Executive Summary
Ten years ago the Institute of Medicine (IOM) declared
that as many as 98,000 people die each year needlessly
because of preventable medical harm, including health
care-acquired infections (See sidebar: Preventable medical
harm). Ten years later, we don’t know if we’ve made any
real progress, and efforts to reduce the harm caused by our
medical care system are few and fragmented. With little
transparency and no public reporting (except where hard
fought state laws now require public reporting of hospi-
tal infections), scarce data does not paint a picture of real
progress.
Based on our review of the scant evidence, we believe
that preventable medical harm still accounts for more than
100,000 deaths each year – a million lives over the past
decade. This statistic by all logic is conservative. For ex-
ample, the Centers for Disease Control and Prevention
(CDC) estimates that hospital-acquired
infections alone kill 99,000 people each
year. This needless death is unaccept-
able, and we must demand action from
our health-care system.
In this report we give the country a fail-
ing grade on progress on select rec-
ommendations we believe necessary
to create a health-care system free of preventable medical
harm.
Few hospitals have adopted well-known systems
to prevent medication errors and the FDA rarely
intervenes.
While the FDA reviews new drug names for potential con-
fusion, it rarely requires name changes of existing drugs
despite high levels of documented confusion among drugs,
which can result in dangerous medication errors. Comput-
erized prescribing and dispensing systems have not been
widely adopted by hospitals or doctors, despite evidence
that they make patients safer.
A national system of accountability through
transparency as recommended by the IOM has not
been created.
While 26 states now require public reporting of some hos-
pital-acquired infections, the medical error reporting cur-
rently in place fails to create external pressure for change.
In most cases hospital-specic information is condential
and under-reporting of errors is not curbed by systematic
validation of the reported data.
No national entity has been empowered to coordinate
and track patient safety improvements.
Ten years after To Err is Human, we have no national entity
comprehensively tracking patient safety events or progress
in reducing medical harm and we are unable to tell if we are
any better off than we were a decade ago. While the federal
Agency for Healthcare Research and Quality attempts to
monitor progress on patient safety, its efforts fall short of
what is needed.
Doctors and other health professionals are not
expected to demonstrate competency.
There has been some piecemeal action on patient safety by
peers and purchasers, but there is no evidence that physi-
cians, nurses, and other health care providers are any more
competent in patient safety practices than they were ten
years ago.
The U.S. health-care system needs nationwide mandatory,
validated and public (MVP) reporting
of preventable health care-acquired in-
fections and medical errors. Medica-
tion errors—cited as a major problem
by the IOM ten years ago—remain a
serious problem today. The FDA, doc-
tors, hospitals, and drug manufacturers
must establish better practices at every
stage of the treatment process to track
and prevent harm from medication errors. Professional
standards regarding patient safety should ensure competent
care. While some progress has been made by private initia-
tives and through purchasing policies, regulators have not
demanded universal competency testing for doctors and
nurses.
Doctors and hospitals raise concerns that public reporting
of medical harm will lead to frivolous lawsuits. But the
best way to prevent claims is to put systems in place to pre-
vent harm. Experience with public reporting in the states
demonstrates the tort concerns about such disclosures is
overstated. With a civil justice system weakened by lim-
ited compensation to harmed patients and inadequate over-
sight of health care, public reporting of preventable medi-
cal harm is today perhaps the only effective accountability
measure we have.
The current health reform debate presents a remarkable op-
portunity for improving access to health care in America
– but that health care should be safe. Patient safety needs to
be a major part of these reforms.
“How much of a problem is patient
safety? The unsettling fact is that no
one knows.”
- Dr. Lucian Leape, March 2008
To Er r i s Hu m a n – To DE l a y i s DE a D l y
Ten years later, a million lives lost, billions of dollars wasted
May 2009
Introduction
In November 1999 the Institute of Medicine (IOM) issued
the report To Err is Human, detailing a problem the pub-
lic knew of only anecdotally: doctors and other health care
professionals can make mistakes. The report also revealed
something that most people didn’t know: the U.S. health-
care system wasn’t doing enough to prevent these mistakes,
and preventable medical errors were killing as many as
98,000 people a year.
1
(See Sidebar: Behind the Statistics:
Real Lives).
“Medical mistakes 8th top killer” screamed the headline in
USA Today.
2
“Medical Errors Blamed for Many Deaths;
As Many as 98,000 a Year in U.S. Linked to Mistakes” re-
ported the front page of the Washington Post.
3
IOM report
co-author Dr. Lucian Leape compared deaths from medical
care to three fully loaded jumbo jets crashing every-other
day, a sound-bite repeated by the New York Times editorial
board.
4
“Experts Say Better Quality Controls Might Save Count-
less Lives. Washington, Are You Listening?” asked the LA
Times Editorial Board.
5
The country certainly was. The story
was featured on three major network news shows the next
morning, and carried in three major news magazines the next
week.
6
A Kaiser Family Foundation survey over the follow-
ing weeks found that more than half of Americans had heard
of the IOM report and Kaiser called the report the “most
closely followed health policy story of 1999.”
7
The IOM report estimated that medical errors cost the U.S.
$17-$29 billion a year, and called for sweeping changes to
the health-care system to improve patient safety (dened
by the IOM as
“freedom from ac-
cidental injury”
8
).
The “combined
goal of the recom-
mendations” said
the IOM, is to “…
make errors costly
to health-care orga-
nizations and pro-
viders, so they are
compelled to take
action to improve
safety.” The IOM
1 Kohn LT, Corrigan JM, Donaldson M, eds. To Err Is Human: Building a Safer Health System. Washing-
ton, DC: Institute of Medicine; 1999, p. 1.
2 Davis, Bob, and Julie Appleby. “Medical mistakes 8th top killer,” USA Today, 11/30/99.
3 Weiss, Rick. “Medical Errors Blamed for Many Deaths; As Many as 98,000 a Year In U.S. Linked to
Mistakes,” The Washington Post, 11/30/99.
4 “Preventing Fatal Medical Errors,” Editorial. New York Times, 12/1/99.
5 “Hospital Error’s High Costs; Experts Say Better Quality Controls Might Save Countless Lives. Wash-
ington, Are You Listening?” Editorial. LA Times, 12/5/99.
6 Dentzer, Susan. “Media Mistakes in Coverage of the Institute of Medicine’s Error Report,” Effective
Clinical Practice, November/December 2000.
7 Kaiser/Harvard Health News Index, November/December 1999 vol.4, No.6. Kaiser Family Founda-
tion. Internet source: http://web.archive.org/www.kff.org/content/2000/1565/HNI+Nov-Dec1999.pdf (Ac-
cessed 4/16/09).
8 Kohn, 1999, p. 58.
also called for a measurable improvement in patient safety,
stating “it would be irresponsible to expect anything less
than a 50% reduction in errors
over ve years.”
9
Within days, the Clinton ad-
ministration asked a federal
task force to examine the
IOM’s recommendations.
10
The task force quickly agreed
with the majority of the
IOM’s ndings.
11
The IOM
report spurred seven hear-
ings on Capitol Hill over the
following three months, and
soon at least ve federal bills
were led regarding medical
errors.
12
Congress allocated
$50 million to the Agency
for Healthcare Research and
Quality (AHRQ) for patient
safety research grants in the
2001 budget, citing the IOM
report as evidence of the need
for work on the problem.
13
Despite this initial urry of
activity, progress slowed once
the media moved on to the
next crisis. When the IOM
published a follow-up report in
March 2001, the release bare-
ly registered.
14
By 2004, the
deadline for the IOM’s goal of
a 50% reduction in errors, no
national medical error report-
ing bills had been passed and
the initial outrage surrounding
the report had faded. Movement towards systematic change
to the health-care system remained “frustratingly slow.”
15
Today our country has an opportunity for dramatic changes
to our fragmented healthcare system. Health reform to en-
sure that all Americans have access to high quality health
care should also include signicant and active mandates to
reduce medical harm.
9 Kohn, 1999, p. 4.
10 “Clinton orders task force to seek reduction in medical errors,” CNN.com, 12/7/1999. Internet source:
http://archives.cnn.com/1999/HEALTH/12/07/medical.errors.02/index.html (Accessed 3/16/09).
11 Doing What Counts for Patient Safety: Federal Actions to Reduce Medical Errors and Their Im-
pact, Report of the Quality Interagency Coordination Task Force, February 2000. Internet Source:
http://www.quic.gov/report/errors6.pdf (Accessed 2/27/09).
12 Kenagy, John W. and Gary C. Stein. “Naming, Labeling, and Packaging of Pharmaceuticals,” Am J
Health-Syst Pharm 58(21):3033-3041, 2001.
Shuren, Allison Weber. “Health care delivery errors: Patient safety falls prey to politics,” J Pediatr Health
Care. 15, 42-44, 2001.
13 106 Congress. “Making Omnibus Consolidated and Emergency Supplemental Appropriations for Fis-
cal Year 2001 Conference Report To Accompany H.R. 4577,” U.S. Washington DC: Governmental Print-
ing Ofce, p. 149.
14 Millenson, M L. “Pushing the profession: how the news media turned patient safety into a priority,”
Qual. Saf. Health Care 2002;11;57-63
15 Leape, Lucian L. and Donald M Berwick. “Five Years After To Err Is Human: What Have We Learned?”
JAMA, May 18, 2005—Vol 293, No. 19.
Authors:
Kevin Jewell & Lisa McGiffert
Project Team:
Consumers Union Staff
Suzanne Henry
Michael McCauley
Daniela Nunez
Eric Charping
Center for Medical Consumers
Arthur Levin
May 2009
Preventable Medical
Harm: The IOM dened
medical error as the failure
of a planned action to be
completed as intended
(error of execution) or
the use of a wrong plan
(including failure to use a
plan) to achieve an aim
(error of planning).
1
Specic
types of medical errors
highlighted in the IOM
report included error in the
administration of treatment,
failure to employ indicated
tests, and avoidable delays
in treatment.
2
The IOM
agreed that many health
care-acquired infections
(HAIs)
3
are preventable,
but relegated discussion
of HAIs to an appendix of
the report. In this report we
use the term preventable
medical harm to explicitly
include both HAIs and other
medical errors.
1 Kohn, 1999, p. 28.
2 Kohn, 1999, p. 36.
3 While the term “healthcare associated in-
fections” is used by medical professionals,
it obscures the cause-and-effect relationship
under discussion. For clarity we use “health-
care acquired infections” to refer to infections
that patients contracted while interacting
with the healthcare system. Most studies and
information about healthcare- acquired infec-
tions focus specically on hospital-acquired
infections.
Page 2
Are we safer today than we were a decade ago? Are we doing
what is necessary to end needless suffering from preventable
medical harm? It is our search for answers to these questions
that drives this report.
The IOM recommended dozens of changes to make our
health-care system safer. In this report we evaluate how the
country has fared on select recommendations we believe
necessary to create a health-care system free of preventable
medical harm.
We evaluate progress on the following IOM recommenda-
tions:
Prevent medication errors: Make the production, regula-
tion, prescribing, and delivery of medications safer.
Create accountability through transparency: Identify
and learn from medical harm through both mandatory and
voluntary reporting systems.
Measure the problem: Establish a ‘national focus’ to track
progress on patient safety.
May 2009
Behind the Statistics: Real Lives
Lewis Blackman was 15 years old when his family brought him
to the hospital for elective surgery. The surgery was predicted
to be short, and he brought the book “Dune” to read while he
recovered. Four days later he was dead; an autopsy revealed
his abdomen was lled with almost three liters of blood and
digestive uids from an undiagnosed perforated ulcer. The ul-
cer had gone undiagnosed by the doctors-in-training attending
Lewis despite indications of trouble for more than thirty hours
prior to his death. Medical experts hired by Lewis’s mother later
said that his symptoms should have suggested a routine blood
test that would have uncovered the problem.
1
A failure to order
an indicated test is a medical error.
2
As we reference the statistics of medical error in this report,
remember that behind each number is the life of someone like
Lewis.
One of the most widely cited statistics from To Err is Human
was the estimate that “at least 44,000 and perhaps as many
as 98,000 Americans die in hospitals each year as a result of
medical errors.”
3
What does this statistic mean? Where did it
come from?
The number used by the IOM was based on two studies which
performed after-the-fact reviews of a sample of medical re-
cords (one study in New York, the other in Utah and Colorado)
to estimate the rate of preventable injuries caused by medical
management. This preventable injury rate was applied to the
33.6 million hospital admissions in the U.S. to calculate the
overall magnitude of medical errors.
The resulting 44,000 - 98,000 estimate has become one of the
most widely cited statistics on medical error, but it is not without
its critics. Several subsequent articles attacked the subjective
nature of the estimate of whether or not a patient death could
be attributed to a particular error.
4
Nevertheless, defenders of the estimate focus on two reasons
that the gures are far more likely an under-estimate of the
magnitude of the problem of preventable medical harm in this
country: First, the chart review process only catches errors that
are recorded in the medical record, and evidence of many er-
rors does not appear in the record. Second, the IOM number
accounts for only medical harm in hospitals, and much health
care is delivered outside of that setting.
5
The medical error rate used to calculate the IOM’s national es-
timate has also been supported by newer studies in Canada,
Australia, and other developed countries. Based on the current
state of knowledge of medical harm, two recent patient safety
textbooks estimate that 5% of hospital admissions experience
some type of adverse error, 30% of which cause consequen-
1 Monk, John. “How a hospital failed a boy who didn’t have to die” The State, Columbia, South
Carolina. Sunday, Jun. 16, 2002, p. A1, A internet source: http://www.lewisblackman.net/ (Accessed
4/2/09).
2 Kohn, 1999, p. 36.
3 Kohn, 1999, p. 26.
4 McDonald CJ, Weiner M, Hui SL. “Deaths due to medical errors are exaggerated in the Institute of
Medicine report.” JAMA. 2000;284:93-5.
See also: Sox HC, Woloshin S. How many deaths are due to medical error? Getting the number right.
Effective Clin Pract. 2000;6:277-283.
5 Quality of Health Care in America Committee “The Institute of Medicine Report on Medical Errors:
Misunderstanding Can Do Harm” Medscape General Medicine 2(3), 2000. 9/19/2000.
Page 3
tial harm.
6
This estimate implies that more than half-a-million
people in the U.S. were harmed by preventable medical errors
last year.
7
Studies of specic errors also suggest the IOM report underesti-
mates the magnitude of medical harm. Some hospital-acquired
infections were identied in the studies used by the IOM, yet
the report hardly mentions them. In an appendix to the report,
CDC statistics are given: 2 million hospital patients and 1.5 mil-
lion long-term care patients are infected each year.
8
Most of
these are now believed to be preventable.
9
A 2007 CDC study
estimated that 99,000 deaths in the US in 2002 were associated
with HAIs.
10
The IOM study found that 7000 deaths each year
are caused by preventable medication errors.
11
The lack of a reliable measurement of medical harm is a major
challenge that must be addressed. But don’t confuse the magni-
tude with the impact. We know the impact of the problem today.
Just ask Lewis Blackman’s family.
6 Wachter, Robert M. Understanding Patient Safety McGraw-Hill 2008, p. 10.
Vincent, Charles, Patient Safety, Elsevier, 2006, p. 42.
7 Author’s calculations. 37.1 M (US Hospital Admissions in 2007) X 10% (of which are adverse events)
X 50% (of which are preventable) X 30% (of which cause consequential harm) = 556,500
2007 US Hospital Admissions Information from “Fast Facts on US Hospitals” Internet Source: http://
www.aha.org/aha/content/2008/pdf/fast_facts_2008.pdf (Accessed 4/3/09)
8 Kohn, 1999, p. 268.
9 The CDC estimates as many as 70% are preventable: Scott, R. Douglas II, “The Direct Medical Cost of
Healthcare-Associated Infections in U.S. Hospitals and the Benets of Prevention,” Division of Health-
care Quality Promotion, National Center for Preparedness, Detection, and Control of Infectious Diseases,
Coordinating Center for Infectious Diseases Centers for Disease Control and Prevention, March 2009,
p.7.
10 Klevens et al. “Estimating Health Care-Associated Infections and Deaths in U.S. Hospitals, 2002,”
Public Health Reports Vol. 122 March–April 2007.
11 Kohn, 1999, p.2
Ernst, Frank R. and Amy J. Grizzle “Drug-Related Morbidity and Mortality: Updating the Cost-of-Illness
Model,” J Am Pharm Assoc 41(2):192-199, 2001.
Expect more: Raise standards for improvements and com-
petency in patient safety for doctors and nurses and health-
care organizations, like hospitals.
Preventing Medication Errors
Implement safe medication practices.
To Err is Human identied medication errors, “as a sub-
stantial source of preventable error in hospitals.”
16
The re-
port recommended stronger oversight by the Food and Drug
Administration (FDA) to address safety issues connected
with drug packaging and labeling, similar named drugs, and
post marketing surveillance by doctors and pharmacists.
17
In
2006 the IOM returned to the issue, publishing “Prevent-
ing Medication Errors,” a report that reiterated many of the
recommendations of the 1999 report and concluded that at
least 1.5 million preventable medication errors cause harm
in the United States each year. The 2006 report estimated
that medication errors in hospitals alone cost $3.5 billion a
year.
18
Drug Confusion Errors
Many medication errors are caused by the confusion of med-
icines with similar names,
such as primidone (a seizure
medication) and prednisone
(an anti-inammatory medi-
cation). For example, the sim-
ilarities of these names led to
the death of an adolescent in
California in a case reported in
2004, despite the fact that the
potential for primidone-pred-
nisone confusion had been
identied three years earlier.
19,20
The confusion continues.
A 2008 report listed Predni-
sone as commonly confused
with 12 other drugs
21
Packaging and design can
also contribute to drug confusion errors. In a high prole
case in 2007 the twin babies of actor Dennis Quaid and his
wife were given 1,000 times the prescribed dose of the blood
thinner heparin. According to Quaid’s testimony before Con-
gress, the couple sued the drug manufacturer, charging that
the manufacturer was negligent in packaging different doses
16 Kohn, 1999, p. 182.
17 Kohn, 1999, p. 136.
18 Aspden P, Wolcott J, Bootman JL, Cronenwett LR (eds), Committee on Identifying and Preventing
Medication Errors: Preventing Medication Errors: Quality Chasm Series. Institute of Medicine of the
National Academies. Washington, National Academy Press, 2006, pp. 112, 117.
19 Pestaner, JP “Fatal mix-up between prednisone and primidone.” Am J Health Syst Pharm. 2004 Aug
1;61(15):1552.
20 “Use Caution—Avoid Confusion” USP Qual Rev. No. 76, March 2001.
21 Hicks, 2008, p. 186
of the product in similar vials with similar blue labels.
22
The
court dismissed the case on jurisdictional grounds and it is
now on appeal. This problem was not new. A year before, a
similar mix-up ocurred when six infants in a newborn inten-
sive care unit at an Indianapolis hospital were given excessive
doses of heparin, leading to the death of three of them, and
two infants at the same Indianapolis hospital had received a
similar overdose in 2001.
23
After the Indianapolis deaths, the
manufacturer issued a letter warning hospitals of the poten-
tial for confusion, but the packaging was not changed for at
least 12 months and the same packaging was still being used
in the hospital treating the Quaid children.
24
After the Quaids
threatened to sue the hospital where their twins were treated,
the hospital agreed to pay the family $750,000 and invested
$100 million in new technology to prevent similar harm in
the future.
25
Regulators ned the hospital for failure to fol-
low its own safety policies.
26
Most victims of medication error do not have the same abil-
ity to drive media attention and prompt action. There were
25,530 look-alike and/or sound-alike drug confusion errors
reported to two drug error reporting systems in the four years
2003-2006; drug labeling and packaging contributed to 7.8%
of look-alike and/or sound-alike errors.
27
With a problem of
this magnitude, we need a systematic solution to address all
of the confusion errors, not just the few that get media at-
tention.
The FDA has tested new drugs for potential name confusion
since 1999 and monitors the market for instances of confu-
sion, but few existing names are changed.
28
In an unusual ac-
tion in 2005 the FDA called for the Alzheimer’s drug Rem-
inyl to be renamed after confusion with the diabetes drug
Amaryl was implicated in two patient deaths. Reminyl was
renamed Razadyne.
29
The current statistics on look-alike/sound-alike error demon-
strates that the FDA’s effort is inadequate. The FDA is con-
ducting a pilot program to expand pre-market drug testing to
include name confusion evaluation by third parties, but the
22 Ornstein, Charles “Dennis Quaid les suit over drug mishap” Los Angeles Times 12/5/2007.
Testimony of Dennis Quaid and Kimberly Quaid Before the Committee on Oversight and Government
Reform of the United States House of Representatives, May 14, 2008; http://oversight.house.gov/docu-
ments/20080514103204.pdf.
23 Martin, Deanna “3rd Ind. preemie infant dies of overdose” Associated Press 9/20/2006 Internet Source:
http://www.boston.com/news/nation/articles/2006/09/20/3rd_baby_dies_from_drug_overdose_in_ind/
(Accessed 4/15/09)
Testimony of Dennis Quaid and Kimberly Quaid, 2008, p. 4.
24 Deutsch, Jonathan “IMPORTANT MEDICATION SAFETY ALERT BAXTER HEPARIN SODIUM
INJECTION 10,000 UNITS/ML AND HEP-LOCK U/P 10 UNITS/ML” Dear Healthcare Provider Let-
ter, Baxter. 2/6/2007. Internet Source: http://www.fda.gov/medwatch/safety/2007/heparin_DHCP_02-06-
2007.pdf (Accessed 4/15/09)
25 Lin, Rong-gong “Dennis Quaid says ‘time is running short,’ is considering suing Cedars-Sinai” LA
Times. 3/28/08 Internet Source: http://articles.latimes.com/2008/mar/28/local/me-quaid28 (Accessed
4/15/09)
“Dennis Quaid’s Medical Nightmare,” The Oprah Winfrey Show. 2/19/2009. Internet Source: http://www.
oprah.com/slideshow/oprahshow/20090219-tows-dennis-quaid/5 (Accessed 4/12/09).
“Dennis & Kimberly Quaid Agree To $750,000 Settlement From Cedars Sinai Medical Center,” Decem-
ber 15, 2008; http://www.accesshollywood.com/dennis-and-kimberly-quaid-agree-to-750000-settlement-
from-cedars-sinai-medical-center_article_12649. (Accessed 5/8/09)
26 “Quaid Hospital Case Closed,” World Entertainment News Network, 1/9/09.
27 Hicks, 2008, pp. 179, 193.
28 Holquist, Carol “How FDA reviews drug names” Drug Topics. 4/2/2001. Internet Source: http://www.
fda.gov/CDER/drug/MedErrors/reviewDrugNames.pdf (Accessed 4/14/2009)
Cohen, Robert, Newhouse news, “What’s in a name,” 8/11/08.
29 Associated Press “J&J changes Alzheimer’s drug name to avoid confusion” April 11, 2005.
May 2009
Page 4
Medication errors include
administering or prescribing
the wrong drug, wrong dose,
or wrong route of adminis-
tration to a patient. These in-
clude cases in which drugs
are provided without regard
to drug allergies or interac-
tions with other medications
the patient may be taking.
1
1 Hicks, R.W., Becker, S.C., & Cousins, D.D.
(Eds.). (2008). MEDMARX data report. A
report on the relationship of drug names and
medication errors in response to the institute
of medicine’s call for action. Rockville, MD:
Center for the Advancement of Patient Safety,
US Pharmacopeia, Preface.
pilot program won’t be nished until 2011, 12 years after the
original IOM report highlighted the problem.
30
In addition,
this FDA effort doesn’t address the problem of look-alike/
sound-alike drugs already
on the market.
Consumers Union be-
lieves the FDA should
use its authority to rig-
orously set and enforce
the naming, labeling,
and packaging standards
necessary to reduce drug
confusion errors among
new and existing drugs.
Another means to reduce
drug confusion errors is
the use of technologies
such as Computerized
Physician Order En-
try (CPOE) systems to
write prescriptions and
Bar-Code Medication
Administration (BCMA)
technology to check that
patients get the right med-
ication. Both technolo-
gies are estimated to cut
medication errors in half
or more.
31
CPOE systems
can identify and warn
prescribing physicians of
medication allergies or
interactions, remove the
challenge of handwrit-
ten records, and provide
decision support on stan-
dardized dosing.
32
CPOE
systems can be electroni-
cally linked to pharma-
cies to directly transmit
prescriptions, a process
called “e-prescribing.”
33
E-prescribing systems can be used by individual doctors in
outpatient settings as well as those working within a hospital
system.
A 2008 survey of American Hospital Association
members found that only 17% had a CPOE system in place
and operational in all units of the hospital. Another 38% had
partially operational systems or plans for systems, but al-
30 Department of Health and Human Services Food and Drug Administration “Pilot Program To Evaluate
Proposed Name Submissions; Concept Paper” Federal Register Vol. 73, No. 195. October 7, 2008.
31 Kohn, 1999, p. 191.
Poon et al. “Medication Dispensing Errors and Potential Adverse Drug Events before and after Implement-
ing Bar Code Technology in the Pharmacy” Annals of Internal Medicine 9/19/2006 Vol. 145, NO.6, pp.
426-434.
32 Wachter, Robert M. Understanding Patient Safety Lange. 2008, pp. 139-140.
33 Virk, Pushwaz et al.”Analyzing Transaction Workows in an ePrescribing System” AMIA 2006 Sym-
posium Proceedings, p.– 1129.
most half (45%) of respondents had no plans to implement a
CPOE system.
34
The IOM called for all health-care provid-
ers to be using e-prescribing by 2010.
35
A federal law passed
in 2008 offers bonus Medicare payments to physicians who
use e-prescribing beginning this year. Doctors not using e-
prescribing will face reductions in Medicare payments in
2012.
36
While technology such as e-prescribing and bar-coding are
not a panacea for medication errors, they hold promise to
improve medication safety. The 2009 economic stimulus bill
provided $19.2 billion for health information technology,
which
may encourage adoption of such systems.
37
Drug Error Reporting Systems
Several voluntary reporting systems collect information on
patient harm from medication, including FDA MedWatch,
the ISMP Medication Errors Reporting Program (ISMP-
MERP), and Quantros MEDMARX. While useful for learn-
ing about medication errors, some researchers believe that
fewer than 1 in 100 are reported to these voluntary systems.
38
Although some state adverse event reporting laws include
medication errors, no national system suitable for tracking
progress on medication errors exists.
In 2008, Bruce Lambert, a Professor at the University of Il-
linois at Chicago, commented:
“Despite all the focus on prevention, there is little evi-
dence of large-scale improvement in the wrong-drug error
rate. We are not suggesting that no one has been success-
ful at minimizing these errors, it is just that few have been
able to demonstrate convincing evidence of success, espe-
cially on a national scale. ... This represents a serious gap
in current knowledge about medication safety. The lack of
a valid, reliable, and efcient method for detecting name
confusion errors is the main reason for this gap in our
knowledge. It is a fundamental principle of quality con-
trol that if a process cannot be measured, it cannot be
improved.” [Emphasis added]
39
The bolded quote sums up a fundamental tenet of this report.
As with other preventable medical harm, the lack of a man-
datory, validated, and public (MVP) reporting system leaves
us in the dark on whether or not we are making meaningful
progress in eliminating preventable medication errors. (The
concept of MVP reporting systems are discussed in more de-
tail below.)
34 Jha AK et al. “Use of Electronic Health Records in U.S. Hospitals.” N Engl J Med. 2009 Mar 25
35 Aspden, 2006, p. 211
36 Park, Carolyne. “Rx by computer moving to state Plan created to cut errors, costs.” Arkansas Dem-
ocrat-Gazette 8/4/08
37 Robert Steinbrook, M.D. “Health Care and the American Recovery and Reinvestment Act,” N Engl J
Med. Vol. 360 n.11 pp. 1057-1060 3/12/2009
38 Hicks, 2008, p. 12.
39 Hicks, 2008, pp. 11-12.
May 2009
Are they Watching?
We may all have the best
of intentions, but knowing
we’re being watched makes a
difference in our actions. At the
beginning of a study of ICU staff
at an Australian hospital, only
12.4% of patient contacts were
preceded by a hand washing.
When informed that their hand
washing was being monitored
and group hand-washing rates
were posted in the ICU, hand
washing occurred prior to 68.3%
of patient contacts, a more than
ve-fold improvement.
1
Publishing the hand washing
rates of individual doctors
would likely have stimulated
even greater improvement. A
study of Wisconsin hospitals
found that hospitals subject to
publicly reported facility-specic
quality measures put more
effort into quality improvement
activities than hospitals
receiving condential reports
on their quality measures. This
was especially true for low-
performing hospitals.
2
1 Tibballs, James “Teaching hospital medical staff
to handwash” Medical Journal of Australia. April
1996. Internet Source: http://www.mja.com.au/
public/issues/apr1/tibballs/tibball.html (Accessed
3/14/09).
2 Hibbard, Judith H., Jean Stockard, and Martin
Tusler “Does Publicizing Hospital Performance
Stimulate Quality Improvement Efforts? Results
from a study in Wisconsin suggest that making per-
formance information public stimulates quality im-
provement.” Health Affairs Vol. 22, No. 2. March/
April 2003.
Page 5
Prevent Medication Errors - Conclusion
Progress on medication errors falls short of the IOM’s vi-
sion. While the FDA reviews new drug names for potential
confusion, high levels of error remain. Electronic prescrib-
ing systems have not been widely adopted, and no national
reporting system for medication mistakes at the facility level
exists that is Mandatory, Validated, and Public.
Create Accountability Through
Transparency
Identify and learn from preventable medical harm
through both mandatory and voluntary reporting sys-
tems.
Imagine two hospitals in your town. One slashed medica-
tion errors in half by investing in a computerized system to
assist doctors with prescription writing (eliminating the no-
torious “doctor scribble” problem). The other cuts costs by
buying cheap ballpoint pens that smudge during prescription
writing, leaving the orders illegible and doubling the rate of
dispensing errors. Do you want to know which hospital is
which?
You are not alone. Ninety-two percent of Americans believe
that hospitals should be required to report serious medical
errors, and 63% believe the reports should be public.
40
The
IOM specically recommended public reporting of harmful
medical errors so that the public could hold local health-care
systems (such as hospitals) accountable and encourage im-
provement.
The IOM panel recommended two separate national report-
ing systems: A mandatory and public reporting system de-
signed to encourage accountability, (i.e. creating external
pressure for change) and a voluntary and condential system
designed to facilitate learning about errors.
41
Progress on reporting since 1999 has been almost entirely
focused on voluntary, condential, or aggregate reporting
systems designed to facilitate learning about errors. Seven-
teen states had established condential reporting systems by
the time a federal framework for such “learning” systems
was created in the Patient Safety and Quality Improvement
Act of 2005.
42
This law prohibits the release of information
about medical harm collected by Patient Safety Organiza-
tions (PSOs) and shields hospitals that report harm.
43
Finally
implemented in 2008, any information collected by PSOs
will not be publicly disclosed by hospital or health-care fa-
cility. Under this system hospitals can learn from their mis-
takes, but you can’t.
The PSO system joins a reporting world crowded with con-
dential, learning-oriented systems. The Joint Commission
(a private membership and accreditation body) collects in-
formation on certain errors causing serious injury or death
in its Sentinel Event Database; the reports are voluntary and
the information collected remains condential.
44
Over 13
years this database has only received 113 reports of serious
hospital-acquired infections, which CDC studies estimate
claim almost 99,000 lives each year.
45
The electronic Patient
Safety Reporting System was developed for Veterans Ad-
ministration facilities. The identities of health-care facilities
reporting to the system are condential and it is operated
40 The Kaiser Family Foundation/Agency for Healthcare Research and Quality/Harvard School of Pub-
lic Health “National Survey on Consumers’ Experiences With Patient Safety and Quality Information”
November 2004.
41 Kohn, 1999, pp. 86-89.
42 CT, FL, GA, KS, MD, ME, NJ, NV, OH, OR, PA, RI, SC, SD, TN, UT, WA.
Rosenthal, Jill and Mary Takach. “2007 Guide to State Adverse Event Reporting Systems,” The National
Academy for State Health Policy. December 2007.
43 “President Signs Patient Safety and Quality Improvement Act of 2005” Press Release. 7/29/2005.
Internet Source: http://georgewbush-whitehouse.archives.gov/news/releases/2005/07/20050729.html (Ac-
cessed 3/16/09).
AHRQ website: http://www.pso.ahrq.gov/regulations/regulations.htm; Patient Safety and Quality Im-
provement Act of 2005, Public Law 109-41 109th Congress, http://www.pso.ahrq.gov/statute/pl109-41.
htm. (Accessed 5/13/09)
44 The Joint Commission. “Sentinel Event Policy and Procedures” July 2007, pp. 1, 6, 7.
45 The Joint Commission. “Sentinel Event Statistics as of: December 31, 2008,” Internet Source: http://
www.jointcommission.org/NR/rdonlyres/241CD6F3-6EF0-4E9C-90AD-7FEAE5EDCEA5/0/SE_
Stats12_08.pdf (Accessed 4/12/2009)
Klevens, 2007.
May 2009
Hospital Acquired Infection Reporting Systems:
Recently passed state hospital infection disclosure laws will
increase public accountability on Healthcare Acquired Infections
(HAIs). Some 25 states now have mandatory reporting systems
for HAIs.
1
All of these states require public disclosure of
hospital-specic rates of select HAIs, and 12 state laws require
systems to validate the data for accuracy.
2
Nebraska, Nevada,
and Arkansas have passed laws that require hospitals to report
infection data to state agencies, but this information is not
disclosed to the public.
3
Preliminary evidence in Pennsylvania – the only state reporting
on all types of HAIs - shows public reporting is an effective tool
for reducing infections: Pennsylvania’s overall infection rate
decreased by eight percent following two consecutive years of
reporting comparable infection data.
Each state has established its own reporting program, although
most are collecting data on similar types of infections via the
CDC’s National Healthcare Safety Network (NHSN), essentially
creating a national standard for collecting information on this
type of medical harm.
4
,
5
While much work remains, progress towards a National MVP
(Mandatory, Validated, and Public at the facility level) reporting
system for HAIs is underway.
1 “Reporting of Hospital Infection Rates,” Consumers Union, October 2008. Internet Source: http://
www.consumersunion.org/campaigns/Map_SHI_state_laws_10-08.pdf. (Accessed 3/27/09)
2 Validating states: CO, FL, IL, NY, NH, OH, OR, PA, RI, SC, TX, VT. Other states may be attempting
to validate the data, but their laws do not specically call for it.
3 “Summary of State Laws on Hospital-Acquired Infections,” Consumers Union, October 2008. In-
ternet Source: http://www.consumersunion.org/campaigns/CU%20Summ%20of%20HAI%20state%20
rpting%20laws%20as%20of%2010-08.pdf
4 Besser, Richard E., “CDC’s role in Preventing Healthcare Associated Infections,” Testimony before
the US House Appropriations Committee, Subcommittee on Labor, Health and Human Services, Edu-
cation, and Related Agencies, 4/1/09, p. 3. http://appropriations.house.gov/Witness_testimony/LHHS/
Richard_Besser_04_01_09.pdf. (Accessed 4/12/09)
Stricof, Rachel, “New York State Approach to Health care Associated Infection Surveillance, Prevention,
and Public Reporting,” p. 3, Testimony before the US House Appropriations Committee, Subcommittee
on Labor, Health and Human Services, Education, and Related Agencies, http://appropriations.house.
gov/Witness_testimony/LHHS/Rachel_Stricof_04_01_09.pdf. (Accessed 4/12/09)
5 GAO, “Health-Care-Associated Infections in Hospitals: An Overview of State Reporting Programs
and Individual Hospital Initiatives to Reduce Certain Infections,” GAO-08-808 (Washington, D.C.: Sept.
2008), pp. 1, 11
.
Page 6
external to the Veterans Administration (by NASA).
46
These
two systems mirror the internal secret systems operated by
most hospitals; a recent survey indicates that 98% of hospi-
tals operate some type of internal reporting system for medi-
cal harm.
47
The voluntary, condential nature of these sys-
tems prevents assessment of whether they have any impact
on the safety of patients.
The National Quality Forum (NQF) is a private membership
group that works to set “national priorities and goals for per-
formance improvement,” and publishes a list of voluntary
consensus standards related to patient safety.
48
In 2002 the
NQF endorsed a list of medical events that should “never
occur.” This list, now formally called the list of “serious re-
portable events” is often referred to as the “Never Event”
list. The list currently contains 28 serious medical errors.
49
State reporting systems based on this list cover only a small
subset of preventable medical harm.
A national mandatory and public reporting system facilitat-
ing public accountability does not exist, although fragment-
ed progress has been made at the state level. As of October
2007, 25 States and the District of Columbia operated some
type of medical error reporting system. Almost half of these
states use a variation of the “Never Event” list to determine
what type of medical harm must be reported.
50
Of the 26 mandatory medical error reporting systems, to
date, only four publicly report facility-specic information
on their websites.
51
Facility specic reporting is essential
to facilitating accountability, and when this report uses the
term “public” reporting, we refer to facility-specic report-
ing. Consider if Consumer Reports tested 50 cars and found
some performed well and others unsafe, but refused to reveal
which cars were which. The public would not be served by
such evaluation. Error information is not useful unless it is
publicly tied to the entity where the harm occurred.
Minnesota is one state that publishes facility-specic infor-
mation about patient harm on a state Minnesota Department
of Health website.
52
Seventy-two percent of Minnesota fa-
cilities surveyed in 2008 felt that the Minnesota error report-
ing law made them safer than they had been when reporting
began in 2003. One respondent said, “(Our) focus was al-
ways on patient safety, however now safety efforts are better
46 Patient Safety Reporting System Website. “Program Overview,” Internet Source: http://www.psrs.arc.
nasa.gov/ashsite/programoverview/index.html (Accessed 4/5/09)
47 Farley et al.”Adverse-event-reporting practices by US hospitals: results of a national survey,” Qual.
Saf. Health Care 2008;17;416-423. (Author’s calculation of weighted average of critical access and non-
critical access respondents.)
48 National Quality Forum “Mission – About - National Quality Forum,” Internet Site: http://www.quali-
tyforum.org/about/mission.asp (accessed 3/16/09).
The National Quality Forum “Safe Practices for Better Healthcare,” 2009. Internet Source: http://www.
qualityforum.org/projects/ongoing/safe-practices/ (Accessed 4/20/09).
49 “National Quality Forum Updates Endorsement of Serious Reportable Events in Healthcare,” Press
Release. National Quality Forum. 10/16/2006 Internet Source: http://www.qualityforum.org/pdf/news/
prSeriousReportableEvents10-15-06.pdf (Accessed 3/31/09)
50 Rosenthal, Jill and Mary Takach. “2007 Guide to State Adverse Event Reporting Systems,” The Na-
tional Academy for State Health Policy. December 2007, p. 1.
51 NY, MN, MA, IN. See http://www.safepatientproject.org/2009/05/state_medical_error_report_lin.
html
Rosenthal, Jill and Mary Takach. “2007 Guide to State Adverse Event Reporting Systems,” The National
Academy for State Health Policy. December 2007.
52 Minnesota Dept. of Health “Patient Safety - Minnesota Dept. of Health” Internet Source: http://www.
health.state.mn.us/patientsafety/ (Accessed 3/16/09).
understood by more of our staff and we prioritize this work
ahead of other work. Data is helping us to create more sense
of urgency for this work.”
53
The magnitude of certain events reported to the Minnesota
system is on the low end of what would be expected from
national estimates of the incidence of medical harm. For
example, only one death and ve signicant disabilities re-
sulting from medication errors in hospitals were reported in
the 2008-reporting year. This may reect underreporting to
the system. Minnesota health ofcials do not perform regu-
lar audits to validate the reporting level, although they do
compare event reports to death records and consumer com-
plaints.
54
Without validation, diligent reporters may appear
to perform more poorly than their peers who simply fail to
report at all. More than half of states with reporting systems
acknowledged that underreporting occurs in their system.
55
Validation, generally through random chart audits or regular
comparison to claims and billing data, counters systematic
underreporting by participants. As of January 2008, only
three states reported performing on-site audits to validate
compliance. (Sixteen states reported using more limited
validation techniques.)
56
Validation programs must be ac-
tive, ongoing and funded to be effective. The New York City
Comptroller recently reported that the state was not suf-
ciently enforcing or funding its reporting system, stating the
ability of the state program “to more broadly improve the
quality of care and reduce unnecessary costs has been seri-
ously compromised” by these shortcomings.
57
We do not have national reporting systems with the three el-
ements needed for accountability: Mandatory, Validated, and
Public at the facility level (MVP). MVP reporting systems
are needed to create the external pressure needed to create
systemic change. (See sidebar: HAI Reporting Systems)
MVP reporting would represent a sea-change in a health-care
system accustomed to hiding errors.
58
Only 14% of doctors
support public reporting of medical errors.
59
Shortly after
the IOM report, the New York Times reported that both the
American Medical Association and the American Hospital
Association “vehemently opposed mandatory reporting of
errors.”
60
Much of this resistance is driven by concerns that
public reporting would lead to frivolous lawsuits.
The best way to prevent negligence claims is to put systems
in place to prevent medical harm. Legal claims are led on
behalf of a small fraction of patients who sustain
53 “Adverse Health Care Events Reporting System: What have we learned? 5-YEAR REVIEW,” Min-
nesota Dept. of Health. January 2009. pp. 2, 9.
54 Lisa McGiffert interview with Diane Rydrych, MN Department of Health. 4/14/09.
55 “Adverse Events In Hospitals: State Reporting Systems,” The Ofce of Inspector General of the Depart-
ment of Health and Human Services. December 2008, pp. 12-13.
56 “Adverse Events In Hospitals: State Reporting Systems,” 2008, pp. 12-13.
57 “The High Cost of Weak Compliance With the New York State Hospital Adverse Event Reporting and
Tracking System” Ofce of New York City Comptroller. 2009, pp. 27-29.
58 Gibson, Rosemary, and Singh, Janardan Prasad, Wall of Silence, 2003, pp. 136-138.
59 Blendon, Robert J., et al. “Views of Practicing Physicians and the Public on Medical Errors” New
England Journal of Medicine. Volume 347:1933-1940, 12/12/2002 Number 24.
60 Pear, Robert “Clinton to Order Steps to Reduce Medical Mistakes,” The New York Times 2/22/2000.
May 2009
Page 7
injury through medical negligence. Of those led, many do
not result in an award.
61
Thus the legal system compensates
patients for a miniscule portion of the injury sustained, and
at tremendous personal cost. While physician and hospital
resistance has slowed adoption of the IOM accountability re-
porting recommendations, experience with reporting lowers
this resistance. A survey of hospital ofcers found that those
in states with mandatory reporting systems were three times
more likely to support facility-specic public reporting than
hospitals without experience with mandatory reporting.
62
Today, many states have passed tort reform laws that signi-
cantly increase the burden on people who have been harmed
by medical care and protect doctors from suits over all but
the most egregious behaviors.
63
With such a weakened civil
justice system, and a weak and inadequate administrative
oversight system in most states, public reporting of prevent-
able medical harm – and the embarrassment that might ac-
company the public release of poor results – is today perhaps
the only accountability measure we have that is both effec-
tive and reliable.
MVP reporting systems are necessary to hold all health-care
facilities equally accountable for patient safety. Consumers
Union recommends mandatory validated and public report-
ing of preventable medical harm (health care-acquired infec-
tions and medical errors), at the state and national level.
Accountability through Transparency - Conclusion
While a network of hospital-acquired infection disclosure
systems is beginning to emerge, the scope of these only
covers a small portion of the HAIs occurring. Medical er-
ror reporting systems currently in place fail to create exter-
nal pressure for change. Most states do not publicly report
facility-specic errors and many do not include a validation
requirement. Twenty-four states do not have any medical er-
ror reporting requirements in place and half of the states do
not require HAI reporting. The federal Patient Safety and
Quality Improvement Act of 2005 is voluntary and keeps the
medical error information gathered by Patient Safety Orga-
nizations condential, thereby removing a key incentive for
safety improvement.
Measure the Problem
Establish a ‘national focus’ to track progress on patient
safety.
When products are connected with deaths, we investigate
whether there are changes in them that might prevent ac-
cidents in the rst place or minimize the harm from acci-
dents when they happen. The seat belt, the child car seat, and
many technical innovations were engineered into cars, for
61 Studdert, David M. et al “Claims, Errors, and Compensation Payments in Medical Malpractice Litiga-
tion” N Engl J Med May 2006;354:2024-33.
62 Weissman et al, “Error Reporting and Disclosure Systems: Views From Hospital Leaders,” JAMA.
2005;293(11):1359-1366.
63 American Tort Reform Association, Medical Liability Reform, http://www.atra.org/issues/index.
php?issue=7338. (Accessed 5/11/09)
Public Citizen, “The Inequitable Impact Of Non-Economic Damage Caps: Three Academic Studies Dem-
onstrate Severely Injured and Female Patients Are Hurt the Most,” 2005.
example, based on this approach to accident prevention. To-
day, a car’s National Highway Trafc Safety Administration
“safety rating” is a key characteristic that buyers examine
before they lay down their money. Health care has enjoyed
no such national safety review.
“There is no cohesive effort to improve safety in health
care,” lamented the IOM in 1999.
64
The report stressed that
the fundamental problem was not that individual doctors
made errors. The fundamental problem was the failure of the
health-care system to monitor these errors, anticipate them,
and minimize the harm to patients. This failure, the IOM
noted, required a national focus on xing the health-care sys-
tem, not just the errors of individual practitioners. The IOM
recommended creation of a Center for Patient Safety within
the federal Agency for Healthcare Research and Quality
(AHRQ).
65
A Simple Checklist.
Consider the case study of one common type of medical
harm — preventable bloodstream infections. In early 2004,
researchers measured catheter-associated infections across
a set of Michigan-afliated Intensive Care Units (ICUs)
and found 7.7 bloodstream infections occurred for ev-
ery 1000 days of catheter use. A statewide safety initiative
called “Michigan Health and Hospital Association (MHA)
Keystone: ICU” set a goal of reducing catheter-associated
bloodstream infections. Inspired by and coordinated with
the successful research of Dr. Peter Pronovost and others
at Johns Hopkins, MHA Keystone instituted a short check-
list of best-practices related to catheter use and empowered
nurses to ensure that doctors were following those practices.
The initiative then tracked catheter-associated bloodstream
infection rates in 103 participating ICUs.
66
The overall results were stunning. Bloodstream infections
across the participating ICUs dropped to 1.4 per 1000 days
of catheter use, less than 20% of the rate prior to implemen-
tation of the checklist and double-checking procedures.
67
MHA Keystone estimates that the initiative saved nearly
1,800 lives over four years.
68
While the aggregate results were impressive, results were
mixed across facilities. A year and a half after the study be-
gan, MHA reported at least 50% of the participating ICUs
had completely eradicated catheter-associated bloodstream
infections. A quarter of the ICUs, however, still had infec-
tion rates of 2.4 per 1000 days or higher.
69
Unfortunately,
MHA Keystone does not identify which facilities lagged
64 Kohn, 1999, p.75.
65 Kohn, 1999, p. 9.
66 Pronovost P, Needham D, Berenholtz S, Sinopoli D, Haitao C, Cosgrove S, et al. An intervention to
decrease catheter-related bloodstream infections in the ICU. N Engl J Med 2006;355:2725-32. Internet
Source: http://content.nejm.org/cgi/content/full/355/26/2725. (Accessed 4/16/09)
Berenholtz SM, Pronovost PJ, Lipsett PA, et al. Eliminating catheter-related bloodstream infections in the
intensive care unit. Crit Care Med 2004;32:2014-2020.
67 Author’s calculations. 1.4/7.7 = 18%. Note that the hospitals used as a starting benchmark were a subset
of the hospitals in the MHA Keystone project.
68 2008 Annual Report. MHA Keystone Center for Patient Safety and Quality, p. 6; four-year estimate
was for 2004-2008.
69 Pronovost, 2006, p. 2730.
May 2009
Page 8
behind, preventing the public from discerning the hospitals
with zero bloodstream infections from the ones without sig-
nicant progress.
While the MHA Keystone: ICU initiative was not an MVP
program as envisioned by this report, the results were widely
reported and the process changes instituted by Keystone are
now the focus of several national initiatives.
70
Dr. Pronovost
won a MacArthur genius award for his work.
71
The project
was recognized as a success in part because it measured the
impact of its work. Without evidence of improvement, the
initiative’s changes may not have been continued by the par-
ticipating ICUs, let alone spurred a national movement to
adopt the process changes.
This is one local example of the type of focus the IOM en-
visioned at the national level. The panel recommended cre-
ation of an agency that would be a ‘national focal point’ on
safety in health care, much the way MHA Keystone: ICU
was a focal point for bloodstream infections in Michigan.
This agency would research and promote best-practices for
patient safety and, crucially, track and report our nation’s
progress towards ending preventable medical harm.
Tracking National Progress.
The AHRQ is the closest federal agency to the IOM’s vi-
sion of a “Center for Patient Safety” coordinating national
resources on patient safety.
72
AHRQ is charged with enhanc-
ing “the quality, appropriateness, effectiveness of health ser-
vices” in the U.S. It funds numerous research projects on
quality and safety and publishes the “National Healthcare
Quality Report,” (NHQR) to discuss and quantify progress
on patient safety.
73
The NHQR estimates national progress on patient safety
primarily through claims data on patients in the Medicare
system, hospital billing data from the states, and various
other sources like vital statistics and census data.
74
It dis-
closes no provider or facility-specic information – all data
is presented in the national or state aggregate. The agency’s
Patient Safety Indicators focus attention mostly on surgical
errors, and does not use data contained in less accessible
forms (such as patient charts).
75
The data is also stale; the
2008 report (published in 2009) discusses patient safety
data only through 2006. Such delays are a chronic problem
with health data and reduce the relevance of the report as a
70 Brody, Jane. E. “A Basic Hospital To-Do List Saves Lives,” The New York Times. January 22, 2008.
Internet Source: http://www.nytimes.com/2008/01/22/health/22brod.html (Accessed 4/16/09)
“AHRQ Awards $3 Million To Help Reduce Central Line-Associated Bloodstream Infections in Hospital
ICUs,” Press Release, October 1, 2008. Agency for Healthcare Research and Quality, Rockville, MD.
Internet Source: http://www.ahrq.gov/news/press/pr2008/clabipr.htm (Accessed 4/12/09)
71 “2008 MacArthur Fellows: Peter Pronovost,” MacArthur Foundation Website. Internet Source: http://
www.macfound.org/fellows/2008/pronovost (Accessed 4/15/09)
72 Kohn, 1999, pp. 78-79.
73 US Code Title 42. CHAPTER 6A, SUBCHAPTER VII, Part A,§ 299. Internet Source: http://www4.
law.cornell.edu/uscode/42/usc_sec_42_00000299----000-.html (accessed 3/16/09).
Agency for Healthcare Research and Quality website. “2007 National Healthcare Quality & Disparities
Reports” Internet Source: http://www.ahrq.gov/qual/qrdr07.htm (Accessed 3/16/09).
74 AHRQ website, NHRQ State Snapshots, Interpretation of Results, Examination of data sources,
http://statesnapshots.ahrq.gov/snaps07/interpretation.jsp?menuId=39&state=ID#examination (Accessed
5/11/09)
75 AHRQ Quality Indicators – Guide to Patient Safety Indicators. Rockville, MD: Agency for Healthcare
Research and Quality, 2003. Version 2.1, Revision 2, (October 22, 2004). AHRQ Pub.03-R203. p. 23.
feedback mechanism. Widespread adoption of health infor-
mation technology would allow more timely and accurate
information from clinical records. In the meantime, much
can be gleaned from claims data and hospitals should be held
accountable for ensuring the data is accurate. The lack of
timely, facility-specic information also limits its use as a
tool for consumers in making health care decisions.
The 2008 NHQR report estimates that patient safety declined
by almost 1% a year over the six years after the IOM report,
but states “[d]ata remain incomplete for a comprehensive
national assessment of patient safety.” In what is an indicator
of how little progress has been made towards accounting for
preventable medical harm, the latest AHRQ report still uses
the IOM’s 1999 work as the best estimate of the magnitude
of medical errors.
76
Without comprehensive measures of progress, we can’t say
if the indicators examined by the NHQR data accurately
represent the state of patient safety as a whole. While 17
of the 38 indicators tracked by AHRQ have declined some-
what over the last six years, in some areas not referenced by
NHQR, there is evidence that patient safety is getting rapid-
ly, not slowly, worse.
77
For example, the number of hospital
discharges with Clostridium difcile-associated infections,
which are primarily regarded as health care-acquired infec-
tions, more than doubled from 2001 to 2005.
78
The 1999 IOM report contemplated tracking national prog-
ress on patient safety through a periodic survey of medical
records, following the methods of the academic research that
provided the basis for the IOM’s original estimate 44,000-
98,000 annual deaths from medical errors.
79
Such a peri-
odic national survey has not been implemented and may
be impractical, although the adoption of electronic medical
records may make such a survey less costly and less labor
intensive.
A national MVP reporting system on preventable medical
harm would be able to ll the measurement role of a national
survey. Variations in current reporting in voluntary systems
may be due to changes in reporting compliance rather than
changes in error rates; validation of mandatory systems
minimizes such variation. A national MVP reporting system
would have the additional benet of tracking progress at the
local, as well as national, level. As discussed above, such a
system does not yet exist.
Measure the Problem - Conclusion
Ten years after To Err is Human, we have no national entity
comprehensively tracking patient safety and we are unable
76 Agency for Healthcare Research and Quality. “2008 National Healthcare Quality Report.” Rockville,
MD: U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality; May
2009. AHRQ Pub. No. 09-0001, pp. 8-9, 101.
77 AHRQ, May 2009, p. 9.
78 Elixhauser, Anne and Michael Jhung. “Clostridium Difcile-Associated Disease in U.S. Hospitals,
1993–2005” Healthcare Cost and Utilization Project Statistical Brief #50. April 2008
79 Kohn, 1999, p. 83.
May 2009
Page 9
to tell if we are any better off than we were a decade ago.
AHRQ is attempting to do this. But without the comprehen-
sive breadth needed to assess the problem, it falls short of
what is needed.
Expect More
Raise standards for competency in patient safety for
health-care professionals (like doctors and nurses) and
health-care organizations (like hospitals).
Professional standards in health care are set by government
agencies, purchasers, and professional peer groups. In 1999,
the IOM recommended a greater focus on patient safety by
regulators, accreditors and purchasers. The report called for
periodic examinations of doctors and nurses to assess “both
competence and knowledge of safety practices.”
80
Over the
past ten years, efforts to improve competency in patient
safety standards have come mostly from the private sector.
These efforts are laudable, but results are fragmented and no
systematic process exists to promote and measure national
improvement.
Fragmented Progress
The Institute for Healthcare Improvement (IHI)’s 100,000
Lives Campaign and subsequent Five Million Lives Cam-
paign were created to stimulate and measure the impact of
improved patient safety practices. This private non-prot
provided tools and technical support to more than 3,700 hos-
pitals, and the doctors and nurses working there, that agreed
to provide IHI with measures of success on at least one of 12
patient safety practices supported by the campaigns.
81
A net-
work of “mentoring” hospitals shared information regarding
methods for system changes. These campaigns introduced
many hospital workers to life-saving practices, though IHI
did not reveal which hospitals implemented which practices.
IHI publicized anecdotal evidence of the positive outcomes
of the campaign, but did not provide the public with the
results at individual hospitals.
82
Even though this was the
broadest patient safety effort of the past decade, the decision
to withhold specic validated results for the public makes it
impossible to assess the full impact it had on improving the
safety of patients.
Not all progress is private. One promising action – with-
holding payments to hospitals when patients are harmed -
was recently initiated by Medicare, the largest health care
purchaser in America. In October 2008, Medicare stopped
paying for certain preventable hospital acquired conditions.
These conditions include several hospital-acquired infec-
tions and some of the “never events” endorsed by the
80 Kohn, 1999, pp. 11-12.
81 “Reaping the Harvest: A Review of the 5 Million Lives Campaign’s First Year… and a Preview of
What’s to Come,” The Institute for Healthcare Improvement. Undated Brochure. Internet Source: http://
www.ihi.org/NR/rdonlyres/A528208C-8B71-4559-BFF3-F1FBDC4CD11C/0/ReapingtheHarvestBro-
chureFINALwebedition.pdf (Accessed 4/12/09).
IHI website, Overview, 5 million Lives Campaign, http://www.ihi.org/IHI/Programs/Campaign/Campaign.
htm?TabId=1 (Accessed 5-11-09).
82 “New Results from IHI Programs” The Institute for Healthcare Improvement Website Internet Source:
http://www.ihi.org/IHI/Results/NewfromIHIPrograms/#HAI (Accessed 4/12/09).
National Quality Forum (NQF), such as surgeries performed
on the wrong patient or part of a body and blood transfusions
with the wrong blood type.
83
Medicare’s no pay policy has increased pressure for account-
ability, and given some time could have a signicant impact
on Medicare patients and costs. Numerous states and private
health plans are following suit by adopting similar no pay
policies for some or all of the Medicare and NQF prevent-
able adverse events.
84
A similar, but private, effort to use purchaser power to im-
prove patient safety began shortly after the publication of To
Err is Human. Several large employers formed The Leap-
frog Group, which now includes many of the nation’s largest
corporations and some public agencies. The group agreed
“to base their purchase of health care on principles that en-
courage quality improvement among providers.”
85
Leapfrog
publishes annual surveys rating the compliance of respond-
ing hospitals with specic quality and safety standards, and
uses the collective purchasing leverage of its members to
stimulate improved quality and safety.
86
Insuring continuing provider competency is an important
step towards creating a safe health-care system, and ongo-
ing competency examination has been adopted by many
specialty licensing boards. The American Board of Medical
Specialties (ABMS) member boards require physicians to
demonstrate specialty-specic skills, knowledge, and use of
best-practice care to maintain their specialist certication.
87
The ABMS has recently added a patient safety self-assess-
ment program to their recertication cycle, although the
standards do not take effect until 2010.
88
Systematic Failure
Despite the action taken by the ABMS, these continuing
competency standards do not apply to the 15% of physicians
not certied by one of the 24 ABMS member boards, or
those physicians ‘grandfathered’ prior to the adoption of the
standards.
89
These remaining doctors, as well as nurses and
other health-care professionals, are primarily licensed at the
83 Tsai, Joyce. “Medicare, insurers to stop reimbursing for errors,” Dallas Business Journal. 10/17/08.
Author’s note: Medicare policy withholds additional payment follow patient harm, but will not pay at all
for wrong surgery: wrong patient, wrong site, wrong procedure.
84 Brown, Jill. “Blue Cross Plans, Providers Work to Develop ‘Never-Events’ Policies,” AIS’s Health
Business Daily. 11/3/08. Internet Source: http://www.aishealth.com/Bnow/hbd110308.html. (Accessed
4/15/09)
Wolke, Anna “Infection Correction,” State Legislatures Magazine. April 2009. Internet Source: http://
www.ncsl.org/magazine/articles/2009/09slapr09_infection.htm#me, p. 22. (Accessed 4/15/09)
85 “Leapfrog Members,” Leapfrog Group Website. Internet Source: http://www.leapfroggroup.org/for_
members/who_are_members. (Accessed 4/12/09)
“The Leapfrog Group Fact Sheet” Internet Source: http://www.leapfroggroup.org/about_us/leapfrog-
factsheet. (Accessed 3/29/09)
86 Leapfrog Group website,“For Members” Leapfrog Group Website. Internet Source: http://www.leap-
froggroup.org/for_members. (Accessed 4/12/09)
Leapfrog Group website, “What does Leapfrog ask hospitals?” http://www.leapfroggroup.org/for_con-
sumers/hospitals_asked_what. (Accessed 3/29/09)
87 “ABMS Maintenance of Certication,” American Board of Medical Specialties. Internet Source: http://
www.abms.org/Maintenance_of_Certication/ABMS_MOC.aspx (Accessed 3/30/09)
88 “New Standards Adopted to Elevate Physician Life-Long Learning Assessment for the ABMS Main-
tenance of Certication (MOC) Program” Press Release 3/26/09. Internet Source: http://www.abms.org/
News_and_Events/Media_Newsroom/Releases/release_NewMOCStandards_03262009.aspx (Accessed
3/30/09)
89 American Board of Medical Specialties. “American Board of Medical Specialties Board Certication
Editorial Background,” Internet Source: http://www.abms.org/news_and_events/media_newsroom/pdf/
abms_editorialbackground.pdf (Accessed 5/6/09)
May 2009
Page 10
state level. The IOM report called for such licensing bodies
to “implement periodic re-examinations and re-licensing of
doctors, nurses, and other key providers, based on both com-
petence and knowledge of safety practices.”
90
No state medical boards require routine testing of skills and
competency.
91
Requirements for license renewal are general-
ly limited to continuing education, despite research indicat-
ing that continuing education alone has little or no impact on
practitioner competency.
92
Once practitioners earns medical
license, they may never have to demonstrate their medical
competency again. Professionals can become incompetent
over time because they don’t keep up with current medical
knowledge, they suffer from drug addiction, alcoholism or
mental illness, or they just weren’t that good in the rst place
and their shortcomings only become evident as they treat
patients day after day. Without ongoing testing, these kinds
of problems may not be recognized by licensing agencies
before serious harm occurs.
This compares poorly with standards in other high-risk
elds. Ironically, New York City police ofcers must dem-
onstrate rearms prociency
in requalication tests at
least twice a year.
93
The Fed-
eral Aviation Administration
(FAA) requires airline pilots
to pass ongoing prociency
testing;
94
this testing is of-
ten implemented through the
use of ight simulators. The
technology for medical care
training simulators exists and
pending federal legislation
envisions more of this kind
of learning.
95
Nevertheless,
the use of medical care simu-
lation to assess physician competency is not widespread and
is not required for maintaining a license.
Hospital accreditation is another systematic attempt, like li-
censing, to ensure competency and adoption of patient safety
standards. The Joint Commission adopted priorities and pro-
tocols to increase patient safety as accreditation standards in
2002.
96
These “National Patient Safety Goals” focus on the
health care delivery process.
97
90 Kohn, 1999, pp. 134-135.
91 Our review of news reports and academic literature on continuing competency failed to identify any
states requiring routine testing of physicians after receiving their initial license. We did nd some refer-
ences to voluntary programs.
92 Swankin, David, Rebecca Arnold LeBuhn, Richard Morrison. “Implementing Continuing Competency
Requirements for Health Care Practitioners” AARP Public Policy Institute, July 2006, p. 9.
93 Rostker et al. “Evaluation of the New York City Police Department Firearm Training and Firearm-
Discharge Review Process,” RAND, 2008, p. xviii.
94 Code of Federal Regulations. 14CFR121.915 (b)(1)(i) U.S. Government Printing Ofce. January 1,
2008. Internet Source:
http://edocket.access.gpo.gov/cfr_2008/janqtr/14cfr121.915.htm (Accessed 3/16/09)
95 The Enhancing SIMULATION Act of 2009, H.R. 855 (111
th
Congress).
Press Release “Forbes Reintroduces Bipartisan Legislation to Reduce Health Care Costs,” 2/17/2009.
96 The Joint Commission Website. “Facts about the National Patient Safety Goals,” Internet Source: http://
www.jointcommission.org/PatientSafety/NationalPatientSafetyGoals/npsg_facts.htm (Accessed 3/16/09)
97 “Standards Improvement Initiative (SII): Chapter Outline; National Patient Safety Goals (NPSG);
Program: Hospital,” Joint Commission. Pre-publication copy. 2008
The progress of individual hospitals on these goals is not
publicly posted, so, after seven years of the program, the
public cannot assess the program’s impact, if any. Several
studies have outlined shortcomings in the ability of the Joint
Commission to detect serious deciencies in its accredita-
tion process. For example, a 2004 study by the GAO found
that the Joint Commission failed to identify 60% of severe
deciencies in infection control procedures identied by
state survey agencies.
98
Finally, national leaders in patient safety remain concerned
about the lack of competency in patient safety. In October
2008, the Lucian Leape Institute, founded by the National
Patient Safety Foundation to provide strategic direction
for the eld of patient safety, held a meeting of experts on
Medical Education Reform.
99
Discussions centered around
the need to change the culture of medical education as well
as the need to educate physicians on best practices sup-
ported by clinical research (“evidence-based medicine”).
The meeting sought ideas for improving the patient safety
competency of doctors, which most participants agreed as
essential to reducing medical
harm. The Institute intends to
issue a report summarizing
the recommendations of the
roundtable, but it has not yet
been released.
100
Expect More - Conclusion
There has been some piece-
meal action on patient safety
by peers and purchasers, but
no comprehensive national
action by regulators, espe-
cially in regards to the IOM’s
practitioner competency recommendations. There is no evi-
dence to assure the public that physicians, nurses, and other
health-care providers are any more competent in patient
safety practices than they were ten years ago. Relying on
private organizations to provide increased awareness and
improved patient safety practices is an arbitrary and frag-
mented process. Nothing is in place to assure the public that
a health-care professional is competent. It is practicing 21st
century medicine with 19th century oversight.
Conclusion:
National Failure on Patient Safety
Almost ten years ago, To Err is Human described the mag-
nitude of the medical error problem in the U.S. health-care
system. Despite a decade of work, we have no reliable evi-
98 GAO “Medicare: CMS Needs Additional Authority to Adequately Oversee Patient Safety in Hospitals”
GAO-04-850 July 2004, p. 14.
See also: OIG “The External Review of Hospital Quality: The Role of Accreditation” OEI-01-97-00051.
July 1999, p. 2.
99 Press Release “Lucian Leape Institute Thought Leaders Dene Strategies for Patient Safety,” National
Patient Safety Foundation. 10/30/08. Internet Source: http://www.npsf.org/pr/pressrel/2008-10-30.php
(Accessed 4/15/09)
100 Comments of conference attendee Lisa McGiffert, Consumers Union. 4/12/09.
May 2009
Page 11
“The publication of To Err Is Human was the vanguard to im-
prove patient safety. Upon nearing the report’s 10-year anni-
versary, little appears to have changed with signicant barriers
encountered when attempting to track progress.”
– Simon C. Mathews and Dr. Peter J. Pronovost, December
2008.
1
1 Mathews, Simon C. and Peter J Pronovost. “Physician Autonomy and Informed Decision Making:
Finding the Balance for Patient Safety and Quality,” JAMA. 2008; 300(24):2913-2915.
dence that we are any better off today. More than 100,000
patients still needlessly die every year in U.S. hospitals and
health-care settings – infected because of sloppy compliance
with basic cleanliness policies, injured by failure to follow
simple checklists for safety – the equivalent of a national
disaster every week of every year.
101
Since the IOM report was issued, there have been count-
less task forces, conferences, editorials, and even episodes
of Oprah focused on patient safety. But action on key recom-
mendations has been sluggish, leaving us without reliable
means to track our progress or hold the local health-care sys-
tems accountable for ending preventable patient harm. We
have failed to make the systematic changes in health care
needed to end preventable medical harm.
Next Steps
Patients, consumer organizations, and advocates alarmed by
the lack of public accountability surrounding patient safety
have issued a Patients’ Call to Action to underscore the need
for implementing the IOM’s key recommendations, includ-
ing:
effective action by the FDA, drug manufactures, hospi-•
tals, doctors, and other health-care providers to prevent
medication errors;
increased accountability through mandatory, validated •
and public reporting of preventable medical harm, in-
cluding health care-acquired infections; and
101 We have adopted the 100,000 annual estimate as the absolute minimum lower boundary of deaths
due to medical harm in hospitals in the United States. This includes 99,000 annual deaths from hospital-
acquired infections estimated by the CDC plus 2,039 deaths among Medicare patients alone from “ac-
cidental puncture or laceration.”
better training in patient safety for doctors and nurses.•
When the IOM sounded the alarm in 1999 it called for im-
mediate action and asked “Must we wait another decade to
be safe in our health system?”
102
Ten years later, we nd our-
selves asking the same question. As the nation begins to re-
form our health-care system, we have an opportunity to take
effective and accountable action to make health care safer
for all Americans. The time to act is now. We cannot wait
another decade.
102 Kohn, 1999, p. 5.
May 2009
Consumers Union’s Safe Patient Project
Consumers Union’s Safe Patient Project (www.safepatientproj-
ect.org) builds on the success of its Stop Hospital Infections
campaign. It seeks to eliminate medical harm in our health-
care system through public disclosure of health-care outcomes
(such as hospital-acquired infection rates and incidents of medi-
cal errors) and information about health-care providers (such
as complaints against and license violations of physicians and
hospitals). The campaign also works to improve drug safety by
ensuring that consumers have full information about prescrip-
tion drugs by strengthening oversight of the FDA and by ending
practices that create conicts of interest, such as drug company
gifts to doctors.
Consumers Union’s Stop Hospital Infections campaign (www.
stophospitalinfections.org) was launched in 2003 and has led a
national consumer movement for public disclosure of infection
rates in hospitals and other health-care facilities. To date, 26
states have enacted laws requiring publication of certain infec-
tion rates by hospital, and eight states have issued reports.
1
1States that have passed laws: AL, CA, CO, CT, DE, FL, IL, MA, MD, MN, MO, NJ, NY, NH, OH, OK,
OR, PA, RI, SC, TN, TX, VA, VT, WA, WVA (from “Summary of State Laws on Hospital-Acquired
Infections,” Consumers Union, May, 2009. Internet Source: http://cu.convio.net/hospital_infection_dis-
closure_laws)
States that have issued reports: CO, FL, MA, MO, NY, PA, SC, VT. (See also: http://www.stophospital-
infections.org/learn.html)
Page 12
Hospital-acquired infections
kill 99,000
Americans
each year
That’s the equivalent of a
jumbo jet full of passengers
crashing every other day
Imagine the government response if planes
started dropping from the sky.
Why isn’t Washington doing more to protect
patients from deadly hospital infections?
Safer patient care must be a priority
in any health reform package
passed by Congress, including:
Increased public •
accountability for hospital
infections and medical errors
Better training in patient •
safety for doctors and nurses
Aggressive action to •
prevent medication errors
To learn more, sign the petition
and fi nd out how you can help
,
visit:
www.SafePatientProject.org
May 2009
