First on this page are two paragraphs that summarize the rest of the page. Below that is a list of studies with summaries and links to them. They are from Health and Human Services, BMJ Quality & Safety, Journal of General Internal Medicine, American College of Physicians, and such like.

(See also Mandatory Reporting in Medicine)

2% rate of accurate reporting of patient harm problems found in this study by The United States Department of Health & Human Services (HHS)
An HHS Report in March 2010 found that reviewed hospitals did not generate incident reports for 93% of adverse events. The 7% of the time when they did generate reports, the information was inaccurate 63% of the time. Which means clinicians are willing to report accurate information about patient harm problems 2% of the time.

The summary is at this link: Reporting Never Events
A pdf of the full report is at this link:
page ii and iii. On Page 22 it mentions the problem of the medical record not containing sufficient documentation to determine that an event even occurred. This will be a problem until information is collected by a third party outside of medicine. Attention needs to be paid to outcomes in order to capture enough information to know even that harm occurred.

Another report from the Department of Health and Human Services called “Hospital Incident Reporting Systems Do Not Capture Most Patient Harm” said “Hospital staff did not report 86 percent of events to incident reporting systems, partly because of staff misperceptions about what constitutes patient harm.”
January 2012 ( OEI-06-09-00091), available at this link:
There needs to be a third party perspective outside of medicine with less resistance to recognizing harm.

A study in the International Journal for Quality in Health Care in 2015 that compared medication errors in two Australian hospitals, identified at audit, detected by staff and reported to an incident system, found no relationship between the number of reported medication incidents and the ‘actual’ rate of prescribing and medication administration errors observed (see medication error reporting on this site).

When things go wrong in medicine, there is almost no accurate reporting, neither from frontline workers nor from administrators (for instance, see Genene Jones).

As of December 2007, almost 50 percent of the hospitals in the U.S. had never reported a single privilege sanction to the National Practitioners Data Bank (NPDB). Before it opened the health care industry’s own prediction was that 10,000 reports per year would be filed, but during its first 17 years in existance the average number of reports made annualy has been only 650.

It is the same in Canada.

One Reason Why

This CDC study found one of the pressures to report falsely. In its “Survey of New York City Resident Physicians on Cause-of-Death Reporting” (covered in more detail in the second half of this page) 70.0% of respondents said they were forced to identify an alternate cause of death when the patient died of septic shock. Almost no accurate reporting in medicine when things go wrong

Septic shock is the leading cause of death in intensive care units. Is this pressure to falsify reports brought by hospitals to protect their reputations at the expense of accurate public knowledge that would enable the public to make safe choices?

Of the respondents who admitted reporting inaccurately 76.8% of the time said the form would not accept the correct cause (for instance, it would not let a clinician enter septic shock as a cause of death unless also able to indentify the cause of the septic shock, much like not being able to report that someone was stabbed to death unless also able to identify the stabber).

To improve that situation, one of the CDC’s suggestions was expanding the acceptable causes of death to include all inpatient diagnoses codes. As much as we would like to see that, that won’t solve the problems caused by computer forms forbidding them to enter the correct cause without also entering information they cannot know, a sly way for an industry to massage data in its own favor.

We have more info on cows than on MRSA infections in our communities“It’s very hard to come up with (prevention strategies) if you don’t have solid data.” – Kevin Kavanagh, MD

Also, expanding the acceptable causes of death to include all diagnosis codes carries its own prejudice by disallowing entering information that no one in medicine has seen fit to make part of the diagnosis codes. Do the latest ICD codes include codes for unfriendly practices like unnecessary treatment, overtreatment, assault, sexual imposition, rape, murder and all the other causes of harm to patients? If there is no code for them, accurate records cannot be made of them, which probably is why there is no code for them.

[When this statement has been made in the presence of health care professionals the response has been versions of, “Hopefully that is not too much of a problem.” Rape and murder are not too much of a problem outside of medicine either, so should we disband the police and courts and district attorneys because it is relatively rare outside of medicine too?]

Were any injured patients at the table representing the interests of the patient community when those codes were established? Our experience is that when patient representatives are at the table, the ones who were selected have been screened to make sure that they are on the same page as the medical community, which means they are people representing medicine’s view of what is regarded as in the interest of patients, not the view of, say, patients who have been injured and so now have had occasion to learn something about it.

What is the ICD-10 code for assault?

JACHO did a study on the frequency of assault, rape and homicide in medicine. There is a crime rate in medicine. But even if there were codes for crimes, it would be career-ending for physicians to report them, so they still wouldn’t, but at least one more excuse for failing to report would have been removed.

When patients are discovered to be experiencing ideations about harming others, if there appears to be a possibility of their acting out they are supposed to be brought swiftly to a place where an assessment can be made and any underlying medical or mental disorder can be treated. But when clinicians go beyond ideation and commit crimes against patients, there is no mechanism to address it. There isn’t even a way to report it. Clinicians respond to that with “Call the police” which shows how out of touch medicine is with this problem. An article in the Journal of Internal Medicine about Patient-Reported Incidents found that none were reported by clinicians. Patients who go to the police find that they almost never will allow patients to file a report. When the police do, and actually look into it, they they find there is nothing about it in the record and dismiss it, as though a criminal, or anyone working with a criminal, would note it in a chart when he/she intentionally caused injuries. See Kashyap.

There is no reason to imagine that the crime rate in medicine is less than the crime rate outside of it, partly because there is so little chance of getting caught. Research shows that the greatest deterrent to crime is the chance of getting caught. There probably is no one with less chance of getting caught than someone in medicine. Perpetrators figure out that everyone in medicine will protect them. Even people outside of it will. And no one believes patients. A journalist called one instance of it “pathological denial on an almost cosmic scale.”

* * *

The rest of this page covers additional studies and additional reasons for the universal failures to report when things go wrong in medicine. If you are an injured patient, you already know it from personal experience and might as well click Next below. If you are someone who doesn’t know it already, the rest of this page could be worth reading. It includes statements like the inspector general of HHS saying that hospital employees do not recognize “what constitutes patient harm” or do not realize that particular events harmed patients.

A community that doesn’t want honest reporting would evolve to make sure of that.

Back – – – – – – – – – – – – Next

* * *

1.5% report rate of adverse events in this study
This fact is cited in:
Making Health Care Safer
A Critical Analysis of Patient Safety Practices
Prepared for:  Agency for Healthcare Research and Quality, Contract No. 290-97-0013
Prepared by: University of California at San Francisco (UCSF)-Stanford University Evidence-based Practice Center

“Most hospitals’ incident reporting systems fail to capture the majority of errors and near misses. Studies of medical services suggest that only 1.5% of all adverse events result in an incident report and only 6% of adverse drug events are identified through traditional incident reporting or a telephone hotline. The American College of Surgeons estimates that incident reports generally capture only 5-30% of adverse events. A study of a general surgery service showed that only 20% of complications on a surgical service ever resulted in discussion at Morbidity and Mortality rounds.”

From Chapter 4. Incident Reporting
Heidi Wald, M.D., University of Pennsylvania School of Medicine
Kaveh G. Shojania, M.D., University of California, San Francisco School of Medicine


The full report of the above was available on line for a few years at at, but now has been archived and will take some digging to access.


BMJ: Neither medical errors nor their costs are visible in the record.
Patients with known conditions were sent for treatment in order to measure the success of caregivers in diagnosing and treating the illnesses. 399 visits were made to 111 physicians in 14 locations.

Missed or unnecessary services cost approximately $174,000 of which only $8,745 (about 5%) was discoverable in the medical records. They just don’t report it. And the people who study it refer to it only as innocent errors.

From an article by Alan Schwartz called Uncharted territory: measuring costs of diagnostic errors outside the medical record in BMJ Quality & Safety.


“Hospital Incident Reporting Systems Do Not Capture Most Patient Harm” is the title of a United States Department of Health and Human Services (HHS) report released in January of 2012. According to it, when Medicare patients are harmed in a hospital 6 out of 7 times hospital employees do not recognize or report it. That is not just events that could lead to harm not being reported. That is harm, actual harm, not being reported as well.

Daniel R. Levinson, the inspector general of HHS, says that hospital employees do not recognize “what constitutes patient harm” or do not realize that particular events harmed patients. And when they do recognize harm they do not recognize that it should be reported. According to an interview in the New York Times (at this link), he said that in some cases employees assumed someone else would report the episode, or they thought it was so common that it did not need to be reported, or “suspected that the events were isolated incidents unlikely to recur.”

This is an expansion of awareness about why there is so little accurate reporting in medicine. Now if only Mr. Levinson could add an understanding of the extent to which people interpret the evidence of their senses in self serving ways, seeing the world differently than a disinterested third party would, making it so that people with conflicts of interest (like health care professionals) never are reliable witnesses. And if after that there were some awareness of how self-interest rewrites memory (for instance, see this study), we might start developing solutions that address the fundamental causes of the problem.

This is in Report (OEI-06-09-00091), dated 1-5-12. A summary is available from the HHS site at this link where there is a link to the PDF of the entire document.


Patient Reporting / Patient-Identified Events

0% of adverse were events reported in “What Can Hospitalized Patients Tell Us About Adverse Events?” in the Journal of General Internal Medicine
Among 228 patients admitted to the medical unit of a Boston teaching hospital, the patient-reported adverse event rate was nearly 9 per 100 admissions. Serious injuries were uncommon, but two thirds were judged preventable. In addition, 4% of patients experienced near misses. Few patient-reported incidents were identified in the medical record, and none were submitted by clinicians to the hospital’s incident-reporting system.
The conclusion of the article was that inpatients can identify adverse events affecting their own care. Many patient-identified events are not captured by the hospital incident reporting system or recorded in the medical record. Engaging hospitalized patients as partners in identifying medical errors and injuries is a potentially promising approach for enhancing patient safety.
All 19 nurse respondents endorsed the statement that medical inpatients can identify problems such as errors and injuries, and that they should continue to be asked about problems, injuries, and errors experienced in the hospital.
A significant limitation of most studies is the ascertainment of adverse events based on chart review alone. Many adverse events are not recorded in the medical record, a finding attributed to variable standards for documentation, clinician unawareness or oversight, and concern about liability exposure.
A study in ambulatory care showed that chart review detected fewer than 11% of adverse drug events. . . patient contact represents an important detection approach.

From an article called “What Can Hospitalized Patients Tell Us About Adverse Events? Learning from Patient-Reported Incidents”
by Saul N. Weingart, MD, PhD, Odelya Pagovich, BA , Daniel Z. Sands, MD, MPH, Joseph M. Li, MD, Mark D. Aronson, MD, Roger B. Davis, ScD, David W. Bates, MD, MSc and Russell S. Phillips, MD
in the Journal of General Internal Medicine, Volume 20, Issue 9, 2005, pages 830-836
The abstract can be viewed at:


Eight Years After IOM Report: Patient Reports are more Reliable than Hospital’s
Daniel R Longo, ScD, Virginia Commonwealth University School of Medicine says that any study that does not take patient reporting into account is severely underestimating the extent of problems in the nation’s hospitals.

Janet Corrigan says that based on the research and knowledge that has accumulated since 1999 regarding root causes and systemic breakdowns, the IOM’s initial estimate is “probably quite low.” She is president and CEO of the National Quality Forum and helped to author the Institute of Medicine report.

When negligence is involved, an adverse event is much less likely to be reported.
Annals of internal medicine ISSN 0003-4819, 2002, vol. 136, no11, pp. 812-816 (16 ref.)
Published by American College of Physicians, Philadelphia, PA,
Authors: THOMAS Eric J. ; STUDDERT David M. ; BRENNAN Troyen A. ;
Authors Affiliations:
Brigham and Women’s Hospital and Harvard School of Public Health, Harvard University, Boston, Massachusetts, Medical University of South Carolina, Charleston, South Carolina, University of TexasHouston Medical School, Houston, Texas,

Background: The data used by the U.S. Institute of Medicine (IOM) to estimate deaths from medical errors come from a study that relied on nurse and physician reviews of medical records to detect the errors.
Objective: To measure the reliability of medical record review for detecting adverse events and negligent adverse events. Design: Medical record review.
Setting: Hospitalizations in Utah and Colorado in 1992.
Measurements: After three independent reviews of 500 medical records, the following were measured: reliability and the effect of varying criteria for reviewer confidence in and reviewer agreement about the presence of adverse events.
Results: Rates for adverse events and for negligent adverse events varied substantially depending on the degree of agreement and the level of confidence that was required among reviewers.
Conclusion: Estimates of adverse event rates from medical record review, including those reported by the Institute of Medicine in its 2000 report on medical errors, are highly sensitive to the degree of consensus and confidence among reviewers.

Patients Identify Undocumented Adverse Events
Telephone interviews with 201 patients after ED discharge identified 10 adverse events that had not been reported in their medical records. That is 5% of patients suffering adverse events with no record made of them.
— Kristi L. Koenig, MD, FACEP in Journal Watch Emergency Medicine September 26, 2008

The Silence
by Michael L. Millenson
in Health Affairs, 22, no. 2 (2003): 103-112
Millenson says that there remains within health care a refusal to confront providers’ responsibility for the problems. He suggests initiating emergency corrective-action comparable to Flexner’s crusade against charlatan medical schools.

“If you don’t concede you have failed, everything is suspect.”
– Vartan Gregorian of the Carnegie Corporation

Physicians pressured and prevented from reporting accurate cause of death.
CDC “Survey of New York City Resident Physicians on Cause-of-Death Reporting, 2010” by Barbara A. Wexelman, MD, MBA; Edward Eden, MD; Keith M. Rose, MD (available at this link)

 Of all respondents 70.0% said they were forced to identify an alternate cause of death when the patient died of septic shock. That is the leading cause of death in intensive care units. So then would the pressure to falsify reports be being brought by hospitals protecting their reputations at the expense of accurate public knowledge that would enable the public to make safe choices?

Of the respondents who reported inaccurately reporting:
76.8% said the system would not accept the correct cause (for instance, a clinician cannot report septic shock as a cause of death unless the cause of the septic shock is identified, much like not being able to report that someone was murdered unless able to identify the murderer),
40.5% said admitting office personnel instructed them to “put something else,”
30.7% said the medical examiner instructed them to do so;

Death certificates contain information critical for protecting public health. Apparently health care professionals think something else is more important a lot of the time.

To improve the situation, among other things the CDC suggests expanding the acceptable causes of death to all inpatient diagnoses codes. This would be another reason to have codes for misdiagnosis and unfriendly practices and other iatrogenic causes of harm. If there is no code for it, they cannot make accurate records when the system will not allow them to report the actual cause of death.

“Disclosing unanticipated outcome to patients: The art and practice”
by Thomas Gallagher, Lucian Leape and others
Journal of Patient Safety, September 2007
“. . . available evidence suggests that open communication of unanticipated outcomes occurs infrequently.”

Infection tracking raises rate of reporting
The federal Agency for Health care Research and Quality reported that the rate of adverse events—a key measure of patient safety defined as unintended harm during medical care—has risen by about 1% in each of the past six years, in part because of a rise in hospital infections. Other studies show that one in seven hospitalized Medicare patients experience one or more adverse events. One in 15 hospitalized children are harmed by medication errors. And those are just the ones that get recorded.


Publicizing errors motivates health care to improve
Judith Hibbard, a professor of health policy at the University of Oregon, says, “We’ve done some research on this. And making it public makes a big difference.” Her research found that hospitals improved safety more when their mistakes were made public than when they were just reported back to the hospital or not reported at all. “The hospitals were primarily motivated to improve because of their public image and reputation,” she said. “Just knowing that they’re not doing great isn’t enough. Unless there’s something at stake, they’re not motivated to change.”

Lisa McGiffert, director of the Safe Patient Project of Consumers Union, is another advocate of public reporting of errors. She said, “Reporting is a form of accountability. And hospitals should be accountable to the public they serve. . . And that should include reliable information about how safe their hospital is, how safe their care is.”

Joe Nocera of the New York times calls it the Heisenberg Journalism Principle. In physics the Heisenberg Uncertainty Principle says that the process of observing subatomic particles affects their behavior. Nocera points out that the process of observing people and institutions affects their behavior too.

According to Diane Suchetka in The Cleveland Plain Dealer, May 11, 2010, some medical safety experts say public announcement of errors is the best way to prevent them.

For peer reviewed academic studies affirming this, see:
1) Ehrenkranz NJ. Surgical wound infection occurrence in clean operations. Am J Med 1981;70:909-14.


2) Cruse PJE, Foord R. The epidemiology of wound infection: a 10-year prospective study of 62,939 wounds. Surg Clin North Am 1980;60:27-40


A possible solution
From Slate, “The Wrong Stuff,”
By Kathryn Schulz, Monday, June 28, 2010

According to James Bagian, director of the Veteran Administration’s National Center for Patient Safety, people report only when they feel it’s in their interest to do so. Since you can’t fix problems you don’t know about, it is important to get reports that reveal what has gone wrong. So how do you get them to report?
In 1998 they did a survey and found that the chief reason people don’t report is embarrassment and humiliation. In response, the VA devised a definition of “blameworthy acts,” which must include either assault, rape, or larceny, or being drunk or on illicit drugs, or purposely doing something unsafe. In the ten months after implementing that definition he says reporting went up 3000 percent.
One must be encouraged by anything that appears to increase reporting, but one also must wonder how much was actually accomplished. Their definition suggests that the only things about which people should or would be embarrassed are, for the most part, crimes, as though there is no reason to be embarrassed about indifference or incompetence or ignorance or anger or jealousy or prejudice or declining physical strength and coordination or any number of other things. And 3000% of what? If the rate of reporting at the beginning was low enough, they might still be reporting only a single digit percentage of what needs to be reported. (See the report higher on this page that found that none of the adverse events reported by patients had been submitted by clinicians to the hospital’s incident-reporting system in “What Can Hospitalized Patients Tell Us About Adverse Events? Learning from Patient-Reported Incidents” in the Journal of General Internal Medicine, Volume 20, Issue 9, 2005, pages 830-836.) A 3000% increase is not much when starting at zero.
On top of that, what has been left off the table in this effort is the most important thing that could be included: crimes like assault and rape and all the other sins in which humans indulge when they believe they can get away with them.
Mr. Bagian continued saying that the only reason to have reporting is to identify vulnerabilities, not to count the number of incidents. Reports are never good for determining incidence or prevalence, because they’re essentially voluntary. Even if you say “You must report,” people will report only when they feel it’s in their interest to do so.

So even he is saying that the reports on which he bases his 3000% number are not reliable for determining numbers like his 3000% number. Any increase in reporting is encouraging, but it would be nice to get a second opinion on just what happened, and how much of it happened, as a result of whatever management practices were implemented at the VA.

For instance, a Government Accountability Office report and a review by the VA Office of Inspector General pointed to weaknesses in policies and oversight of medical equipment at VA facilities after the above figures were published.

“Despite changes to improve VA’s oversight with selected reprocessing requirements, weaknesses still exist,” the GAO report states, “weaknesses [that] render VA unable to systematically identify and address noncompliance with the requirements. . .” (

So how reliable is the information he put out stating what large improvements have been made? I suspect not very. Which is the way it is throughout medicine in matters like this. They keep managing to persuade themselves that they have made great strides in spite of none of the numbers changing, numbers like the rate of unnecessary death, injury, misdiagnosis, etc.

In addition to being director of the Veteran Administration’s National Center for Patient Safety, James Bagian is, among other things, an engineer, an anesthesiologist, a NASA astronaut, and a mountain rescue instructor.

This is not new

The example of Dr. Benjamin Rush 200 years ago shows that none of this is new. What is new is that for the first time in history there is a practical way to gather from patients the information that people in medicine never have reported and probably never will. For instance, does someone really think a pedophile or other sex abuser is going to file a report him or herself? If you think their colleagues will report them, you haven’t been a victim.

“Real power begins where secrecy begins.”
– Hannah Arendt

Arendt's "Origins of Totalitarianism"The determination of health care professionals to cover up sins and errors at all costs is born out by the scholarly journal articles cited on this page. Studies done by everyone from AHRQ to HHS to the Annals of Internal Medicine show how little accurate reporting there is in medicine. The rate at which adverse events accurately are reported can be referred to as 2% or 3%. Injured patients have known this for decades. Health care professionals almost universally are in denial about it. They have hidden the problems so well that even they themselves cannot see them.

“The immediate silence from the physicians, nurses and the hospital’s chief executive officer was deafening. I felt connected with the staff, having worked for an HMO in the area for a number of years. Now, all I was seeing was a classic pattern of denial and defense.”

From: The aftermath of a ‘never event’ – A child’s unexplained death and a system seemingly designed to thwart justice
By: Dale Ann Micalizzi
Modern Healthcare March 3, 2008

In the same breath that health care professionals insist that they and their colleagues report everything, they say that “of course you have to protect yourself from lawsuits” – which is done by not reporting anything.

In the book Justin’s Hope (links to another site) Dale Ann Micalizzi said: “I never wanted lawyers involved. I never wanted to question a physician’s judgment or a hospital’s care. There was no other option available to us. . . “

“. . . it isn’t worth a lawyer’s time to accept such a case unless the evidence is overwhelming and an easy settlement attainable.”

Of the patients with legitimate grievances, only one in some thousands can get a lawyer (see Studdert). And even they have to sign gag agreements. So the patient community does not learn where most injuries are incurred. Neither does the medical community. Caregivers do not share their failures with each other except in rare circumstances. An inept clinician easily can define his/her experiences in a way that prevents even his/her own recognition of having a poor success rate. Arendt's "Origins of Totalitarianism"No one disinterested is keeping track.

“The reason why the totalitarian regimes can get so far toward realizing a fictitious, topsy-turvy world is that the outside nontotalitarian world . . . indulges also in wishful thinking and shirks reality. . .”
Hannah Arendt

Back – – – – – – – – – – – – Next

(See also Mandatory Reporting in Medicine)
When records appear to be incomplete, the patient and their family or other advocates can turn to the Office of Civil Rights at the Department of Health and Human Services, Rhodes says.