Department of Veterans Affairs

Office of Inspector General

Healthcare Inspection
Management of Disruptive Patient
Behavior at VA Medical Facilities

Report No. 11-02585-129 VA Office of Inspector General

Washington, DC 20420

March 7, 2013

To Report Suspected Wrongdoing in VA Programs and Operations: Telephone: 1-800-488-8244 E-Mail: [email protected] Web site: www.va.gov/oig

Management of Disruptive Patient Behavior at VA Medical Facilities

Table of Contents

Page

Executive Summary ..............................................................................................i

Introduction .......................................................................................................... 1
Purpose............................................................................................................................. 1
Background...................................................................................................................... 1
Scope and Methodology .................................................................................................. 5
Results and Conclusions .................................................................................... 6
Overview of Disruptive Behavior Committees ............................................................... 6
Issue 1. Description of Disruptive Behavior Management at Selected Facilities .......... 6
Issue 2. Data Collection and Analysis of Incidents ...................................................... 15
Issue 3. Assignment of Patient Record Flags ............................................................... 17
Appendixes A. Under Secretary for Health Comments.................................................................... 20
B. OIG Contact and Staff Acknowledgments............................................................... 24
C. Report Distribution................................................................................................... 25

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

Executive Summary

The VA Office of Inspector General Office of Healthcare Inspections conducted a review to assess how VA medical facilities manage patients who display disruptive and violent behaviors. The objectives of the review were to determine what criteria facilities use to identify and address disruptive patient behavior; whether facilities have structured approaches and procedures to follow up on disruptive behavior; and how well facilities track, trend, and analyze disruptive behavior. According to the U.S. Bureau of Labor Statistics, health care workers are nearly 5 times more likely than workers in other industries to be victims of workplace violence, and many acts of violence are perpetrated by patients. Employees at VHA facilities are also at increased risk, and balancing the rights and health care needs of violent and disruptive patients with the health and safety of other patients, visitors, and staff is a significant challenge for VHA facilities. We interviewed officials from VHAs Behavioral Threat and Management Program and National Center for Ethics in Health Care and VA Police Service. We also reviewed the policies and procedures of Disruptive Behavior Committees (DBCs) at 30 randomly selected VHA facilities. We interviewed DBC representatives and reviewed the electronic health records of all 889 unique patients referred to DBCs at the 30 facilities in fiscal year (FY) 2011. We found significant differences in how VHA facilities define disruptive behavior, document incidents and interventions, and employ interventions to prevent and/or minimize risk of further incidents. VHA does not have a comprehensive definition of what constitutes disruptive behavior. As a result, facilities have applied their own definitions, which are generally not consistent from facility to facility. The lack of a common definition and terminology diminishes VHAs capacity to collect and analyze national data on disruptive behavior and identify system problems and training needs that could prevent or minimize disruptive behavior. VHA also has not established guidelines for what information facilities should document regarding disruptive incidents and how or where this information should be documented, and it is unclear what should be documented in a patients electronic health record so that providers have information needed for safe, quality health care. Lastly, facilities used a broad range of interventions to address disruptive behavior, all of which appeared to comply with the Federal regulation. DBC officials at 12 of the 30 sampled facilities reported that they were not tracking or trending data on disruptive incidents, and officials at 9 facilities reported that they collect only limited data, such as the overall number of referrals. Although VHA policy requires facilities to collect and analyze data, the policy does not provide guidance on what specific information facilities need to collect or methods for collecting data, such as a

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

national reporting system or template formats. Furthermore, according to DBC officials at several facilities, they do not have sufficient administrative resources to meet this policy requirement. We also found significant delays in assigning Category I Patient Record Flags (PRFs), which are intended to alert VHA employees to patient behavior that may pose an immediate threat to other patients, facility employees, and visitors. In FY 2011, the median number of days it took the 30 sampled facilities to assign Category I PRFs was 30, with a range of 0 to 469 days. About 15 percent of the PRFs were not assigned until 100 days after the triggering incidents. Although VHA policy on PRFs does not include a timeliness standard for assigning PRFs, sound clinical and business practice suggests that because Category I PRFs are intended to alert employees of immediate threats or risks, they should be assigned as soon after the triggering events as feasible. We recommended that the Under Secretary for Health ensure that VHA program officials provide guidance on what constitutes disruptive behavior and establish common terminology for VHA facilities, develop guidelines for what information facilities should document about disruptive incidents and where this information should be documented, and provide guidance to VHA facilities on collecting and analyzing data on disruptive incidents. We also recommended that the Under Secretary for Health consider implementing a national reporting system or data collection template for disruptive patient incidents and ensure that VHA facilities implement procedures to improve the timeliness of assigning Category I PRFs to alert VHA employees to patients who may pose an immediate threat. The Under Secretary for Health concurred with the findings and recommendations. See Appendix A (pages 2023) for the full text of his comments. We will follow up on the corrective actions until all recommendations have been fully implemented.

JOHN D. DAIGH, JR., M.D.

Assistant Inspector General for
Healthcare Inspections

VA Office of Inspector General

ii

Management of Disruptive Patient Behavior at VA Medical Facilities

Introduction

Purpose
The VA Office of Inspector General (OIG) Office of Healthcare Inspections performed a review to assess how VA medical facilities are managing patients who display disruptive and violent behaviors. The objectives of the review were to determine: What criteria facilities use to identify and address disruptive patient behavior Whether facilities have structured approaches and procedures to follow up on disruptive patient behavior How well facilities track, trend, and analyze disruptive patient behavior

Background
In recent years, high profile violent incidents have prompted health care facilities, including VA facilities, to revisit and revise their policies and procedures for responding to and managing violent patients. Unfortunately, violence in health care settings is not a new phenomenon. According to the U.S. Bureau of Labor Statistics (BLS), during the 5-year period 20032007, injuries related to assaults and violent acts, and resulting in lost workdays, increased every year for workers in the health care and social assistance industry. BLS reported, A worker in health care and social assistance is nearly 5 times more likely to be the victim of a nonfatal assault or violent act by person than the average worker in all industries combined.1 While most media coverage, statistical studies, and professional literature focus on the more egregious and injurious acts of violence in health care settings, such as physical and sexual assault and homicide, other acts of aggression and disruptive behavior are also prevalent. In a 2007 survey of over 3,400 emergency department nurses, over 50 percent of the respondents reported being spit on, hit, pushed/shoved, and/or scratched. Over 70 percent reported being yelled and/or cursed at, intimidated, and harassed with sexual language or innuendo.2 Recommendations and guidelines for handling violent and disruptive patients have been published by an array of organizations including the Occupational Safety and Health Administration, the Centers for Disease Control and Prevention (CDC), and the World Health Organization. The guidelines focus primarily on identifying risk factors and implementing prevention and mitigation strategies. According to the National Institute
1

Janocha, Jill A. and Ryan T. Smith - U.S. Bureau of Labor Statistics, Workplace Safety and Health in the Health Care and Social Assistance Industry, 20032007, Compensation and Working Conditions Online, accessed at www.bls.gov/opub/cwc/print/sh20100825ar01p1.htm on 8/23/12. 2 Gacki-Smith, Jessica, Altair M. Juarez, and Lara Boyett, Violence Against Nurses Working in US Emergency Departments, The Journal of Nursing Administration, 39:7/8, July/August 2009, pp. 340349.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

for Occupational Safety and Health (NIOSH), a research agency of the CDC, there are many risk factors for patient violence and disruptive behavior, such as (but not limited to): Volatile people, especially if they are under the influence of drugs or alcohol or have a history of violence or certain psychotic diagnoses Under staffing Long waits for service Overcrowded, uncomfortable waiting rooms Poor environmental design Inadequate security Lack of staff training and policies for preventing and managing crises Balancing the rights and health care needs of violent and disruptive patients with the health and safety of other patients, visitors, and staff is a significant challenge to health care organizations including VA. In some private sector settings, medical providers may fire patients who exhibit violent, disruptive, and rude behavior.3 Under Federal regulation, however, this is not an option for VA medical facilities. Although, VA facilities may limit the time, place, and/or manner of providing services to violent and disruptive patients, they must continue to offer the full range of medical services to which a Veteran is eligible.4 Veterans Health Administration (VHA) Policy. The primary policy source on violent and disruptive patient behaviors is contained in VHAs policy on patient record flags (PRFs). PRFs are a clinical tool facilities may use to mark or flag a patients electronic health record (EHR) to alert clinical staff that the patients behavior or medical status may pose an immediate threat to either the patients safety or the safety of other patients or staff. VHA uses two types of PRFs. Category I PRFs are used to identify patients who are high-risk for violent or disruptive behavior and are shared across all known VHA treating facilities. Category II PRFs are used to identify patients who are at risk for other reasons, such as drug seeking behavior, history of wandering, or spinal cord injuries. Category II PRFs are not shared across facilities; they are local flags.5 VHAs PRF policy focuses mainly on the placement of alert flags and only briefly addresses program requirements for managing violent and disruptive patients. The policy requires facilities to establish a Disruptive Behavior Committee (DBC), which is intended

3 4

Harris, Steven M., Take care when firing a patient, American Medical News, February 4, 2008.
U.S. Code of Federal Regulations (CFR), Title 38: Pensions, Bonuses, and Veterans Relief, Part 17 Medical,
Section 17.107 VA responses to disruptive behavior of patients.
5 VHA Directive 2010-053, Patient Record Flags, December 3, 2010.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

to focus on reducing the risk of patient violence toward employees and others at the facility, by: Coordinating with clinicians and recommending amendments to patient treatment plans to reduce patient risk of violence Making recommendations about and following up on the placement of PRFs Collecting and analyzing incidents of patient disruptive, threatening, or violent behavior Assessing the risk of violence in individual patients Identifying system problems, as well as training needs, related to the prevention and management of disruptive behavior Recommending to facility leadership other actions related to the problem of patient violence According to a VHA program official, the DBC is intended for clinical decision making in response to disruptive incidents. The DBC must be chaired by a senior clinician, such as psychiatrist or psychologist, who has knowledge of and experience in assessing violence risk. Other members should include representatives from VA Police, Health Information Management Service or Privacy Office, Patient Safety or Risk Management, Patient Advocacy, Union Safety Committee, and facility areas or departments that are at high risk for violence (as appropriate). In addition to the PRF policy, in September 2012, VHA issued a new directive intended to establish a unified policy describing the management of all individuals in VHA facilities whose behavior has, or could, jeopardize the health or safety of others, undermine a culture of safety in VHA, or otherwise interfere with the delivery of health care at the facility.6 This policy broadly addresses incidents and behaviors committed by not only patients, but also patient family members, employees, contractors, and visitors. It also includes specific focus on sexual assaults in VA facilities. VHA Program Responsibility and Oversight. Program management and guidance related to violent and disruptive patients comes from VHAs Office of Public Health and Environmental Hazards Behavioral Threat Management Program. Although the Director of the Behavioral Threat Management Program does not have direct oversight authority over the facility-based programs, the Director is responsible for developing policy within VHA. The Director also provides technical assistance and training to staff at VA medical facilities. Oversight of the DBCs is the responsibility of local medical facility leadership, usually the facility chief of staff.

VHA Directive 2012-026, Sexual Assaults and Other Defined Public Safety Incidents in Veterans Health Administration (VHA) Facilities, September 27, 2012.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

VHA does not maintain a national database or statistics on violent and disruptive patient incidents, although incidents involving police intervention are tracked by VA police. In 2010, at the request of the Office of Public Health and Environmental Hazards, VHAs Healthcare Analysis and Information Group conducted a survey of VA medical facilities to learn more about their programs to manage violent and disruptive patients. The four survey objectives were to: Determine current practices in relation to seriously disruptive, threatening, violent, and seriously non-compliant patient behaviors. Determine the extent to which medical facilities experienced disruptions to medical care due to difficult patients, as evidenced by numbers of patients banned or barred and/or self separated from care during FY 2009. Evaluate the impact of proposed changes in Federal regulation, which previously allowed facilities to ban or bar patients from nonemergent care. Ascertain alternative approaches to managing these challenging patients. All 140 VA medical facilities surveyed responded. The survey found that in fiscal year (FY) 2009: All 140 facilities had established DBCs (or equivalents). One-hundred thirty-eight (99 percent) facilities referred a total of 6,472 cases to the DBCs; 2 facilities reported no referrals. Of the 6,472 cases, 1,599 (25 percent) resulted in a Category I PRF, with 640 of the cases requiring police presence during subsequent patient visits. Thirty-two (23 percent) facilities reported banning or barring patients from their facilities; 10 of the 32 provided patients fee basis care. One-hundred twenty-six (90 percent) facilities reported using the EHR to document disruptive behavior. Prior OIG Reviews. Since 2004, the OIG has published two reports that addressed patient violenceHealthcare Program Evaluation: Veterans Health Administrations Management of Violent Patients7 and Combined Assessment Program Summary Report: Management of Workplace Violence in Veterans Health Administration Facilities.8 In the first report, issued in 2004, we made several recommendations to address both how facilities respond to violent incidents that are in progress, and how facilities should collect and analyze violent incidents data to develop prevention strategies. We also made
Healthcare Inspection Healthcare Program Evaluation: Veterans Health Administrations Management of
Violent Patients, Report No. 02-01747-139, May 3, 2004.
8 Combined Assessment Program Summary Report: Management of Workplace Violence in Veterans Health
Administration Facilities, Report No. 11-00215-194, June 14, 2012.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

a recommendation related to internal communication of patient risks. The second report, issued in 2012, addressed the broader issue of workplace violence, which includes not only patient violence but also violence committed by employees and visitors. This report included several recommendations to implement comprehensive national guidance on managing disruptive and violent behaviors, to monitor compliance with policies, and to provide specialized training to employees who work in high-risk areas. In addition, in 2010, we published a hotline report, Healthcare Inspection: Review of Quality of Care at a VA Medical Center (Report 10-03237-41, December 9, 2010), in which we concluded that the DBC of a medical center could have been more consistent in how it responded to a patient threat, more patient-centered in how it communicated its conclusions to the patient, and more thorough in how it documented discussions and deliberations supporting its decision to move the patients care from a community based outpatient clinic located near his home to the main facility located farther away.

Scope and Methodology

To address the review objectives, we interviewed officials from VHAs Behavioral Threat Management Program and Prevention and Management of Disruptive Behaviors Program, the VHA National Center for Ethics in Health Care, and VA Police Services. We also reviewed relevant VA and VHA policies and procedures, Federal regulation, and medical literature. To gain a better understanding of operational aspects of DBCs, we randomly selected 30 VA medical facilities, representing 18 Veterans Integrated Service Networks (VISNs), for further review. For these 30 facilities, we reviewed applicable policies and procedures, meetings minutes, and referral lists pertaining to the DBCs. We interviewed DBC officials and other program officials, as appropriate. We also reviewed the EHRs for all 889 patients who were reportedly referred to DBCs at the 30 facilities in FY 2011. We did not review disruptive incidents perpetrated by visitors, staff, or other non-VA patients. Furthermore, our review did not address how facilities respond to immediate incidents or crises (often referred to as Code Greens or Code Purples); rather, we focused on how facilities respond after such incidents and/or to prevent incidents. We conducted the review in accordance with Quality Standards for Inspection and Evaluation published by the Council of the Inspectors General on Integrity and Efficiency.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

Results and Conclusions

Overview of Disruptive Behavior Committees

All 30 facilities in our sample had active DBCs. DBC chairpersons were primarily clinical staff, including mental health service chiefs, staff psychiatrists and psychologists, quality management officials, a psychiatric nurse, and a chief of occupational health. At one facility, an administrative officer to the chief of staff served as the DBC chairperson. All 30 sites reported that representatives from VA Police (typically the Chief or Assistant Chief) are members of the DBC, and VA Police are a primary source of referrals to the DBCs. At several facilities, VA Police take the lead in collecting and trending data on patient incidents. In addition, at least one facility reported that its Chief of Police participated in a VISN-level DBC workgroup to identify best practices and standardize approaches across the VISN. In FY 2011, the 30 facilities reported at least 980 disruptive incident referrals to DBCs, involving 889 unique patients (indicating that some patients had multiple referrals during this period).9 The number of unique patient referrals by site ranged from 1 to 104, with a median of 27 per site. The 889 unique patient referrals do not include DBC reviews of existing patient record flags, flags transferred from other facilities, or DBC deliberations on incidents that occurred prior to or after FY 2011. Common sources of referrals included VA Police; clinical providers; and administrative staff, including reception area and telephone clerks. Occasionally, the DBCs also received referrals from other sources, such as shuttle bus drivers, fee-basis providers, and employees of the Veterans Benefits Administration who work at nearby regional offices. Referrals were made to the DBCs using a variety of reporting mechanisms, including consult requests in EHRs, emails, telephone calls, in-person reports, local DBC referral forms, and incident and police reports. At some facilities, referrals were generated when members of the DBC responded to violent patient incidents as part of an incident response team. None of the 30 facilities had a universally used, standard reporting mechanism.

Issue 1: Description of Disruptive Behavior Management at Selected Facilities

Finding
VHA facilities vary significantly in how they identify and manage disruptive patient behavior, especially in regards to defining disruptive behavior, documenting incidents
9

We use the phrase, at least because, in their tracking and reporting, some facilities combined incidents into one referral or reported multiple incidents without providing specific numbers.

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

and interventions in EHRs, and employing interventions to prevent and/or minimize the risk of further incidents.
VHA Facilities Use Various Definitions of Disruptive Behavior

Although VHA policies addressing DBCs provide general definitions of disruptive behaviors, they do not provide detailed guidance on what specific actions constitute disruptive behavior. In an attachment about threat assessments, VHAs PRF policy includes examples of common sentinel events that should lead to a violence risk assessment. Examples include: Physical violence against patients or staff Documented acts of repeated violence against others Credible reports of verbal threats of harm against specific individuals, patients, staff, or VA property Reports of possession of weapons or objects used as weapons in a health care facility Documented history of repeated nuisance, disruptive, or larcenous behavior that disrupts the environment of care Documented history of repeated sexual harassment toward patients or staff The Federal regulation addressing disruptive behavior also does not define what constitutes disruptive behavior.10 Instead, it states that a facility may restrict the time, place, and/or manner of the provision of medical care when it is determined that a patients behavior has jeopardized or could jeopardize the health or safety of other patients, VA staff, or guests at the facility, or otherwise interfere with the delivery of safe medical care to another patient at the facility. Without clear guidance from VHA or Federal regulation, VHA facilities have generally defined or described for their own use what constitutes disruptive behavior. Our review of facility policies for the 30 sampled sites found that 21 policies included descriptions, examples, and/or definitions of disruptive behaviorthe remaining 9 policies did not address this topic. As shown in the following example, some facilities provided broad definitions of the terms, disruptive, threat, and violence.11 1. Disruptive any incident in which the delivery of care or services is interrupted or impeded.

Code of Federal Regulations; Title 38: Pensions, Bonuses, and Veterans Relief; Part 17 Medical; Section 17.107 VA response to disruptive behavior of patients.
11 VA Healthcare Upstate New York Network Memorandum 10N2-061-11, Management of Disruptive,
Threatening or Violent Behavior, November 15, 2011.

10

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

2. Threat any verbal or non-verbal expression of an intention to inflict pain or injury or to cause annoyance or alarm. 3. Violence any physical force exerted to violate, damage, or abuse another person, equipment, or property. A small number of facilities provided more detailed definitions and examples, as well as examples of what disruptive behavior does not include, such as a patient exercising his or her right to decline treatment. The following is an example of more detailed guidance provided to staff at one facility:12 a. DISRUPTIVE BEHAVIOR: For the purposes of this policy, disruptive behavior is defined as: (1) Behavior by patients, patient families, patient representatives, employees, volunteers, visitors, and all other persons on VA grounds that is intimidating, threatening, or dangerous and may pose a threat to the health or safety of other patients, VA employees, or visitors at the facility; (2) Behavior that interferes with the delivery of safe medical care to other patients at the facility; or (3) Behavior that impedes the operations of the facility. (4) Specific examples of disruptive behavior include, but are not limited to: verbal abuse such as name-calling, racial or ethnic epithets, sexual harassment, loud or profane language; direct, indirect, or implied threats; physical abuse (e.g., bumping, shoving, slapping, striking, or inappropriate touching); unwanted approaches toward or contact with others; possession or brandishing of weapons; persistent or intense outbursts; or disruptive behavior to such a degree that it interferes with the ability of other patients to access care (e.g., excessive calls to the 24-hour call center without a clear medical need, leaving repeated voicemail messages on a providers voice mail in such numbers that no other patients are able to leave messages). (5) Disruptive behavior may be exhibited in a personal encounter or deployed in any media, including telephone calls or messages, email, website postings, social media, video, or written or printed form. (6) Disruptive behavior is defined by whether its occurrence would create fear in a reasonable person or would be perceived by a reasonable person to interfere with the delivery of health care or the performance of employee duties.

Charles George VA Medical Center, VAMC Memorandum 637-2011-11-84, Disruptive Behavior Committee, dated November 28, 2011.

12

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

(7) The definition of disruptive behavior does not depend upon the disruptive persons stated intentionality or justification for his/her behavior, the presence of psychological or physical impairment, whether the person has decision-making capacity, or whether the person later expresses remorse or any apology. (8) Disruptive behavior does not include non-adherent behavior.13 Our review of EHRs for the 889 unique patients referred to DBCs in FY 2011 found a broad range of incidents and behaviors resulting in referrals, as shown in Table 1 below.
Table 1. Types of Incidents and Behaviors Referred to Disruptive Behavior Committees in FY 2011
Type of Incident or Behavior Percent of Unique Patient Referrals* (N=889)

Verbal Aggression or Verbal Attack Other Drug Seeking Behavior Physical Violence Toward a Person Threat of Harm (No Weapon Mentioned) Unspecified or Vague Threat Threat of Harm with a Weapon Threat of Suicide Threat of Property Damage Weapon Brought on VA Property Active Suicide Attempt

56.6% 21.5% 20.7% 16.8% 15.0% 9.4% 7.9% 6.4% 2.2% 1.6% 1.2%

Source: VA OIG (*Note: Percentages do not equal 100 because 45 percent of the unique patient referrals involved multiple incidents or behaviors.)

Verbal aggression or verbal attack, such as screaming or swearing at facility staff or using racial or ethnic slurs, was most commonly reported (57 percent). About 21 percent of referrals were for drug seeking behavior, including demanding narcotics, doctor shopping, or not complying with narcotic contracts. The Other category included a wide range of incidents or behaviors ranging from dangerous and life threatening to annoying. Examples include setting a hospital bed on fire, pulling a fire alarm, throwing a chair through a window, slamming doors, theft from the VA canteen, venting frustration about VA services and/or wait times, threatening lawsuits or to have people fired, and frequent unwarranted visits to the emergency department or telephone calls to facility staff. Our discussions with VA and VHA officials, as well as DBC officials at the 30 sampled sites, combined with our review of facility policies and patient EHRs, found that at all facilities, incidents of physical violence, property damage, and threats of harm were
Non-adherent behavior refers to situations where patients may not follow or comply with recommended treatment plans or prescribed therapies.
13

VA Office of Inspector General

Management of Disruptive Patient Behavior at VA Medical Facilities

expected to be reported to the DBC. However, officials we spoke to expressed disagreement, and sometimes ambivalence, about the gray areas, such as patients who call excessively and take up staff time, patients who are unduly loud or frequently vocalize displeasure with their care or choice of facility providers, patients who routinely use profanity in their communications, or patients who are drug seeking. VA and VHA officials provided mixed views about how specific the guidance should be on what constitutes disruptive behavior. Some officials told us that they did not support having guidelines that are too specificinstead, they believe employees should report any actions or incidents in which they feel at risk. The officials concern is that overly prescriptive guidance will discourage employees from reporting potentially dangerous incidents and/or incidents with the potential to escalate. Other officials supported having more consistent guidelines and expressed the opinion that over-reporting may happen for certain types of issues because staff are not comfortable and/or not sufficiently trained to respond to patients with challenging behaviors. According to the Director of VA Police Service, the challenge is finding a balance between employees feeling safe and supported by VA Police but also learning and using tools, techniques, and customer service approaches to prevent escalation of behaviors. Providing clear guidance to facility staff may help to ensure that staff understand their role in supporting patients rights to choose and participate in their care, including nonadherence to treatment plans, and avoid making referrals related to patients exercising their rights or for issues that could be addressed through better customer service. Furthermore, having a more uniform definition of what constitutes disruptive behavior, may strengthen VHA program officials ability to collect, aggregate, and trend systemwide data on incidents and behaviors, which could be useful in developing programs to address staff training, behavioral interventions and prevention, risk management, and care delivery issues. Although VA Police reporting systems capture information on many patient incidents and behaviors, it is important to note that this information is limited in regards to disruptive patient behavior because not all incidents referred to DBCs necessarily require VA Police intervention and/or result in Uniform Offense Reports (UORs).
VHA Facilities Use Different Approaches for Documenting Disruptive Patient Behavior in EHRs

VHA policy addressing DBCs only refers to documentation requirements for placing behavioral flags in patient records. It does not address requirements for other disruptive incidents that do not result in PRFs. In our reviews of facility policies and patient EHRs and discussions with VHA and DBC officials, we focused on the question: After a disruptive incident, what, if anything, should be documented in a patients EHR?

VA Office of Inspector General

10

Management of Disruptive Patient Behavior at VA Medical Facilities

Facility policies varied in how they addressed documentation. Of the 30 facility policies we reviewed, only 6 specifically included requirements to document incidents in EHRs. Sixteen policies addressed documentation requirements only in cases where PRFs are entered in EHRs. The remaining eight policies included no discussion of documentation requirements (in either patient EHRs or elsewhere). Our review of EHRs for patients referred to DBCs in FY 2011 found wide variation in what facilities documented in regards to disruptive incidents, referrals to DBCs, deliberations by DBCs, and the recommendations or actions of DBCs. The incidents resulting in referrals to DBCs were commonly documented, but little else was. Table 2 shows what facilities documented for the 889 unique patient referrals.
Table 2. Documentation of Disruptive Incidents in Electronic Health Records
Topic of Documentation Percent of Unique Patient Referrals (N=889)

Incident (Resulting in DBC Referral) Referral to DBC DBC Deliberations DBC Recommendations/Actions
Source: VA OIG

77.6% 18.4% 12.9% 30.1%

For the majority of unique patient referrals, we found documentation in the EHRs describing the incidents that resulted in referral to DBCs. This documentation was most often contained in progress and administrative notes written by treating providers, clinic or unit nurses, social workers, intake or telephone staff, or other clinical staff who witnessed incidents. However, we found significantly less documentation of referrals to DBCs; DBC deliberations, such as risk assessments or discussions of factors considered by DBCs; and resulting DBC recommendations or actions. Documentation of referrals mostly occurred in progress notes and/or consult requests. Only 3 of the 30 facilities used standard progress note titles for referrals. When documented, DBC deliberations were generally found in PRFs and addressed the factors DBCs considered in recommending PRFs. We found formal risk assessments in less than 2 percent of the EHRs. Recommendations or actions by the DBC were documented in progress notes, consult requests, and/or PRFs. DBC officials noted that the majority of documentation related to disruptive incidents is kept in records outside of EHRs and usually maintained by the DBC chairperson or VA Police. These records typically include documents such as referrals to the DBC via template forms, emails, and reports of contact, as well as incident reports, risk assessments, UORs, and letters or other contacts to Veterans. Most DBC officials considered DBC meeting minutes to be the official documentation of DBC deliberations and recommendations.

VA Office of Inspector General

11

Management of Disruptive Patient Behavior at VA Medical Facilities

VHA and DBC officials and other facility staff shared different perspectives on what they believed should be documented in EHRs. One perspective is that patients may be stigmatized by putting too much information about disruptive incidents in their EHRs, resulting in providers being unwilling to provide care or impeding provider-patient communications. An alternative perspective is that because the EHR serves as a repository of clinically relevant patient information and a communication instrument between clinicians, it should include information about incidents that may affect the delivery of care, especially considering that Veterans often seek care at multiple VHA facilities. Officials we spoke to agreed that, like any other EHR documentation, documentation of disruptive patient incidents should be objective and factual, and documentation of DBC recommendations or actions should be justified by the relevant facts. The underlying question remains as to what should be documented in EHRs regarding disruptive patient behavior and how should it be documented. Our EHR reviews and discussions with VHA officials support that there is no clear guidance or agreement on this issue. The risk of not having clear documentation guidelines is that EHRs may not give providers and other VHA clinicians the information they need to fully address patients needs; to identify and act on escalating behaviors occurring over a period of time and/or observed by different clinical staff; or to implement or follow through on interventional strategies to reduce the risk of harm and injury to staff, visitors, or other patients.
VHA Facilities Use a Variety of Interventions for Patients Referred to DBCs

In responding to disruptive patient incidents, Federal regulation permits facilities to restrict the time, place, and/or manner of care, such as requiring a patient to receive care at a parent facility instead of a community based outpatient clinic or receive care on a fee-basis in the community. However, facilities may not terminate VA care for eligible Veterans. Specific restrictions identified in regulation include, but are not limited to: Specifying the hours in which non-emergent care will be provided Arranging for care and services to be provided in a particular patient care area Arranging for care and services to be provided at a specific site of care Specifying the provider and/or other personnel who will be involved in a patients care Authorizing VA providers to terminate an encounter immediately if certain behaviors occur Our review of facility policies and interviews with DBC officials found that facilities use a variety of interventions for patients referred to DBCs and that they generally complied with Federal requirements. Most DBC officials reported that interventions are

VA Office of Inspector General

12

Management of Disruptive Patient Behavior at VA Medical Facilities

determined on a case-by-case basis, taking into consideration the severity of the incident, whether anyone was injured, prior history of incidents, diagnoses, and other relevant psychosocial factors. DBC officials at 18 of the 30 sampled facilities reported using some type of clinical or law enforcement risk/threat assessment tool to assist in determining appropriate interventions; although, they reported using these tools for only selected cases, not all cases. DBC officials at the other 12 facilities did not use formal assessment tools; instead, they reported that they used their clinical (or law enforcement) judgment to assess risks and determine appropriate interventions. We found documentation of interventions in the EHRs for 268 (30 percent) of the 889 unique patient referrals. This is due, in part, to the documentation issues discussed previously, but it may also reflect that DBCs do not make recommendations or take action on all referrals because actions are not always warranted. Instead, as several DBC officials told us, they often opt to watch and wait when a patient exhibits disruptive behavior and either has no prior incidents or has other mitigating factors that may have contributed to the behavior (such as diagnosis or psychosocial factors). Table 3 shows the common interventions documented by facilities in the EHRs.
Table 3. Interventions in Response to Disruptive Behavior for
Documented Unique Patient Referrals

Percent of Unique Patient Referrals with Documented Interventions (n=268)

Documented Intervention

Category I Patient Record Flags (National) Other Police Check-In Required Letter to Patient Category II Patient Record Flags (Local) Police Escort Required No Action Care Transferred to Main Facility or Barred from CBOC Counseling of Patient Clinical Intervention Health Care Agreement Clinical Warnings Banned from Main Facility

64.6% 26.5% 25.4% 12.7% 11.9% 11.9% 10.4% 9.0% 4.9% 4.9% 3.0% 1.1% 1.1%

Source: VA OIG (*Note: Percentages do not equal 100 because multiple facilities may use multiple interventions to address disruptive behavior.)

Category I PRFs, which are national EHR flags indicating high risk for violent or disruptive behavior, were documented most frequently. Although these Category I PRFs accounted for almost 65 percent of all documented interventions, it should be noted that Category I PRFs were used as interventions for approximately 20 percent of all the unique patient referrals (889). Requiring patients to check in with VA Police prior to appointments was also commonly used (25 percent), as were warning letters to patients
VA Office of Inspector General 13

Management of Disruptive Patient Behavior at VA Medical Facilities

(13 percent) and police escorts or standby for appointments (12 percent). Facilities also used Category II PRFs, which are local flags that indicate potential patient risks, most frequently for disruptive incidents involving drug-seeking behavior. Interventions in the Other category included assigning new primary care providers, assigning providers and staff of a different gender, assigning care coordinators or case managers, and requiring the presence of two staff members during patient care encounters. Our review found very few instances of banning patients from facilities (1 percent), and in these instances, patients were offered services either at other VA facilities or on a fee basis in the community. We found no documented instances where facilities banned patients from VHA care entirely. However, several DBC officials acknowledged that in the past, this practice did occur and that their facility policies had not yet been updated to reflect current Federal regulation, which prohibits this practice. Interestingly, although we were unable to fully quantify it, we found from our policy reviews, interviews, and EHR reviews that some facilities tended to view disruptive behavior as indicators of underlying psychosocial and/or patient satisfaction concerns and preferred clinical interventions, such as changing providers, interviewing and counseling patients, using Healthcare Agreements, or assigning care coordinators. Other facilities appeared to favor behavioral approaches that more frequently involved police check-ins, police escorts to appointments, and restricted access to services. While both approaches are consistent with VHA policy and Federal regulation, this observation just serves as a reminder that DBCs are considered clinical committees and that interventions determined by DBCs should be individualized and narrowly tailored to address the patients disruptive behavior.14 Evaluation of a patients disruptive behavior should also consider whether the behavior resulted from the patients individual fears, preferences, or perceived needs.15

Conclusion
Our review found significant variation in how VHA facilities identify and manage disruptive patient behavior. VHA does not have a system wide, commonly accepted definition of what constitutes disruptive behavior. As a result, most facilities have applied their own definitions of disruptive behavior, and the terminology used by facilities to describe incidents is not consistent between facilities. VHA also has not established guidelines for what information should be documented in regards to disruptive incidents and how or where this information should be documented. While most facilities documented disruptive incidents in EHRs, information about referrals to DBCs, DBC deliberations, and resulting recommendations was documented to a lesser degree. This raises the question of what should be documented in the EHRthe primary communication tool between providersto give providers the information they need to
14 15

Title 38, Code of Federal Regulations, Part 17, Responding to Disruptive Patients. ibid.

VA Office of Inspector General

14

Management of Disruptive Patient Behavior at VA Medical Facilities

provide safe, quality health care but also to minimize the potential for stigmatizing patients. Lastly, facilities used a wide range of interventions to address disruptive behavior; however, all of the interventions appeared to comply with the Federal regulation. Recommendation 1: We recommended that the Under Secretary for Health ensure that VHA program officials provide additional guidance on what constitutes disruptive behavior and establish common terminology. Recommendation 2: We recommended that the Under Secretary for Health ensure that VHA program officials develop guidelines for what information VHA facilities should document regarding disruptive incidents and where this information should be documented.

Issue 2: Data Collection and Analysis of Incidents

Findings
VHA facilities do not consistently collect and analyze data on disruptive incidents. VHA policy requires DBCs to collect and analyze incidents of patient disruptive, threatening, or violent behavior and to identify system problems and training needs related to the prevention and management of disruptive behavior. However, the policy does not provide guidance on what specific information facilities need to collect or methods for collecting data, such as a national reporting system or template formats. In our interview with DBC officials at 30 facilities, 12 reported that they were not tracking or trending data on disruptive incidents. Officials at the remaining 18 facilities reported that they collected information to varying degrees9 of the 18 facilities collect data on multiple factors, such as dates/times of specific incidents, locations, patients/staff involved, reasons, injuries, and resulting interventions. The other nine facilities collect data on just a limited number of factors, such as the overall number of referrals or the type of referral. Of the nine facilities that collect data on multiple factors, officials at seven reported that they have used the data to identify trends to help improve staff training and address other system issues. Our review of referral data obtained from the 30 facilities for FY 2011 supported what they told us. Some facilities submitted detailed spreadsheets containing information on the dates and times of incidents, patient names, locations of incidents (for example, units, clinics, or administrative areas), descriptive details of incidents, and descriptions of the interventions recommended and/or taken in response to the incidents. Other facilities provided us with lists that included just the names of patients, dates of incidents (or referrals to the DBC), and general descriptions of the incidents (for example, Disruptive Behavior or Physical Violence).

VA Office of Inspector General

15

Management of Disruptive Patient Behavior at VA Medical Facilities

DBC officials cited several reasons for why they did not consistently collect and analyze data on disruptive incidents and/or only collected limited data, including not having enough referrals to perform meaningful trend analysis or relying on VA Police or the Patient Safety Manager to perform this function. Several DBC officials cited the lack of administrative support to the DBC as a significant challenge to collecting and analyzing data and using the results to recommend system changes to facility leadership. At most of the 30 facilities, DBC chairpersons are clinical staff, and their DBC function is a collateral one, which limits the time they can spend on the administrative aspects of managing the DBC. Those facilities that do collect and analyze incidents of disruptive behavior report that this information has been useful to identify human resources, environmental, and system factors that may help to reduce or prevent incidents. DBC officials report that this information is especially helpful in developing and enhancing staff training and focusing on individual staff members, units, or clinics where additional training might be warranted. Other examples of information facilities have learned from analyzing collected DBC data include: A facility determined that many referrals were due to changes in providers in the behavioral health unit. A facility found that a newly hired staff member required additional training on how to manage and de-escalate disruptive patients. A facility determined that a high number of incidents occurred in its beneficiary travel office, so it implemented a new process by assigning a volunteer to assist Veterans with paperwork. Several facilities identified high use of narcotics for chronic pain as a contributing factor to disruptive incidents; they are using this information to look at their pain management programs and/or prescribing practices. NIOSH has identified numerous risk factors for disruptive patient behavior, such as long wait times, understaffing, and poor employee training. (See page 2 for more detailed list of the NIOSH factors.) Without sufficient time, administrative support, or tools to collect and analyze data on disruptive incidents, VHA facilities do not have the complete information they need to identify their unique risk factors and implement process, system, and structural improvements that may prevent or minimize disruptive behavior incidents.

Conclusion
VHA facilities do not consistently collect and analyze data on disruptive incidents, as required by VHA policy. Without this information, facilities with higher incidence rates and VHA program officials do not have sufficient information to identify system problems and training needs related to the prevention and management of disruptive

VA Office of Inspector General

16

Management of Disruptive Patient Behavior at VA Medical Facilities

behavior. VHA policy does not provide guidance on what specific information facilities need to collect or methods for collecting data, such as a national reporting system or template formats. Furthermore, DBC officials at several facilities reported that they do not have adequate administrative resources to meet this policy requirement. Recommendation 3: We recommended that the Under Secretary for Health ensure that VHA program officials provide guidance to VHA facilities on collecting and analyzing data on disruptive incidents. Recommendation 4: We recommended that the Under Secretary for Health consider implementing a national reporting system or data collection template for disruptive patient incidents.

Issue 3: Assignment of Patient Record Flags

Findings
Although not part of our original objectives, during our review of EHRs, we found significant delays in assigning Category I PRFs to patients deemed to be at high risk. According to VHA policy, a Category I PRF is intended to alert VHA employees to patients whose behavior may pose an immediate threat to the patient, other patients, facility employees, and visitors. Facilities are responsible for establishing a process for requesting, assigning, reviewing, and evaluating PRFs. Typically, DBCs recommend when PRFs should be assigned, and facility chiefs of staff make final determinations about flag placements in the EHRs. Actual flag assignments in the EHRs are usually the responsibility of the DBC chairperson. Although VHA policy indicates that Category I PRFs are intended for patients who pose an immediate threat, it does not include a timeliness standard for assigning flags. However, given the intent of Category I PRFs, sound clinical and business practice would suggest that facilities assign PRFs in EHRs as soon as possible after triggering events in order to alert staff who have a need to know about potential risks and recommended interventions. We reviewed the 173 Category I PRFs assigned to patients at the sampled sites in FY 2011. As noted under Issue 1, these PRFs accounted for almost 65 percent of all documented interventions (268) and approximately 20 percent of all the unique patient referrals (889). Of the 173 PRFs, 142 (82 percent) were newly assigned in FY 2011, and 20 (12 percent) were PRFs continued from prior periods. For the remaining 11 PRFs, the EHRs did not contain sufficient information about either the incidents or the PRF assignments for us to classify them.

VA Office of Inspector General

17

Management of Disruptive Patient Behavior at VA Medical Facilities

For the 142 newly assigned Category I PRFs, we found that the elapsed number of days between the actual assignments in the EHRs and the incidents that triggered the PRFs varied significantly. The median number of elapsed days was 30, with a range of 0 to 469 days. Chart 1 shows the elapsed days for assigning new PRFs for the sampled facilities.
Chart 1. Elapsed Days from Triggering Incident to
Patient Record Flag Assignment in Electronic Health Records

ElapsedDays, 11 49,32.4% ElapsedDays, 50 100, 23.9% ElapsedDays, 1 10,18.3% ElapsedDays, 0,10.6% ElapsedDays, >100,14.8%

0
Source: VA OIG

About 29 percent of the PRFs were assigned within 10 days of the triggering incident, including about 11 percent that were assigned on the day of the incident. However, facilities took more than 100 days to assign about 15 percent of the PRFs. This data is concerning because Category I PRFs are the primary tool VHA uses to provide nationwide notification that a patient is considered high risk for violent behavior. If facilities are delaying PRF assignments due to inefficient processes, internal communications, or other factors, VHA providers and other staff (both at the assigning facility, as well as other VHA facilities patients may visit) do not have the information they need to provide quality care in safe environments. In addition, such significant delays in placing PRFs puts into question the validity of the initial flag decision.

Conclusion
VHA facilities were not timely in assigning a significant number of Category I PRFs, which are intended to alert VHA employees to patients whose behavior may pose an immediate threat to the patient, other patients, and facility employees and visitors. More than a third of PRFs were assigned 50 days or more after triggering incidents, and about 15 percent were not assigned until 100 days after the triggering incidents. Although VHA policy on PRFs does not include a timeliness standard for assigning PRFs, sound

VA Office of Inspector General

18

Management of Disruptive Patient Behavior at VA Medical Facilities

clinical and business practice suggest that flags intended to alert employees of immediate threats or risks should be assigned as soon after the triggering events as feasible. Recommendation 5: We recommended that the Under Secretary for Health ensure that VHA facilities implement procedures to ensure more timely assignment of Category I PRFs.

Comments
The Under Secretary for Health concurred with all the findings and recommendations. See Appendix A (pages 2023) for the full text of his comments. We will follow up on the corrective actions until all recommendations have been fully implemented.

VA Office of Inspector General

19

Management of Disruptive Patient Behavior at VA Medical Facilities Appendix A

Under Secretary for Health Comments

VA Office of Inspector General

20

Management of Disruptive Patient Behavior at VA Medical Facilities

Under Secretary for Health Comments to Office of Inspector Generals Report The following comments are submitted in response to the recommendations in the Office of Inspector Generals report: OIG Recommendations Recommendation 1: We recommended that the Under Secretary for Health ensure that VHA program officials provide additional guidance on what constitutes disruptive behavior and establish common terminology. VHA Comments Concur Using the definition of patient disruptive behavior published in 38 CFR 17.107 as a foundation for additional guidance, VHA will provide specific examples of what patient behaviors do and do not constitute disruptive behaviors in order to establish common terminology across VHA facilities. The definition of, and additional guidance regarding, patient-generated disruptive behaviors will be disseminated to VHA Patient Disruptive Behavior Committee Chairs. Status: In Process Target Completion Date: June 30, 2013

Recommendation 2: We recommended that the Under Secretary for Health ensure that VHA program officials develop guidelines for what information VHA facilities should document regarding disruptive incidents and where this information should be documented. VHA Comments Concur As part of the roll-out of a national, secure, web-based, standardized disruptive behavior incident reporting and data management system (see response to Recommendation 4 below), VHA will develop guidelines for what information will be documented, and where, regarding patient disruptive behavior incidents. Status: In Process Target Completion Date: See Recommendation 4

VA Office of Inspector General

21

Healthcare Inspection, Quality of Care Issues, at the Anywhere VA Healthcare System, Anywhere, State

Recommendation 3: We recommended that the Under Secretary for Health ensure that VHA program officials provide guidance to VHA facilities on collecting and analyzing data on disruptive incidents. VHA Comments Concur As part of the roll-out of a national, secure, web-based, standardized disruptive behavior incident reporting and data management system (see response to Recommendation 4 below), VHA will develop guidelines regarding the collection and analysis of data on patient disruptive behavior incidents. Status: In Process Target Completion Date: See Recommendation 4

Recommendation 4: We recommended that the Under Secretary for Health consider implementing a national reporting system or data collection template for disruptive patient incidents. VHA Comments Concur VHA currently is piloting, in Veterans Integrated Service Networks (VISN) 8 and 19, a disruptive behavior reporting and data management system developed in VISN 5. The system is secure, available to all VHA employees via the Intranet homepage at the facility, includes data tracking and incident management features (providing a mechanism for standardizing data collection and analysis concerns in Recommendation 3), and has the potential to be upgraded to include electronic health record notation features (providing a mechanism for standardizing the documentation issues raised in Recommendation 2). The Office of Informatics and Analytics (OIA) and the Office of Analytics and Business Intelligence (OABI) have agreed to support the pilot, reporting system upgrade, and national roll-out of this system in collaboration with VHAs Behavioral Threat Management Program office. Status: In Process Target Completion Dates: VISN Pilots JanuaryJune 2013 Pilot Review and Upgrade Recommendations Summer 2013 OABI Upgrades Fall 2013 Plan for VISN-by-VISN Systematic Roll-Out December 31, 2013

VA Office of Inspector General

22

Healthcare Inspection, Quality of Care Issues, at the Anywhere VA Healthcare System, Anywhere, State

Recommendation 5: We recommended that the Under Secretary for Health ensure that VHA facilities implement procedures to ensure more timely assignment of Category I PRFs. VHA Comments Concur VHA will develop guidance regarding the maximum allowable timelines involved in the process of assessing the appropriateness and implementation of a Category I Behavioral Patient Record Flag to address violence risk posed by patient disruptive behavior(s) and disseminate to Disruptive Behavior Committee Chairs. Status: In Process Target Completion Date: June 30, 2013

VA Office of Inspector General

23

Management of Disruptive Patient Behavior at VA Medical Facilities Appendix B

OIG Contact and Staff Acknowledgments

Contact Contributors
For more information about this report, please contact the Office of Inspector General at (202) 461-4720. Claire McDonald, MPA, Project Leader Michael Shepherd, MD Jeanne Martin, PharmD Annette Acosta, MN, RN Limin Clegg, PhD Sheila Cooley, RN Jeffrey Joppie, BS Elaine Kahigian, RN, JD Frank Keslof, EMT, MHA Karen McGoff-Yost, LCSW Glen Pickens, RN, MHSM Clarissa Reynolds, CNHA, MBA Jarvis Yu, MS

VA Office of Inspector General

24

Management of Disruptive Patient Behavior at VA Medical Facilities Appendix C

Report Distribution
VA Distribution Office of the Secretary Veterans Health Administration Assistant Secretaries Office of General Counsel Non-VA Distribution House Committee on Veterans Affairs House Committee on Veterans Affairs, Subcommittee on Health House Committee on Oversight and Government Reform House Appropriations Subcommittee on Military Construction, Veterans Affairs, and Related Agencies Senate Committee on Veterans Affairs Senate Committee on Homeland Security and Governmental Affairs Senate Appropriations Subcommittee on Military Construction, Veterans Affairs, and Related Agencies National Veterans Service Organizations Government Accountability Office Office of Management and Budget

This report is available on our web site at www.va.gov/oig

VA Office of Inspector General

25