Risk Management
Preventing sexual attacks in healthcare facilities: Risk management considerations By John D. Banja, PhD Reports or allegations of sexual attacks in healthcare facilities are extremely upsetting and sometimes not given the attention they deserve. In June 2011, the United States Government Accountability Office (GAO) issued a remarkable report on sexual attacks occurring in Veterans Affairs (VA) health facilities that not only raised awareness of the magnitude of the problem but that detailed numerous system weaknesses in VA facilities that might have enabled such attacks. This article discusses some of the GAO’s findings as well as other instances of sexual attacks, such as occurred in the criminal prosecution of Paul Serdula, a former health professional who might have sexually assaulted hundreds of women. Some of Serdula’s victims have subsequently sued in civil court, charging Serdula’s employers with lack of supervision and raising the possibility of serial sexual attacks such as his evolving into large-scale patient safety disasters. This article will review certain ethical and legal considerations bearing on the liability of healthcare facilities in which sexual attacks are alleged to have occurred. Following a discussion of how two courts have used the legal construct of “foreseeability” in determining a healthcare facility’s liability when an employee is charged with sexual assault, the article will conclude with a host of patient safety recommendations aimed at discouraging or deterring the occurrence of sexual attacks. I N TR O D U C TI O N Because allegations of sexual attacks in hospitals and clinics are shocking, embarrassing, and sometimes made with less than persuasive evidence or credibility, they have sometimes been dismissed by leadership as exaggerated, imagined, or simply lies.1 Yet, in 2011, the United States Government Accountability Office (GAO) produced a report on “Actions Needed to Prevent Sexual Assaults and Other Safety Incidents” that surveyed safety incidents occurring from January 2007 through July 2010 at the nation’s Department of Veterans Affairs (VA) medical facilities.2 The report noted that
© 2014 American Society for Healthcare Risk Management of the American Hospital Association Published online in Wiley Online Library (wileyonlinelibrary.com) • DOI: 10.1002/jhrm.21130 AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
5
284 sexual assault incidents were reported to VA police from January 2007 through July 2010, not including reports that may never have been created, never sent, or were lost by local VA police or the VA’s Office of Security and Law Enforcement. The cases involved allegations of rape, inappropriate touching, “forceful” medical examinations, oral sex, and other types of nonconsensual sexual assaults. Not including assaults that involved outsiders, visitors, or unknowns, 89 of the assaults were patienton-patient, 46 were employee-on-patient, and 85 were patient-on-employee. The report subsequently triggered a hearing by the House Veterans Affairs Committee on the safety of VA patients and employees3 that only added to the national concern about the increasing rate of violence in our nation’s health facilities.
than 42% of all rapes and sexual assaults are reported to law enforcement, largely because of victim embarrassment, fears about causing familial distress or the victim’s being blamed as the provocateur, the victim’s refusing to press charges, or the victim’s being convinced by others that she agreed to the sexual contact.9 Additionally, the GAO report speculated that patient-on-patient attacks are probably vastly underreported, as health professionals sometimes attribute the attack to the patient attacker’s diminished psychological condition and, hence, dismiss his or her responsibility for the incident.2 Furthermore, one has only to do the most cursory Internet search of “sexual abuse in hospitals” to read about dozens of harrowing and sensationalistic reports of the same. Virtually all of them involve health professionals sexually attacking patients rather than the more common patient-on-patient attack. However, while there might be a temptation to dismiss these stories as sensationalistic or aberrations, patient safety personnel must be keenly sensitive to the fact that a lone sexual predator on staff can cause a risk management disaster.
Of course, health professionals are explicitly tasked with protecting the welfare of their patients, such that when patients (or visitors or staff ) provide evidence that they suffered a sexual attack in a healthcare facility, suspicions are aroused about the adequacy of protection they were afforded. Since Hippocrates, health professionals have been obliged to Consider the recent Paul Serdula do no harm, which includes taking case. Between 2005 and 2009, reasonable precautions to prevent Serdula was a certified registered unnecessary harm befalling those 4 nurse anesthetist who sexuentrusted to them. In accordance with Risk managers hardly need ally molested possibly hundreds the Hippocratic tradition, the Joint of women whom he sedated while Commission’s Hospital Accreditation reminding of the moral delivering anesthesia in and around Standards have reminded facilities over metropolitan Atlanta health facilithe years that patients have the right obligation to keep patients ties.10 When police raided his home “to be free from neglect; exploitation; in November 2009, they found safe from harm. and verbal, mental, physical, and sexual 5(p2) hundreds of videos of him sexuFurthermore, the Joint abuse.” ally assaulting anesthetized patients, Commission includes rape—which is some of whom were identifiable. defined as “unconsented sexual conThe videos were taken by Serdula tact”—as a sentinel event, meaning that himself, who, once he had rendered it signals the need for immediate inves6(pSE-5) these women unconscious, molested The National Quality Forum, tigation and response. them and photographed the acts with his cell phone. another well-known standards-setting organization, Serdula was eventually convicted in January 2011 of 34 includes sexual abuse/assault among its “Never Events,” criminal counts and sentenced to life in prison plus 25 implying not just that sexual attacks shouldn’t happen but 11 7 Subsequently, multiple civil suits have been filed years. that they must never happen. by some of Serdula’s victims, who are seeking damages for Risk managers hardly need reminding of the moral obliprofessional negligence and negligent supervision from the gation to keep patients safe from harm. Yet, if the GAO facilities at which they were assaulted. report on the frequency of sexual attacks is in any way Collectively, these data, findings, and anecdotal reports representative of what occurs in our nation’s hospitals not only suggest that instances of sexual attacks in health and clinics, preventing various forms of sexual violence facilities occur more often than we’d care to think, but to patients may be much harder than it seems. The Joint that a single individual like Paul Serdula can cause a largeCommission has recently confirmed the distressing frescale risk management event that can require years of quency of sexual attacks, noting that assault, rape, and litigation and enormous resources to settle. The objective homicide are consistently among the top 10 types of senof this article is not to offer risk management procedures tinel events reported to the Commission’s database and pertaining to the aftermath of sexual assault allegations, that violence in healthcare settings remains significantly 8 but rather to consider preventive strategies that might disunderreported. The GAO report also admits that their courage their occurrence. Because it is a central construct findings likely reflect underreporting, but underreporting in assessing the healthcare provider’s liability in preventing of sexual violence is precisely what one would expect: A sexual attack or in responding to allegations of them, the 2009 Bureau of Justice statistics bulletin noted that fewer 6
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
legal construct of “foreseeability” will be examined in what immediately follows. The article will then conclude with a series of risk management considerations that might help reduce the risk of these unspeakable acts.
“ FORE S EEABILITY” AN D SEXU A L ATTA C K S I N HE ALTH C ARE FAC ILITIES Sexual assault in healthcare facilities is hardly an unknown complaint in civil court as plaintiffs alleging sexual attacks have advanced various causes of action against healthcare entities over the years. Space does not allow an analysis of these causes of action (eg, premises liability, negligent supervision, vicarious liability), nor is it the objective of this article to advise on legal theories for managing these kinds of claims. Rather, and especially in the hope that patient safety personnel will find it more relevant, the following section will consider some of the ways that courts have used the legal notion of “foreseeability” in their rulings on the liability of healthcare facilities with regard to protecting personnel and patients from sexual attacks or abuse. Foreseeability is a key—indeed, possibly the most fundamental—construct in negligence law because it explicitly speaks to the professional’s duty to protect clients from the onset of any unreasonable or unjustifiable harm.12 Foreseeability denotes the health professional’s awareness of whatever risks are reasonably known to be present in a course of treatment or therapy (including the environment in which that treatment occurs) and serves to trigger the professional’s obligation to take reasonable measures to protect patients given the (foreseeable) harm potential of those (foreseeable) risks.13 (See Exhibit 1 for various ethical and legal characterizations of foreseeability.)
As persons trained in law well know, foreseeability is theoretically linked with “causation” because Y is a foreseeable event if I know or should know that Y will likely result from doing X. Events are unforeseeable if the chain of causation resulting in Y is beyond the comprehension of a reasonably reflective or cautious person.12,13 Thus, when it comes to assessing legal liability for an adverse event, defendants will often plead that the event in question was unforeseeable because its occurrence was entirely outside the realm of normal expectation, or the event’s causational factors were unpredictable, or the adversity was simply beyond the defendant’s control. Alternatively, when harm is reasonably foreseeable but the actor did nothing to prevent its onset, one would intuitively be inclined to hold him or her liable, at least to some degree, for whatever harm materialized. In instances that contemplate the liability of healthcare facilities given the risk of sexual attacks, courts have focused the notion of foreseeability in two different directions: one understands foreseeability in terms of the event’s occurrence likelihood, while the other understands foreseeability as the capacity of a reasonably cautious person to recognize that a contemplated action or practice may prove unduly harmful. Consider the following case examples, both from the state of Washington. In Kaltreider v. Lake Chelan Community Hospital, Elizabeth Kaltreider was receiving inpatient treatment at Lake Chelan Community Hospital (LCCH) for alcohol dependency.14 While there, she began a flirtatious relationship that voluntarily progressed to a sexual one with an LCCH nurse, George Menard. Upon her discharge, Kaltreider wanted to continue the relationship but Menard apparently did not. Kaltreider proceeded to file suit against the hospital, claiming that, among other
Exhibit 1: Ethical and Legal Characterizations of “Foreseeability”
“[F]oreseeability may be morally grounded in our shared acceptance of a behavioral norm of a reflective, cautious person.”12(p12) “[T]he capacity of a reasonable prudent person to contemplate that acting in a certain way may produce the type of hazard that actually does result.”12(p12) “[F]oreseeability includes risks that an actor may not know but reasonably should, commonly explained in constructive knowledge terms as risks the actor ‘should have known,’ meaning that prudence sometimes requires actors to investigate and evaluate possibilities of hidden or inchoate risk.”12(p17) An occurrence is foreseeable “if it is the kind of occurrence that a reasonable person ought to have anticipated under the circumstances.”13(pp1498–1499) “Foreseeability is an attempt to predict at what point the ‘causal’ relationship between [one person’s ] conduct and [another person’s] injury is too attenuated, remote, or freakish to justify imposing liability.”13(p1499) Foreseeability is “an aspect of the scope of a person’s duty of reasonable care.”13(p1500)
DOI: 10.1002/jhrm
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
7
things, she was owed a duty of protection from Menard’s sexual misconduct. The court’s majority eventually upheld a lower court’s ruling that dismissed Kaltreider’s claim on summary judgment, holding that Menard’s conduct was not reasonably foreseeable. That is, because LCCH had no reason to suspect that Menard would sexually approach Kaltreider—ie, his conduct was not “foreseeable” given his previous work history—and also because Kaltreider apparently consented to the relationship, the court ruled that LCCH had no duty to “protect” her.
the opinion of Niece’s expert that such unsupervised contact is unwise, and Legislative recognition of the problem of sexual abuse in residential care facilities, all demonstrate that sexual abuse by staff at a group home for developmentally disabled persons may be a foreseeable hazard against which reasonable precautions must be taken.15(p427)
What distinguishes Niece from Kaltreider is how each court understood foreseeability in terms of the plaintiff ’s degree of vulnerability. Because Lori Niece was profoundly unable to protect herself from sexual attack, the Interestingly, the Kaltreider decision made numerous referSupreme Court of Washington concluded that Elmview ences to another Washington case decided a dozen years Group Home had a heightened responsibility to proearlier: Dory Niece, as Guardian Petitioner, v. Elmview tect her “from a universe of possible harms.” Elizabeth Group Home, Respondent.15 In that decision, however, the Kaltreider, however, not only voluntarily consented to Washington Supreme Court affirmed a lower court’s rulher relationship with George Menard but appeared able ing that the sexual attack of a caregiver at the Elmview to discourage, fight off, call for help, or otherwise resist Group Home “was not legally unforeseeable”15(p422) and any unwanted advances Menard made toward her. In fact, that “the special relationship between the group home and a dissenting justice in Kaltreider believed that Elizabeth its vulnerable residents gives rise to a duty of reasonable indeed had a “recognized vulnerability to sexual miscare.”15(p422) conduct of which the hospital was aware or should have been aware,”14(para7) irrespective of the The facts in Niece were straightforhospital’s knowledge about George ward: Lori Niece was born with cerMenard’s sexual intentions or previebral palsy that involved severe mobilous work history. This dissenting ity, communicational, and cognitive judge understood Elizabeth Kaltreider impairments. She resided at Elmview to be psychologically vulnerable to Group Home, which was a licensed sexual advances and, on that basis, Sexual attacks are often private provider of residential care for thought that her sexual contacts with persons with developmental disabilities. Menard were foreseeable. enabled or facilitated by While there, she was sexually attacked by an Elmview employee, Kleber Although it is impossible to know various system weaknesses. Quevodo. Quevodo had no criminal how any court will rule on foreseehistory and had received positive refability issues involving sexual attack, erences from his previous employer the legal and ethical intuition that such that Elmview had no reason to connects “foreseeability” to the anticipate or foresee his sexual assault vulnerability of a patient to sexual on Niece. At the time of his assault, attack seems eminently reasonable. Quevodo was the only Elmview staff Determining the foreseeability of sexmember on duty, although Elmview previously had a poliual attack purely on the basis of whether a facility had any cy forbidding male staff from being left alone with female reason to believe that an employee might sexually abuse a residents—a policy that originated because of prior sexual patient, or whether it was “likely” or “probable” that such assaults on residents by another Elmview employee. By an attack might occur, seems a “lazy” rather than “pruthe time of Quevodo’s assault on Niece, however, Elmview dent” person’s account of foreseeability. The Niece ruling had abandoned that policy. At bench trial, Niece’s expert suggests that healthcare facilities might have to take addiopined that unsupervised contact with residents like tional steps to protect obviously vulnerable patients from Quevodo’s violated the standard of care for group homes assault or abuse. Prevention strategies should take this with disabled residents. The Washington Supreme Court observation to heart because if a facility routinely leaves eventually held that: males alone with females who are profoundly unable to defend themselves or call for help, it seems only a matter Given Niece’s total inability to take care of herself, Elmview of time before a sexual attack will occur. was responsible for every aspect of her well being. This responsibility gives rise to a duty to protect Niece and other similarly vulnerable residents from a universe of possible R I S K MA N A GE ME N T S TR ATEG I ES harms. … Quevodo’s assault was not legally unforeseeable as It will come as no surprise to risk managers that a promilong as the possibility of sexual assaults on residents by staff nent variable influencing the likelihood of sexual attacks was within the general field of danger which should have is the condition of the physical and human factors envibeen anticipated. … The prior sexual assaults at Elmview, ronment in which attacks occurred. Sexual attacks are the earlier policy against unsupervised contact with residents, 8
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
often enabled or facilitated by various system weaknesses that fail to discourage or halt their instigation and trajectories. Four variables that figured prominently in the GAO’s investigations of sexual abuse were: (1) information systems that omitted data on individuals’ work or legal histories that might alert staff to the possibility of a sexual attack; (2) policies or decisions that put patients or personnel at heightened risk of attack; (3) the quality of surveillance and safety monitoring systems; and (4) staff training issues. Recommendations that patient safety personnel might implement to deter sexual attacks, especially as gleaned from the GAO study, are as follows: 1. Healthcare facilities should vigilantly examine patient and employee records and documentation for information that might alert staff to an individual’s propensity for inappropriate sexual encounters and address any form of sexually inappropriate behavior as soon as it appears. Typically, healthcare facilities do not use tools that specifically assess patients’ potential for sexual assault but instead rely on assessments and observations by clinical staff. Nevertheless, clinical personnel who know or should know that a patient has a history of violence must be exquisitely attentive to that possibility. Given increasing rates of violence in our nation’s health facilities, the GAO report emphasized that personnel should scrutinize biopsychosocial forms, medical records, and nursing or suicide risk assessments for such indications, and have formal processes in place that alert treatment and security staff accordingly. The GAO study recommended flagging such patients’ electronic medical records so that all staff are aware of those persons’ potential for violence and that increased levels of observation might be required. As will be further discussed, such reports need to be consistently documented and made uniformly and reliably accessible across the system. A less formal but possibly even more important consideration is for a health organization’s leadership to cultivate environments where staff feel comfortable in speaking up about odd, eccentric, or frankly dangerous behaviors their peers exhibit. In matters pertaining to sexual concerns, these behaviors might include inappropriate (or “creepy”) conversation, touching, staring, excessive clinical attention involving visitation, and so forth. Importantly, the literature on “speaking up” reports that staff often are reluctant to speak up about their peers’ behaviors because they fear retaliation from doing so, and they can easily rationalize their silence with “nothing bad has happened so far” or “it’s not my job.”16 Yet the interests of patient safety demand that healthcare providers maintain the safety of their environments, which can include protecting and even rewarding staff who justifiably speak up, and committing organizational resources to training employees to speak up and to manage emotionally difficult conversations. 2. Facilities should be extremely wary of policies or practices that allow male employees or visitors to be left alone with other patients, especially with incapacitated females. Studies DOI: 10.1002/jhrm
suggest that males perpetrate at least 90% of sexual attacks.17 About 75% to 90% of these are male-on-female, although one study of sexual allegations in a psychiatric facility reported that 21% of them were male-on-male.18 On the one hand, then, prudence would dictate that any practice that presumably or allegedly leaves helpless female patients—such as the ones Paul Serdula victimized—alone with men should be reevaluated and revised. On the other hand, if there is any lesson learned from the account of foreseeability above, it is that the more vulnerable a particular patient is to sexual attack regardless of his or her gender, the more a health organization may be expected to protect that individual. On this account and because sedated or unconscious patients of either gender are frequently left alone with health professionals, their care providers must maintain a keen vigilance toward their safety. Thus, health professionals should not assume that all “visitors” in patients’ rooms are legitimate but should determine their identities and visitation rights. Also, patients who may require increased observation for the threat of violence can be physically located nearer rather than farther from the nurse’s station or be within the nursing staff ’s line of sight. Especially because health professionals will inevitably leave vulnerable patients alone, facilities must take reasonable steps to discourage any kind of violence toward them as much as possible, which can include installing and maintaining appropriate monitoring or surveillance technology. 3. Healthcare facilities should upgrade and maintain their monitoring and surveillance technology and insist that security and treatment staff are attentive to them. The GAO study performed on-site safety evaluations at 5 Veterans Health Administration (VHA) facilities and found that among some of them: • Police and command personnel did not monitor camera feeds continuously; at some facilities the police office was unattended part of the time, usually because officers had other duties such as serving as telephone operators or emergency dispatchers for the entire hospital. • At 4 of the 5 facilities, stationary and computer-based panic alarm systems in the inpatient mental health units alerted only police and security personnel at the control center, but not staff at the units where the alarms originated; at the other facility, the stationary alarms notified only the unit nursing station staff and not the police, who had to be notified separately. • Video feed at one hospital unit was incompatible with systems installed in other parts of the medical facility and could not be monitored by control center security personnel. • At some of the facilities, police or control center personnel could not always pinpoint the location of an alarm that was activated in testing.
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
9
• At one facility, a test call from an emergency call box went to the hospital’s central telephone operator rather than to the police command-and-control center; another test call was erroneously identified as coming from an elevator rather than from the correct location. • At one facility, almost 20% of the stationary panic alarms in the inpatient mental health units were inoperable. • At another facility, the recent relocation of the facility’s computers caused errors in pinpointing the location of the alarm being tested. • One site failed to properly document tests conducted of their alarm systems, which is a required element of the VHA’s “Environment of Care Checklist.” • Distributing whistles to personnel to signal assaults might not be effective: One nurse was grabbed by the throat and couldn’t summon assistance; staff asserted an interest in having portable panic alarms. • VHA police and command-and-control centers were seriously understaffed at the majority of the facilities surveyed. • At one of the sites, when the alarms were activated in a test, neither staff in the unit nor police knew how to turn them off. • Architectural design can be critical in thwarting attacks: At one VHA facility, a patient reached over the counter
of the unit’s inpatient mental health nursing station and stabbed a nurse in her neck, shoulder, and leg with a pen; shoulder-height plexiglass barricades were included in the redesign of the unit. These findings are listed as risk management recommendations in Exhibit 2. Additionally, safety equipment such as metal detectors, fences, locks, windows, and doors should be in good working order in view of the ways sexual predators can exploit them. Organizations that are recognized for excellent patient safety practices are known to be relentless in anticipating how things might go wrong and exercise continuous vigilance and response to the potential for adversity, especially involving malfunctioning equipment or substandard operating practices.19 4. Staff training should be an integral part of any organizational effort aimed at reducing the risk of sexual attacks. Consider the following recommendations, which are inspired by the GAO report: First, healthcare facilities should have a definition of sexual assault that is understood and applied throughout the organization. Remarkably, an organization as bureaucratic and seemingly integrated as the VA did not have such a definition at the time of the GAO report, such that VA facilities made up their own definitions or simply had none (see Exhibit 3). The problem with individualized definitions is not only their inconsistent meaning and (mis)application across the system but their tendency to result in highly subjective interpretations of
Exhibit 2: Monitoring and Surveillance Recommendations2
• • • • • • • • • • • •
10
Designated personnel must monitor camera feeds continuously or as reasonably required. Alarms must be checked to ensure that they alert relevant staff, which usually includes both clinical staff and security personnel (or police). Video feeds from each of the clinical units in a facility must be system-compatible such that each one can be monitored by security personnel at the central control desk. Security personnel must be able to locate accurately the geographical location of a security alarm that is activated. Calls from emergency call boxes must go to an appropriate party, which will usually be a police or command and control center. All panic alarms must be operational and personnel must be trained in their use. Computer assisted location of activated alarms must be checked for accuracy. Documentation on the maintenance or upkeep of alarms must be monitored for compliance. Institutions should consider equipping staff with portable panic alarms rather than other devices that may require a physical operation difficult to perform in the event of a physical attack. Staffing of security personnel must be adequate. Relevant staff must be knowledgeable in the operation of alarms. Architectural design, especially including the construction of nursing units and environments where staff might be stationary, should be considered from the standpoint of affording reasonable safety for treating personnel.
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
Exhibit 3: Varying Definitions of Sexual Assault at VHA Facilities2(p19)
Facility A: “Sexual violation of a person (male or female) by use of force, threat, or intimidation [that] is committed without the consent of the person assaulted. The violent act may or may not include penetration and may be [an] oral, anal, or vaginal violation.” Facility B: No definition Facility C: “Conduct of a sexual or indecent nature toward another person that is accompanied by actual or threatened physical force or that induces fear, shame, or mental suffering.” Facility D: “Includes incest, oral copulation, penetration, rape, sexual assault, sexual battery, and sodomy which occurs without the consent of a person, or when a person is not capable of giving consent. Sexual abuse also means acts of a sexual nature committed in the presence of a vulnerable adult without that person’s informed consent. It includes, but is not limited to, the acts defined in a state statute, fondling, exposure of a vulnerable adult’s sexual organs, or the use of a vulnerable adult to solicit for or engage in prostitution or sexual performance.” Facility E: “Sexual assault is sexual conduct of ANY kind against a person’s will, brought about by force, threats, or coercion.” Note: The Joint Commission addresses sexual attack through its sentinel event policies. There, sexual attacks are understood as “rape” which is defined as “unconsented sexual contact involving a patient and another patient, staff member, or other perpetrator while being treated or on the premises of the hospital, including oral, vaginal or anal penetration or fondling of the patient’s sex organ(s) by another individual’s hand, sex organ, or object.” 6(pSE-5)
sexual misconduct. For example, and with reference to the definitions in Exhibit 3, when does a friendly hug become “fondling” or a stare become an amorous or inappropriate glance? Second, a healthcare facility should have a consensually adopted list of what incidents are to be reported to risk management, leadership, and law enforcement. Note that if a facility does not have a clear and robust definition of sexual assault, then reporting such incidents may vary enormously. The GAO report repeatedly found instances of sexual attacks being reported to one level of leadership, such as the nursing unit, but not another, such as the police. Very importantly, staff should be trained in recognizing the perpetration of sexual behavior that could count as a felony since felonies must be reported to law enforcement. Such reporting is mandated regardless of whether staff believe the behavior is related to the patient’s condition. Facilities should maintain friendly and cooperative arrangements with local police, who can advise on the relevant reporting laws in that jurisdiction. Third, healthcare organizations, especially large, integrated healthcare systems like the VHA’s, should have information-sharing practices, systemwide monitoring, and tracking mechanisms such that all network facilities can be aware of and notified about individuals who might perpetrate a sexual attack. Fourth, facilities should use investigations of sexual attacks to train staff and remediate system weaknesses. For at least a decade, risk management personnel have been DOI: 10.1002/jhrm
advised to approach individuals who have been harmed by adverse events with a description of system changes that have been instituted to lessen the future risk of such events. Obviously, a facility that has witnessed previous instances of sexual attack but made no organizational, environmental, or policy changes in their aftermath may well draw a court’s ire in the wake of another attack. Fifth, facilities should maintain a centralized database on sexual attacks. Just as the absence of a systemwide definition of sexual assault compromises efforts to reduce the frequency of its occurrence, so does the absence of a centralized database compromise efforts to review or analyze reports of sexual assaults, to track and trend patient safety incidents, and to launch appropriate education and training efforts. Fragmented informational systems often mean that a system continues to respond to reports of sexual attacks on a case-by-case basis and fails to take advantage of any historical lessons learned or to record and study whether subsequent system changes are working.
C O N C LU S I O N Allegations and substantiated reports of sexual attacks in our nation’s healthcare facilities (or anywhere else) are acutely distressing for all concerned. But given their psychologically upsetting nature, reports of such attacks might cause professionals to whom the reports are made to minimize or dismiss them—often because the allegation itself constitutes a situation that people just want to “go away.”20 If the GAO report is
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
11
in any way representative of the state of our nation’s capacity to protect personnel and patients from sexual attack, that capacity might be considerably less than acceptable. The objective of this article has been to briefly characterize health professionals’ obligations in preventing sexual attacks in healthcare facilities and to consider some risk management strategies that might discourage such occurrences or at least diminish their harm. Because healthcare systems never run optimally and because it is often impossible to know when sexual predators are in our midst, risk managers must impress personnel with the need for continued vigilance; the adoption and maintenance of adequate technology that might deter sexual attacks; and the knowledge required to maintain a safe environment for patients and personnel.
RE FE REN C ES 1. Aronfeld SM. Sexual assault by a health care provider. Trial. December 1, 2009;45(12):46–50. 2. United States Government Accountability Office. VA health care: Actions needed to prevent sexual assaults and other safety incidents. Washington, DC: Government Accountability Office; 2011. GAO11-530. http://www.gao.gov/assets/320/319342.pdf. Accessed September 29, 2013. 3. Maze R. VA vows to address sexual assaults in hospitals. Marine Corps Times. June 8, 2011. http:// www.marinecorpstimes.com/article/20110608 /NEWS/106080320/VA-vows-address-sexual-assaults -hospitals. Accessed September 29, 2013. 4. Beauchamp TL, Childress JF. Principles of Biomedical Ethics. 6th ed. Oxford, UK: Oxford University Press; 2009:149–155. 5. Joint Commission. Preventing violence in the health care setting. Issue 45, June 3, 2010. http://www .jointcommission.org/assets/1/18/sea_45.pdf. Accessed September 29, 2013. 6. Joint Commission. Ethics, rights, and responsibilities. 2008 Hospital Accreditation Standards. Chicago, IL: Joint Commission; 2008. 7. Joint Commission. Sentinel events. CAMH Refreshed Care. Chicago, IL: Joint Commission; 2011. 8. National Quality Forum. Serious reportable events in healthcare. 2011 Update: A Consensus Report. Washington, DC, 2011. http://www.prweb.com /releases/2011/6/prweb8568166.htm. Accessed September 29, 2013. 9. Rand MR. National crime victimization survey: criminal victimization. Bureau of Justice Statistics Bulletin. 12
September 2009, NCJ 227777. http://www.bjs.gov /content/pub/pdf/cv08.pdf. Accessed September 29, 2013. 10. Atlanta nurse charged with molesting, sodomizing patients under anesthesia. FoxNews.com. November 30, 2009. http://www.foxnews.com/story/2009 /11/30/atlanta-nurse-charged-with-molesting -sodomizing-patients-under-anesthesia/. Accessed September 29, 2013. 11. Seward C. Dental practice faces suits over videotaped sexual assaults. Atlanta Journal-Constitution. May 12, 2012. http://www.ajc.com/news/news/local /dental-practice-faces-suits-over-videotaped-sexual /nQTmW/. Accessed September 29, 2013. 12. Owen DG. Figuring foreseeability. In: Third Restatement: A symposium on the third restatement of torts; April 3, 2009; Wake Forest University School of Law. 3-31 Draft. http://tortssymposium.law.wfu .edu/papers/owen.pdf. Accessed September 4, 2013. 13. VerSteeg R. Perspectives on foreseeability in the law of contracts and torts: the relationship between “intervening causes” and “impossibility.” Mich. St. L. Rev. 2011:1497–1519. 14. Kaltreider v. Lake Chelan Community Hospital (249 P.3d 182). 15. Dory Niece, as Guardian Petitioner, v. Elmview Group Home, Respondent (929 P2d 420). 16. Maxfield D, Grenny J, Patterson K, McMillan R, Switzler A. Silence kills: the seven crucial conversations for healthcare. VitalSmarts, L.C.; 2005. http://www .aacn.org/WD/Practice/Docs/PublicPolicy /SilenceKills.pdf. Accessed September 29, 2013. 17. Wikipedia. Rape statistics. http://en.wikipedia.org /wiki/Rape_statistics. Accessed September 29, 2013. 18. Berland D, Guskin K. Patient allegations of sexual abuse against psychiatric hospital staff. Gen Hosp Psych. 1994;16(5):335–339. 19. Reason J. The Human Contribution. Burlington, VT: Ashgate; 2008:276–280. 20. Banja J. Medical Errors and Medical Narcissism. Sudbury, MA: Jones and Bartlett; 2005.
A BO U T TH E A U TH O R John D. Banja, PhD, is a professor in the department of rehabilitation medicine and a medical ethicist at Emory University in Atlanta, Georgia. He is an ASHRM member and has participated often as a faculty member at ASHRM training conferences. He consults frequently on medical malpractice cases and can be reached at [email protected].
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
Preventing sexual attacks in healthcare facilities: Risk management considerations By John D. Banja, PhD Reports or allegations of sexual attacks in healthcare facilities are extremely upsetting and sometimes not given the attention they deserve. In June 2011, the United States Government Accountability Office (GAO) issued a remarkable report on sexual attacks occurring in Veterans Affairs (VA) health facilities that not only raised awareness of the magnitude of the problem but that detailed numerous system weaknesses in VA facilities that might have enabled such attacks. This article discusses some of the GAO’s findings as well as other instances of sexual attacks, such as occurred in the criminal prosecution of Paul Serdula, a former health professional who might have sexually assaulted hundreds of women. Some of Serdula’s victims have subsequently sued in civil court, charging Serdula’s employers with lack of supervision and raising the possibility of serial sexual attacks such as his evolving into large-scale patient safety disasters. This article will review certain ethical and legal considerations bearing on the liability of healthcare facilities in which sexual attacks are alleged to have occurred. Following a discussion of how two courts have used the legal construct of “foreseeability” in determining a healthcare facility’s liability when an employee is charged with sexual assault, the article will conclude with a host of patient safety recommendations aimed at discouraging or deterring the occurrence of sexual attacks. I N TR O D U C TI O N Because allegations of sexual attacks in hospitals and clinics are shocking, embarrassing, and sometimes made with less than persuasive evidence or credibility, they have sometimes been dismissed by leadership as exaggerated, imagined, or simply lies.1 Yet, in 2011, the United States Government Accountability Office (GAO) produced a report on “Actions Needed to Prevent Sexual Assaults and Other Safety Incidents” that surveyed safety incidents occurring from January 2007 through July 2010 at the nation’s Department of Veterans Affairs (VA) medical facilities.2 The report noted that
© 2014 American Society for Healthcare Risk Management of the American Hospital Association Published online in Wiley Online Library (wileyonlinelibrary.com) • DOI: 10.1002/jhrm.21130 AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
5
284 sexual assault incidents were reported to VA police from January 2007 through July 2010, not including reports that may never have been created, never sent, or were lost by local VA police or the VA’s Office of Security and Law Enforcement. The cases involved allegations of rape, inappropriate touching, “forceful” medical examinations, oral sex, and other types of nonconsensual sexual assaults. Not including assaults that involved outsiders, visitors, or unknowns, 89 of the assaults were patienton-patient, 46 were employee-on-patient, and 85 were patient-on-employee. The report subsequently triggered a hearing by the House Veterans Affairs Committee on the safety of VA patients and employees3 that only added to the national concern about the increasing rate of violence in our nation’s health facilities.
than 42% of all rapes and sexual assaults are reported to law enforcement, largely because of victim embarrassment, fears about causing familial distress or the victim’s being blamed as the provocateur, the victim’s refusing to press charges, or the victim’s being convinced by others that she agreed to the sexual contact.9 Additionally, the GAO report speculated that patient-on-patient attacks are probably vastly underreported, as health professionals sometimes attribute the attack to the patient attacker’s diminished psychological condition and, hence, dismiss his or her responsibility for the incident.2 Furthermore, one has only to do the most cursory Internet search of “sexual abuse in hospitals” to read about dozens of harrowing and sensationalistic reports of the same. Virtually all of them involve health professionals sexually attacking patients rather than the more common patient-on-patient attack. However, while there might be a temptation to dismiss these stories as sensationalistic or aberrations, patient safety personnel must be keenly sensitive to the fact that a lone sexual predator on staff can cause a risk management disaster.
Of course, health professionals are explicitly tasked with protecting the welfare of their patients, such that when patients (or visitors or staff ) provide evidence that they suffered a sexual attack in a healthcare facility, suspicions are aroused about the adequacy of protection they were afforded. Since Hippocrates, health professionals have been obliged to Consider the recent Paul Serdula do no harm, which includes taking case. Between 2005 and 2009, reasonable precautions to prevent Serdula was a certified registered unnecessary harm befalling those 4 nurse anesthetist who sexuentrusted to them. In accordance with Risk managers hardly need ally molested possibly hundreds the Hippocratic tradition, the Joint of women whom he sedated while Commission’s Hospital Accreditation reminding of the moral delivering anesthesia in and around Standards have reminded facilities over metropolitan Atlanta health facilithe years that patients have the right obligation to keep patients ties.10 When police raided his home “to be free from neglect; exploitation; in November 2009, they found safe from harm. and verbal, mental, physical, and sexual 5(p2) hundreds of videos of him sexuFurthermore, the Joint abuse.” ally assaulting anesthetized patients, Commission includes rape—which is some of whom were identifiable. defined as “unconsented sexual conThe videos were taken by Serdula tact”—as a sentinel event, meaning that himself, who, once he had rendered it signals the need for immediate inves6(pSE-5) these women unconscious, molested The National Quality Forum, tigation and response. them and photographed the acts with his cell phone. another well-known standards-setting organization, Serdula was eventually convicted in January 2011 of 34 includes sexual abuse/assault among its “Never Events,” criminal counts and sentenced to life in prison plus 25 implying not just that sexual attacks shouldn’t happen but 11 7 Subsequently, multiple civil suits have been filed years. that they must never happen. by some of Serdula’s victims, who are seeking damages for Risk managers hardly need reminding of the moral obliprofessional negligence and negligent supervision from the gation to keep patients safe from harm. Yet, if the GAO facilities at which they were assaulted. report on the frequency of sexual attacks is in any way Collectively, these data, findings, and anecdotal reports representative of what occurs in our nation’s hospitals not only suggest that instances of sexual attacks in health and clinics, preventing various forms of sexual violence facilities occur more often than we’d care to think, but to patients may be much harder than it seems. The Joint that a single individual like Paul Serdula can cause a largeCommission has recently confirmed the distressing frescale risk management event that can require years of quency of sexual attacks, noting that assault, rape, and litigation and enormous resources to settle. The objective homicide are consistently among the top 10 types of senof this article is not to offer risk management procedures tinel events reported to the Commission’s database and pertaining to the aftermath of sexual assault allegations, that violence in healthcare settings remains significantly 8 but rather to consider preventive strategies that might disunderreported. The GAO report also admits that their courage their occurrence. Because it is a central construct findings likely reflect underreporting, but underreporting in assessing the healthcare provider’s liability in preventing of sexual violence is precisely what one would expect: A sexual attack or in responding to allegations of them, the 2009 Bureau of Justice statistics bulletin noted that fewer 6
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
legal construct of “foreseeability” will be examined in what immediately follows. The article will then conclude with a series of risk management considerations that might help reduce the risk of these unspeakable acts.
“ FORE S EEABILITY” AN D SEXU A L ATTA C K S I N HE ALTH C ARE FAC ILITIES Sexual assault in healthcare facilities is hardly an unknown complaint in civil court as plaintiffs alleging sexual attacks have advanced various causes of action against healthcare entities over the years. Space does not allow an analysis of these causes of action (eg, premises liability, negligent supervision, vicarious liability), nor is it the objective of this article to advise on legal theories for managing these kinds of claims. Rather, and especially in the hope that patient safety personnel will find it more relevant, the following section will consider some of the ways that courts have used the legal notion of “foreseeability” in their rulings on the liability of healthcare facilities with regard to protecting personnel and patients from sexual attacks or abuse. Foreseeability is a key—indeed, possibly the most fundamental—construct in negligence law because it explicitly speaks to the professional’s duty to protect clients from the onset of any unreasonable or unjustifiable harm.12 Foreseeability denotes the health professional’s awareness of whatever risks are reasonably known to be present in a course of treatment or therapy (including the environment in which that treatment occurs) and serves to trigger the professional’s obligation to take reasonable measures to protect patients given the (foreseeable) harm potential of those (foreseeable) risks.13 (See Exhibit 1 for various ethical and legal characterizations of foreseeability.)
As persons trained in law well know, foreseeability is theoretically linked with “causation” because Y is a foreseeable event if I know or should know that Y will likely result from doing X. Events are unforeseeable if the chain of causation resulting in Y is beyond the comprehension of a reasonably reflective or cautious person.12,13 Thus, when it comes to assessing legal liability for an adverse event, defendants will often plead that the event in question was unforeseeable because its occurrence was entirely outside the realm of normal expectation, or the event’s causational factors were unpredictable, or the adversity was simply beyond the defendant’s control. Alternatively, when harm is reasonably foreseeable but the actor did nothing to prevent its onset, one would intuitively be inclined to hold him or her liable, at least to some degree, for whatever harm materialized. In instances that contemplate the liability of healthcare facilities given the risk of sexual attacks, courts have focused the notion of foreseeability in two different directions: one understands foreseeability in terms of the event’s occurrence likelihood, while the other understands foreseeability as the capacity of a reasonably cautious person to recognize that a contemplated action or practice may prove unduly harmful. Consider the following case examples, both from the state of Washington. In Kaltreider v. Lake Chelan Community Hospital, Elizabeth Kaltreider was receiving inpatient treatment at Lake Chelan Community Hospital (LCCH) for alcohol dependency.14 While there, she began a flirtatious relationship that voluntarily progressed to a sexual one with an LCCH nurse, George Menard. Upon her discharge, Kaltreider wanted to continue the relationship but Menard apparently did not. Kaltreider proceeded to file suit against the hospital, claiming that, among other
Exhibit 1: Ethical and Legal Characterizations of “Foreseeability”
“[F]oreseeability may be morally grounded in our shared acceptance of a behavioral norm of a reflective, cautious person.”12(p12) “[T]he capacity of a reasonable prudent person to contemplate that acting in a certain way may produce the type of hazard that actually does result.”12(p12) “[F]oreseeability includes risks that an actor may not know but reasonably should, commonly explained in constructive knowledge terms as risks the actor ‘should have known,’ meaning that prudence sometimes requires actors to investigate and evaluate possibilities of hidden or inchoate risk.”12(p17) An occurrence is foreseeable “if it is the kind of occurrence that a reasonable person ought to have anticipated under the circumstances.”13(pp1498–1499) “Foreseeability is an attempt to predict at what point the ‘causal’ relationship between [one person’s ] conduct and [another person’s] injury is too attenuated, remote, or freakish to justify imposing liability.”13(p1499) Foreseeability is “an aspect of the scope of a person’s duty of reasonable care.”13(p1500)
DOI: 10.1002/jhrm
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
7
things, she was owed a duty of protection from Menard’s sexual misconduct. The court’s majority eventually upheld a lower court’s ruling that dismissed Kaltreider’s claim on summary judgment, holding that Menard’s conduct was not reasonably foreseeable. That is, because LCCH had no reason to suspect that Menard would sexually approach Kaltreider—ie, his conduct was not “foreseeable” given his previous work history—and also because Kaltreider apparently consented to the relationship, the court ruled that LCCH had no duty to “protect” her.
the opinion of Niece’s expert that such unsupervised contact is unwise, and Legislative recognition of the problem of sexual abuse in residential care facilities, all demonstrate that sexual abuse by staff at a group home for developmentally disabled persons may be a foreseeable hazard against which reasonable precautions must be taken.15(p427)
What distinguishes Niece from Kaltreider is how each court understood foreseeability in terms of the plaintiff ’s degree of vulnerability. Because Lori Niece was profoundly unable to protect herself from sexual attack, the Interestingly, the Kaltreider decision made numerous referSupreme Court of Washington concluded that Elmview ences to another Washington case decided a dozen years Group Home had a heightened responsibility to proearlier: Dory Niece, as Guardian Petitioner, v. Elmview tect her “from a universe of possible harms.” Elizabeth Group Home, Respondent.15 In that decision, however, the Kaltreider, however, not only voluntarily consented to Washington Supreme Court affirmed a lower court’s rulher relationship with George Menard but appeared able ing that the sexual attack of a caregiver at the Elmview to discourage, fight off, call for help, or otherwise resist Group Home “was not legally unforeseeable”15(p422) and any unwanted advances Menard made toward her. In fact, that “the special relationship between the group home and a dissenting justice in Kaltreider believed that Elizabeth its vulnerable residents gives rise to a duty of reasonable indeed had a “recognized vulnerability to sexual miscare.”15(p422) conduct of which the hospital was aware or should have been aware,”14(para7) irrespective of the The facts in Niece were straightforhospital’s knowledge about George ward: Lori Niece was born with cerMenard’s sexual intentions or previebral palsy that involved severe mobilous work history. This dissenting ity, communicational, and cognitive judge understood Elizabeth Kaltreider impairments. She resided at Elmview to be psychologically vulnerable to Group Home, which was a licensed sexual advances and, on that basis, Sexual attacks are often private provider of residential care for thought that her sexual contacts with persons with developmental disabilities. Menard were foreseeable. enabled or facilitated by While there, she was sexually attacked by an Elmview employee, Kleber Although it is impossible to know various system weaknesses. Quevodo. Quevodo had no criminal how any court will rule on foreseehistory and had received positive refability issues involving sexual attack, erences from his previous employer the legal and ethical intuition that such that Elmview had no reason to connects “foreseeability” to the anticipate or foresee his sexual assault vulnerability of a patient to sexual on Niece. At the time of his assault, attack seems eminently reasonable. Quevodo was the only Elmview staff Determining the foreseeability of sexmember on duty, although Elmview previously had a poliual attack purely on the basis of whether a facility had any cy forbidding male staff from being left alone with female reason to believe that an employee might sexually abuse a residents—a policy that originated because of prior sexual patient, or whether it was “likely” or “probable” that such assaults on residents by another Elmview employee. By an attack might occur, seems a “lazy” rather than “pruthe time of Quevodo’s assault on Niece, however, Elmview dent” person’s account of foreseeability. The Niece ruling had abandoned that policy. At bench trial, Niece’s expert suggests that healthcare facilities might have to take addiopined that unsupervised contact with residents like tional steps to protect obviously vulnerable patients from Quevodo’s violated the standard of care for group homes assault or abuse. Prevention strategies should take this with disabled residents. The Washington Supreme Court observation to heart because if a facility routinely leaves eventually held that: males alone with females who are profoundly unable to defend themselves or call for help, it seems only a matter Given Niece’s total inability to take care of herself, Elmview of time before a sexual attack will occur. was responsible for every aspect of her well being. This responsibility gives rise to a duty to protect Niece and other similarly vulnerable residents from a universe of possible R I S K MA N A GE ME N T S TR ATEG I ES harms. … Quevodo’s assault was not legally unforeseeable as It will come as no surprise to risk managers that a promilong as the possibility of sexual assaults on residents by staff nent variable influencing the likelihood of sexual attacks was within the general field of danger which should have is the condition of the physical and human factors envibeen anticipated. … The prior sexual assaults at Elmview, ronment in which attacks occurred. Sexual attacks are the earlier policy against unsupervised contact with residents, 8
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
often enabled or facilitated by various system weaknesses that fail to discourage or halt their instigation and trajectories. Four variables that figured prominently in the GAO’s investigations of sexual abuse were: (1) information systems that omitted data on individuals’ work or legal histories that might alert staff to the possibility of a sexual attack; (2) policies or decisions that put patients or personnel at heightened risk of attack; (3) the quality of surveillance and safety monitoring systems; and (4) staff training issues. Recommendations that patient safety personnel might implement to deter sexual attacks, especially as gleaned from the GAO study, are as follows: 1. Healthcare facilities should vigilantly examine patient and employee records and documentation for information that might alert staff to an individual’s propensity for inappropriate sexual encounters and address any form of sexually inappropriate behavior as soon as it appears. Typically, healthcare facilities do not use tools that specifically assess patients’ potential for sexual assault but instead rely on assessments and observations by clinical staff. Nevertheless, clinical personnel who know or should know that a patient has a history of violence must be exquisitely attentive to that possibility. Given increasing rates of violence in our nation’s health facilities, the GAO report emphasized that personnel should scrutinize biopsychosocial forms, medical records, and nursing or suicide risk assessments for such indications, and have formal processes in place that alert treatment and security staff accordingly. The GAO study recommended flagging such patients’ electronic medical records so that all staff are aware of those persons’ potential for violence and that increased levels of observation might be required. As will be further discussed, such reports need to be consistently documented and made uniformly and reliably accessible across the system. A less formal but possibly even more important consideration is for a health organization’s leadership to cultivate environments where staff feel comfortable in speaking up about odd, eccentric, or frankly dangerous behaviors their peers exhibit. In matters pertaining to sexual concerns, these behaviors might include inappropriate (or “creepy”) conversation, touching, staring, excessive clinical attention involving visitation, and so forth. Importantly, the literature on “speaking up” reports that staff often are reluctant to speak up about their peers’ behaviors because they fear retaliation from doing so, and they can easily rationalize their silence with “nothing bad has happened so far” or “it’s not my job.”16 Yet the interests of patient safety demand that healthcare providers maintain the safety of their environments, which can include protecting and even rewarding staff who justifiably speak up, and committing organizational resources to training employees to speak up and to manage emotionally difficult conversations. 2. Facilities should be extremely wary of policies or practices that allow male employees or visitors to be left alone with other patients, especially with incapacitated females. Studies DOI: 10.1002/jhrm
suggest that males perpetrate at least 90% of sexual attacks.17 About 75% to 90% of these are male-on-female, although one study of sexual allegations in a psychiatric facility reported that 21% of them were male-on-male.18 On the one hand, then, prudence would dictate that any practice that presumably or allegedly leaves helpless female patients—such as the ones Paul Serdula victimized—alone with men should be reevaluated and revised. On the other hand, if there is any lesson learned from the account of foreseeability above, it is that the more vulnerable a particular patient is to sexual attack regardless of his or her gender, the more a health organization may be expected to protect that individual. On this account and because sedated or unconscious patients of either gender are frequently left alone with health professionals, their care providers must maintain a keen vigilance toward their safety. Thus, health professionals should not assume that all “visitors” in patients’ rooms are legitimate but should determine their identities and visitation rights. Also, patients who may require increased observation for the threat of violence can be physically located nearer rather than farther from the nurse’s station or be within the nursing staff ’s line of sight. Especially because health professionals will inevitably leave vulnerable patients alone, facilities must take reasonable steps to discourage any kind of violence toward them as much as possible, which can include installing and maintaining appropriate monitoring or surveillance technology. 3. Healthcare facilities should upgrade and maintain their monitoring and surveillance technology and insist that security and treatment staff are attentive to them. The GAO study performed on-site safety evaluations at 5 Veterans Health Administration (VHA) facilities and found that among some of them: • Police and command personnel did not monitor camera feeds continuously; at some facilities the police office was unattended part of the time, usually because officers had other duties such as serving as telephone operators or emergency dispatchers for the entire hospital. • At 4 of the 5 facilities, stationary and computer-based panic alarm systems in the inpatient mental health units alerted only police and security personnel at the control center, but not staff at the units where the alarms originated; at the other facility, the stationary alarms notified only the unit nursing station staff and not the police, who had to be notified separately. • Video feed at one hospital unit was incompatible with systems installed in other parts of the medical facility and could not be monitored by control center security personnel. • At some of the facilities, police or control center personnel could not always pinpoint the location of an alarm that was activated in testing.
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
9
• At one facility, a test call from an emergency call box went to the hospital’s central telephone operator rather than to the police command-and-control center; another test call was erroneously identified as coming from an elevator rather than from the correct location. • At one facility, almost 20% of the stationary panic alarms in the inpatient mental health units were inoperable. • At another facility, the recent relocation of the facility’s computers caused errors in pinpointing the location of the alarm being tested. • One site failed to properly document tests conducted of their alarm systems, which is a required element of the VHA’s “Environment of Care Checklist.” • Distributing whistles to personnel to signal assaults might not be effective: One nurse was grabbed by the throat and couldn’t summon assistance; staff asserted an interest in having portable panic alarms. • VHA police and command-and-control centers were seriously understaffed at the majority of the facilities surveyed. • At one of the sites, when the alarms were activated in a test, neither staff in the unit nor police knew how to turn them off. • Architectural design can be critical in thwarting attacks: At one VHA facility, a patient reached over the counter
of the unit’s inpatient mental health nursing station and stabbed a nurse in her neck, shoulder, and leg with a pen; shoulder-height plexiglass barricades were included in the redesign of the unit. These findings are listed as risk management recommendations in Exhibit 2. Additionally, safety equipment such as metal detectors, fences, locks, windows, and doors should be in good working order in view of the ways sexual predators can exploit them. Organizations that are recognized for excellent patient safety practices are known to be relentless in anticipating how things might go wrong and exercise continuous vigilance and response to the potential for adversity, especially involving malfunctioning equipment or substandard operating practices.19 4. Staff training should be an integral part of any organizational effort aimed at reducing the risk of sexual attacks. Consider the following recommendations, which are inspired by the GAO report: First, healthcare facilities should have a definition of sexual assault that is understood and applied throughout the organization. Remarkably, an organization as bureaucratic and seemingly integrated as the VA did not have such a definition at the time of the GAO report, such that VA facilities made up their own definitions or simply had none (see Exhibit 3). The problem with individualized definitions is not only their inconsistent meaning and (mis)application across the system but their tendency to result in highly subjective interpretations of
Exhibit 2: Monitoring and Surveillance Recommendations2
• • • • • • • • • • • •
10
Designated personnel must monitor camera feeds continuously or as reasonably required. Alarms must be checked to ensure that they alert relevant staff, which usually includes both clinical staff and security personnel (or police). Video feeds from each of the clinical units in a facility must be system-compatible such that each one can be monitored by security personnel at the central control desk. Security personnel must be able to locate accurately the geographical location of a security alarm that is activated. Calls from emergency call boxes must go to an appropriate party, which will usually be a police or command and control center. All panic alarms must be operational and personnel must be trained in their use. Computer assisted location of activated alarms must be checked for accuracy. Documentation on the maintenance or upkeep of alarms must be monitored for compliance. Institutions should consider equipping staff with portable panic alarms rather than other devices that may require a physical operation difficult to perform in the event of a physical attack. Staffing of security personnel must be adequate. Relevant staff must be knowledgeable in the operation of alarms. Architectural design, especially including the construction of nursing units and environments where staff might be stationary, should be considered from the standpoint of affording reasonable safety for treating personnel.
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
Exhibit 3: Varying Definitions of Sexual Assault at VHA Facilities2(p19)
Facility A: “Sexual violation of a person (male or female) by use of force, threat, or intimidation [that] is committed without the consent of the person assaulted. The violent act may or may not include penetration and may be [an] oral, anal, or vaginal violation.” Facility B: No definition Facility C: “Conduct of a sexual or indecent nature toward another person that is accompanied by actual or threatened physical force or that induces fear, shame, or mental suffering.” Facility D: “Includes incest, oral copulation, penetration, rape, sexual assault, sexual battery, and sodomy which occurs without the consent of a person, or when a person is not capable of giving consent. Sexual abuse also means acts of a sexual nature committed in the presence of a vulnerable adult without that person’s informed consent. It includes, but is not limited to, the acts defined in a state statute, fondling, exposure of a vulnerable adult’s sexual organs, or the use of a vulnerable adult to solicit for or engage in prostitution or sexual performance.” Facility E: “Sexual assault is sexual conduct of ANY kind against a person’s will, brought about by force, threats, or coercion.” Note: The Joint Commission addresses sexual attack through its sentinel event policies. There, sexual attacks are understood as “rape” which is defined as “unconsented sexual contact involving a patient and another patient, staff member, or other perpetrator while being treated or on the premises of the hospital, including oral, vaginal or anal penetration or fondling of the patient’s sex organ(s) by another individual’s hand, sex organ, or object.” 6(pSE-5)
sexual misconduct. For example, and with reference to the definitions in Exhibit 3, when does a friendly hug become “fondling” or a stare become an amorous or inappropriate glance? Second, a healthcare facility should have a consensually adopted list of what incidents are to be reported to risk management, leadership, and law enforcement. Note that if a facility does not have a clear and robust definition of sexual assault, then reporting such incidents may vary enormously. The GAO report repeatedly found instances of sexual attacks being reported to one level of leadership, such as the nursing unit, but not another, such as the police. Very importantly, staff should be trained in recognizing the perpetration of sexual behavior that could count as a felony since felonies must be reported to law enforcement. Such reporting is mandated regardless of whether staff believe the behavior is related to the patient’s condition. Facilities should maintain friendly and cooperative arrangements with local police, who can advise on the relevant reporting laws in that jurisdiction. Third, healthcare organizations, especially large, integrated healthcare systems like the VHA’s, should have information-sharing practices, systemwide monitoring, and tracking mechanisms such that all network facilities can be aware of and notified about individuals who might perpetrate a sexual attack. Fourth, facilities should use investigations of sexual attacks to train staff and remediate system weaknesses. For at least a decade, risk management personnel have been DOI: 10.1002/jhrm
advised to approach individuals who have been harmed by adverse events with a description of system changes that have been instituted to lessen the future risk of such events. Obviously, a facility that has witnessed previous instances of sexual attack but made no organizational, environmental, or policy changes in their aftermath may well draw a court’s ire in the wake of another attack. Fifth, facilities should maintain a centralized database on sexual attacks. Just as the absence of a systemwide definition of sexual assault compromises efforts to reduce the frequency of its occurrence, so does the absence of a centralized database compromise efforts to review or analyze reports of sexual assaults, to track and trend patient safety incidents, and to launch appropriate education and training efforts. Fragmented informational systems often mean that a system continues to respond to reports of sexual attacks on a case-by-case basis and fails to take advantage of any historical lessons learned or to record and study whether subsequent system changes are working.
C O N C LU S I O N Allegations and substantiated reports of sexual attacks in our nation’s healthcare facilities (or anywhere else) are acutely distressing for all concerned. But given their psychologically upsetting nature, reports of such attacks might cause professionals to whom the reports are made to minimize or dismiss them—often because the allegation itself constitutes a situation that people just want to “go away.”20 If the GAO report is
AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
11
in any way representative of the state of our nation’s capacity to protect personnel and patients from sexual attack, that capacity might be considerably less than acceptable. The objective of this article has been to briefly characterize health professionals’ obligations in preventing sexual attacks in healthcare facilities and to consider some risk management strategies that might discourage such occurrences or at least diminish their harm. Because healthcare systems never run optimally and because it is often impossible to know when sexual predators are in our midst, risk managers must impress personnel with the need for continued vigilance; the adoption and maintenance of adequate technology that might deter sexual attacks; and the knowledge required to maintain a safe environment for patients and personnel.
RE FE REN C ES 1. Aronfeld SM. Sexual assault by a health care provider. Trial. December 1, 2009;45(12):46–50. 2. United States Government Accountability Office. VA health care: Actions needed to prevent sexual assaults and other safety incidents. Washington, DC: Government Accountability Office; 2011. GAO11-530. http://www.gao.gov/assets/320/319342.pdf. Accessed September 29, 2013. 3. Maze R. VA vows to address sexual assaults in hospitals. Marine Corps Times. June 8, 2011. http:// www.marinecorpstimes.com/article/20110608 /NEWS/106080320/VA-vows-address-sexual-assaults -hospitals. Accessed September 29, 2013. 4. Beauchamp TL, Childress JF. Principles of Biomedical Ethics. 6th ed. Oxford, UK: Oxford University Press; 2009:149–155. 5. Joint Commission. Preventing violence in the health care setting. Issue 45, June 3, 2010. http://www .jointcommission.org/assets/1/18/sea_45.pdf. Accessed September 29, 2013. 6. Joint Commission. Ethics, rights, and responsibilities. 2008 Hospital Accreditation Standards. Chicago, IL: Joint Commission; 2008. 7. Joint Commission. Sentinel events. CAMH Refreshed Care. Chicago, IL: Joint Commission; 2011. 8. National Quality Forum. Serious reportable events in healthcare. 2011 Update: A Consensus Report. Washington, DC, 2011. http://www.prweb.com /releases/2011/6/prweb8568166.htm. Accessed September 29, 2013. 9. Rand MR. National crime victimization survey: criminal victimization. Bureau of Justice Statistics Bulletin. 12
September 2009, NCJ 227777. http://www.bjs.gov /content/pub/pdf/cv08.pdf. Accessed September 29, 2013. 10. Atlanta nurse charged with molesting, sodomizing patients under anesthesia. FoxNews.com. November 30, 2009. http://www.foxnews.com/story/2009 /11/30/atlanta-nurse-charged-with-molesting -sodomizing-patients-under-anesthesia/. Accessed September 29, 2013. 11. Seward C. Dental practice faces suits over videotaped sexual assaults. Atlanta Journal-Constitution. May 12, 2012. http://www.ajc.com/news/news/local /dental-practice-faces-suits-over-videotaped-sexual /nQTmW/. Accessed September 29, 2013. 12. Owen DG. Figuring foreseeability. In: Third Restatement: A symposium on the third restatement of torts; April 3, 2009; Wake Forest University School of Law. 3-31 Draft. http://tortssymposium.law.wfu .edu/papers/owen.pdf. Accessed September 4, 2013. 13. VerSteeg R. Perspectives on foreseeability in the law of contracts and torts: the relationship between “intervening causes” and “impossibility.” Mich. St. L. Rev. 2011:1497–1519. 14. Kaltreider v. Lake Chelan Community Hospital (249 P.3d 182). 15. Dory Niece, as Guardian Petitioner, v. Elmview Group Home, Respondent (929 P2d 420). 16. Maxfield D, Grenny J, Patterson K, McMillan R, Switzler A. Silence kills: the seven crucial conversations for healthcare. VitalSmarts, L.C.; 2005. http://www .aacn.org/WD/Practice/Docs/PublicPolicy /SilenceKills.pdf. Accessed September 29, 2013. 17. Wikipedia. Rape statistics. http://en.wikipedia.org /wiki/Rape_statistics. Accessed September 29, 2013. 18. Berland D, Guskin K. Patient allegations of sexual abuse against psychiatric hospital staff. Gen Hosp Psych. 1994;16(5):335–339. 19. Reason J. The Human Contribution. Burlington, VT: Ashgate; 2008:276–280. 20. Banja J. Medical Errors and Medical Narcissism. Sudbury, MA: Jones and Bartlett; 2005.
A BO U T TH E A U TH O R John D. Banja, PhD, is a professor in the department of rehabilitation medicine and a medical ethicist at Emory University in Atlanta, Georgia. He is an ASHRM member and has participated often as a faculty member at ASHRM training conferences. He consults frequently on medical malpractice cases and can be reached at [email protected].
JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 33, NUMBER 3
DOI: 10.1002/jhrm
