Report on Police Interactions with Persons in Mental Health Crisis

A draft of the document reproduced below, “Report on Police Interactions with Persons in Mental Health Crisis,” was circulated Feb. 14 during a press conference called by the Portland Police Bureau in response to the death of Brad Morgan, who was shot and killed by Officer Dave Scott and Sgt. John Holbrook on Jan. 25 after calling 911 to announce he was planning to suicide.
The document is framed as evidence and product of the PPB’s commitment to “transparent review,” and also, apparently, serves as an occasion to amplify their contention that shooting persons with mental illness is an excusable, unavoidable result of factors beyond their control — with the scholarly flourish of charts and footnotes.
The report was not publicly released until now.
A fully formatted PDF version of this report as created by the PPB is available here.

Report on Police Interactions with

Persons in Mental Health Crisis

Prepared for:

Mayor Sam Adams; Commissioner Amanda Fritz; Chief Michael Reese


Dr. Mark Kaplan, Portland State University; Rachel Philofsky, M.S., Multnomah County; Joan Rice, Multnomah County; Bill Osborne, Multnomah County; Patrick Jones, City of Portland Bureau of Emergency Communication; Dr. William Nunley, Cascadia Behavioral Healthcare, Care Oregon & The Bazelon Center for Mental Health Law; Dr. Derald Walker, Cascadia Behavioral Healthcare; Dr. Maggie Bennington Davis, Cascadia Behavioral Healthcare

Prepared by:

Greg Stewart, Sergeant, Portland Police Bureau Crime Analysis Unit; Liesbeth Gerritsen, Ph.D., Portland Police Bureau Training Division; Emma Covelli, M.S., Portland Police Bureau Crime Analysis Unit; Kris Henning, Ph.D., Portland State University College of Urban & Public Affairs


March 21st, 2012


The Portland Police Bureau recognizes and respects the integrity and value of human life. We understand that the decision to use force is the most important decision that an officer will make in the course of his/her career. In addition to the potential harm caused to the intended person, the use of force will emotionally, physically and psychologically impact the officer involved, the subject to whom force was directed (even if the person was uninjured physically), and the family and friends of both. The use of force can impact the community as a whole.1 These incidents have the potential to damage the legitimacy of the police in the eyes of the public and to break the bond of trust between the police and the community. Recognizing the impact the use of force has on the community, the Bureau is committed to a transparent review of these incidents in order to understand the root causes of these encounters and to improve our response. This commitment has helped the Bureau reduce the number of officer-involved shootings from ten in 1998 to four in 2011. It has also helped the Bureau to substantially reduce the use of any type of force.

There are still instances where officers are placed in situations requiring them to use force. On January 25, 2012, officers of the Portland Police Bureau (PPB) responded to a call of a suicidal subject armed with a handgun (later determined to be a replica) which culminated in the death of Brad Lee Morgan. In response to this incident, Mayor Sam Adams, Chief Michael Reese and Commissioner Amanda Fritz asked for a review of suicide calls and related data in our community. This led to a larger investigation of what factors may be influencing police interactions with persons in mental health crisis.2 The purpose of this report is to present the results of this investigation.

Executive Summary

Police-community interactions that involve a mental health crisis occur for multiple reasons. These calls fluctuate in frequency due to broader societal factors, such as growing homeless populations, changes in care for those with mental illness, and increasing demand due to structural factors (such as financial crisis). Understanding what influences the occurrence of these incidents, whether these incidents are increasing or decreasing over time, and what the outcomes of these incidents are is critical for future planning.

The Portland Police Bureau serves a community that has one of the highest rates of homelessness in the country and has suicide rates greater than the national average and many large counties, such as Los Angeles, San Francisco, and Philadelphia. Between 2001 and 2011, suicide calls to BOEC in the City of Portland grew from 630 calls to 1,2003 per year, a dramatic rise in call volume. Some of this rise in call volume may be due to increased demand for services, such as mental health and addiction services.

This increase in call volume for the police has been mirrored by a corresponding increase in adults provided with mental health services by Multnomah County. In 2004, 5,292 adults were served by Multnomah County mental health providers. By 2011, this number had grown to 10,062 persons served. Other indicators discussed later in this report, point toward increasing contacts between persons in mental health crisis and police.

Despite the increase in these contacts, use of force by Portland Police Officers has decreased substantially over the last four years.4  It is also important to note that the vast majority of calls for service involving a mental health crisis are successfully managed without the use of force and to the benefit of the person needing help.

The following list contains changes over time in potential indicators related to police contact with persons in mental health crisis: 5

  • Since 2004, Multnomah County has expanded mental health services provided to adults by 90% (this does not necessarily indicate increased need but demonstrates improved service delivery by Multnomah County and may indicate increased demand).
  • Since 2001, aggregate suicides have increased by more than 33% (although rates have risen less and decreased considerably since the mid-1990s).
  • Suicide calls responded to by the Portland Police Bureau have risen 90% since 2001.
  • Attempted suicides (not calls but actual attempts) have risen by nearly 13% since 2001.
  • Welfare check calls (a broad range of calls, only some of which are related to mental health) have risen 39% since 2001.
  • Investigations by Multnomah County’s Investigative Commitment Program (ICP) have risen nearly 70% since 1998 (this is exceptionally important as the ICP investigates holds for mental health issues prior to civil commitment and is an excellent indicator of increased demand related to issues surrounding mental health).


For purposes of this report, “persons in mental health crisis” encompasses a wide range of individuals, all of whom are in a distressed situation when contacting police. This may include persons with a severe and persistent mental illness, but it is important to recognize that this group is not homogeneous. This report attempts to identify those factors which may contribute to increased contact between the police and persons in mental health crisis.

Police officer assisted holds are either peace officer custody holds or Director’s custody holds. Peace officer holds are directly placed by the police on individuals who the officer believes are “a danger to themselves or others.” Director’s custody holds are placed by trained mental health professionals who are specifically authorized by the local Mental Health Authority to order persons to be taken into custody pursuant to ORS 426.233.

For purposes of this report, police officer-assisted holds include all police holds placed by Bureau officers and director’s holds in which the police assist (normally by transporting). The person held is transported to a medical facility (normally an emergency room) and then evaluated by a medical staff.


This preliminary report contains an analysis of Bureau of Emergency Communication (BOEC) call data between 2001 and 2011 related to suicide6 and other call types, an analysis of reports on the mental health system, statistical data related to suicide,7 and some published research. Interviews were conducted with BOEC staff, Police Bureau staff, community experts on suicide and mental health issues, as well as individuals involved in community mental health. Involuntary commitment, ICP, hold data was obtained from Multnomah County,8 as was data related to mental health services provided by the county. Some of the information gathered, such as comparison of officer-involved shootings, is taken from reports compiled by other entities (i.e. other agencies, published scientific journals etc.).

This report is broken into five sections. The first section details challenges particular to Multnomah County resulting in a large demand for mental health resources. The second section examines how this demand has increased over time, in particular for law enforcement. The third section details the outcomes of the increased interaction of police with persons in mental health crisis and includes a short review of the scientific literature around police use of force and persons with a mental illness. The fourth section explains some of the services and programs currently in place in the City of Portland. Finally, there is a list of recommendations for future analysis related to this issue.

Demand Levels for Multnomah County

Multnomah County Services to Adults

Since 2004, Multnomah County has nearly doubled the number of individuals provided with mental health services. In 2004, 5,292 adults were served by Multnomah County mental health providers. By 2011, this number had grown to 10,062 persons served.9

This increase in service provision does not necessarily indicate an increase in demand, but is consistent with other indicators pointing toward increasing need for mental health and addiction services.

Figure 1

Community members also list mental health and addiction issues as having great impact on the community. In a 2011 report, Sorvari & Mowlds found there, “was broad agreement across demographic lines that mental health, child abuse, and domestic violence were the three issues with the greatest impact on the community’s health.” This indicates a broad consensus that issues related to mental health are impacting the citizens of this county.


Suicide Rates in Multnomah County are Unusually High

Unfortunately and despite the efforts of mental health responders, Multnomah County has a high rate of suicide. Some experts believe that this can be an indicator of the workload placed on police when interacting with persons in mental health crisis.

In May 2010, Dr. Bill McConnell, Officer Kelly Dunn and Helynaa Brooke presented a report on workload estimates to the San Francisco Mental Health Board and the San Francisco Police Department (McConnell, Dunn, & Brooke, 2010). They identified a number of measures (including suicide rates and homelessness) which are potentially associated with police interactions with persons with a serious mental illness. The following is an excerpt from this report discussing how suicide rates may impact Officer-Involved-Shootings (OIS):

“Based on this data, more police effort would be predicted for San Francisco than most other counties. Also, the suicide rate may be particularly relevant for Officer-Involved Shootings (OIS) because a higher suicide rate in a jurisdiction might increase the likelihood of OISs as a function of suicide-by-cop.”

The data for Figure 2 is taken from the San Francisco report with the exception of the information for Multnomah County, which was added for this report.10 It displays the suicide rates per 10,000 people for selected California and Non-California counties in 2006.

Figure 2

The report notes that Sacramento County and Marin County’s suicides are inflated by the fact that they share jurisdiction of the Golden Gate Bridge. Multnomah County had a higher suicide rate (in 2006) than any other county in the study. The study also notes that suicides may be especially relevant when considering the potential for “suicide by cop.”

Given the high rates of suicide, it would be prudent for the Bureau to expect a higher level of interaction with individuals in these crisis situations. This trend appears to be ongoing. In 2009, the Oregon Health Authority reported that suicide rates in Oregon exceeded the national average by 35% (Stone, 2010). Again, these indicators would support the need to proactively refine the response to persons diagnosed with a severe mental illness as well as persons in mental health crisis.

Deinstitutionalization of Persons with Mental Illness in Oregon

Since the 1950s, there has been a national trend toward the deinstitutionalization of persons with mental illness. Between 1955 and 2005, public psychiatric beds in Oregon fell from a rate of 292.9 per 100k population to 19.2 per 100k population, consistent with the national trend toward deinstitutionalization.  A report by the Treatment Advocacy
Center, a national nonprofit organization dedicated to eliminating barriers to the timely and effective treatment of severe mental illness, cites an expert consensus of a need for 50 beds per 100k population (few states can meet this criteria) (Torrey, Entsminger, Geller, Stanely, & Jaffe).

As with the other indicators mentioned in this report, the impact of deinstitutionalization would support a continued increase in the demands placed on public safety as it relates to community mental health. Figure 3 demonstrates the change in the number of public psychiatric beds:Figure 3

*Note: the 50 bed per 100k is a recommendation by the Treatment Advocacy Center and does not represent universal opinion.

It is important to note that the Portland Police Bureau is not advocating for an increased use of institutionalization. In the words of one person interviewed, the 1950s were the, “dark ages,” of mental health care. The reduction in the use of institutionalization has had beneficial effects for the system as a whole (reduced costs, improved treatment of persons with mental illness and improved support for persons with mental illness civil rights). It has also been advocated for by groups, such as the Bazelon Center for Mental Health Law. The Bureau is only attempting to identify trends which may cause an increase in the number of interactions between police officers and persons in mental health crisis.

Oregon has One of the Highest Rates of Homelessness in the Nation

Oregon is tied with Hawaii for the highest state rate of homelessness in the United States (Homelessness Research Institute, 2012). While one report or time period may be an anomaly, this does not appear to be the case with regard to the issue of homelessness in Oregon. In 2009, Oregon had the third highest homeless rate in the nation (U.S. Department of Housing and Urban Development, 2010). Interestingly, this same report noted that Washington State has among the lowest rates of homelessness in the nation. Furthermore, between 2009 to 2011 the homeless population in Multnomah County increased between 7% and 9% (Kristina Smock Consulting, 2011). This increase in homelessness may be associated with increased interaction between police and persons in a mental health crisis.

The level of homelessness may impact police contacts with persons in mental health crisis because as much as 25% of all homeless persons suffer from mental illness (National Coalition for the Homeless, 2009). This report also noted the intersection between homelessness, mental illness and substance abuse. In Multnomah County, 50% of homeless persons report suffering from a disability (physical, cognitive/developmental disability, substance abuse or mental health issue) (Kristina Smock Consulting, 2011).

While most persons with mental illness are not violent, persons with mental illness who abuse substances can be at increased risk to behave violently (Elbogen & Johnson, 2009).

Mental health issues can also be exacerbated by conditions of homelessness and being homeless can increase the likelihood of having contact with the police. Homelessness may be a particularly relevant in predicting the volume of interactions between persons with mental illness and police.11

The National Institute of Mental Health funded a monograph (Epperson, Wolff, Morgan, Fisher, Frueh, & Huening, 2011) examining the over-representation of persons with serious mental illness in the criminal justice system and exploring current and future interventions aimed at addressing this issue. This report highlights the importance of other criminogenic risks which impact the rate at which persons with a serious mental illness interact with the criminal justice system. This report advocates a “person-place framework” which address both individual level characteristics (such as mental illness, addiction, antisocial characteristic and/or poverty) with place-based characteristics (such
as homelessness, criminal activity in the environment which may offer the person opportunities to commit crime and/or the available social services). Homelessness is identified in this report as an environmental factor which may be a driver for increased interaction between police and persons with serious mental illness.

Increasing Demand for Police Contact

Increase in the Number of Suicides12

Over the last decade, there has been an increase in the number of suicides in Multnomah County. Figure 4 displays the number of suicidal deaths in Multnomah County during this time frame:

Figure 4

Figure 4Completed suicides hit a low in 2001 and have since increased. It is important to note that 2001 saw a number of events (financial crisis, the closure of the CTC) which may have led to this increase. Alternately, the suicide rate (number of suicides controlling for population) is decreasing (Sorvari & Mowlds, 2011). This drop in rate is a testament to the work of individuals in the mental health system, but the fact remains that police are increasingly called to a greater number of suicides and calls involving suicidal persons.

It is important to note that Multnomah County has experienced significant population growth during this period.13  This can lead to increased number of completed suicides at an aggregate (total number of suicides) level while the rate (suicides controlling for increased population) drops. Suicide rates have also dropped since 1996 (Multnomah County Health Department and Public Health, 2009) (See Appendix A). However, when considering the potential number of suicidal persons contacted by police it is appropriate to use aggregate numbers. Similarly, there are fewer police officers now in the City of Portland than in 2001, so the number of suicides (or suicidal persons) any individual officer is likely to encounter has grown. This may also explain the general perception among police that this problem is growing over time. The increase in completed suicides may correspond to other issues related to increased contact between police and persons in a mental health crisis. This trend also mirrors a national trend, with the aggregate number of suicides growing nationally.

Increase of Suicide Calls to BOEC

Since 2001, calls to BOEC regarding suicides from within the City of Portland have grown by more than 90%. In 2011, BOEC received 1,200 calls involving suicidal subjects. These calls involved situations where someone completed a suicide, attempted suicide or threatened to complete a suicide. Some of this increase may be due to change in how mental health calls are coded by BOEC (see Appendix A, Decline in Mental Health Calls), however the increases call volume cannot account for the entire change. It does demonstrate that police officers are now responding to nearly twice as many of these calls compared to a decade ago.

Figure 5 shows the steady increase in call volume over time.14

Figure 5Preliminary numbers as of Feb. 7, 2012

It is important to note that the increase in attempted or threatened suicide calls is substantially greater than the increase in completed suicides. This may mean that additional factors, beyond suicidality are influencing the number of these calls.

This data is consistent with a 2009 Multnomah County Health Department Assessment (Multnomah County Health Department and Public Health, 2009). This report found that hospitalized suicide attempts were increasing for both men and women (although the increase was not statistically significant for men) between 2000 and 2006 (See Appendix B)

Calls Resulting in an Attempted Suicide Code

It is important to note that most calls involving persons in a mental health crisis (even if dispatched as an attempted or threatened suicide) do not result in an actual suicide or attempted suicide. In fact, officers and mental health professionals are normally able to resolve calls involving persons in mental health crisis without the situation resulting in a suicide or attempted suicide. Despite the positive outcomes, there has still been a rise (although less precipitous) in attempted suicides documented by the Portland Police.

Figure 6

*Attempted suicide documented in the Portland Police Data System

The number of attempted suicides has increased, but not to the same degree as suicide calls. This is consistent with the number of suicides in Multnomah County and it is possible that alternate services provided by Multnomah County are helping to avert actual suicide and suicide attempts by persons in mental health crisis. It may also indicate that the rise in attempted or threatened suicides is not due entirely to persons intent on completing a suicide. However, the fact that all of these indicators triangulate in a positive direction is strong evidence of increased demand for police to respond to assist persons in mental health crisis.

Increase in Welfare Check Calls to BOEC

Welfare check calls cover a broad range of activity. These calls include everything from checking on a stranded motorist to responding to persons in crisis (physical or mental health). Therefore, it is difficult to use this measure as a reliable indicator of issues related to mental health in the absence of supporting evidence that changes to how calls
are dispatched may be impacting the relative percentage of these calls involving mental health. Interviews with BOEC indicate that this may be the case. It appears the calls involving mental health crisis may be increasingly coded as welfare checks. However, the evidence for this is anecdotal in nature. Figure 7 displays changes in welfare check calls over time:

Figure 7Clearly, the number of these calls has increased substantially over time. There were 2,272 more welfare check calls in 2011 than 2010. Additionally, we cannot directly attribute this change increased contact between police and persons in a mental health crisis. Despite this caveat, this increase is consistent with our other indicators examined in this report and there is support for its inclusion as a measure.

Involuntary Commitments are Increasing in Multnomah County

In Oregon, a police officer, a medical doctor, or mental health professionals can place temporary holds on people who are a danger to themselves or others. Prior to a civil commitment hearing, holds are investigated by Involuntary Commitment Investigators. Examining the trends in the amount of involuntary commitments can be an indicator for
whether police are coming in contact with persons with mental illness more or less often. In 2004, the Department of Community Human Services (DCHS) Office of Research and Evaluation released a report, titled, “Canary in the Mineshaft: Tracking Changes in Involuntary Commitment Investigator Workload.” (Multnomah County, 2004) This report details changes in mental health services in the county. From the executive summary:

The 2004 DCHS report highlights the impact of the closure of Dammasch State Hospital in 1995 and the Crisis Triage Center in 2001. It also highlights budgetary issues, such as changes to the Oregon Health Plan. Multnomah County holds investigated by the Involuntary Commitment Program have increased from 2,823 in 1994 to 4,226 in 2011.

Figure 8

The Multnomah County CTC was closed in 2001 as a result of the Multnomah County mental health system redesign. While unintended, the closure resulted in more individuals being evaluated in emergency departments than in urgent walk-in with a corresponding increase in mental health holds placed by emergency department physicians.

The rate of civil commitments to ICP investigations has declined as well. In fiscal year 2005, there were 304 civil commitments to 3,871 ICP investigations (around 8% of ICP investigations resulted in civil commitment). By 2011 this had fallen to 254 civil commitments for 4,226 ICP investigations (or roughly 6% of ICP investigations resulting in a civil commitment). Interviews with ICP investigators reveal that a changing legal environment may be impacting the civil commitment process.15

This change (from 2,488 ICP investigations in 1998 to 4,226 investigations in 2011) may be attributable to system infrastructure changes and opposed to changes in the number of persons in mental health crisis. Interviews with individuals associated with the Crisis Triage Center reveal that the center may have diverted a large number out of the ICP process.

The increase in involuntary commitment investigations is a statewide pattern and has been occurring over the last 20 years.16  From 1983 to 2003, statewide investigations for involuntary commitments grew from 3,996 to 8,315, while the number of actual commitments fell from 1165 to 785. When accounting for increased population, the civil commitment rate per 100,000 citizens fell from 45 per 100k in 1983 to 22 per 100k in 2003 (Bloom, 2006). This drop is, at least partially, the result of increased concern around supporting the civil rights of persons with mental illness to be treated in the least restrictive setting possible. However, decreased use of commitment may result in increased contact between police and persons in a mental health crisis.


Repeated Police Contact

Repeat contact may increase the potential for an individual to have force used against them. Portland Police officers frequently have to place multiple holds on the same person. Multnomah County commitment services data spanning 2005 through 2011 shows that 23% of the individuals placed on holds annually have 2 or more holds in a calendar year.

To demonstrate this effect over time, the PPB Crime Analysis Unit tracked all individuals listed in the Portland Police Data System as having been placed on a hold in 200917 through to February 14, 2012. Figure 9 displays the number of repeated police officer assisted holds placed on individuals in this cohort.18

Figure 9Most of the individuals from 2009 required only one police officer-assisted hold. However, more than 26% (219 people) in this group required two or more police officer assisted holds between 2009 and 2012. These 219 people required 617 police officer assisted holds (more than half of all police officer assisted holds placed on this group). Figure 10 displays this:

Figure 10

Based on this data it would appear that more than half the police officer-assisted holds are repeat holds (at least for the time period examined). It may be possible that individuals contacted by police are more likely to require repeat police officer-assisted holds, or alternately, that system changes since 2004 have driven up the volume of repeat police officer-assisted holds.

The reduction in civil commitments may benefit the mental health system by diverting people out of the few remaining state hospital beds. It also supports the rights of persons diagnosed with a mental illness to be treated in the least restrictive setting as required by the Olmstead Act and the Civil Rights of Institutionalized Persons Act. However, the lack of funding for an adequate array of community-based services most likely contributes to the increase in contacts between police and people in mental health crisis.

As noted in the 2004 Multnomah County DCHS report, there is a precarious balance between protecting individuals’ physical well-being and protecting their civil liberties. It is not our intent to comment on where this line should be drawn. However, these policies do impact the likelihood of police officers responding to a person in a mental health crisis. This increased contact may be driving use of force in these situations.

Use of Force During Holds and Custodial Contacts19

Between 2009 and 2012, force was used in less than 5% of all custodial contacts with individuals who required a police officer-assisted hold in 2009. Repeat contact may be negatively impacting the potential for an individual to have force used against them.

There is almost no difference in the percentage of force used in custodial contacts requiring force for all persons with a police officer assisted holds (in 2009) and persons arrested in 2009 (this includes arrests and police officer assisted holds). Figure 11 displays the percentage of custodial contacts between 2009 and February 14, 2012 which resulted in force:20

Figure 11

On an individual level it is more likely that force will be used against individuals with 3 or more police officer-assisted holds. Figure 12 displays this:

Figure 12This may seem counterintuitive as there is less chance in any given contact of police using force against an individual with 3 or more police officer assisted holds during the study period (Figure 9) yet those same individuals are more likely to eventually have force used against them overall. This appears to be due to having more repeated contact with the police in general. Figure 13 shows the average number of contacts for each group:

Figure 13

To clarify, officers used force less frequently when contacting individuals with three or more holds, but were much more likely to contact them multiple times. Individuals held 3 or more times had an exceptionally high rate of contact, 9.3 police officer-assisted holds or arrests compared with 2.5 police officer-assisted holds or arrests for individuals arrested in 2009.21

As mentioned above, the majority of arrests were for lower level offenses. This would lead to an expectation of lower levels of force. Individuals who have been put on a hold are by definition in a state where they present “a danger to themselves or others.” While the methodology and definitions are not completely analogous, Skeem and Bibeau (Skeem & Bibeau, 2008) conducted a study of police reports for Crisis Intervention Team officers in Memphis, Tennessee, between March 2003 and May 2005. In reviewing the 655 reports, they found that 45% involved a suicidal crisis, 26% involved a threat to others. This report found that officers used forced in 15% of the 189 events “posing serious or extreme risk of violence” and 6% of the total events. It is important to note that their definition of force included handcuffing, which is different from the Bureau’s definition. However, handcuffing accounted for only 4 of the 36 incidents in which force was used. We cannot directly compare the 6% of force used in these incident to the Bureau’s average as the definitions are not consistent!  However, this study provides context to the discussion surrounding use of force and persons in a mental health crisis. These incidents have the potential to escalate to a point where force is necessary.

What do we know about person with a mental illness and use of force by police?

General studies of police use of force (not specific to persons with a mental illness) find that the degree of hostility and aggressive or threatening behavior on the part of the suspect is the strongest predictor of police use of force (Hickman, Piquero, & Garner, 2008) and that force is generally commensurate with resistance (Gallo, Collyer, & Gallagher, 2008). Macro level studies have found a significant relationship between predatory crime and police use of deadly force (MacDonald, Kaminski, Alpert, & Tennenbaum, 2001). This is consistent with other studies examining specifically the use of force by police against persons with mental illness (Kaminski, DiGiovanni, & Downs, 2004; Johnson, 2011). These studies generally find that the impact of mental illness on these encounters is minimal and that other factors are more influential in determining when and how force is used.

Studies examining Crisis Intervention Teams and the use of force have had inconsistent findings. In a study using training scenarios to gauge officers force preferences in containing a schizophrenic subject found that officer opted for less force over the course of the encounter and perceived force as being a less-effective option (Compton, et al., 2011). This would support the utility for CIT-type programs in reducing police use of force against persons with mental illness; however, other studies have had contradictory findings. A study examing the effects of the implementation of a CIT program and SWAT callouts, (Compton, Berivan, Oliva, & Boyce, 2009), found that the introducton of CIT-trained officers had no significant correlation with the number of SWAT callouts in Atlanta, Georgia.

Another study examined how CIT officers in a given police district impacted use of force after controlling for a broad range of variables (Moribito, et al., 2012). Matching two pilot districts with a high saturation of CIT officers against two districts with a low saturation of CIT officers the study found used regression analysis to analyze how officers in the high saturation CIT districts used force compared with the low saturation districts controlling for factors such as crime and available resources. This study used an officer’s self-reported interaction with the last person with mental illness he/she contacted to determine the nature of the officer’s force and the person with mental illness’s resistance. The study found that an interaction between CIT and suspect demeanor as well as district characteristics and subject resistance impacted the likelihood of the use of force (which for this study included verbal commands). There was evidence that the factors influencing the use of force against persons with mental illness resembled those generally associated with use of force. This analysis is consistent with the general findings regarding police use of force against individuals with a mental illness.

Related studies on crisis negotiation have also introduced questions on the effectiveness of communication with individuals intent on “suicide by cop.” Mohandie & Meloy, 2010, (Mohandie & Meloy, 2010) conducted a study of 84 hostage, barricaded or “jumper” incidents which resulted in an officer-involved shooting. They found that, while a primary assumption of most negotiation strategies is that additional time will deescalate the stituation, this may not be the case in situations where the individual does not have the will to live. In these situations the subject may regard the negotiator instrumentally (as a possible tool for suicide). In these situations, the passage of time and additional communication may not aid in building the kind of rapport necessary to defuse the situation peacefully. This is important in regard to persons with mental illness as nearly half of the sample in this study (n=84) had probable or definate mental health history. It is also important to note that the authors recognize that there may be substantial limitations to the generalizablity of this study.

A review of the scientific literature surrounding police use of force supports the hypothesis that force is generally commensurate to resistance, that a suspect’s mental impairment may not independently impact police use of force and that police use force to overcome hostility, aggression and percieved threat. Studies of CIT yield mixed results on their impact in reducing use of force or crisis situations requiring deployment of a SWAT unit and studies of crisis negotiation find that there may be situations in which a person is in a state which renders effective crisis communication unfruitful.

If the literature is correct it would support the necessity of system-level support in reducing the number of use of force incidents involving police and persons in a mental health crisis. This concept is supported by our general understanding regarding the necessity of prevention in reducing harmful outcomes. The potential importance of system level influences on police use of force is consistent with other findings related to improving general (not force specific) outcomes of interactions involving persons in a mental health crisis. For instance, resources such as a police friendly single point of intake may be highly infuential in the process, such as improving mental health diversion from jail (Steadman, Stainbrook, Griffin, Draine, Dupont, & Horey, A Specialized Crisis Response Site as a Core Element of Police-Based Diversion Programs, 2001) .

Deadly Force

There has been significant public interest surrounding the use of deadly force by Portland Police officers. It is important that the public be well-informed regarding the use of any force by the police. Between 1998 and 2011, the use of deadly force (officer-involved shootings or OIS) by Portland Police has trended downward. It is important to note that the majority of these incidents do not appear to involve persons with severe and persistent mental illness (SPMI). A study of 8 shootings occurring between January 1, 2010, and January 2, 2011, was only able to identity 2 of the 8 subjects in the shootings as suffering from a SPMI, while at least one other individual appeared to be suffering from depression. A history indicative of substance abuse issues appeared in nearly all the incidents (Stewart & King, 2011). However, the lack of available information regarding mental health status limits the confidence in those findings. Figure 14 illustrates the decrease in officer-involved shootings since 1998:

Figure 14

*This chart includes all incidents in which a PPB officer discharged their firearm at another person. In some of these incidents, no one was injured.

While the actual number of shootings in Portland may have decreased, it is important to review the rates across different police departments to see if Portland has an abnormally high number of police shootings. Obtaining this data is difficult and current data from other jurisdictions was not available at the time this report was written. However, the San Francisco Police Department has prepared an excellent report on officer-involved shootings (Various, 2010). As previously mentioned, this study was conducted by the San Francisco Police Department and was done independent of this report. The following graph (Figure 15) is taken directly from this report:

Figure 15

Figure 15

As indicated in Figure 15, the Portland Police Bureau has fewer shootings per one million in population than most of the other agencies examined. The chart compares the dates of January 1, 2005 to August 27, 2009.22

Overall Decline in the Use of any type of Force

The use of any type of force by Portland Police officers has also fallen in recent years. Between 2008 and 2011, there was a substantial drop in police use of force both as a percent of contacts and as a percent of arrests.  Figure 16 displays this trend over time:

Figure 16

Figure 16Portland Police use force in less than 0.3% of all contacts and less than 4% of all arrests.

The Police Bureau includes incidents where officers point a firearm at a person during an arrest as a use of force. For example, if officers respond to a burglary and point a firearm during the arrest this is counted as force. Actual use of physical force (use of Taser or physically taking a person to the ground etc.) occurs in about 2% of all arrests. Figure 17 displays the declining trend in use of force.

Figure 17

Figure 17


Numbers for 2011 are preliminary.

Please note the drop in nearly all types of force usage since 2008.

Police-Assisted Holds

Police-assisted holds related to mental health do not follow the same pattern as suicide calls. As figure 18 displays from 1998 to 2011, the total number of police officer-assisted holds remains constant (and drops relative to population):

Figure 18

Figure 18

Police officer-assisted holds peaked in 2000. These holds dropped slightly in 2001 before falling precipitously in 2002.23 This coincided with the closure of the Crisis Triage Center in mid-2001. The closure of the Crisis Triage Center (CTC), a police friendly drop-off site, is cited repeatedly in interviews as impacting police services to persons in a mental health crisis.

One officer, who worked in an area with a large number of individuals with mental illness, said that after the closure of CTC, officers went to great lengths to avoid utilizing holds. Officers cited the increased time to process individuals in an emergency room environment, the sense that emergency rooms did not want to treat individuals with a mental illness and expressed concerns around individuals with both mental health and substance abuse issues. A significant (if not majority) of individuals contacted in crisis by police are under the influence of an intoxicating substance when experiencing a mental health crisis. There currently is a lack of resources for individuals with a SPMI experiencing a behavior crisis with co-occurring substance abuse issues. Follow-up interviews with mental health professionals and officers reveal that most emergency rooms inform first responders that use of emergency departments for intoxicated individuals regardless of mental health status is not appropriate. While the emergency rooms may take individuals whose substance use has impaired their physical health, officers report a reluctance on the part of emergency rooms to accept individuals who are experiencing a mental health crisis while intoxicated. Officers also report the dual occurrence of intoxication and mental health crisis as being very common. Due to a lack of alternatives, individuals who are intoxicated and in a mental health crisis may be taken to jail or a detox center in place of receiving treatment.

Officers may also have utilized arrests in lieu of placing holds for mental health. This is not an issue isolated to the Bureau. Steadman, Stainbrook, Griffin, Draine, Dupont, & Horey, 2001, found in a study which specifically mentions the Crisis Triage Center, that “It appears that 24-hour specialized crisis response sites with no-refusal policies, appropriate legal foundations, and real linkages to community-based services are a key element in successful pre-booking jail diversion programs for individuals with serious mental illness and substance abuse problems.”

The closure of the CTC did not occur in a vacuum and other services, including expanded community-based 24-hour-a-day programs and crisis services, became available. It is possible that police, assisted by increases in community-based services such as Project Respond are doing a better job of limiting the use of police officer-assisted holds. Alternately, it may be that utilizing emergency rooms is more difficult, time-consuming and viewed as less effective, which discourages officers from using holds, with arrest or detox (for intoxicated individuals) becoming more common. It is likely a combination of these factors which has resulted in the flat number of police officer-assisted holds.


A number of studies have found that persons with severe mental illness are victimized at a much higher rate than the community at large (Silver, 2002; Salyers, et al., 2001; Teplin, McClelland, Abram, & Weiner, 2005). Given the difficulty involved with the exchange of information, there is currently a lack of understanding locally on the scope of this issue (from the prespective of the Portland Police Bureau). However, there is no reasons to suspect Multnomah County in general or the City of Portland in particular differs from the nation. Given the apparent increase in contacts between officers and persons in a mental health crisis, it is reasonable to surmise that the number of individual being victimized may be increasing.

Emergency Services

The Portland Police Bureau has made significant strides over the last several years in developing the skills of officers and improving resources to assist with mental health calls. The Bureau hired Dr. Liesbeth Gerritsen, a mental health professional, and assigned her to assist with training all officers, and to act as a liaison between the Bureau and mental health partners. Dr. Gerritsen works closely with the Police Bureau’s Crisis Negotiation Team and with persons who have mental health issues. She also assists families impacted by a mental health crisis. Dr. Gerritsen works with the Bureau’s Neighborhood Response Team officers and jail personnel to help problem solve issues surrounding people with mental illness who have frequent contact with police and Cascadia’s Project Respond.

Additionally, over the past few years, the Bureau has partnered on important projects such as Safer PDX (in conjunction with the Bazelon Center for Mental Health Law) and SAMHSA’s GAINS Center (hosted by Multnomah County), and Multnomah County’s Local Public Safety Coordinating Council. These groups and others are intensively reviewing and researching the interface between the mental health system and the criminal justice system.

One of the most widely debated changes to our practices is the Bureau’s decision in 2008 to train all Operations Branch police officers and sergeants in Crisis Intervention Training (CIT). In the past the Bureau trained only a small group of volunteer officers, and the protocol was for a primary officer to be dispatched to the scene and call for a CIT officer if the situation required it. But often, the CIT-trained officers on duty would be in another precinct or on another call and unable to respond immediately.

Portland has joined an ongoing national debate among mental health professionals, academics, police agencies, and people and families impacted by mental illnesses about what “best practice” is for CIT. There is no data that clearly supports using one model over the other; the CIT International Board is split as to which approach is best. There is general agreement, however, that cities should tailor their police response to people in mental health crises according to the unique and specific characteristics of each community. Our partners in the mental health services support having all of our patrol officers trained in CIT. We believe a core competency as a police officer is the ability to effectively communicate with a person in a mental health crisis.

The Bureau has made other changes, including the creation of a Mobile Crisis Unit, which pairs a mental health professional with a police officer. This unit responds to people with mental health issues who have frequent contact with the police or are at high risk to become a danger to the community. Additionally, the Portland Police Bureau is working in partnership with Multnomah County to have 9-1-1 mental health calls diverted to the Multnomah County crisis line when appropriate.

The City of Portland has also engaged in the provision of social services related to mental health, homelessness and addiction. Other city programs include funding from the Housing Bureau for mental health and other services targeted toward individuals with severe mental illness, recreational programs through the Parks Bureau for people with physical and/or mental health-related challenges, approximately $1 million annually to help fund services related to Hooper Detoxification Center and nearly $2 million annually to fund the Service Coordination Team. A list of these services can be found in Appendix C.

Despite the continued attempts at improving police response to people in crisis, several system-wide gaps remain. With the closure of the Crisis Triage Center in 2001, Multnomah County no longer has a 24/7 psychiatric triage facility with a “no refusal” policy for police. Police officers take people who are a “danger to self or others” to hospital emergency departments. Hospital emergency rooms, with their bright lights and loud stimuli, are not optimal for people in a psychiatric crisis. In addition, people in crisis sometimes have to sit in the back of a police car for long periods of time while waiting for space to open in a hospital emergency department.

While the county has opened the Multnomah County Crisis Assessment and Treatment Center (CATC), it does not provide the same service as the former Crisis Triage Center. The CATC does not have capacity to accept clients who are combative or assaultive. In addition, the police cannot directly transport persons to the CATC regardless of acuity level. Multnomah County’s Crisis Assessment and Treatment Center (CATC) provides additional capacity to accept insured and uninsured individuals in a mental health crisis who would otherwise have been to a hospital. More than 300 admissions to the CATC have occurred since opening in June of 2011. Half of the admissions to this program are to uninsured individuals in our community. Unfortunately officers report that this program, while filling a valuable niche, does not meet their needs.


There are a number of factors contributing to the continued increase in interactions between public safety personnel and persons in mental health crisis. While some of these are contradictory (the flat number of police officer-assisted holds and the decline in mental health calls), overall the evidence supports the hypothesis that contact between police and persons in a mental health crisis have grown.

A number of factors may be influencing this increased contact. These factors include the closure of Dammasch State Hospital, the closure of the Crisis Triage Center, deinstitutionalization of people with mental illness coupled with a lack of adequate community care resources, and a lack of housing for persons suffering severe and persistent mental illness.

To meet this crisis, the Police Bureau, BOEC and other public safety partners will need to continue to innovate and explore additional options in how they assist persons in mental health crisis. To date the Police Bureau has provided every single officer through Crisis Intervention training with the core competency necessary to help people in mental health crisis, instituted a mobile crisis unit consisting of an officer and a mental health professional, developed excellent relationship with mental health partners, and instituted a more robust review of force. These changes will help the Police Bureau effectively respond to this challenge and better serve the community.

Next Steps

This report highlights several issues surrounding police interactions with persons with mental health crisis. Based on an analysis of the data and interviews with community partners the Bureau may benefit from additional research into the following areas:

  • The creation of a police-friendly (no refusal) facility to a quick-release, priorityaccess facility, that will accept persons with co-occurring mental health crisis and substance abuse.
    • This is recommendation is consistent with a 2010 Multnomah County report titled, “Sequential Intercept Mapping and Taking Action for Change” commonly known as “The GAINS” report (Policy Research Associates, 2010).
    • It is also consistent with recommendations in, “Improving Responses to People with Mental Illness, Tailoring Law Enforcement Initiatives to Individual Jurisdictions,” (Reuland, Draper, & Norton, 2010).
  • Assess where we are with the implementation of recommendations associated with the “Sequential Intercept Mapping and Taking Action for Change.” The institution of a system-wide review of outcomes related to all mental health holds, with an emphasis on holds which do not qualify for a notice of mental illness resulting in an involuntary hospitalization or admission to the CATC.
  • This should be re-occurring (possibly quarterly) and outcomes should be shared with police officers to help improve their responses to these issues.
  • Continued exploration of BOEC coordinating with the mental health system (this is currently occurring but should be supported).
  • Renewed promotion of the dedicated police line in the Multnomah County 24 hour Mental Health Call Center available to officers wanting immediate access to mental health information, if available, during the course of an encounter.
  • The police and mental health community should explore how the approximately 3,200 holds which police are not involved with, intersect with the criminal justice system. Are the police contacting the individuals, but are unaware of their status in the mental health system and what can be done to limit these contacts?
    • This would not require the exchange of individual level data but could be done at the aggregate level. The criminal justice system utilizes DSS-J for similar strategic level tracking and this system may serve as a model for the aggregation of this type of data.
  • The Bureau should continue to explore improved responses to persons in mental health crisis. Current issues include:
    • Improved tracking (currently being implemented)
    • Examining the co-occurrence of substance abuse with mental illness and/or behavior crisis.
    • Implementing the Mobile Crisis Unit(s) in the most effective way possible.
    • Additional analysis on how to more effectively address the victimization of persons with a mental illness.


¹ See PPB Directive 1010.10 – Deadly Force Sanctity of Human Life

² Sergeant Greg Stewart of the Crime Analysis Unit (CAU) and Dr. Liesbeth Gerritsen, coordinator of the Crisis Intervention Training Program conducted a review of suicide calls and police interactions with persons in mental health crisis. It quickly became apparent that the scope of the review would need to be expanded. This led to the creation of this report.

³ Provided by PPB Statistical Support Unit’s record of suicide calls. Data from prior to 2001 not included because of odd variability in the number of calls may indicate a change in data definitions or some other variable between 1999 and 2000. It is important to note that this data is subject to changing definitions. Therefore, it is important to triangulate data sources and not rely overly on one individual piece of data. This report examines a number of different factors which may indicate increased contact between police and person in a mental health crisis.

4 See Figure 11. Force use before 2008 is not included due to changes in how force was recorded making comparisons difficult.

5 Note: different time frames are used depending on the availability and reliability of data. For instance, approximately 150 suicide calls were received in 1999 and this jumped to more than 500 in 2000. This indicates a change in how calls were typed. Because this change appears related to how calls are coded, data is only used from 2001 forward. Other data prior to 2001 was used (if it was reliable), especially as it relates to the Crisis Triage Center, which was closed in July 2001. Police data is of mixed reliability as legal changes, public opinion regarding appropriate language and new challenges faced by the police may alter how data is coded.

6 Obtained via the PPB Statistical Support Unit and queries of the Bureau’s SQL server.


8 The Bureau would like to thank Bill Osborne, Joan Rice and Rachel Philofsky of Multnomah County, Dr Mark Kaplan, Dr. Kris Henning and Dr. Brian Renauer of Portland State University, Dr. Bill Nunely of Cascadia Behavioral Health Care, Care Oregon and the Bazelon Center, Dr. Derald Walker and Dr. Maggie Bennington-Davis of Cascadia Behavioral Health Care as well as many others who provided essential input for this document. Any mistakes or omissions are the fault of the author.

9 Data provided by Multnomah County

10 Population data for Multnomah County 2006 came from the Portland State University Population Research Center. Suicide figures for Multnomah County were found at—Oregon Health Authority Vital Statistics.

11 Homeless individuals may be at an increased risk of police contact generally. Homelessness poses criminogenic risk irrespective of mental health status and this risk may be interacting with mental health issues to drive this increased contact. They may also be at increased risk of using emergency rooms and/or being incarcerated. All of these options can prove expensive. Additional study in this area may prove fruitful.

12 Note: Suicides, Suicide calls and attempted suicides are tracked from 2001 due to reliability issues for suicide call data prior to 2001. Appendix A has suicide data from 1993 to 2009 (the entire time range available at the Oregon Vital Statistics website).

13 It is also important to note that the aggregate number of completed suicides in 2009 is comparable to figures in the late 1990s (111 completed suicides in 2009 and 104 completed suicides in 1999) and the rate was actually higher in 1999.

14 Numbers provided by the Portland Police Bureau Statistical Support Unit

15 Additional research into this phenomena could prove enlightening. It may also be beneficial to track outcomes for both committed and non-committed individuals. There is currently a dearth of readily available (at least to public safety workers) information regarding these outcomes. While privacy requirements may prevent the collection of individual level data there could still be aggregate level analysis similar to Multnomah County’s Decisions Support System Justice (DSS-J). There may be a lack of follow-up on cases which do not result in a commitment but there is currently little publicly available data on these outcomes.

16 Note: an involuntary mental health hold investigation is distinct from an involuntary commitment hearing. An involuntary commitment hearing is a judicial process in which the person placed on a hold goes before a judge in a civil hearing.

17 This group includes only individuals who had one or more holds for a mental health crisis in which the PPB either places the hold or assisted a community partner in placing a hold.

18 This does include all the holds placed during that time period. It only tracks individuals held in 2009.

19 Footnote 1920 This chart compares individuals with a hold in 2009 as a cohort against all individuals arrested in 2009. Their activity was tracked forward through to February 14, 2012 1900 hours (the most up to date information at the time this data was gathered). It does not include all activity but treats each group (i.e. 2009 holds and 2009 arrests). Arrests were counted using individual case numbers with involvement code AR (arrested) or CH (charged). Holds were counted using individual caes numbers with involvement ME (mentally ill, cared for). Force was determined by counting individual case numbers associated with CRN (individual identifiers).

21 It is important to remember that individuals held 3 times had at least 3 contacts by definition (the holds).

22 It is important to recognize that the 6 shootings in 2010 and 4 shootings in 2011 could change Portland’s relative position. Portland’s number of shootings is not inconsistent with other northwest locations. Seattle PD had 5 OIS in both 20 Spokane County, WA, experienced 7 OIS from August 2010 to January 2011.

23 Hold definitions may not have remained constant over time. There is no record of changes if they occurred. These definitions, if changed, would have resulted in a more confining definition (moving from simply assisting persons with mental illness to placing a hold). In practical terms, this means that these numbers may under-represent demand changes over time.


Bloom, J. D. (2006). Civil Commitment Is Disappearing in Oregon. The Journal of the American Academy of Psychiatry and the Law , 534-537. Compton, M. T., Berivan, D., Oliva, J. R., & Boyce, T. (2009). Crisis Intervention Team Training and Special Weapons and Tactics Callouts in an Urban Police Department. PSYCHIATRIC SERVICES, 831-833.

Compton, M. T., Demir Neubert, B., Broussard, B., McGriff, J. A., Morgan, R., & Oliva, J. R. (2011). Use of Force Preferences and Perceived Effectiveness of Actions Among Crisis Intervention Team (CIT) Police Officers and Non-CIT Officers in an Escalating Psychiatric Crisis Involving a Subject With Schizophrenia. Schizophrenia Bulletin, 37(4):737-745.

Elbogen, E. B., & Johnson, S. C. (2009). The Intricate Link Between Violence and Mental Disorder. Arch Gen Psychiatry, 152-161.

Epperson, M., Wolff, N., Morgan, R., Fisher, W., Frueh, B. C., & Huening, J. (2011). The Next Generation of Behavior Health and Criminal Justice Intervestions:Improving Outcomes by Improving Interventions. National Institute of Mental Health. Gallo, F., Collyer, C. E., & Gallagher, P. L. (2008).Prevalence of Force by Police in Rhode Island Jurisdictions. Criminal Justice Review , 33(4):480-501.

Hickman, M. J., Piquero, A. R., & Garner, J. H. (2008). Toward a National Estimate of Police Use of Force. Criminology and Public Policy, 7(4):564-604. Homelessness Research Institute. (2012). The State of Homelessness in America 2012. Washington D.C.: National Alliance to End Homelessness.

Johnson, R. R. (2011). Suspect Mental Disorder and Police Use of Force. Criminal Justice and Behavior, 38(2):127-145.

Kaminski, R. J., DiGiovanni, C., & Downs, R. (2004). The Use of Force between the Police and Persons with Impared Judgement. Police Quarterly , 7(3):311-338.

Kristina Smock Consulting. (2011). 2011 Point-in-Time. Portland: City of Portland and Multnomah County.

MacDonald, J. M., Kaminski, R. J., Alpert, G. P., & Tennenbaum, A. M. (2001). The Temporal Relationship Between Police Killing of Civilians and Criminal Homicide: A Refined Version of the Danger Perception Theory. Crime & Delinquency, 47(2):155-172.

McConnell, B., Dunn, K., & Brooke, H. (2010). San Francisco Police Department Workload Estimates Related to Providing Services to the Seriously Mentally Ill Population. San Francisco: San Francico Police Department and The San Francisco Mental Health Board.

Mohandie, K., & Meloy, J. R. (2010). Hostage and Barricade Incidents within an Officer-Involved-Shooting Sample: Suicide by Cop, Intervention Efficacy, and Descriptive Characteristics. Journal of Police Crisis Negotiations, 10(1):101-115.

Moribito, M. S., Kerr, A. N., Watson, A., Draine, J., Ottati, V., & Angel, B. (2012). Crisis Intervention Teams and People With Mental Illness : Exploring the Factors That Influence the Use of Force. Crime & Delinquency , 57-77.

Multnomah County. (2004). Canary in the Mineshaft: Tracking Changes in Involuntary Commitment Investigator Workload. Portland: Multnomah County.

Multnomah County DCHS Office of Research and Evaluation. (2004). Canary in the Mineshaft: Tracking Changes in Involuntary Commitment Investigator Workload. Portland: Multnomah County.

Multnomah County Health Department and Public Health. (2009). Community Health Assessment Quarterly (Volume 4, Issue 2). Multnomah County: Multnomah County.

National Coalition for the Homeless. (2009). Mental Illness and Homelessness. Washington, D.C.: National Coalition for the Homeless.

Policy Research Associates. (2010). Sequential Intercept Mapping and Taking Action for Change. Multnomah County: Multnomah County.

Reuland, M., Draper, L., & Norton, B. (2010). Improving Responses to People with Mental Illness. New York: Bureau of Justice Administration and Council of State Governments Justice Center.

Salyers, M. P., Goodman, L. A., Mueser, K. T., Rosenberg, S. D., Swartz, M., Essock, S. M., et al. (2001). Recent Victimization in Women and Men With Severe Mental Illness: Prevalence and Correlates. Journal of Traumatic Stress, 14(4):615-632.

Silver, E. (2002). Mental disorder and violent victimization: the mediating role of involvement in conflicted social relationships. Criminology, 40:191-211.

Skeem, J., & Bibeau, L. (2008). How Does Violence Potential Relate to Crisis Intervention Team Responses to Emergencies. Psychiatric Services , 59(2):201-204.

Sorvari, C., & Mowlds, E. (2011). Multnomah County Community Health Assessment Mobilizing for Action through Planning and Partnerships to Identify Health-Related Priorities. Multnomah County: Multnomah County Health Department.

Steadman, H. J., Stainbrook, K. A., Griffin, P., Draine, J., Dupont, R., & Horey, C. (2001). A Specialized Crisis Response Site as a Core Element of Police-Based Diversion Programs. Psychiatric Services, 52(2):219-222.

Steadman, H. J., Stainbrook, K. A., Griffin, P., Draine, J., Dupont, R., & Horey, C. (2001). A Specialized Crisis Response Site as a Core Element of Police-Based Diversion Programs. Psychiatric Services, 52(2):219-222.

Stewart, G. J., & King, R. (2011, March 8th). Officer Involved Shootings. Presentation to Downtown Public Safety Action Committee . Portland, Oregon, United States. Stone, C. (2010, Sept 9). Oregon Health Authority News Release. Retrieved February 21, 2012, from

Teplin, L. A., McClelland, G. M., Abram, K. M., & Weiner, D. A. (2005). Crime Victimization in Adults With Severe Mental Illness Comparison with the National Crime Victimization Survey. Arch Gen. Psychiatry, 62(8):911.

Torrey, E. F., Entsminger, K., Geller, J., Stanely, J., & Jaffe, D. J. The Shortage of Public Hospital Beds for Mentally Ill Persons. Arlington: Treatment Advocacy Center.

U.S. Department of Housing and Urban Development. (2010). The 2009 Annual Homeless Assessment Report To Congress. U.S. Department ofHousing and Urban Development.

Various, S. F. (2010, January 20). Officer-Involved Shootings A Five-Year Study.  Retrieved February 21st, 2012, from San Francisco Police Department:

Appendix A - Suicides in Multnomah County

Appendix A - Suicide Death Rates by Sex


Appendix B - Suicide Hospitalization Rates by Sex

While the inclusion of this data may seem redundant it is important to note that most of the available data (especially to public safety due to privacy concerns) is unreliable. This increases the importance of triangulating data sources to improve the overall reliability of this report.

Appendix C
Decline in Mental Health Calls Provides an Explanation for Only a Small Portion of the Increase in Suicide and Welfare Check Calls

BOEC previously used a code for calls involving a mental health issue but not necessarily suicidal behavior. The following figure displays changes in this call type over time:

Appendix C - Mental Health CallsThis chart shows a precipitous drop in call volume and is in stark contrast to the increase in suicide calls. Interviews with BOEC shed light on this discrepancy. BOEC has increasingly discouraged the use of this call type, instead focusing on the behavior in question. It appears that this call type was removed entirely with the introduction of the new dispatching system (hence the low number of calls in 2011). This change may be driving part of the increase in suicidal calls (as some percentage of these calls may have become suicidal type calls), but the volume is not sufficient to account for the entire increase in attempted suicide calls. Furthermore, interviews with BOEC reveal that the bulk of these may have become welfare checks type calls. Consistent with this interpretation, welfare check calls have grown substantially over the last decade.

Appendix C - City Services Related to MH