Comparing Safety Outcomes In Police Use-of-force Cases For .

Comparing safety outcomes inpolice use-of-force cases forlaw enforcement agencies that havedeployed Conducted Energy Devicesand a matched comparison groupthat have not:A quasi-experimental evaluationSeptember 2009Report submitted to the National Institute of JusticePERF Project Staff:Bruce Taylor, Ph.D., Research Director Daniel Woods, Associate Bruce Kubu, Senior Associate Chris Koper, Ph.D., Deputy Research Director Bill Tegeler, Deputy Director, Management Services Jason Cheney, Associate Mary Martinez, Associate James Cronin, Senior Associate Kristin Kappelman, Associate

Comparing safety outcomes inpolice use-of-force cases forlaw enforcement agencies that have deployedConducted Energy Devices and a matchedcomparison group that have not:A quasi-experimental evaluationSeptember 2009Report submitted to the National Institute of JusticePERF Project Staff:Bruce Taylor, Ph.D., Research DirectorDaniel Woods, AssociateBruce Kubu, Senior AssociateChris Koper, Ph.D., Deputy Research DirectorBill Tegeler, Deputy Director, Management ServicesJason Cheney, AssociateMary Martinez, AssociateJames Cronin, Senior AssociateKristin Kappelman, Associate

This study was funded by the National Institute of Justice (Grant # 2006-IJ-CX-0028).The views expressed are those of the authors and do not necessarily represent the viewsor the official position of the National Institute of Justice or any other organization.Police Executive Research Forum, Washington, D.C. 20036Copyright 2009 by Police Executive Research ForumAll rights reservedEdited by Craig FischerCover and interior design by Dave Williams

Table of ContentsAbstract .1Executive Summary .2Chapter 1: Introduction .9Chapter 2: Literature Review.12Chapter 3: Research Design and Methods .21Chapter 4: Study Results.35Chapter 5: Discussion and Conclusion .59References.72Appendices.79

AbstractHow law enforcement agencies (LEAs)manage the use-of-force by officers isperhaps one of the most important tasks thatthey will undertake. One weapon that hasbeen advanced as a way to reduce injuriesfor officers and suspects is the ConductedEnergy Device (CED). The purpose of ourproject, conducted from late 2006 to 2008,was to produce scientifically valid resultsthat will inform LEA executives’ decisionsregarding CED use. The goal of our studywas to produce practical information thatcan help LEAs establish guidelines thatassist in the effective design of CEDdeployment programs that support increasedsafety for officers and citizens. Weconducted one of the first quasi-experimentsto compare LEAs with CED deployment(n 7) to a set of matched LEAs (n 6) thatdo not deploy CEDs on a variety of safetyoutcomes, controlling for a variety ofincident factors (force used by officer, timeframe of incident, suspect race/gender/age,suspect resistant behavior, and suspectweapon use) and agency-level factors(agency policy on CEDs, size/density ofLEA, and population density forjurisdiction). For the LEAs that deployedCEDs, we collected two years of data beforeCED deployment and two years of data afterCED deployment. For the non-CED sites,we collected four years of data over asimilar period.Overall, we found that the CED siteswere associated with improved safetyoutcomes when compared to a group ofmatched non-CED sites on six of nine safetymeasures, including reductions in (1) officerinjuries, (2–3) suspect injuries and severeinjuries, (4–5) officers and suspectsreceiving injuries requiring medicalattention, and (6) suspects receiving aninjury that resulted in the suspect beingtaken to a hospital or other medical facility.(We refer to this last category as“hospitalization,” although we have no dataon the extent to which officers or suspectswho went to a hospital or other medicalfacility were admitted and stayed overnight,as opposed to simply receiving an evaluationor treatment and being released.)Also within CED agencies, in somecases the actual use of a CED by an officeris associated with improved safety outcomescompared to other less-lethal weapons. Forfive of the eight comparisons, the caseswhere an officer used a CED wereassociated with the lowest or second lowestrate of injuries, injuries requiring medicalattention, or injuries officer was taken to amedical facility such as hospital or medicalclinic for treatment of an injury due to a useof-force incident requiring “hospitalization”(see comment in previous paragraph). Therewere no differences between the CED andthe non-CED sites on the outcomes of thenumber of suspect deaths, officer severeinjuries, and officer injuries requiringhospitalization.The evidence from our study suggeststhat CEDs can be an effective weapon inhelping prevent or minimize physicalstruggles in use-of-force cases. LEAs shouldconsider the utility of the CED as a way toavoid up-close combative situations andreduce injuries to officers and suspects.PERF’s Quasi-Experimental Evaluation on Deployment of Less Lethal Weapons1

Executive SummaryThe management of police officers’ useof force is perhaps one of the mostimportant tasks that a law enforcementagency (LEA) will undertake. LEAexecutives have to make important policydecisions on the types of force that will beauthorized, technologies to deliver thatforce, and when and how often various typesof force can be used. One of the keyobjectives in managing force is designingapproaches to reduce incidents of police useof force and the injuries associated withforce. One weapon that has been advancedas a way to reduce injuries for officers andsuspects is the Conducted Energy Device(CED). Law enforcement executives havebeen overwhelmed with questions about theeffectiveness of CEDs and the safety ofthese devices. The lack of availableinformation and a full understanding of theeffects of using CEDs has hampered theability of police executives to makeinformed policy decisions about thesedevices. Police executives have beenprovided with little independent scientificevidence and guidance on the impact ofusing CEDs. While decades of research havedocumented the nature and extent of theforce used by police and the conditions andcorrelates that affect its application (Smith etal., 2007), little research has been doneisolating the effects of using CEDs oninjuries to suspects and officers.Project purpose, goalsand objectives:The purpose of our project was to producescientifically valid results that will informLEA executives’ decisions regarding the useof CEDs. The goal of our study was toproduce practical information that can helplaw enforcement executives make gooddecisions about whether to deploy CEDs,and if a decision is made to deploy them, tohelp the agencies develop CED policy andprocedural guidelines that provide increasedsafety for officers and citizens. In order toaccomplish this goal, our objective was toconduct an evaluation comparing LEAs thathave deployed CEDs to a matched group ofLEAs that have not deployed CEDs in termsof officer and suspect safety during use-offorce incidents.Research design:Our team used a quasi-experimental design(QED) to compare departments with CEDdeployment (n 7) to a set of matcheddepartments (n 6) that do not deploy CEDson a variety of outcomes. With our QED, weare able to isolate the safety outcomes to beexpected if a department deploys CEDs,controlling for a variety of relatedorganizational and individual/incident-levelfactors.A key element for all QEDs is theprocess used to select a comparison group.In our study, we used a matching design.CED (n 7) and non-CED (n 6) sites werematched based on violent crime levels,PERF’s Quasi-Experimental Evaluation on Deployment of Less Lethal Weapons2

police activity, agency size, and populationsize of jurisdiction. The inclusion of 13departments allows us not only to assessincident-level factors, but also someimportant departmental/organizational-levelfactors that could affect outcomes. Ourstudy is one of the first to examine LEAsthat use CEDs to matched LEAs that do notuse CEDs.We collected four years of data on allincidents of use of force for all of theparticipating departments. For the LEAs thatdeployed CEDs, we collected at least twoyears of data before CED deployment andtwo years of data after CED deployment.For the LEAs that did not deploy CEDs, wecollected at least four years of data over asimilar period. While the focus of our studywas on the use CEDs, we also collected dataon all use-of-force incidents (not just CEDcases) and examined the range of weaponsand unarmed tactics that the police employin exerting force to arrest suspects.Site participants:Our selection of cities was based on amatching analysis using a PERF nationallyrepresentative survey on use of force. Weobtained our data from seven sites that havedeployed CEDs and six sites that have notdeployed CEDs.Overall, we believe our CED and nonCED sites are comparable. We collecteddata from fairly comparable periods for theCED and non-CED sites, within a year ortwo. And while some differences emerged inour assessment of the comparability of ourCED and non-CED sites, most of thedifferences were relatively small and did notseem to introduce any substantivelyimportant biases. When combined with ourmultivariate analyses, we believe that wehave a reasonably comparable group of CEDand non-CED sites with results that areinterpretable.Data analytic approach:We conducted a series of analysescomparing CED and non-CED sites,including bivariate analyses to describe thebasic raw differences between the CED andnon-CED sites on our outcome measures,and a variety of multivariate analyses toattempt to assess the viability of thebivariate results and control for possiblealternative explanations for the earlier rawdifferences. Our first multivariate analyseswere done using logistic regression to isolatethe effects of CED deployment on oursafety-related outcomes where we includedthe following independent/control variables:Whether the agency deploys CEDs, the timeframe of the incident, an interaction of CEDmultiplied by time-frame, suspect race,suspect gender, suspect age, whether thesuspect used resistant behavior, and whetherthe suspect had a weapon at the forceincident.One of the concerns with examiningmulti-site data is that the individual use-offorce cases we analyze are clustered within13 departments. In our study, individualcases of weapon use by officers are nestedwithin specific police departments that havevarious policy guidelines on the use of force.Ignoring the nested structure of our data canpotentially lead to biased estimates. Toaddress this clustering issue we used twoapproaches. First, we conducted a modifiedlogistic regression with a robust varianceestimator to adjust for within-clustercorrelation. However, with this approach wedo not get aggregate-level coefficients to seethe exact effects of aggregate-levelconditions on our individual results. Toexamine and observe the effects ofaggregate-level factors, we conducted amulti-level analysis using HierarchicalLinear Modeling (HLM). While werecognize our limited statistical power toconduct HLM analyses (n 13 LEAs), we arePERF’s Quasi-Experimental Evaluation on Deployment of Less Lethal Weapons3