Mental Health: Improper Restraint or Seclusion Use Places People at Risk
(Letter Report, 09/07/1999, GAO/HEHS-99-176).

Pursuant to a congressional request, GAO provided information on
inpatient and residential treatment facilities' use of restraints or
seclusion as a means of treating mental patients, focusing on the: (1)
dangers of restraint and seclusion, the extent to which restraint and
seclusion are used in inpatient and residential treatment facilities for
individuals with mental illness or mental retardation, and the number of
related injuries and deaths; (2) federal and state policies that govern
the use of restraint and seclusion in inpatient or residential treatment
facilities for individuals with mental illness or mental retardation;
and (3) experiences of states that have instituted regulations and
reporting requirements to address the use of restraint and seclusion.

GAO noted that: (1) improper restraint and seclusion can be dangerous to
both people receiving treatment and staff, but the full extent of
related injuries and deaths is unknown; (2) there is no comprehensive
reporting system to track such injuries and deaths or the rates of
restraint and seclusion use by facility; (3) GAO's telephone survey of
51 state Protection and Advocacy agencies (P&A) found that only 15
states have any systematic reporting to alert these agencies to any
deaths that occur among individuals in residential treatment settings;
(4) even these reporting systems are not comprehensive, because most
agencies that receive reports get them only from state facilities; (5)
on the basis of the partial information available from these 51
agencies, GAO identified 24 deaths associated with restraint or
seclusion during fiscal year 1998; (6) because reporting is so
fragmentary, GAO believes many more deaths related to restraint or
seclusion may occur; (7) data on use of restraint and seclusion are also
fragmentary because most facilities are not required to report these
data to oversight agencies; (8) federal and state regulations governing
restraint and seclusion for individuals with mental illness and mental
retardation are inconsistent across types of facilities; (9) the federal
government regulates the use of restraint and seclusion in nursing homes
and state Intermediate Care Facilities for the Mentally Retarded, but
until recently, no federal regulations governed their use in other
facilities, such as psychiatric hospitals, residential treatment centers
for children, or community group homes; (10) in July 1999, the Health
Care Financing Administration (HCFA) issued an interim final rule with
revised Medicare conditions of participation for hospitals that address
restraint and seclusion use; (11) although this is a positive first
step, people in residential treatment centers and group homes
participating in the Medicaid Home and Community-Based Waiver program
have limited federal protection; (12) while some states have regulations
in place governing the use of restraint and seclusion, often these
regulations do not apply to privately operated facilities; and (13) on
the basis of the experience of several states, having regulatory
protections and reporting requirements can reduce the use of restraint
and seclusion and improve safety for patients and staff.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-99-176
     TITLE:  Mental Health: Improper Restraint or Seclusion Use Places
	     People at Risk
      DATE:  09/07/1999
   SUBJECT:  Mental care facilities
	     Mental health care services
	     Mental hospitals
	     Mental illnesses
	     Persons with disabilities
	     Reporting requirements
	     State law
	     Safety standards
	     Safety regulation
IDENTIFIER:  Medicaid Home and Community Based Waiver Program
	     Medicare Program
	     Medicaid Program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************

Cover
================================================================ COVER

Report to Congressional Requesters

September 1999

MENTAL HEALTH - IMPROPER RESTRAINT
OR SECLUSION USE PLACES PEOPLE AT
RISK

GAO/HEHS-99-176

Use of Restraint and Seclusion

(101818)

Abbreviations
=============================================================== ABBREV

  HCFA - Health Care Financing Administration
  HHS - Department of Health and Human Services
  ICF/MR - Intermediate Care Facility for the Mentally Retarded
  JCAHO - Joint Commission on the Accreditation of Healthcare
     Organizations
  P&A - Protection and Advocacy agency
  SAMHSA - Substance Abuse and Mental Health Services Administration

Letter
=============================================================== LETTER

B-282597

September 7, 1999

The Honorable Joseph I.  Lieberman
The Honorable Christopher J.  Dodd
The Honorable Daniel Patrick Moynihan
The Honorable Pete V.  Domenici
United States Senate

The Honorable Nancy L.  Johnson
The Honorable Christopher Shays
The Honorable James H.  Maloney
The Honorable Sam Gejdenson
The Honorable Rosa L.  DeLauro
The Honorable Pete Stark
House of Representatives

People with serious mental illness or mental retardation are among
the country's most vulnerable citizens.  About 5.5 million adults
experience severe mental illness each year--with about 240,000
requiring either inpatient treatment in mental hospitals or
residential treatment in centers or group homes.  An estimated
120,000 individuals with mental retardation lived in intermediate
care facilities, while about 240,000 others lived in smaller
residential settings in 1998.  While states, insurance companies, and
patients and their families pay for some of this treatment, Medicare,
the federal health insurance program for the elderly and disabled,
and Medicaid, the federal and state health insurance program for the
poor, also pay for treatment of eligible individuals. 

Patient advocates and recent press coverage report that some of these
individuals are at risk of injury or death in inpatient or
residential treatment facilities as a result of improper restraint or
seclusion practices.  The Hartford Courant reported that patient
deaths were related to the use of restraint or seclusion\1 in 142
cases over the past 10 years in several types of residential
treatment settings across the country.\2

Concern over these reports has led to the introduction of proposed
legislation and your request that we conduct a study to assist you in
your legislative deliberations.  Specifically, you asked us to

  -- determine the dangers of restraint and seclusion, the extent to
     which restraint and seclusion are used in inpatient and
     residential treatment facilities for individuals with mental
     illness or mental retardation, and the number of related
     injuries and deaths;

  -- identify the federal and state policies that govern the use of
     restraint and seclusion in inpatient or residential treatment
     facilities for individuals with mental illness or mental
     retardation; and

  -- describe the experiences of states that have instituted
     regulations and reporting requirements to address the use of
     restraint and seclusion. 

To do this study, we reviewed federal regulations for Medicaid and
Medicare and regulations from selected states that affect individuals
with mental illness or mental retardation in inpatient or residential
treatment facilities.  Following meetings with experts, provider
representatives, patient advocates, and government officials, we
identified and reviewed relevant data sources.  To gain at least a
partial indication of the scope of the problem, we obtained data on
the number of deaths related to restraint or seclusion investigated
by the Protection and Advocacy agencies (P&A)\3 in all 50 states and
the District of Columbia in fiscal year 1998.  To obtain insights
into the effects of different reporting requirements and other
policies regarding restraint and seclusion use, we conducted field
work in Delaware, Massachusetts, New York, and Pennsylvania--states
that either have reduced restraint use in their public mental health
facilities or have imposed more comprehensive reporting requirements. 
We also met with officials from the Health Care Financing
Administration (HCFA), the federal agency that administers Medicare
and Medicaid, and the Substance Abuse and Mental Health Services
Administration (SAMHSA). 

As agreed, we focused our inquiry on the population receiving
services for mental illness or mental retardation in residential
facilities that receive public funding, primarily from Medicare and
Medicaid.  We did not specifically address outpatient treatment
programs, sheltered workshops, schools, drug and alcohol
rehabilitation programs, or correctional facilities.  We excluded
from our review restraints used to facilitate medical procedures,
prevent interference with medical equipment such as feeding tubes, or
provide postural support.  We did not independently audit the rates
of restraint use provided to us by states cited in this report.  We
conducted our work between March and July 1999 in accordance with
generally accepted government auditing standards. 

--------------------
\1 Restraint is the partial or total immobilization of a person
through the use of drugs, mechanical devices such as leather cuffs,
or physical holding by another person.  Seclusion is involuntary
confinement in a room that the person is physically prevented from
leaving. 

\2 The Hartford Courant (Oct.  11-15, 1998). 

\3 P&As for individuals with mental illness were established or
designated by states pursuant to the Protection and Advocacy for
Individuals With Mental Illness Act of 1986, as amended, 42 U.S.C. 
10801 et seq. 

   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Improper restraint and seclusion can be dangerous to both people
receiving treatment and staff, but the full extent of related
injuries and deaths is unknown.  There is no comprehensive reporting
system to track such injuries and deaths or the rates of restraint
and seclusion use by facility.  Our telephone survey of 51 state P&As
found that only 15 states have any systematic reporting to alert
these agencies to any deaths that occur among individuals in
residential treatment settings.  Even these reporting systems are not
comprehensive, because most agencies that receive reports get them
only from state facilities.  Additionally, P&As sometimes have
difficulty getting access to medical records, which prevents them
from thoroughly investigating such incidents.  On the basis of the
partial information available from these 51 agencies, we identified
24 deaths associated with restraint or seclusion during fiscal year
1998.  Because reporting is so fragmentary, we believe many more
deaths related to restraint or seclusion may occur.  Data on use of
restraint and seclusion are also fragmentary because most facilities
are not required to report these data to oversight agencies. 

Federal and state regulations governing restraint and seclusion for
individuals with mental illness and mental retardation are
inconsistent across types of facilities.  The federal government
regulates the use of restraint and seclusion in nursing homes and
Intermediate Care Facilities for the Mentally Retarded, but until
recently, no federal regulations governed their use in other
facilities, such as psychiatric hospitals, residential treatment
centers for children, or community group homes.  In July 1999, HCFA
issued an interim final rule with revised Medicare conditions of
participation for hospitals that address restraint and seclusion use. 
Although this is a positive step, people in residential treatment
centers and group homes participating in the Medicaid Home and
Community-Based Waiver program have limited federal protection. 
While some states have regulations in place governing the use of
restraint and seclusion, often these regulations do not apply to
privately operated facilities. 

On the basis of the experience of several states, having regulatory
protections and reporting requirements can reduce the use of
restraint and seclusion and improve safety for patients and staff. 
For example, Pennsylvania reduced the use of restraint and seclusion
by over 90 percent between 1993 and 1999 in state mental health
facilities.  And Delaware's state Intermediate Care Facility for the
Mentally Retarded introduced an initiative under Medicaid that
reduced the state's restraint use by 81 percent between 1994 and
1997.  Typically, successful strategies to reduce the use of
restraint and seclusion have similar components:  defined principles
and policies that clearly outline when and how restraint or seclusion
may be used; strong management commitment and leadership; a
requirement to report the use of restraint or seclusion; staff
training in safe use of, and alternatives to, restraint and
seclusion; and oversight and monitoring.  To improve patient safety,
we believe HCFA should, at a minimum, consider extending the same
policies--tailored to the needs of individuals--on the use of
restraint and seclusion that now protect individuals in long-term
care and hospitals to people in any treatment setting funded by
Medicare and Medicaid.  We also recommend that HCFA improve reporting
of restraint and seclusion use and any related deaths or injuries and
require staff training in safely applying restraint or seclusion as
well as alternative methods for dealing with potentially violent
situations. 

   BACKGROUND
------------------------------------------------------------ Letter :2

Clinicians, providers, and patient advocates generally agree that
when patients lose control to the extent that they or others are at
imminent risk of physical harm, staff may legitimately restrain or
seclude them on an emergency basis.  Far less agreement exists about
the use of restraint and seclusion in any other situation.  For
people with psychiatric problems, some clinicians consider seclusion
to be an appropriate early intervention strategy to reduce
overstimulation, teach self-control, and protect the treatment
setting.  For people with mental retardation, seclusion is generally
not considered appropriate, but some clinicians consider restraint to
be a legitimate part of a behavioral treatment plan, for example, as
a way to reduce self-injuring behavior.  However, many patient
advocates, state mental health program officials, and representatives
of the psychiatric nursing profession disagree.  While they accept
that restraint may be needed in some cases, they consider it an
emergency response to a treatment failure to be used only as a last
resort. 

People with mental illness or mental retardation may receive
residential treatment, and may be subject to restraint and seclusion,
in a variety of settings.  People with psychiatric conditions may
receive inpatient treatment in traditional state hospitals, private
psychiatric hospitals, or community hospitals with psychiatric units. 
Many of the advocates and clinicians we met with indicated that
deinstitutionalization of individuals with less serious mental
illness has resulted in an inpatient population with more severe
mental illness. 

Federal funding, primarily federal Medicare and federal/state
Medicaid programs, accounts for about 40 percent of the revenue for
mental health treatment facilities.  Medicare provides limited mental
health coverage for individuals over age 65 and those under 65 who
are disabled.  In 1994, Medicare spent about $4.5 billion for mental
health services in either private psychiatric hospitals or general
hospitals. 

The Medicaid program covers children with mental illness under the
age of 21 and, at state option, adults aged 65 and older with mental
illness and adults and children with mental retardation.  Medicaid
provides inpatient mental health services for children under 21 years
old in general hospitals, psychiatric hospitals, and nonhospital
settings.  Individuals aged 65 and older may receive inpatient mental
health services in a hospital or nursing home.  Medicaid spending for
inpatient psychiatric treatment totaled over $2 billion in fiscal
year 1996.  In fiscal year 1996, Medicaid spent about $9.6 billion
for Intermediate Care Facilities for the Mentally Retarded (ICF/MR),
which provide long-term residential care and treatment for people
with mental retardation.  In addition, Medicaid covers care for
children with mental illness and adults and children with mental
retardation in less restrictive settings via the home and
community-based waiver program.  These waivers allow states the
flexibility to cover a broader range of services in less restrictive
settings such as group homes.  State Medicaid programs spent $5.6
billion in federal and state funding on home and community-based
waiver services in fiscal year 1996, some of which was used to
provide residential treatment for this population. 

HCFA defines federal requirements for facilities to participate in
the Medicare and Medicaid programs.  For long-term care and ICF/MR
facilities, HCFA contracts with states to survey facilities and
certify that they meet federal requirements.  Most general and
psychiatric hospitals are accredited by the Joint Commission on the
Accreditation of Healthcare Organizations (JCAHO) or other
accrediting bodies, and HCFA accepts this as proof of meeting federal
requirements for these facilities.  For two additional conditions of
participation for psychiatric hospitals, a separate state survey is
periodically performed. 

Following discovery of severe patient neglect and abuse at a
state-run facility for individuals with mental retardation in New
York, the Congress in 1975 enacted what has become known as the
Developmental Disabilities Assistance and Bill of Rights Act.  This
act requires states, as a condition for receiving federal assistance,
to have in effect a protection and advocacy system for people with
developmental disabilities.  In 1985, congressional hearings detailed
reports of appalling conditions in psychiatric hospitals, and the
following year the Congress enacted what is known today as the
Protection and Advocacy for Individuals With Mental Illness Act. 
This law requires states to establish or designate P&As for people
with mental illness.  Most P&As are independent of state government,
though a few are state agencies.  In most states, the same P&A serves
both individuals with mental illness and those with mental
retardation.  The P&As are charged with investigating reports of
abuse or neglect of individuals with developmental disabilities or
mental illness in institutional care and are empowered to pursue
legal and administrative remedies. 

   INCOMPLETE REPORTING LEAVES
   FULL EXTENT OF PATIENT RISK
   UNKNOWN
------------------------------------------------------------ Letter :3

While restraint and seclusion use can injure patients and staff, the
full extent of that risk is not known because reporting is so
fragmentary.  Our survey of state P&As identified 24 deaths during
fiscal year 1998 related to restraint or seclusion.  However, this is
likely to understate the problem, because the lack of comprehensive
reporting makes it impossible to determine all deaths in which
restraint or seclusion was a factor.  Of 51 P&As, only 15 receive
reports of deaths in residential treatment settings on a systematic
basis, and many P&As reported having difficulty obtaining the
documents needed to pursue their investigations.  HCFA requires
reporting of deaths and the use of restraint and seclusion in some,
but not all, types of Medicaid or Medicare residential facilities
that serve adults and children with mental illness and mental
retardation.  State reporting requirements vary, and not all states
require facilities to report restraint-related deaths to the state
licensing authority.  JCAHO--the principal accrediting body for
Medicare-certified hospitalsï¿½encourages voluntary reporting of
sentinel events such as deaths and injuries related to restraint or
seclusion and collects information on the sentinel events reported. 
It does not compile data on the use of restraint and seclusion in its
accredited facilities. 

      RESTRAINT AND SECLUSION USE
      CAN INJURE PATIENTS AND
      STAFF
---------------------------------------------------------- Letter :3.1

Restraint and seclusion can be dangerous to individuals in treatment
settings because restraining them can involve physical struggling,
pressure on the chest, or other interruptions in breathing.  JCAHO
reviewed 20 restraint-related deaths and found that in 40 percent the
cause of death was asphyxiation, while strangulation, cardiac arrest,
or fire caused the remainder.  Among the deaths reported by the
Hartford Courant as cases in which restraint or seclusion was a
factor, the causes of death included asphyxia, cardiac complications,
drug overdoses or interactions, blunt trauma, strangulation or
choking, fire/smoke inhalation, and aspiration (breathing vomit into
the lungs). 

Recent incidents reported by the National Alliance for the Mentally
Ill and P&As included the following: 

  -- A 36-year-old man died in November 1998 from cardio-respiratory
     failure caused by extreme agitation after being restrained by
     eight staff members and bound in leather restraints. 

  -- In May 1997, a 35-year-old man who was doing yard work with a
     staff member at a group home became agitated, pushed the staff
     member, and walked away.  The staff member pursued the man and
     placed him in a basket hold.  (A basket hold consists of
     crisscrossing a person's arms over his or her chest and holding
     them from behind.  This hold compresses the chest and also
     prevents the staff member from observing the person's face and
     breathing).  The staff member wrapped his arms around the man's
     chest and took this man down to his knees, then face down on the
     ground.  This action compressed the man's chest and killed him. 

Children are subjected to restraint and seclusion at higher rates
than adults and also are at greater risk of injury.  Several of the
states that took part in a study sponsored by the Center for Mental
Health Services indicated they had higher restraint rates for
children, including one state in which children in state-run
inpatient facilities were restrained four times more frequently than
adults.  Children are smaller and weaker than adults, so staff who
are used to overpowering adults may apply too much pressure or force
when restraining children.  The following cases reported by the
National Alliance for the Mentally Ill illustrate the dangers of
restraint to children: 

  -- In February 1999, a 16-year-old girl died of respiratory arrest
     in California while being restrained by four staff members with
     her face on the floor. 

  -- The use of basket holds was involved in the deaths of a
     17-year-old girl in a Florida residential treatment center in
     November 1998 and a 9-year-old boy, who died in March 1999 in
     North Carolina after being restrained in a basket hold following
     a period of seclusion. 

People are at particular risk if they have a combination of
conditions, such as both mental retardation and mental illness, or
mental illness and substance abuse.  People with both mental illness
and mental retardation often are not in specialized programs to
address their unique needs and instead may be placed in either
psychiatric hospitals or facilities for people with mental
retardation only.  In one state hospital system, treatment plans for
these patients included extensive use of restraint and seclusion,
including several patients who were kept in either restraint or
seclusion 24 hours a day.  These and other practices were the subject
of a class action suit, which resulted in implementation of a
monitoring procedure. 

Many advocates we spoke with indicated that restraining individuals
who are on certain medications can be risky.  For example, a commonly
prescribed antidepressant may result in metabolic problems when a
patient's movement is restricted, which may lead to life-threatening
hyperthermia.  Clinicians have postulated that potentially fatal
cardiac arrhythmia can result from the combination of certain drugs
and the adrenaline produced by an individual's agitation and physical
struggle while being restrained.  For example, a 48-year-old man in
Texas was placed in a straitjacket and tied to a chair.  Although
15-minute checks were required, they were not performed, and he was
found dead the next day.  The cause of death was listed as an
overdose of imipramine, an antidepressant.  The medical examiner
stated that the restraints contributed to his death by affecting his
ability to metabolize the medicine. 

The use of restraint or seclusion also can result in serious injury
or abuse.  During fiscal year 1998, P&As received about 1,000
complaints regarding restraint and seclusion and documented numerous
instances of bruising and broken bones.  In one instance, a
24-year-old man suffered a severe fracture in the right arm while
being put into restraints by staff.  He was subsequently left in
four-point restraints\4 for 12 hours, despite his requests for
medical attention.  Other examples of excessive or abusive restraint
use reported by the National Alliance for the Mentally Ill and the
P&As included the following: 

  -- An 18-year-old man in a New York psychiatric hospital was
     tackled to the floor by five staff members, hit in the face, and
     then placed in restraints in February 1998. 

  -- A woman was kept in seclusion for over 30 hours in an Oregon
     hospital in December 1998 without being allowed to use the
     bathroom or contact a relative.  She eventually began screaming,
     and staff held her down by placing a knee to her neck and
     injected her with medicine. 

  -- In February 1999, a man in a Missouri state psychiatric hospital
     was restrained for 21 days and secluded for 30 days.  As a
     result of the long periods of restraint, he developed kidney
     problems and lost muscle tone. 

Even if no physical injury is sustained, people in treatment settings
can be severely traumatized during restraint, especially those who
had been sexually abused in the past.  A Massachusetts task force
investigating this issue reported that research indicates at least
half of all women treated in psychiatric settings have a history of
physical or sexual abuse.  The task force found that the use of
restraints on people who have been abused often results in those
people reexperiencing the trauma and causes setbacks in treatment. 
The task force's report recommended that staff should identify
patients who have been abused and use only certain forms of restraint
and seclusion on these patients when necessary, avoiding forms such
as mechanical restraints that place a person in a spread-eagle
position.\5

While the people in treatment are at risk during episodes of
restraint and seclusion, health care workers can also be severely
injured.  Studies continually show that the occupation of mental
health care worker is dangerous,\6 with one study demonstrating that
it can be more dangerous than that of construction workers.\7 One
study found that the largest percentage of patient assaults on staff
members occurs during restraint or seclusion incidents,\8 and another
documented that most staff injuries are sustained when staff are
trying to control patient violence.\9

--------------------
\4 Four-point restraints immobilize a person on a bed with a cuff
around each wrist and ankle. 

\5 Elaine Carmen, Bill Crane, Margaret Dunnicliff, and others, Report
of the Task Force on the Restraint and Seclusion of Persons Who Have
Been Physically or Sexually Abused (Boston, Mass.:  Massachusetts
Department of Mental Health, Jan.  25, 1996). 

\6 D.J.  Drummond and others, "Hospital Violence Reduction Among
High-Risk Patients," Journal of the American Medical Association,
261(17) (1989), pp.  2531-34. 

\7 Jane A.  Lipscomb, "Violence Toward Health Care Workers:  An
Emerging Occupational Hazard," American Association of Occupational
Health Nurses, Inc.  (AAOHN) Journal, 40(5) (May 1992), pp.  219-28. 

\8 J.R.  Lion and others, "Nursing Aides and Violence," American
Journal of Psychiatry, 24 (1981), pp.  40-43. 

\9 H.  Carmel and M.  Hunt, "Staff Injuries From Inpatient Violence,"
Hospital and Community Psychiatry, 40(1) (1989), pp.  41-46. 

      FULL EXTENT OF DEATHS AND
      INJURIES IS UNKNOWN BECAUSE
      REPORTING TO P&AS IS
      INCOMPLETE
---------------------------------------------------------- Letter :3.2

The exact number of deaths each year in which restraint or seclusion
was a factor is not known because reporting is fragmentary.  We
contacted the P&As for each state and the District of Columbia and
asked them to identify people in treatment settings who died in
fiscal year 1998 related to the use of restraint or seclusion.  The
P&As told us that restraint or seclusion was a factor in 24 of the
deaths they reported in fiscal year 1998.  But this number is likely
to be understated, because P&A officials told us they do not learn of
all deaths that may be related to restraint or seclusion.  Even
though they are charged with the responsibility to protect the
state's inpatient mentally ill population, only 15 of the 51 P&As
receive any kind of systematic reporting of such deaths from their
respective states or from psychiatric facilities.  Of the 15, 9
receive death reports for state facilities only and not for private
facilities. 

Because of the lack of reporting requirements in most states, P&As
learn about deaths on an ad hoc basis through complaints from family,
patients, and staff, as well as through on-site monitoring.  For
example, the Texas P&A did not find out about one 1995
restraint-related death until 1998, when it was reported in a
telephone call from a staff member at the facility.  Even including
these other methods, only 22 of these agencies had deaths reported to
them by any means.  Of the 1,203 deaths reported to the P&As in
fiscal year 1998, over two-thirds were reported by just five states,
and no deaths were reported to the P&As in 28 states.  P&As
investigated 376 of these 1,203 deaths.  The P&A in New York--a state
in which comprehensive reporting of such deaths is
required--accounted for almost one-third of all the death
investigations, while four other agencies investigated a combined 107
deaths. 

Even if a P&A learns about a death, some P&A officials told us that
it is often difficult to obtain incident reports and medical records
to determine whether restraint or seclusion played a role in the
patient's death.  According to some P&A officials, health facilities
often claim that these records are part of the peer review process\10
and thus are protected from disclosure under state law.  A major
concern of many P&As we talked with was the need to litigate to
obtain access to records that are critical for them to properly
investigate a case.  We were told that in some cases, litigation over
access to records used up the agencies' limited resources, further
delaying investigative efforts. 

Many P&As indicated that they face even greater obstacles in
obtaining information when the death occurred at a private facility. 
Obtaining information from private facilities is becoming
increasingly important as more mental health patients leave
state-operated facilities to receive residential treatment in other
settings, which may include facilities reimbursed through managed
care.  However, while many state agencies may gather data from their
own state's facilities, private psychiatric facilities usually are
not required to report data to either the states or the P&As.  In
1984, 71 percent of the deaths reported in New York were in state
facilities.  In 1998, 78 percent of the reports came from private
facilities.  According to the New York State official in charge of
investigations, this shift reflected, more than any other factor, a
change in where individuals receiving treatment resided.  New York is
one of the few states that requires reporting of deaths from both
public and private facilities.  Without information from private
facilities, this official said that the effectiveness of the state's
reporting system would be severely limited. 

In our survey, P&A officials told us that their ability to conduct
investigations is also hindered by limited resources.  Thus, even
when they are aware of a death, it is possible that no investigation
will take place because of a lack of funds or staff.  For example,
the Pennsylvania P&A officials stated that they limit their outreach
efforts to the public hospital system because this system already
provides more cases than they can handle.  Many of the other P&As
also noted that their representatives have only limited involvement
with the private mental health system. 

--------------------
\10 The peer review process refers to the inquiry by a committee
within the facility composed of medical personnel, which reviews
incidents to determine how quality of care can be improved or whether
professional standards were met.  Most states have laws providing
that the records of these committees are confidential and not
accessible to parties who may want to sue the provider involved, but
also providing that original documents cannot be protected just
because they were considered by the peer review committee. 

      FEDERAL AND OTHER STATE
      REPORTING REQUIREMENTS ARE
      NOT COMPREHENSIVE
---------------------------------------------------------- Letter :3.3

Federal reporting requirements differ by type of facility.  On July
2, 1999, HCFA issued an interim final rule to revise the conditions
of hospital participation in Medicare and Medicaid.\11 Effective
August 2, 1999, it requires hospitals to report to HCFA deaths that
occurred during--or can be reasonably assumed to be related
to--restraint or seclusion.  This regulation covers all hospitals but
not other facilities that receive Medicare or Medicaid funds to
provide treatment services to individuals with mental illness or
mental retardation.  ICF/MR and nursing home surveyors check and
report to HCFA on use of restraints at the time of the facility's
survey.  Federal regulations now require hospitals to track and
report on the use of restraint and seclusion.  But there are no
federal reporting requirements on restraint or seclusion use for
other types of facilities.  These facilities include community-based
group homes and day treatment centers funded under the Medicaid
waiver program and residential treatment centers for children.  Yet,
these settings are providing services to a growing number of
individuals.  Although federal regulations that implement the home
and community-based waiver program do not specifically address
reporting requirements for abuse and neglect (including the use of
restraint and seclusion), states are required to make annual reports
to HCFA on the impact of their waiver programs on the health and
welfare of the participants.  HCFA is in the process of developing
regulations that will address the use of restraint and seclusion in
nonhospital settings that provide inpatient mental health services to
children under the age of 21.  These regulations are expected to
include reporting requirements. 

Most states do not comprehensively track either restraint use or
related injuries.  Further, JCAHO recently surveyed states regarding
their requirements to report ï¿½sentinel events,ï¿½ defined as unexpected
occurrences involving death or serious physical or psychological
injury, or the risk of such injury or death.  While the results are
preliminary, of the 34 states that responded to the JCAHO survey,
only 16 and the District of Columbia indicated that they had a law
that requires some type of sentinel event reporting to a state
agency.  Our study of the P&As found that only 11 states track
restraint use in private psychiatric facilities. 

The Center for Mental Health Services, within the Department of
Health and Human Services' (HHS) SAMHSA, has collected limited data
on restraint and seclusion rates from five state psychiatric hospital
systems as part of a study to determine the feasibility of tracking
these and other performance measures.  Though the data are not
directly comparable among the five state hospital systems because of
inconsistent definitions, it is clear that use of restraint and
seclusion varies widely.  For the five hospital systems, restraint
use ranged from 0.6 to 48.1 episodes per 1,000 patient days, and the
use of seclusion ranged from 0.2 to 29.1 episodes per 1,000 patient
days.  Likewise, data collected by New York State in its own review
of its public psychiatric facilities in 1998 showed a wide range in
restraint and seclusion use.  Restraint use ranged from 0.01 to 4.7
episodes per 1,000 patient days, and seclusion use ranged from 0 to
8.76 episodes per 1,000 patient days. 

--------------------
\11 Federal Register, Vol.  64, No.  127, 36070 (July 2, 1999). 

      ACCREDITATION PROCESS RELIES
      ON VOLUNTARY REPORTING,
      WHICH TENDS TO BE INCOMPLETE
---------------------------------------------------------- Letter :3.4

Accreditation surveys are accepted by the federal government as proof
that hospitals and psychiatric facilities meet requirements for
participating in Medicare and Medicaid.  While several agencies
accredit providers of residential psychiatric care, JCAHO is the
principal accrediting body for Medicare-certified hospitals,
accrediting about 80 percent of these facilities.  It also accredits
many residential treatment facilities. 

JCAHO does not require hospitals to report sentinel events--such as
deaths related to restraint and seclusion.  JCAHO's sentinel event
guidelines, issued in 1996, encourage voluntary reporting and
encourage the hospital to conduct a root cause analysis for quality
improvement.  If a hospital does not elect to report a sentinel
event, JCAHO's expectation is that the hospital will still conduct a
root cause analysis of the event for its internal use.  If JCAHO
finds out about a sentinel event that has not been reported, an
accredited facility has 45 days to conduct a root cause analysis. 
JCAHO representatives said that the goal of this system is to be
nonpunitive and to foster self-examination that can lead to quality
improvement.  The American Hospital Association, whose
representatives said they believed mandatory reporting would
encourage staff to cover up incidents, believes this is an effective
approach. 

JCAHO's policy requires hospitals to record restraint and seclusion
use in patient records, which are subject to review during the
accreditation process.  However, the policy does not require
hospitals to report these data, nor does JCAHO compile data on these
rates.  Twenty-one of the restraint- or seclusion-related deaths
reported by the Hartford Courant occurred in JCAHO-accredited
facilities since 1996, when its new sentinel event reporting system
was established.  Fifteen of these deaths were reported to JCAHO as
sentinel events, three were not, and JCAHO did not have enough
information to be certain about whether or not three others had been
reported. 

An additional problem with giving facilities choices about reporting
is that it can limit the information available for independent review
and lead to fewer cases being investigated.  When New York first
instituted a statewide reporting system, it allowed mental health
facilities to decide whether a death was due to natural causes or
unnatural causes, such as restraint or seclusion, and should be
reported.  In 1977, it reversed this decision and began requiring
mental health facilities to report all deaths to the P&A.  State
officials said that this change was made because providers used their
discretion under the former policy to decide that deaths were the
result of some other cause, even if the patient had been restrained
or secluded during the incident.  New York shifted the determination
of which deaths were related to seclusion and restraint to the P&A
because of concern that hospitals have a tendency to underreport
suspicious deaths. 

   POLICIES GOVERNING RESTRAINT
   AND SECLUSION USE VARY AMONG
   PROGRAMS, STATES, AND
   FACILITIES
------------------------------------------------------------ Letter :4

Policies covering the use of restraint and seclusion vary among
federal programs, states, and types of facilities.  Until recently,
individuals had federal regulatory protection against improper
restraint and seclusion only if they resided in nursing homes or
ICF/MRs.  Effective in August 1999, HCFA incorporated into the
hospital conditions of participation patient rights provisions, which
address restraint and seclusion.  In addition, states are required to
ensure the health and welfare of home and community-based waiver
participants.  However, current regulations do not protect people
receiving psychiatric care at nonhospital providers such as
residential treatment centers, day treatment centers, and group
homes.  States have varying degrees of regulatory protection for
people receiving care in residential settings, but sometimes those
regulations cover only state-run facilities.  JCAHO addresses
restraint and seclusion in its accreditation process.  While hospital
industry spokespersons see accreditation as an effective means of
ensuring appropriate use of restraint and seclusion, many patient
advocates are concerned that the accreditation process alone does not
sufficiently protect individuals in treatment settings. 

      FEDERAL REQUIREMENTS DO NOT
      ADDRESS RESTRAINT AND
      SECLUSION USE FOR ALL
      PROVIDERS
---------------------------------------------------------- Letter :4.1

Federal regulations governing two types of facilities establish
affirmative rights for individuals to be free from restraint--except
under specific circumstances--and seclusion.  Residents in
long-term-care facilities that participate in Medicare or Medicaid
have the right to be ï¿½free from any physical or chemical restraints
imposed for purposes of discipline or convenience, and not required
to treat the resident's medical symptoms,ï¿½ and may not be placed in
seclusion.  ICF/MRs must ï¿½ensure that clients are free from
unnecessary drugs and physical restraints and are provided active
treatment to reduce dependency on drugs and physical restraints.ï¿½ The
ICF/MR regulations specify that restraint may be employed only as
part of an individual behavioral teaching program, as an emergency
measure when necessary to protect the individual or others from
injury, or to facilitate medical treatment. 

Federal requirements for Medicaid home and community-based waiver
programs do not specifically address restraint or seclusion, but do
require that the state applying for the waiver provide satisfactory
assurances that necessary safeguards are in place to protect the
health and welfare of the recipients of the services.  States must
adopt standards to meet the safeguard requirement that reflect each
state's approach to ensuring quality care and safety for the program
participants.  These standards may include professional licensing
standards, certification for group homes, and local building and
safety codes. 

As mentioned earlier, HCFA recently added federal guidelines on
restraint and seclusion use for hospitals, including psychiatric
hospitals.  On July 2, 1999, HCFA published its interim final rule
with revised conditions of participation for hospitals, effective on
August 2, 1999.  These conditions will also apply to psychiatric
hospitals, because psychiatric hospitals that participate in Medicare
must meet the same conditions of participation all hospitals must
meet, along with two additional conditions addressing medical records
and staffing.  The interim final rule establishes the right for
patients to be free from restraint or seclusion as a means of
coercion, discipline, or staff convenience.  These measures may be
used only for medical or surgical care or in emergency situations to
ensure the patient's physical safety and after less restrictive
interventions have been found ineffective to protect the patient or
others from harm. 

HCFA is currently reviewing whether the revised conditions of
participation should apply to residential treatment centers for
children.  These providers are rapidly replacing hospitals in
treating children with psychiatric disorders and are a less
restrictive alternative to a hospital for children whose illness is
less acute but who still require a therapeutic residential
environment.  All providers who receive Medicaid funding to treat
such children must provide ï¿½active treatmentï¿½ according to a plan of
care developed by an interdisciplinary professional team.  HCFA
issued proposed regulations in 1994 with conditions of participation
for residential treatment centers, but these regulations have not
been finalized. 

      DEGREE OF PATIENT PROTECTION
      VARIES AMONG STATES
---------------------------------------------------------- Letter :4.2

States have varying degrees of regulation and oversight for restraint
and seclusion.  They set licensing standards, survey facilities for
compliance with the standards, contract with private providers for
state-funded services, and provide care directly in state-run
facilities.  Some states have different standards for their state-run
facilities than for private providers.  Private psychiatric hospitals
are frequently not subject to the same degree of oversight as the
state-run facilities.  Even some states with extensive regulation of
their public hospitals--such as Pennsylvania and New York--have not
imposed the same requirements on the private sector. 

Individuals with mental illness or mental retardation residing in
state-operated facilities have certain basic rights that have been
recognized by federal courts.  In a case involving a man with mental
retardation confined to a Pennsylvania state institution, the U.S. 
Supreme Court held that institutionalized people have
constitutionally protected rights to safety and freedom from undue
bodily restraint.\12 In determining whether restraint is reasonable,
the Court indicated that the proper inquiry is whether professional
judgment was exercised when the restraint was ordered.  Further, a
federal district court issued detailed standards to address
conditions in three Alabama state treatment facilities.\13

These include the requirement that written orders for restraint or
seclusion be prepared by a physician or qualified health care
professional after evaluating the individual in treatment and are
valid only for 24 hours.  Emergency imposition of restraint or
seclusion in the absence of a written order may last only an hour. 
People in restraint or seclusion must have their physical and
psychiatric conditions assessed hourly and must be allowed to use the
bathroom every hour.  Some states have chosen to incorporate the
principles of this case into their own laws, which often vary as to
the type of professional who is authorized to order an emergency
restraint, the maximum length of time orders are valid, and the
frequency of required monitoring. 

--------------------
\12 Youngberg v.  Romeo, 457 U.S.  307 (1982). 

\13 Wyatt v.  Stickney, 344 F.  Supp.  373 (M.D.  Ala., 1973). 

      ACCREDITATION PROCESS LACKS
      SPECIFICS
---------------------------------------------------------- Letter :4.3

The accreditation process plays an important role in promoting
industry standards and quality improvement.  However, representatives
of health care providers and family advocates differed about whether
the accreditation process alone is sufficient to protect patients. 

JCAHO, which accredits about 80 percent of the hospitals that
participate in Medicare, has developed standards on the appropriate
use of restraint or seclusion.  JCAHO applies the same standards for
the use of restraint and seclusion to both hospitals and nonhospital
behavioral health care treatment facilities.  As part of the
accreditation survey, JCAHO surveyors conduct record reviews to
determine whether restraint or seclusion are used and documented
according to its standards and facility policy.  Routine JCAHO
surveys are conducted every 3 years.  JCAHO conducts random,
unannounced surveys on 5 percent of its accredited providers and
infrequently conducts unannounced surveys for cause.  JCAHO reports
that since adopting its current policy on voluntary reporting of
sentinel events, it has received reports of 24 restraint-related
deaths in facilities it has accredited.  On the basis of these
reports, it published a Sentinel Event Alert in November 1998 with
its summary of the root cause analyses of 20 restraint-related deaths
from its sentinel event database. 

Representatives of health care provider organizations told us that
the accreditation process is the most appropriate way to ensure that
patients are protected from the improper use of restraint and
seclusion.  They believe that a voluntary review process that does
not involve mandatory disclosure allows the facility to address any
systemic clinical problems and develop quality improvement plans for
the future.  For that reason, they believed that additional
regulation is not needed. 

In contrast, many advocates are concerned that the accreditation
process is not sufficient to establish consistent patient protection. 
Although JCAHO surveyors tour facilities and talk with patients and
staff to better understand care issues at a facility, advocates noted
that the overall process emphasizes paperwork reviews, which can miss
ongoing quality-of-care problems. 

   STATES HAVE LOWERED RESTRAINT
   AND SECLUSION USE THROUGH
   REGULATION, REPORTING,
   TRAINING, AND STAFFING
------------------------------------------------------------ Letter :5

Several states have lowered restraint and seclusion use in their
public psychiatric health systems and have instituted reporting
requirements.  Providers, advocates, and state officials indicated
that management commitment to patient protection, regulation,
reporting, and monitoring have led to increased patient and employee
safety.  However, they believe a program to reduce restraint rates
also requires effective training programs for staff, adequate
staffing, and independent oversight. 

      STATE REGULATION AND
      REPORTING HAVE LED TO LESS
      USE OF RESTRAINT
---------------------------------------------------------- Letter :5.1

In the last several years, Delaware, Massachusetts, New York, and
Pennsylvania have adopted strategies to reduce restraint use in their
public mental health or mental retardation service systems.  The
officials we met with at the state health departments indicated that
the primary reason for their success in reducing restraint use is
management commitment to achieving this goal.  Management philosophy,
not patient acuity, was the most important factor in determining
restraint use at different state hospitals, according to a 1994 study
conducted by the New York Commission on Quality of Care.\14
Management can take responsibility for shaping the overall culture in
which restraint and seclusion are either considered routine practice
or last-resort measures.  An integral part of this commitment is a
clearly delineated set of policies and procedures for staff to follow
governing the use of restraint and seclusion. 

For example, in Pennsylvania, the deputy secretary for mental health
emphasized to all hospital administrators and staff that restraint
and seclusion are not treatment but rather represent an emergency
response to a treatment failure that resulted in an individual's loss
of control.  The state Department of Mental Health issued policies
specifying that restraint or seclusion may be used only after all
other interventions have failed and when there is imminent danger of
physical harm to the individual or others.  A physician must make an
on-site assessment within 30 minutes.  According to state officials,
despite initial opposition to these restrictions within the
facilities, the Department's emphasis on maintaining adequate
staffing levels and improving crisis management training allowed it
to gain the support of psychiatrists and direct care workers. 
Pennsylvania, which administers a system of 10 facilities with over
3,000 individuals with psychiatric problems in residence, was able to
reduce both restraint and seclusion hours by over 90 percent between
1993 and 1999. 

Reporting requirements play a central role in lowering restraint use
and improving safety for people in treatment settings.  Officials in
New York and Pennsylvania stated that accurate and complete reporting
allows hospital administrators to compare their facilities with
others and focus on quality improvement within their facilities. 
This creates an incentive for administrators with high restraint
rates to find ways to reduce them so they are more in line with their
peers.  A 1999 survey by the National Association of State Mental
Health Program Directors indicated that 18 states currently require
reporting on restraint and/or seclusion use in their public
hospitals. 

In addition to tracking restraint rates, reporting of deaths or other
sentinel events to an independent agency can contribute to improved
safety for people in treatment settings.  New York is unique among
states in its long-standing comprehensive reporting requirement.  All
licensed hospitals that provide inpatient psychiatric care must
report all deaths to the Commission on Quality of Care as well as the
relevant state agency and indicate whether the individual had been
restrained or secluded within 24 hours of death.  Mandatory reporting
and investigation allow an independent entity to analyze events at
multiple facilities.  Because the Commission and other agencies
review information from the entire state, they can determine whether
incidents that appear to be isolated events from the perspective of
individual providers are actually part of a pattern.  For example,
comprehensive incident reviews led to the discovery that two
authorized restraints--the prone wrap-up and the use of a towel to
prevent biting or spitting--were associated with injuries and deaths
throughout the state.\15 As a result of these analyses, these two
types of restraint were banned. 

Some industry and physician representatives have expressed concern
that mandatory reporting requirements could thwart provider efforts
to gather information and analyze adverse outcomes.  Their concern is
that mandatory reporting to an independent body will make employees
more likely to cover up their mistakes.  This, in turn, would limit a
hospital's ability to gather all the facts it needs to identify
weaknesses that can be changed to improve care for future patients. 
However, according to a hospital industry representative in New York,
hospitals have not found this to occur in response to the state's
mandatory reporting requirement.  In fact, according to this
official, the requirement has been in place so long now that
hospitals have accepted it as a normal part of doing business. 

Another concern cited by providers relates to the increased
administrative burden associated with a new reporting requirement. 
However, all JCAHO-accredited facilities already must document
restraint or seclusion use in patient records.  In addition, both
public and private providers are currently developing performance
measures to better track quality of care. 

Restraint and seclusion use is one measure being tested by both the
public and the private sector to determine behavioral health care
quality.  The main public sector effort consists of a multiphase
feasibility study by HHS' Center for Mental Health Services.  It
began by assessing five states' capacity to measure numerous
demographic and quality indicators within their state-run psychiatric
systems and is expanding to 16 states.  On the private sector side,
JCAHO has initiated a major data compilation system--
ORYX--which will ultimately include all accredited facilities.  The
goal of both these projects is to help facilities improve care by
tracking performance measures and be able to evaluate their own
performance over time, as well as compare themselves with similar
facilities.  Although both systems include restraint and seclusion
use, not all hospitals participating in ORYX have opted to track this
measure. 

--------------------
\14 New York State Commission on Quality of Care for the Mentally
Disabled, Restraint and Seclusion Practices in New York State
Psychiatric Facilities (Albany, N.Y.:  1994). 

\15 The use of a towel had been authorized by certain hospitals as a
precaution against biting and spitting during take-down and restraint
to protect staff against possible infection.  On banning the use of
this procedure, the Commission indicated that no objects should ever
be placed over or near a patient's face because of the danger of
asphyxiation and recommended that staff wear gloves and masks and, if
necessary, wrap the patient in a ï¿½calming blanketï¿½ as a safe barrier. 
The prone wrap-up consisted of immobilizing a person in a face-down
position. 

      TRAINING AND ADEQUATE STAFF
      RATIOS HELP DECREASE
      RESTRAINT AND SECLUSION USE
---------------------------------------------------------- Letter :5.2

Clinicians, advocates, labor unions representing mental health
workers, program administrators, and providers consistently stress
that training and adequate staff-to-patient ratios are essential to
safely minimize use of restraint and seclusion.  To safely use
restraint or seclusion when there is no other option, staff need
training in how to put individuals in and take them out of restraint. 
This would include training in monitoring a restrained individual's
physical condition.  To reduce restraint and seclusion use, nurses
and other direct-care staff need to have effective alternative
methods for handling potentially violent individuals.  In the states
we visited, training programs that address how to handle potentially
violent or aggressive individuals were an integral part of the effort
to safely reduce reliance on restraint and seclusion.  In its interim
final rule implementing new hospital conditions for participation in
Medicare and Medicaid, HCFA has added requirements that hospitals
train their staffs in alternative techniques to lessen reliance on
the use of restraint and seclusion, but these requirements do not
extend to residential treatment centers or group homes. 

New York, Massachusetts, Delaware, and Pennsylvania initiated
training programs that emphasize crisis prevention.  The goal of the
training was to give staff the skills to assess potentially violent
situations and intervene early to help individuals regain control. 
State officials as well as labor union representatives stressed that
direct-care staff must be trained in alternative techniques if a
facility is serious about reducing restraint and seclusion.  Federal
officials emphasized that training should differentiate between
techniques suitable for children and those for adults. 

Officials at Delaware's ICF/MR told us that staff and patient
injuries decreased after staff had been trained in alternate ways of
managing patient behavior.  According to a patient advocate,
Delaware's emphasis on reducing restraint rates was precipitated by a
1994 restraint-related death in the state ICF/MR.  Following
implementation of a new training program that emphasized training in
crisis prevention and new priorities by management, this facility
reduced the number of emergency restrictive procedures by 81 percent
between 1994 and 1997, with the number of procedures per resident
falling from 1.38 to .29 during that time.  Along with this reduction
in restraint, the number of major injuries to residents fell by 78
percent and resident behavior improved.  A psychologist from
Delaware's ICF/MR noted that once staff have experienced success in
calming a resident through alternate means when restraint would have
otherwise been used, the new techniques become self-reinforcing
because staff prefer to use less drastic measures. 

The mental health program officials we met with indicated that
training in alternatives to restraint and seclusion and maintaining
adequate staff levels are costly, but they can save money in the long
run by creating a safer treatment and work environment.  Data from
state hospitals in New York indicated that usually facilities with
higher restraint and seclusion rates had higher rates of staff injury
and lost staff time.  A New York State official noted that many of
the injuries classified as assaults actually took place during
restraint and seclusion procedures.  Staff training has been found to
save the state money by directly reducing the frequency of
restraint-related staff injuries, which represent costs of sick leave
and overtime payments for staff to cover the shifts. 

      INDEPENDENT OVERSIGHT AND
      INVESTIGATION CONTRIBUTE TO
      PATIENT SAFETY
---------------------------------------------------------- Letter :5.3

Advocates and state administrators we interviewed often expressed the
view that the most effective monitoring system involves a combination
of internal and external oversight.  Medicare and Medicaid generally
require providers to have internal quality-of-care monitoring and
assessment programs.  JCAHO requires accredited facilities to have
quality improvement processes and to investigate the causes of
sentinel events internally.  HHS' Office of Inspector General
recently reported that the most effective system involves a balanced
combination of peer review to emphasize quality improvement and
independent regulatory oversight to ensure compliance with basic
patient safety standards.\16

External monitors complement internal quality control systems by
providing an independent perspective.  In addition to accreditation
or state licensing surveyors and P&As, some states allow trained lay
monitors to visit mental health facilities unannounced and assess
environmental conditions.  In Delaware, for example, if a monitor
reports a concern about conditions in the state psychiatric hospital,
the facility must respond within 10 days.  Because staff at the
facilities know the reports are reviewed and acted on by management,
they sometimes inform monitors about concerns that affect patient
care, such as low staffing.  In some cases, courts have appointed
independent monitors to ensure compliance with specific requirements
and safeguarding of basic patient rights in facilities that have had
serious problems. 

Investigations into specific events may be conducted by each
facility, by a peer review committee, state or federal authorities,
law enforcement agencies, or the P&As.  Some P&A directors believe
the outside review of state mental health systems is necessary to
ensure an objective look at problems within state-operated
facilities.  In Massachusetts, investigators informed us that the
peer review process and official investigations occur independently. 
However, many P&As from other states indicated that their
investigations are hampered when providers seek to preclude access to
all documents under review by the peer review committee.  The degree
to which P&A agencies can investigate deaths or injuries also depends
on each agency's priorities, relationship with the state government,
and resources.  The New York P&A is a state agency that operates
independently to review all deaths and conduct investigations.  In
Massachusetts, the P&A lacks sufficient staff to conduct individual
death investigations, but it receives reports of all deaths and
monitors the state agencies to ensure that they investigate incidents
satisfactorily.  The Massachusetts P&A staff indicated that the state
system incorporates sufficient ï¿½checks and balancesï¿½ to ensure
independent review of both state-operated and private providers. 

--------------------
\16 HHS Office of Inspector General, The External Review of Hospital
Quality:  A Call for Greater Accountability (Washington, D.C.:  HHS,
July 20, 1999). 

   CONCLUSIONS
------------------------------------------------------------ Letter :6

People with mental illness or mental retardation in residential
settings are among the most vulnerable members of our society. 
Protecting them from abuse and injury is a responsibility of the
federal government, the states, the treatment facilities, and the P&A
system.  However, the safeguards currently in place are not
comprehensive and fail to fully ensure the rights and safety of these
individuals. 

The use of restraint and seclusion represents a significant risk to
such individuals, but without more comprehensive reporting to a state
licensing body, P&A, or the federal government, the total number of
injuries and deaths each year will not be known.  Because few states
require comprehensive reporting of such events to P&As, we believe
that many more deaths occur than those reported to the P&As. 
Although P&As are charged with the responsibility to help protect
people with mental illness and mental retardation, in some states
they lack the information and resources to do so. 

The federal government does not have consistent requirements on
reporting such injuries and deaths of patients.  HCFA's
implementation of the new conditions of participation for Medicare
and Medicaid hospitals includes a reporting requirement on deaths
related to restraint or seclusion.  However, this requirement does
not apply to all Medicare- and Medicaid-funded facilities that serve
people with mental illness and mental retardation.  Reporting
injuries and deaths allows authorities to comprehensively review
patterns and identify particularly dangerous practices, and thus it
is an important step in reducing such incidents and improving safety. 
This can complement a facility's own efforts to analyze a sentinel
event and change its policies and procedures to prevent the
occurrence of similar deaths or injuries in the future. 

The experience of several states shows that use of restraint and
seclusion can be reduced and that people receiving treatment and
staff are safer as a result.  Successful strategies such as clear
guidelines and a comprehensive reporting requirement; commitment by
management; adequate staffing levels; and staff training in the safe
use of, and alternatives to, restraint and seclusion are key.  Among
the new conditions of participation for Medicare hospitals, HCFA's
requirement that hospitals train their staff in alternatives to
restraint and seclusion is a step in the right direction. 

In addition to not consistently requiring reporting, the federal
government has not implemented consistent policies on the use of
restraint and seclusion for Medicare- and Medicaid-reimbursed
facilities.  As a result, protections against the improper use of
restraint and seclusion vary widely depending on the program and
facility.  While patient protections are now included in federal
regulations governing hospitals, nursing homes, and ICF/MRs, a
significant and growing number of individuals living in other
residential settings such as residential treatment centers and group
homes lack such safeguards.  Although state regulations may offer
some protections, this protection is still not consistent among all
individuals whose treatment is funded through either Medicare or
Medicaid.  We believe that HCFA's new conditions of participation for
Medicare hospitals is a positive action, but it does not fully
protect all people with mental illness and mental retardation served
by the Medicare and Medicaid programs. 

   RECOMMENDATIONS
------------------------------------------------------------ Letter :7

We recommend that the HCFA Administrator ensure that protections
regarding the use of restraint and seclusion are consistent by
extending to all individuals receiving treatment in Medicare- and
Medicaid-funded facilities, including those in facilities covered
under a waiver program, protections such as those currently in place
for these individuals in hospitals, nursing homes, and ICF/MRs. 
These protections include a right to be free from any physical or
chemical restraints or seclusion imposed for the purposes of
coercion, discipline, or staff convenience and to receive active
treatment to reduce dependency on chemical or physical restraint or
seclusion.  Restraint or seclusion would be used only for medical or
surgical care, as part of an individual behavioral teaching program
that is intended to lead to a less restrictive means of managing and
eliminating the behavior for which the restraint is applied, or in
emergency situations when necessary to ensure the individual's or
others' physical safety and after less restrictive interventions have
been ineffective to protect the individual or others from harm. 

We also recommend that the HCFA Administrator mandate that any
hospital or residential facility that treats individuals with mental
illness or mental retardation, as a requirement for receiving
Medicare and Medicaid funds,

  -- using a uniform reporting protocol, report promptly to the state
     licensing body and the appropriate P&A all deaths and serious
     injuries among individuals with mental illness or mental
     retardation to facilitate effective investigation, and indicate
     whether restraint or seclusion was used during or immediately
     before death or injury;

  -- maintain records to document the facility's use of restraint and
     seclusion and report rates of use to HCFA periodically, using
     common definitions; and

  -- ensure that staff regularly receive training and refresher
     courses in safe methods to handle agitated or potentially
     violent individuals, including alternative methods to restraint
     or seclusion, and document staffs' receipt of the training. 

   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

We provided draft copies of this report to HHS, two of its
agencies--HCFA and SAMHSA--and JCAHO for comment.  SAMHSA and JCAHO
officials provided oral comments, and HCFA officials provided written
comments (see app.) HHS was unable to provide written comments in
time to be included in the report.  The reviewing officials also
suggested some technical corrections, which we incorporated in the
report where appropriate. 

Generally, HCFA agreed with the report's contents and concurred with
our recommendations.  HCFA stated that it believed people should be
free from inappropriate use of restraint and seclusion when receiving
care.  In response to our recommendations, HCFA detailed the steps it
has taken and plans to take to improve the quality of care.  Chief
among the actions cited was HCFA's recent interim final rule
establishing new conditions of participation for hospitals.  In
addition, HCFA referred to its collaborative work with HHS' Office of
Inspector General and its intent to work with others such as state
agencies, JCAHO, and SAMHSA to further ensure patient safety in other
treatment settings. 

Regarding our recommendation that patient protections should apply to
individuals with mental illness and mental retardation in all
Medicare- and Medicaid-funded facilities, HCFA stated that it is
studying the advisability of applying the restraint and seclusion
standards in the new conditions of participation for hospitals to
other providers.  However, HCFA cautioned that implementation of
these protections must be tailored to the specific treatment setting
in question.  While we agree that HCFA may need to take into account
differences among treatment programs in establishing patient
protections, we believe that HCFA should take action to ensure the
safety of all individuals with mental illness and mental retardation,
regardless of the setting in which they receive treatment. 

In response to our recommendation that all deaths and serious
injuries be reported to an outside agency, HCFA cited its new
conditions of participation for hospitals, which require reporting of
a death if the death is due to restraint or seclusion.  However, we
note that the experience of New York State suggests that if
facilities are given the option of determining whether a death is
related to restraint or seclusion, they do not always report all
deaths that might be related.  We believe it is important to report
all deaths and serious injuries of people in restraint or seclusion
in order to allow an independent monitor to determine whether the
death or injury was related to restraint or seclusion. 

We had suggested that HCFA maintain a database on rates of restraint
and seclusion by facility.  In response, HCFA noted that it is
planning to work with the Food and Drug Administration and SAMHSA to
determine the best way to implement this record keeping.  We consider
this interagency coordination to be an appropriate approach for HCFA
to use and therefore removed the suggestion that HCFA maintain the
database.  However, we believe it is important that one of these
agencies maintains such data.  Finally, HCFA agreed with our
recommendation on staff training, emphasizing that it has such a
provision now for hospitals and nursing homes.  HCFA said it will
examine requirements for state Medicaid agencies to ensure that
similar training requirements are in place for staff in treatment
facilities paid under Medicaid's home and community-based waiver. 

SAMHSA officials generally agreed with the report's contents.  The
officials agreed with the need for a comprehensive reporting system. 
HCFA has indicated its intent to work with SAMHSA and others to
develop an appropriate reporting mechanism.  SAMHSA officials also
suggested that they would work with HCFA and others to identify and
communicate best practices for facilities and providers to use to
avoid the use of restraint and seclusion. 

SAMHSA officials also said that, given the P&As' current lack of
resources, if additional duties are imposed on them, they will
require increased funding from the Congress.  We highlighted the lack
of resources of the P&As in the report and believe that additional
responsibilities without commensurate resources would not result in
improved patient protection. 

JCAHO officials emphasized that reporting sentinel events was only a
first step in preventing deaths and injuries.  For reporting to have
any effect, it has to be paired with an in-depth analysis of the root
cause that led to the death or injury, followed by procedures and
practices to reduce risk in the future.  JCAHO officials emphasized
that their approach--voluntary reporting with root cause
analysis--promoted a culture in both inpatient and residential
settings that was conducive to quality improvements.  In JCAHO's
opinion, mandatory reporting would not be an effective solution
unless it was accompanied by requirements for a root cause analysis
and corrective action.  Furthermore, JCAHO believes that requirements
for mandatory reporting should include safeguards so that facilities
would not lose current state protections against disclosure of peer
review results.  We agree that reporting is only part of a successful
strategy to lower restraint and seclusion use, as evidenced by the
experiences of the states we visited.  States such as Pennsylvania
and New York have used mandatory reporting to help lower restraint
use in their state hospital systems. 

JCAHO officials also emphasized that their organization has worked to
develop standards relating to restraint and seclusion use for
facilities JCAHO accredits.  One of JCAHO's goals in recent years has
been to lower the use of restraint and seclusion.  According to these
officials, JCAHO's standards have served as a model for others,
including protections in the new interim final rule for hospitals. 
Furthermore, in response to advocates' criticism of JCAHO's survey
process, the officials stated that surveys have changed in recent
years to a more in-depth evaluation of a facility's operational
compliance with JCAHO accreditation standards, including standards
that relate to restraint or seclusion. 

JCAHO officials also stressed that restraint and seclusion are
dangerous to both staff members and individuals being treated.  They
stated that while staff training in alternative techniques to avoid
restraint use was important, staff should also be trained in the
proper application and removal of restraints and in how to monitor
individuals in restraint or seclusion.  We agree that staff training
is a crucial element not only in reducing the use of restraint or
seclusion but also in ensuring the proper use of these techniques. 

---------------------------------------------------------- Letter :8.1

As agreed with your offices, unless you release its contents earlier,
we plan no further distribution of this report for 30 days.  At that
time, we will make copies available to other congressional committees
and Members of the Congress with an interest in these matters; the
Honorable Donna E.  Shalala, Secretary of Health and Human Services;
the Honorable Nancy-Ann Min DeParle, Administrator of HCFA; and
Bernard Arons, M.D., Director of the Center for Mental Health
Services, SAMHSA. 

This report was prepared by Frank Putallaz, Suzanne Rubins, and
Sondra Schwartz under the direction of Sheila Avruch, Assistant
Director.  Please call William J.  Scanlon, Director, Health
Financing and Public Health Issues, at (202) 512-7114, or me at (312)
220-7600 if you or your staffs have any questions about this
information. 

Leslie G.  Aronovitz
Associate Director
Health Financing and
 Public Health Issues

(See figure in printed edition.)Appendix
COMMENTS FROM THE HEALTH CARE
FINANCING ADMINISTRATION
============================================================== Letter 

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

*** End of document. ***