumption based on 
prescriptions dispensed;[Footnote 25] 

* data from DEA's internal system for tracking controlled substances 
transactions, known as the Automation of Reports and Consolidated 
Orders System (ARCOS); 

* past histories of quota granted for each substance from YERS/QMS; 

* estimates of the projected medical, scientific, and reserve stock 
needs provided by FDA's Controlled Substances Staff;[Footnote 26] 

* information regarding new and discontinued drug products containing 
schedule II substances from FDA; and: 

* data on the diversion of controlled substances, such as information 
from case seizures and national databases of drug evidence. 

By May 1 of each year, DEA is required to propose APQs in the Federal 
Register for the next year and mail a copy of its proposal to persons 
registered as bulk manufacturers of the basic class.[Footnote 27] Any 
interested person may file comments on or objections to the notice, 
and, after consideration of such comments or objections, DEA must then 
establish APQs in the Federal Register and mail a copy of its 
decisions to manufacturers of the basic class.[Footnote 28] DEA may 
increase or decrease the APQ at any point due to fluctuations in 
demand and other factors, such as changes in accepted medical use of a 
substance for treatment or availability of raw materials for use in 
manufacturing.[Footnote 29] DEA monitors the various data sources used 
to establish the APQ, as well as an FDA website containing information 
on drug shortages, to identify any possible changes in supply. 
[Footnote 30] Generally, the agency revises the APQ midway through the 
year to account for any of these changes, as well as to account for 
any changes in anticipated need communicated by quota applicants. The 
process for revising the APQ is similar to the one used to establish 
it. Beginning with 2013 APQs, DEA established an additional 25 percent 
reserve in APQs in the event of natural disaster or other unforeseen 
event that could result in substantial disruption to the amount of 
basic classes of schedule II substances available for manufacture. 

Bulk manufacturing quotas limit the amount of a basic class of 
schedule I or II controlled substance that an individual bulk 
manufacturer can manufacture through the extraction or synthesis of 
plant material or other controlled substances. Each bulk manufacturer 
must apply to DEA to obtain bulk manufacturing quota for a specific 
substance class and the bulk manufacturing quotas granted to 
manufacturers in sum cannot exceed the APQs established by DEA. 

Procurement quotas limit the amount of a basic class of schedule I or 
II controlled substance that an individual manufacturer can procure 
from a manufacturer of bulk raw materials in order to manufacture into 
dosage forms of a drug or into other substances.[Footnote 31] The sum 
of procurement quotas determines the amount of bulk material that 
needs to be produced. 

DEA does not have the authority to issue quotas for specific products 
or to require manufacturers to use their quota for specific products, 
but rather establishes the quotas for the basic class of controlled 
substance. For example, DEA may grant a quota of 10,000 grams of 
amphetamine to a manufacturer, but it is up to the manufacturer to 
determine how much of that quota will be put towards the various 
strengths and formulations of its drug products. 

In establishing bulk manufacturing and procurement quotas for 
individual manufacturers, DEA considers the same types of information 
it uses to establish APQs. However, the agency's focus in reviewing 
the information is relative to the requesting manufacturer for that 
particular substance, rather than the total annual need, as is done 
for the APQ. 

The CSA and DEA's implementing regulations designate specific dates by 
which manufacturers must apply for quotas, as well as dates by which 
DEA must establish quotas (see figure 1). Manufacturers apply for 
quotas in the current year, which we term the application year. Their 
application is for use of the quota in the next calendar year, which 
we term the production year. Manufacturers may request a revision to 
their bulk manufacturing or procurement quota at any point.[Footnote 
32] When manufacturers apply for a revision to their quota during the 
production year, we term this a supplemental quota application. 

Figure 1: Timeline for Manufacturers to Apply for Quotas and for the 
Drug Enforcement Administration (DEA) to Propose or Establish Quotas: 

[Refer to PDF for image: timeline] 

Application year (Current calendar year) (e.g. 2011): 

On or before April 1st: 
DEA's regulations require manufacturers to submit procurement quota 
applications to DEA for the production year. 

On or before May 1st: 
DEA's regulations require manufacturers to submit bulk manufacturing 
quota applications to DEA for the production year. 

On or before May 1st: 
DEA's regulations require the agency to propose aggregate production 
quotas (APQ) for the production year in the Federal Register. 

No required date: 
After consideration of comments, DEA is to establish APQs for the 
production year. 

At any point: 
DEA can increase or decrease APQs. 

On or before July 1st: 
DEA's regulations require the agency to establish bulk manufacturing 
and procurement quotas for manufacturers for the production year[A]. 

Production year (Next calendar year) (e.g. 2012): 

On or after January 1st: 
Manufacturers can use procurement and bulk manufacturing quotas 
established for that production year. 

Source: GAO analysis of the Controlled Substances Act (CSA) and DEA's 
regulations. GAO-15-202. 

Notes: Unused quota from one year cannot be carried over to the 
following production year. For example, any remaining quota authorized 
for 2011 cannot be used to obtain controlled substances for 
procurement quotas and cannot be used to manufacture controlled 
substances for bulk manufacturing quotas in 2012. A bulk manufacturer, 
however, may extract or synthesize active pharmaceutical ingredient in 
one year and hold it in inventory until any subsequent year. 

Pursuant to section 1005 of the Food and Drug Administration Safety 
and Innovation Act, DEA is to review quota requests from manufacturers 
that relate to shortages of drugs containing schedule II controlled 
substances and respond within 30 days. Pub. L. No. 112-144, 126 Stat. 
993, 1105 (2012) (codified at 21 U.S.C. § 826(h)). Manufacturers can 
apply to DEA for more quota at any point, but there is no required 
time frame for DEA to respond to such supplemental quota requests 
unless a request relates to a drug in shortage. 

[A] In contrast to DEA's regulations, the CSA requires DEA to issue 
bulk manufacturing quotas to manufacturers on or before October 1 for 
the production year. 21 U.S.C. § 826(c). 

[End of figure] 

DEA launched an electronic system, known as YERS/QMS, in 2008 for 
manufacturers to submit end of year reports to the agency. Starting in 
2011, manufacturers were able to use YERS/QMS to submit quota 
applications to the agency electronically.[Footnote 33] YERS/QMS 
serves as the official record for the quota process. The system allows 
DEA to track manufacturers' quota applications--including the date of 
the quota application, the amount of quota requested, the amount of 
quota previously authorized by DEA in prior years, and information 
about the products that will be manufactured with the requested quota--
throughout its review process within the Office of Diversion Control. 
Additionally, YERS/QMS records DEA's decision about how much quota to 
authorize and the date on which DEA sends a decision letter to the 
manufacturer notifying it of the agency's quota decision. 

FDA's Role in Preventing and Resolving Drug Shortages and Coordination 
with DEA: 

Consistent with its mission of protecting public health, FDA 
established the Drug Shortage Program in 1999--now known as the Drug 
Shortage Staff (DSS)--to help prevent, alleviate, and resolve 
shortages. DSS compiles information on drug shortages received from 
manufacturers and the public, including the reasons for the shortages. 
[Footnote 34] FDA defines a shortage as a period of time when the 
demand or projected demand for the drug within the United States 
exceeds the supply of the drug.[Footnote 35] DSS verifies the 
existence of a shortage by (1) determining if the current demand for a 
product is stable or increasing based on historical data from IMS 
Health; (2) contacting all manufacturers of a given drug to 
investigate supply issues and to obtain information on inventory, 
manufacturing schedules, and any changes to ordering patterns; (3) 
evaluating product distribution at the wholesale level, if needed; and 
(4) assessing information obtained from IMS Health, manufacturers, and 
wholesalers for the particular product reported to have a potential or 
actual shortage and for any acceptable alternative products.[Footnote 
36] DSS tracks drug shortages until their resolution, and posts 
information on its website about current and resolved shortages. FDA 
also obtains information from the American Society of Health-System 
Pharmacists, which becomes aware of drug shortages through voluntary 
reporting from hospitals, pharmacists, and other health care 
providers. FDA responds to shortages by taking actions within its 
authority to address the underlying causes and to enhance product 
availability. For example, FDA may offer feedback on a manufacturers' 
proposed approach to responding to quality concerns. 

FDASIA, enacted in July 2012, contains several provisions related to 
drug shortages that require coordination between FDA and DEA. When FDA 
is notified of a supply disruption of a drug that is life-supporting, 
life-sustaining, or intended for use in the prevention or treatment of 
a debilitating disease or condition, and that contains a controlled 
substance subject to quotas, FDASIA requires that FDA request, if FDA 
determines that it is necessary, that DEA increase quotas applicable 
to that controlled substance.[Footnote 37] Similarly, when FDA has 
determined that a schedule II drug is in shortage in the United 
States, manufacturers may also make such requests by submitting quota 
applications and FDASIA requires that DEA respond to these requests 
from manufacturers within 30 days.[Footnote 38] Regardless of whether 
the request came from FDA or manufacturers, if DEA determines that 
issuing additional quota is not necessary to address a shortage it 
must provide a written response, which FDA is to post on its website. 
[Footnote 39] 

Additionally, FDASIA requires both FDA and DEA to describe their 
coordination on efforts to prevent or alleviate drug shortages in 
annual reports to Congress.[Footnote 40] FDASIA also requires FDA to 
maintain an up-to-date list of drugs that it determines to be in 
shortage, which, subject to public health, trade secret, and 
confidentiality provisions, must be publicly available.[Footnote 41] 
To meet this requirement, FDA posts information about drug shortages 
on its website, including the name of the drug and the reason for the 
shortage. 

Performance Measures and Internal Controls: 

Under the Government Performance and Results Act of 1993 (GPRA), as 
amended by the GPRA Modernization Act of 2010, agencies are 
statutorily required to prepare annual performance plans that 
articulate goals for the upcoming fiscal year and must include 
indicators that the agency will use to measure its performance. 
[Footnote 42] Standards for internal control in the federal government 
and federal guidance on performance management also call for agencies 
to establish objectives and performance measures for programs that 
clearly link to agency-wide goals.[Footnote 43] As our prior work has 
shown, performance measures are a key tool to help managers assess 
progress toward the stated agency-wide strategic goals and program-
level objectives.[Footnote 44] Leading agencies seek to establish 
clear hierarchies of results-oriented performance measures and targets 
to ensure that the agency is achieving its mission and strategic 
goals. Such performance measures help agencies make resource decisions 
and foster accountability to communicate agency progress to Congress 
and the public. 

Internal control, which is synonymous with management control, 
comprises the plans, methods, and procedures used to ensure an agency 
is meeting its missions, goals, and objectives, which supports 
performance-based management.[Footnote 45] Such controls help agency 
program managers achieve desired results and provide reasonable 
assurance that program objectives are being achieved through, among 
other things, effective and efficient use of agency resources. In 
addition, internal control standards in the federal government call 
for agencies to communicate with and obtain information from external 
stakeholders, compare actual performance to planned or expected 
results, review performance measures and indicators, compare and 
assess different sets of data to establish relationships and then take 
appropriate action, and to document all internal controls, 
transactions, and other significant events. In addition, we have 
previously reported on the importance of interagency collaboration. 
[Footnote 46] Collaboration can be broadly defined as any joint 
activity that is intended to produce more public value than could be 
produced when organizations act alone. 

Shortages of Drugs Containing Controlled Substances Rose Sharply After 
2007; Many Involved Generic Pain Relievers and Drugs Manufactured by 
Few Companies: 

Shortages of drugs containing controlled substances have increased 
sharply in recent years; such shortages lasted for nearly a year, on 
average, and the majority of drugs were in shortage multiple times. 
Half of the shortages of drugs containing controlled substances 
identified as critical by UUDIS were analgesics (pain relievers) and 
most involved generic drugs. 

Shortages of Drugs Containing Controlled Substances Have Spiked in 
Recent Years; Most Were in Shortage Multiple Times and for Nearly 1 
Year on Average: 

Of the 168 shortages of drugs containing controlled substances that 
were reported from January 2001 through June 2013, nearly 70 percent 
began between 2008 and 2013.[Footnote 47] The number of new shortages 
reported each year peaked in 2009--with 28 that year--and has somewhat 
declined since then, though the number still remains significantly 
above pre-2008 levels. (See figure 2.) The recent increase in these 
shortages mimics the pattern we have found for shortages of all drugs, 
which have also sharply increased since 2007 and subsequently declined 
after peaking in 2011. The 168 shortages of drugs containing 
controlled substances represent a small percentage of the total 1,575 
drug shortages reported from January 2001 through June 2013. 

Figure 2: New Shortages of Drugs Containing Controlled Substances and 
All Drugs Reported January 2001 through June 2013, by Year: 

[Refer to PDF for image: combined vertical bar and line graph] 

Year: 2001; 
Number of shortages of drugs containing controlled substances: 12; 
Number of shortages of all drugs: 102. 

Year: 2002; 
Number of shortages of drugs containing controlled substances: 6; 
Number of shortages of all drugs: 81. 

Year: 2003; 
Number of shortages of drugs containing controlled substances: 10; 
Number of shortages of all drugs: 71. 

Year: 2004; 
Number of shortages of drugs containing controlled substances: 5; 
Number of shortages of all drugs: 54. 

Year: 2005; 
Number of shortages of drugs containing controlled substances: 3; 
Number of shortages of all drugs: 71, 

Year: 2006; 
Number of shortages of drugs containing controlled substances: 5; 
Number of shortages of all drugs: 64. 

Year: 2007; 
Number of shortages of drugs containing controlled substances: 10; 
Number of shortages of all drugs: 114. 

Year: 2008; 
Number of shortages of drugs containing controlled substances: 20; 
Number of shortages of all drugs: 138. 

Year: 2009; 
Number of shortages of drugs containing controlled substances: 28; 
Number of shortages of all drugs: 157. 

Year: 2010; 
Number of shortages of drugs containing controlled substances: 21; 
Number of shortages of all drugs: 200. 

Year: 2011; 
Number of shortages of drugs containing controlled substances: 22; 
Number of shortages of all drugs: 255. 

Year: 2012; 
Number of shortages of drugs containing controlled substances: 19; 
Number of shortages of all drugs: 195. 

Year: 2013 (through June 30); 
Number of shortages of drugs containing controlled substances: 7; 
Number of shortages of all drugs: 73, 

Note: A shortage is counted as "reported" in the year that the 
University of Utah Drug Information Service is first notified of the 
shortage. The totals for shortages of all drugs reported for 2007 
through 2011 may differ slightly from what we reported in our 2011 
report due to minor modifications by the University of Utah Drug 
Information Service to its data since that report. See [hyperlink, 
http://www.gao.gov/products/GAO-12-116], 16. 

[End of figure] 

The vast majority of shortages of drugs containing controlled 
substances--96 percent--lasted longer than 1 month and some of these 
shortages spanned multiple years. (See table 1.) The average duration 
was nearly 1 year; however, some were much longer. For instance, a 
single shortage of lorazepam injection, a medication used to treat 
anxiety, lasted slightly more than 5 years. 

Table 1: Summary of the Duration of Shortages of Drugs Containing 
Controlled Substances, January 2001 through June 2013: 

Duration of shortages: Shortages lasting less than 30 days; 
Number of shortages: 6; 
Percent[A]: 4%; 
Average duration[A] (in days): 17. 

Duration of shortages: Shortages lasting between 30 and 365 days; 
Number of shortages: 96; 
Percent[A]: 61%; 
Average duration[A] (in days): 160. 

Duration of shortages: Shortages lasting more than 365 days; 
Number of shortages: 55; 
Percent[A]: 35%; 
Average duration[A] (in days): 729. 

Duration of shortages: Shortages lasting zero days[A]; 
Number of shortages: 11; 
Percent[A]: 0; 
Average duration[A] (in days): 0. 

Duration of shortages: Total; 
Number of shortages: 168; 
Percent[A]: 100%; 
Average duration[A] (in days): 354. 

Source: GAO analysis of University of Utah Drug Information Service 
data. GAO-15-202. 

Note: 

[A] We excluded 11 shortages of the 168 shortages of drugs containing 
controlled substances identified by the University of Utah Drug 
Information Service from January 2001 through June 2013 from this 
analysis because these shortages lasted 0 days, leaving 157 shortages. 
The majority of these excluded shortages represented manufacturers' 
decisions to discontinue their production of a drug. 

[End of table] 

Most shortages--143 of 168--involved drugs that were in short supply 
multiple times, ranging from two to seven times. Forty-five different 
drugs containing controlled substances were reported to be in shortage 
multiple times from January 2001 through June 2013, representing 143 
individual shortages. Overall, these 45 different drugs were in short 
supply for an average of about 3 years. For instance, oxycodone oral 
solution, a drug that is used to treat moderate to severe pain, was in 
shortage for the longest combined amount of time over the course of 
four different shortages, with a combined duration of over 8.5 years. 

Drugs containing schedule II substances accounted for more than half 
of the shortages of drugs containing controlled substances that were 
reported from January 2001 through June 2013. Fifty-seven percent of 
shortages involved schedule II substances, while the remaining 
schedules--schedules III, IV, and V--comprised 15, 26, and 2 percent 
of shortages, respectively. Shortages of drugs containing schedule II, 
III, and IV substances ranged on average from 356 days to 359 days. 
However, shortages of schedule V drugs were shorter, lasting an 
average of 228 days. 

Though the number of shortages reported has declined in recent years, 
the number of active shortages of drugs containing controlled 
substances remains high. (See figure 3.) The active shortage total for 
each year includes both (1) new shortages reported that year and (2) 
shortages that started in a prior year that were still ongoing. For 
example, a shortage reported in July 2010 and resolved in March 2012 
would be counted as an active shortage in three different years (2010, 
2011, and 2012). Rather than peaking in 2009, as the number of new 
shortages reported did, the number of active shortages has remained 
more or less constant, with 37 to 42 active shortages of drugs 
containing controlled substances per year from 2009 to 2012. For 
instance, in 2012, there were 19 new shortages reported that year and 
another 23 shortages that started in prior years and remained ongoing 
through some of the year. As of June 2013, there were 31 active 
shortages of drugs containing controlled substances. 

Figure 3: Number of Active Shortages of Drugs Containing Controlled 
Substances, January 2001 through June 2013, by Year: 

[Refer to PDF for image: stacked vertical bar graph] 

Year: 2001; 
New shortages, by year reported: 12; 
Ongoing shortages, which began in prior years: 0. 

Year: 2002; 
New shortages, by year reported: 6; 
Ongoing shortages, which began in prior years: 9. 

Year: 2003; 
New shortages, by year reported: 10; 
Ongoing shortages, which began in prior years: 12. 

Year: 2004; 
New shortages, by year reported: 5; 
Ongoing shortages, which began in prior years: 15. 

Year: 2005; 
New shortages, by year reported: 3; 
Ongoing shortages, which began in prior years: 7. 

Year: 2006; 
New shortages, by year reported: 5; 
Ongoing shortages, which began in prior years: 5. 

Year: 2007; 
New shortages, by year reported: 10; 
Ongoing shortages, which began in prior years: 4. 

Year: 2008; 
New shortages, by year reported: 20; 
Ongoing shortages, which began in prior years: 6. 

Year: 2009; 
New shortages, by year reported: 28; 
Ongoing shortages, which began in prior years: 9. 

Year: 2010; 
New shortages, by year reported: 21; 
Ongoing shortages, which began in prior years: 20. 

Year: 2011; 
New shortages, by year reported: 22; 
Ongoing shortages, which began in prior years: 19. 

Year: 2012; 
New shortages, by year reported: 19; 
Ongoing shortages, which began in prior years: 23. 

Year: 2013; 
New shortages, by year reported: 7; 
Ongoing shortages, which began in prior years: 24. 

Source: GAO analysis of University of Utah Drug Information Service 
data. GAO-15-202. 

Note: The active shortage total for each year includes (1) new 
shortages reported that year and (2) shortages that were reported in a 
prior year that remained ongoing. 

[End of figure] 

Half of Critical Shortages of Drugs Containing Controlled Substances 
Were of Pain Relievers and Many Involved Generic Products: 

Of the 87 shortages of drugs containing controlled substances 
identified as critical by UUDIS from January 2001 through June 2013, 
half involved pain relievers (analgesics) (44 of 87). These shortages 
were identified as critical by UUDIS because alternative medicines 
were not available, the shortages affected multiple manufacturers, or 
it received multiple reports of shortages from different provider 
institutions. There were also shortages of anti-anxiety medications, 
sedative hypnotic drugs, and stimulants. (See table 2.) All of these 
types of drugs belong to the broader class of drugs that affect the 
central nervous system, and are used to treat seizures, manage 
anxiety, and to provide sedation and pain reduction, among other 
things.[Footnote 48] 

Table 2: Summary of Critical Shortages of Drugs Containing Controlled 
Substances by Therapeutic Class, January 2001 through June 2013: 

Therapeutic class: Analgesic; 
Number of critical shortages: 44; 
Percent of critical shortages: 51. 

Therapeutic class: Antianxiety; 
Number of critical shortages: 7; 
Percent of critical shortages: 8. 

Therapeutic class: Sedative hypnotic; 
Number of critical shortages: 8; 
Percent of critical shortages: 9. 

Therapeutic class: Stimulant; 
Number of critical shortages: 5; 
Percent of critical shortages: 6. 

Therapeutic class: Other; 
Number of critical shortages: 23; 
Percent of critical shortages: 26. 

Therapeutic class: Total; 
Number of critical shortages: 87; 
Percent of critical shortages: 100. 

Source: GAO analysis of University of Utah Drug Information Service 
and Truven Health Analytics (Red Book). GAO-15-202. 

Note: This figure reflects 87 of the 168 (52 percent) shortages of 
drugs containing controlled substances reported during this time 
period; it is limited to the shortages the University of Utah Drug 
Information Service identified as critical. Red Book is a compendium 
published by Truven Health Analytics that includes information about 
the characteristics of drug products. 

[End of table] 

Most critical shortages of drugs containing controlled substances--68 
of 87 (78 percent)--involved a generic drug. This 78 percent includes 
38 critical shortages of drugs available only in generic form and 30 
shortages of drugs available in both brand-name and generic forms. 
Another 19 critical shortages were of drugs available only in brand-
name form (22 percent). Our analysis also showed that 40 critical 
shortages (46 percent) involved generic pain relievers. These 40 
critical shortages include 19 shortages of analgesics only available 
in generic form (22 percent) and 21 shortages of analgesics available 
in brand-name and generic form (24 percent). (See figure 4.) 

Figure 4: Distribution of Critical Shortages of Drugs Containing 
Controlled Substances by Therapeutic Class and Product Type, January 
2001 through June 2013: 

[Refer to PDF for image: pie-chart] 

Analgesic drugs: 46%; 40 analgesic drugs available in generic only or 
brand-name and generic form: 
Only available in generic form: 22%; 19 drugs; 
Available in brand-name and generic form: 24%; 21 drugs; 
Only available in brand-name form: 5%; 4 drugs; 
Nonanalgesic drugs: 
Only available in generic form: 22%; 19 drugs; 
Available in brand-name and generic form: 10%; 9 drugs; 
Only available in brand-name form: 17%; 15 drugs. 

Source: GAO analysis of data from the University of Utah Drug 
Information Service and Truven Health Analytics (Red Book). GAO-15-202. 

Note: This figure reflects 87 of the 168 (52 percent) shortages of 
drugs containing controlled substances reported during this time 
period; it is limited to the shortages the University of Utah Drug 
Information Service identified as critical. Non-analgesics include all 
other therapeutic classes, such as anesthetics and anti-anxiety 
medications. Red Book is a compendium published by Truven Health 
Analytics that includes information about the characteristics of drug 
products. 

[End of figure] 

Most critical shortages of drugs containing controlled substances--79 
shortages (91 percent)--involved drugs that were manufactured by four 
or fewer companies and 39 of those shortages involved drugs 
manufactured by only one company. The remaining 8 shortages involve 
drugs that were manufactured by five or more companies (9 percent). 

Finally, injectable drugs were in short supply slightly more often 
than orally administered drugs during this time period, with 44 
critical shortages of injectable drugs (51 percent) and 35 critical 
shortages of orally administered drugs (40 percent). The remaining 
shortages involved drugs with other routes of administration, such as 
those administered topically or via the rectum. 

Shortages of Drugs Containing Controlled Substances Can Compromise 
Patient Care and Create Challenges for Providers: 

According to provider and patient organizations, during shortages of 
drugs containing controlled substances, patients may receive less 
effective care, experience medication errors, or not receive treatment 
at all. Provider organizations also report that providers encounter 
serious challenges when such drugs are in short supply, requiring them 
to spend time and resources to mitigate the effects of shortages. 

Shortages Can Result in Less Effective Care, Medication Errors, and 
Delayed or Canceled Treatment: 

FDA reports that drug shortages can pose a significant threat to the 
public health by adversely affecting care for patients. According to 
FDA, drug shortages can delay or deny needed care for patients. Drug 
shortages can also lead prescribers to use second-line alternatives, 
which may be less effective or pose additional risks. In addition, 
provider and patient organizations we spoke with told us that 
shortages of drugs containing controlled substances have resulted in 
less effective patient care. For example, representatives from an EMS 
organization told us that some providers limited the amount of pain 
medication administered to patients in order to make their supply of 
the drugs last longer, resulting in patients with unmanaged pain. In 
addition, some providers often have to use substitutes during 
shortages that involve undesirable side effects or can be less 
effective, according to provider organizations.[Footnote 49] 
Representatives from the American Society of Anesthesiologists told us 
that when providers have to use an alternative anesthesia medication 
that is not the preferred treatment, patients can experience longer 
recovery times, delayed awakening, or postoperative nausea or 
vomiting. Medications can affect patients differently and, as a 
result, finding an effective substitute during a shortage can be 
complicated. For instance, according to provider and patient 
organizations focused on mental health issues, providers treating 
patients with ADHD may need to try different drugs containing other 
controlled substances and at varied doses to find one that works, 
although perhaps not as effectively as the patient's usual medication. 
Even switching from an extended-release formulation to an immediate-
release formulation of the same drug can have a significant effect on 
a child with ADHD if the child's school does not have a nurse to 
supervise the child taking medication during the day, according to the 
director of UUDIS. One organization representing patients and families 
affected by this disorder, Children and Adults with Attention 
Deficit/Hyperactivity Disorder, found that 18 percent of survey 
respondents had to change medications during a shortage of ADHD 
medications. Patients may also be seriously affected by changing to a 
different route of administration of the same drug during shortages. 
Patients receiving hospice care or near the end of life may require 
injections because they cannot swallow a pill for example. Therefore, 
during a shortage of a morphine injection, clinicians may have to 
switch the patient to another injectable pain reliever because the 
patient cannot swallow the morphine tablet, according to the director 
of UUDIS. 

Patients are at increased risk of experiencing medication errors 
during these shortages. Providers may have to use drugs in unfamiliar 
dosages or concentrations during shortages of drugs containing 
controlled substances, which can increase the risk of medication 
errors and lead to serious consequences for patients. For instance, in 
its survey of health care practitioners, the Institute for Safe 
Medication Practices found several instances of overdose and death 
from providers incorrectly converting doses for pain relievers during 
a morphine shortage.[Footnote 50] Errors administering sedatives have 
also led to patients receiving too much or too little medication, 
resulting in unintended consciousness during surgery, agitation, and 
increased recovery times. EMS organizations also noted that the 
providers they represent are particularly at risk for administering 
medication incorrectly as they typically have less experience dealing 
with different medications, doses, or concentrations than hospital-
based providers. Additionally, representatives of an EMS organization 
added that EMS providers deliver care in a mobile environment, which 
further increases the risk of administering medication incorrectly. 

Patients may have to delay or forgo care because of shortages of drugs 
containing controlled substances. For instance, provider organizations 
told us that providers have had to cancel or delay surgeries and 
procedures during shortages of anesthesia drugs. EMS associations we 
spoke with also reported that patients were at risk when drugs that 
are used to treat seizures were in short supply and there were no 
available substitutes. 

In addition to compromised care, patients faced other hardships. For 
example, parents spent significant time tracking down ADHD medications 
that were in shortage for their children, according to provider and 
patient organizations focused on mental health issues, such as 
Children and Adults with Attention Deficit/Hyperactivity Disorder and 
the American Academy of Child and Adolescent Psychiatry. In 
particular, these organizations reported that some parents of children 
with ADHD called more than 30 pharmacies to find their children's 
medication. Representatives of mental health organizations also 
reported that patients faced increased costs during shortages. Some 
paid higher prices for their drugs when generic ADHD medications were 
in short supply and only the more expensive brand-name drug was 
available. In other instances, patients found they had to visit their 
physicians more often to obtain new prescriptions when pharmacies with 
dwindling supplies could only partially fill their orders, which then 
resulted in additional fees from the physician to obtain a new 
prescription, according to mental health organizations.[Footnote 51] 

Providers Face Challenges Mitigating the Effects of Shortages of Drugs 
Containing Controlled Substances: 

FDA reports and provider organizations told us that shortages of drugs 
containing controlled substances burden providers, as the providers 
must spend time and resources addressing such shortages rather than 
providing care to patients. For example, some providers have had to 
add full-time staff to manage the effects of shortages and track down 
medications, according to provider organizations. Officials from a 
provider organization said that those who could not afford the cost of 
additional personnel may have had to divert staff from patient care. 
EMS organizations said that shortages can be particularly challenging 
for EMS organizations that typically have fewer staff or resources 
than large hospitals to dedicate to finding drugs that are in short 
supply. These reports corroborate research on the cost of all drug 
shortages; researchers at the University of Michigan, in collaboration 
with the American Society of Health-System Pharmacists, found that the 
labor costs for managing all drug shortages for U.S. hospitals is 
approximately $216 million annually.[Footnote 52] 

In trying to mitigate the effects of shortages of drugs containing 
controlled substances, providers may face situations that pose safety 
and efficacy concerns, and create ethical dilemmas. For example, 
providers may opt to turn to expired medications. FDA states that 
expired medications can be less effective or pose risks due to a 
change in chemical composition or decrease in potency. In limited 
cases FDA has exercised enforcement discretion to enable use of 
expired drugs to alleviate the effect of a shortage. Some states allow 
the use of expired drugs and providers in those states often pay for 
expensive tests to assess an expired drug's stability and efficacy, 
according to provider organizations.[Footnote 53] Obtaining drugs 
containing controlled substances in short supply through compounding 
pharmacies is another avenue providers may take that can pose risks to 
patients and can cost significantly more than noncompounded drugs. 
[Footnote 54] Compounded drugs can raise safety and efficacy concerns 
because the pharmacies that produce such drugs may be exempt from some 
FDA requirements that apply to drug manufacturers.[Footnote 55] 
Provider organizations also said drugs containing controlled 
substances from compounding pharmacies are significantly more 
expensive than traditional suppliers and have a shorter shelf live, 
which can result in wasted product. 

EMS providers may face additional challenges during shortages of drugs 
containing controlled substances, compared to other providers. EMS 
organizations said that their providers typically obtain medications 
either through contracts with hospitals or directly from distributors. 
However, provider organizations said that during periods of short 
supply, it may be a low priority for hospitals to ensure that EMS 
providers receive drugs in shortage because hospitals will typically 
fill their own supply first. EMS organizations added that distributors 
are more likely to cater to hospitals and other, larger providers 
because EMS providers are comparatively small purchasers.[Footnote 56] 
Additionally, EMS organizations reported some confusion about whether 
the providers they represent can transfer medications to other 
affiliated EMS providers because of federal and state laws.[Footnote 
57] In addition, the vast majority of EMS providers are governed by 
state or local clinical protocols and, in some cases, medication 
formularies designate the specific doses and concentrations of 
medications that are permitted for use.[Footnote 58] As a result, EMS 
providers in those states and localities are likely to continue to 
grapple with short supplies, unless state laws or regulations enable 
them to use different medications, doses, or concentrations than what 
is allowed under existing formularies or protocols. 

DEA Has Not Effectively Administered the Quota Process Due to Missed 
Time Frames and Weak Internal Controls: 

DEA did not meet its required time frames for establishing quotas for 
classes of schedule II substances, and manufacturers report that the 
lack of timeliness has caused or exacerbated shortages of some of 
their drugs containing controlled substances. Additionally, our work 
shows that DEA's lack of internal controls, such as controls to ensure 
data reliability, performance measures, and monitoring of performance, 
may hinder the agency's ability to ensure an adequate and 
uninterrupted supply of controlled substances. 

DEA Did Not Meet Required Time Frames for Establishing Quotas for 
Schedule II Substances; Manufacturers Reported Quotas Caused or 
Exacerbated Shortages of Some of Their Products: 

Our examination of DEA's Federal Register notices proposing and 
establishing APQs found that DEA has not met its required time frames 
for more than a decade. Specifically, DEA has not met the time frames 
provided in its regulations for proposing APQs for classes of schedule 
II substances and establishing annual bulk manufacturing and 
procurement quotas for individual manufacturers since at least 2001. 

* Proposing APQs: DEA did not propose APQs in the Federal Register on 
or before May 1st, as required by DEA's regulations, for any year from 
2001 through 2014.[Footnote 59] On average, DEA proposed APQs about 5 
months after the May 1 deadline specified in its regulations. 
Additionally, though there is no requirement for when DEA must 
establish the APQ in the Federal Register, for 10 of the 14 years from 
2001 through 2014, DEA established APQs in the Federal Register in 
December of the application year, generally in the last two weeks of 
the year.[Footnote 60] For one year (2005), DEA established the APQs 
in the Federal Register in January of the production year. DEA does 
not inform manufacturers of their quota authorizations until after 
APQs have been established for the production year; therefore, the 
date the APQ is established affects when manufacturers receive their 
quotas. 

* Establishing annual bulk manufacturing and procurement quotas: 
According to our analysis of Federal Register notices, DEA did not 
meet the requirements specified in the agency's regulations for 
establishing bulk manufacturing and procurement quotas by July 1, for 
any year from 2001 through 2014.[Footnote 61] From our analysis of a 
random probability sample of YERS/QMS source documents, we estimate 
that DEA took an average of 182 days in 2011 and 178 in 2012 beyond 
July 1 to establish annual bulk manufacturing and procurement quotas. 
[Footnote 62] In addition to not meeting the time frames in DEA's 
regulations, our analysis of Federal Register notices found that 
neither did DEA meet the CSA requirement for establishing bulk 
manufacturing quotas by October 1 from 2001 through 2013, though it 
did meet this deadline recently.[Footnote 63] For 2014 annual bulk 
manufacturing quotas, DEA established manufacturers' bulk 
manufacturing quotas in September 2013. 

Moreover, manufacturers have expressed concerns about DEA's timeliness 
in establishing quotas and have asserted that the amount of time it 
takes DEA to respond to their quota applications has contributed to 
shortages of some drugs containing controlled substances. We obtained 
FDA data on the manufacturer-reported causes of 40 shortages of drugs 
containing schedule II controlled substances from January 2010 through 
June 2013.[Footnote 64] These data show that 10 manufacturers reported 
to FDA that 7 of the 40 shortages of drugs containing schedule II 
substances during this time period were caused by problems related to 
quota.[Footnote 65] The remaining 33 shortages of drugs containing 
schedule II substances were caused by other factors, such as 
manufacturing delays, capacity issues, and product quality issues, 
according to information reported to FDA by manufacturers.[Footnote 
66] After reviewing this information, we contacted representatives 
from all of these companies. Many told us that DEA's lack of 
timeliness in establishing manufacturers' quotas has caused or 
exacerbated shortages of their products containing schedule II 
substances. In particular, the representatives said that DEA's timing 
of establishing annual bulk manufacturing and procurement quotas, 
which has generally been in late December of the application year, 
after it establishes the APQ, does not provide manufacturers with 
enough time to plan for production and order the raw material or API 
needed to start manufacturing their products at the beginning of the 
production year in January. Representatives from dosage form 
manufacturers report that these late decisions leave manufacturers 
operating solely with what is left in their inventory for the first 
few months of the production year, which may be limited because 
manufacturers operate in a lean manufacturing environment where they 
carry as little inventory as possible. 

Additionally, manufacturer representatives said that DEA's lack of 
timeliness in responding to supplemental quota applications submitted 
during the production year has also caused or exacerbated shortages of 
their products. Based on our analysis of a probability sample of YERS/ 
QMS source documents, we estimate that it took DEA an average of 58 
days to respond to supplemental quota applications in 2011 and 2012. 
Within the source documents we reviewed, DEA's fastest response to a 
supplemental application was 7 days and its longest response took 
about 5 months. While there are no required time frames for DEA to 
respond to supplemental applications in the CSA or DEA's regulations, 
as of July 2012, FDASIA has required DEA to respond to manufacturers' 
quota requests that relate to shortages caused by quotas within 30 
days.[Footnote 67] Although this 30-day time frame was not yet a 
requirement in 2011 and much of 2012, it is worth noting that it often 
took longer for DEA to respond, suggesting this requirement could pose 
a challenge during a future shortage. We estimate that DEA responded 
within 30 days to 21 percent of supplemental applications in 2011 and 
2012. Furthermore, we found that DEA took 10 days longer, on average, 
to respond to supplemental quota applications submitted by 
manufacturers that reported shortages caused by quota to FDA in 2011, 
the year in which these shortages began, than other supplemental 
applications (see table 3). In 2012, however, we found no statistical 
difference between the estimated amount of time DEA took to respond to 
both groups of supplemental applications. 

Table 3: The Drug Enforcement Administration's (DEA) Estimated 
Response Times for Supplemental Applications in Year-End Reporting and 
Quota Management System (YERS/QMS) Source Documents, 2011 and 2012: 

YERS/QMS source documents: 2011; 
Applications submitted by manufacturers who reported shortages caused 
by quota: 
Yes: Average number of days: 73; 
No: Average number of days: 63[A]. 

YERS/QMS source documents: 2012; 
Applications submitted by manufacturers who reported shortages caused 
by quota: 
Yes: Average number of days: 51; 
No: Average number of days: 54[B]. 

YERS/QMS source documents: Both years; 
Applications submitted by manufacturers who reported shortages caused 
by quota: 
Yes: Average number of days: 59; 
No: Average number of days: 58[C]. 

Source: GAO analysis of YERS/QMS source documents. GAO-15-202. 

Notes: This table reflects our review of a stratified probability 
sample of 440 source documents from YERS/QMS. This sample included all 
of the 146 records in 2011 and 2012 that are associated with drug 
shortages manufacturers reported to the Food and Drug Administration 
(FDA) were caused by quotas. We consider YERS/QMS records to be 
associated with these shortages if the record contains a quota 
application that was (a) submitted by a manufacturer who reported 
quota caused a shortage of their product to FDA; (b) for a substance 
that was used in the products reportedly in shortage because of quota; 
and (c) was submitted during the time of the reported shortage caused 
by quota. Manufacturers may or may not have reported to DEA that their 
quota application was related to a shortage. We also analyzed a random 
sample of 294 records stratified by year that were not associated with 
any drug shortages reportedly caused by quotas--180 and 114 records 
from 2011 and 2012, respectively. 

[A] The margin of error at the 95 percent confidence level for 
supplemental applications submitted by manufacturers who did not 
report shortages caused by quota in 2011 is plus or minus 10.9 percent. 

[B] The margin of error at the 95 percent confidence level for 
supplemental applications submitted by manufacturers who did not 
report shortages caused by quota in 2012 is plus or minus 13.7 percent. 

[C] The margin of error at the 95 percent confidence level for 
supplemental applications submitted by manufacturers who did not 
report shortages caused by quota in 2011 and 2012 combined is plus or 
minus 8.8 percent. 

[End of table] 

DEA officials acknowledged that the agency has not met the time frames 
for proposing and establishing quotas, as required by the agency's 
regulations. DEA officials cited inadequate staffing of the Quota Unit 
as the reason why DEA has not met its requirements for proposing APQs 
and establishing manufacturers' bulk manufacturing and procurement 
quotas for schedule II substances. In particular, DEA officials noted 
that the agency's workload with respect to quotas has increased 
substantially and that drug manufacturing has become more complex 
since the CSA was enacted and DEA's regulations were established in 
the 1970s. For example, DEA officials said that when the CSA was 
enacted and DEA's regulations were initially established, dozens of 
companies applied for quota each year, whereas now hundreds of 
companies apply for quota. Additionally, DEA reports that 
manufacturers submit substantially more applications to DEA than they 
have in the past. In 2012, for example, DEA received over 3,000 quota 
applications, and officials said that each application necessitates 
its own evaluation and analysis for each substance as it relates to 
the application. In addition to an increasing workload, DEA officials 
said that the agency has faced hiring challenges. In particular, it 
has been unable to hire additional qualified scientists for the Quota 
Unit because DEA has not found candidates with the right skills who 
are also able to pass the background checks required for employment. 
They also said that offers to qualified candidates have been declined. 
DEA officials noted that as of October 2014, the agency was still in 
the process of trying to hire two additional scientific staff for the 
Quota Unit, which would increase the staffing level to 12 (8 quota 
scientists and 4 administrative staff). 

DEA officials added that the quality and timeliness of manufacturers' 
quota applications affects DEA's ability to meet its required time 
frames for proposing APQs and establishing manufacturers' quotas. For 
example, DEA estimates that the agency follows up with manufacturers 
on about half of the quota applications to get clarification on the 
information provided or to obtain additional information. DEA 
officials said that manufacturers also continue to revise their annual 
applications as manufacturing conditions change throughout the year, 
such as when manufacturers gain or lose business or encounter a 
manufacturing problem that destroys part of their inventory of 
controlled substances and they seek to replace what was lost.[Footnote 
68] From our analysis of a probability sample of YERS/QMS source 
documents, we estimate that although 80 percent of annual applications 
in 2011 and 65 percent of annual applications in 2012 were submitted 
in the first two quarters of the application year, another 19 percent 
of annual applications in 2011 and 22 percent of annual applications 
in 2012 were submitted in the fourth quarter of the application year. 
This may represent the first annual application submitted by a 
manufacturer for the production year or subsequent revisions because 
YERS/QMS contains both the first annual application submitted and 
subsequent revisions. DEA officials said that they do not consider 
manufacturers to be noncompliant with their time frames when they 
submit annual quota applications after the deadlines. They said they 
recognize that it is important for manufacturers to submit revisions 
as conditions change, to ensure they receive appropriate amounts of 
quota. Our review of a sample of YERS/QMS source documents shows that 
manufacturers often submit annual bulk manufacturing and procurement 
quota applications throughout the year. Despite this, DEA officials 
said that they expect the agency will be compliant with its CSA time 
frames in the future. DEA officials said that moving to an electronic 
data system for the quota process--YERS/QMS--has helped the agency to 
be more compliant with its time frames, as well as the reorganization 
of staff into a separate Quota Unit in 2011.[Footnote 69] They noted 
that the agency has already made progress in meeting its CSA-required 
time frames for establishing annual bulk manufacturing quotas in 2014. 

Although DEA acknowledges it has generally not adhered to the quota 
setting time frames required by the CSA and the agency's regulations 
in the last 14 years, agency officials do not agree that quotas can 
cause shortages of drugs containing schedule II substances. In 
particular, according to DEA, shortages cannot be related to quotas 
established by the agency because DEA authorizes quotas at the basic 
class level for a substance--such as amphetamine or morphine. DEA 
cannot authorize quota for specific drug products containing those 
substances, as this is precluded by the CSA. DEA officials said the 
agency has no control over what specific drug products manufacturers 
actually produce with the quota authorized. For example, if DEA 
authorized 10,000 grams of morphine to a manufacturer, the 
manufacturer--and not DEA--then decides how to authorize the quota 
amongst its various products that contain different formulations, 
dosage forms, and concentrations. 

We cannot confirm whether DEA's lack of timeliness in establishing 
annual and supplemental quotas has caused or exacerbated shortages. 
However, by not responding to annual applications in accordance with 
the time frames required by its regulations or the CSA and by not 
acting promptly on supplemental applications, DEA may hinder 
manufacturers' ability to manufacture drugs that contain schedule II 
controlled substances that may help prevent or resolve a shortage. 

DEA's Weak Internal Controls Jeopardize Its Ability to Effectively 
Manage the Quota Process and Ensure an Adequate Supply of Controlled 
Substances for Legitimate Medical Use: 

DEA does not have sufficient internal controls to effectively oversee 
and manage the quota process and therefore may not be able to 
appropriately respond to quota applications related to shortages when 
they arise. In particular, DEA does not have adequate controls to 
ensure the reliability of the YERS/QMS system that it uses to track 
manufacturers' quota applications and record its quota decisions. DEA 
officials told us there are no systematic quality checks to ensure 
that the data in YERS/QMS are accurate, such as electronic checks for 
data outliers or manual comparisons of YERS/QMS data to source 
documents (i.e., manufacturers' quota applications and corresponding 
DEA decision letters) to identify inconsistent information. Instead, 
DEA officials said that when manufacturers submit their quota 
applications electronically, YERS/QMS automatically checks to ensure 
that the manufacturer applying for quota has a valid DEA registration. 
Additionally, DEA officials said they added an electronic prompt to 
YERS/QMS that asks manufacturers to reexamine and confirm quota 
requests that are for smaller amounts than they have already received 
for the production year. Officials told us that the Quota Unit also 
reviews the information submitted by manufacturers and will contact 
them when information entered looks inconsistent or out of the 
ordinary. DEA also reports that managers verify that the information 
entered into YERS/QMS by the Quota Unit is accurate. Though DEA takes 
some steps to ensure the accuracy of its YERS/QMS data, the agency has 
not systematically checked to make sure that its efforts are 
sufficient or having the intended effect. Lacking systematic data 
checks is inconsistent with federal standards for internal control, 
which calls for agencies to have appropriate control activities in 
place, such as periodic data checks, to ensure that the data used by 
the agency for decision making are accurate.[Footnote 70] Without 
controls in place to ensure the accuracy of YERS/QMS data, DEA cannot 
provide reasonable assurance that it has correct information in its 
official record of the dates and amounts of its quota decisions beyond 
the individual electronic and hard copies of decision letters it keeps 
for retention purposes. Further, inaccuracies in YERS/QMS data will 
also affect the accuracy of future quota applications submitted by 
manufacturers, as YERS/QMS automatically populates information on 
quotas previously authorized by DEA from the prior 2 years into a new 
quota application. 

In our review of YERS/QMS data from 2011 and 2012, we found a large 
number of errors in 2011 and a significant reduction in the number of 
errors in 2012. Based on our analysis of a probability sample of YERS/ 
QMS data, we estimate that 44 percent of the records in 2011 and 10 
percent of the records in 2012, each of which corresponds to one quota 
application and DEA decision letter, contained at least one data field 
with incorrect data, such as incorrect dates or amounts of quota 
requested or authorized.[Footnote 71] In particular, we estimate that 
errors in the quota application fields declined substantially in 2012, 
with, for example, errors in the total amount of quota requested field 
dropping from 13 percent of records in 2011 to 3 percent of records in 
2012. See table 4 for the specific fields in which we found errors in 
our review for years 2011 and 2012. 

Table 4: Estimated Percent of Errors in the Drug Enforcement 
Administration's (DEA) Year-End Reporting and Quota Management System 
(YERS/QMS): 

Quota application: 

Data field: Application year; 
Estimated percent of records with errors[A]: 
2011: 0%; 
2012: 0%. 

Data field: Company name; 
Estimated percent of records with errors[A]: 
2011: 1%; 
2012: 0%. 

Data field: Substance name; 
Estimated percent of records with errors[A]: 
2011: 0%; 
2012: 1%. 

Data field: Quota type; 
Estimated percent of records with errors[A]: 
2011: 1%; 
2012: 0%. 

Data field: Date submitted; 
Estimated percent of records with errors[A]: 
2011: 23%; 
2012: 1%. 

Data field: Total amount of quota; 
Estimated percent of records with errors[A]: 
2011: 13%; 
2012: 3%. 

Decision letter: 

Data field: Application year; 
Estimated percent of records with errors[A]: 
2011: 0%; 
2012: 0%. 

Data field: Company name; 
Estimated percent of records with errors[A]: 
2011: 1%; 
2012: 0%. 

Data field: Substance name; 
Estimated percent of records with errors[A]: 
2011: 1%; 
2012: 0%. 

Data field: Quota type; 
Estimated percent of records with errors[A]: 
2011: 1%; 
2012: 0%. 

Data field: Date mailed; 
Estimated percent of records with errors[A]: 
2011: 9%; 
2012: 3%. 

Data field: Total amount of quota; 
Estimated percent of records with errors[A]: 
2011: 5%; 
2012: 2%. 

Total: 
Estimated percent of records with errors[A]: 
2011: 44%[B]; 
2012: 10%[B]. 

Source: GAO analysis of YERS/QMS data and source documents obtained 
from DEA. GAO-15-202. 

Notes: This table reflects our review of a stratified random sample of 
440 source records from YERS/QMS. The sample included all of the 146 
records in 2011 and 2012 that are associated with drug shortages 
manufacturers reported to the Food and Drug Administration (FDA) were 
caused by quotas. We consider YERS/QMS records to be associated with 
these shortages if the record contains a quota application that was 
(a) submitted by a manufacturer who reported quota caused a shortage 
of their product to FDA; (b) for a substance that was used in the 
products reportedly in shortage because of quota; and (c) was 
submitted during the time of the reported shortage caused by quota. 
Manufacturers may or may not have reported to DEA that their quota 
application was related to a shortage. We also analyzed a random 
sample of 294 records stratified by year that were not associated with 
any drug shortages reportedly caused by quotas--180 and 114 records 
from 2011 and 2012, respectively. 

[A] All percentage estimates have margins of error at the 95 percent 
confidence level of plus or minus 10 percentage points or less. 

[B] Rows may not sum to total because some records have errors in more 
than one data field. Therefore, such records are counted towards the 
total number of errors for multiple data fields but count only once 
for the total number of records with any error. 

[End of table] 

DEA officials said 2011 was a transitional year for YERS/QMS as 
manufacturers first started submitting quota applications 
electronically, and they expected the information for 2012 and beyond 
to be more accurate. DEA officials noted that they consider the 
information submitted to YERS/QMS by manufacturers to belong to the 
manufacturers and therefore DEA officials do not correct this 
information, even when they learn that it is not accurate. Instead, 
DEA officials said that they ask manufacturers to submit a revised 
application and they mark the original application as withdrawn, as 
well as make a note in YERS/QMS that the application will be corrected 
by manufacturers. Nonetheless, we removed applications marked as 
withdrawn from our analysis and we still found errors in the YERS/QMS 
data for manufacturers' quota applications, as well as in the data for 
DEA's quota decisions. As YERS/QMS data is the official record of how 
much quota is requested by manufacturers and authorized by DEA, 
maintaining accurate records is important for the management of the 
quota setting process. 

DEA also does not have agency-wide performance measures related to 
establishing quotas or ensuring an adequate and uninterrupted supply 
of controlled substances to assess how well the program is carrying 
out its responsibilities.[Footnote 72] Federal internal control 
standards state that performance measures are an important management 
tool and that such measures provide critical information to managers 
about agency performance and program outcomes.[Footnote 73] Such 
performance information can demonstrate progress toward goals and help 
program managers determine how to allocate resources among competing 
priorities. DEA officials said that performance measures related to 
establishing quotas or ensuring an adequate and uninterrupted supply 
of controlled substances would be inappropriate because of the 
complexity of individual quota applications and the difficulty of 
projecting the number of quota applications for future years. 
Nonetheless, performance measures related to quotas--such as the 
percentage of applications that require follow-up by DEA because of 
missing documentation--are a critical internal control, without which 
DEA cannot determine if its process for establishing quotas is having 
the desired outcome of ensuring an adequate and uninterrupted supply 
of controlled substances for legitimate medical use. In the absence of 
such performance measures, DEA is missing important information for 
program managers to use when making decisions about program resources 
and the agency cannot effectively demonstrate program results. 

In addition, DEA officials told us that they do not monitor or analyze 
available data from YERS/QMS to assess its administration of the quota 
process, which federal standards for internal control state is an 
important component of managing a program.[Footnote 74] DEA officials 
said that the agency does not monitor or analyze these data and that 
they do not use YERS/QMS to produce aggregate information on 
timeliness or other performance metrics and the agency has no plans to 
do so as of October 2014. Absent such analysis, DEA is unable to 
evaluate its responses to manufacturers' quota applications or to 
understand the nature of its workload. For example, whereas DEA 
previously withheld some of the amphetamine APQ to authorize later in 
the year, officials told us that the agency initially granted more of 
the amphetamine APQ in 2012 to manufacturers than it had previously, 
to see if manufacturers would still claim that there was a shortage of 
their products. However, DEA officials said they did not monitor the 
effect of this decision, which they said was unprecedented, and they 
were therefore unsure if it had an effect on shortages or the agency's 
workload. Officials were unable to say if this change resulted in 
manufacturers submitting fewer supplemental applications for 
amphetamine that year. Additionally, DEA officials cited 
manufacturers' incomplete applications and frequent revisions as 
affecting the agency's ability to meet its required time frames. 
However without monitoring or analyzing data to identify how 
frequently manufacturers submit incomplete and revised applications, 
DEA lacks information to assess the effect on the agency's workload. 
This lack of monitoring also prevents DEA from determining whether a 
few manufacturers are repeatedly submitting incomplete applications or 
if there is a larger, systemic issue to address. DEA therefore cannot 
identify appropriate means to help prevent such issues in the future, 
such as through in-depth training or guidance. 

Finally, DEA officials told us that the agency does not have 
protocols, policies, or other documentation to manage how the agency 
processes quota applications and establishes APQs and annual and 
supplemental quotas for individual manufacturers.[Footnote 75] 
Instead, DEA officials told us that the agency exclusively relies on 
its regulations and the CSA to serve as guidance on how to conduct 
these activities. According to federal internal control standards, 
agencies should have written documentation, such as detailed policies, 
procedures, and practices, to fit their agency's operations and to 
ensure that they are built into, and an integral part of, operations. 
The absence of written guidance may also pose a risk to the 
consistency of DEA's future operations in the event of staff turnover, 
changes in administration, or any other disruption that could lead to 
a loss of institutional knowledge.[Footnote 76] 

The need for detailed policies, procedures, and practices is 
particularly important because the activities conducted by Quota Unit 
staff are very complex, requiring staff to weigh data from at least 
five different sources and to make recommendations about how to 
authorize the APQ among various manufacturers. Having additional 
guidance for these activities is key to help ensure that staff in the 
Quota Unit process applications consistently and that they are 
adhering to DEA's regulations and the CSA when setting quotas. For 
instance, Quota Unit staff analyze and reconcile data from multiple 
sources in order to recommend quotas that meet estimated legitimate 
medical need. One such data source is ARCOS, DEA's system for tracking 
the distribution of controlled substances, which DEA uses in 
determining how much quota to authorize. Officials told us that there 
are many instances in which ARCOS data are unreliable and therefore it 
may contradict information from other sources used by quota 
scientists, such as the sales information manufacturers provide in 
their quota applications.[Footnote 77] Although quota scientists may 
be faced with contradictory information, it is left to the discretion 
of the Quota Unit staff to determine how to reconcile conflicting data 
sources, as there are no written policies or guidance and this level 
of detail is not included in DEA's regulations or the CSA. Moreover, 
DEA officials told us that the reliability of different data sources 
can vary based on the substance and the quota applicant. They noted 
that in some cases, manufacturer-reported sales data may not be as 
reliable as data on the number of retail prescriptions dispensed, 
whereas for a different substance or quota applicant, the reverse may 
be true. Nonetheless, despite the need to consider data sources 
differently relative to the substance class or quota applicant, there 
is nothing in the CSA or DEA's regulations about how to do this and 
DEA has no additional guidance on the matter. 

Additionally, when asked how they train new staff if there is no 
written guidance about the quota process, DEA officials said that they 
train new Quota Unit staff about the CSA and DEA's regulations through 
on-the-job training and using presentations created for industry. 
There are no specific documents developed for the exclusive purposes 
of training new staff. DEA officials said that the presentations for 
industry give new Quota Unit staff an overview of the quota process 
and that everything else must come from on-the-job training, as each 
function within the Quota Unit is unique and each quota application is 
different from the next. DEA officials told us that they ensure 
consistency in their decision-making by having the Deputy Assistant 
Administrator of the Office of Diversion Control review and authorize 
every quota decision that is made. However, given the volume of quota 
applications that DEA processes each year--over 3,000 in 2012--it is 
unreasonable to expect that the Quota Unit staff can make consistent 
decisions with no written guidance. Similarly it is not reasonable to 
assume that one senior manager can devote sufficient time to review 
these many staff decisions and ensure that quotas are set in 
accordance with DEA's regulations and the CSA. Further, the lack of 
written guidance poses a risk to the continuity of the agency's 
operations should personnel leave or be reassigned. 

DEA and FDA Have Not Established a Sufficiently Collaborative 
Relationship, Which Could Hinder Their Abilities to Effectively 
Coordinate and Address Future Shortages: 

FDASIA contains provisions that require DEA and FDA to coordinate 
their activities where additional quota may be needed to address a 
shortage of a drug containing controlled substances, though no such 
opportunity has occurred since the law's enactment in 2012. In 
addition, FDASIA requires each agency to report to Congress annually 
describing the coordination between the two agencies on efforts to 
prevent or alleviate shortages. Although both agencies report that 
their working relationship has improved in recent years--namely by 
establishing a point of contact at DEA who is responsible for 
coordinating with FDA's DSS--the agencies' use of essential 
collaborative practices to enhance their abilities to successfully 
coordinate to resolve or mitigate shortages has been limited.[Footnote 
78] We have previously identified key practices that agencies need for 
enhanced and sustained collaboration, including defining and 
articulating a common outcome and establishing compatible policies, 
procedures, and other means to operate across agency boundaries. 
[Footnote 79] Running throughout these practices are a number of 
factors, such as trust, that are important elements for a 
collaborative working relationship. 

One barrier to effective coordination between the two agencies is that 
DEA and FDA have not defined and articulated a common outcome related 
to shortages of drugs containing controlled substances. FDASIA gives 
FDA the authority to create and maintain a list of drug shortages. 
[Footnote 80] However, the two agencies sometimes disagree about 
whether a shortage exists. This is a fundamental concern because 
FDASIA also contains provisions that require the two agencies to work 
together to address shortages of drugs subject to quotas. According to 
DEA, FDA has posted shortages on its website when, from DEA's 
perspective, there is no shortage. DEA often views drugs containing 
the same substance class as viable substitutes for one another, even 
when the drugs are different strengths or formulations. Officials told 
us that there is no shortage, from DEA's perspective, as long as there 
is quota available to manufacture a given controlled substance, 
regardless of which manufacturers are producing the product and which 
strengths or formulations are available. For instance, DEA would not 
consider there to be a shortage of 1 milliliter vials of hydromorphone 
in the 10 milligram/milliliter concentration if the 5 milliliter vials 
were still available. Separately, FDA evaluates potential substitutes 
on a case by case basis to determine whether drugs are clinically 
interchangeable in order to assess if there is enough supply of a drug 
to meet demand. To determine clinical interchangeability, FDA states 
that it relies upon clinical expertise from within the agency, as well 
as consultation with outside medical professional associations. For 
example, for ADHD drugs, both the immediate release and extended 
release formulations are needed and are not substitutes for each 
other. According to FDA, some patients are well maintained on the 
extended release formulation alone, some patients require the 
immediate release formulation at certain points in the day to treat 
their specific symptoms, and some patients require both the extended 
release formulation as well as the immediate release formulation at 
specific times of the day to control their symptoms. FDA may or may 
not consider drugs of different strengths or formulations to be 
substitutes for one another and FDA's determination about whether 
drugs are clinically interchangeable has not always aligned with DEA's 
perspective, leading DEA to assert that some of the shortages listed 
on FDA's website are not actual shortages. By not reaching an 
agreement about what constitutes a drug shortage, the agencies are 
lacking a key practice of effective coordination and it is unclear if 
they will be able to work together to respond to shortages of drugs 
containing controlled substances caused by quotas. 

The agencies' collaboration is further impaired by a lack of trust on 
the part of DEA regarding the shortages listed on FDA's drug shortage 
website. DEA officials told us that they do not believe FDA 
appropriately validates or investigates the shortages it lists on its 
website, which they believe encourages abuse by manufacturers and 
distributors. Specifically, DEA officials said that they are concerned 
that manufacturers may falsely report shortages to FDA when they do 
not receive the amount of quota they requested, rather than because 
there are actual supply disruptions. As DEA does not consider FDA to 
have independently confirmed the existence of a shortage, DEA 
officials said that it considers FDA's website to be an incentive for 
manufacturers to file self-serving and misleading reports. As a 
result, despite the fact that DEA officials told us they monitor FDA's 
website, they still consider it to be neither useful nor accurate for 
DEA's purposes. When we spoke to FDA about its drug shortage website, 
officials told us that the agency takes multiple steps to verify the 
existence of a shortage before placing it on its website, including 
confirming the supply disruption with the manufacturer of the product, 
checking with manufacturers of related products to confirm that they 
cannot cover the decrease in supply, and analyzing market sales data 
to compare current supply with historical demand.[Footnote 81] 

Another barrier to effective collaboration is the lack of compatible 
policies, procedures, and other means to operate across agency 
boundaries, including mutually agreed upon time frames for DEA to 
respond if FDA notifies DEA of a shortage caused by quota. Although 
FDA established policies and procedures in September 2014 for 
requesting a quota adjustment from DEA if it determines that it is 
necessary to address a shortage, DEA officials said that the agency 
has not and does not plan to establish formal policies or procedures 
to coordinate the agency's response with FDA.[Footnote 82] Such 
actions are not consistent with key practices for effective 
collaboration. It is important to note that while FDASIA directs DEA 
to respond within 30 days to manufacturers that request additional 
quota pertaining to a drug in shortage that is on FDA's drug shortage 
list, the law does not specify how quickly DEA must respond to a 
request from FDA to address a shortage of a life-sustaining drug and 
the agencies do not have an agreement in place regarding this matter. 
A time frame for DEA to respond to an FDA request is particularly 
important, given that FDA has determined that there is a shortage of a 
life-sustaining drug that an increase in quota is necessary to 
address, and DEA may not necessarily agree with FDA's determination of 
a shortage. Federal standards for internal control also state that 
communicating information within a specified time frame that enables 
entities to carry out their responsibilities is an important control 
activity.[Footnote 83] DEA officials told us that the notifications it 
received from FDA prior to FDASIA were not useful. These notifications 
included copies of manufacturers' quota applications and FDA's 
assessment that the application was, or could be, related to a 
shortage. DEA maintained that, because FDA was not providing it with 
any new information, FDA's notifications were not helpful. Further, 
DEA officials said that the applications may or may not be related to 
a shortage from DEA's perspective. 

Finally, both agencies stated that they are subject to restrictions on 
exchanging the proprietary information they receive from drug 
manufacturers. Both agencies agree that sharing such information with 
one another would be helpful in preventing and mitigating shortages of 
drugs containing controlled substances. For example, DEA officials 
said that it would be helpful to be informed in advance of FDA's 
decisions about upcoming drug approvals that may help alleviate a 
shortage. According to DEA, this could provide it with early 
notification about the need to quickly verify information provided by 
manufacturers in quota applications for drugs that appear to be 
nearing FDA approval. DEA officials added that such information 
sharing would make it easier for the two agencies to communicate about 
drug shortages and quotas in a meaningful way. 

Although DEA and FDA have long had memorandums of understanding (MOU) 
in place, these documents do not address the topic of drug shortages. 
The agencies have been working for more than 2 years on developing a 
new MOU, which would replace the original documents that were executed 
in the 1970s. As of January 20, 2015, the new MOU had not been 
finalized and the two agencies report that they are continuing to work 
towards outlining mutually agreeable terms. Officials from both 
agencies explained that they expect the new MOU would establish the 
terms for sharing and safeguarding the proprietary information that 
they are currently prohibited from exchanging and that may ultimately 
allow the agencies to work together more effectively to address 
shortages. However, DEA officials said that they do not expect the new 
MOU to specify what types of information may be shared, including 
information related to a shortage of a drug containing controlled 
substances or the time frames for sharing such information. FDA 
officials indicated that once the new MOU is in place the two agencies 
will still need to take additional steps before they can begin to 
share information, such as developing a work plan for sharing 
information related to shortages of drugs containing controlled 
substances. As the agencies will need to take additional steps before 
they can begin exchanging information related to shortages after the 
MOU is in place, it is uncertain whether FDA and DEA will be able to 
expeditiously address shortages of drugs containing controlled 
substances, should one occur in the near future. 

Conclusions: 

Prescription drugs containing controlled substances are routinely used 
in inpatient, outpatient, and emergency settings. DEA is responsible 
for preventing, detecting, and investigating the diversion of these 
substances, while nonetheless ensuring an adequate and uninterrupted 
supply for legitimate needs. Because DEA maintains a closed system of 
distribution that limits the availability of certain substance classes 
by establishing aggregate production, bulk manufacturing, and 
procurement quotas, the agency's management of the quota process is of 
critical importance to public health. 

Our work shows that DEA is not well prepared to expeditiously respond 
to future shortages. DEA has not met its required time frames for 
establishing quotas for more than a decade. DEA also lacks sufficient 
internal controls to ensure the reliability of the data it uses to 
establish quotas, which we found led to errors in its data system. 
Moreover, DEA does not monitor the data it collects and has no 
established performance measures related to either setting quotas in a 
timely manner or ensuring an adequate and uninterrupted supply of 
controlled substances for legitimate medical use. Its lack of written 
policies and procedures for a complex process poses a risk to the 
continuity of its future operations. 

Because FDA also plays a vital role in preventing, mitigating, and 
resolving drug shortages, its collaboration with DEA is key to 
responding to shortages. FDA adopted the definition of a drug shortage 
that is specified in FDASIA and was assigned authority to maintain the 
list of drugs in shortage. Moreover, FDASIA directs the two agencies 
to work together to coordinate certain activities related to 
shortages. However, DEA and FDA are not able to effectively 
collaborate due to fundamental disagreement over whether any given 
shortage exists. DEA has made it clear it does not trust FDA's 
information, as it does not consider many of the shortages that FDA 
verifies to be legitimate. In addition, DEA has not developed any 
formal policies or procedures to coordinate its response to a shortage 
with FDA. The two agencies have been operating under an MOU that has 
been unchanged for about 40 years and that does not address the topic 
of drug shortages. Although it is under revision, this process began 
over 2 years ago and remains incomplete. Despite FDASIA's provisions 
directing the two agencies to coordinate and report on their responses 
to drug shortages, it is not clear that the new MOU will facilitate 
this. Given that the agencies have indicated that they will still need 
to take additional steps before they can begin exchanging information 
after the MOU is in place, they may not be able to provide a quick 
response should a shortage occur in the near future. 

While we cannot establish a causal relationship between shortages of 
drugs containing controlled substances and DEA's management of the 
quota setting process, the shortcomings we have identified prevent DEA 
from having reasonable assurance that it is prepared to help ensure an 
adequate and uninterrupted supply of these drugs for legitimate 
medical need, and to avert or address future shortages. This approach 
to the management of an important process is untenable and poses a 
risk to public health. 

Recommendations for Executive Action: 

To ensure that DEA is best positioned to administer the quota process 
to ensure an adequate and uninterrupted supply of controlled 
substances for legitimate medical use and respond to shortages of 
drugs containing controlled substances, we recommend that the 
Administrator of DEA take the following five actions: 

* establish controls, including periodic data checks, to ensure that 
the YERS/QMS data accurately reflect both manufacturers' quota 
submissions and DEA's decisions; 

* establish performance measures for DEA related to quotas and 
ensuring an adequate and uninterrupted supply of controlled substances 
for legitimate medical use; 

* monitor and analyze YERS/QMS data to assess DEA's administration of 
the quota process; 

* establish internal policies for processing quota applications and 
setting aggregate, annual, and supplemental quotas to ensure that 
staff conduct these activities consistently and in accordance with the 
CSA and agency's regulations; and: 

* expeditiously establish formal policies and procedures to facilitate 
coordination with FDA as directed by FDASIA, including a specific time 
frame in which DEA will be expected to respond to FDA requests to 
expedite shortage-related quota applications. 

To strengthen DEA's and FDA's ability to respond to shortages of drugs 
containing controlled substances, we recommend that the Administrator 
of DEA and the Commissioner of FDA take the following two actions: 

* promptly update the MOU between the two agencies, and: 

* either in the MOU or in a separate agreement, specifically outline 
what information the agencies will share, and time frames for sharing 
such information, in response to a potential or existing drug shortage. 

Agency Comments and Our Evaluation: 

We provided a draft of this report for comment to HHS and DOJ. In 
addition, we provided excerpts of this report to UUDIS. We received 
written comments from HHS and DEA, which are reproduced in full in 
appendix IV and V, respectively. Both agencies also provided technical 
comments, which we incorporated as appropriate. UUDIS did not have any 
comments. 

HHS agreed with the two recommendations applicable to FDA. 
Specifically, it agreed with our recommendation that it promptly 
update its MOU with DEA. It also agreed with our recommendation that 
it and DEA should specifically outline what information they will 
share and the time frames for doing so, in response to a potential or 
existing drug shortage. HHS also stated that FDA is actively working 
to finalize the new MOU with DEA. 

DEA did not explicitly agree or disagree with the seven 
recommendations we made to it, but commented on each. In some 
instances, its comments indicate that it is either supportive of 
strengthening its management of the quota process and improving 
coordination with FDA, or that it has already taken steps consistent 
with our recommendations. In one instance, it did not respond directly 
to our recommendation. DEA also raised multiple objections to our work 
and described what it characterized as flaws and weaknesses in our 
draft report. In particular, DEA stated concerns with (1) our 
understanding of quotas, (2) definition of a shortage, (3) the title 
of our draft report, (4) our methods of conducting investigation and 
data analysis, and (5) our conclusions. 

Understanding of quotas. DEA stated that we fail to understand that it 
does not have the authority to issue quotas for individual dosage 
forms and that it cannot require a manufacturer to manufacture or 
distribute its products, whether API or final dosage form. This report 
reflects these facts, specifically stating that "DEA does not have the 
authority to issue quotas for specific products or to require 
manufacturers to use their quota for specific products." DEA concluded 
that because the agency does not have this authority, its quota 
decisions have no effect on the availability of individual dosage 
forms containing substances that are subject to quotas. We disagree 
with this conclusion because manufacturers must receive authorization 
from DEA to obtain controlled substances subject to quotas, and so the 
amounts of quota authorized by DEA and the timing of its decisions 
directly affect when manufacturers are able to obtain the substances 
necessary to manufacture their products. We understand that 
manufacturers are ultimately responsible for what they manufacture 
with the quota authorized by DEA, but their decisions are made within 
the confines of the quota granted by DEA. 

Definition of a shortage. DEA stated that our draft report is 
misleading because we fail to account for the fact that the definition 
of a "shortage" means different things to different entities. DEA 
suggested that our approach is flawed because we focus on individual 
drug products rather than the basic class of controlled substances. We 
do so because this is the level that affects patients and providers. 
Thus, we used data from UUDIS and FDA, both of which track shortages 
of individual drug products. UUDIS is regarded as the most 
comprehensive and reliable source of drug shortage information between 
January 2001 through June 2013 and FDA has been mandated by statute to 
identify and monitor drug shortages. Drugs within the same substance 
class are not necessarily clinically interchangeable, according to FDA 
and UUDIS. For example, as we describe in this report, there can be 
significant implications for patients who have to switch from an 
extended-release formulation to an immediate-release formulation of 
the same drug and patients who have trouble swallowing may not be able 
to use a tablet formulation if an injection is in short supply. 

Additionally, DEA criticized our use of FDA's revised definition of a 
drug shortage, which FDA adopted in September 2014. DEA stated that 
this definition, which FDA adopted to be consistent with FDAISA, is 
"materially different" from the definition FDA used from 2006 through 
2012. DEA implied that, by doing so, our analysis was flawed in that 
we inappropriately applied the new definition to our analysis of 
YERS/QMS records from 2011 and 2012--a period predating FDA's revised 
definition. We disagree. Our analysis of certain YERS/QMS records was 
premised on particular shortages of drugs containing controlled 
substances that occurred between January 2010 and June 2013. We 
collected data from FDA beginning in October 2012 to specifically 
identify the manufacturer-reported causes of shortages of drugs 
containing controlled substances. We completed our collection of FDA's 
data in August 2013. We could not have applied FDA's revised 
definition of a drug shortage to our efforts as, at that time, the 
revised definition was not yet adopted by FDA. We requested the 
YERS/QMS source documents, including the entire population of records 
in 2011 and 2012 that were associated with drug shortages 
manufacturers reported to FDA that were caused by quotas, in June 
2014. Therefore, FDA's revised definition had no bearing on our 
YERS/QMS analysis. However, we include and discuss FDA's revised 
definition in this report as it is relevant to FDA's and DEA's 
implementation of FDASIA provisions that require coordination between 
both agencies in response to certain drug shortages. It is also 
important to note that, although DEA asserts that FDA's revised 
definition is "materially different" than its prior one, as we note in 
this report, FDA does not consider the revised definition to be 
substantively different nor does it alter how FDA works to resolve 
drug shortages. 

With respect to DEA's and FDA's different views on when a shortage 
exists, our report explicitly recognized that DEA and FDA are not able 
to effectively collaborate due to this fundamental disagreement. In 
fact, this was the basis for our recommendation that DEA and FDA take 
steps to improve the coordination required of them under FDASIA. 

Title of our draft report. DEA stated that the title of our draft 
report was inconsistent with the our findings: (1) that a causal 
relationship could not be established between shortages of drugs 
containing controlled substances and DEA's lack of timeliness in 
establishing annual and supplemental quotas and (2) that we could not 
prove that DEA's collaboration with FDA would hinder their collective 
abilities to effectively coordinate regarding future shortages. We 
disagree with DEA's conclusion that the quota process and its 
collaboration with FDA play no role in, and cannot cause or 
exacerbate, drug shortages. As this report shows, DEA does not use 
effective approaches for managing the quota process or collaborating 
with FDA. We continue to believe that DEA should improve its 
administration of the quota process and enhance coordination with FDA 
to be in a position to fulfill its statutory responsibilities. 
Nevertheless, we modified the title of this report to emphasize the 
need for process and collaboration improvements, rather than the link 
between drug shortages and the process and collaboration shortcomings. 

Method of conducting investigation and data analysis. DEA stated that 
our methodology was flawed because we did not attempt to independently 
determine whether there were shortages of drugs subject to quotas and 
whether quotas actually caused such shortages. In its comments, DEA 
stated that two of its data systems--ARCOS and YERS/QMS--would have 
provided us with such information. This stance contradicts information 
that DEA consistently provided throughout the course of our work, and 
we disagree with DEA's position for several reasons. 

* First, we sought to obtain ARCOS and YERS/QMS data from DEA early in 
our review and were met with strong opposition from agency officials. 
We requested ARCOS data in order to track the drugs that manufacturers 
reported to be in shortage to examine their movement through the 
supply chain before, during, and after the reported shortage and 
YERS/QMS data to analyze DEA's timeliness and responsiveness in 
authorizing quota. DEA officials said that ARCOS is used to monitor 
the diversion of drugs and it is not used nor intended to be used to 
identify shortages of specific drugs. DEA officials also told us that 
there is too much complexity in the supply chain to perform a 
meaningful analysis and that it would be inappropriate to use ARCOS 
data for our purpose. DEA officials also insisted that supplying us 
with ARCOS data would be burdensome and an inefficient use of their 
agency's resources. Similarly, DEA officials objected to providing us 
with the YERS/QMS data we requested. As DEA stated in its comments, 
negotiations to obtain these data were "protracted and contentious." 
Our audit was ultimately delayed for over a year as we attempted to 
obtain these ARCOS and YERS/QMS data. However, following the 
intervention of senior DOJ management officials, DEA eventually 
provided the requested information from both ARCOS and YERS/QMS. 

* Second, although DEA indicated in its comments that ARCOS and YERS/ 
QMS data could have provided us with a "full view" of the distribution 
of schedule II controlled substances throughout the supply chain, this 
is inconsistent with information the agency provided throughout the 
course of our work and what we ultimately found through our analyses. 
DEA officials repeatedly informed us that, apart from their concerns 
with providing us with the ARCOS data, they objected to our planned 
analysis because of significant and inherent inaccuracies in the data. 
Agency officials explained that, because the ARCOS data are self-
reported by registrants, the data contain many errors, particularly in 
the quantity field that records the number of packages, weight, or 
volume being reported in each transaction. DEA officials told us that 
they consider these data to be so unreliable they only use data from 
ARCOS as a preliminary tool when beginning diversion investigations, 
and that they must ultimately verify the data against source documents 
available at the registrants' business locations. DEA advised us that 
ARCOS data were unreliable; however, to maintain our independence, it 
was necessary for us to obtain a sample of data to complete our own 
assessment. 

Once we obtained the requested data from both ARCOS and YERS/QMS, we 
conducted a preliminary analysis on each data set. We identified 
significant data reliability issues with both sets of data and 
determined that both ARCOS and YERS/QMS were unreliable for our 
purposes. Specifically, regarding ARCOS, our analysis of the data 
sample identified drastically different numbers of transactions and 
amounts reported as transferred between manufacturers and distributors 
for any given quarter between 2010 through 2012. Both manufacturers 
and distributors are required to report the number of transactions of 
these substances and the amounts transferred to ARCOS, so we would 
therefore expect that the number of transactions and amounts reported 
as transferred between both groups to be similar. DEA officials could 
not provide explanations for these inconsistencies. In addition, it 
was not possible for us to use source documents to verify all of the 
transactions in our sample because it was beyond the two year 
retention requirement for registrants' original documentation for many 
of the transactions. Because our tests on the sample we obtained 
showed the ARCOS data to be unreliable for our purposes, there was no 
point in further analyzing ARCOS data or in requesting additional data. 

Our assessment of YERS/QMS also determined that these data were not 
reliable for our purposes. However, we were able to analyze a sample 
of YERS/QMS data by taking certain methodological steps. Specifically, 
to correct for the errors that we found in the data, we analyzed the 
information contained in the source documents for our sample of YERS/ 
QMS records for certain fields. Overall this sample represented about 
15 percent of the total YERS/QMS records that we obtained from DEA in 
2011 and 2012.[Footnote 84] We found a significant number of errors in 
the data from 2011 (44 percent overall) and less in the data from 2012 
(10 percent overall). Although we had correct information from the 
source documents, we determined that it was not appropriate for us to 
use all of the fields contained in those documents because we 
continued to identify inconsistencies, including in the source 
documents themselves. For instance, within the source documents 
associated with a particular supplemental application--a 
manufacturer's supplemental quota application and DEA's decision 
letter--the manufacturer may have requested the additional amount of 
quota it was seeking (e.g., an additional 500 grams in addition to 500 
grams it previously received), but DEA's corresponding decision letter 
authorized the total amount of quota for that year (e.g., 1,000 
grams). Therefore, it would appear that DEA authorized 200 percent of 
the quota requested in this case, according to the source documents, 
when in fact it authorized 100 percent of the request. Due to these 
inconsistencies, we did not pursue analyzing the amount of quota that 
DEA authorized for manufacturers in 2011 and 2012. DEA states in its 
comments that we should have analyzed the amounts of quota authorized 
by DEA for the substance classes that had drugs in shortage in 2011 
and 2012 to verify the manufacturers' claims that quotas caused or 
exacerbated shortages. However, the data reliability concerns we 
identified within the quota amount fields, particularly for 2011 data 
when 6 of the 7 quota related shortages began, prevented the 
feasibility of a meaningful analysis. 

In addition to criticizing our methodology for not fully utilizing 
data from YERS/QMS and ARCOS to independently establish causation 
between a specific shortage and the quotas DEA established for a basic 
class of schedule II controlled substance, DEA stated that our draft 
report was misleading because of our extensive description of the 
nature and magnitude of drug shortages of all controlled substances 
and their consequences. We do so because the scope of this report is 
not limited to drugs containing controlled substances that are subject 
to quota, but includes other prescription drugs containing controlled 
substances. It is important to note that we identify specific examples 
of shortages of schedule II drugs that are subject to quota--such as 
morphine, fentanyl (anesthesia), amphetamine and methylphenidate (ADHD 
drugs)--and the impact of these shortages on patients and providers in 
this report. As we note in this report, quota was not reported to be 
the primary cause of all shortages of drugs that contain schedule II 
substances. Indeed, not all drugs containing controlled substances are 
subject to quota. Other primary causes included manufacturing delays, 
capacity issues, and product quality issues, which are the same 
factors that cause shortages of drugs generally. However, because the 
availability of quota could be a factor in shortages of those drugs 
that do contain controlled substances subject to quota, we considered 
it both necessary and relevant to include such information. We 
therefore provide FDA data on the manufacturer-reported causes of 
shortages of their products containing schedule II substances, 
including quota. 

Lastly, we believe that DEA may have misconstrued the intent of our 
study. DEA implies that our purpose was to evaluate whether it 
authorized adequate amounts of quota and whether manufacturers 
provided adequate justification for their quota requests. However, our 
purpose was to examine the trends and characteristics of drug 
shortages of controlled substance and their affect on patients and 
providers, as well as assess DEA's administration of the quota 
process, including its internal controls and compliance with required 
time frames. The purpose of our study was not to determine whether 
DEA's decisions were correct or manufacturers' justifications were 
sufficient. In using ARCOS and YERS/QMS data, we sought to obtain 
information about where certain controlled substances were in the 
supply chain before, during, and after a shortage and DEA's timeliness 
and responsiveness in responding to manufacturers' quota applications. 
Data reliability concerns with both data sets prevented us from fully 
accomplishing such analyses. 

Our conclusions. DEA stated that our conclusions unfairly link the 
quota process to diminished patient care by describing the trends of 
shortages of drugs containing controlled substances and then 
evaluating DEA's administration of the quota process that only applies 
to a subset of those substances. These are two separate issues. This 
report does not say, nor do we believe, that quotas are the sole cause 
of shortages of drugs containing schedule II substances. However, we 
disagree with DEA's contention that the actions it takes in setting 
quotas at the class level would have no bearing on the drug products 
that are made with those substances. The shortcomings that we 
identified in DEA's administration of the quota process prevent it 
from having reasonable assurance that it is providing for an adequate 
and uninterrupted supply of these drugs for legitimate medical need, 
and calls into question DEA's ability to avert or address future 
shortages of medically necessary products. We agree that there are 
many factors that affect and ultimately cause drug shortages and 
include these in this report. However, simply because not all drugs 
that have been in shortage are subject to quotas does not mean that 
quotas cannot cause, contribute to, or exacerbate shortages of drugs 
containing schedule II substances. 

DEA also questioned our conclusion that by not acting promptly to 
respond to supplemental quota applications, it may hinder 
manufacturers' ability to produce schedule II drugs that may help 
prevent or resolve a shortage. DEA said this is not possible because 
manufacturers' business decisions are involved. We recognize that 
manufacturers must apply for quota, and then use it to manufacture and 
ultimately distribute the appropriate drugs for there to be any impact 
on drug shortages. However, DEA's actions also play a part when quota 
is involved. DEA's lack of timeliness in responding to manufacturers' 
supplemental quota applications--taking nearly 60 days on average--for 
both shortage and non-shortage related applications is problematic and 
not in the interest of public health. We continue to believe DEA 
should implement our recommendations to ensure that it is positioned 
to comply with the statutorily mandated time frame of responding to 
applicable requests within 30 days. 

DEA's response to our recommendations. Regarding our first 
recommendation that DEA establish controls to ensure that the YERS/QMS 
data accurately reflect both manufacturers' quota submissions and 
DEA's decisions, DEA described existing flags and reviews in place to 
ensure the information that is entered by manufacturers and DEA staff 
is accurate. While we understand that DEA takes some steps to help 
ensure the accuracy of YERS/QMS data, it does not perform sufficient 
checks to ensure the accuracy of the information. For example, DEA 
does not compare a sample of YERS/QMS data to the source documents 
(manufacturer's application and DEA decision letter) to identify if 
its existing checks are adequate. Although the accuracy of the data 
improved from 2011 to 2012, we nonetheless found a 10 percent error 
rate in the 2012 data. In particular, some of the fields with the 
highest rates of error in the 2012 data were in the fields for the 
amount of quota requested or authorized, which have implications for 
the availability of controlled substances subject to quota. We 
continue to believe that taking additional steps to reduce this error 
rate, as we recommended, is important. 

Regarding our second recommendation to establish performance measures 
related to quotas, DEA said it recognizes the value in establishing 
performance measures for personnel who review quota applications. 
While we agree that it can be useful to develop performance measures 
for personnel, our recommendation is intended to be more expansive 
than that and should include performance measures at the program or 
agency level. For example, as we note in this report, DEA could set 
goals and then measure the number of manufacturers who submit 
incomplete or revised quota applications, both factors that DEA cited 
as affecting its ability to meet its required time frames. DEA could 
also establish goals and track the percentage of annual applications 
that DEA responds to in accordance with its required time frames. It 
is important that DEA set such measures to assess its performance in 
achieving its mission of ensuring an adequate and uninterrupted supply 
of controlled substances as it does for its diversion-related mission. 

For our third recommendation about monitoring and analyzing YERS/QMS 
data, DEA said it agrees that such actions are important to ensuring 
proper administration of the quota process. DEA also provided examples 
of how the agency uses YERS/QMS data in setting APQs and bulk 
manufacturing and procurement quotas. However, the agency did not 
commit to using the data to monitor its administration of the quota 
process or its performance. Such information could be of vital 
importance to determine the feasibility of potential performance 
measures, as we recommended, or to evaluate DEA's efforts in meeting 
its performance measures once developed, such as the extent to which 
DEA is meeting its required time frames for responding to annual quota 
applications. 

In response to our fourth recommendation that DEA establish internal 
policies for processing quota applications, DEA stated that it has 
already established such policies and procedures. However, the agency 
did not provide documentation of such policies and procedures despite 
repeated requests. To the contrary, DEA officials told us that the 
agency relies exclusively on its regulations and the CSA in 
establishing APQs and manufacturers' quotas. As we state in this 
report, this is inconsistent with federal standards for internal 
control, and we believe that policies are still needed because of the 
complexity of the work involved in setting quotas. 

In response to our fifth recommendation that DEA expeditiously 
establish formal policies and procedures to facilitate coordination 
with FDA as directed by FDASIA, DEA stated that it will follow the 
requirements of FDASIA to respond to FDA's requests within 30 days. 
However, FDASIA does not specify a time frame in which DEA must 
respond to FDA's requests to increase quota; this 30-day time frame is 
only applicable to requests that DEA receives from manufacturers. 
Furthermore, DEA has not historically met its required time frames for 
establishing quotas from either the CSA or its regulations. Therefore, 
we believe that policies and procedures are needed to establish such a 
time frame and specify how DEA will respond to such a request from 
FDA. We understand that DEA has received no such request from FDA 
since the enactment of FDASIA; however, we believe that it is 
imprudent to wait for a shortage that has serious implications for 
public health before determining how to proceed. 

Regarding our sixth recommendation that DEA promptly update the MOU 
with FDA, DEA stated that it had begun negotiating the terms of the 
new information sharing agreement when this review began in 2012, 
which we note in this report. DEA said that the final version of this 
MOU has been routed for signature within the agency as of December 
2014. 

Lastly, in response to our seventh recommendation that, either in the 
MOU or in a separate agreement, DEA and FDA specifically outline what 
information the agencies will share and the time frames for sharing 
such information in response to a potential or existing drug shortage, 
DEA noted that it has already engaged in discussions with FDA to 
determine the specific procedures for sharing drug shortage related 
information pursuant to FDASIA. We believe that reaching a specific 
agreement on information sharing will be beneficial to both agencies. 

As agreed with your offices, 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 the Department of Health and Human Services and 
the Attorney General, and other interested parties. In addition, the 
report will be available at no charge on GAO's 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 at crossem@gao.gov. Contact points for 
our Office of Congressional Relations and Office of Public Affairs can 
be found on the last page of this report. Other major contributors to 
this report are listed in appendix VI. 

Signed by: 

Marcia Crosse: 
Director, Health Care: 

[End of section] 

Appendix I: List of Provider and Patient Organizations Interviewed: 

Provider Organizations: 

1. American Academy of Child and Adolescent Psychiatry: 

2. American Academy of Pediatrics: 

3. American College of Emergency Physicians: 

4. American Hospital Association: 

5. American Psychiatric Association: 

6. American Society of Anesthesiologists: 

7. American Society of Health-System Pharmacists: 

8. Association of Critical Care Transport: 

9. Institute for Safe Medication Practices: 

10. International Association of Fire Chiefs: 

11. National Association of Boards of Pharmacy: 

12. National Association of Emergency Medical Technicians: 

13. National Association of State EMS Officials: 

14. National Community Pharmacists Association: 

Patient Organizations: 

1. Children and Adults with Attention Deficit/Hyperactivity Disorder: 

2. National Alliance on Mental Illness: 

[End of section] 

Appendix II: Classes of Schedule II Controlled Substances with Medical 
Use: 

The Drug Enforcement Administration (DEA) establishes quotas on the 
quantity of schedule II substance classes that may be produced in the 
United States in any given calendar year.[Footnote 85] We consider 
classes of schedule II controlled substances with medical use to be 
those substances that are contained in Food and Drug Administration 
(FDA) approved products currently marketed for human use, and are not 
considered precursors, reference standards, or metabolites, as 
reported by DEA.[Footnote 86] Those substance classes include the 
following: 

Alfentanil; 
Amobarbital; 
Amphetamine; 
Cocaine; 
Codeine; 
Dihydrocodeine; 
Diphenoxylate; 
Fentanyl; 
Hydrocodone; 
Hydromorphone; 
Levorphanol; 
Lisdexamfetamine; 
Meperidine; 
Methadone; 
Methamphetamine; 
Methylphenidate; 
Morphine; 
Nabilone; 
Opium; 
Oxycodone; 
Oxymorphone; 
Pentobarbital; 
Remifentanil; 
Secobarbital; 
Sufentanil; 
Tapentadol. 

[End of table] 

[End of section] 

Appendix III: Scope and Methodology - Analysis of the Year-End 
Reporting and Quota Management System Data: 

As part of our work examining the Drug Enforcement Administration's 
(DEA) quota process, we analyzed data from DEA's Year-End Reporting 
and Quota Management System (YERS/QMS) for certain controlled 
substances with medical use to assess the reliability of the system 
and DEA's timeliness in responding to manufacturers' quota 
applications. This appendix provides further detail on our methods for 
determining the reliability of the data and analyzing the data. 

To determine the reliability of data from DEA's YERS/QMS, DEA provided 
data from the system for 2011 and 2012, the most recent years for 
which data were available when we began our analysis. We discussed 
these data with the officials responsible for maintaining YERS/QMS and 
examined the data for obvious errors and values outside of expected 
ranges. We removed applications that were marked as "withdrawn," as 
DEA does not authorize quota for these applications.[Footnote 87] We 
also removed applications that appeared to be duplicates, which we 
identified by searching for applications that had the same registrant, 
drug, date submitted, and quota type (e.g., bulk manufacturing or 
procurement). We searched for applications that had the word 
"duplicate" in the notes field of YERS/QMS as well, which DEA uses to 
record additional information about quota applications. Additionally, 
we removed applications that were not for schedule II substance 
classes with medical use.[Footnote 88] 

We then selected a sample of 442 records, from the total of 2,982 
records, from YERS/QMS to verify against source documents, each of 
which includes a manufacturer's quota application for a particular 
substance and the corresponding DEA decision letter informing the 
manufacturer of its quota authorization (see table 5). This sample 
included the entire population of records in 2011 and 2012 that are 
associated with drug shortages manufacturers reported to the Food and 
Drug Administration (FDA) were caused by quotas, of which there were 
146. We consider YERS/QMS records to be associated with these 
shortages if the record contains a quota application that was (a) 
submitted by a manufacturer who reported to FDA that quota caused a 
shortage of its product; (b) for a substance that was used in the 
products reportedly in shortage because of quota; and (c) was 
submitted during the time of the reported shortage caused by quota. 
For instance, if a manufacturer reported a shortage of a drug 
containing a controlled substance to FDA beginning in 2011 and ending 
in 2012, and said that the shortage was caused by quota issues, we 
considered all applications from that manufacturer for that controlled 
substance in 2011 and 2012 to be related to a shortage reportedly 
caused by a quota issue. We also selected a random sample of 296 
records stratified by year--182 and 114 records from 2011 and 2012, 
respectively. We included more records from 2011 because we believed 
those records would be harder for DEA to locate, as DEA told us that 
there were a larger percentage of hard-copy applications, as opposed 
to electronic applications, in 2011 than 2012. Because we followed a 
probability procedure based on random selections, our sample is only 
one of a large number of samples that we might have drawn. Since each 
sample could have provided different estimates, we express our 
confidence in the precision of our particular sample's results as a 
relative margin of error based on a 95 percent confidence interval 
(e.g., plus or minus 10 percentage points). This is the interval that 
would be expected to contain the actual population value for 95 
percent of the samples we could have drawn. 

Table 5: Records Sampled from the Drug Enforcement Administration's 
Year-End Reporting and Quota Management System (YERS/QMS) for Schedule 
II Substances with Medical Use: 

Records associated with shortages reportedly due to quota (2011 and 
2012): 
Number of records in population: 146; 
Number of records in sample: 146. 

Records not associated with shortages reportedly due to quota (2011): 
Number of records in population: 1,419; 
Number of records in sample: 182. 

Records not associated with shortages reportedly due to quota (2012): 
Number of records in population: 1,417; 
Number of records in sample: 114. 

Total: 
Number of records in population: 2,982; 
Number of records in sample: 442. 

Source: GAO analysis of YERS/QMS data. GAO-15-202. 

[End of table] 

From our sample of 442 YERS/QMS records for which we requested source 
documents from DEA, we analyzed source documents for 440 records. DEA 
officials said that they could not locate documents for one record and 
another record was found to be ineligible based on our review. 
Noncertainty sample sizes were inflated to account for an expected 
location rate of 50 percent for 2011 and 80 for 2012, to assure enough 
located files would be available for analysis. Because DEA told us 
that some paper applications had been put into storage due to office 
renovations and were difficult to retrieve, we assumed we would not be 
able to locate the full sample of records. DEA officials told us that 
there were more paper records--as opposed to electronic records--
submitted in 2011 than 2012. Based on that information, we assumed a 
lower location rate in 2011 than 2012. In order to estimate population 
quantities, we analyzed our sample data using survey analysis software 
that accounts for our stratified random selection design and uses 
analysis weights in order to make the results of our sample 
representative of all YERS/QMS applications for schedule II substances 
with medical use in 2011 and 2012. 

In order to confirm that we were comparing the source documents to the 
correct record in YERS/QMS, we matched the documents based on the 
manufacturer and substance, and at least one of the following fields: 
date application was submitted; total amount requested; date decision 
letter was mailed; and total amount authorized.[Footnote 89] Using 
this methodology, we were able to match 440 records in our sample. To 
verify the accuracy of the YERS/QMS data for these 440 records, we 
compared the data in YERS/QMS against the information in DEA's source 
documents (quota applications and decision letters). We used the 
following fields in the quota application to validate or correct the 
data in YERS/QMS: quota year, company name, substance name, quota type 
(e.g., bulk manufacturing or procurement), the date submitted, and 
total amount requested. We used the following fields in the decision 
letters provided by DEA to validate or correct the data in YERS/QMS: 
quota year, company name, substance name, quota type (e.g., bulk 
manufacturing or procurement), date mailed, and total amount 
authorized. We calculated the number of fields with errors (e.g., 
number of errors in the date submitted field) and the number of 
records with errors (i.e., a record with an error in any field). We 
also weighted the results of our error calculations in order to make 
the results representative of all YERS/QMS records in 2011 and 2012. 
Based on our review of these source documents, we determined that the 
YERS/QMS data we obtained from DEA were unreliable for our purposes. 

Because of the reliability concerns with the YERS/QMS data, we used 
the information contained in the 440 YERS/QMS source documents for our 
analysis of DEA's timeliness in responding to manufacturers' quota 
applications and when manufacturers submitted annual quota 
applications. In order to separate annual applications and 
supplemental applications, which have different requirements for 
timeliness, we considered annual applications to be applications 
submitted in the year prior to when the quota would be used (e.g., an 
application submitted in May 2010 for 2011 quota). Applications 
submitted in the same year as when the quota would be used were 
considered supplemental. For annual applications, we analyzed DEA's 
compliance with the required deadlines in the Controlled Substances 
Act and DEA's regulations, including how often DEA met the deadlines 
for our sample of records and how many days passed when it did not 
meet the deadlines.[Footnote 90] We also analyzed when manufacturers 
submitted their annual quota applications. For supplemental 
applications, we analyzed the amount of time that it took DEA to 
respond to supplemental applications in our sample. We used a two-
tailed t-test to determine whether there was a statistical difference 
in the amount of time DEA took to respond to supplemental applications 
submitted by manufacturers that reported shortages due to quota 
compared to all other supplemental applications. For our analysis of 
these 440 YERS/QMS source documents, we express our confidence in the 
precision of our particular sample's results the same way in which we 
did for our analysis of the errors in the YERS/QMS data, as we 
described above. 

[End of section] 

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

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

December 29, 2014: 

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, "Controlled Substances: Better Management of 
the Quota Process and Enhanced Coordination between DEA and FDA Needed 
to Address Drug Shortages" (GAO-15-202). 

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

Sincerely, 

Signed by: 

Jim R. Esquea: 
Assistant Secretary for Legislation: 

Attachment: 

General Comments Of The Department Of Health And Human Services (HHS) 
On The Government Accountability Office's Draft Report Entitled: 
Controlled Substances: Better Management Of The Quota Process And 
Enhanced Coordination Between DEA and FDA Needed To Address Drug 
Shortages (GAO-15-202): 

The U.S. Department of Health and Human Services (HHS) appreciates the 
opportunity to review and comment on this draft report. 

GAO Recommendation: 

To strengthen the Drug Enforcement Administration's (DEA) and the U.S. 
Food and Drug Administration's (FDA) ability to respond to shortages 
of drugs containing controlled substances, we recommend that the 
Administrator of DEA and the Commissioner of FDA promptly update the 
memorandum of understanding (MOU) between the two agencies and either 
in the MOU or in a separate agreement, specifically outline what 
information the agencies will share, and timeframes for sharing such 
information, in response to a potential or existing drug shortage. 

HHS Response: 

HHS concurs with this recommendation and is actively working to 
finalize the new MOU with the DEA. 

[End of section] 

Appendix V: Comments from the Drug Enforcement Administration: 

U.S. Department of Justice: 
Drug Enforcement Administration: 
[hyperlink, http://www.dea.gov] 

December 29, 2014: 

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

Dear Ms. Crosse: 

The Drug Enforcement Administration (DEA) provides the following 
comments to the Government Accountability Office (GAO) report entitled 
"Controlled Substances: Better Management of the Quota Process and 
Enhanced Coordination between DEA and FDA Needed to Address Drug 
Shortages" (GAO-15-202). 

It is important to note that the titled conclusion of the report is 
inconsistent with the GAO finding that it cannot establish either a 
"causal relationship between shortages of drugs containing controlled 
substances and DEA's management of the quota setting process" (Draft, 
p. 45) or that DEA coordination with the U.S. Food and Drug 
Administration (FDA) adversely affected the availability of drug 
products containing controlled substances.[Footnote 1] 

Introduction: 

The DEA agrees that prescription drug abuse is a nationwide epidemic 
and more must be done to prevent, detect, and deter the diversion of 
pharmaceutical controlled substances that supply drug addiction and 
abuse. The DEA role in this effort is as the primary agency 
responsible for coordinating the drug law enforcement activities of 
the United States. The Diversion Control Program (DCP) is a strategic 
component of the DEA's law enforcement mission. The DEA Office of 
Diversion Control administers the DCP and implements and enforces 
Titles H and III of the Comprehensive Drug Abuse Prevention and 
Control Act of 1970, as amended. 21 U.S.C. 801-971. Titles II and HI 
are referred to as the "Controlled Substances Act" and the "Controlled 
Substances Import and Export Act," respectively, but are collectively 
referred to as the "Controlled Substances Act" or the "CSA." The CSA 
and its implementing regulations are designed to prevent, detect, and 
deter the diversion of controlled substances and listed chemicals into 
the illicit market while establishing the total quantity of each basic 
class[Footnote 2] of controlled substance in schedules I and II and 
for ephedrine, pseudoephedrine, and phenylpropanolamine to be 
manufactured each year to provide for the estimated medical, 
scientific, research, and industrial needs of the United States, for 
lawful export requirements, and for the establishment and maintenance 
of reserve stocks. This is a delicate balance. 

One way that DEA provides for the estimated medical, scientific, 
research, and industrial needs of the United States is to establish an 
aggregate, nationwide quota for each basic class of schedule I and II 
controlled substance (referred to as the Aggregate Production Quota, 
or "APQ") and to authorize individual quotas (referred to as 
manufacturing quota and procurement quota). It is important to 
understand that DEA authorizes quota only at the manufacturer level 
for those entities that manufacture active pharmaceutical ingredients 
(API), those entities that manufacture substances into dosage forms, 
and those entities that repackage or re-label drug products that 
contain schedule I or II controlled substances. Once the aggregate 
quota is established and a particular manufacturer is authorized to 
manufacture a specific amount of a basic class of controlled 
substance, DEA cannot require the manufacturer to manufacture API or a 
specific drug Product, or distribute that substance down through the 
supply chain. Furthermore, a bulk manufacturer may extract or 
synthesize API in an authorized calendar year, and hold it in 
inventory until any subsequent calendar year. Of equal importance, the 
CSA prohibits DEA from establishing quotas in terms of individual 
pharmaceutical dosage forms prepared from or containing such a 
controlled substance. 21 U.S.C. § 826(a). These limitations are 
critical to understanding the effect that quota can have on the 
availability of a specific drug product at the retail level or at the 
emergency medical service (EMS) provider level. The failure to 
appreciate these limitations is the fatal flaw in the GAO report. 

Another fundamental weakness in the GAO report is the failure to 
account for the fact that "shortage" means different things to 
different entities, and without accounting for this distinction, the 
GAO report is misleading with respect to the effect that the DEA quota 
process can have on patient care. To identify trends in shortages of 
drugs containing controlled substances, GAO analyzed University of 
Utah Drug Information Service (UUDIS) data. UUDIS broadly defines a 
"shortage" as a supply issue that affects how pharmacies prepare and 
dispense a product or that influences patient care when prescribers 
must choose an alternative therapy because of a supply issue. A UUDIS 
"critical shortage" occurs when alternative medications are 
unavailable, the shortages affect multiple manufacturers, or the 
shortages are widely reported. In addition, UUDIS information is based 
on national drug codes (NDCs) rather than the basic class of 
controlled substance contained within a specific drug product. NDCs 
are identifiers that are unique to a particular manufacturer, drug 
product, dosage form, dosage strength, and package size. Accordingly, 
a single basic class of controlled substance will be represented by 
many different NDCs. Statistics and analysis based on the NDC, rather 
than the basic class of controlled substance, could dramatically 
distort the actual number of shortages that could be attributed to 
quota. 

From September 2006 through July 2012, the FDA defined a "drug 
shortage" as a "situation in which the total supply of all clinically 
interchangeable versions of an FDA-regulated drug is inadequate to 
meet the current or projected demand at the user level."[Footnote 3] 
(FDA CDER MAPP 6003.1, Sept. 26, 2006). FDA changed their definition 
in September 2014 to align with the definition in the 2012 Food and 
Drug Administration Safety and Innovation Act (FDASIA) to "a period of 
time when the demand or projected demand for the drug within the 
United States exceeds the supply of the drug." Although this new 
definition is materially different from the definition applicable 
during the period in review (2011 and 2012), the GAO report uses the 
2014 definition in its analysis. 

In contrast to the UUDIS and FDA, which view shortages in the context 
of patient-level availability, DEA views shortages in the context of 
manufacturer-level quota. Accordingly, a shortage within DEA's 
jurisdiction is the lack of sufficient quota available for bulk or 
dosage form manufacturers to manufacture a basic class of a schedule I 
or II controlled substance. This perspective is the result of the CSA 
prohibition against establishing quotas in terms of individual 
pharmaceutical dosage forms prepared from or containing a controlled 
substance, together with the mandate to limit the supply of controlled 
substances available for diversion, and the inability to require a 
manufacturer to manufacture a specific API or drug product or require 
the distribution of controlled substance drug products downstream. 
Accordingly, if alternative forms of a drug product are available 
(e.g., brand, generic, or other clinically interchangeable drug 
products), or if manufacturers have quota authorization, the DEA 
cannot remedy any patient-level shortage (e.g., inadequate supply at 
the retail level or EMS provider level) by increasing the aggregate or 
authorizing more quota at the manufacturer level. GAO fails to 
emphasize and account for these fundamental distinctions in its review 
of the potential effect of the quota process on the availability of 
drug products containing schedule II controlled substances. 

Some simplified examples can illustrate how the above distinctions can 
skew shortage conclusions. If there is an unmet patient need for 
hydromorphone 10 mg/mL, 1 mL vials, UUDIS and FDA would qualify it as 
a shortage. However, DEA would not consider there to be a shortage 
within its jurisdiction if the 10 mg/mL, 5 mL vials are available. 
Similarly, DEA would not consider there to be a shortage if the brand 
name version of a particular drug product is unavailable, if the 
generic version of the drug product is available. Also, there would be 
no shortage in DEA's jurisdiction if there is hydromorphone quota 
available in the APQ (i.e., the annual APQ has not been exhausted) and 
manufacturers are not requesting additional quota, or if manufacturers 
with authorized hydromorphone quota are not manufacturing their quota 
allotment or are not distributing the manufactured hydromorphone 
downstream. 

Further comments regarding the GAO report are focused on the following 
three main areas and are discussed below: method of conducting 
investigation; method of data analysis; and GAO conclusions. 

Method of conducting investigation: 

Generally, the Congressional requesters sought to "better understand 
the impact of DEA quotas. on patients with emergency medical and 
critical care conditions and traumatic injuries, and the extent to 
which DEA policies and regulations may impede the ability of 
physicians and health care providers to mitigate a shortage of a drug 
on any of the applicable schedules." Specifically, GAO was asked to 
particularly focus on shortages of drugs containing controlled 
substances used by EMS providers and to treat attention deficit 
hyperactivity disorder (ADHD). GAO did not evaluate the flow of 
specific controlled substances from the point of quota request and 
authorization to manufacture, or from the point of manufacture to 
distribution to the retail level, using available data from the 
Automation of Reports and Consolidated Orders System (ARCOS) and Year 
End Reporting and Quota Management System (YERS/QMS). 

GAO was aware of which manufacturers self-reported shortages to FDA 
claiming the shortage was due to quota, and which specific drug 
products were reported in shortage due to quota, yet GAO did not 
investigate each manufacturer's quota allotment or usage, the 
manufacturer's manufacturing and distribution practices, whether the 
manufacturer provided adequate justification for quota, or if the 
manufacturer asked for quota before or after reporting the shortage. 
In fact, GAO did not attempt to determine whether "shortages" actually 
existed because of a lack of quota. GAO was not without the tools to 
determine the answer to this question. After a protracted and 
contentious negotiation regarding the data that DEA could release to 
GAO to conduct this investigation, DEA and GAO reached an agreement 
wherein DEA provided specified information from ARCOS and YERS/QMS. 
ARCOS and YERS/QMS data could have provided GAO with a full view of 
the distribution of schedule II controlled substances by manufacturers 
and distributors, from the point of bulk manufacture, to dosage form 
manufacture, and down to distribution to the retail level. Instead, 
GAO simply reported manufacturers' anecdotal complaints about the 
quota process' effect on shortages. 

Amphetamine is a schedule II controlled substance and is used to treat 
ADHD, among other things, and it was reported to be in shortage during 
the review period. GAO requested ARCOS information pertaining to 17 
specific NDCs, all of the amphetamine basic class. However, the 17 
requested NDCs represented only a small fraction of the available 
market supply. The DEA estimated that there were 48 other amphetamine-
containing drug products, 25 of which were manufactured and 
distributed during the period under review. GAO did not request ARCOS 
data on the other available amphetamine-containing products on the 
market, and GAO did not discuss in its report any findings relative to 
the 17 requested NDCs. The ARCOS information, combined with 
information from YERS/QMS, is crucial to determining whether 
sufficient API was manufactured, whether the API was distributed 
downstream by bulk manufacturers, whether dosage form manufacturers 
were manufacturing drug products in accordance with their quota 
applications, whether dosage form manufacturers were distributing drug 
products downstream and if so, where, and whether controlled 
substances were being held at the manufacturer level or destroyed 
rather than placed into the supply chain. 

DEA is confident that a review of ARCOS and YERS/QMS data would have 
established that DEA' s administration of the quota process did not 
cause or exacerbate any shortages of drug products used to treat ADHD 
in 2011. In 2011, DEA increased the APQ for amphetamine salts by 6,700 
kg. A review of the ARCOS and YERS/QMS data for amphetamine salts 
showed that manufacturers subsequently requested, and DEA authorized, 
only a very small percentage of this increase. In addition, a 
significant number of amphetamine dosage units were destroyed 
throughout 2011, as well as a substantial amount of raw material, and 
millions of dosage units of ADHD drug treatment products still 
remained at the distributor and retail level at the end of 2011. 

Close review of the ARCOS data would have also refuted manufacturers' 
assertions about the effects of DEA' s timing to establish quotas. For 
example, manufacturer representatives reported to GAO that the 
timeline for establishing quotas does not provide manufacturers with 
enough time to plan for production and order the raw material or API 
needed to start manufacturing their products at the beginning of the 
production year. Representatives reported to GAO that they operated 
solely with what is left in their inventory for the first few months 
of the production year, "which may be limited because manufacturers 
operate in a lean manufacturing environment where they carry as little 
inventory as possible." (Draft, p. 30). This statement from 
manufacturers is conflicting. Manufacturers complained that they do 
not have sufficient inventory because of quota and must operate on 
what is solely remaining in inventory, but then go on to state their 
business choice to operate in a lean environment where they carry as 
little inventory as possible. Even so, manufacturers may manufacture 
API and procure raw material at any time during the year, and not 
distribute it until the next calendar year because DEA regulations 
provide for an inventory allowance. 

In addition, a review of ARCOS data would have been critical to 
determining whether the DEA's processing of supplemental quota 
applications in 2011 caused or exacerbated FDA-reported drug 
shortages, as alleged by manufacturers. (Draft, p. 30-31). GAO's 
probability sample of YERS/QMS source documents showed that it took 
DEA an average of 58 days to respond to supplemental quota 
applications in 2011 and 2012. (Draft, p. 30-31). GAO also reported 
that it took DEA 10 days longer, on average, to respond to 
supplemental quota applications submitted by manufacturers that 
reported shortages caused by quota in 2011. However, as discussed 
above, a review of ARCOS and YERS/QMS information would have 
established that amphetamine manufacturers only sought authorization 
to manufacture a very small percentage of the mid-year increase. 

Method of Data Analysis: 

The report's extensive description of the nature and magnitude of 
shortages (Draft, p. 17) is misleading as it uses a very broad 
definition of "shortage," using data from two different sources to 
quantify and explain the consequences of shortages, and then ties 
these consequences to the very small number of schedule II drug 
product "shortages" without ever establishing causation between the 
specific shortage and the quotas for the specific basic class of 
controlled substance. 

GAO found that approximately 10% (168 of 1,575) of the UUDIS shortages 
from January 2001 through June 2013 involved drug products containing 
a controlled substance (Draft, p. 17); of these, 57% (96) involved 
drug products containing schedule II controlled substances (Draft, p. 
19), or approximately 6% (96 of 1,575) of the total number of UUDIS 
shortages. Because UUDIS information is presented according to NDC 
rather than the basic class, the results (96 shortages of schedule II 
controlled substances from 2001 to 2013) can dramatically distort the 
actual number of shortages that could have been attributed to lack of 
quota in a particular basic class of controlled substance. The results 
can also be misleading because UUDIS counts a shortage as a period of 
time; as a result, 45 different drugs containing controlled substances 
were reported to be in shortage multiple times from January 2001 
through June 2013, representing 143 individual shortages. (Draft, p. 
19). The data could also be distorted by the fact that GAO analyzed 
data from YERS/QMS for 2011 and 2012, rather than 2001 to 2013. 
Analyzing the information regarding the specific drug products and the 
specific basic class of controlled substance represented by the 96 
NDCs, as well as the calendar year that the substances were reported 
in shortage would have helped to determine the role, if any, that 
quota played in any shortage. 

GAO also reported that critical shortages represented 52% (87 of 168) 
of all shortages of drugs containing controlled substances. (Draft, p. 
4, n.6; p. 21). Of the 87 shortages containing controlled substances 
identified as critical by UUDIS from January 2001 through June 2013, 
half (44 of 87) involved pain relievers (analgesics). (Draft, p. 20-
21). Analgesics can be controlled in schedule II, III, IV, or V. 
However, GAO does not state whether these products contained schedule 
II controlled substances subject to quota, or schedule III through V 
controlled substances not subject to quota. This information, along 
with the NDCs and basic class of controlled substance involved, would 
be important in determining the role, if any, quota played in any 
shortage, particularly with respect to the UUDIS "critical shortages," 
because the applicable criteria (alternative medications are 
unavailable, the shortages affect multiple manufacturers, or the 
shortages are widely reported) are more likely to implicate quota than 
a standard shortage (a supply issue that affects how pharmacies 
prepare and dispense a product or that influences patient care when 
prescribers must choose an alternative therapy because of a supply 
issue). 

Even so, for the period January 2010 to June 2013, GAO reported that 
there were 40 FDA-reported shortages of drug products containing 
schedule II controlled substances, and of those, only seven were 
alleged to have been caused or exacerbated by quotas. (Draft, p. 30). 
The remaining 33 reported shortages of drugs containing schedule H 
controlled substances were caused by other factors that cause 
shortages of drugs generally such as manufacturing delays, capacity 
issues, and product quality issues. (Draft, p.30). GAO does not state 
whether any of these seven shortages occurred during the period in 
review, 2011 and 2012, nor does GAO indicate which basic class of 
controlled substance was involved, or whether each shortage involved a 
different basic class of controlled substance or if a single basic 
class of controlled substance was involved in several reported 
shortages. However, GAO contacted the 10 manufacturers that reported 
the seven shortages from January 2010 to June 2013, and reported that 
"many" manufacturers stated DEA's lack of timeliness in establishing 
quotas caused or exacerbated shortages of their drug products. It does 
not appear that GAO verified these statements with the data it 
obtained from YERS/QMS or ARCOS. Rather, the cause of these self-
reported shortages was substantiated by collecting anecdotal 
information from manufacturers. 

Some drug products specifically mentioned in the report were in 
shortage due to reasons other than quota. For example, beginning in 
2010, a major manufacturer of injectable drug products containing 
controlled substances voluntarily shut down certain of its production 
lines and slowed the release of products in certain manufacturing 
facilities as a result of certain quality issues cited by the FDA. 
Such interruptions adversely impacted, and continue to adversely 
impact, the manufacturer's ability to manufacture and sell its 
products. The availability of all injectables were adversely affected, 
including substances specifically mentioned in the GAO report as 
having significant deleterious effects on patient care as a result of 
shortage, such as fentanyl, hydromorphone, and morphine—all schedule 
II substances subject to quota. Review of the quota data would have 
shown that when new manufacturers submitted quota applications to meet 
the new demand, DEA verified. with FDA the supply disruption and acted 
quickly to authorize quota to the new manufacturers. 

In addition, some drug products emphasized by GAO when it reported the 
effects of drug shortages on treatment and patient care were not drug 
products subject to quota. For example, GAO references an American 
Society of Anesthesiologists' survey regarding the effects of drug 
shortages on anesthesiologists. (Draft, p.24, n.49). The highest 
frequency of reported shortages were fentanyl (66%), thiopental (40%), 
succinylcholine (21%), propofol (19%), and pancuronium (15%). As 
discussed above, fentanyl shortages were due to manufacturing issues. 
Thiopental is a schedule III controlled substance and thus not subject 
to quotas; and the remaining three substances are non-controlled 
substances. In another example, lorazepam injection is a schedule IV 
controlled substance, and GAO highlighted the adverse consequences of 
its shortage, indicating that a single shortage of it lasted slightly 
more than 5 years. (Draft, p.18). Another consideration GAO ignores is 
that the benzodiazepines are primarily imported and not manufactured 
in the United States. Finally, GAO reports that oxycodone oral 
solution (Draft, p.19), a drug GAO reports is used to treat moderate 
to severe pain, was in shortage for the longest combined amount of 
time from January 2001 through June 2013. However, certain oxycodone 
oral solution drug products were not FDA-approved drugs and could not 
be lawfully manufactured or distributed until FDA approval in 
September, 2014. 

GAO Conclusions: 

Failure to utilize the available information as discussed above, and 
failure to evaluate and analyze the causes of specific controlled 
substance shortages lead to an analysis that unfairly linked the quota 
process to diminished patient care. Even though GAO could not find 
that shortages occurred because of a lack of quota or because of DEA's 
administration of the quota process, GAO makes several inferences 
about a relationship between drug shortages and the quota process. 
This was accomplished because GAO begins its report with identifying 
trends in shortages from January 2001 through June 2013, and then 
examines DEA's administration of the quota process, thereby suggesting 
the effect of the quota process on shortages. However, GAO only 
evaluated quota data for 2011 and 2012, and its evaluation is not 
generalizable to other years. 

As discussed above, only seven of 40 FDA-reported shortages of drug 
products containing schedule II controlled substances were alleged to 
have been caused or exacerbated by quotas. GAO reports that it cannot 
confirm that DEA's lack of timeliness caused or exacerbated shortages. 
However, the tools were available to GAO to refute the specific claims 
that DEA's administration of the quota process caused or exacerbated 
shortages. 

DEA is confident that its administration of the quota process did not 
affect a shortage during the period in review because drug product 
shortages are not limited to products that contain schedule II 
controlled substances. In fact, for the period January 2010 to June 
2013, only 18% (7 of 40) of FDA-reported schedule H drug product 
shortages implicated quotas. Also, UUDIS data shows that from January 
2001 through June 2013, approximately 43% of all reported controlled 
substance shortages were present in schedule III through V drug 
products, where quotas are not involved. (Draft, p.19). In addition, 
GAO found that, from January 2001 through June 2013, the number of new 
controlled substance shortages reported each year peaked in 2009 and 
then declined. (Draft, p. 17-18; fig. 2). The increase in these 
shortages mimics the pattern found for shortages of all drugs, 
indicating that the same factors affecting shortages of all drugs are 
also the same factors affecting shortages of drugs containing 
controlled substances. It is more likely that a common denominator (or 
a combination of common denominators) are effecting the similar 
patterns in shortages amongst controlled substances and non-controlled 
substances; as well as amongst schedule II controlled substances and 
schedule III through V controlled substances. 

GAO concluded that by not acting "promptly" on supplemental 
applications, DEA may hinder manufacturers' ability to manufacture 
schedule II drugs that may help prevent or resolve a shortage. 
However, as explained above, even if DEA increased the APQ or 
authorized additional manufacturing or procurement quota, 
manufacturers must apply for it and actually use it to manufacture the 
drug products in shortage, and then distribute those products 
downstream—activities that DEA cannot compel. 

DEA Response to Recommendations: 

GAO Recommendation (1): Establish controls, such as periodic data 
checks, to ensure that the YERS/QMS data accurately reflect both 
manufacturers' quota submissions and DEA's decisions. 

Response: The GAO report found that the data error rate was 
substantially improved from the initial year the process became 
electronic to the second year (2011 to 2012), dropping from 45% to 
10%. (Draft, p. 35). DEA has established policies and procedures to 
ensure data is accurate. In order to determine the timeliness of 
responses to submitted requests, there are a number of computer-
generated dates, including date submitted, date assigned for review, 
and date review complete. In order to determine accuracy in quota 
values being requested and granted there are a series of system-
generated flags in YERS/QMS. The flags guide and verify data provided 
by applicants; and there are flags for internal review, including when 
a quantity greater than requested is entered. Managers review 
worksheets for accuracy in summarizing the analysis of the data and 
supporting documentation provided by the applicant. They then verify 
that the values contained in the working documents are accurately 
entered into YERS/QMS. Upon final authorization, managers close the 
application in YERS/QMS after ensuring that the dates mailed are 
entered as the authorization letters are scanned and sent to the 
applicant (via email and U.S. Postal mail). YERS/QMS has a flag to 
ensure that the date entered is correct. 

GAO Recommendation (2): Establish performance measures for DEA related 
to quotas and ensuring an adequate and uninterrupted supply of 
controlled substances for legitimate medical use. 

Response: DEA recognizes the value in establishing performance 
measures for personnel reviewing quota applications and will determine 
whether performance is measurable with regard to processing quotas. 
Many factors determine how quickly and how accurately a quota 
application is reviewed and a quota recommended. For example, a single 
quota application is for one specific basic class; however, each quota 
request may involve quota for more than one specific drug product 
containing that basic class. The reasonable amount of time to evaluate 
each application is highly dependent on how many different factors 
affect a single request. 

GAO Recommendation (3): Monitor and analyze YERS/QMS data to assess 
DEA's administration of the quota process. 

Response: DEA agrees that monitoring and analyzing YERS/QMS data is 
important to ensuring proper administration of the quota process. The 
YERS/QMS data are integrally related for manufacturing and procurement 
quotas applications and responses. The data are reviewed and monitored 
constantly when analyzing each quota application. For example, with 
the APQ set as the maximum of each basic class to be manufactured each 
year, the quota review process of every manufacturing quota 
application checks the APQ, amounts issued, pending and remaining. In 
addition, the manufacturing quota data are analyzed and used with 
other sources to establish and revise the annual APQs. 

GAO Recommendation (4): Establish internal policies for processing 
quota applications and setting aggregate, annual, and supplemental 
quotas to ensure that staff conduct these activities consistently and 
in accordance with the CSA and agency's regulations. 

Response: DEA has established policies and procedures for staff 
administering the quota procedures. In addition, beginning with 2013 
APQs, DEA included an additional 25% of the estimated medical, 
scientific, and research needs as part of the amount necessary to 
ensure the establishment and maintenance of reserve stocks. DEA 
expects that maintaining this reserve in the aggregate production 
quotas will mitigate adverse public affects if an unforeseen event 
results in substantial disruption to the amount of controlled 
substances available to provide for legitimate public need. 

GAO Recommendation (5): Expeditiously establish formal policies and 
procedures to facilitate coordination with FDA as directed by FDASIA, 
including a specific timeframe in which DEA will be expected to 
respond to FDA requests to expedite shortage-related quota 
applications. 

Response: DEA shall follow the requirements of FDASIA to respond to 
requests within 30 days. It should be noted that no special requests 
for expedited quota review have been forwarded to DEA since enactment 
of FDASIA in July, 2012. As previously conveyed to GAO, DEA and FDA 
began negotiating the terms of a new information sharing agreement 
before this engagement commenced. As of December 15, 2014, the final 
memorandum of agreement has been routed for signature within DEA. 

GAO Recommendation (6): Promptly update the MOU between the two 
agencies. 

Response: As previously conveyed to GAO, DEA and FDA began negotiating 
the terms of a new information sharing agreement before this 
engagement commenced. As of December 15, 2014, the final memorandum of 
agreement has been routed for signature within DEA. 

GAO Recommendation (7): Either in the MOU or in a separate agreement, 
specifically outline what information the agencies will share, and 
timeframes for sharing such information, in response to a potential or 
existing drug shortage. 

Response: As previously conveyed to GAO, DEA and FDA began negotiating 
the terms of a new information sharing agreement before this 
engagement commenced. As of December 15, 2014, the final memorandum of 
agreement has been routed for signature within DEA. DEA and FDA have 
engaged in discussions to determine the specific procedures by which 
information regarding drug shortages shall be exchanged, pursuant to 
FDASIA. These procedures will be memorialized in a mutually agreeable 
workplan. 

Conclusion: 

There can be no doubt that drug shortages adversely affect the public 
health. Drug shortages occur across the continuum of pharmaceutical 
characteristics, e.g., brand, generic, controlled, non-controlled, 
over-the-counter, dosage forms and dosage strengths, analgesics, 
sedatives, stimulants. Shortages can be caused by a variety of 
factors, as GAO previously reported in 2011 and 2014. Determining the 
relationship between retail and EMS level drug product shortages and 
manufacturing quota is a multifaceted undertaking that particularly 
requires an understanding of controlled substance manufacturing and 
distribution practices, an appreciation of how competitive contractual 
agreements affect the actions of manufacturers, distributors, and 
patent owners, and how these dynamics influence the annual forecasting 
of quota need. 

DEA remains committed to establishing production quotas for each basic 
class of controlled substance in schedule II to be manufactured each 
year to provide for the estimated medical, scientific, research, and 
industrial needs of the United States, for lawful export requirements, 
and for the establishment and maintenance of reserve stocks. 
Accordingly, DEA appreciates the GAO finding that it cannot establish 
a causal relationship between shortages of drugs containing controlled 
substances and DEA' s management of the quota setting process. 

Should you have any questions regarding this matter or our comments, 
please contact Michael A. Dixon, Acting Deputy Chief Inspector, Office 
of Inspections, at (202) 307-4007. 

Sincerely, 

Signed by: 

Joseph T. Rannazzisi: 
Deputy Assistant Administrator: 
Office of Diversion Control: 

cc: 
Richard P. Theis: 
Director, Audit Liaison Group: 
Internal Review and Evaluation Office: 
Justice Management Division: 

Comment letter footnotes: 

[1] See Draft, p. 34 (stating GAO "cannot confirm whether DEA's lack 
of timeliness in establishing annual and supplemental quotas has 
caused or exacerbated shortages"); p. 40 (stating "DEA and FDA have 
not established a sufficiently collaborative relationship. which could 
hinder their abilities to effectively coordinate future shortages") 
(emphasis added). 

[2] "Basic class" means, as to controlled substances listed in 
Schedules I and II: 
(1) Each of the opiates, including its isomers, esters, ethers, salts, 
and salts of isomers, esters, and ethers whenever the existence of 
such isomers, esters, ethers, and salts is possible within the 
specific chemical designation, listed in § 1308.11(b) of this chapter;
(2) Each of the opium derivatives, including its salts, isomers, and 
salts of isomers whenever the existence of such salts, isomers, and 
salts of isomers is possible within the specific chemical designation, 
listed in §1308.11(c) of this chapter;
(3) Each of the hallucinogenic substances, including its salts, 
isomers, and salts of isomers whenever the existence of
such salts, isomers, and salts of isomers is possible within the 
specific chemical designation, listed in §1308.11(d) of this chapter;
(4) Each of the following substances, whether produced directly or 
indirectly by extraction from substances of vegetable origin, or 
independently by means of chemical synthesis, or by a combination of 
extraction and chemical synthesis: 
(i) Opium, including raw opium, opium extracts, opium fluid extracts, 
powdered opium, granulated opium, deodorized opium and tincture of 
opium; (ii) Apomorphine; (iii) Codeine; (iv) Etorphine hydrochloride;
(v) Ethylmorphine; (vi) Hydrocodone; (vii) Hydromorphone; (viii) 
Metopon; (ix) Morphine; (x) Oxycodone; (xi) Oxymorphone; (xii) Thebaine;
(xiii) Mixed alkaloids of opium listed in §1308.12(b)(2) of this 
chapter; (xiv) Cocaine; and; (xv) Ecgonine; 
(5) Each of the opiates, including its isomers, esters, ethers, salts, 
and salts of isomers, esters, and ethers whenever the existence of 
such isomers, esters, ethers, and salts is possible within the 
specific chemical designation, listed in §1308.12(c) of this chapter; 
and; 
(6) Methamphetarnine, its salts, isomers, and salts of its isomers;
(7) Amphetamine, its salts, optical isomers, and salts of its optical 
isomers;(8) Phenmetrazine and its salts;
(9) Methylphenidate;
(10) Each of the substances having a depressant effect on the central 
nervous system, including its salts, isomers, and salts of isomers 
whenever the existence of such salts, isomers, and salts of isomers is 
possible within the specific chemical designation, listed in 
§1308.12(e) of this chapter. 21 C.F.R. § 1300.01(b). 

[3] The GAO report discusses DEA's concern with the reliability of 
information posted on FDA's drug shortage website. During the period 
under review, DEA was concerned that FDA did not adhere to the 
applicable definition of "drug shortage" because FDA was listing drug 
products in shortage when clinically interchangeable drug products 
were available. Other inaccuracies included reporting that some 
distributors experienced a shortage due to quota or lack of API, yet 
distributors do not receive quota. 

[End of section] 

Appendix VI: 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, Geri Redican-Bigott, Assistant 
Director; Zhi Boon; Jessica Farb; Cathleen Hamann; Rebecca 
Hendrickson; Tom Jessor; Eileen Larence; Jan Montgomery; Lisa Motley; 
Leslie Powell; Dan Ries; Janet Temko-Blinder; Sonya Vartivarian; and 
Eric Wedum made key contributions to this report. 

[End of section] 

Footnotes: 

[1] A controlled substance is one that is included in one of five 
schedules under the Controlled Substances Act. A controlled substance 
is placed in a respective schedule based on whether it has a currently 
accepted medical use in the United States and its potential for abuse 
and physical or psychological dependence. 21 U.S.C. §§ 802(6), 812. 

[2] We have previously reported on trends in the number of drug 
shortages and types of drugs most often in shortage, the causes of 
such shortages, and FDA's ability to respond to them. GAO, Drug 
Shortages: FDA's Ability to Respond Should Be Strengthened, 
[hyperlink, http://www.gao.gov/products/GAO-12-116] (Washington, D.C.: 
Nov. 21, 2011), and Drug Shortages: Public Health Threat Continues, 
Despite Efforts to Help Ensure Product Availability, [hyperlink, 
http://www.gao.gov/products/GAO-14-194] (Washington, D.C.: Feb. 10, 
2014). 

[3] In addition, DEA is also required to provide for lawful export 
requirements, and for the establishment and maintenance of reserve 
stocks. 21 U.S.C. § 826; 21 C.F.R. § 1303.11. 

[4] UUDIS broadly defines a shortage as a supply issue that affects 
how pharmacies prepare and dispense a product or that influences 
patient care when prescribers must choose an alternative therapy 
because of supply issues. Once UUDIS identifies a shortage, it 
generally does not consider a shortage to be resolved until the drug 
is available again in all strengths and package sizes from all 
manufacturers that currently produce the drug. For example, UUDIS 
could be notified of a shortage involving three manufacturers: 
Manufacturer A has no product available; Manufacturers B and C still 
do, but have limited supply of certain package sizes. According to a 
UUDIS official, UUDIS would consider the shortage to be resolved (1) 
when Manufacturers A, B, and C all have all strengths and package 
sizes back in stock; (2) if Manufacturer A decides to discontinue its 
product, when Manufacturers B and Manufacturer C both have all 
strengths and package sizes back in stock; or (3) when UUDIS obtains 
other information indicating that a shortage has been resolved, such 
as FDA telling UUDIS that Manufacturers B and C have increased supply 
and all market need has been met. Our analysis focuses on shortages of 
prescription drugs, so we excluded shortages of over-the-counter 
drugs, biologics (including vaccines), medical devices, and orally 
administered vitamins from our analysis even though UUDIS also tracks 
and includes these shortages in its data. 

[5] See [hyperlink, http://www.gao.gov/products/GAO-12-116] and 
[hyperlink, http://www.gao.gov/products/GAO-14-194]. At the time we 
did work for our 2011 report, FDA did not have a database containing 
information on drug shortages. Since then, FDA developed its own 
database to track shortages. We reported concerns with FDA's 
management of this database in 2014, see [hyperlink, 
http://www.gao.gov/products/GAO-14-194]. For this report, we used the 
drug shortage data maintained by UUDIS because the time period we 
reviewed includes years that predate the development of FDA's 
database. It also allows us to provide information on the trends and 
characteristics of drug shortages comparable to that which we 
presented in our 2011 and 2014 reports. 

[6] These shortages represented 52 percent of all shortages of drugs 
containing controlled substances reported during this period. 

[7] UUDIS creates a drug shortage bulletin for all shortages that it 
identifies as critical. Each bulletin is publicly posted on the 
American Society of Health-System Pharmacists' website and describes 
the reason for the shortage; any estimated resupply dates; any related 
shortages; and the national drug codes associated with the shortage. A 
national drug code is a unique identifier, though one drug can have 
multiple national drug codes associated with it. For example, a drug 
made by one manufacturer, in one strength, but in three package sizes 
would have a different national drug code for each of the three 
package sizes. UUDIS does not consistently track the national drug 
codes associated with shortages that it has determined are not 
critical. 

[8] Red Book is a compendium published by Truven Health Analytics that 
includes information about the characteristics of drug products. 

[9] We consider certain classes of controlled substances with medical 
use to be those substances subject to quotas that are contained in FDA-
approved products that are currently marketed for human use. 

[10] See GAO, Standards for Internal Control in the Federal 
Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: 
November 1999). Among other things, these standards address the need 
for federal agencies to establish internal controls and to monitor the 
performance of their programs. The establishment and review of 
performance measures and indicators is a control activity. Monitoring 
internal controls allows agencies to assess the quality of performance 
over time. We have also reported on the importance of establishing 
performance measures that demonstrate how well a program is achieving 
its goals. See GAO, Executive Guide: Effectively Implementing the 
Government Performance and Results Act, [hyperlink, 
http://www.gao.gov/products/GAO/GGD-96-118] (Washington, D.C.: June 
1996). 

[11] Dosage form manufacturers produce drugs in finished dosage form, 
which is a drug product in the form in which it will be administered 
to a patient, such as a tablet or capsule. Finished dosage forms are 
manufactured using active pharmaceutical ingredients, such as 
controlled substances, and excipients, such as fillers or 
preservatives. An active pharmaceutical ingredient is any element that 
is intended to provide pharmacological activity or other direct effect 
in the diagnosis, cure, mitigation, treatment, or prevention of 
disease, or to affect the structure or any function of the human body. 
An excipient is an inactive ingredient or component that is not the 
active pharmaceutical ingredient and serves as the vehicle or medium 
for the active ingredient. 

[12] Bulk manufacturers produce API from raw materials for use by 
dosage form manufacturers in making finished dosage forms for 
consumption. For example, bulk manufacturers may obtain poppy raw 
materials to manufacture morphine API. 

[13] Pub. L. No. 112-144, tit. X, 126 Stat. 993, 1099-1108 (2012). 

[14] See GAO, Results-Oriented Government: Practices That Can Help 
Enhance and Sustain Collaboration among Federal Agencies, [hyperlink, 
http://www.gao.gov/products/GAO-06-15] (Washington, D.C.: Oct. 21, 
2005). For that report, we reviewed relevant literature and 
interviewed experts in the area of collaboration to identify key 
practices that can help enhance and sustain federal agency 
collaboration. 

[15] Completion of our audit was delayed significantly because of 
DEA's refusal to comply with our requests for information from and 
about YERS/QMS for over a year. We requested access to YERS/QMS 
materials and data starting in January 2013, but did not obtain needed 
information until March 2014. 

[16] Diversion control activities refer to activities related to the 
registration and control of the manufacture, distribution, dispensing, 
importation, and exportation of controlled substances and listed 
chemicals. 21 U.S.C. § 886a(2)(B). See 21 C.F.R. pt. 1301 for more 
information on DEA's registration requirements and fees. 

[17] 21 U.S.C. § 886a. 

[18] DEA's March 2012 registration fee adjustment took place nearly 6 
years after its prior adjustment. Controlled Substances and List I 
Chemical Registration and Reregistration Fees, 77 Fed. Reg. 15234-
15250 (Mar. 15, 2012). 

[19] In addition to DEA's role in regulating controlled substances, 
states may also establish laws that regulate controlled substances. 
State laws must be in accordance with the CSA, but may be more 
restrictive. 

[20] Although schedule I substances have no accepted medical use in 
the United States, DEA issues quotas for these substances for 
scientific, industrial, and research purposes, including clinical 
trials. We do not include schedule I substances in our review of DEA's 
quota process because they have no accepted medical use in the United 
States. 

[21] In addition to each basic class of schedule I and II substances, 
DEA is also mandated to establish quotas for list I chemicals 
ephedrine, pseudoephedrine, and phenylpropanolamine. 21 U.S.C. § 826. 
In setting quotas for these chemicals, DEA establishes an assessment 
of annual needs, which is to provide for the estimated medical, 
scientific, research, and industrial needs of the United States, 
lawful export requirements, and the establishment and maintenance of 
reserve stocks. Registrants apply for import quotas to import list I 
chemicals in to the United States. We did not include quotas for list 
I chemicals within the scope of this report because DEA officials said 
that only a small amount of these substances are used to manufacture 
schedule II substances. 

[22] These schedules include drugs such as Tylenol with codeine®, 
Valium®, and Xanax®. 

[23] Although DEA's regulations use the term "individual manufacturing 
quotas," DEA officials refer to them as "bulk manufacturing quotas." 
Accordingly, we use the term "bulk manufacturing quotas" throughout 
our discussion. Companies that obtain bulk manufacturing quotas 
register with DEA as bulk manufacturers. 

[24] In this report, we refer to the Office of Diversion Control's UN 
Reporting and Quota Section as the Quota Unit. 

[25] DEA obtains prescription data from IMS Health, a company that 
collects and analyzes health care data, to determine the number of 
prescriptions and total amount of drugs purchased by retail 
establishments for dispensing for each substance. 

[26] FDA also uses data from IMS Health to develop its estimates of 
projected use in the United States. 

[27] 21 C.F.R. § 1303.11. 

[28] 21 C.F.R. § 1303.11. See e.g., Controlled Substances: Established 
Aggregate Production Quotas for 2012, 76 Fed. Reg. 78044 (Dec. 15, 
2011). 

[29] 21 C.F.R. § 1303.13. 

[30] DEA also monitors letters and telephone calls it receives from 
individuals who indicate that they cannot get their prescriptions 
filled for specific products or drugs from their local pharmacies. 

[31] We refer to companies that obtain procurement quotas as dosage 
form manufacturers. 

[32] 21 C.F.R. §§ 1303.12(d), 1303.25(a). DEA also may at any time 
reduce a manufacturer's bulk manufacturing quota to prevent exceeding 
the APQ. 21 C.F.R. § 1303.26. 

[33] Manufacturers can also submit hard copy paper applications by 
mail, in which case DEA officials will then manually enter the 
information from the applications into YERS/QMS. 

[34] With the enactment of FDASIA, manufacturers must notify FDA at 
least 6 months prior to the date of a discontinuance or interruption 
(or as soon as possible if 6 months is not feasible) in the 
manufacture of a drug that is life supporting, life sustaining, or 
used to treat debilitating health issues. Pub. L. No. 112-144, § 1001, 
126 Stat. at 1099 (codified at 21 U.S.C. § 356c(a)-(b)). 

[35] Pub. L. No. 112-144 § 1001, 126 Stat. at 1100 (codified at 21 
U.S.C. § 356c(h)(2)). In September 2014, FDA changed its definition of 
a drug shortage to be consistent with FDASIA. According to FDA, this 
is not a substantive change in definition and does not alter how FDA 
works on matters regarding drug shortages. In general, FDA focuses on 
shortages of medically necessary drugs that have a significant effect 
on public health. 

[36] FDA, Manual of Policies and Procedures, Drug Shortage Management, 
MAPP 4190.1, Rev. 2, (Sept. 3, 2014). 

[37] Pub. L. No. 112-144, § 1001, 126 Stat. at 1099-1100 (codified at 
21 U.S.C. § 356c). 

[38] This requirement applies where the manufacturer's request 
pertains to a schedule II controlled substance that is on the list of 
drugs in shortage maintained by FDA under section 506E of the Federal 
Food, Drug, and Cosmetic Act (established by section 1004 of FDASIA). 
Pub. L. No. 112-144, § 1005, 126 Stat. at 1105 (codified at 21 U.S.C. 
§ 826(h)). 

[39] Id. §§ 1001, 1005, 126 Stat. at 1099-1101, 1105 (codified at 21 
U.S.C. §§ 356c(e), 826(h)). 

[40] Id. §§ 1002, 1006, 126 Stat. at 1102, 1105-06 (codified at 21 
U.S.C. §§ 356c-1, 826a). 

[41] Id. § 1004, 126 Stat. at 1104-05 (codified at 21 U.S.C. § 356e). 

[42] 31 U.S.C. § 1115(b). 

[43] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[44] For example, see our findings in GAO, Managing for Results: 
Enhancing Agency Use of Performance Information for Management 
Decision Making, [hyperlink, http://www.gao.gov/products/GAO-05-927] 
(Washington, D.C.; Sept. 9, 2005); Results-Oriented Cultures: Creating 
a Clear Linkage between Individual Performance and Organizational 
Success, [hyperlink, http://www.gao.gov/products/GAO-03-488] 
(Washington, D.C.; Mar. 14, 2003); Agencies' Annual Performance Plans 
Under the Results Act: An Assessment Guide to Facilitate Congressional 
Decisionmaking, [hyperlink, 
http://www.gao.gov/products/GAO/GGD/AIMD-10.1.18] (Washington, D.C.; 
February 1998); and Program Performance Measures: Federal Agency 
Collection and Use of Performance Data, [hyperlink, 
http://www.gao.gov/products/GAO/GGD-92-65 (Washington, D.C.; May 4, 
1992). 

[45] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[46] [hyperlink, http://www.gao.gov/products/GAO-06-15]. 

[47] A shortage is counted in the total for "reported" shortages in 
the year that UUDIS is first notified of the shortage. For example, a 
shortage reported in July 2010 and resolved in March 2012 would be 
counted as a reported shortage in 2010. It would not be counted as a 
reported shortage in either 2011 or 2012. 

[48] Central nervous system and anesthesia drugs represented the 
therapeutic class with the highest frequency of all critical drug 
shortages reported from January 2009 through June 2013. See 
[hyperlink, http://www.gao.gov/products/GAO-12-116] and [hyperlink, 
http://www.gao.gov/products/GAO-14-194]. 

[49] An American Society of Anesthesiologists' survey revealed that 
over 95 percent of its members used an alternative medication because 
of a shortage and another 50 percent modified a procedure in some way 
because of a shortage. The survey also showed that the anesthesia 
drugs with the highest reported frequency of shortage were fentanyl 
and thiopental, both of which are controlled substances, as well as 
succinylcholine, propofol, and pancuronium, which are not controlled 
substances. The American Society of Anesthesiologists, 2012 ASA Drug 
Shortage Survey Results, accessed January 20, 2015, [hyperlink, 
http://asahq.org/advocacy/federal-activities/regulatory-activity/drug-
shortages#Drug%20Shortages%20Survey]. 

[50] Institute for Safe Medication Practices, Drug Shortages: National 
Survey Reveals High Level of Frustration, Low Level of Safety, 
accessed October 1, 2012, [hyperlink, 
http://www.ismp.org/newsletters/acutecare/articles/20100923.asp]. 

[51] The CSA precludes prescriptions for such schedule II ADHD 
medications from being refilled, so, according to mental health 
organizations, patients needed to see their physician more frequently 
when they were not able to get their prescriptions filled completely. 
The CSA establishes various requirements that govern prescriptions of 
controlled substances depending upon the schedule of the drug. For 
example, schedule II substances, including amphetamine used to treat 
ADHD, cannot be dispensed without a written prescription of a 
practitioner, except in emergency situations, and cannot be refilled. 
21 U.S.C. § 829(a). Schedule III and IV substances require a 
prescription; however, the prescription can be filled or refilled for 
up to 6 months after the date of the prescription and refilled up to 
five times. 21 U.S.C. § 829(b). 

[52] R. Kaakeh, B.V. Sweet, C. Reilly, et al., "Impact of Drug 
Shortages on U.S. Health Systems," American Journal of Health-System 
Pharmacy, vol. 68, no. 19 (2011). This estimate is based on the number 
of shortages in 2010, when the number of shortages of all drugs was 
about half of what it was in 2013, according to data from UUDIS. 

[53] For instance, in July 2013, Pennsylvania's Bureau of Emergency 
Medical Services instituted a procedure to allow EMS providers to 
apply for an exemption that allows them to use certain drugs for 6 
months beyond the drugs' expiration date. 

[54] Drug compounding is the process by which a pharmacist or doctor 
combines, mixes, or alters ingredients to create a drug tailored to 
the needs of an individual patient. 

[55] Drugs compounded by a licensed pharmacist in a state licensed 
pharmacy or federal facility, or by a licensed physician, and that 
meet the conditions of section 503A of the Federal Food, Drug, and 
Cosmetic Act, 21 U.S.C. § 353a, can qualify for exemptions from the 
FDA approval requirements, the requirement to label products with 
adequate directions for use, and the requirement to comply with 
current good manufacturing practice requirements. The exemptions in 
section 503A are only available if the licensed pharmacist or licensed 
physician obtains a valid prescription for an identified individual 
patient among other things. 

Although drug compounding is a traditional component of the practice 
of pharmacy, concerns have been raised by FDA and others that some 
pharmacies were going beyond traditional drug compounding for 
individual patients by producing large quantities of compounded drugs. 
The Drug Quality and Security Act created a new section 503B of the 
Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 353b, under which a 
compounder can register as an "outsourcing facility." Drugs compounded 
in an outsourcing facility by or under the direct supervision of a 
licensed pharmacist, and that meet the conditions in section 503B of 
the Federal Food, Drug, and Cosmetic Act, can qualify for exemptions 
from the FDA drug approval requirements and the requirement to label 
products with adequate directions for use. Such outsourcing facilities 
remain subject to current good manufacturing practice requirements. 

[56] The drug distributors we spoke with told us that during a 
shortage they generally use an allocation system in which customers 
receive a percentage of past orders, in order to ensure equitable 
distribution among their customers. However, one distributor said that 
customers who buy more drugs may receive a higher percentage of their 
order than other customers though all customers would be limited in 
the quantities that they receive. 

[57] EMS providers must comply with federal laws when handling 
controlled substances, and may also have to comply with a variety of 
additional state and local requirements. EMS provider organizations 
reported confusion among EMS agencies about how to comply with federal 
requirements related to controlled substances. As a result, officials 
from the Federal Interagency Committee on EMS said they are working 
with DEA to clarify DEA regulations and guidance that apply to EMS 
agencies. 

[58] Most EMS providers are required to follow certain protocols that 
govern how they administer care and, in some cases, adhere to 
medication formularies, both of which can limit the medications used 
by these providers. State and local governments regulate the practice 
of EMS by licensing EMS providers and establishing standards of care, 
including what medications can be administered by EMS providers. 
According to the 2011 National EMS Assessment, prepared for the 
Federal Interagency Committee on Emergency Medical Services, 48 states 
have either state or locally developed patient care protocols for EMS 
providers (2 states did not provide information). Additionally, 25 
states have medication lists or formularies for EMS professionals, 
according to this assessment. 

[59] 21 C.F.R. § 1303.11. The CSA does not designate when DEA is to 
propose or establish APQs. 

[60] We define "application year" as the year in which manufacturers 
apply for annual quota for the next calendar year. We define 
"production year" to be the year in which manufacturers can use their 
quota. 

[61] 21 C.F.R. §§ 1303.12 and 1303.21. 

[62] All percentage estimates from the sample of source documents have 
margins of error at the 95 percent confidence level of plus or minus 
10 percentage points or less, unless otherwise noted. All numerical 
estimates other than percentages have margins of error of plus or 
minus 10 percent or less of the value of those numerical estimates, 
unless otherwise noted. Our sample of YERS/QMS source documents 
includes the entire population of applications submitted by 
manufacturers that reported shortages caused by quota to FDA and a 
probability sample of all other applications within each year (2011 
and 2012). 

[63] 21 U.S.C. § 826(c). 

[64] The FDA data reflect information reported by manufacturers to FDA 
that is subsequently analyzed and categorized by the agency. 

[65] These 7 shortages were of drugs that included the substance 
classes of amphetamine, methylphenidate, or oxycodone, and began in 
either 2010 or 2011. According to FDA's data, there were several 
instances of shortages of drugs containing each of these substances 
between January 2010 and June 2013 and some, but not all, were 
reportedly caused by quotas. 

[66] We previously reported that these are the same factors that cause 
shortages of drugs generally. See [hyperlink, 
http://www.gao.gov/products/GAO-14-194] for more information on the 
reasons for shortages of all drugs. 

[67] Pub. L. No. 112-144, § 1005, 126 Stat. at 1105 (codified at 21 
U.S.C. § 826(h)). 

[68] DEA's regulations authorize manufacturers to apply for revisions 
to their quotas at any point. 21 C.F.R. §§ 1303.12(d) and 1303.25(a). 

[69] DEA officials report that this change is provisional as it is 
pending finalization. 

[70] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[71] This difference is significant with a margin of error at the 95 
percent confidence level of plus or minus 10 percentage points. 

[72] DEA does have agency-wide performance measures to assess its 
performance of reducing the diversion of licit drugs, such as the 
number of scheduled investigations completed each year. We have 
previously reported on these measures and recommended that DEA 
reassess its set of performance measures for the Diversion Control 
Program to identify ways to enhance the measures' link to the outcome 
of reducing diversion. See GAO, Prescription Drug Control: DEA Has 
Enhanced Efforts to Combat Diversion, but Could Better Assess and 
Report Program Results, [hyperlink, 
http://www.gao.gov/products/GAO-11-744] (Washington, D.C.: Aug. 26, 
2011). 

[73] Standards for Internal Control for the Federal Government state 
that the establishment and review of performance measures and 
indicators is a control activity and that monitoring internal controls 
allows agencies to assess the quality of performance over time. See 
[hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1], 14, 20. 
We have also reported on the importance of establishing performance 
measures that demonstrate how well a program is achieving its goals. 
See [hyperlink, http://www.gao.gov/products/GAO/GGD-96-118]. 

[74] Standards for Internal Control for the Federal Government state 
that managers need to compare actual performance to planned or 
expected results throughout the organization and analyze significant 
differences. See [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1], 13. 

[75] Standards for Internal Control for the Federal Government state 
that all internal controls, transactions, and other significant events 
need to be clearly documented, and the documentation should be readily 
available for examination. The documentation should appear in 
management directives, administrative policies, or operating manuals 
and may be in paper or electronic form. See [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1], 15. 

[76] We have previously reported on the risk to the management and 
operational continuity at agencies due to a lack of written policies 
and procedures. See, for example, GAO, Social Security Disability: 
Management Controls Needed to Strengthen Demonstration Projects, 
[hyperlink, http://www.gao.gov/products/GAO-08-1053] (Washington, 
D.C.: Sept. 26, 2008). 

[77] In the course of our work, DEA officials told us that ARCOS data 
were unreliable unless verified by DEA officials against source 
documents located at registrants' place of business. 

[78] Collaboration can be broadly defined as any joint activity that 
is intended to produce more public value than could be produced when 
the organizations act alone, including coordination. 

[79] See [hyperlink, http://www.gao.gov/products/GAO-06-15]. 

[80] Pub. L. No. 112-144, § 1004, 126 Stat. at 1104-05 (codified at 21 
U.S.C. § 356e) (directing the Secretary of Health and Human Services 
to "maintain an up-to-date list of drugs that are determined by the 
Secretary to be in shortages in the United States" among other things). 

[81] For more information on the steps that FDA takes to verify the 
existence of a drug shortage see [hyperlink, 
http://www.gao.gov/products/GAO-14-194]. 

[82] See FDA, Manual of Policies and Procedures, Drug Shortage 
Management, MAPP 4190.1 Rev. 2 (Sept. 3, 2014). 

[83] See [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1], 
18. 

[84] According to DEA officials, YERS/QMS data prior to 2011 were not 
available and our data request was made before 2013 was complete. 

[85] DEA also establishes quotas for schedule I substances, but we did 
not include them in our report because such substance classes have no 
currently accepted medical use in treatment in the United States. 

[86] A precursor is a chemical that is transformed into another 
compound as in the course of a chemical reaction. For example, the 
controlled substance noroxymorphone is a precursor used to manufacture 
opiates. A metabolite is a substance produced during metabolism or a 
result of metabolism. For example, the substance ecgonine is a 
metabolite of cocaine as it is generated in the extraction of cocaine 
from coca leaf. 

[87] DEA told us that manufacturers often withdraw an application if 
it has an error or if a manufacturer loses a contract and no longer 
needs quota for a particular substance. 

[88] See appendix II for the list of schedule II substances that we 
consider to have medical use. 

[89] DEA matched source documents to the corresponding YERS/QMS 
records for 8 of the 440 records for which they provided source 
documents. 

[90] See 21 U.S.C. § 826(c); 21 C.F.R. §§ 1303.12 and 1303.21. 

[End of section] 

GAO's Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

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

Order by Phone: 

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

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

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

Connect with GAO: 

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

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

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

Congressional Relations: 

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

Public Affairs: 

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

[End of document]