Take accreditation seriously; your future may depend on it
Take accreditation seriously; your future may depend on it
By David L. Freedman, MD, JD, FAAEM, Emergency Medicine Physician, Chelsea Community Hospital, Chelsea, MI; Attorney, Miller, Canfield, Paddock & Stone PLC, Ann Arbor, MI.
Most physicians realize that JCAHO stands for the Joint Commission on Accreditation of Healthcare Organizations. However, few realize the crucial importance that JCAHO accreditation has for hospitals. Everyone knows that a JCAHO survey visit means that previously unlocked doors and cabinets must be locked, medications routinely left accessible in patient examination and treatment rooms must be removed, etc. We also all know that as soon as the surveyor leaves, we tend to return to "business as usual." Physicians may not think JCAHO is that important, or that its accreditation and survey visits are something to joke about. Not so. As a general rule, a hospital cannot survive without Medicare and, for all practical purposes, loss of JCAHO accreditation means no Medicare reimbursement. In short, the survival of most hospitals depends on JCAHO accreditation.
Based in Oakbrook Terrace, IL, JCAHO (sometimes referred to as the "Joint Commission") is the oldest private accreditation agency in the United States, having been formed in 1951 as an "independent," non-profit organization by the American College of Physicians, the American College of Surgeons, the American Hospital Association, the American Medical Association, and the Canadian Medical Association (the Canadian Medical Association later withdrew in 1959 to form its own accrediting organization in Canada, and the American Dental Association joined as a corporate member in 1979). JCAHO began offering hospital accreditation in 1953 and now has accreditation programs for a variety of health care organizations in addition to hospitals. Currently, JCAHO evaluates and accredits more than 19,500 health care organizations in the United States. Its stated mission is to "improve the quality of care provided to the public through the provision of health care accreditation and related services that support performance improvement in health care organizations."1 Joint Commission accreditation is crucial for most hospitals. Hospitals love to issue press releases announcing that they have achieved Accreditation or, better yet, Accreditation with Commendation. This information is also excellent material for the hospital’s advertising and public relations campaigns. This is not, however, the most important aspect of JCAHO accreditation. The Medicare Conditions of Participation provide that the government may accept accreditation by an organization approved by the Health Care Financing Administration (HCFA) (e.g., JCAHO and American Osteopathic Association), as evidence that the organization meets Medicare requirements for participation. As a result, accreditation by JCAHO results in "deemed status" for an accredited hospital. That is, a hospital is deemed to have met the Medicare Conditions of Participation because of its JCAHO accreditation.2 In addition to HCFA deemed status, some states recognize private accreditation in their licensing process. The alternative to JCAHO (or AOA) accreditation for hospitals is review by a state agency. This is not the preferred method, as evidenced by the fact that approximately 80% of the 6,200 hospitals that participate in Medicare opt for Joint Commission review and accreditation.
The Joint Commission introduced its Sentinel Event Policy in 1996. This policy has been revised repeatedly since its inception in response to various criticisms and remains a subject of significant concern for hospitals and other health care organizations. The majority of this article will consist of a discussion of JCAHO’s Sentinel Event Policy. While most reported hospital Sentinel Events fortunately do not take place in the emergency department, we can certainly learn from adverse events that occur elsewhere in the hospital and adopt systems and processes to avoid their occurrence in our emergency departments.
The discussion in this article is primarily focused on hospitals. JCAHO accreditation, which began for hospitals, is now available for many other health care facilities and organizations (e.g., health care networks, home care organizations, nursing homes and other long-term care facilities, behavioral health care organizations, ambulatory care providers, and clinical laboratories), and can be as crucial to them as it is for hospitals. Purchasers of health care, other than the federal government, may also use JCAHO accreditation as a measure of quality and require that organizations that provide such care for them be accredited. For example, Xerox and Pepsi Co. now mandate that the managed care organizations with which they contract must be accredited.3 States may also rely on the private accreditation process as part of their licensing regimen. For example, in Florida, for an HMO to maintain its certificate of authority, it must become accredited by an approved accrediting organization within two years of its initial receipt of a certificate.4
There are an increasing number of other private accrediting agencies currently providing accreditation for various health care organizations. These include: the National Committee for Quality Assurance (NCQA), which accredits MCOs, new health plans, and managed behavioral health organizations, and credentials verification organizations (CVOs) and physician organizations; the American Accreditation Health Care Commission (AAHCC), which accredits health utilization management programs, health networks, network practitioner credentialing programs, workers’ compensation utilization management services, workers’ compensation networks, and CVOs; the American Osteopathic Association (AOA); the American Medical Association (AMA); the Community Health Accreditation Program (CHAP); and the Commission on Accreditation of Rehabilitation Facilities (CARF).5
Seven JCAHO Accreditation Levels
Accreditation with Commendation. The highest level of accreditation and one which we see so commonly in hospitals’ advertising materials. Accreditation with Commendation requires a minimum score of 90 and, if this level is achieved, no follow-up monitoring of the hospital is required.
Accreditation. A hospital that complies with JCAHO’s standards in all performance areas is granted Accreditation.
Accreditation with Type 1 Recommendations. If a hospital is not sufficiently compliant in a specific area, it will be given Accreditation with Type 1 Recommendations. In order for the hospital to maintain accreditation, it must correct the Type 1 Recommendations within a specific period of time.
Provisional Accreditation. A hospital may be preliminarily surveyed as to selected standards and, if those standards are met, receive Provisional Accreditation. A full survey will be conducted approximately six months later, at which time an official accreditation decision will be made.
Conditional Accreditation. An organization that is not in substantial compliance with JCAHO standards, but that JCAHO believes is capable of compliance, may be given Conditional Accreditation. In order to receive this level of accreditation, the hospital must submit a plan of correction which is judged to be appropriate by JCAHO and must demonstrate improvement at the time of a follow-up survey in six months. At that time, a final accreditation decision is made.
Preliminary Nonaccreditation. If a hospital is not in compliance with JCAHO standards, or if JCAHO has made a decision to withdraw the hospital’s accreditation, it will receive Preliminary Nonaccreditation. This category is used if there is additional information that might be relevant to a final accreditation decision.
Nonaccreditation. A hospital receives Nonaccreditation when accreditation is denied because of noncompliance with JCAHO standards, when JCAHO withdraws a hospital’s accreditation, or when the hospital withdraws from the JCAHO accreditation process.
Accreditation Watch. It is not an official accreditation category, but rather an accreditation status. It was added in conjunction with the institution of JCAHO’s Sentinel Event Policy. The implications of Accreditation Watch are discussed below in the Root Cause Analysis section.
Evidence of Negligence
Failure to meet a JCAHO accreditation standard does not itself establish that a hospital, or its personnel, were negligent. However, noncompliance with a JCAHO accreditation standard may, depending on the state, be used as evidence of the hospital’s negligence. The leading case on this issue is Darling v Charleston Community Memorial Hospital, a case taught to all students of health law.6
In Darling, a lawsuit was brought against a hospital on behalf of an 18-year-old boy alleging negligent medical and hospital treatment. The physician had been dismissed just prior to trial, presumably because of a settlement. On Nov. 5, 1960, the plaintiff fractured his leg while playing in a college football game. He was taken to the hospital emergency room where Dr. Alexander was on "emergency call that day." With the assistance of other hospital personnel, Dr. Alexander applied traction and subsequently placed the leg in a plaster cast.
"Not long after the application of the cast plaintiff was in great pain and his toes, which protruded from the cast, became swollen and dark in color. They eventually became cold and insensitive."7 On the evening of the next day, the doctor "notched" the cast around the toes, and "on the afternoon of the next day he cut the cast approximately three inches up from the foot."8 It was not until another two days had passed that the physician split the sides of the cast. In splitting the cast, the physician lacerated the patient’s leg on both sides. This was, however, the least of the problems in this case. Upon splitting the cast, blood and other "seepage" were observed by the nurses and others and "there was a stench in the room, which one witness said was the worst he had smelled since World War II."9
On Nov. 19, two weeks after admission, the boy was transferred to Barnes Hospital in St. Louis where he was under the care of Dr. Fred Reynolds, the Head of Orthopedic Surgery at Washington University School of Medicine and Barnes Hospital. Dr. Reynolds determined that there was considerable ischemic necrosis in the leg, caused by "swelling or hemorrhaging of the leg against the construction of the cast."10 After several surgeries were performed in an attempt to save the leg, the boy underwent a below-the-knee amputation of his leg.
The apparent negligence of the physician in treating the fracture immediate casting of a fracture, in particular, one that might have been predisposed from the onset to result in a compartment syndrome, and then failing to recognize what appears, in retrospect, to have been obvious signs of ischemia is not the focus of our interest in this case. This is the leading case establishing, among other things, that in some situations, a hospital may have direct liability for not preventing staff physicians from providing negligent treatment to patients in the hospital, including physicians who are independent members of the medical staff, not employees or agents of the hospital.
The plaintiff contended that the hospital was negligent in permitting Dr. Alexander to do orthopedic work of the kind required in this case, and not requiring him to review his operative procedures to bring them up to date; in failing, through its medical staff, to exercise adequate supervision over the case, especially since Dr. Alexander had been placed on emergency duty by the hospital; and in not requiring consultation, particularly after complications had developed.11
There was also an allegation that the nurses, who were hospital employees, were negligent in their care of the patient.
The plaintiff argued that, because the hospital was a licensed and accredited hospital, the "licensing regulations, accreditation standards, and its own bylaws define the hospital’s duty, and that an infraction of them imposes liability for the resulting injury."12 As to its potential liability for the actions of its physicians, it was the hospital’s position that:
"The extent of the duty of a hospital with respect to actual medical care of a professional nature such as is furnished by a physician is to use reasonable care in selecting medical doctors. When such care in the selection of the staff is accomplished, and nothing indicates that a physician so selected is incompetent or that such incompetence should have been discovered, more cannot be expected from the hospital administration."13
The court addressed two issues of interest to our discussion: 1) the effect to be given to evidence concerning the community standard of care; and 2) "the effect to be given to hospital regulations adopted by the State Department of Public Health under the Hospital Licensing Act . . . , to the Standards for Hospital Accreditation of the American Hospital Association, and to the bylaws of the defendant."14
As to the first issue, the Court held that evidence of the community standard of care was relevant in deciding whether the hospital was negligent; however, it was not conclusive. We will defer further discussion of this interesting issue to another day. Our primary interest at this time is the Court’s opinion on the second issue, in particular, the relevance of private accreditation standards as evidence of the standard of care.
In deciding the relevance of accreditation standards (as well as licensing regulations and bylaws), the Court analogized to its holding as to industry custom. "In the present case the regulations, standards, and bylaws which the plaintiff introduced into evidence, performed much the same function as did evidence of custom."15 As such, the accreditation standards were relevant as to the standard of care and were properly admitted into evidence at trial, but were not conclusively determinative of the standard of care.
This case stands for a number of principles. Of primary interest to our topic this month is the Court’s holding that accreditation standards may be introduced as evidence of the standard of care. A more significant holding of the Court, which we must at least mention, is that hospitals have direct responsibilities to patients and may be held directly liable to patients if the hospital is negligent in carrying out its non-delegable duties. That is, a hospital’s liability is no longer limited to vicarious liability liability based upon the fact that the individual who was negligent was an employee or agent of the hospital.
Sentinel Event Policy
JCAHO instituted its Sentinel Event Policy in 1996 after several well-publicized health care treatment errors during 1995 at accredited health care organizations focused attention on patient injury and quality of care (e.g., amputation of the wrong leg).16 The policy has undergone multiple modifications since its inception and further refinement of the policy is likely to continue on an ongoing basis. The purpose of JCAHO’s Sentinel Event Policy is twofold. First, facilities are encouraged to voluntarily self-report significant adverse events, are required to conduct Root Cause Analysis in an effort to identify the underlying cause, or causes, of the Sentinel Event, and are required to develop systems and procedures to prevent, or at least decrease the frequency of, similar adverse events in the future. This is the "Action Plan." Second, JCAHO hopes that, by collecting reports of significant adverse events and analyzing the subsequent Root Cause Analyses on a national basis, it will be better able to understand how and why such events occur, and will be able to develop procedures to prevent similar events in the future.
There are two basic components of the Sentinel Event Policy: 1) the identification and voluntary self-reporting of "Sentinel Events;" and 2) the use of "Root Cause Analysis" to identify the basic cause or causes of the event. JCAHO classifies certain events as "sentinel" because "they signal the need for immediate investigation and response."17 JCAHO’s definition of "Sentinel Event" has undergone several modifications since 1996. Currently, a Sentinel Event is defined as: an unexpected occurrence or variation involving death or serious physical or psychological injury, or the risk thereof. Serious injury specifically includes loss of limb or function. The event is called "sentinel" because it sends a signal or sounds a warning that requires immediate attention.18
The definition was clarified in May 1998, at which time JCAHO stated that Sentinel Events are subject to review by JCAHO and may be reported on a voluntary basis if they "result in an unanticipated death or major permanent loss of function, not relating to the patient’s illness or underlying condition," or if the event is one of the specific events listed in the following paragraph. 19
"Major permanent loss of function’ means sensory, motor, physiologic, or intellectual impairment not present on admission requiring continued treatment or life-style change. When major permanent loss of function’ cannot be immediately determined, reporting is not expected until either the patient is discharged with continued major loss of function, or two weeks have elapsed with persistent major loss of function, whichever occurs first." 20
The following events are specifically included in the definition of reviewable Sentinel Events: suicide of a patient in a setting where the patient receives around-the-clock care, such as a hospital, a residential treatment center, or a crisis stabilization center; infant abduction or discharge to the wrong family; rape; hemolytic transfusion reaction involving administration of blood or blood products with major blood group incompatibilities; or surgery on the wrong patient or body part.21
JCAHO distinguishes between adverse outcomes that are related to the natural course of a patient’s illness or underlying condition (not reportable) and death or permanent loss of function that is associated with the patient’s treatment, or lack thereof, of the condition (voluntarily reportable).22 Examples of voluntarily reportable Sentinel Events are listed in Table 1.
| TABLE 1: Voluntary Sentinel Events |
| • Any patient death, paralysis, coma, or other major permanent loss of function associated with a medication error. |
| • Any suicide of a patient in a setting where the patient is housed around-the-clock, including suicides following elopement from such a setting. |
| • Any elopement, i.e., unauthorized departure, of a patient from an around-the-clock care setting resulting in a temporally related death (suicide or homicide) or major permanent loss of function. |
| • Any procedure on the wrong patient, wrong side of the body, or wrong organ. |
| • Any intrapartum (related to the birth process) maternal death. |
| • Any perinatal death unrelated to a congenital condition in an infant having a birth weight greater than 2500 grams. |
| • Assault, homicide, or other crime resulting in patient death or major permanent loss of function. |
| • A patient fall that results in death or major permanent loss of function as a direct result of the injuries sustained in the fall. |
| • Hemolytic transfusion reaction involving major blood group incompatabilities. |
| NOTE: An adverse outcome that is directly related to the natural course of the patient's illness or underlying condition, e.g., terminal illness present at the time of presentation, is not reportable except for suicide in, or following elopement from, a 24-hour care setting (see preceding list). |
| (Source: "Accreditation Committee Approves Examples of Voluntarily Reportable Sentinel Events," Sentinel Event Alert, May 11, 1998.) |
| Examples of events that JCAHO considers not to be reportable are listed in Table 2. |
| TABLE 2: Nonreportable Events |
| 1. Any "near miss." |
| 2. Full return of limb or bodily function to the same level as prior to the adverse event by discharge or within two weeks of the initial loss of said function. |
| 3. Any sentinel event that has not affected a recipient of care (patient, client, resident). |
| 4. Medication errors that do not result in death or major permanent loss of function. |
| 5. Suicide other than in an around-the-clock care setting or following elopement from such a setting. |
| 6. A death or loss of function following a discharge "against medical advice (AMA)." |
| 7. Unsuccessful suicide attempts. |
| 8. Unintentionally retained foreign body without major permanent loss of function. |
| 9. Minor degrees of hemolysis with no clinical sequelae. |
| (Source: "Accreditation Committee Approves Examples of Voluntarily Reportable Sentinel Events," Sentinel Event Alert, May 11, 1998.) |
Once a Sentinel Event has been identified, the basic cause or causes should be identified through the use of Root Cause Analysis: "a process for identifying the most basic or causal factor or factors that underlie variation in performance, including the occurrence of a sentinel event." 23
Initially, the Sentinel Event Policy provided that, in response to a serious adverse event, JCAHO could immediately put an accredited facility on Conditional Accreditation, if the facility was determined to be accountable for the event. JCAHO made a significant change in its policy in October 1996, when it replaced its Conditional Accreditation approach with a new accreditation status, "Accreditation Watch."24 Under this revised version, if JCAHO became aware of a Sentinel Event, from whatever source or sources, the policy called for JCAHO to conduct an on-site investigation of the event and the organization could be placed on Accreditation Watch. An organization could still maintain its level of accreditation while being on Accreditation Watch. The fact that an entity had been placed on Accreditation Watch was not confidential and, in response to an inquiry, JCAHO would acknowledge that it was aware of the event and was working with the entity to address the concern.25
The Sentinel Event Policy was again modified, effective April 1998. In this modification, the definition of what events were considered significant enough to be reviewable was modified and a five-day self-report opportunity was added. The five-day self-report provision did not last long it was eliminated effective July 1, 1998. The incentive for self-reporting had been avoidance, in most cases, of the imposition of Accreditation Watch status and the requirement of an on-site inspection. Under the July 1, 1998, revision of the policy, in response to an inquiry during the 30-day period allowed for Root Cause Analysis, JCAHO would report the accreditation status of the organization without referencing the Sentinel Event. However, if asked specifically about the Sentinel Event, JCAHO would disclose that it was working with the organization in reviewing an event. The policy has since been revised to extend the period for completion of Root Cause Analysis from 30 to 45 days.
JCAHO instituted additional changes, also effective April 1, 1998, to address concerns raised by health care organizations about the confidentiality of information provided to JCAHO in connection with Root Cause Analysis. These included:
1. Withholding of the names of patients and caregivers involved in the Sentinel Event from inclusion in Root Cause Analysis documents;
2. Root Cause Analysis materials (i.e., identification of where responsibility was felt to lie, obviously potentially damaging information to disclose in a malpractice case) would be separated from the organization’s Action Plan, describing the organization’s strategies for future risk reduction (potentially protected from admission into evidence at a trial);
3. Root Cause Analysis materials would be returned to the organization once data was reviewed by JCAHO for inclusion in its Sentinel Event database;
4. Any copies of Root Cause Analysis documents sent to JCAHO would be destroyed after being reviewed;
5. Satisfactory Action Plans would be returned to the organization;
6. Inclusion of language in future Terms of Agreement documents for accreditation that would formally acknowledge JCAHO as a participating entity in the health care organization’s quality monitoring and improvement activities, the purpose being to attempt to bring JCAHO’s activities as to the Sentinel Event and Root Cause Analysis under the protection of state peer-review confidentiality statutes; [How effective this strategy might be remains an open question and will certainly vary among the states.]
7. Establishment of a telephone hotline; and
8. Making available on-site assistance for an organization in preparing a Root Cause Analysis at a fee limited to direct costs.26
Root Cause Analysis
JCAHO’s prescribed method for determining the underlying cause or causes of a Sentinel Event is the Root Cause Analysis. A Root Cause Analysis "a) focuses primarily on systems and processes, not individual performance; b) progresses from special causes in clinical processes to common causes in organizational process; c) repeatedly digs deeper’ to underlying organization systems and process that can be altered to reduce the likelihood of human error in the future; and d) identifies changes that should be made in systems and process, either through redesign or development of new systems or process, that would reduce the likelihood of such events occurring in the future.27
Root Cause Analysis was not an original concept for JCAHO, or for anyone in the health care field, for that matter. Rather, the concept was borrowed from other industries, most notably engineering and aviation. 28,29 Causation analysis (JCAHO uses the term "Root Cause Analysis") has long been used in these, and other, industries as a tool to improve quality. Everyone is, of course, familiar with the use by pilots of mandatory checklists on airplanes, one of the primary safety measures developed in the aviation industry in response to causation analysis.
Root Cause Analysis involves a comprehensive investigation of the systems and processes in place at the time the Sentinel Event occurred, an evaluation of why they failed, and development of a plan of correction (the "Action Plan") involving new systems and processes designed to prevent similar adverse events in the future. JCAHO’s framework for an organization’s Root Cause Analysis and development of an Action Plan is outlined in Table 3.
| TABLE 3: Action Plan |
| 1. What happened (a detailed description of the Sentinel Event)? |
| 2. Why did it happen? What were the most proximate factors? |
| 3. The process or activity in which the event occurred. |
| A. Human factors. |
| B. Equipment factors. |
| C. Controllable environmental factors. |
| D. Uncontrollable external factors. |
| E. Other. |
| 4. Why did that (i.e., the proximate factor) happen? What systems and processes underlie those proximate factors? |
| A. Human resource issues. |
| B. Information management issues. |
| C. Environmental management issues. |
| F. Leadership issues. Corporate culture. |
| G. Encouragement of communication. |
| H. Clear communication of priorities. |
| I. Uncontrollable factors. |
| (Source: www.jcaho.org/sentinel/sentevnt_main.html) |
Organizations now have 45 days to complete their Root Cause Analysis, whether they self-report or JCAHO becomes aware of the Sentinel Event in some other way (patient complaint, media report, etc.). However, if an accredited organization decides not to share information regarding a reportable Sentinel Event with JCAHO, the organization will be placed on Accreditation Watch and ultimately risk loss of its accreditation.30
During the 45-day analysis period, if JCAHO receives an inquiry about the accreditation status of an organization that has experienced a reviewable Sentinel Event, the organization’s accreditation status will be reported in the usual manner without making reference to the Sentinel Event. If the inquirer specifically references the Sentinel Event, JCAHO will acknowledge that it is aware of the event and is working with the organization through the Sentinel Event review process.31 If, on the other hand, the organization refuses to cooperate with JCAHO regarding the Sentinel Event and is placed on Accreditation Watch, JCAHO considers this publicly disclosable information.
Self-Reporting Sentinel Events
If an organization is contemplating self-reporting a Sentinel Event, the first step that should be taken is obvious. Consult with experienced counsel before making the report or doing anything else. In conjunction with counsel, the following actions should be considered:
1. Commit sufficient resources to accomplish a credible Root Cause Analysis within the 45-day deadline;
2. Focus the Root Cause Analysis on systems and processes, not on individuals. JCAHO has no interest in critiques of individual practitioners and does not involve itself in review of the competence or conduct of individuals. Rather, JCAHO desires to focus exclusively on systems issues and the development of processes that will prevent, or catch and correct human errors;
3. Assign a lead person to communicate with JCAHO. JCAHO requires that this person be an officer of the organization. This individual should work under the auspices of counsel to 1) avoid legal error (e.g., disclosing privileged material which need not be disclosed), and 2) attempt to protect as much material as possible through attorney-client privilege;
4. All other individuals should be instructed not to talk with JCAHO. They should deal exclusively with the designated lead person and counsel;
5. Counsel should, in conjunction with JCAHO’s general counsel, draft and enter into a written agreement that incorporates JCAHO into the hospital’s internal peer review process in an effort to avoid disclosure of peer review material to JCAHO resulting in a waiver of peer review confidentiality under state law;
6. Redact patient and practitioner identity from all self-report, Root Cause Analysis or Action Plan materials;
7. Consultation with the hospital’s liability insurer, if any;
8. Whether the hospital self-reports or not, a Root Cause Analysis should be promptly completed;
9. If an on-site review or interview takes place, counsel should be present; and
10. If a decision is made not to cooperate with JCAHO regarding a Sentinel Event, counsel should communicate with JCAHO. Counsel will attempt to put the hospital in the best possible pre-litigation position should JCAHO place the hospital on Accreditation Watch and the hospital wants to consider taking legal action against JCAHO.32
Confidentiality
The degree of confidentiality with which Root Cause Analyses, and other communications with JCAHO related to Sentinel Events, will be kept confidential is of major concern to health care attorneys. Included in this concern, and perhaps most important, is whether information shared with JCAHO will lose its peer review privilege and become discoverable in a subsequent malpractice action, should one be filed. The law on this issue varies from state to state and is far from clear. In addition, the Sentinel Event Policy and Root Cause Analysis requirement are new enough that there is not a significant body of reported appellate cases specifically addressing the discoverability of these materials.
Modifications were made in the Sentinel Event Policy effective July 1, 1998, in an effort to address concerns about the possible waiver of confidentiality of materials included in an organization’s Root Cause Analysis.33 An option was made available for an organization which had reported a Sentinel Event to have an on-site review or interview, instead of sending Root Cause Analysis materials to JCAHO. As a result, no materials, that would otherwise have been protected under state law as peer review materials, would be at risk of losing peer review protection because of having been disclosed in writing outside the organization’s peer review committees. Unfortunately, just because the information is not given to JCAHO in writing does not eliminate the possibility that peer review protection might be waived. Accredited organizations are truly caught "between a rock and a hard place" in this regard. On the one hand, they can reveal statutorily protected confidential peer review information to JCAHO and face the possibility of losing evidentiary protection for the information. On the other hand, refusal to share the information with JCAHO might result in the organization being placed on Accreditation Watch and, worse yet, possibly losing its accreditation. Most organizations understandably choose the first option, reasoning that the potential risk of having to reveal peer review materials if a malpractice lawsuit is filed is a lesser evil than the more certain risk of being placed on Accreditation Watch, or losing accredited status, a devastating possibility.
The possible loss of confidentiality of Sentinel Event and Root Cause Analysis materials remains a major concern of health care attorneys. As discussed above, the primary concern is whether the sharing of otherwise confidential peer review materials with JCAHO, as a result of the Sentinel Event Policy, will result in a waiver of confidentiality, as otherwise provided in state peer review statutes. There is currently no federal protection of this material and passage of such legislation in the near future does not appear likely. Therefore, because peer review protection varies from state to state, the possible effect of sharing this confidential material with JCAHO also varies from state to state. A 50-state survey completed by the Credentialing and Peer Review Committee of the American Health Lawyers Association found considerable variation in the evidentiary protection for such material among the states.34 Most states do not directly address this issue by specifically including JCAHO in the list of committees or organizations that are entitled to peer review protection. An exception is Maine, where the peer review privilege statute specifically includes JCAHO.35
If an organization is notified by JCAHO that it has become aware of a Sentinel Event, or the organization believes a Sentinel Event has occurred, it should immediately consult with counsel that is knowledgeable regarding peer review protection law (as well as other potential protections for the process and materials, e.g., attorney-client privilege) and in dealing with JCAHO. Generally, the investigation and gathering of materials should be done under the auspices of an attorney, in an attempt to bring as much material as possible under the umbrella of attorney-client privilege. It is, of course, true that simply passing documents through an attorney does not itself make them privileged. However, it is worth a try. In addition, any existing attorney-client privilege will be waived if the material is shared with a third-party (e.g., JCAHO), unless that third-party is an agent of the attorney.
If the hospital is not self-insured, the attorney handling the situation for the hospital may want to contact the general counsel for the hospital’s liability insurer if there is any concern that the hospital’s decision to reveal confidential peer review material to JCAHO might jeopardize the hospital’s liability coverage. Revealing confidential information that is potentially damaging to the defense of a liability claim, and thereby waving peer review protection, will make it more difficult for the insurer to defend a subsequent liability claim. As a result, such conduct might be prohibited under the terms of the organization’s insurance policy and might result in the insurer issuing a "reservation of rights" letter to the hospital, indicating that, while it will defend the suit, the insurer reserves its right to contest indemnity coverage.
JCAHO is on record that it will notify the involved organization if it receives a discovery demand and that it is committed to "vigorously resisting, and to helping individual organizations resist, the disclosure of confidential information related to sentinel events."
Sentinel Event Analysis by JCAHO
In February 1998, after review of 200 Sentinel Events, JCAHO determined that the most common category of Sentinel Event was medication error and, of those, the most frequently implicated drug was potassium chloride.36 JCAHO reviewed 10 incidents of patient death due to the misadministration of potassium chloride, eight of these being the result of direct injection of concentrated potassium chloride. Rather than dismissing this as something that could never happen in your department, it is important to understand that this does in fact occur, and systems must be in place to prevent such avoidable disasters. The philosophy has to be that, we are only human, therefore we will make mistakes. As a result, since human error is inevitable, systems must be designed to catch those errors before they result in disaster.
In the case of the potassium chloride deaths, JCAHO determined through its analysis of the Root Cause Analysis data that a contributing factor was the availability of concentrated potassium chloride on the nursing units. In six of the eight cases, potassium was somehow mistaken for another medication, primarily due to similarities in packaging and labeling, most commonly heparin or furosemide. Based upon these data, JCAHO recommended that facilities not make concentrated potassium chloride available outside the pharmacy "unless appropriate specific safeguards are in place."37
On Aug. 28, 1998, JCAHO reported that, pursuant to its Sentinel Event Policy, it had reviewed 15 cases related to surgery at the wrong site (e.g., left leg instead or right).38 Evaluation of Root Cause Analyses revealed that wrong-site surgery was most common during orthopedic procedures (10), followed by urologic procedures (3), and neurosurgical procedures (2). Risk factors for wrong site surgery were identified as:
— more than one surgeon involved in the case, either because multiple procedures were contemplated or because the care of the patient was transferred to another surgeon;
— multiple procedures were conducted on the same patient during a single trip to the operating room, especially when the procedures were on different sides of the patient;
— unusual time pressures, related to an unusual start time or pressure to speed up the preoperative procedures; and
— unusual patient characteristics such as physical deformity or massive obesity that might alter the usual process for equipment set-up or positioning of the patient.
The proximate causes identified by the involved hospitals most often related to communication, preoperative assessment of the patient, and the procedures used to verify the operative site. The first of these risk factors is particularly applicable to emergency physicians (i.e., shift change). JCAHO made the following recommendations to reduce the risk of wrong-site surgery:
• Clearly mark the operative site and involve the patient in the marking process;
• Require an oral verification of the correct site in the operating room by each member of the surgical team;
• Develop a verification checklist that includes all documents referencing the intended operative procedure and site, including the medical record, x-rays and other imaging studies and their reports, the informed consent document, the operating room record, and the anesthesia record; and direct observation of the marked operative site on the patient;
• Personal involvement of the surgeon in obtaining informed consent; and
• Ongoing monitoring that verification procedures are followed for high-risk procedures.39
While these recommendations are specifically directed at operating room procedures by surgeons, lessons for the emergency department can be learned. While I have never put a chest tube in the wrong side of the chest, I always worried about the possibility, and I know it has happened. These operating room strategies may be useful in avoiding wrong side procedures in the emergency department, as well as in the operating room (e.g., chest tube placement, thoracentesis, etc.).
On Nov. 18, 1998, JCAHO announced the results of its review of restraint deaths, a category of Sentinel Events of obvious relevance to the emergency department.40 In 40% of the cases, the cause of death was asphyxiation. The remainder were caused by strangulation, cardiac arrest, or fire. Risk factors for restraint deaths were determined to be: restraining of patients who smoke; restraining of patients with deformities that preclude the proper application of the restraining device (especially vest restraints); restraining a patient in the supine position may predispose the patient to aspiration; restraining a patient in the prone position may predispose the patient to suffocation; and restraining a patient in a room that is not under continuous observation by staff.
Proximate causes of the deaths included: patient assessment, such as incomplete medical assessment or incomplete examination of the individual (e.g., failure to identify contraband, such as matches); inadequate care planning, such as alternatives not fully considered, restraints used as punishment, and inappropriate room or unit assignment; lack of patient observation procedures or practices; staff-related factors, such as insufficient orientation or training, competency review or credentialing, or insufficient staffing levels; and equipment-related factors, such as use of split side rails without side rail protectors, use of two-point rather than four-point restraints, use of a high-neck vest, incorrect application of a restraining device, or a monitor or an alarm not working or not being used when appropriate. 41 JCAHO’s recommended strategies to reduce the risk of restraint deaths are listed in Table 4.
| TABLE 4: Recommended Strategies |
| 1. Redouble efforts to reduce the use of physical restraint and therapeutic hold through the use of risk assessment and early intervention with less restrictive measures. |
| 2. Revise procedures for assessing the medical condition of psychiatric patients. |
| 3. Enhance staff orientation/education regarding alternatives to physical restraints and proper application of restraints or therapeutic holding. |
| 4. Consider age, sex, and gender of patients when setting therapeutic hold policies. |
| 5. Revise the staffing model. |
| 6. Develop structure procedures for consistent application of restraints. |
| 7. Continuously observe any patient that is restrained. |
| 8. If a patient must be restrained in the supine position, ensure that the head is free to rotate to the side and, when possible, the head of the bed is elevated to minimize the risk of aspiration. |
| 9. If a patient must be restrained in the prone position, ensure that the airway is unobstructed at all times (for example, do not cover or "bury" the patient's face). Also, ensure that expansion of the patient's lungs is not restricted by excessive pressure on the patient's back (special caution is required for children, elderly patients and very obese patients). |
| 10. Do not restrain a patient in a bed with unprotected split side rails. |
| 11. Discontinue use of certain types of restraints, such as high vests and waist restraints. |
| 12. Ensure that all smoking materials are removed from patient's access, including access from family and friends. |
| (Source: "Preventing Restraint Deaths," Sentinel Event Alert, Nov. 18, 1998.) |
JCAHO has also analyzed and made risk-reduction recommendations regarding blood transfusion errors,42 infant abductions,43 and inpatient suicides.44
As of July 22, 1999, JCAHO had reviewed a total of 562 Sentinel Events. Of these, 63.3% occurred in general hospitals, 55% were self-reported, and 78% involved the death of a patient.
Criticism of the Private Accreditation Process
There has been growing criticism by both private commentators and the government of the government’s reliance on private accreditation and the adequacy of the private accreditation process. This criticism includes concerns regarding both the adequacy of the accrediting organization’s standards and the adequacy of their enforcement (e.g., concerns about the propriety of announcing survey visits), as well as the level of independence and objectivity of the accrediting agencies. This latter concern includes questions about having a single organization serve both as advisor and enforcer, and the overall independence of accrediting organizations such as JCAHO. Finally, questions have been raised about the adequacy of enforcement mechanisms, whether they are used often enough, and the adequacy of the public’s access to survey and investigation information.
The Department of Health and Human Services, Office of Inspector General, and Office of Evaluation and Inspection ("OEI") published a comprehensive assessment of The Review of Hospital Quality in July 1999. Copies of this report, consisting of four parts (A Call for Greater Accountability, The Role of Accreditation, The Role of Medicare Certification, and Holding the Reviewers Accountable) may be obtained from the OEI’s website, www.dhhs.gov/progorg/oei. OEI’s abstract of its findings summarizes its findings.
On July 20, we released four final inspection reports assessing the key roles in hospital quality oversight played by the Joint Commission for Accreditation of Healthcare Organizations, the State survey and certification agencies, and HCFA. Overall, the reports conclude that while the system of oversight that HCFA relies upon has some strengths, it also has deficiencies that warrant serious attention. HCFA, we documented, does little at present to hold either the Joint Commission or the State survey agencies accountable for their performance.
The reports call for HCFA to exert leadership in addressing the shortcomings. First, as a guiding principle, we urged HCFA to steer the external review process so that it represented a balance between the educationally oriented approaches of the Joint Commission and the enforcement-oriented approaches of the State agencies. Then we offer two sets of recommendations. In the first, we presented a number of steps HCFA should take to hold both the Joint Commission and the States more fully accountable for their performance in reviewing hospitals. Second, we called for HCFA to determine the appropriate minimum cycle for conducting certification surveys of non-accredited hospitals. HCFA responded positively to our report by presenting a detailed hospital oversight plan that incorporates our recommendations and an accompanying strategy for hospital performance measurement. In-depth discussion of the findings and recommendations of this assessment are beyond the scope of this article.
Conclusion
JCAHO accreditation is all-important for most hospitals, and many other health care organizations. Part of the accreditation process is the proper handling of Sentinel Events. A hospital’s level of accreditation is important as a public relations tool, but, more importantly, a hospital’s participation in Medicare is often dependent on its JCAHO accreditation. Individual physicians may not think knowledge of JCAHO standards and procedures is important. But this is a mistake. While hospital administrative personnel, some of them physicians, will do the lion’s share of work in preparing for JCAHO surveys and in handling other communications with JCAHO (e.g., response to Sentinel Events), they cannot do this effectively without the cooperation of the medical staff.
Sentinel Events will occur in the emergency department; hopefully, only rarely in your emergency department. While it is not necessary for physicians to be able to decide whether a given adverse outcome constitutes a reviewable Sentinel Event, it is important for all physicians to know what adverse outcomes might constitute Sentinel Events and, when such an event occurs, to notify appropriate hospital personnel and cooperate in investigating its causes.
References
1. www.jcaho.org/who_23_are.html
2. 42 CFR § 1395bb.
3. "Accreditation of Health Care Organizations," Health Law Manual ("HLM"), 1998 (Aspen), 1.
4 Fla. Admin. Code §§ 59A-12.0071 & 59A-12.0072.
5. HLM, supra note 3, at 3
6. 33 Ill.2d 326, 211 N.E.2d 253 (Ill. 1965).
7. 211 N.E.2d 253, 255
8. Id.
9. Id.
10. 211 N.E.2d 253, 256
11. Id.
12. Id.
13. Id.
14. Id.
15. 211 N.E.2d 253, 257
16. Fay A. Rozovsky, JD, MPH, DFASHRM, et al, "The JCAHO Sentinel Event Policy. Concerns and Alternatives," Health Law Digest ("HLD"), 26(5), 4.
17. www.jcaho.org/sentinel/se_pp.html
18. "Sentinel Events. Evaluating Causes and Planning Improvement." Oakbrook Terrace, Illinois. JCAHO, 1998.
19. "Board of Commissioners Affirms Support for Sentinel Event Policy," Sentinel Event Alert, May 1, 1998.
20. Id.
21. "Accreditation Committee Approves Examples of Voluntarily Reportable Sentinel Events," Sentinel Event Alert, May 11, 1998.
22 Id.
23. "Conducting a Root Cause Analysis in Response to a Sentinel Event." Oakbrook Terrace, Illinois. JCAHO, 1996.
24. "The Sentinel Event Policy. A Work in Progress," Perspectives, November/December 1996, 1.
25. "Rationale Guiding the Evaluation of Sentinel Events." Joint Commission Perspectives, May/June 1996.
26. HLD supra note 16, at 5
27. HLD supra note 16, at 3
28. "Sentinel Events. Evaluating Cause and Planning Improvement," Perspectives, November/December 1996, 29.
29. Id. at 115.
30. "Joint Commission Statement Regarding the Sentinel Event Policy." Sentinel Event Alert, May 1, 1998.
31. www.jcaho.org/sentinel/se_pp.html
32. HLD supra note 16, at 9.
33. "Joint Commission Statement Regarding the Sentinel Event Policy," Sentinel Event Alert, May 1, 1998.
34. HLD supra note 16, at 6.
35. Me. Rev. Stat. Ann. Tit. 32 § 2599 and § 3296.
36. "Medication Error Prevention Potassium Chloride," Sentinel Event Alert, Feb. 27, 1998.
37. Id.
38. "Lessons Learned. Wrong Site Surgery," Sentinel Event Alert, Aug. 28, 1998.
39. Id.
40. "Preventing Restraint Deaths," Sentinel Event Alert, Nov. 18, 1998.
41. Id.
42. "Blood Transfusion Errors. Preventing Future Occurrences," Sentinel Event Alert, Aug. 30, 1999.
43. "Infant Abductions. Preventing Future Occurrences," Sentinel Event Alert, April 9, 1999.
44. "Inpatient Suicides. Recommendations for Prevention," Sentinel Event Alert, Nov. 6, 1998.
Physician CME Questions
21. JCAHO’s Sentinel Event Policy provides for:
a. root cause analysis.
b. mandatory self-reporting of sentinel events.
c. development of an action plan.
d. both A and C.
22 Using its Sentinel Event databank, JCAHO has analyzed the proximate causes of various types of adverse events and made recommendations designed to decrease their occurrence. These events include:
a. medication error.
b. wrong site surgery.
c. restraint deaths.
d. all of the above.
23. JCAHO Accreditation levels include:
a. Accreditation with Merit.
b. Accreditation with Commendation.
c. Accreditation without Deficiency.
d. all of the above.
24. JCAHO accreditation standards are:
a. never admissible in a negligence lawsuit.
b. in some states, admissible as evidence of the standard of care.
c. always determinative of the standard of care.
d. none of the above.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.