TY - JOUR
AU - Brandl, Steven G
AB - Police use of force has been, and remains, one of the most critical and controversial issues in policing. One way to address this issue, and perhaps make the police a little less controversial, is to identify situations where much force is used, force that may be deemed excessive. Policy and training may then be developed to reduce the frequency of these situations or to otherwise lesson the likelihood of force in these situations. For example, in the 1990s, research established that excessive force was a potential risk of motor vehicle pursuits; officers, fuelled by a combination of fear, adrenaline, and excitement, often used more force than necessary upon termination of motor vehicle pursuits (Alpert et al., 1997). Other studies demonstrated that motor vehicle pursuits resulted in significant risk of injury and death to subjects, officers, and bystanders (e.g. Alpert and Dunham, 1988; Alpert, 1997; Hill, 2002). Collectively, these findings led many in the field to conclude that the risks of motor vehicle pursuits outweighed their benefits in the case of nonviolent crime (Kaminski et al., 2012) and led to significant policy change. Many departments either adopted restrictive or prohibitive pursuit policies with the goal of reducing the frequency of motor vehicle pursuits and resulting injuries, deaths, excessive force, and liability (Alpert and Dunham, 1988). More recently, the risks associated with foot pursuits have been examined (Kaminski et al., 2012). Many questioned whether foot pursuits could be similarly dangerous and problematic as motor vehicle pursuits (Bohrer et al., 2000; Pinizzotto et al., 2002; Bobb, 2003, 2005; International Association of Chiefs of Police, 2003; Kaminski, 2007; Graham, 2009; Kaminski and Alpert, 2013; Plohetski, 2013). These studies have shown that police use of force is more likely during foot pursuit-related arrests than in arrest situations generally (Kaminski, 2007). Specifically, Kaminski et al. (2004) found that compared to other arrests, foot pursuits increased the likelihood of police using of force by 345%. The greater likelihood of force also increases the risk of injuries and assaults in arrest situations involving foot pursuits compared to other arrest situations, leading Kaminski (2007) to conclude that ‘…compared to arrests generally, foot pursuits appear to be a higher risk activity’ (p. 69). That said, many questions remain. One important unanswered question concerns the impact of foot pursuits on the amount of force used. While we know that force is more likely in foot pursuit arrest situations than other arrest situations, we do not know if foot pursuits enhance the likelihood of more force being used, relative to other situations where force is used. Do police use more force in force encounters involving foot pursuits? Similarly, are use of force incidents involving foot pursuits more likely to result in injury to the officer(s) and/or subject(s) involved than other use of force incidents? These are worthwhile questions to address. If the answers are yes, it highlights another situation where force may generate much controversy, especially if it is excessive, and policy may be developed to address these risks. If the answer is no, the merits of foot pursuits may be more accurately assessed on issues other than their association with the application of force. The frequency of foot pursuits1 calls additional importance to the issue. In this study, we examine 660 use of force incidents that occurred in a large Midwestern police department to determine the impact of foot pursuits on the amount of force used by officers. All else equal, we predict that more force is used in use of force situations involving foot pursuits, resulting in the greater likelihood of subject and officer injury, than in other force situations. Literature review In this study, we are interested in force-in-foot pursuits, but the parallel research on vehicle pursuits may be informative. Studies have established that the risk of use of force and injury are great in motor vehicle pursuits (Alpert et al., 1997). The link between force and vehicle pursuits can be explained in several ways. First, these are stressful events. Pursuits are dangerous; subjects put themselves, officers, and potentially innocent bystanders in danger when they flee. This danger creates an emotional (e.g. fear, anger) and physiological response (i.e. adrenaline release) that may lead officers to overdo it and, as stated by Alpert et al. (1997), ‘pull the subject out the vent window’ upon conclusion of the pursuit (p. 371). Another explanation for the use of force in situations in which subjects flee is symbolic; subject flight represents an ‘affront’ to the officer. It is a direct challenge to the officer’s authority. As Van Maanen (1978) described: In a very real sense, the patrolman-to-citizen exchanges are moral contests in which the authority of the state is either confirmed, denied, or left in doubt …. An affront, as used here, is a challenge to the policeman’s authority, control, and definition of the immediate situation … an affront is simply a response on the part of the other which indicates to them that their position and authority in the interaction is not being taken seriously. (p. 229) Skolnick and Fyfe (1993) provided a similar explanation for the use of force following subject flight, arguing that fleeing motorists commit a ‘cardinal sin’ when they flee and challenge the officer’s authority, making themselves ‘… prime candidates for painful lessons at the ends of police nightsticks’ (p. 111). In short, to deny an officer’s authority is to make oneself vulnerable to retaliatory violence once apprehended. For similar and additional reasons, it is possible that more force is used in force incidents involving foot pursuits than in other force situations. Whether by vehicle or foot, the act of fleeing symbolizes an unambiguous affront to officers’ authority and a direct challenge to their control over the situation (Van Maanen, 1978). Fleeing may represent resistance beyond what is normally expected from subjects. This form of resistance may only be stopped through risky actions on the part of the officer, actions with a high probability of causing injuries to officers2 such as running on uneven terrain, circumventing obstacles, confronting dogs, and perhaps most importantly, taking extraordinary actions to get physical control over the subject (e.g. tackling a subject on the run). Foot pursuit situations present high levels of uncertainty for officers, often unique compared to other force situations. Most significantly, maintaining constant visual contact with subjects even when in close proximity is often not possible thereby making weapon possession a paramount but undetermined threat. These factors could result in overwhelming force being used upon the subjects’ apprehension. A subject who flees may also have the physical capabilities to continue resistance once apprehended, thereby necessitating the application of more force by officers to successfully incapacitate the subject. It is also possible that a foot pursuit would lead to more officers being involved in the incident (chasing, searching) and correspondingly the opportunity for more officers to use force upon apprehension of the subject. As noted though, to date no studies have examined how the amount of force used in foot pursuit situations compares to that used in other force situations. However, studies have examined other potentially related issues relating to the risks of foot pursuits (Bohrer et al., 2000; Pinizzotto et al., 2002; Bobb, 2003, 2005; International Association of Chiefs of Police, 2003; Kaminski, 2007; Graham, 2009; Kaminski et al., 2012; Kaminski and Alpert, 2013; Plohetski, 2013). For example, in an attempt to more clearly specify the frequency of force in foot pursuit situations, Kaminski (2007) administered a web-based survey to deputies in the Richland County, South Carolina, Sheriff’s Department (RCSD). The survey asked officers who were ranked lieutenant and below to recall foot pursuits during three different time frames (i.e. lifetime, last 6 months, most recent foot pursuit). Officers reported that the use of force was common in foot pursuits. About 50% of foot pursuits involved the use of force, and in about 15% of the pursuits the officer used a weapon. With regard to officer injuries in foot pursuit situations, the Kaminski (2007) study also revealed that police officers were intentionally injured by subjects in 10% of incidents (i.e. assaulted) and accidentally injured (e.g. trip and fall) in 14% of incidents over a 6-month period. Moreover, the costs of injuries sustained during foot pursuits were quite high; officers needed medical attention often, missed many days of work, and spent even more days working in a reduced capacity (Kaminski, 2007). Although relatively few foot pursuits involved assaults on officers (8%), this is higher than the rate of assault on police in other arrest situations (Adams, 1999), leading Kaminski (2007) to conclude that foot pursuits are a higher risk activity than arrest situations more generally. To examine officer and subject injuries in foot pursuit incidents, Kaminski et al. (2012) analysed 267-foot pursuits from the Los Angeles County Sheriff’s Department. The authors found that 17% of foot pursuits resulted in injury to officers and 60% resulted in injuries to subjects. The authors also examined the effects of demographic and situational factors on officer and subject injuries (as well as the severity of injury among subjects). The analyses showed that officers were more likely to be injured when the officer used empty-hand tactics and when they were assaulted by the fleeing subject. Subjects were more likely to be injured when officers used empty-hand tactics, Conducted Energy Devices (CEDs), or canines; when one or more of the subjects were Hispanic; and when more than one officer was on scene. Subjects were less likely to be injured if impaired. The severity of subject injury was greater when more officers were on scene and when force was delivered via a police canine (Kaminski et al., 2012). There is also evidence that foot pursuits are implicated in a significant proportion of lethal force cases, ranging from 12% to 48% of cases of police shootings of civilians (Bobb, 2003, 2005; Simpson, 2007; Graham, 2009). The most recent of these analyses (based on data from Chicago and Austin, TX) suggest that around one-third of police shootings occur during foot pursuit incidents (Plohetski, 2013; Caputo et al., 2016). However, for a more accurate assessment of risk, Kaminski et al. (2012) explain that one should analyse foot pursuits and determine the number that involves shootings and/or fatalities (i.e. the ‘base-rate problem’; see Garner and Clemmer, 1986; Kaminski and Sorensen, 1995; Lichtenberg and Smith, 2001). Kaminski (2006, 2010), using this approach, found that about 1–3% of foot pursuits involve shootings. Brandl and Stroshine (2017), in their analyses predicting the use of Tasers and OC spray, examined the types of force used in force situations, including those that involved a foot pursuit. Their analyses showed that when subjects fled, officers were significantly more likely to use Tasers than OC spray to facilitate capture and arrest. From this finding, one might be tempted to conclude that foot pursuits lead to ‘more’ or ‘greater’ force used, since Tasers are often placed higher on use of force continua than OC spray (Alpert and Dunham, 2010; Taylor and Woods, 2010; Terrill and Paoline, 2013). However, it might simply be that Tasers provide a tactical advantage over OC spray in foot pursuit situations, since the chance of ‘blowback’ or collateral damage would likely be greater in foot pursuits (Bertilsson et al., 2017). Since Brandl and Stroshine (2017) did not directly test the impact of foot pursuits on the amount of force used, or on weapon use more generally, questions remain. In sum, the research has established that foot pursuits are associated with a high likelihood of force (Kaminski et al., 2004) and injury (Kaminski, 2007) in arrest situations, and the use of different types of force (Brandl and Stroshine, 2017) in force situations. However, we do not know if foot pursuits result in the use of more force compared to other force situations. We also need to better understand the relationship between foot pursuits, officer injury, and subject injury in use of force incidents. In this study, we use two measures of force to examine whether more force is used in foot pursuit situations than other force situations. We hypothesize the following: H1: More force is used in use-of-force situations that involve foot pursuits than in use-of-force situations that do not involve foot pursuits. Given that foot pursuits are broadcast to officers, the perceived seriousness of such incidents, and the possible on-going nature of the pursuit which may allow officers time to arrive on-scene, more officers may respond to the pursuit and participate in it, including in the apprehension and incapacitation of the fleeing subject through the use of force. H2: Officers are more likely to use a weapon3 in force situations involving foot pursuits than in other use of force situations. Prior research has shown that Taser use is common in foot pursuit situations (Brandl and Stroshine, 2017). The use of Tasers, as well as other weapons (i.e. batons, OC spray, flashlights), are indicative of the application of more force. Weapons could be used to stop the pursuit or to control the subject after the pursuit has ended. Many departments (including the study department) authorize officers to use weapons when confronted with this type of resistance. We also examine whether foot pursuits—all else equal–make officer and subject injury more likely in use of force situations. Do foot pursuits exert a unique and distinct effect on officer and subject injuries? We hypothesize the following: H3: Officer injury is more likely in use-of-force situations that involve foot pursuits than in use-of-force situations that do not involve foot pursuits. H4: Subject injury is more likely in use-of-force situations that involve foot pursuits than in use-of-force situations that do not involve foot pursuits. Foot pursuits are known to pose threats to officer safety (Brandl, 1996; Brandl and Stroshine, 2003, 2012; Kaminski, 2007). There is good reason to believe that use of force situations involving foot pursuits may be more hazardous to officers and subjects: not only are officers facing subject resistance that could escalate, but there are also physical hazards at play: officers and subjects may face uneven terrain, limited visibility, and possibly inclement weather (e.g. snow, rain) causing dangerous pursuit conditions (e.g. icy sidewalks, slippery grass). Method Data The data for this study were obtained from a large municipal police department. At the time of the study (2018), the department employed approximately 2,000 sworn officers, about 1,200 of whom were patrol officers. The police department served a population of approximately 600,000; 40% of the population was African American and 10% was Latino. The study department does not have a foot pursuit policy; it is up to officer discretion whether to give chase by foot. The data for the study were obtained from a case management system used by the study police department. The resulting database contained a comprehensive list of variables on each use of force incident recorded by the department including information about the incident, officers, and subjects. These data were based on use of force reports which were completed by supervisory officers whenever a use of force incident occurred. According to the official policy of the department at the time of the study, a report was required to be completed by a supervisory officer whenever an officer: discharged a firearm; used a baton in the line of duty; discharged an irritant, chemical, or inflammatory agent; deployed an Electronic Control Device; used physical force that involved focused strikes, diffused strikes, or decentralizations to the ground; used any type of force in which a person is injured or claimed injury, whether or not the injury was immediately visible. Along with the departmental use of force report, a narrative of the incident was also written by the supervisory officer based on interviews of officers as well as subjects and witnesses. For this study, all of the narratives were reviewed and additional data were coded from them (e.g. type of subject resistance). Only incidents where force was used against people were included.4 Cases involving unintentional force (e.g. officer accidentally discharges firearm) and cases where supervisors reported threats of force (e.g. ‘Stop resisting or I’ll tase you’) were excluded.5 In total, in 2018, there were 612 incidents that involved the intentional use of force in any of the above circumstances. Variables Dependent variables The primary issue of interest in this study was a comparison of the amount of force used in foot pursuit situations versus other force situations. We use two distinct measures to represent ‘more’ force (see Table 1). First, consistent with prior research (Terrill and Paoline, 2017), we provide a measure of cumulative force used at the scene. Cumulative force is an additive scale created by summing all the values of various types of force used by officer(s) during the use of force encounter (see Figure 1 for types of force and corresponding coding values). As shown in Figure 1, force was assessed according to the five levels of force as distinguished in the study department’s use of force policy; these levels included no force, dialogue, physical force, control alternatives, protective alternatives, and deadly force. For example, if a subject was ‘tased’ twice by officer(s) and then decentralized, our measure of cumulative force would have a value of 8 (i.e. 3 + 3 + 2). Alternatively, if an officer struck a subject once and then deployed his pepper spray, the cumulative force measure would have a value of 5 (i.e. 2 + 3). This measure of force takes into account the number of officers who used force by counting all the instances of force used by officers. It should be noted that our measure of force does not include threats to use a weapon. Cumulative force ranged in value from 0 to 19, with a mean value of 5.69 (SD = 3.07). Figure 1: Open in new tabDownload slide Resistance and force coding scheme. Figure 1: Open in new tabDownload slide Resistance and force coding scheme. Table 1: Descriptive Statistics and t-Test Results Variable . . . . . . Independent variables . . All . No . Yes . t-value . (n = 612) . (n = 473) . (n = 139) . (df) .  % Black officers M 11.71 12.10 11.07 0.37 SD (28.37) (28.72) (28.54) (592) Range 0–100  % Male officers M 96.09 95.67 97.61 −1.53 SD (14.66) (15.36) (12.13) (277.95) Range 0–100  Mean officer age M 34.63 35.35 32.14 5.02*** SD (7.15) (7.23) (6.40) (254.42) Range 22–58  % Black subjects M 80.14 79.52 82.73 −0.83 SD (39.89) (40.35) (37.93) (585) Range 0–100  % Male subjects M 86.57 84.84 92.09 −2.55* SD (33.80) (35.49) (27 09) (296.44) Range 0–100  Mean subject age M 28.42 28.85 26.65 2.14* SD (10.55) (10.82) (9.58) (590) Range 11–68  Believed armed M 0.28 0.25 0.36 −2.44* SD (0.44) (0.43) (0.48) (210.17) Range 0–1  Disrespect M 0.36 0.42 0.14 7.43*** SD (0.47) (0.49) (0.35) (317.46) Range 0–1  Suicidal M 0.07 0.08 0.02 3.36*** SD (0.25) (0.27) (0.15) (439.04) Range 0–1  Mentally impaired M 0.15 0.17 0.08 3.07** SD (0.35) (0.37) (0.27) (311.81) Range 0–1  Under the influence M 0.27 0.27 0.19 2.02* SD (0.45) (0.45) (0.40) (253.51) Range 0–1  Proactive encounter M 0.31 0.30 0.39 −1.93 SD (0.46) (0.46) (0.49) (216.90) Range 0–1  Threaten police or others M 0.20 0.24 0.08 5.18*** SD (0.40) (0.43) (0.27) (361.65) Range 0–1  Encounter recorded (BWC, dashcam) M 0.86 0.86 0.86 0.11 SD (0.35) (0.35) (0.35) (593) Range 0–1  Highest resistance M 3.15 3.19 3.02 2.95** SD (0.72) (0.79) (0.49) (371.04) Range 0–5 Dependent variables  Cumulative force M 5.69 5.68 5.60 0.24 SD (3.07) (3.09) (3.02) (593) Range 0–19  Weapon use M 0.15 0.11 0.22 −2.68** SD (0.35) (0.32) (0.41) (190.60) Range 0–1  Officer injury M 0.16 0.16 0.14 0.77 SD (0.36) (0.37) (0.34) (593) Range 0–1  Subject injury M 0.52 0.49 0.63 −3.01** SD (0.49) (0.49) (0.47) (235.53) Range 0–1 Variable . . . . . . Independent variables . . All . No . Yes . t-value . (n = 612) . (n = 473) . (n = 139) . (df) .  % Black officers M 11.71 12.10 11.07 0.37 SD (28.37) (28.72) (28.54) (592) Range 0–100  % Male officers M 96.09 95.67 97.61 −1.53 SD (14.66) (15.36) (12.13) (277.95) Range 0–100  Mean officer age M 34.63 35.35 32.14 5.02*** SD (7.15) (7.23) (6.40) (254.42) Range 22–58  % Black subjects M 80.14 79.52 82.73 −0.83 SD (39.89) (40.35) (37.93) (585) Range 0–100  % Male subjects M 86.57 84.84 92.09 −2.55* SD (33.80) (35.49) (27 09) (296.44) Range 0–100  Mean subject age M 28.42 28.85 26.65 2.14* SD (10.55) (10.82) (9.58) (590) Range 11–68  Believed armed M 0.28 0.25 0.36 −2.44* SD (0.44) (0.43) (0.48) (210.17) Range 0–1  Disrespect M 0.36 0.42 0.14 7.43*** SD (0.47) (0.49) (0.35) (317.46) Range 0–1  Suicidal M 0.07 0.08 0.02 3.36*** SD (0.25) (0.27) (0.15) (439.04) Range 0–1  Mentally impaired M 0.15 0.17 0.08 3.07** SD (0.35) (0.37) (0.27) (311.81) Range 0–1  Under the influence M 0.27 0.27 0.19 2.02* SD (0.45) (0.45) (0.40) (253.51) Range 0–1  Proactive encounter M 0.31 0.30 0.39 −1.93 SD (0.46) (0.46) (0.49) (216.90) Range 0–1  Threaten police or others M 0.20 0.24 0.08 5.18*** SD (0.40) (0.43) (0.27) (361.65) Range 0–1  Encounter recorded (BWC, dashcam) M 0.86 0.86 0.86 0.11 SD (0.35) (0.35) (0.35) (593) Range 0–1  Highest resistance M 3.15 3.19 3.02 2.95** SD (0.72) (0.79) (0.49) (371.04) Range 0–5 Dependent variables  Cumulative force M 5.69 5.68 5.60 0.24 SD (3.07) (3.09) (3.02) (593) Range 0–19  Weapon use M 0.15 0.11 0.22 −2.68** SD (0.35) (0.32) (0.41) (190.60) Range 0–1  Officer injury M 0.16 0.16 0.14 0.77 SD (0.36) (0.37) (0.34) (593) Range 0–1  Subject injury M 0.52 0.49 0.63 −3.01** SD (0.49) (0.49) (0.47) (235.53) Range 0–1 Open in new tab Table 1: Descriptive Statistics and t-Test Results Variable . . . . . . Independent variables . . All . No . Yes . t-value . (n = 612) . (n = 473) . (n = 139) . (df) .  % Black officers M 11.71 12.10 11.07 0.37 SD (28.37) (28.72) (28.54) (592) Range 0–100  % Male officers M 96.09 95.67 97.61 −1.53 SD (14.66) (15.36) (12.13) (277.95) Range 0–100  Mean officer age M 34.63 35.35 32.14 5.02*** SD (7.15) (7.23) (6.40) (254.42) Range 22–58  % Black subjects M 80.14 79.52 82.73 −0.83 SD (39.89) (40.35) (37.93) (585) Range 0–100  % Male subjects M 86.57 84.84 92.09 −2.55* SD (33.80) (35.49) (27 09) (296.44) Range 0–100  Mean subject age M 28.42 28.85 26.65 2.14* SD (10.55) (10.82) (9.58) (590) Range 11–68  Believed armed M 0.28 0.25 0.36 −2.44* SD (0.44) (0.43) (0.48) (210.17) Range 0–1  Disrespect M 0.36 0.42 0.14 7.43*** SD (0.47) (0.49) (0.35) (317.46) Range 0–1  Suicidal M 0.07 0.08 0.02 3.36*** SD (0.25) (0.27) (0.15) (439.04) Range 0–1  Mentally impaired M 0.15 0.17 0.08 3.07** SD (0.35) (0.37) (0.27) (311.81) Range 0–1  Under the influence M 0.27 0.27 0.19 2.02* SD (0.45) (0.45) (0.40) (253.51) Range 0–1  Proactive encounter M 0.31 0.30 0.39 −1.93 SD (0.46) (0.46) (0.49) (216.90) Range 0–1  Threaten police or others M 0.20 0.24 0.08 5.18*** SD (0.40) (0.43) (0.27) (361.65) Range 0–1  Encounter recorded (BWC, dashcam) M 0.86 0.86 0.86 0.11 SD (0.35) (0.35) (0.35) (593) Range 0–1  Highest resistance M 3.15 3.19 3.02 2.95** SD (0.72) (0.79) (0.49) (371.04) Range 0–5 Dependent variables  Cumulative force M 5.69 5.68 5.60 0.24 SD (3.07) (3.09) (3.02) (593) Range 0–19  Weapon use M 0.15 0.11 0.22 −2.68** SD (0.35) (0.32) (0.41) (190.60) Range 0–1  Officer injury M 0.16 0.16 0.14 0.77 SD (0.36) (0.37) (0.34) (593) Range 0–1  Subject injury M 0.52 0.49 0.63 −3.01** SD (0.49) (0.49) (0.47) (235.53) Range 0–1 Variable . . . . . . Independent variables . . All . No . Yes . t-value . (n = 612) . (n = 473) . (n = 139) . (df) .  % Black officers M 11.71 12.10 11.07 0.37 SD (28.37) (28.72) (28.54) (592) Range 0–100  % Male officers M 96.09 95.67 97.61 −1.53 SD (14.66) (15.36) (12.13) (277.95) Range 0–100  Mean officer age M 34.63 35.35 32.14 5.02*** SD (7.15) (7.23) (6.40) (254.42) Range 22–58  % Black subjects M 80.14 79.52 82.73 −0.83 SD (39.89) (40.35) (37.93) (585) Range 0–100  % Male subjects M 86.57 84.84 92.09 −2.55* SD (33.80) (35.49) (27 09) (296.44) Range 0–100  Mean subject age M 28.42 28.85 26.65 2.14* SD (10.55) (10.82) (9.58) (590) Range 11–68  Believed armed M 0.28 0.25 0.36 −2.44* SD (0.44) (0.43) (0.48) (210.17) Range 0–1  Disrespect M 0.36 0.42 0.14 7.43*** SD (0.47) (0.49) (0.35) (317.46) Range 0–1  Suicidal M 0.07 0.08 0.02 3.36*** SD (0.25) (0.27) (0.15) (439.04) Range 0–1  Mentally impaired M 0.15 0.17 0.08 3.07** SD (0.35) (0.37) (0.27) (311.81) Range 0–1  Under the influence M 0.27 0.27 0.19 2.02* SD (0.45) (0.45) (0.40) (253.51) Range 0–1  Proactive encounter M 0.31 0.30 0.39 −1.93 SD (0.46) (0.46) (0.49) (216.90) Range 0–1  Threaten police or others M 0.20 0.24 0.08 5.18*** SD (0.40) (0.43) (0.27) (361.65) Range 0–1  Encounter recorded (BWC, dashcam) M 0.86 0.86 0.86 0.11 SD (0.35) (0.35) (0.35) (593) Range 0–1  Highest resistance M 3.15 3.19 3.02 2.95** SD (0.72) (0.79) (0.49) (371.04) Range 0–5 Dependent variables  Cumulative force M 5.69 5.68 5.60 0.24 SD (3.07) (3.09) (3.02) (593) Range 0–19  Weapon use M 0.15 0.11 0.22 −2.68** SD (0.35) (0.32) (0.41) (190.60) Range 0–1  Officer injury M 0.16 0.16 0.14 0.77 SD (0.36) (0.37) (0.34) (593) Range 0–1  Subject injury M 0.52 0.49 0.63 −3.01** SD (0.49) (0.49) (0.47) (235.53) Range 0–1 Open in new tab Second, we measured weapon use (yes/no) as a measure of force. Fifteen per cent of use of force incidents involved a weapon. Since weapons are placed above most forms of physical force (i.e. soft- and hard-hands tactics) on use of force continua (e.g. Terrill and Paoline, 2013), weapon use is thought to represent more force. It must be highlighted here that with these indicators we are able to measure the amount of force used in a situation, we are not able to identify instances of excessive force. Where there is more force there may be excessive force, but not necessarily. Subjects who flee on foot may have physical capabilities and motivations that require officers to legitimately use greater amounts of force to gain control over subjects. Regardless, instances of ‘more force’ may be viewed as problematic and as ‘excessive’ by the public and for this reason, these situations are important to identify. Further, given the objectives of this study, we are not interested in simply measuring the highest level of force used; rather with the two measures of force used here (cumulative force and weapon use) we are interested in measuring the amount of force used. Simply analysing the highest level of force used by officer(s) would miss much of the complexity of the construct (e.g. consider the difference between an incident where seven officers used physical force and a Taser resulting in subject injuries and an incident where one officer used a Taser on a subject and that person was not injured). Our last two dependent measures are dichotomous (yes/no) measures of injury, one for officers and the other for subjects.6 All subject and officer injuries included in the study were related and relevant to the use of force or the apprehension of the subject. If officers or subjects were injured when officers used force, the injury was recorded. If officers or subjects were injured when subjects resisted apprehension in any way, the injury was recorded. So, for instance, if a subject (or officer) was injured as a result of the application of force (e.g. a decentralization), the injury was recorded. Similarly, if a subject was injured fleeing from an officer or an officer was injured pursuing a subject, the injury was recorded. In this study, an injury was sustained by an officer in 16% (n = 94) of cases. Subjects were injured more frequently,7 in just over half (52%; n = 302) of all cases. Independent variables Foot pursuits were the key independent variable of interest. For our purposes, a foot pursuit occurred whenever a subject fled on foot in an effort to avoid apprehension or get away from the officer. A foot pursuit could last only a few feet or could cover a long distance. Attempts to escape the control of officers (e.g. subjects struggled to flee) were not considered foot pursuits, nor were foot pursuits that were either immediately preceded or followed by motor vehicle pursuits. Approximately 23% of incidents (n = 139) involved a foot pursuit. To assess the impact of foot pursuits on the amount of force used, we control for other variables that could also impact the amount of force used and officer and subjects’ injuries (see Table 2) Table 2: OLS regression model of force used . Cumulative force . Independent Variables . B . Beta . SE . Foot pursuit 0.18 0.03 0.32 Per cent black officers 0.01 0.09 0.01* Per cent male officers 0.01 0.05 0.01 Mean officer age −0.03 −0.07 0.02 Per cent subjects black −0.00 −0.02 0.00 Per cent subjects male 0.01 0.08 0.00* Mean subject age 0.00 0.01 0.01 Believed armed 0.44 0.06 0.29 Disrespect 0.67 0.10 0.30* Suicidal 0.80 0.06 0.54 Mentally impaired 0.09 0.01 0.39 Under the influence 0.45 0.07 0.29 Proactive encounter −0.06 −0.01 0.30 Threaten police or others 0.29 0.04 0.35 Encounter recorded −0.08 −0.01 0.36 Highest resistance 0.71 0.17 0.19** Constant 2.20 1.37 N 589 F 3.29** df 16 R2 0.08 . Cumulative force . Independent Variables . B . Beta . SE . Foot pursuit 0.18 0.03 0.32 Per cent black officers 0.01 0.09 0.01* Per cent male officers 0.01 0.05 0.01 Mean officer age −0.03 −0.07 0.02 Per cent subjects black −0.00 −0.02 0.00 Per cent subjects male 0.01 0.08 0.00* Mean subject age 0.00 0.01 0.01 Believed armed 0.44 0.06 0.29 Disrespect 0.67 0.10 0.30* Suicidal 0.80 0.06 0.54 Mentally impaired 0.09 0.01 0.39 Under the influence 0.45 0.07 0.29 Proactive encounter −0.06 −0.01 0.30 Threaten police or others 0.29 0.04 0.35 Encounter recorded −0.08 −0.01 0.36 Highest resistance 0.71 0.17 0.19** Constant 2.20 1.37 N 589 F 3.29** df 16 R2 0.08 * p < 0.05, ** p < 0.001. Open in new tab Table 2: OLS regression model of force used . Cumulative force . Independent Variables . B . Beta . SE . Foot pursuit 0.18 0.03 0.32 Per cent black officers 0.01 0.09 0.01* Per cent male officers 0.01 0.05 0.01 Mean officer age −0.03 −0.07 0.02 Per cent subjects black −0.00 −0.02 0.00 Per cent subjects male 0.01 0.08 0.00* Mean subject age 0.00 0.01 0.01 Believed armed 0.44 0.06 0.29 Disrespect 0.67 0.10 0.30* Suicidal 0.80 0.06 0.54 Mentally impaired 0.09 0.01 0.39 Under the influence 0.45 0.07 0.29 Proactive encounter −0.06 −0.01 0.30 Threaten police or others 0.29 0.04 0.35 Encounter recorded −0.08 −0.01 0.36 Highest resistance 0.71 0.17 0.19** Constant 2.20 1.37 N 589 F 3.29** df 16 R2 0.08 . Cumulative force . Independent Variables . B . Beta . SE . Foot pursuit 0.18 0.03 0.32 Per cent black officers 0.01 0.09 0.01* Per cent male officers 0.01 0.05 0.01 Mean officer age −0.03 −0.07 0.02 Per cent subjects black −0.00 −0.02 0.00 Per cent subjects male 0.01 0.08 0.00* Mean subject age 0.00 0.01 0.01 Believed armed 0.44 0.06 0.29 Disrespect 0.67 0.10 0.30* Suicidal 0.80 0.06 0.54 Mentally impaired 0.09 0.01 0.39 Under the influence 0.45 0.07 0.29 Proactive encounter −0.06 −0.01 0.30 Threaten police or others 0.29 0.04 0.35 Encounter recorded −0.08 −0.01 0.36 Highest resistance 0.71 0.17 0.19** Constant 2.20 1.37 N 589 F 3.29** df 16 R2 0.08 * p < 0.05, ** p < 0.001. Open in new tab Officer demographic variables included race (Black/non-Black8), gender (male/female), and age (in years). For each incident, to account for incidents that involved multiple officers, we created variables to specify the per cent of Black officers, per cent of male officers, and mean officer age. Treating the variables in this manner reflects the racial, gender, and age dynamics among officers, as well as the hierarchical nature of the data.9 This approach has been used by others in prior research (e.g. Kaminski et al., 2012). Most officers were non-Black (88%), male (96%), and relatively young (a mean age of 34.6 years). Demographic characteristics of subjects are also depicted in Table 2 and consisted of gender (male/female), race (Black/non-Black), and age (in years). As with officers, variables were created to represent the percentage of Black subjects, percentage of male subjects, and mean age of subjects involved in the incident. Most subjects were Black (80%), male (87%), and young (a mean age of 28.4 years). In the few cases with multiple subjects, there was very little variation among the subjects in race, gender, or age. Next, we considered situational characteristics that have been shown in previous studies to consistently predict the use of force. Data were collected regarding whether a subject was believed armed (yes/no), disrespectful (yes/no), suicidal (yes/no), in crisis or having a mental health episode (yes/no), and under the influence of alcohol or drugs (yes/no). Subjects were believed to be carrying a weapon in 28% of cases. Subjects were disrespectful to officers in 36% of cases, suicidal in 7% of cases, and mentally impaired about 15% of the time. Another 27% of the time, subjects were under the influence of alcohol or drugs. ‘Disrespect’ was usually in the form of words spoken to officers (e.g. ‘fuck you police’) or gestures. Next, we created a variable that indicated whether the encounter was the result of proactive activity (yes/no) on the part of the officer(s) involved. Proactive encounters included traffic stops and officer-initiated field interviews and accounted for 31% of all use of force encounters included in this study. Another situational variable was whether the subject threatened the police or others (yes/no). Subjects made such threats in 20% of use of force encounters. We were also able to capture whether the encounter was recorded (either by body-worn camera or dashboard camera). The overwhelming majority of encounters (86%) were recorded, typically via body-worn camera. Finally, we include one measure of subject resistance: the highest level of resistance encountered. This variable was an ordinal measure comprised of six categories (see Figure 1). The average level of resistance encountered was 3.15 (SD = 0.70), representing a moderate level of resistance (such as attempts to evade police attempts at control, including flight). Analytic techniques Given the purposes of this study, we first conducted analyses (t-tests) to compare use of force situations in which foot pursuits occur with those where there was not a pursuit. Then, we estimated regression equations to examine the impact of foot pursuits on our four measures of force. First, how do the 139 use of force incidents that involved a foot pursuit differ from the 456 force incidents where a pursuit did not occur? As seen in Table 1, officers involved in foot pursuits were likely to be younger (t = 5.02; p < 0.001) than officers involved in use of force situations where a foot pursuit did not occur. Subjects were more likely to be male (t = −2.55; p < 0.05), younger (t = 2.14; p < 0.05), and believed armed (t = −2.44; p < 0.05) compared to subjects who did not flee in force situations. Subjects who fled were less likely to exhibit disrespect (t = 7.43; p < 0.01), be suicidal (t = 3.36; p < 0.001), mentally impaired (t = 3.07; p < 0.001), under the influence (t = 2.02, p < 0.05), threatening of police or others (t = 5.18; p < 0.001), and especially resistant (t = 2.95; p < 0.01). These findings are not surprising. Young males who are armed may have the physical capabilities to flee, as well as the motivation (they may be about to be arrested and/or are fearful of the police). Likewise, subjects who are suicidal, mentally impaired, under the influence, or threatening of police or others are in crisis, making flight less likely. Looking at our dependent variables, two of the four measures were significantly related to foot pursuits. Foot pursuit situations were more likely to involve the use of weapon (t= −2.34; p < 0.01) and a greater likelihood of subject injury (t = −3.01; p < 0.01) compared to non-pursuit use of force situations. Next, regression equations were estimated to identify factors that predict greater use of force; these results are shown in Tables 2 and 3.10 Table 2 shows the results of an ordinary least square (OLS) regression with cumulative force as the dependent variable.11 The model predicting cumulative force was statistically significant (F = 3.29; p < 0.001 with df= 16). Foot pursuit was not significant; controlling for other factors, use of force incidents that involved subject flight was no more forceful than incidents that did not involve flight. Instead, more force was used when officer(s) were Black and subject(s) were male, disrespectful, and highly resistant. Table 3: Logistic regression model of force used . Weapon used . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . Foot pursuit 0.98 0.29 2.67*** Per cent black officers 0.01 0.00 1.01** Per cent male officers 0.01 0.02 1.01 Mean officer age 0.01 0.02 1.01 Per cent black subjects 0.00 0.00 1.02 Per cent male subjects 0.00 0.00 1.00 Mean subject age 0.02 0.01 1.01 Believed armed 0.59 0.28 1.81* Disrespect 0.15 0.30 1.16 Suicidal 0.10 0.50 1.11 Mentally impaired 0.58 0.34 1.78 Under the influence 0.30 0.28 1.34 Proactive encounter 0.40 0.30 1.49 Threaten police or others 0.57 0.32 1.76 Encounter recorded 0.26 0.38 1.30 Highest resistance 0.31 0.20 1.36 Constant −6.23 1.51 0.00*** −2 Log likelihood 448.40 R2 (Nagelkerke) 0.11 Model chi-square (df) 38.09 (16) *** N 587 . Weapon used . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . Foot pursuit 0.98 0.29 2.67*** Per cent black officers 0.01 0.00 1.01** Per cent male officers 0.01 0.02 1.01 Mean officer age 0.01 0.02 1.01 Per cent black subjects 0.00 0.00 1.02 Per cent male subjects 0.00 0.00 1.00 Mean subject age 0.02 0.01 1.01 Believed armed 0.59 0.28 1.81* Disrespect 0.15 0.30 1.16 Suicidal 0.10 0.50 1.11 Mentally impaired 0.58 0.34 1.78 Under the influence 0.30 0.28 1.34 Proactive encounter 0.40 0.30 1.49 Threaten police or others 0.57 0.32 1.76 Encounter recorded 0.26 0.38 1.30 Highest resistance 0.31 0.20 1.36 Constant −6.23 1.51 0.00*** −2 Log likelihood 448.40 R2 (Nagelkerke) 0.11 Model chi-square (df) 38.09 (16) *** N 587 * p < 0.05 ** p < 0.01 *** p < 0.001. Open in new tab Table 3: Logistic regression model of force used . Weapon used . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . Foot pursuit 0.98 0.29 2.67*** Per cent black officers 0.01 0.00 1.01** Per cent male officers 0.01 0.02 1.01 Mean officer age 0.01 0.02 1.01 Per cent black subjects 0.00 0.00 1.02 Per cent male subjects 0.00 0.00 1.00 Mean subject age 0.02 0.01 1.01 Believed armed 0.59 0.28 1.81* Disrespect 0.15 0.30 1.16 Suicidal 0.10 0.50 1.11 Mentally impaired 0.58 0.34 1.78 Under the influence 0.30 0.28 1.34 Proactive encounter 0.40 0.30 1.49 Threaten police or others 0.57 0.32 1.76 Encounter recorded 0.26 0.38 1.30 Highest resistance 0.31 0.20 1.36 Constant −6.23 1.51 0.00*** −2 Log likelihood 448.40 R2 (Nagelkerke) 0.11 Model chi-square (df) 38.09 (16) *** N 587 . Weapon used . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . Foot pursuit 0.98 0.29 2.67*** Per cent black officers 0.01 0.00 1.01** Per cent male officers 0.01 0.02 1.01 Mean officer age 0.01 0.02 1.01 Per cent black subjects 0.00 0.00 1.02 Per cent male subjects 0.00 0.00 1.00 Mean subject age 0.02 0.01 1.01 Believed armed 0.59 0.28 1.81* Disrespect 0.15 0.30 1.16 Suicidal 0.10 0.50 1.11 Mentally impaired 0.58 0.34 1.78 Under the influence 0.30 0.28 1.34 Proactive encounter 0.40 0.30 1.49 Threaten police or others 0.57 0.32 1.76 Encounter recorded 0.26 0.38 1.30 Highest resistance 0.31 0.20 1.36 Constant −6.23 1.51 0.00*** −2 Log likelihood 448.40 R2 (Nagelkerke) 0.11 Model chi-square (df) 38.09 (16) *** N 587 * p < 0.05 ** p < 0.01 *** p < 0.001. Open in new tab Table 3 presents the results of a logistic regression analysis with weapon use (yes/no) as the dependent variable. A test of the full model predicting weapon use against a constant-only model was significant (chi-square = 38.09, p < 0.001, with df = 16). When the incident involved a foot pursuit, the odds of weapon use nearly tripled (odds ratio = 2.67; p < 0.001). Two other independent variables were significant predictors of weapon use: Weapon use was significantly more likely in use of force encounters involving more Black officers (odds ratio = 1.01; p < 0.01) and against subjects who were believed to be armed (odds ratio = 1.81; p < 0.05). Table 4 presents the results of two logistic regression analyses with officer injury (yes/no) as the dependent variable. In the first, we examined the impact of the same set of predictor variables in the other analyses in this paper, with the addition of cumulative force (chi-square = 34.15; p < 0.01; with df = 17). In the second regression, we analyse the impact of officer, subject, and situational characteristics and weapon use on officer injury (chi-square = 35.27; p < 0.01; with df = 17). Three variables were consistent across injury models. Subject race significantly increased the likelihood of officer injury in both injury models (though it should be noted this variable was substantively weak since the odds ratios were barely above 1.0). In both models, officer injury was less likely when subject(s)threatened police or others (odds ratios = 0.42 and 0.44; p < 0.05) and when the encounter was recorded (odds ratio = 0.53; p < 0.05). In the second officer injury model (including weapon use as a predictor variable), officer injury was significantly more likely when subject(s) resisted police(odds ratio = 1.54, p > 0.05). In neither equation did a foot pursuit significantly affect the likelihood of officer injury. Table 4: Logistic regression models of officer injury . Officer injury . Officer injury . . 0 = no . 1 = yes . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit −0.12 0.31 0.89 Foot pursuit −0.04 0.31 0.96 Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.01 0.01 0.99 Per cent male officers −0.01 0.01 0.99 Mean officer age 0.01 0.02 1.01 Mean officer age 0.00 0.02 1.00 Per cent black subjects 0.01 0.00 1.01* Per cent black subjects 0.01 0.00 1.01* Per cent male subjects −0.00 0.00 1.00 Per cent male subjects −0.00 0.00 1.00 Mean subject age 0.02 0.01 1.02 Mean subject age 0.02 0.01 1.02 Believed armed −0.61 0.31 0.55 Believed armed −0.54 0.30 0.58 Disrespect 0.31 0.27 1.36 Disrespect 0.32 0.27 1.38 Suicidal −0.46 0.53 0.63 Suicidal −0.22 0.52 0.80 Mentally impaired 0.41 0.34 1.51 Mentally impaired 0.38 0.34 1.46 Under the influence 0.18 0.28 1.120 Under the influence 0.26 0.28 1.30 Proactive encounter −0.10 0.30 0.91 Proactive encounter −0.07 0.29 0.94 Threaten police or others −0.82 0.34 0.44* Threaten police or others −0.86 0.35 0.42*** Encounter recorded −0.64 0.31 0.53* Encounter recorded −0.64 0.31 0.53 Cumulative resistance 0.05 0.05 1.06 Highest resistance 0.43 0.21 1.54* Cumulative force 0.04 0.07 1.04 Weapon use −0.12 0.36 0.89 Constant −2.03 1.16 0.13* Constant −2.77 1.29 0.06* −2 Log likelihood 463.63 −2 Log likelihood 470.01 R2 (Nagelkerke) 0.12 R2 (Nagelkerke) 0.10 Model chi-square (df) 41.64*** Model chi-square (df) 35.27** N 612 N 612 . Officer injury . Officer injury . . 0 = no . 1 = yes . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit −0.12 0.31 0.89 Foot pursuit −0.04 0.31 0.96 Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.01 0.01 0.99 Per cent male officers −0.01 0.01 0.99 Mean officer age 0.01 0.02 1.01 Mean officer age 0.00 0.02 1.00 Per cent black subjects 0.01 0.00 1.01* Per cent black subjects 0.01 0.00 1.01* Per cent male subjects −0.00 0.00 1.00 Per cent male subjects −0.00 0.00 1.00 Mean subject age 0.02 0.01 1.02 Mean subject age 0.02 0.01 1.02 Believed armed −0.61 0.31 0.55 Believed armed −0.54 0.30 0.58 Disrespect 0.31 0.27 1.36 Disrespect 0.32 0.27 1.38 Suicidal −0.46 0.53 0.63 Suicidal −0.22 0.52 0.80 Mentally impaired 0.41 0.34 1.51 Mentally impaired 0.38 0.34 1.46 Under the influence 0.18 0.28 1.120 Under the influence 0.26 0.28 1.30 Proactive encounter −0.10 0.30 0.91 Proactive encounter −0.07 0.29 0.94 Threaten police or others −0.82 0.34 0.44* Threaten police or others −0.86 0.35 0.42*** Encounter recorded −0.64 0.31 0.53* Encounter recorded −0.64 0.31 0.53 Cumulative resistance 0.05 0.05 1.06 Highest resistance 0.43 0.21 1.54* Cumulative force 0.04 0.07 1.04 Weapon use −0.12 0.36 0.89 Constant −2.03 1.16 0.13* Constant −2.77 1.29 0.06* −2 Log likelihood 463.63 −2 Log likelihood 470.01 R2 (Nagelkerke) 0.12 R2 (Nagelkerke) 0.10 Model chi-square (df) 41.64*** Model chi-square (df) 35.27** N 612 N 612 * p < 0.05 ** p < 0.01 *** p < 0.001. Open in new tab Table 4: Logistic regression models of officer injury . Officer injury . Officer injury . . 0 = no . 1 = yes . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit −0.12 0.31 0.89 Foot pursuit −0.04 0.31 0.96 Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.01 0.01 0.99 Per cent male officers −0.01 0.01 0.99 Mean officer age 0.01 0.02 1.01 Mean officer age 0.00 0.02 1.00 Per cent black subjects 0.01 0.00 1.01* Per cent black subjects 0.01 0.00 1.01* Per cent male subjects −0.00 0.00 1.00 Per cent male subjects −0.00 0.00 1.00 Mean subject age 0.02 0.01 1.02 Mean subject age 0.02 0.01 1.02 Believed armed −0.61 0.31 0.55 Believed armed −0.54 0.30 0.58 Disrespect 0.31 0.27 1.36 Disrespect 0.32 0.27 1.38 Suicidal −0.46 0.53 0.63 Suicidal −0.22 0.52 0.80 Mentally impaired 0.41 0.34 1.51 Mentally impaired 0.38 0.34 1.46 Under the influence 0.18 0.28 1.120 Under the influence 0.26 0.28 1.30 Proactive encounter −0.10 0.30 0.91 Proactive encounter −0.07 0.29 0.94 Threaten police or others −0.82 0.34 0.44* Threaten police or others −0.86 0.35 0.42*** Encounter recorded −0.64 0.31 0.53* Encounter recorded −0.64 0.31 0.53 Cumulative resistance 0.05 0.05 1.06 Highest resistance 0.43 0.21 1.54* Cumulative force 0.04 0.07 1.04 Weapon use −0.12 0.36 0.89 Constant −2.03 1.16 0.13* Constant −2.77 1.29 0.06* −2 Log likelihood 463.63 −2 Log likelihood 470.01 R2 (Nagelkerke) 0.12 R2 (Nagelkerke) 0.10 Model chi-square (df) 41.64*** Model chi-square (df) 35.27** N 612 N 612 . Officer injury . Officer injury . . 0 = no . 1 = yes . 0 = no . 1 = yes . Independent Variables . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit −0.12 0.31 0.89 Foot pursuit −0.04 0.31 0.96 Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.01 0.01 0.99 Per cent male officers −0.01 0.01 0.99 Mean officer age 0.01 0.02 1.01 Mean officer age 0.00 0.02 1.00 Per cent black subjects 0.01 0.00 1.01* Per cent black subjects 0.01 0.00 1.01* Per cent male subjects −0.00 0.00 1.00 Per cent male subjects −0.00 0.00 1.00 Mean subject age 0.02 0.01 1.02 Mean subject age 0.02 0.01 1.02 Believed armed −0.61 0.31 0.55 Believed armed −0.54 0.30 0.58 Disrespect 0.31 0.27 1.36 Disrespect 0.32 0.27 1.38 Suicidal −0.46 0.53 0.63 Suicidal −0.22 0.52 0.80 Mentally impaired 0.41 0.34 1.51 Mentally impaired 0.38 0.34 1.46 Under the influence 0.18 0.28 1.120 Under the influence 0.26 0.28 1.30 Proactive encounter −0.10 0.30 0.91 Proactive encounter −0.07 0.29 0.94 Threaten police or others −0.82 0.34 0.44* Threaten police or others −0.86 0.35 0.42*** Encounter recorded −0.64 0.31 0.53* Encounter recorded −0.64 0.31 0.53 Cumulative resistance 0.05 0.05 1.06 Highest resistance 0.43 0.21 1.54* Cumulative force 0.04 0.07 1.04 Weapon use −0.12 0.36 0.89 Constant −2.03 1.16 0.13* Constant −2.77 1.29 0.06* −2 Log likelihood 463.63 −2 Log likelihood 470.01 R2 (Nagelkerke) 0.12 R2 (Nagelkerke) 0.10 Model chi-square (df) 41.64*** Model chi-square (df) 35.27** N 612 N 612 * p < 0.05 ** p < 0.01 *** p < 0.001. Open in new tab In Table 5, we present results of parallel analyses conducted with subject injury as the dependent variable. Both our models predicting subject injury revealed a significant relationship between the foot pursuit variable and subject injury. When the encounter involved a foot pursuit, subject injury was nearly twice as common (odds ratios of 1.92 and 1.67; p < 0.05) as in use of force incidents not involving a foot pursuit. Not surprisingly, both measures of ‘more’ force (i.e. cumulative force, weapon use) significantly increased the likelihood of subject injury. While cumulative force made subject injury 1.28 times more likely (p < 0.01), weapon use increased the odds of subject injury by more than four (odds ratio = 4.52; p < 0.001). Table 5: Logistic regression models of subject injury . Subject injury . Subject injury . . 0 = no . 1 = yes . . 0 = no . 1 = yes . . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit 0.65 0.23 1.92** Foot pursuit 0.51 0.23 1.67* Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.00 0.01 1.00 Per cent male officers −0.00 0.01 1.00 Mean officer age 0.01 0.01 1.01 Mean officer age 0.01 0.01 1.01 Per cent black subjects 0.00 0.00 1.00 Per cent black subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Mean subject age −0.01 0.01 1.00 Mean subject age −0.01 0.01 0.99 Believed armed 0.35 0.21 1.42 Believed armed 0.33 0.21 1.39 Disrespect 0.13 0.21 1.14 Disrespect 0.10 0.22 1.11 Suicidal 0.04 0.37 1.04 Suicidal 0.17 0.38 1.19 Mentally impaired 0.08 0.27 1.08 Mentally impaired −0.08 0.28 0.92 Under the influence 0.19 0.21 1.21 Under the influence 0.22 0.22 1.25 Proactive encounter 0.13 0.21 1.14 Proactive encounter −0.01 0.22 0.99 Threaten police or others 0.11 0.24 1.12 Threaten police or others −0.11 0.25 0.90 Encounter recorded 0.13 0.25 1.14 Encounter recorded 0.06 0.26 1.06 Cumulative resistance −0.04 0.04 0.96 Highest resistance 0.18 0.13 1.20 Cumulative force 0.16 0.06 1.18** Weapon use 1.51 0.32 4.52*** Constant −1.28 0.89 0.28 Constant −1.12 0.98 0.33 −2 Log likelihood 738.60 −2 Log likelihood 722.33 R2 (Nagelkerke) 0.08 R2 (Nagelkerke) 0.12 Model chi-square (df) 34.15 (17) ** Model chi-square (df) 50.41 (17) N 612 N 612 . Subject injury . Subject injury . . 0 = no . 1 = yes . . 0 = no . 1 = yes . . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit 0.65 0.23 1.92** Foot pursuit 0.51 0.23 1.67* Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.00 0.01 1.00 Per cent male officers −0.00 0.01 1.00 Mean officer age 0.01 0.01 1.01 Mean officer age 0.01 0.01 1.01 Per cent black subjects 0.00 0.00 1.00 Per cent black subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Mean subject age −0.01 0.01 1.00 Mean subject age −0.01 0.01 0.99 Believed armed 0.35 0.21 1.42 Believed armed 0.33 0.21 1.39 Disrespect 0.13 0.21 1.14 Disrespect 0.10 0.22 1.11 Suicidal 0.04 0.37 1.04 Suicidal 0.17 0.38 1.19 Mentally impaired 0.08 0.27 1.08 Mentally impaired −0.08 0.28 0.92 Under the influence 0.19 0.21 1.21 Under the influence 0.22 0.22 1.25 Proactive encounter 0.13 0.21 1.14 Proactive encounter −0.01 0.22 0.99 Threaten police or others 0.11 0.24 1.12 Threaten police or others −0.11 0.25 0.90 Encounter recorded 0.13 0.25 1.14 Encounter recorded 0.06 0.26 1.06 Cumulative resistance −0.04 0.04 0.96 Highest resistance 0.18 0.13 1.20 Cumulative force 0.16 0.06 1.18** Weapon use 1.51 0.32 4.52*** Constant −1.28 0.89 0.28 Constant −1.12 0.98 0.33 −2 Log likelihood 738.60 −2 Log likelihood 722.33 R2 (Nagelkerke) 0.08 R2 (Nagelkerke) 0.12 Model chi-square (df) 34.15 (17) ** Model chi-square (df) 50.41 (17) N 612 N 612 * p < 0.05 ** p < 0.01, *** p < 0.001. Open in new tab Table 5: Logistic regression models of subject injury . Subject injury . Subject injury . . 0 = no . 1 = yes . . 0 = no . 1 = yes . . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit 0.65 0.23 1.92** Foot pursuit 0.51 0.23 1.67* Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.00 0.01 1.00 Per cent male officers −0.00 0.01 1.00 Mean officer age 0.01 0.01 1.01 Mean officer age 0.01 0.01 1.01 Per cent black subjects 0.00 0.00 1.00 Per cent black subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Mean subject age −0.01 0.01 1.00 Mean subject age −0.01 0.01 0.99 Believed armed 0.35 0.21 1.42 Believed armed 0.33 0.21 1.39 Disrespect 0.13 0.21 1.14 Disrespect 0.10 0.22 1.11 Suicidal 0.04 0.37 1.04 Suicidal 0.17 0.38 1.19 Mentally impaired 0.08 0.27 1.08 Mentally impaired −0.08 0.28 0.92 Under the influence 0.19 0.21 1.21 Under the influence 0.22 0.22 1.25 Proactive encounter 0.13 0.21 1.14 Proactive encounter −0.01 0.22 0.99 Threaten police or others 0.11 0.24 1.12 Threaten police or others −0.11 0.25 0.90 Encounter recorded 0.13 0.25 1.14 Encounter recorded 0.06 0.26 1.06 Cumulative resistance −0.04 0.04 0.96 Highest resistance 0.18 0.13 1.20 Cumulative force 0.16 0.06 1.18** Weapon use 1.51 0.32 4.52*** Constant −1.28 0.89 0.28 Constant −1.12 0.98 0.33 −2 Log likelihood 738.60 −2 Log likelihood 722.33 R2 (Nagelkerke) 0.08 R2 (Nagelkerke) 0.12 Model chi-square (df) 34.15 (17) ** Model chi-square (df) 50.41 (17) N 612 N 612 . Subject injury . Subject injury . . 0 = no . 1 = yes . . 0 = no . 1 = yes . . B . SE . Odds ratio . . B . SE . Odds ratio . Foot pursuit 0.65 0.23 1.92** Foot pursuit 0.51 0.23 1.67* Per cent black officers 0.00 0.00 1.00 Per cent black officers 0.00 0.00 1.00 Per cent male officers −0.00 0.01 1.00 Per cent male officers −0.00 0.01 1.00 Mean officer age 0.01 0.01 1.01 Mean officer age 0.01 0.01 1.01 Per cent black subjects 0.00 0.00 1.00 Per cent black subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Per cent male subjects 0.00 0.00 1.00 Mean subject age −0.01 0.01 1.00 Mean subject age −0.01 0.01 0.99 Believed armed 0.35 0.21 1.42 Believed armed 0.33 0.21 1.39 Disrespect 0.13 0.21 1.14 Disrespect 0.10 0.22 1.11 Suicidal 0.04 0.37 1.04 Suicidal 0.17 0.38 1.19 Mentally impaired 0.08 0.27 1.08 Mentally impaired −0.08 0.28 0.92 Under the influence 0.19 0.21 1.21 Under the influence 0.22 0.22 1.25 Proactive encounter 0.13 0.21 1.14 Proactive encounter −0.01 0.22 0.99 Threaten police or others 0.11 0.24 1.12 Threaten police or others −0.11 0.25 0.90 Encounter recorded 0.13 0.25 1.14 Encounter recorded 0.06 0.26 1.06 Cumulative resistance −0.04 0.04 0.96 Highest resistance 0.18 0.13 1.20 Cumulative force 0.16 0.06 1.18** Weapon use 1.51 0.32 4.52*** Constant −1.28 0.89 0.28 Constant −1.12 0.98 0.33 −2 Log likelihood 738.60 −2 Log likelihood 722.33 R2 (Nagelkerke) 0.08 R2 (Nagelkerke) 0.12 Model chi-square (df) 34.15 (17) ** Model chi-square (df) 50.41 (17) N 612 N 612 * p < 0.05 ** p < 0.01, *** p < 0.001. Open in new tab Discussion This study aimed to better assess the potential hazards of foot pursuits in policing, specifically as they pertain to the police use of force. In doing so, the risks of force in foot pursuits vis-à-vis the risks of force in other situations were assessed. Similar to research regarding the use of force and vehicle pursuits, we hypothesized that more force would be used in foot pursuit force situations compared to force situations that did not involve a fleeing subject. Overall, we found mixed support for this hypothesis. While foot pursuits are not at higher risk for the application of greater cumulative amounts of force than other situations where force is used, we found that foot pursuits did increase the likelihood of weapon use. We also hypothesized that use of force situations involving subject flight would result in greater officer and subject injuries. Once again, we found mixed support for this hypothesis: Foot pursuits increased the likelihood of injuries to subjects but not to officers. This apparent inconsistency makes sense given that foot pursuits increased the likelihood of weapon use. Police use of weapons has an adverse impact on subjects but not officers. It appears that rather than increasing the overall amount of force used, officers are more likely to use different types of force (weapons) in foot pursuit situations. As such, use of force in foot pursuit situations appears to be more nuanced than what has been demonstrated in the findings of research on the vehicle pursuit-force relationship. How can one make sense of these findings? First, previous research on vehicle pursuits and force has not examined the types of force most likely to be used by officers. Perhaps in vehicle pursuits too, officers are more likely to use weapons than in other use of force situations. This remains an unanswered empirical question. Second, it is important to consider the similarities and differences in the structure of foot pursuits, vehicle pursuits, and other use of force situations. Foot pursuits require greater physical abilities (aerobic) and exertion than motor vehicle pursuits, as well as many other force situations. Therefore, fewer officers may be involved in the foot pursuit and the use of force because some officers may simply be physically unable to pursue its termination. With fewer officers comes less cumulative force. Relatedly, perhaps officers may be less likely to use more cumulative force because by the time the subject is apprehended the officer(s) may be exhausted; they are physically spent from the pursuit. Perhaps the same goes for the subject; with no further resistance from the subject, ‘more force’ on the part of officer(s) is not necessary. In addition, or alternatively, the use of a weapon by officer(s) may serve as a practical and tactical advantage in potentially exhausting foot pursuit situations. Use of a weapon may allow officers to gain compliance over a subject who has demonstrated the willingness and ability to escape the control of officers. Under these circumstances, officer(s) may be more likely to substitute physical force with force delivered via a weapon. The third possibility that cannot be discounted or underestimated here is that our findings conflict with those relating to vehicle pursuits and force simply because the well-established relationship between vehicle pursuits and more force is no longer valid. Perhaps in the new era of police body-worn cameras (BWCs) foot pursuits and vehicle pursuits are no longer associated with more force, at least in a statistically significant way. Further, foot pursuits are high-visibility events known to officers and supervisors as they occur. Simply stated, in this post-Ferguson context of BWCs, officers may feel as though they are ‘under the microscope’ and less likely to use great amounts of force in all situations but especially highly visible foot and vehicle pursuits. At the time of this study, some of the officers in this department were equipped with BWCs. It may only take one officer with a camera to discourage the use of force by other officers involved in a foot pursuit. Plus, given safety, liability, and supervisory concerns, officers are probably more likely to make sure the camera is operating in this type of situation, further decreasing the likelihood of more force being used. Another explanation may have to do with our measurement of force, and our indicators of ‘more force.’ Force did not include that delivered via soft hands (grabs and holds) and was based on police-reported incidents. Commonly there is variation in studies on how force is measured and the type of force data analysed. The ways in which we measured these factors may have had effects on the results obtained here. Policy implications and directions for future research From a practical and policy perspective, our results represent good news and are also potentially concerning. The good news is that officers are no more likely to be injured and officers do not use more overall force in foot pursuit situations compared to other force situations. What is potentially concerning is that subjects are more likely to be injured and officers are more likely to use weapons in foot pursuits compared to other use of force situations. And although officers are not more likely to be injured in foot pursuit force situations than in other force situations, the proportion who are injured is still substantively significant (16%). In light of these findings, there is enough reason to pay more attention to foot pursuits in terms of policy and future research. However, given the current state of research on foot pursuits, and the use of force in foot pursuits specifically, we need to know more about foot pursuits before reasonable and defensible policy can be developed. As a starting point to assess officer and subject injuries, and other risks, associated with foot pursuits, departments should maintain a database of foot pursuits (Bobb, 2003, 2005). The department in this study did not have a foot pursuit policy nor did it systematically document all foot pursuits. It was even impossible for the department to simply flag all incidents that involved a foot pursuit. Analysis of such data could be a basis for the management of risks associated with foot pursuits and a prerequisite for a data-informed foot pursuit policy. By documenting information on foot pursuits such as their length (e.g. physical distance or time), the conditions under which they occur (weather, time of day, whether preceded by a motor vehicle pursuit), the characteristics of officers and subjects who engage in them (e.g. weight, height, sex, fitness level of officers), whether force is used (and the amount and type of force used), and whether injuries occur, departments could have a much richer understanding of foot pursuits and the ways policy might be used to reduce the risks associated with them. Foot pursuit policies remain uncommon in policing (only about 15% of departments have foot pursuit policies in place; R. J. Kaminski and J. Rojek 2015), but the formulation of such policies may be appropriate. From such databases, future research could analyse data where foot pursuits are the reference population (instead of use of force situations). This was also recommended by Kaminski et al. (2012) to determine the relative risk of shootings/fatalities in foot pursuit situations. Again, of particular policy interest is how the length of foot pursuits and the conditions under which they occur affect force decisions and injuries. Similarly, researchers should examine the characteristics of involved officers and subjects and how these factors affect whether force is used, the utility of weapon use in these situations, and whether injuries are sustained. It is especially critical to understand how and why subjects are injured in foot pursuit situations and how officers’ use of a weapon may contribute to these injuries. This additional information would allow for a better understanding of foot pursuits and their potential benefits or hazards. Finally, we suggest that researchers conduct a direct comparison of foot and vehicle pursuits in the same jurisdiction at the same time. While there were not enough cases of vehicle pursuits to provide meaningful analyses in the current study (at least in part due to a restrictive vehicle pursuit policy), but this would be desirable in the future. A direct comparison would allow for more definitive conclusions to be drawn about the risks associated with each type of pursuit and provide an update on often dated findings regarding pursuits and force. While this study represents an important step towards providing a greater understanding of the impact of foot pursuits on use of force situations, it has limitations. Most notably, this study was conducted in one large department over a 1-year time period, force incidents did not include soft hand techniques, and the data were based on officer and supervisory reports. As such, the generalizability of the findings may be questioned. External validity is always an empirical issue; the results depend on replication for verification. An important research goal is to make policing safer and perhaps a bit less controversial. To do so, a better understanding of police use of force and the circumstances in which it is used is necessary. This study represents a step towards this goal. Footnotes 1 In 2018, police made an estimated 10.3M arrests (FBI, 2019). It is estimated that between 6% (Garner and Maxwell, 1999) and 9% (Kaminski et al., 2004) of arrests situations involve the attempted flight of subjects, with foot pursuits slightly more common than vehicular pursuits (Garner and Maxwell, 1999; Kaminski et al., 2004). This translates to several hundred thousand foot pursuits a year. 2 Research has shown that when officers are injured on-duty, 12% to 14% of the time it is during a foot pursuit incident (Brandl, 1996; Brandl and Stroshine, 2003, 2012). 3 In this study, ‘weapon’ refers to less lethal weapons, including Tasers, OC spray, batons, and flashlights. 4 In 20 incidents, force was used against a dog. These cases were excluded from the current study. For purposes of content, in 2018 there were five incidents where officers discharged their firearms at subjects (3 subjects were struck, 2 of those subjects were killed). These two cases represented the subject fatalities for the year. None of these incidents involved a foot pursuit. Also, three officers were killed in the line of duty in 2018 (two were murdered, one was killed in a vehicle accident). None of these incidents involved a foot pursuit. Prior to these fatalities, the most recent line-of-duty death in this department was in 1996 when an officer was shot and killed by a suspect during a foot pursuit. 5 Our measure of force does not include threats to use weapons. Clearly there is a substantial difference between threatening to use a weapon and actually deploying that weapon when quantifying the amount of force used in an incident. Threats to use a weapon were not required to be reported by officers. It should be noted that when the 48 incidents that involved a threat to use a weapon were included, there were negligible changes in the results. 6 While not ideal, the dichotomous treatment of these variables is not uncommon (e.g., Smith et al., 2007; Paoline et al., 2012) and is necessary given how the study agency classified these variables. Most injuries to officers (64%) and subjects (91%) were minor in nature, including contusions, abrasions, lacerations, and bruises. Consistent with prior research, serious injuries such as broken bone(s), teeth, dislocated joints, and knife wounds were very uncommon (Henriquez, 1999; Alpert and Dunham, 2000; Smith and Petrocelli, 2002; Kaminski et al., 2004). 7 If a subject was injured from Taser probes or burned by Tasers in stun mode, the subject was considered to have been injured and coded accordingly. Our measure of injury also includes eye or respiratory irritation as a result of the use of OC spray. We note this is controversial (e.g., see Paoline et al., 2012; Kaminski et al., 2015), however, additional analyses (available on request) revealed that excluding injuries related to Tasers or OC spray led to no substantive differences in the outcomes examined. 8 The ‘non-Black’ category included White, Hispanic/Latinx, Native American, and Asian/Pacific Islander. While preferable, there were too few “other” cases to measure race/ethnicity ordinally (i.e., White, Black, Other). 9 The data is hierarchical, in that subjects and officers are nested within use of force encounters. While most encounters involved one officer (54%) and one subject (97%), there were some that involved more than one officer or subject. Providing the percentage of Black officers/subjects, percentage of male officers/subjects, and the mean age of officers and subjects are one way to deal with the nested nature of the data given that our sample size precluded hierarchical modeling of the data. 10 Collinearity diagnostics were conducted on the independent variables in the multivariate analyses. Variance inflation factor scores, condition index numbers and variance proportions were all within conventional limits (Neter et al., 1996; Kennedy, 2003). Regression diagnostics including tests for heteroskedasticity were also performed and revealed no cause for concern. 11 Since cumulative force is a skewed variable (i.e., fewer cases as cumulative force increases) that is over-dispersed (i.e., the conditional variance exceeds the conditional mean), a negative binomial model might be more appropriate. However, results indicated no substantive difference between the results obtained via negative binomial model and an OLS regression (analyses available upon request). For ease of interpretation, OLS results are presented here. References Adams K. ( 1999 ). ‘What We Know about Police Use of Force.’ In Adams K. (ed.), Use of Force by Police: Overview of National and Local Data . Washington, DC : National Institute of Justice , pp. 1 – 14 . http://catalog.hathitrust.org/Record/003485418 (accessed 25 June 2021). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Alpert G. P. ( 1997 ). Police Pursuit: Policies and Training; Research in Brief . Washington, DC : National Institute of Justice . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Alpert G. P. , Dunham R. ( 1988 ). ‘ Research on Police Pursuits: Applications for Law Enforcement .’ American Journal of Police 7 : 123 – 131 . Google Scholar OpenURL Placeholder Text WorldCat Alpert G. P. , Dunham R. G. ( 2000 ). Analysis of Police Use-of-Force Data . Washington, DC : National Institute of Justice . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Alpert G. P. , Dunham R. G. ( 2010 ). ‘ Policy and Training Recommendations Related to Police Use of CEDs: Overview of Findings from a Comprehensive National Study.’ Police Quarterly 13 ( 3 ): 235 – 259 . doi:10.1177/1098611110373993 (accessed 15 May 2020). Google Scholar OpenURL Placeholder Text WorldCat Alpert G. , Kenney D. J., Dunham R. ( 1997 ). ‘ Police Pursuits and the Use of Force: Recognizing and Managing “the Pucker Factor” – a Research Note .’ Justice Quarterly 14 ( 2 ): 371 – 385 . Google Scholar Crossref Search ADS WorldCat Bertilsson J. , Petersson U., Fredriksson P. J., Magnusson M., Fransson P. A. ( 2017 ). ‘ Use of Pepper Spray in Policing: Retrospective Study of Situational Characteristics and Implications for Violent Situations .’ Police Practice & Research 18 ( 4 ): 391 – 406 . doi:10.1080/15614263.2017.1288119 (accessed 19 October 2019). Google Scholar Crossref Search ADS WorldCat Bobb M. J. ( 2003 ). The Los Angeles County Sheriff's Department 16th Semiannual Report . Los Angeles, CA : Police Assessment Resource Center . http://www.parc.info/los_angeles_county_sheriffs_department.chtml (accessed 3 March 2019). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bobb M. J. ( 2005 ). The Los Angeles County Sheriff's Department 19th Semiannual Report . Los Angeles, CA : Police Assessment Resource Center . http://www.parc.info/los_angeles_county_sheriffs_department.chtml (accessed 3 March 2019). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bohrer S. , Davis E., Garrity T. ( 2000 ). ‘ Establishing a Foot Pursuit Policy: Running into Danger .’ FBI Law Enforcement Bulletin 69 ( 5 ): 10 – 15 . Google Scholar OpenURL Placeholder Text WorldCat Brandl S. G. ( 1996 ). ‘ In the Line of Duty: A Descriptive Analysis of Police Assaults and Accidents .’ Journal of Criminal Justice 24 ( 3 ): 255 – 265 . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=true&db=i3h&AN=9606164088&site=ehost-live (accessed 7 July 2020). Google Scholar Crossref Search ADS WorldCat Brandl S. G. , Stroshine M. S. ( 2003 ). ‘ Toward an Understanding of the Physical Hazards of Police Work .’ Police Quarterly 6 ( 2 ): 172 – 191 . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=true&db=i3h&AN=9976364&site=ehost-live (accessed 16 June 2021). Google Scholar Crossref Search ADS WorldCat Brandl S. G. , Stroshine M. S. ( 2012 ). ‘ The Physical Hazards of Police Work Revisited .’ Police Quarterly 15 ( 3 ): 262 – 282 . doi:10.1177/1098611112447757 (accessed 16 June 2021). Google Scholar Crossref Search ADS WorldCat Brandl S. G. , Stroshine M. S. ( 2017 ). ‘ Oleoresin Capsicum Spray and TASERs .’ Criminal Justice Policy Review 28 ( 3 ): 279 – 306 . doi:10.1177/0887403415578732 (accessed 16 June 2021). Google Scholar Crossref Search ADS WorldCat Caputo A. , Richard J. S., Meisner J. ( 2016 , November 12,). ‘Third of Police Shootings Started with Foot Chases, Tribune Analysis Finds.’ Chicago Tribune. https://www.chicagotribune.com/news/watchdog/ct-chicago-police-shooting-foot-chase-met-20160907-story.html (accessed 21 February 2020). FBI . ( 2019 ). 2018: Crime in the United States. https://ucr.fbi.gov/crime-in-the-u.s/2018/crime-in-the-u.s.-2018/topic-pages/persons-arrested (accessed 20 May 2020). Garner J. , Clemmer E. ( 1986 ). Danger to Police in Domestic Disturbances–a New Look (Research in Brief). Washington, DC : National Institute of Justice . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Garner J. , Maxwell C. ( 1999 ). ‘Measuring the Amount of Force Used by and against the Police in Six Jurisdictions.’ In Use of Force by Police: Overview of National and Local Data . Washington, DC : U.S. Department of Justice , pp. 22 – 44 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Graham T. ( 2009 , November 16). ‘Lawsuit Shines Spotlight on Police Foot Pursuits.’ The Philadelphia Enquirer. http://www.policeone/legal/articles/1966072-Lawsuit-shines-spotlight-on-Philly-police-foot-pursuits/ (accessed 21 February 2020). Henriquez M. ( 1999 ). ‘IACP National Database Project on Police Use of Force.’ In Adams, K. editor. Use of Force by Police: Overview of National and Local Data . Washington, DC : National Institute of Justice, Bureau of Justice Statistics. pp. 19-24 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hill J. ( 2002 ). ‘ High-Speed Police Pursuits: Danger, Dynamics, and Risk Reduction .’ The FBI Law Enforcement Bulletin 71 ( 7 ): 14 – 18 . Google Scholar OpenURL Placeholder Text WorldCat International Association of Chiefs of Police . ( 2003 ). Foot Pursuit Model Policy. http://www.theiacp.org/PublicationsGuides/Projects/ViolenceAgainstthePolice/Checklist/Policies/tabid/990/Default.aspx (accessed 21 February 2020). Kaminski R. J. ( 2006 ). An Exploratory Study of Police Foot Pursuits in the Richland County Sheriff's Department. http://www.cas.sc.edu/crju/footsurvey.html (accessed 26 January 2019). Kaminski R. J. ( 2007 ). ‘ Police Foot Pursuits and Officer Safety .’ Law Enforcement Executive Forum 7 ( 3 ): 59 – 72 . Google Scholar OpenURL Placeholder Text WorldCat Kaminski R. J. ( 2010 ). A Descriptive Analysis of Foot Pursuits in the Los Angeles County Sheriff's Department . Los Angeles, CA . www.cas.sc.edu/crju/research.html (accessed 26 January 2019). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kaminski R. J. , DiGiovanni C., Downs R. ( 2004 ). ‘ The Use of Force between the Police and Persons with Impaired Judgment .’ Police Quarterly 7 ( 3 ): 311 – 338 . Google Scholar Crossref Search ADS WorldCat Kaminski R. J. , Rojek J., Smith H. P., Alpert G. P. ( 2012 ). ‘ Correlates of Foot Pursuit Injuries in the Los Angeles County Sheriff's Department.’ Police Quarterly 15 ( 2 ): 177 – 196 . Google Scholar OpenURL Placeholder Text WorldCat Kaminski R. J. , Sorensen D. W. M. ( 1995 ). ‘ A Multivariate Analysis of Individual, Situational, and Environmental Factors Associated with Police Assault Injuries .’ American Journal of Police 14 ( 3/4 ): 3 – 48 . Google Scholar Crossref Search ADS WorldCat Kaminski R. J. , Alpert G. P. ( 2013 ). ‘ Recent Findings on Police Foot Pursuits ’. Research in Brief.’ Police Chief 80 ( 1 ): 14 .http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=truedb=i3hAN=84864707site=ehost-live (accessed 13 August 2021). Google Scholar OpenURL Placeholder Text WorldCat Kaminski R. J. , Engel R. S., Rojek J., Smith M. R., Alpert G. ( 2015 ). ‘ A Quantum of Force: The Consequences of Counting Routine Conducted Energy Weapon Punctures as Injuries.’ Justice Quarterly 32 ( 4 ): 598 – 625 . doi:10.1080/07418825.2013.788729 (accessed 13 August 2021). Google Scholar Crossref Search ADS WorldCat Kennedy P. ( 2003 ). A Guide to Econometrics . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lichtenberg I. D. , Smith A. ( 2001 ). ‘ How Dangerous Are Routine Police-Citizen Traffic Stops? A Research Note .’ Journal of Criminal Justice 29 ( 5 ): 419 – 428 . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=truedb=i3hAN=5524949site=ehost-live (accessed 23 May 2019). Google Scholar Crossref Search ADS WorldCat Neter J. , Kutner M., Nachtsheim C., Wasserman W. ( 1996 ). Applied Linear Statistical Models , 4th edn. Chicago, IL : Irwin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Paoline E. A. III, Terrill W., Ingram J. R. ( 2012 ). ‘ Police Use of Force and Officer Injuries: Comparing Conducted Energy Devices (CEDs) to Hands- and Weapon-Based Tactics .’ Police Quarterly 15 ( 2 ): 115 – 136 . Google Scholar OpenURL Placeholder Text WorldCat Pinizzotto A. J. , Davis E. F., Miller C. E. III. ( 2002 ). ‘ Accidentally Dead: Accidental Line-of-Duty Deaths of Law Enforcement Officers ’. FBI Law Enforcement Bulletin 71 ( 7 ): 8 – 13 . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=truedb=i3hAN=11707604site=ehost-live (accessed 28 February 2020). Google Scholar OpenURL Placeholder Text WorldCat Plohetski T. ( 2013 , September 8). ‘Dangerous Pursuit: Austin Police Foot Chases Can Lead to Injuries, Shootings.’ Austin American-Statesman. https://www.statesman.com/news/20130908/dangerous-pursuit-austin-police-foot-chases-can-lead-to-injuries-shootings (accessed 21 February 2020). Simpson D. ( 2007 , March 5). ‘Police Foot Chases Not as Simple as Shown on TV.’ Atlanta Journal-Constitution. http://www.newslibrary.com/sites/ajc/archives (accessed 21 February 2020). Skolnick J. H. , Fyfe J. J. ( 1993 ). Above the Law: Police and the Excessive Use of Force . New York : Maxwell Macmillan International . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=true&db=cat02292a&AN=mar.b1692997&site=eds-live (accessed 17 July 2021). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smith M. R. , Petrocelli M. ( 2002 ). ‘ The Effectiveness of Force Used by Police in Making Arrests .’ Police Practice and Research: An International Journal 3 ( 3 ): 201 – 215 . Google Scholar Crossref Search ADS WorldCat Smith M. R. , Kaminski R. J., Rojek J., Alpert G. P., Mathis J. ( 2007 ). ‘ The Impact of Conducted Energy Devices and Other Types of Force and Resistance on Officer and Suspect Injuries .’ Policing: An International Journal 30 ( 3 ): 423 – 446 . http://0-search.ebscohost.com.libus.csd.mu.edu/login.aspx?direct=truedb=i3hAN=26435866site=ehost-live (accessed 15 September 2019). Google Scholar Crossref Search ADS WorldCat Taylor B. , Woods D. J. ( 2010 ). ‘ Injuries to Officers and Suspects in Police Use-of-Force Cases: A Quasi-Experimental Evaluation .’ Police Quarterly 13 ( 3 ): 260 – 289 . Google Scholar Crossref Search ADS WorldCat Terrill W. , Paoline E. A. III. ( 2013 ). ‘ Examining Less Lethal Force Policy and the Force Continuum: Results from a National Use-of-Force Study .’ Police Quarterly 16 ( 1 ): 38 – 65 . Google Scholar Crossref Search ADS WorldCat Terrill W. , Paoline E. A. ( 2017 ). ‘ Police Use of Less Lethal Force: Does Administrative Policy Matter? ’ Justice Quarterly 34 ( 2 ): 193 – 216 . Google Scholar Crossref Search ADS WorldCat Van Maanen J. ( 1978 ). ‘The Asshole. ’ In Policing: A View from the Street . Santa Monica, CA : Goodyear Publishing , pp. 221 – 238 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © The Author(s) 2021. Published by Oxford University Press. All rights reserved. For permissions please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - The Impact of Foot Pursuits on Police Use of Force
JF - Policing:A Journal of Policy and Practice
DO - 10.1093/police/paab054
DA - 2021-09-11
UR - https://www.deepdyve.com/lp/oxford-university-press/the-impact-of-foot-pursuits-on-police-use-of-force-jCLYy0Jwhn
SP - 2355
EP - 2371
VL - 15
IS - 4
DP - DeepDyve
ER -