GAO 15 211, Antipsychotic Drug Use 211 668221

User Manual: 15-211

Open the PDF directly: View PDF .
Page Count: 50

Download
Open PDF In Browser	View PDF

United States Government Accountability Office

Report to Congressional Requesters

January 2015

ANTIPSYCHOTIC
DRUG USE
HHS Has Initiatives to
Reduce Use among
Older Adults in
Nursing Homes, but
Should Expand Efforts
to Other Settings

GAO-15-211

January 2015

ANTIPSYCHOTIC DRUG USE

Highlights of GAO-15-211, a report to
congressional requesters

HHS Has Initiatives to Reduce Use among Older
Adults in Nursing Homes, but Should Expand Efforts
to Other Settings

Why GAO Did This Study

What GAO Found

Dementia affects millions of older
adults, causing behavioral symptoms
such as mood changes, loss of
communication, and agitation.
Concerns have been raised about the
use of antipsychotic drugs to address
the behavioral symptoms of the
disease, primarily due to the FDA’s
boxed warning that these drugs may
cause an increased risk of death when
used by older adults with dementia and
the drugs are not approved for this
use.

Antipsychotic drugs are frequently prescribed to older adults with dementia.
GAO’s analysis found that about one-third of older adults with dementia who
spent more than 100 days in a nursing home in 2012 were prescribed an
antipsychotic, according to data from Medicare’s prescription drug program, also
known as Medicare Part D. Among Medicare Part D enrollees with dementia
living outside of a nursing home that same year, about 14 percent were
prescribed an antipsychotic. (See figure.)

GAO was asked to examine
psychotropic drug prescribing for older
adult nursing home residents. In this
report, GAO examined (1) to what
extent antipsychotic drugs are
prescribed for older adults with
dementia living inside and outside
nursing homes, (2) what is known from
selected experts and published
research about factors contributing to
the such prescribing, and (3) to what
extent HHS has taken action to reduce
the use of antipsychotic drugs by older
adults with dementia. GAO analyzed
multiple data sources including 2012
Medicare Part D drug event claims and
nursing home assessment data;
reviewed research and relevant federal
guidance and regulations; and
interviewed experts and HHS officials.

What GAO Recommends
GAO recommends that HHS expand
its outreach and educational efforts
aimed at reducing antipsychotic drug
use among older adults with dementia
to include those residing outside of
nursing homes by updating the
National Alzheimer’s Plan. HHS
concurred with this recommendation.

View GAO-15-211. For more information,
contact Katherine M. Iritani, (202) 512-7114,
iritanik@gao.gov.

Proportion of Older Adult Medicare Part D Enrollees Outside of the Nursing Home Diagnosed
with Dementia Who Were Prescribed an Antipsychotic in 2012

Note: GAO excluded individuals diagnosed with schizophrenia or bipolar disorder because the Food
and Drug Administration (FDA) has approved certain antipsychotic drugs for the treatment of these
conditions.

Experts and research identified patient agitation or delusions, as well as certain
setting-specific characteristics, as factors contributing to the prescribing of
antipsychotics to older adults. For example, experts GAO spoke with noted that
antipsychotic drugs are often initiated in hospital settings and carried over when
older adults are admitted to a nursing home. In addition, experts and research
have reported that nursing home staff levels, particularly low staff levels, lead to
higher antipsychotic drug use.
Agencies within the Department of Health and Human Services (HHS) have
taken several actions to address antipsychotic drug use by older adults in nursing
homes, as described in HHS’s National Alzheimer’s Plan; however, none have
been directed to settings outside of nursing homes, such as assisted living
facilities or individuals’ homes. While the National Alzheimer’s Plan has a goal to
improve dementia care for all individuals regardless of residence, HHS officials
said that efforts to reduce antipsychotic use have not focused on care settings
outside nursing homes, though HHS has done work to support family caregivers
in general. Stakeholders GAO spoke to indicated that educational efforts similar
to those provided for nursing homes should be extended to other settings.
Extending educational efforts to caregivers and providers outside of the nursing
home could help lower the use of antipsychotics among older adults with
dementia living both inside and outside of nursing homes.
United States Government Accountability Office

Contents

Letter

1
Background
Approximately 33 Percent of Older Adult Medicare Part D
Enrollees with Dementia Who Resided in a Nursing Home, and
14 Percent Outside of a Nursing Home, Were Prescribed
Antipsychotic Drugs in 2012
Experts and Research Commonly Cited Certain Patient and
Setting-Specific Factors Contributing to the Prescribing of
Antipsychotic Drugs to Older Adults
HHS Agencies’ Actions Focused on Reducing Antipsychotic Drug
Use by Older Adults with Dementia Target Nursing Home
Residents, Not Those in Other Settings
Conclusion
Recommendation for Executive Action
Agency Comments

10
21
26
32
33
33

Appendix I

Scope and Methodology for Data Analyses

Appendix II

Literature Review

Appendix III

Comments from the Department of Health and Human Services

Appendix IV

GAO Contact and Staff Acknowledgments

Table 1: Number and Percent of Older Adult Medicare Part D
Enrollees Diagnosed with Dementia Who Had a Long
Stay in a Nursing Home and Were Prescribed an
Antipsychotic in 2012, by Characteristic
Table 2: Antipsychotic Drug Prescribing among Older Adult
Nursing Home Residents with a Dementia Diagnosis,
2012

Tables

Page i

GAO-15-211 Antipsychotic Drugs and Older Adults

Table 3: Number and Percent of Older Adult Medicare Part D
Enrollees Diagnosed with Dementia Who Spent No Time
in a Nursing Home and Were Prescribed an Antipsychotic
in 2012, by Characteristic
Table 4: Medicare Part D Plan Payments for Older Adult Enrollees
Who Used an Antipsychotic in 2012, by Setting and
Diagnosis Category
Table 5: Number and Percent of Older Adult Medicare Part D
Enrollees with Dementia and Antipsychotic Drug
Prescriptions, and Medicare Part D Plan Payments for
Those Prescriptions in 2012, by Antipsychotic Drug
Table 6: Antipsychotic Drugs Included in GAO Analysis, by
Generation

17
19

20
36

Figures
Figure 1: Proportion of Older Adult Nursing Home Residents
Diagnosed with Dementia Who Were Prescribed an
Antipsychotic in 2012
Figure 2: Proportion of Older Adult Medicare Part D Enrollees
Outside of the Nursing Home Diagnosed with Dementia
Who Were Prescribed an Antipsychotic in 2012

Page ii

13
15

GAO-15-211 Antipsychotic Drugs and Older Adults

Abbreviations
ACL
AHRQ
CERT
CMS
FDA
HHS
MBSF
MDS
NDC
NIH
NPPES
OIG
PASRR
PDE
QIO

Administration for Community Living
Agency for Healthcare Research and Quality
Center for Education & Research on Therapeutics
Centers for Medicare & Medicaid Services
Food and Drug Administration
Department of Health and Human Services
Master Beneficiary Summary File
Long Term Care Minimum Data Set
national drug code
National Institutes of Health
National Plan and Provider Enumeration System
Office of Inspector General
Preadmission Screening and Resident Review
Medicare Part D Prescription Drug Event
Quality Improvement Organization

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

Page iii

GAO-15-211 Antipsychotic Drugs and Older Adults

441 G St. N.W.
Washington, DC 20548

January 30, 2015
The Honorable Ron Johnson
Chairman
The Honorable Thomas R. Carper
Ranking Member
Committee on Homeland Security and Governmental Affairs
United States Senate
The Honorable Susan M. Collins
Chairman
Special Committee on Aging
United States Senate
Dementia affected almost 15 percent of older adults in the United States
in 2010, according to an estimate from the RAND Corporation. 1 RAND
estimated that the total monetary cost for caring for individuals with
dementia was between $157 billion and $215 billion, of which the
Medicare program paid around $11 billion. 2 Furthermore, RAND
estimated that dementia costs will more than double by 2040 due to the
aging of the American population. While dementia is most commonly
associated with a decline in memory, it can also cause changes in mood
or personality, loss of communication, and, at times, agitation or
aggression. These behavioral symptoms can become challenging for
caregivers, both at home and in institutions such as nursing homes. To
manage these behaviors, antipsychotic drugs are sometimes prescribed.
Although experts indicate that an appropriate rate of prescribing of
antipsychotic drugs cannot be determined given that the clinical decision
to prescribe antipsychotics is based on individual patient factors, many

This estimate applies to adults aged 71 and older. See Michael D. Hurd, et. al, “Monetary
Costs of Dementia in the United States,” New England Journal of Medicine, vol. 368,
no. 14 (2013).
Dementia is not a specific illness, but a term that describes symptoms including memory
decline. Specific types of dementia, and diseases that can cause dementia, include
Alzheimer’s disease (the most common type), vascular dementia, and Parkinson’s
disease dementias.
2
Monetary costs include out-of-pocket spending, Medicare spending, nursing home
spending, and costs of in-home care. The higher estimate includes forgone wages from
informal, unpaid care.

Page 1

GAO-15-211 Antipsychotic Drugs and Older Adults

believe that prescribing rates for these drugs, particularly for nursing
home residents with dementia, have been too high.
Concerns have been raised about the use of antipsychotic drugs to
address behavioral symptoms—such as agitation or aggression—in older
adults. While a large proportion of prescriptions for antipsychotic drugs in
this population are used to treat behavioral symptoms, these drugs are
not approved for this use, and antipsychotic drugs have health risks. 3 In
2011, the Department of Health and Human Services (HHS) Office of
Inspector General (OIG) reported that a large percentage—88 percent—
of a 2007 sample of 1.4 million Medicare claims for newer antipsychotic
drugs for older adult nursing home residents was associated with a
dementia diagnosis. The Food and Drug Administration (FDA)—tasked
with approving drugs to be marketed in the United States—has not
approved these drugs to treat the behavioral symptoms of dementia,
although it has generally approved antipsychotic drugs to treat
schizophrenia and bipolar disorder. 4 Furthermore, these drugs have been
found to cause falls and other adverse events, including an increased risk
of death, among older adults with a diagnosis of dementia. 5 All
antipsychotic drugs carry an FDA-required boxed warning stating that
they are associated with an increased risk of death when used to treat
older adults with dementia-related psychosis. 6 Physicians are not
prohibited from prescribing antipsychotic drugs in the presence of
dementia despite the boxed warning, nor are they prohibited generally
from prescribing a drug for uses other than what the FDA has approved.
Otherwise known as off-label prescribing, this practice is common in the

We define older adults as those around the age of 65 and older.

Certain antipsychotic drugs are also FDA-approved for the treatment of Tourette
syndrome and adjunct treatment of major depressive disorder.

When the FDA or drug manufacturers determine that a drug may lead to death or serious
injury when used in certain situations, FDA may require that the product’s label include a
boxed warning.
6
Psychosis is characterized by a loss of touch with reality and can be expressed through
delusions and hallucinations. While the boxed warning is specific to dementia-related
psychosis, antipsychotic drugs are often used to treat other behavioral symptoms of
dementia. For the purposes of our report, we examine antipsychotic use among older
adults with dementia, broadly.

Page 2

GAO-15-211 Antipsychotic Drugs and Older Adults

United States, and Medicare Part D covers drugs prescribed off-label in
some cases. 7
Less is known about antipsychotic prescribing in settings outside of the
nursing home, but one study estimated that out of a sample of older
adults with dementia living outside of a nursing home, close to one in five
took an antipsychotic between 2002 and 2004. 8 The federal government
has an interest in improving dementia care for individuals living both
inside and outside nursing homes. The National Alzheimer’s Project Act,
enacted in January 2011, required HHS to establish the National
Alzheimer’s Project to create and maintain an integrated plan to
overcome Alzheimer’s disease, including related dementias. 9 HHS
developed its first National Alzheimer’s Plan in 2012, and it includes a
number of actions to improve systems of care and service delivery to
individuals with dementia. 10
You asked us to examine psychotropic drug prescribing for older adult
nursing home residents. In this report, we examine (1) to what extent
antipsychotic drugs are prescribed for older adults with dementia living in
and outside of nursing homes and what Medicare Part D plans paid for
these drugs; (2) what is known from selected experts and published

Medicare Part D is an optional outpatient prescription drug benefit offered by Medicare—
the federally financed health insurance program for persons aged 65 and over, individuals
under age 65 with certain disabilities, and individuals with end-stage renal disease
administered by HHS’s Centers for Medicare & Medicaid Services (CMS). CMS contracts
with private companies—plan sponsors—to provide benefits under Medicare Part D.
Antipsychotic drugs—a type of psychotropic drug—are also sometimes paid for through
Medicare Parts A (hospital benefit) and B (physician office benefit).
Medicare Part D reimbursement criteria require that drugs be used for medically accepted
conditions; this includes FDA-approved conditions as well as conditions, which may or
may not be off-label, supported by three specific medical compendia: the American
Hospital Formulary Service Drug Information, the United States Pharmacopaiea-Drug
Information, and the DrugDEX Information System.
8

Y. Rhee, J. G. Cernansky, L. L. Emanuel, C. G. Chang, and J. W. Shega, “Psychotropic
Medication Burden and Factors Associated with Antipsychotic Use: An Analysis of a
Population-Based Sample of Community-Dwelling Older Persons with Dementia,” Journal
of the American Geriatrics Society, vol. 59, no. 11 (2011).
9

Pub. L. No. 111-375, § 2(d)(2), 124 Stat. 4100 (Jan. 4, 2011).

While the National Alzheimer’s Plan specifically mentions Alzheimer’s disease, it also
addresses related dementias, as required under the act. The National Alzheimer’s Plan is
updated annually.

Page 3

GAO-15-211 Antipsychotic Drugs and Older Adults

research about factors contributing to the prescribing of antipsychotic
drugs to older adults with dementia; and (3) to what extent HHS has
taken action to reduce the use of antipsychotic drugs in older adults with
dementia.
To estimate the extent to which older adults residing inside and outside of
nursing homes are prescribed antipsychotic drugs, we first analyzed
Medicare Part D Prescription Drug Event (PDE) data for individuals with
dementia in 2012. 11 We used the Medicare Part D PDE data because
Medicare is the primary source of insurance coverage for individuals over
the age of 65, and approximately 63 percent of Medicare beneficiaries
were enrolled in Medicare Part D in 2012. To identify individuals living in
nursing homes, we combined the PDE claims data with 2012 data from
the Long Term Care Minimum Data Set (MDS), 12 which includes nursing
home assessments for all individuals living in nursing homes, regardless
of insurance coverage. We also used data from the Medicare Master
Beneficiary Summary File (MBSF), 13 as well as the Medicare Part D Risk
File to identify diagnoses, including dementia diagnoses and diagnoses
for certain conditions for which FDA has approved the use of
antipsychotics drugs. 14 We excluded from our estimates individuals with
dementia also diagnosed with one of these FDA-approved conditions for
antipsychotic drugs—schizophrenia and bipolar disorder. We define an
individual as having been prescribed an antipsychotic drug if they were
prescribed at least one prescription for an antipsychotic drug during the
year, regardless of how many days’ supply are covered by the
prescription. Within the nursing home population, our analysis of PDE
claims specifically identified those with a long stay in the nursing home—
11

The PDE data contain pharmacy claims for all prescription drugs dispensed to Medicare
Part D beneficiaries.

We define locations outside of the nursing home as any location an individual may reside
that it is not a nursing home. This may include assisted living facilities or an individual’s
home.
12

The MDS contains information from nursing-home resident assessments for all
Medicare or Medicaid certified nursing homes, regardless of payer. In 2012, over 15,600
nursing homes were Medicare and Medicaid certified and included in the MDS. Medicaid
is the joint federal-state program that finances health care for low-income individuals.

MBSF includes data on enrollment, spending, and use of services for all Medicare
beneficiaries.

The Medicare Part D Risk File contains enrollee information such as age, gender, and
diagnoses from the previous year.

Page 4

GAO-15-211 Antipsychotic Drugs and Older Adults

defined by the Centers for Medicare & Medicaid Services (CMS) as more
than 100 days—because drugs for individuals with short stays—100 days
or less—are generally covered under Medicare Part A, not Part D. We
identified antipsychotic prescriptions in the PDE claims data by examining
relevant national drug codes (NDC) using a list of generic names for
antipsychotic drugs. 15 We disaggregated the data to examine certain
characteristics, such as gender, age, and geographic location.
To supplement our analysis of the Medicare Part D data for the nursing
home population, we also analyzed data on antipsychotic prescribing and
diagnoses among nursing home residents available in the MDS. This
allowed us to look at a more comprehensive population of nursing home
residents—all residents in a Medicare or Medicaid certified nursing
home—and to examine prescribing rates by length of stay. 16 For this
analysis, we determined an individual was prescribed an antipsychotic
drug if any nursing home assessment during 2012 indicated the resident
took an antipsychotic drug during the previous 7 days. In addition to
excluding residents with dementia also diagnosed with schizophrenia and
bipolar disorder, we also excluded residents with Tourette syndrome, a
condition for which FDA has approved the use of certain antipsychotics,
as well as Huntington’s disease, a condition for which CMS guidance has
recognized antipsychotics as an acceptable treatment. 17 Individuals
with dementia and at least one of these diagnoses accounted for about
7 percent of nursing home residents with dementia overall.
To identify what Medicare Part D plans paid for antipsychotic drugs
prescribed to older adults with dementia in 2012, we identified individuals
with dementia using the Medicare Part D Risk File, and calculated plan
payments for those enrollees using the PDE claims data. We also
calculated plan payments for the most commonly prescribed antipsychotic
drugs, and used the National Plan and Provider Enumeration System
(NPPES) to identify the breakdown of prescriber specialties listed on

NDCs uniquely identify specific drug products for a given manufacturer.

Nearly all of the 15,700 nursing homes providing long term care services in the United
States participated in the Medicare and Medicaid programs in 2012.

We were unable to exclude Medicare Part D enrollees with Tourette syndrome and
Huntington’s disease for the Medicare PDE analysis because the Medicare Part D Risk
File does not contain information on whether an enrollee has been diagnosed with these
conditions.

Page 5

GAO-15-211 Antipsychotic Drugs and Older Adults

antipsychotic drug claims under Medicare Part D in 2012 to calculate the
share of plan payments for prescriptions from the specialties with the
most antipsychotic prescribing for individuals with dementia. 18 For more
details on the data analyses, see appendix I.
We ensured the reliability of the MDS data, Medicare PDE claims data,
Medicare Part D Risk File data, MBSF data, and NPPES data used in this
report by performing appropriate electronic data checks, reviewing
relevant documentation, and interviewing officials and representatives
knowledgeable about the data, where necessary. We found the data were
sufficiently reliable for the purpose of our analyses.
To examine what is known from selected experts and published research
about factors contributing to the prescribing of antipsychotic drugs to
older adults with dementia, we interviewed experts in the field of dementia
care and conducted a literature review. We used a snowball sampling
approach to identify industry, provider, and advocacy groups, as well as
research experts, with experience or work on the subject of antipsychotic
drug use among older adults. We interviewed two to five groups within
each category of expert groups and asked them about contributing
factors. We also conducted a literature review to identify original research
on factors associated with prescribing antipsychotic drugs to older adults.
We searched for relevant articles published in peer-reviewed journals
from January 2009 through March 2014 and excluded international
research. We also included articles published within our timeframe that
were identified through our interviews. We found a total of 42 articles that
met our inclusion criteria; after reviewing those articles, we found
contributing factors cited in 18 articles. For more details on the literature
review and a list of the articles identified, see appendix II.
To identify the extent to which HHS has taken action to reduce
prescribing of antipsychotic drugs to older adults, we spoke with officials
from agencies within HHS, including CMS, FDA, the Administration for
Community Living (ACL), 19 the National Institutes of Health’s (NIH)

NPPES is a list of all Medicare providers, including unique provider identifiers,
maintained by CMS.

ACL was created in 2012 to bring together key HHS organizations and offices—the
Administration on Aging, the Office on Disability, and the Administration on Developmental
Disabilities—into a single agency to provide support for individuals in the community with
functional needs, such as seniors with dementia.

Page 6

GAO-15-211 Antipsychotic Drugs and Older Adults

National Institute on Aging and National Institute of Mental Health, and
the Agency for Healthcare Research and Quality (AHRQ). We also
reviewed relevant federal regulations, CMS nursing home guidance,
agency Web pages, and other federal documents, such as the National
Alzheimer’s Plan.
We conducted this performance audit from January 2014 through January
2015 in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our findings
and conclusions based on our audit objectives.

Background
Antipsychotic Drug
Classification and FDAApproved Uses

Antipsychotic drugs are classified into two sub-groups. The first group, or
generation, of antipsychotic drugs—also known as “conventional” or
“typical” antipsychotic drugs—was developed in the mid-1950s. Examples
include haloperidol (Haldol®) and loxapine (Loxitane®). The second
generation of antipsychotic drugs, known as “atypical” antipsychotics, was
developed in the 1980s. Examples include aripiprazole (Abilify®) and
risperidone (Risperdal®). Atypical antipsychotics became more popular
upon their entry into the market due to the initial belief that these drugs
caused fewer side effects than the conventional antipsychotics. Each
antipsychotic drug has its own set of FDA-approved indications. 20 The
vast majority of antipsychotic drugs are FDA-approved for the treatment
of schizophrenia, and most atypical antipsychotic drugs are FDAapproved for the treatment of bipolar disorder. In addition, some
antipsychotics are FDA-approved for the treatment of Tourette syndrome.
CMS guidance to state nursing home surveyors also recognizes
antipsychotics as an acceptable treatment for conditions for which the
drugs have not been FDA-approved, such as for the treatment of
Huntington’s disease.

FDA-approved indications may include, for example, use of a drug to treat a particular
diagnosed condition, or symptom, in specific populations and under certain conditions.

Page 7

GAO-15-211 Antipsychotic Drugs and Older Adults

In 2005, FDA recognized the risks associated with atypical antipsychotic
drugs and required those drugs to have a boxed warning, citing a higher
risk of death related to use among those with dementia. In 2008, FDA
recognized similar risks for conventional antipsychotic drugs and required
the same boxed warning. Besides the risks described in the boxed
warning, use of antipsychotic drugs carries risks of other side effects,
such as sedation, hypotension, movement disorders, and metabolic
syndrome issues.

Treatment Approaches for
Behavioral Symptoms of
Dementia

Clinical guidelines consistently suggest the use of antipsychotic drugs for
the treatment of the behavioral symptoms of dementia only when other,
non-pharmacological attempts to ameliorate the behaviors have failed,
and the individuals pose a threat to themselves or to others. 21 For
example, AMDA–The Society for Post-Acute and Long-Term Care
Medicine suggests first assessing the scope and severity of the behavior
and identifying any environmental triggers for the behavior. A medical
evaluation may determine whether the behavioral symptoms are
associated with another medical condition, such as under-treated arthritis
pain or constipation. In its clinical guideline, AMDA cited conflicting
evidence surrounding the effectiveness of antipsychotic drugs in treating
the behavioral symptoms of dementia. 22 It noted one evidence review that
found significant improvement in symptoms with the treatment of certain
atypical antipsychotic drugs, but also noted that other reviews signaled
there were no significant differences attributable to atypical antipsychotic
drugs.
Other non-pharmacological interventions that can be attempted prior to
the use of antipsychotic drugs may focus on emotions, sensory

See American Psychiatric Association, Practice Guideline for the Treatment of Patients
With Alzheimer’s Disease and Other Dementias (October 2007); American Geriatrics
Society, “Updated Beers Criteria for Potentially Inappropriate Medication Use in Older
Adults,” Journal of the American Geriatrics Society, vol. 60, no. 4 (2012); and American
Medical Directors Association, Excerpts from the Dementia in the Long Term Care Setting
Clinical Practice Guideline: American Medical Directors Association (Columbia, Md.:
2012).

American Medical Directors Association, Excerpts from the Dementia in the Long Term
Care Setting Clinical Practice Guideline.

Page 8

GAO-15-211 Antipsychotic Drugs and Older Adults

stimulation, behavior management, or other psychosocial factors. 23 An
example of an emotion-oriented approach is Reminiscence Therapy,
which involves the recollection of past experiences through old materials
with the intention of enhancing group interaction and reducing
depression. An example of a sensory stimulation approach is Snoezelen
Therapy, which typically involves introducing the individual to a room full
of objects designed to stimulate multiple senses, including sight, hearing,
touch, taste, and smell. This intervention is based on the theory that
behavioral symptoms may stem from sensory deprivation. A 2012 white
paper published by the Alliance for Aging Research and the
Administration on Aging, a part of the ACL, noted that advancements
have been made with regards to the evidence base supporting some nonpharmacological interventions, but that evidence-based interventions are
not widely implemented. 24 Experts referenced in the white paper identified
the need for clearer information about the interventions, such as a system
to classify what interventions exist and who might benefit from those
interventions. Experts also noted that additional research is needed to
develop effective interventions.

Federal Nursing Home
Standards and
Regulations

Federal law requires nursing homes to meet federal quality and safety
standards, set by CMS, to participate in the Medicare and Medicaid
programs. CMS regulations require nursing homes to ensure that
residents’ drug therapy regimens are free from unnecessary drugs, such
as medications provided in excessive doses, for excessive durations, or
without adequate indications for use. 25 Nursing facility staff must assess
each resident’s functional capacity upon admission to the facility and

See Department of Veterans Affairs Health Services Research & Development Service,
A Systematic Evidence Review of Non-pharmacological Interventions for Behavioral
Symptoms of Dementia (Washington, D.C.: March 2011).

K. Maslow, Translating Innovation to Impact: Evidence-Based Interventions to Support
People with Alzheimer’s Disease and Their Caregivers at Home and in the Community, a
report prepared at the request of the Alliance for Aging Research, the Administration on
Aging, and MetLife Foundation (Washington, D.C.: The Alliance for Aging Research,
September 2012).

42 C.F.R. § 483.25(l). Federal law also requires that nursing homes protect and promote
residents’ rights to be free from chemical restraints, defined as any drug used for
discipline or convenience and not required to treat the resident’s medical symptoms.
42 U.S.C. § 1395i-3(c)(1)(A)(ii).

Page 9

GAO-15-211 Antipsychotic Drugs and Older Adults

periodically thereafter, and provide each resident a written care plan. 26
Based on these assessments, nursing homes must ensure that
antipsychotics are prescribed only when necessary to treat a specific
condition diagnosed and documented in the patient’s record, and that
residents who use antipsychotic drugs receive gradual dose reductions
and behavioral interventions, unless clinically contraindicated. Part of the
nursing home survey process, otherwise known as nursing home
inspections, involves audits of these care plans and assessments. 27

Approximately
33 Percent of Older
Adult Medicare Part D
Enrollees with
Dementia Who
Resided in a Nursing
Home, and
14 Percent Outside of
a Nursing Home,
Were Prescribed
Antipsychotic Drugs
in 2012

About one-third of older adult Medicare Part D enrollees with dementia
who spent over 100 days in a nursing home were prescribed an
antipsychotic drug in 2012. Among those Medicare Part D enrollees with
dementia who spent no time in a nursing home in 2012, we found that
about 14 percent were prescribed an antipsychotic. In total, Medicare
Part D plans paid roughly $363 million in 2012 for antipsychotic drugs
prescribed for older adult Medicare Part D enrollees with dementia.

In addition to any other requirements a state may have established, it must operate a
Preadmission Screening and Resident Review (PASRR) program, approved by CMS as
part of its state Medicaid plan, that provides for the evaluation of all individuals with
serious mental illness who apply to reside in Medicaid certified nursing homes. According
to CMS, if PASRR recommends that a nursing home is the appropriate setting, necessary
treatment, such as antipsychotic drugs, should be specifically listed on the PASRR
recommendations. If an individual does not have a PASRR recommendation for
antipsychotic drugs, additional scrutiny may be triggered if an antipsychotic drug is
prescribed.

All nursing homes that participate in the Medicare and Medicaid programs are subject to
periodic surveys to ensure that they are in compliance with federal quality standards. CMS
contracts with state survey agencies to conduct the surveys. Surveys must include an
adequate number of residents with dementia who are receiving an antipsychotic drug.

Page 10

GAO-15-211 Antipsychotic Drugs and Older Adults

Roughly One-third of
Older Adult Nursing Home
Residents with Dementia
Were Prescribed an
Antipsychotic Drug in 2012

We found that about 33 percent of Medicare beneficiaries with dementia
who were enrolled in a Part D plan and had a long stay in a nursing
home—defined as over 100 cumulative days—were prescribed an
antipsychotic in 2012. 28 (See table 1.) We also found that prescribing
rates for Medicare Part D enrollees with dementia who were nursing
home residents varied somewhat by resident characteristic:
•

Male enrollees were slightly more likely to have been prescribed
an antipsychotic drug than female enrollees—about 36 percent and
32 percent, respectively.

•

The prescribing rate declined as Medicare Part D enrollee age
increased. For example, about 41 percent of those Medicare Part D
enrollees aged 66 to 74 received an antipsychotic prescription,
compared to 29 percent of those enrollees aged 85 and older who
were prescribed an antipsychotic drug.

•

The prescribing rate for antipsychotic drugs was highest for enrollees
in the South, and lowest for enrollees in the West.

28
Medicare Part D prescriptions for short-stay nursing home enrollees—enrollees with
100 days or less in the nursing home—may be low because often Medicare Part A covers
drugs administered during short, post-acute stays in nursing homes.

Page 11

GAO-15-211 Antipsychotic Drugs and Older Adults

Table 1: Number and Percent of Older Adult Medicare Part D Enrollees Diagnosed
with Dementia Who Had a Long Stay in a Nursing Home and Were Prescribed an
Antipsychotic in 2012, by Characteristic
Enrollees with dementia in the long-stay
a
nursing home setting

Gender
Age

Number without
antipsychotic
prescription

Number with
antipsychotic
prescription

Percent with
antipsychotic
prescription

Total

268,486

131,480

33%

Female

206,215

95,911

Male

62,271

35,569

66-74

28,888

19,842

75-84

89,131

51,014

85+
Census location

150,467

60,624

Midwest

73,906

33,743

Northeast

68,648

31,120

South

93,026

52,938

West

32,906

13,679

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) Medicare Part D data. | GAO-15-211

Notes: Percentages are rounded to the nearest whole number. Enrollees outside of the 50 states and
the District of Columbia were excluded, as were enrollees with less than 12 months of Medicare
Part D enrollment, those whose Medicare coverage began on or after January 1, 2011, and those
who passed away in 2012. We included only those that lived through 2012 because antipsychotics
can be used in the hospice or palliative setting to make patients more comfortable at the end of their
lives. We also excluded enrollees with dementia who were also diagnosed with schizophrenia or
bipolar disorder because FDA has approved certain antipsychotics for the treatment of these two
conditions. Diagnostic information was identified using Medicare Part D Risk File data, and includes
only diagnoses from the previous year. Enrollees are considered as having dementia if one of the
following diagnoses were present in the 2012 Medicare Part D Risk File: a general dementia
diagnosis, a diagnosis of Alzheimer’s disease, or a diagnosis of Parkinson’s disease.
a

If a Medicare Part D enrollee had a nursing home stay that was longer than 100 days, they were
considered a long-stay nursing home enrollee; data on all other enrollees with shorter stays in nursing
homes are not included in this table.

We found slightly lower rates of antipsychotic drug prescribing when we
restricted our analysis to those enrollees with three or more 30-day
supply prescriptions during 2012. Specifically, about 28 percent of longstay Medicare Part D enrollees with dementia were given three or more
30-day supply prescriptions for an antipsychotic drug over the course of
2012. We also found that the majority of prescriptions given to those longstay Medicare Part D enrollees with dementia—about 68 percent—were
for seven or more 30-day supplies of the drug, while only 3 percent were
for less than one 30-day supply.

Page 12

GAO-15-211 Antipsychotic Drugs and Older Adults

Consistent with the findings for Medicare Part D enrollees, our analysis of
MDS data showed that approximately 30 percent of all older adult nursing
home residents—regardless of enrollment in Medicare Part D—with a
dementia diagnosis were prescribed an antipsychotic drug at some point
during their 2012 nursing home stay. 29 (See fig. 1.) Residents with
dementia accounted for a significant proportion of all nursing home
residents. In 2012, about 38 percent, or almost 1.1 million of the
2.8 million nursing home residents that year, were diagnosed with
dementia.
Figure 1: Proportion of Older Adult Nursing Home Residents Diagnosed with
Dementia Who Were Prescribed an Antipsychotic in 2012

Notes: Data on antipsychotic drugs were missing for 204 residents with dementia. Residents in
facilities outside of the 50 states and the District of Columbia were excluded, as were residents who
passed away in the facility in 2012. We included only those that lived through 2012 because
antipsychotics can be used in the hospice or palliative setting to make residents more comfortable at
the end of their lives. We also excluded residents with dementia who were also diagnosed with
schizophrenia, bipolar disorder, Huntington’s disease, and Tourette syndrome because FDA has
approved certain antipsychotics for the treatment of schizophrenia, bipolar disorder, and Tourette
syndrome, and CMS guidance recognizes antipsychotics as an acceptable treatment for Huntington’s
disease. Diagnostic information was identified using all assessments with a target date in 2012 for a
given resident. The initiation of the antipsychotic prescription could have occurred prior to the nursing
home stay or during the nursing home stay.
a

A resident is considered as having dementia if one of the following diagnoses were present on any of
the resident’s assessments in 2012: a general dementia diagnosis, a diagnosis of Alzheimer’s
disease, or a diagnosis of Parkinson’s disease.

To examine this more comprehensive population of nursing home residents, we
identified prescribing rates based on nursing home assessments through the MDS.

Page 13

GAO-15-211 Antipsychotic Drugs and Older Adults

Examining this more comprehensive database of nursing home residents
also allowed us to compare the antipsychotic drug prescribing rates of
long-stay residents and short-stay residents—those residents who spent
100 days or less in the nursing home. The proportion of residents
diagnosed with dementia who were prescribed an antipsychotic drug
was greater for long-stay residents than for short-stay residents (about
33 percent versus 23 percent, respectively). (See table 2.) Variation in
prescribing rates across resident characteristics was similar to the
variation found in the Medicare Part D enrollee long-stay nursing home
population.
Table 2: Antipsychotic Drug Prescribing among Older Adult Nursing Home
Residents with a Dementia Diagnosis, 2012
a

Residents with dementia

All
b

Length of stay

Number without
antipsychotic
prescription

Number with
antipsychotic
prescription

Percent with
antipsychotic
prescription

741,926

313,311

30%

Short stay

285,570

86,471

Long stay

456,356

226,840

Gender

Female

520,729

208,516

Male

221,145

104,785

Age

65-74

75,251

43,080

75-84

240,158

115,043

85+

429,247

155,188

Midwest

199,391

77,804

Northeast

177,365

72,313

South

259,121

126,145

West

106,049

37,049

Census location

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) nursing home assessment data. | GAO-15-211

Notes: Percentages are rounded to the nearest whole number. The gender of some residents was
unidentified, and those residents are not included in the gender breakdown. Residents in facilities
outside of the 50 states and the District of Columbia were excluded, as were residents who passed
away in the facility in 2012. We included only those that lived through 2012 because antipsychotics
can be used in the hospice or palliative setting to make residents more comfortable at the end of their
lives. We also excluded residents with dementia who were also diagnosed with schizophrenia, bipolar
disorder, Huntington’s disease, and Tourette syndrome because FDA has approved certain
antipsychotics for the treatment of schizophrenia, bipolar disorder, and Tourette syndrome, and CMS
guidance recognizes antipsychotics as an acceptable treatment for Huntington’s disease. Diagnostic
information was identified using all assessments with a target date in 2012 for a given resident. The
initiation of the antipsychotic prescription could have occurred prior to the nursing home stay or during
the nursing home stay.
a

Page 14

GAO-15-211 Antipsychotic Drugs and Older Adults

If a resident had a nursing home stay that was longer than 100 days, they were considered a longstay resident; all other residents were considered short-stay residents.

One in Seven Older Adult
Medicare Part D Enrollees
with Dementia Living
Outside of Nursing Homes
Were Prescribed an
Antipsychotic Drug in 2012

Of those Medicare Part D enrollees with dementia in settings outside of
the nursing home, about one in seven (14 percent) were prescribed an
antipsychotic. (See fig. 2.) Roughly 1.2 million of the 20.2 million older
adult Medicare Part D enrollees living outside of a nursing home in 2012
had a diagnosis of dementia—just above 6 percent.

Figure 2: Proportion of Older Adult Medicare Part D Enrollees Outside of the
Nursing Home Diagnosed with Dementia Who Were Prescribed an Antipsychotic in
2012

Notes: Enrollees outside of the 50 states and the District of Columbia were excluded, as were
enrollees with less than 12 months of Medicare Part D enrollment, those whose Medicare coverage
began on or after January 1, 2011, and those who passed away in 2012. We included only those that
lived through 2012 because antipsychotics can be used in the hospice or palliative setting to make
patients more comfortable at the end of their lives. We also excluded enrollees with dementia who
were also diagnosed with schizophrenia or bipolar disorder because FDA has approved certain
antipsychotics for the treatment of these two conditions. Diagnostic information was identified using
Medicare Part D Risk File data, and includes only diagnoses from the previous year.
a

Page 15

GAO-15-211 Antipsychotic Drugs and Older Adults

The rate of antipsychotic drug prescribing among older adult Medicare
Part D enrollees with dementia was lower for those living outside of
nursing homes, compared to those living in nursing homes, given that
residents of nursing homes are generally sicker than those living outside
of nursing homes. We also found that the pattern of variation in
antipsychotic drug prescribing for Medicare Part D enrollees outside of a
nursing home for certain characteristics was different from the pattern of
variation found in the nursing home population.
•

The proportion of Medicare Part D enrollees outside of nursing homes
diagnosed with dementia who were prescribed an antipsychotic drug
was higher for older enrollees—the opposite of the pattern found in
the nursing home setting. (See table 3.)

•

The prescribing rate was also higher for female enrollees outside of
the nursing home than for male enrollees, whereas the opposite was
true in the nursing home setting.

•

The prescribing rate for enrollees with dementia outside of the nursing
home changed less depending on enrollee location than those in
nursing homes.

Page 16

GAO-15-211 Antipsychotic Drugs and Older Adults

Total
Gender
Age

Census location

Number without
antipsychotic
prescription

Number with
antipsychotic
prescription

Percent with
antipsychotic
prescription

1,056,433

170,286

14%

Female

677,304

119,779

Male

379,129

50,507

66-74

238,542

33,328

75-84

468,323

73,171

85+

349,568

63,787

Midwest

225,473

30,913

Northeast

204,389

34,316

South

384,800

65,196

West

241,771

39,861

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) Medicare Part D data. | GAO-15-211

Notes: Percentages are rounded to the nearest whole number. Enrollees outside of the 50 states
and the District of Columbia were excluded, as were enrollees with less than 12 months of Medicare
Part D enrollment, those whose Medicare coverage began on or after January 1, 2011, and those
who passed away in 2012. We included only those that lived through 2012 because antipsychotics
can be used in the hospice or palliative setting to make patients more comfortable at the end of their
lives. We also excluded enrollees with dementia who were also diagnosed with schizophrenia or
bipolar disorder because FDA has approved certain antipsychotics for the treatment of these two
conditions. Diagnostic information was identified using Medicare Part D Risk File data, and includes
only diagnoses from the previous year. Enrollees are considered as having dementia if one of the
following diagnoses were present in the 2012 Medicare Part D Risk File: a general dementia
diagnosis, a diagnosis of Alzheimer’s disease, or a diagnosis of Parkinson’s disease.

We found slightly lower rates of antipsychotic drug prescribing for
Medicare Part D enrollees outside of the nursing home when we
restricted our analysis to those enrollees with three or more 30-day
supply prescriptions. Specifically, about 11 percent of enrollees outside of
the nursing home received three or more prescriptions for antipsychotic
drugs over the course of 2012. About 58 percent of antipsychotic
prescriptions for Medicare Part D enrollees with dementia living outside of
a nursing home were for seven or more 30-day supplies of the drug, while
only 3 percent were for less than a 30-day supply.

Page 17

GAO-15-211 Antipsychotic Drugs and Older Adults

Medicare Part D Plan
Payments for
Antipsychotic Drugs
Prescribed to Older Adult
Enrollees with Dementia
Totaled $363 Million in
2012

Medicare Part D plans paid roughly $363 million in 2012 for antipsychotic
drugs used by Medicare Part D enrollees with dementia aged 66 and
older. 30 (See table 4.) Medicare Part D spending on antipsychotic drugs
for Medicare Part D enrollees living outside of a nursing home with a
dementia diagnosis totaled almost $171 million in 2012, the same as
spending for long-stay nursing home enrollees with dementia. Payments
for short-stay nursing home enrollees may be low because often
Medicare Part A covers drugs administered during short, post-acute stays
in nursing homes. Medicare Part D plans consistently spent more than
double on antipsychotic prescriptions for female enrollees than for male
enrollees; as reported in table 1, the number of female Medicare Part D
enrollees using antipsychotic drugs was also over two times that of
males.

Medicare Part D plans paid almost $1.2 billion in 2012 for antipsychotic drugs used by
Medicare Part D enrollees aged 66 and older, regardless of dementia diagnosis. We
excluded new enrollees, including all 65-year-olds, because the Medicare Part D Risk File
did not have diagnostic data for these groups.

Page 18

GAO-15-211 Antipsychotic Drugs and Older Adults

Table 4: Medicare Part D Plan Payments for Older Adult Enrollees Who Used an Antipsychotic in 2012, by Setting and
Diagnosis Category
In millions
Enrollees in the long-stay
a
nursing home setting

Age

Total, including short-stay
nursing home

Total spending
on antipsychotic
drugs

Enrollees
with
b
dementia

Total spending
on antipsychotic
drugs

Enrollees
with
dementia

Total spending
on antipsychotic
drugs

Enrollees
with
dementia

$367

$171

$748

$171

$1,185

$363

Female

259

124

521

122

830

261

Male

109

227

356

102

66-74

132

446

611

75-84

138

210

373

151

201

130

All
Gender

Enrollees outside of the
nursing home setting

85+

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) Medicare Part D data. | GAO-15-211

Notes: Enrollees in and outside of the nursing home cannot be directly compared because the
severity of dementia—and thus, the level of treatment acuity—needed for enrollees in the nursing
home is much greater than that for enrollees outside of the nursing home. Enrollees outside of the 50
states and the District of Columbia were excluded, as were enrollees with less than 12 months of
Medicare Part D enrollment, those whose Medicare coverage began on or after January 1, 2011, and
those who passed away in 2012. We included only those that lived through 2012 because
antipsychotics can be used in the hospice or palliative setting to make patients more comfortable at
the end of their lives. Diagnostic information was identified using Medicare Part D Risk File data, and
includes only diagnoses from the previous year.
a

If a Medicare Part D enrollee had a nursing home stay that was longer than 100 days, they were
considered a long-stay nursing home enrollee; all other enrollees who spent time in a nursing home
were considered short-stay enrollees.

Enrollees are considered as having dementia if one of the following diagnoses were present in the
2012 Medicare Part D Risk File: a general dementia diagnosis, a diagnosis of Alzheimer’s disease, or
a diagnosis of Parkinson’s disease. For this dementia-specific calculation, we excluded enrollees with
dementia who were also diagnosed with schizophrenia or bipolar disorder because FDA has
approved certain antipsychotics for the treatment of these two conditions.

Internal medicine, family medicine, and psychiatry or neurology
physicians prescribed the greatest proportion of antipsychotic drug
prescriptions for older adult Medicare Part D enrollees with dementia—
about 82 percent in total. Antipsychotic drugs prescribed by these
specialties also made up about 82 percent of the Medicare Part D plan
payments for antipsychotic drugs—almost $298 million in plan payments.
Antipsychotic prescriptions from internal medicine physicians comprised
36 percent of Medicare Part D plan payments for antipsychotic drugs,
while family medicine and psychiatry or neurology prescriptions
comprised about 30 and 16 percent, respectively. Nurse practitioner and

Page 19

GAO-15-211 Antipsychotic Drugs and Older Adults

physician assistant prescriptions collectively accounted for almost
5 percent of antipsychotic drug claim payments, while the remaining
13 percent encompassed many specialties. 31
Quetiapine Fumarate, Risperidone, and Olanzapine were the most
commonly prescribed antipsychotic drugs for older adult Medicare Part D
enrollees with dementia in 2012, comprising approximately $246 million in
plan payments. (See table 5.) Haloperidol and Aripiprazole were also
commonly prescribed; these two drugs were prescribed to almost 9 and
6 percent of Medicare Part D enrollees with dementia, respectively.
Table 5: Number and Percent of Older Adult Medicare Part D Enrollees with
Dementia and Antipsychotic Drug Prescriptions, and Medicare Part D Plan
Payments for Those Prescriptions in 2012, by Antipsychotic Drug
Antipsychotic drug

Number

Percent

Payments in millions

Quetiapine Fumarate*

146,868

48.7%

$158

Risperidone*

100,108

33.2

Olanzapine*

38,458

12.7

Haloperidol

26,761

8.9

Aripiprazole*

18,015

6.0

Ziprasidone HCL*

5,031

1.7

All Other

9,270

3.0

Source: GAO analysis of Centers for Medicare & Medicaid Services (CMS) Medicare Part D data. | GAO-15-211

*denotes atypical antipsychotic drugs
Notes: Percentages are rounded to the nearest tenth of a percent and do not add up to 100,
suggesting that some enrollees had prescriptions for multiple antipsychotic drugs. Enrollees
outside of the 50 states and the District of Columbia were excluded, as were enrollees with less than
12 months of Part D enrollment, those whose Medicare coverage began on or after January 1, 2011,
and those who passed away in 2012. We included only those that lived through 2012 because
antipsychotics can be used in the hospice or palliative setting to make patients more comfortable at
the end of their lives. We also excluded enrollees with dementia who were also diagnosed with
schizophrenia or bipolar disorder because FDA has approved certain antipsychotics for the treatment
of these two conditions. Diagnostic information was identified using Medicare Part D Risk File data,
and includes only diagnoses from the previous year. Enrollees are considered as having dementia if
one of the following diagnoses were present in the 2012 Medicare Part D Risk File: a general
dementia diagnosis, a diagnosis of Alzheimer’s disease, or a diagnosis of Parkinson’s disease. This
table does not include prescriptions for enrollees with a short stay in the nursing home because
prescriptions during a short stay may be covered by Medicare Part A.

About 5 percent of prescriptions contained a provider identifier that did not match with
the NPPES data, and 1 percent of prescriptions contained a provider identifier for which
the taxonomy code listed did not match a taxonomy code listed by the Washington
Publishing Company, the creator of healthcare provider taxonomy codes.

Page 20

GAO-15-211 Antipsychotic Drugs and Older Adults

Experts and
Research Commonly
Cited Certain Patient
and Setting-Specific
Factors Contributing
to the Prescribing of
Antipsychotic Drugs
to Older Adults

Experts we spoke with and research we reviewed commonly identified
certain factors that are specific to the patient that contribute to
antipsychotic prescribing, such as patient agitation or delusions. Experts
and research also identified certain contributing factors that are specific to
settings, such as to nursing homes or hospitals.

Experts and Research
Identified Patient-Specific
Factors, Such As Agitation
or Delusions, That
Contribute to the Decision
to Prescribe Antipsychotic
Drugs

The majority of experts we spoke with and some research articles we
reviewed highlighted agitation, aggression, or exhibiting a risk to oneself
or others as factors that contribute to the decision to prescribe
antipsychotics. For example, in a study examining the MDS from 1999 to
2006 in eight states, 51 percent of aggressive nursing home residents
diagnosed with dementia were prescribed antipsychotic drugs in 2006, as
opposed to 39 percent of residents with behavioral symptoms but who
were not aggressive during that same time period. 32 The study suggested
that aggressive residents may have been more likely to be prescribed
antipsychotics because of the greater risk of injury associated with the
aggressive behavior. This is consistent with findings from our analysis of
nursing home assessment data; we found that, of residents diagnosed
with dementia and documented as being a risk to themselves or others,
61 percent had an antipsychotic drug prescription in 2012.
Many experts we interviewed identified other situations that may warrant
the use of antipsychotics despite their risk, such as patients experiencing
frightening delusions or hallucinations that cause the patient to act out in
ways that may be violent or harmful. Several experts noted that
individuals experiencing these psychotic and other behaviors may be
suffering from distress and are more likely to be prescribed antipsychotic

S. Crystal, M. Olfson, C. Huang, H. Pincus, and T. Gerhard, “Broadened Use of Atypical
Antipsychotics: Safety, Effectiveness, and Policy Challenges,” Health Affairs, vol. 28, no. 5
(2009).

Page 21

GAO-15-211 Antipsychotic Drugs and Older Adults

drugs to ease their distress and improve their quality of life. 33 For
example, individuals may injure themselves or strike another resident or
staff member because of delusions that these people intend to kill them.
A few research articles identified psychotic behaviors as a contributing
factor. For instance, one study that examined medical records of more
than 200 nursing home residents with dementia found that 47 percent of
residents who were on an antipsychotic also had a diagnosis of
psychosis. 34
The research we reviewed also cited other specific patient characteristics
associated with higher antipsychotic use in dementia patients. Patient
characteristics such as age, gender, race or ethnicity, and psychiatric
diagnoses were associated with higher antipsychotic prescribing in
several articles. For example, in one study of nursing home assessments
and Medicaid drug claims from seven states, researchers found that
nursing home residents with psychiatric co-morbidities, such as anxiety
and depression without psychosis, were more likely to be prescribed
antipsychotic drugs. 35 Male gender was also mentioned as a patient
characteristic associated with higher antipsychotic prescribing in three
research articles. 36 In our analyses of 2012 Medicare data, males had a
higher prescribing rate in the nursing home, while females had a higher
rate outside of the nursing home. Finally, one article found that black
nursing home residents were more likely to be prescribed antipsychotic

Psychotic behaviors may be associated with multiple conditions, including schizophrenia
or dementia. Antipsychotic drugs are FDA-approved for the treatment of schizophrenia but
not approved for the treatment of behavioral symptoms of dementia.

Healthcare Management Solutions, LLC and the Meyers Primary Care Institute at the
University of Massachusetts Medical School. Antipsychotic Drug Use Project Final Report
(Columbia, Md.: January 2013).

See J. A. Lucas, S. Chakravarty, J. R. Bowblis, T. Gerhard, E. Kalay, E. K. Paek, and
S. Crystal, “Antipsychotic Medication Use in Nursing Homes: A Proposed Measure of
Quality,” International Journal of Geriatric Psychiatry (2014).

36
See Lucas, et al, “Antipsychotic Medication Use”; Y. Chen, B. A. Briesacher, T. S. Field,
J. Tija, D. T. Lau, and J. H. Gurwitz, “Unexplained Variation across US Nursing Homes in
Antipsychotic Prescribing Rates,” Archives of Internal Medicine, vol. 170, no.1 (2010); and
P. Kamble, H. Chen, J. T. Shere, and R. R. Aparasu, “Use of Antipsychotics Among
Elderly Nursing Home Residents with Dementia in the U.S.: An Analysis of National
Survey Data,” Drugs & Aging. vol. 26, no. 6 (2009).

Page 22

GAO-15-211 Antipsychotic Drugs and Older Adults

drugs, while another article found that black residents were less likely to
receive them when compared to white residents. 37

Experts and Research
Identified Setting-Specific
Factors That Contribute to
Antipsychotic Prescribing

Experts and research identified factors within the setting that an individual
visits or resides in, such as nursing homes or hospitals, as contributing to
the decision to prescribe antipsychotic drugs to older adults. Among
nursing homes, experts and research cited factors, including the culture
of the facility, the level of staff training and education, and the number of
staff at the nursing home, as contributing to the decision to prescribe
antipsychotic drugs to older adults. Specifically, nursing home
leadership—such as administrators and medical directors—and culture
were cited by half of the experts and two of the research articles. An
expert told us that when the leadership of the nursing home believes it is
broadly acceptable to provide antipsychotic drugs to residents with
dementia, this belief spreads throughout the facility. One study examining
variation in antipsychotic use in nursing homes looked at the pharmacy
claims and nursing home assessments of more than 16,000 residents in
1,257 nursing homes. 38 The study found that new nursing home residents
admitted to facilities with high antipsychotic prescribing rates were
1.4 times more likely to receive antipsychotics, even after controlling for
patient-specific factors.
In addition to nursing home culture and leadership, many experts and two
research articles identified staff or prescriber education and training on
antipsychotic prescribing for individuals with dementia as affecting
antipsychotic drug prescribing. One industry group we spoke with
indicated that physician training specifically regarding older adults with
dementia in nursing homes and knowledge of related federal regulations
are often lacking. Similarly, a study in 68 nursing homes in Connecticut
examining knowledge of nursing home leaders and staff, who often set
the tone for prescribing antipsychotic drugs and observing patients’
behavioral symptoms, found most of the certified nursing assistants—
96 percent—were not aware of the serious risks to residents that can

See Lucas, et al, “Antipsychotic Medication Use” and Chen, et al, “Unexplained
Variation Across U.S. Nursing Homes.”

See Chen, et al, “Unexplained Variation across U.S. Nursing Homes.”

Page 23

GAO-15-211 Antipsychotic Drugs and Older Adults

result from antipsychotic use. 39 The study also found that 56 percent of
direct-care staff believed medications worked well to manage resident
behavior. Another article reported that antipsychotic drug prescribing for
individuals with dementia decreased from 20.3 to 15.4 percent in one
nursing home after the implementation of an educational in-service
training designed to reduce the inappropriate use of antipsychotic
prescribing and increase documentation of non-pharmacological
interventions. 40 In expert interviews, education of staff was identified as a
factor that can contribute to minimizing unnecessary antipsychotic
prescribing. One provider group noted that, in order to reduce
antipsychotic use, a facility would need to invest in professional training
for staff in a way that provides information about adequate alternatives to
antipsychotic drugs.
Nursing home staffing levels, specifically low staff levels, were also cited
as a contributing factor to antipsychotic drug use in one research article
and by a few experts. For example, one study examined more than 5,000
nursing homes and 561,000 residents by linking 2009 and 2010
prescription drug claims to the Nursing Home Compare database to
identify a nationwide pattern of antipsychotic drug use. 41 The study found
the nursing homes with the highest quintiles of antipsychotic drug use had
significantly less staff than those with the lowest quintiles. 42 An expert
group noted that nursing homes with less staff may not have enough
activities and oversight for the patients, which in turn may make the
nursing home residents susceptible to higher antipsychotic drug use.

C. A. Lemay, K. M. Mazor, T. S. Field, J. Donovan, A. Kanaan, B. A. Briesacher, S. Foy,
L. R. Harrold, J. H. Gurwitz, and J. Tjia, “Knowledge of and Perceived Need for Evidencebased Education about Antipsychotic Medications among Nursing Home Leadership and
Staff,” Journal of the American Medical Directors Association, vol. 14, no.12 (2013).

K. Watson-Wolfe, E. Galik, J. Klinedinst, and N. Brandt, “Application of the Antipsychotic
Use in Dementia Assessment Audit Tool to Facilitate Appropriate Antipsychotic Use in
Long Term Care Residents with Dementia,” Geriatric Nursing, vol. 35 (2014).

Nursing Home Compare is a consumer tool designed to help individuals choose a
nursing home for themselves or someone for whom they are caring by displaying
information related to the quality of specific nursing homes.

B. A. Briesacher, J. Tjia, T. Field, K. M. Mazor, J. L. Donovan, A. O. Kanaan, and
J. H. Gurwitz, “Nationwide Variation in Nursing Home Antipsychotic Use, Staffing and
Quality of Care,” Pharmacoepidemiology and Drug Safety, vol. 21 (2012).

Page 24

GAO-15-211 Antipsychotic Drugs and Older Adults

In addition, the majority of experts we spoke with told us that entering a
nursing home from a hospital is a factor leading to higher antipsychotic
prescribing in the nursing home. These experts agreed that antipsychotic
drugs are often initiated in hospital settings and carried over to nursing
home settings. One industry group we spoke with noted that individuals
with dementia go to the hospital frequently and can be prescribed an
antipsychotic drug if they exhibit disruptive behavior. Another industry
group attributed the actual prescribing of antipsychotic drugs to hospital
care culture and stated that the prescribing of antipsychotics is a common
practice in hospitals for treating individuals with dementia. A research
study that examined the medical charts of 73 residents in seven nursing
homes found 84 percent of the residents that had been admitted to the
nursing home from the hospital were admitted on at least one
psychoactive medication—including antipsychotics. 43
Finally, experts we spoke with indicated that caregivers’ frustration with
the behavior of individuals with dementia can lead to requests for
antipsychotic drugs. For example, an advocacy group we spoke with
mentioned that a caregiver may request an antipsychotic drug for an
individual with dementia in an effort to keep them in the home. The
individual with dementia may not recognize their relative, which can
cause them agitation. To keep the individual calm so that they can stay in
the home and not be placed in a nursing home, an antipsychotic
medication may be prescribed. Representatives from another provider
group explained that when an individual with dementia has an unmet
need, they may also appear to be in distress, which may cause the
caregiver to become frustrated because they do not know how to relieve
this distress.

43
V. A. Molinari, D. A. Chiriboga, L. G. Branch, J. Schinka, L. Schonfeld, L. Kos,
W. L. Mills, J. Krok, and K. Hyer, “Reasons for Psychiatric Medication Prescription
for New Nursing Home Residents,” Aging and Mental Health, vol. 15, no. 7 (2011).

Page 25

GAO-15-211 Antipsychotic Drugs and Older Adults

HHS Agencies’
Actions Focused on
Reducing
Antipsychotic Drug
Use by Older Adults
with Dementia Target
Nursing Home
Residents, Not Those
in Other Settings

HHS agencies, including CMS, AHRQ, and NIH, have taken actions to
address antipsychotic drug use by older adults with dementia in nursing
homes. However, HHS has done little to address antipsychotic drug use
among older adults with dementia living in settings outside of the nursing
home.

HHS Has Several
Initiatives Focused on
Reducing Antipsychotic
Drug Use by Older Adult
Nursing Home Residents

Under the National Plan to Address Alzheimer’s Disease, HHS has a goal
to expand support for people with Alzheimer’s disease and their families
with emphasis on maintaining the dignity, safety, and rights for those
suffering from this disease. 44 To reach this goal, HHS outlined several
actions, including monitoring, reporting, and reducing the use of
antipsychotics drugs by older adults in nursing homes. CMS has taken
the lead in carrying out this work. Other HHS agencies have also done
work related to reducing antipsychotic drug use in nursing homes.

CMS Actions

In 2012, CMS launched the National Partnership to Improve Dementia
Care in Nursing Homes with federal and state agencies, nursing homes,
providers, and advocacy organizations. 45 This was in response to several
reports dating back to 2001 published by the HHS Inspector General and
advocate concerns about the persistently high rate of antipsychotic drug
use and quality of care provided to nursing home residents with dementia.
The National Partnership began with an initial goal of reducing the
national prevalence of antipsychotic drug use in long-stay nursing home
residents by at least 15 percent by December 31, 2012. CMS used
publicly reported measures from the Nursing Home Compare website to

Alzheimer’s disease is the most common type of dementia. While the National
Alzheimer’s Plan specifically mentions Alzheimer’s disease, it also addresses related
dementias.

Throughout this document we use the term “National Partnership” to refer to the
National Partnership to Improve Dementia Care in Nursing Homes.

Page 26

GAO-15-211 Antipsychotic Drugs and Older Adults

track the progress of the National Partnership and, according to officials,
to reach out to those states and individual facilities with high prescribing
rates. In the fourth quarter of 2011, which was deemed the baseline,
23.8 percent of long-stay nursing home residents nationwide were
prescribed an antipsychotic drug. While the National Partnership did not
reach its target reduction in 2012, by the end of 2013 the national use
rate decreased to 20.2 percent, a 15.1 percent reduction. 46 The majority
of states showed some improvements in their rates; however some states
showed much more improvement than others. For example, Delaware
showed a 27 percent reduction—from 21.3 to 15.5 percent—in the
prevalence of antipsychotic drug use from 2011 through 2013, while
Nevada saw a smaller reduction of 2.7 percent—from 20.3 to
19.7 percent—during the same period. The National Partnership is
working with state coalitions, as well as nursing homes to reduce this rate
even further. In September 2014, CMS established a new set of national
goals to reduce the use antipsychotic drugs in long-stay nursing home
residents by 25 percent by the end of 2015 and 30 percent by the end of
2016, which, assuming a baseline of 23.8 percent, would lead to a
prescribing rate of 16.7 percent. Beginning in January 2015, CMS’s FiveStar Quality Rating System for nursing homes will be based, in part, on
this measure of the extent to which antipsychotic drugs are used in the
nursing home. The Five-Star Quality Rating System provides a way for
consumers to compare nursing homes on the Medicare Web site.
Previously, the measure was displayed, but not included in the calculation
of each nursing home’s overall quality score.
The National Partnership works with state-based coalitions and consumer
advocates to educate and promote a re-thinking of dementia care in
nursing homes, with a focus on person-centered care. 47 In addition to
state-based coalitions, the National Partnership also includes Quality
Improvement Organizations (QIO), which are state-based Medicare
contractors tasked with promoting the delivery of quality services to

The calculations CMS uses to measure antipsychotic drug use in nursing homes are
somewhat different than the calculations used for the information provided in earlier
sections of the report. For example, CMS’s measure is not specific to individuals with a
dementia diagnosis.

Person-centered care is an approach to care that focuses on residents as individuals
and supports caregivers working most closely with them. It involves a continual process of
listening, testing new approaches, and changing routines and organizational approaches
in an effort to individualize and de-institutionalize the care environment.

Page 27

GAO-15-211 Antipsychotic Drugs and Older Adults

Medicare beneficiaries, and Advancing Excellence in America’s Nursing
Homes Campaign, a major initiative of the Advancing Excellence in Long
Term Care Collaborative. 48 The National Partnership includes regular
conference calls with states, regions, and advocates, and presentations
by experts in the field, to share best practices and brainstorm ways to
improve dementia care in their facilities.
In addition, CMS has taken four additional actions that aim to reduce
antipsychotic drug use among older adults in nursing homes:
•

CMS provided additional guidance and mandatory training around
behavioral health and dementia care from 2012 through 2013 to the
state surveyors responsible for reviewing and assessing nursing
homes. This was done in order to improve surveyors’ ability to identify
the use of unnecessary drugs, including inappropriate use of
antipsychotic drugs.

•

QIOs have focused some of their efforts on reducing antipsychotic
drug use in nursing homes. For example, beginning in 2013, the QIOs
provided training to nearly 5,000 nursing homes on the appropriate
use of antipsychotic medications.

•

CMS recently concluded pilots of a new dementia-focused survey that
examines the use of antipsychotic drugs to older adults with dementia
living in nursing homes. CMS reported that the focused survey pilot
results will allow the agency to gain new insight about the current
survey process, including how the process can be streamlined to
more efficiently and accurately identify and cite deficient practices as
well as to recognize successful dementia care programs. The pilot
consisted of onsite, targeted surveys of dementia care practices in
five nursing homes in each of five states.

•

CMS began reporting the rate of chronic use of atypical antipsychotic
drugs by older adult Medicare beneficiaries living in nursing homes for

CMS is required to contract with one QIO for each of the 50 states, the District of
Columbia, Puerto Rico, and the U.S. Virgin Islands. The statutory mission of the QIO is to
promote the effectiveness, efficiency, economy, and quality of services delivered to
Medicare beneficiaries and to ensure that those services are reasonable and necessary.
42 U.S.C. §§ 1320c-2,1320c-3.
The mission of the Advancing Excellence in Long Term Care Collaborative is to help
nursing homes improve residents’ quality of care and quality of life.

Page 28

GAO-15-211 Antipsychotic Drugs and Older Adults

Medicare Part D plans in 2013. 49 This information is publicly available
on the Medicare Part D Compare Website, which is used by Medicare
beneficiaries comparing Medicare Part D plans. The measure used for
Medicare Part D plans differs in a few respects from the measure
used to assess nursing homes. First, the Medicare Part D measure
examines chronic use, defined as having at least 3 months or more of
a prescription for an atypical antipsychotic drug, whereas the nursing
home measure includes any use. Additionally, the Medicare Part D
measure only includes atypical antipsychotic drugs, compared to the
nursing home measure, which includes all antipsychotic drugs. Of the
421 Medicare Part D plans reporting in 2012, the rate of use among
Medicare Part D enrollees residing in nursing homes ranged from 0 to
almost 64 percent. The average among all Medicare Part D plans in
2012 was approximately 22 percent of enrollees residing in nursing
homes having at least 3 months or more of a prescription. CMS told
us that variation in antipsychotic prescribing among Medicare Part D
plans may be explained by the prescribing practice in the plan’s
service area, nursing home willingness to allow the use of
antipsychotic drugs for the behavioral symptoms of dementia, resident
need, and success in implementing interventions to reduce the
inappropriate use of antipsychotic drugs.

Actions of Other HHS
Agencies

In addition to CMS actions, AHRQ and NIH have awarded research
grants for work related to antipsychotic drug use by older adults with
dementia in nursing homes.
•

AHRQ has funded individual grants for work related to antipsychotic
drug use in nursing homes through its Center for Evidence and
Practice Improvement and the Centers for Education & Research on
Therapeutics (CERT) program. For example, in 2011, CERT funded
several project centers for a 5-year period to study a broad range of
health care issues, including Rutgers University, which studied
patterns of antipsychotic drug use, along with the safety and
effectiveness of antipsychotic drug use for individuals living in nursing
homes.

This metric specifically measures the percent of Medicare Part D beneficiaries 65 years
and older who are continuously enrolled in a nursing home and who received an atypical
antipsychotic medication. Medicare Part D Compare defines chronic use as any
beneficiary who has received at least a 90-day supply of atypical antipsychotic
medication(s) during a nursing home stay.

Page 29

GAO-15-211 Antipsychotic Drugs and Older Adults

•

Within the NIH, the National Institute on Aging and the National
Institute of Mental Health have also funded related research, including
a number of studies examining the safety of antipsychotic drugs in
older adults.

Some stakeholders and other provider groups we spoke with expressed
overall support of HHS’s efforts, while others cautioned that the emphasis
should not curtail access to those individuals who need antipsychotic
drugs. Specifically, stakeholders indicated that the collaboration between
public and private organizations, as part of the National Partnership,
along with the sharing of practices aimed at reducing antipsychotic drug
use, contributed to the campaign’s success. Stakeholders also mentioned
that the National Partnership allowed nursing homes to pay attention and
start talking about issues related to antipsychotic drug use. Some
stakeholders further indicated that HHS’s initiatives have brought focus to
the issue of antipsychotic drug use in older adults in nursing homes.
Conversely, other groups and individuals involved in HHS’s efforts
expressed concern that the emphasis on reducing antipsychotic drug use
in nursing homes could result in some individuals who need these
medications not receiving them. One researcher we spoke with noted that
because nursing homes’ use of antipsychotic drug use is measured and
publicly reported, these facilities may be worried about their antipsychotic
drug rate and focus on the bottom-line number instead of what is good for
the individual. CMS officials told us that they are careful in their
messaging to acknowledge that antipsychotic drugs have a useful
prescribing purpose and therefore will never be totally eliminated. They
are working with providers to develop a comprehensive view of what a
patient potentially needs, emphasizing that using antipsychotic drugs
should not be the first-line intervention.

HHS Has Taken Little
Action to Educate and
Provide Outreach to
Reduce Antipsychotic
Drug Use among Older
Adults Residing Outside of
Nursing Homes

While the National Alzheimer’s Plan was established to improve care for
all individuals with dementia regardless of the setting where they reside,
HHS efforts related to reducing antipsychotic drug use among older
adults have primarily focused on those living in nursing homes with less
activity geared toward those living outside of nursing homes. HHS
officials noted that the focus has been on reducing antipsychotic drug use
rates in nursing homes for a variety of reasons, including the severity of
dementia among nursing home residents and the agency’s responsibility
to ensure appropriate training of nursing home staff. However, the risk of
antipsychotic drugs to older adults is not specific to those in nursing
homes. Furthermore, we found that 1 in 7 Medicare Part D enrollees with

Page 30

GAO-15-211 Antipsychotic Drugs and Older Adults

dementia outside of the nursing home were prescribed an antipsychotic
drug in 2012.
We identified one activity by HHS’s ACL that examined a topic related to
the use of antipsychotic drugs, specifically the use of nonpharmacological interventions in the treatment of individuals with
dementia. In 2012, ACL partnered with a research group to conduct a
study on non-pharmacological treatments and care practices for
individuals with dementia and their caregivers. 50 The study results were
presented in a white paper and disseminated on the ACL’s Web page.
ACL also included the study results in a newsletter distributed to state
organizations on aging. ACL officials also told us that they participate in
the National Partnership as a stakeholder organization, including
reviewing the training materials that were distributed to nursing homes.
However, ACL officials told us that none of their other past activities have
dealt specifically with reducing antipsychotic drug use among older adults
outside of nursing homes.
While ACL has not focused on reducing antipsychotic drug use among
older adults outside of nursing homes, ACL is responsible for other parts
of the National Alzheimer’s Plan related to improving dementia care in the
community. ACL partners with national groups to share information on
dementia-related issues such as caring for minority populations with
dementia and preventing elder abuse and neglect. As part of this work,
ACL works with organizations, such as the Alliance for Aging Research
and the National Family Caregiver Alliance, to share research, host
webinars and presentations, and promote issues through social media.
ACL also funds grants for state long-term care ombudsmen that are
responsible for advocating for older adults living in nursing homes,
assisted living facilities, and other residential settings for older adults.
Stakeholder groups we spoke to indicated that educational efforts similar
to those provided under the National Partnership should be extended to
those providing care to older adults in other settings, such as hospitals
and assisted living facilities. Some stakeholders noted that some of the
same material regarding non-pharmacological interventions could be
shared with caregivers in these other care settings.

See K. Maslow, “Translating Innovation to Impact.”

Page 31

GAO-15-211 Antipsychotic Drugs and Older Adults

Many experts we spoke with said that many nursing home residents
come to the nursing home already on an antipsychotic drug. Extending
educational efforts to caregivers and providers outside of the nursing
home could help lower the use of antipsychotics among older adults with
dementia living both inside and outside of nursing homes.

Conclusion

The decision to prescribe an antipsychotic drug to an older adult with
dementia is dependent on a number of factors, according to experts in the
field, and must take into account the possible benefits of managing
behavioral symptoms associated with dementia against potential adverse
health risks. In some cases, the benefits to prescribing the drugs may
outweigh the risks. HHS has taken important steps to educate and inform
nursing home providers and staff on the need to reduce unnecessary
antipsychotic drug use and ways to incorporate non-pharmacological
practices into their care to address the behavioral symptoms associated
with dementia. However, similar efforts have not been directed toward
caregivers of older adults living outside of nursing homes, such as those
in assisted living facilities and private residences. Targeting this segment
of the population is equally important given that over 1.2 million Medicare
Part D enrollees living outside of nursing homes were diagnosed with
dementia in 2012 and Medicare Part D pays for antipsychotic drugs
prescribed to these individuals. While the extent of unnecessary
prescribing of antipsychotic drugs is unknown, older adults with dementia
living outside of nursing homes are also at risk of the same dangers
associated with taking antipsychotics drugs as residents of nursing
homes. In fact, the National Alzheimer’s Project Act was not limited to the
nursing home setting, but calls upon HHS to develop and implement an
integrated national plan to address dementia. HHS’s National Alzheimer’s
Plan addresses antipsychotic drug prescribing in nursing homes only,
however, and HHS activities to reduce such drug use have primarily
focused on older adults residing in nursing homes. Given that HHS does
not specifically target its outreach and education efforts relating to
antipsychotic drug use to settings other than nursing homes, older adults
living outside of nursing homes, their caregivers, and their clinicians in
these settings may not have access to the same resources about
alternative approaches to care. By expanding its outreach and
educational efforts to settings outside nursing homes, HHS may be able
to help reduce any unnecessary reliance on antipsychotic drugs for the
treatment of behavioral symptoms of dementia for all older adults
regardless of their residential setting.

Page 32

GAO-15-211 Antipsychotic Drugs and Older Adults

Recommendation for
Executive Action

We recommend that the Secretary of HHS expand its outreach and
educational efforts aimed at reducing antipsychotic drug use among older
adults with dementia to include those residing outside of nursing homes
by updating the National Alzheimer’s Plan.

Agency Comments

We provided a draft of this report to HHS for comment. In its written
response, reproduced in appendix III, HHS concurred with our
recommendation, stating that the agency will support efforts to update the
National Alzheimer’s Plan through continued participation on the Federal
National Alzheimer’s Project Act Advisory Council. HHS also provided
technical comments that we incorporated as appropriate.
As agreed with your offices, unless you publicly announce the contents of
this report earlier, we plan no further distribution of it until 30 days from its
date. At that time, we will send copies to the Secretary of Health and
Human Services. In addition, the report will be available at no charge on
the GAO website at http://www.gao.gov.
If you or your staffs have any questions about this report, please contact
me at (202) 512-7114 or iritanik@gao.gov. Contact points for our Offices
of Congressional Relations and Public Affairs may be found on the last
page of this report. GAO staff who made key contributions to this report
are listed in appendix IV.

Katherine M. Iritani
Director, Health Care

Page 33

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix I: Scope and Methodology for Data
Analyses
Appendix I: Scope and Methodology for Data
Analyses

This appendix describes our methodology for analyzing the 2012
prescribing of antipsychotic drugs for older adults with dementia in
nursing homes and other settings, as well as for analyzing Medicare
Part D plan payments for these antipsychotic drug prescriptions. It also
describes our efforts to ensure the reliability of the data.

Analyses of Antipsychotic
Drug Prescribing for Older
Adults with Dementia

We used two primary data sources to examine antipsychotic drug
prescribing for older adults with dementia: the Medicare Part D
Prescription Drug Event (PDE) data to identify antipsychotic drug
prescribing for Medicare Part D enrollees in and outside of the nursing
home, 1 and the Long Term Care Minimum Data Set (MDS) to identify
antipsychotic drug prescribing for all nursing home residents, regardless
of Medicare Part D enrollment. 2

Medicare Part D Prescription
Drug Event data analyses

To estimate the extent to which older adults residing inside and outside of
nursing homes are prescribed antipsychotic drugs, we first analyzed 2012
PDE data for individuals with dementia. 3 We used the Medicare Part D
PDE data because Medicare is the primary source of insurance coverage
for individuals over the age of 65 and approximately 63 percent of
Medicare beneficiaries were enrolled in Medicare Part D in 2012. To
identify individuals living in nursing homes, we combined the PDE claims
data with data from the MDS, which includes nursing home assessments
for all individuals living in nursing homes, regardless of insurance
coverage. We also used data from the Medicare Master Beneficiary
Summary File (MBSF), 4 as well as the Medicare Part D Risk File to
identify diagnoses, including dementia diagnoses and diagnoses for
certain conditions for which the Food and Drug Administration (FDA) has
approved the use of antipsychotics drugs. 5 We excluded from our

The PDE data contain pharmacy claims for all prescription drugs dispensed to Medicare
Part D beneficiaries.
2

The MDS contains information from nursing home resident assessments for all Medicare
or Medicaid certified nursing homes, regardless of payer. In 2012, over 15,600 nursing
homes were Medicare and Medicaid certified and included in the MDS.
3

The 2012 PDE data were the most recent data available at the time of our review.

MBSF includes data on enrollment, spending, and use of services for all Medicare
beneficiaries.
5

The Medicare Part D Risk File contains enrollee information such as age, gender, and
diagnoses from the previous year.

Page 34

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix I: Scope and Methodology for Data
Analyses

estimates individuals with dementia also diagnosed with one of these
FDA-approved conditions for antipsychotic drugs—schizophrenia and
bipolar disorder. The Medicare Part D Risk File contains diagnoses based
on claims from the previous year for each enrollee, so our diagnosis
categories may be conservative estimates as they did not take into
account longer-standing or newer diagnoses. We also excluded enrollees
with outlier data, enrollees with less than 12 months of Medicare Part D
enrollment in 2012, and those enrollees who died in 2012 because they
did not have complete Medicare Part D data for the entire year. Finally,
we excluded enrollees who resided outside of the 50 states and the
District of Columbia.
For these analyses, we define an individual as having been prescribed an
antipsychotic drug if they were prescribed at least one prescription for an
antipsychotic drug during the year, regardless of how many days supply
are covered by the prescription. We identified relevant national drug
codes (NDC) using a list of generic names for antipsychotic drugs, 6 and,
using those codes, we determined the number and percent of Medicare
Part D enrollees who were prescribed an antipsychotic drug in 2012. 7 The
specific drugs included are listed in table 6.

NDCs uniquely identify specific drug products for a given manufacturer.

7
We did so using Red Book, a compendium published by Truven Health Analytics that
includes information about the characteristics of drug products. Some included drugs have
been discontinued. We excluded NDCs for drugs with a route of administration that was
not oral or sublingual. We also excluded drugs that contain antipsychotics but are not
classified as antipsychotic.

Page 35

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix I: Scope and Methodology for Data
Analyses

Table 6: Antipsychotic Drugs Included in GAO Analysis, by Generation
First generation (conventional)

Second generation (atypical)

Chlorpromazine Hydrochloride

Aripiprazole

Fluphenazine Hydrochloride

Asenapine

Haloperidol; Haloperidol Lactate

Clozapine

Loxapine Hydrochloride (discontinued); Loxapine
Succinate

Iloperidone

Mesoridazine Besylate (discontinued)

Lurasidone Hydrochloride

Molindone Hydrochloride (discontinued)

Olanzapine

Perphenazine

Paliperidone

Pimozide

Quetiapine Fumarate

Promazine Hydrochloride (discontinued)

Risperidone

Thioridazine (discontinued); Thioridazine
Hydrochloride

Ziprasidone Hydrochloride

Thiothixene; Thiothixene Hydrochloride (discontinued)
Trifluoperazine Hydrochloride
Source: GAO summary. | GAO-15-211

Notes: Antipsychotic drugs are classified into two sub-groups. The first generation of antipsychotic
drugs, also known as “conventional” antipsychotic drugs, was developed in the mid-1950s. The
second generation of antipsychotic drugs, also known as “atypical” antipsychotics, was developed in
the 1980s. We excluded NDCs for drugs that are not classified as antipsychotics and do not have an
oral or sublingual route of administration.

Within the nursing home population, our analysis of PDE data specifically
identified those with a long stay in the nursing home—defined by the
Centers for Medicare & Medicaid Services (CMS) as more than
100 days—because drugs for individuals with short stays—100 days or
less—are generally covered under Medicare Part A, not Part D. We
disaggregated the data to examine certain characteristics, such as
gender, age, and geographic location.

Long Term Care Minimum
Data Set analysis

To supplement our analysis of the Medicare Part D data for the nursing
home population, we also analyzed 2012 data on antipsychotic
prescribing and diagnoses among nursing home residents available in the
MDS. This allowed us to look at a more comprehensive population of
nursing home residents—all residents in a Medicare or Medicaid certified
nursing home—and to examine prescribing rates by length of stay, using
steps identified by CMS based on dates reported in the nursing home

Page 36

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix I: Scope and Methodology for Data
Analyses

assessments. 8 In addition to excluding residents with dementia also
diagnosed with schizophrenia and bipolar disorder, we excluded residents
with Tourette syndrome, a condition for which FDA has approved the use
of certain antipsychotics, as well as Huntington’s disease, a condition for
which CMS guidance has recognized antipsychotics as an acceptable
treatment. 9 Individuals with both dementia and at least one of these
diagnoses accounted for about 7 percent of nursing home residents with
dementia overall. We also excluded residents with outlier identification
codes or other outlier data, residents under the age of 65, and residents
in facilities outside of the 50 states and the District of Columbia. We
included only those residents that lived through 2012 so that there was a
complete year of data for each resident and because antipsychotic drugs
can be used in a hospice setting to make residents more comfortable at
the end of their lives. 10 For this analysis, we determined an individual was
prescribed an antipsychotic drug if any nursing home assessment during
2012 indicated the resident took an antipsychotic drug during the previous
7 days, and we include any instance where antipsychotic use is
documented. 11 We disaggregated the data to examine certain
characteristics, such as gender, age, and geographic location.

Medicare Part D Plan
Payments

To identify what Medicare Part D plans paid for antipsychotic drugs
prescribed to older adults with dementia in 2012, we identified individuals
with dementia using the Medicare Part D Risk File, and calculated plan

8
We looked at rates for those residents with a long stay, defined as more than 100
cumulative days in a facility, and a short stay, defined as all other residents.
9
It was not possible to exclude other conditions on this basis such as nausea and severe
depression refractory to other therapies without a medical record review. Therefore, these
conditions are not included in our exclusions. In addition, we were unable to exclude
Medicare Part D enrollees with Tourette syndrome and Huntington’s disease for the
Medicare PDE analysis because the Medicare Part D Risk File does not contain
information on whether an enrollee has been diagnosed with these conditions. Of nursing
home residents in 2012, approximately 0.06 percent were diagnosed with Huntington’s
disease and 0.02 percent were diagnosed with Tourette syndrome.
10

We conducted a sensitivity analysis to determine whether the prescribing rate changed
as a result of including residents who died in the facility in 2012, and the prescribing rate
did not change significantly.

This differs from the CMS measures of antipsychotic drug use among nursing home
residents. For example, the CMS measure for residents that spend 100 days or less in a
nursing home does not include instances of antipsychotic drug use documented on an
initial assessment.

Page 37

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix I: Scope and Methodology for Data
Analyses

payments for those enrollees using the PDE claims data. We also
calculated plan payments for the most commonly prescribed antipsychotic
drugs, and used the National Plan and Provider Enumeration System
(NPPES) to identify the breakdown of prescriber specialties listed on
antipsychotic drug claims under Medicare Part D in 2012 to calculate the
share of plan payments for prescriptions from the specialties with the
most antipsychotic prescribing for individuals with dementia. 12

Data Reliability and Audit
Standards

We ensured the reliability of the MDS data, Medicare PDE data, Medicare
Part D Risk File data, MBSF data, Red Book data, and NPPES data used
in this report by performing appropriate electronic data checks, reviewing
relevant documentation, and interviewing officials and representatives
knowledgeable about the data, where necessary. We found the data were
sufficiently reliable for the purpose of our analyses.
We conducted this performance audit from January 2014 through January
2015 in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our findings
and conclusions based on our audit objectives.

NPPES is a list of all Medicare providers, including unique provider identifiers,
maintained by CMS.

Page 38

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix II: Literature Review
Appendix II: Literature Review

To identify what is known from published research about factors
contributing to the prescribing of antipsychotic drugs to older adults with
dementia, we conducted a literature search among recently published
articles; specifically, we searched for relevant articles published from
January 1, 2009, through March 31, 2014. We conducted a structured
search of various databases for relevant peer reviewed and industry
journals including MEDLINE, BIOSIS Previews, and ProQuest. Key terms
included various combinations of “antipsychotic,” “dementia,” “elderly,”
“older adults,” “nursing homes,” “community,” “assisted living,” “home
health,” “medication management,” and “medication monitoring.” From all
database sources, we identified 386 articles. We first reviewed the
abstracts for each of these articles for relevancy in identifying contributing
factors related to the use of antipsychotic drugs both inside and outside of
nursing homes. For those articles we found relevant, we reviewed the full
article and excluded those where the research (1) was conducted outside
the United States; (2) included individuals less than 65 years of age; or
(3) was an editorial submission. We added one article that could be linked
to original research outside of the research cited in the article. After
excluding these articles and including others, 42 articles remained: 22
focused on nursing homes; 11 focused on settings outside of nursing
homes; 7 focused on both settings; and in 2 articles, the settings were
either unclear or undetermined. Articles were then coded by analysts
according to whether they identified contributing factors for use of
antipsychotic drugs. We found 18 that contained detailed reasons that
contribute to antipsychotic drug use among older adults:
Bowblis, J. R., S. Crystal, O. Intrator, and J. A. Lucas. “Response to
Regulatory Stringency: The Case of Antipsychotic Medication Use in
Nursing Homes.” Health Economics, vol. 21 (2012).
Briesacher, B. A., J. Tjia, T. Field, K. M. Mazor, J. L. Donovan,
A. O. Kanaan, L. R. Harrold, C. A. Lemay, and J. H. Gurwitz. “
Nationwide Variation in Nursing Home Antipsychotic Use, Staffing
and Quality of Care.” Abstracts of the 28th ICPE 2012, (2012).
Briesacher, B. A., J. Tjia, T. Field, D. Peterson, and J. H. Gurwitz.
“Antipsychotic use Among Nursing Home Residents.” The Journal of
American Medical Association, vol. 309, no. 5 (2013).
Chen, Y., B. A. Briesacher, T. S. Field, J. Tjia, D. T. Lau, and
J. H. Gurwitz. “Unexplained Variation across U.S. Nursing Homes in
Antipsychotic Prescribing Rates.” Archives of Internal Medicine, vol. 170,
no. 1 (2010).

Page 39

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix II: Literature Review

Crystal, S., M. Oflson, C. Huang, H. Pincus, and T. Gerhard. “Broadened
Use of Atypical Antipsychotics: Safety, Effectiveness, and Policy
Challenges: Expanded Use of these Medications, Frequently Off-label,
Has Often Outstripped the Evidence Base for the Diverse Range of
Patients Who Are Treated with Them.” Health Affairs, vol. 28, no. 5
(2009).
Department of Health and Human Services – Office of Inspector General,
Medicare Atypical Antipsychotic Drug Claims for Elderly Nursing Home
Residents,” OEI-07-08-00150, May 2011.
Fung, V., M. Price, A. B. Busch, M. B. Landrum, B. Fireman,
A. Nierenberg, W. H. Dow, R. Hui, R. Frank, J. P. Newhouse, and
J. Hsu. “Adverse Clinical Events among Medicare Beneficiaries Using
Antipsychotic Drugs: Linking Health Insurance Benefits and Clinical
Needs.” Medical Care, vol. 51, no. 7 (2013).
Healthcare Management Solutions, LLC and the Meyers Primary Care
Institute at the University of Massachusetts Medical School. Antipsychotic
Drug Use Project Final Report (Columbia, Md.: January 2013).
Kamble, P., J. Sherer, H. Chen, and R. Aparasu. “Off-Label Use of
Second-Generation Antipsychotic Agents among Elderly Nursing Home
Residents.” Psychiatric Services, vol. 61, no. 2 (2010).
Kamble, P., H. Chen, J. T. Sherer, and R. R. Aparasu. “Use of
Antipsychotics among Elderly Nursing Home Residents with Dementia in
the United States: An Analysis of National Survey Data.” Drugs & Aging,
vol. 26, no. 6 (2009).
Lemay, C. A., K. M. Mazor, T. S. Field, J. Donovan, A. Kananaan,
B. A. Briesacher, S. Foy, L. R. Harrold, J. H. Gurwitz, and J. Tjia.
“Knowledge of and Perceived Need for Evidence-Based Education
about Antipsychotic Medications among Nursing Home Leadership
and Staff.” The Journal of American Medical Directors Association,
vol. 14, no. 12 (2013).
Lucas, J. A., S. Chakravarty, J. R. Bowblis, T. Gerhard, E. Kalay,
E. K. Paek, and S. Crystal. “Antipsychotic Medication Use in Nursing
Homes: A Proposed Measure of Quality.” International Journal of
Geriatric Psychiatry (2014).

Page 40

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix II: Literature Review

Molinari, V. A., D. A. Chiriboga, L. G. Branch, J. Schinka, L. Shonfeld,
L. Kos, W. L. Mills, J. Krok, and K. Hyer. “Reasons for Psychiatric
Medication Prescription for New Nursing Home Residents.” Aging &
Mental Health, vol. 15, no.7 (2011).
Rhee, Y., J. G. Cernansky, L. L. Emanuel, C. G. Chang, and J. W. Shega.
“Psychotropic Medication Burden and Factors Associated with
Antipsychotic Use: An Analysis of a Population-Based Sample of
Community-Dwelling Older Persons with Dementia.” The Journal of
American Geriatrics Society, no. 59 (2011).
Saad, M., M. Cassagnol, and E. Ahmed. “The Impact of FDA’s Warning
on the Use of Antipsychotics in Clinical Practice: A Survey.” The
Consultant Pharmacist, vol. 25, no. 11 (2010).
Sapra, M., A. Varma, R. Sethi, I. Vahia, M. Chowdhury, K. Kim, and
R. Herbertson. “Utilization of Antipsychotics in Ambulatory Elderly with
Dementia in an Outpatient Setting.” Federal Practitioner, (2012).
Tjia, J., T. Field, C. Lemay, K. Mazor, M. Pandolfi, A. Spenard, S. Ho,
A. Kanaan, J. Donovan, J. H. Gurwitz, and B. Briesacher. “Antipsychotic
Use in Nursing Homes Varies By Psychiatric Consultant.” Medical Care,
vol. 52, no. 3. (2014).
Watson-Wolfe, K., E. Galik, J. Klinedinst, and N. Brandt. “Application of
the Antipsychotic Use in Dementia Assessment Audit Tool to Facilitate
Appropriate Antipsychotic Use in Long Term Care Residents with
Dementia.” Geriatric Nursing, vol. 35 (2014).

Page 41

GAO-15-211 Antipsychotic Drugs and Older Adults

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

Page 42

GAO-15-211 Antipsychotic Drugs and Older Adults

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

Page 43

GAO-15-211 Antipsychotic Drugs and Older Adults

Appendix IV: GAO Contact and Staff
Acknowledgments
Appendix IV: GAO Contact and Staff
Acknowledgments

GAO Contact

Katherine M. Iritani, (202) 512-7114 or iritanik@gao.gov

Staff
Acknowledgments

In addition to the contact named above, Lori Achman, Assistant
Director; Todd D. Anderson; Shaunessye D. Curry; Leia Dickerson;
Sandra George; Kate Nast Jones; Ashley Nurhussein-Patterson;
and Laurie Pachter made key contributions to this report.

(291176)

Page 44

GAO-15-211 Antipsychotic Drugs and Older Adults

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 (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 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,
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 www.gao.gov.

To Report Fraud,
Waste, and Abuse in
Federal Programs

Contact:
Website: 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) 5124400, 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

Please Print on Recycled Paper.

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.7
Linearized                      : Yes
Author                          : U.S. Government Accountability Office, http://www.gao.gov
Company                         : GAO
Create Date                     : 2015:01:27 17:01:05-05:00
Modify Date                     : 2015:02:27 16:28:40-05:00
Source Modified                 : D:20150127220035
Tagged PDF                      : Yes
XMP Toolkit                     : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:08:04
Metadata Date                   : 2015:02:27 16:28:40-05:00
Creator Tool                    : Acrobat PDFMaker 9.1 for Word
Document ID                     : uuid:c2012dff-c06b-4bb4-b334-1f2ba44baf6f
Instance ID                     : uuid:6f1abc27-2816-41a2-b0f1-131af6d82595
Subject                         : 2
Format                          : application/pdf
Title                           : GAO-15-211, Antipsychotic Drug Use: HHS Has Initiatives to Reduce Use among Older Adults in Nursing Homes, but Should Expand Efforts to Other Settings
Creator                         : U.S. Government Accountability Office, http://www.gao.gov
Producer                        : Adobe PDF Library 9.0
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 50

EXIF Metadata provided by EXIF.tools

GAO 15 211, Antipsychotic Drug Use 211 668221

Navigation menu

Versions of this User Manual:

Views

Navigation