Police foot-pursuit policies, practices and training: Findings from a national survey

Abstract

Despite increased concerns about the hazards associated with foot pursuits, research on the topic is nascent. This study examines data from a national sample of large law enforcement agencies. Findings indicate that, in spite of the publication of a model foot pursuit policy by the International Association of Chiefs of Police in 2003 and safety concerns raised by law enforcement experts and the media, the majority of agencies surveyed did not have a written foot pursuit policy and most reported engaging in practices deemed risky to the police and the public. Few differences regarding policies, practices and training were observed by agency type and size, but the findings indicate several significant differences by region.

Full text

1
Police foot-pursuit policies, practices and training: Findings from a national survey
(Unpublished Manuscript, rev. 8-10-2015)
Robert J. Kaminski
University of South Carolina
Jeff Rojek
University of Texas at El Paso

2
Abstract
Despite increased concerns about the hazards associated with foot pursuits, research on the topic
is nascent. This study examines data from a national sample of large law enforcement agencies.
Findings indicate that, in spite of the publication of a model foot pursuit policy by the
International Association of Chiefs of Police in 2003 and safety concerns raised by law
enforcement experts and the media, the majority of agencies surveyed did not have a written foot
pursuit policy and most reported engaging in practices deemed risky to the police and the public.
Few differences regarding policies, practices and training were observed by agency type and
size, but the findings indicate several significant differences by region.

3
INTRODUCTION
Although a substantial research literature exists on the hazards associated with police high-
speed motor vehicle pursuits (Alpert & Dunham, 1988; Alpert, Kenney, Dunham, & Smith,
2000; Charles, Falcone & Wells, 1992; Crew 1992, 1999; Lum & Fachner, 2008; Nugent, 1990;
Oechsli, 1992), empirical research on police foot pursuits is in its infancy (Kaminski, 2007;
Kaminski & Alpert, 2013; Kaminski, Rojek, Smith, & Alpert, 2012). Although the risks of death
and serious injury associated with vehicle pursuits are certainly greater, concerns regarding the
hazards of more frequent foot pursuits have increased in recent years (Bobb, 2003, 2005; Bohrer,
Davis, & Garrity, 2000; Graham, 2009; International Association of Chiefs of Police, 2003;
Joyner, 2010; Kaminski, 2007; Pinizzotto, Davis, & Miller, 2002; Plohetski, 2013; Russ, 2010).
These include accidental officer injuries, fatal and nonfatal assaults on officers, and risks to
public safety.
One of the leading concerns expressed by special counsels as well as the media is the
seemingly high number of officer-involved shootings associated with this tactic. For examples, a
review of 33 incidents in which individuals were shot and killed by deputies in DeKalb County
(GA) during 2001-2006 found that 12% involved foot pursuits (Simpson, 2007). Reviews
conducted by the Special Counsel to the Los Angeles County Sheriff's Department (LASD)
found that 22% to 27% of officer-involved shooting incidents involved foot pursuits (Bobb,
2003, 2005). In Austin, Texas, an investigation by the Austin American-Statesman of over two-
dozen police shootings between 2008 and 2013 found that 30% involved foot pursuits (Plohetski,
2013), and an internal review by the Philadelphia Police Department found that 48% of officer-
involved shootings between 1998 and 2003 involved foot pursuits (Graham, 2009).

4
While the likelihood of a shooting during any given foot pursuit is extremely low (Kaminski
et al., 2012), the absolute number of foot pursuit-related shootings is of concern. This realization,
along with other considerations, led the LASD Special Counsel to recommend that the LASD
implement safer foot-pursuit tactics, a restrictive foot pursuit policy, and the creation of a
dedicated foot- and motor-vehicle pursuit database (Bobb, 2003, 2005). The mayor of one town
even implemented a total ban on all pursuits, if only temporarily, after some its officers were
injured accidentally during foot chases and others were involved in motor vehicle pursuit-related
crashes (Adcox, 2009).
Although foot-pursuit related shootings that may result in death or serious injury to officers,
suspects and bystanders is of major concern, it is important to consider other less serious but
more frequent outcomes as well, such as the incidence of less-lethal force and nonfatal injuries to
officers and suspects during foot pursuits. For example, research has shown that that foot
pursuits are associated with a high likelihood of the use of force (Alpert, Kenney & Dunham,
1997; Kaminski et al., 2004) and substantial losses in officer work productivity due to accidental
and force-related nonfatal injuries (Kaminski, 2007).
To reduce the risk of foot pursuit-related fatalities and injuries, in the early 2000s law
enforcement experts and the International Association of Chiefs of Police (IACP) published
recommended guidelines, restrictions and tactics designed to enhance officer and public safety
(Bohrer et al., 2000; International Association of Chiefs of Police, 2003; Pinizzotto et al., 2002).
For example, the IACP’s 2003 model foot pursuit policy recommends that officers terminate foot
pursuits when acting alone, when losing sight of a suspect, when a suspect enters a building,
structure or an isolated area, when communication with dispatch and/or backup officers is lost,
when information is available that would likely lead to apprehension at a later time, and when an

5
officer is unsure of his or her own location or direction of travel. The policy also recommends
that lone officers not try to overtake a fleeing suspect to make an arrest. Rather, the officer
should maintain sight of the suspect and coordinate with backup using a strategy of containment
(setting up a perimeter) and/or other alternatives (aerial surveillance, canine search, area
saturation, etc.). When two or more officers are actively in pursuit of a suspect, the IACP
recommended that they not separate (partner splitting) unless they remain in sight of each other
and maintain communication (see http://www.tacp.org/getdoc/0b850b0f-7570-4a25-a460-
c37996807d8b/Foot_Pursuit_Policy).
Despite the release of the IACP model policy and safety recommendations made by others
(Bobb, 2003, 2005; Bohrer et al., 2000; Pinizzotto et al., 2002), it is unknown how many law
enforcement agencies have since adopted them. To help fill this gap in knowledge, we surveyed
several hundred law enforcement agencies in the United States to determine if they have written
policies governing foot pursuits, whether they practice recommended foot-pursuit tactics, and
whether training specific to foot pursuits is provided. Details regarding the survey methodology
follow a review of the previous research.
PREVIOUS RESEARCH
As noted earlier, there have been few dedicated studies of police foot pursuits. General
studies on pursuits, police use of force and overall hazards associated with police work provide
some limited information regarding foot pursuits. For example, a descriptive analysis of foot
pursuits pursuant to motor-vehicle pursuits in three agencies found that officers used force in
13% or 51 of 407 arrests (Alpert, Kenney & Dunham, 1997). A study of 2,046 arrests in a large
municipal agency in the southeastern United States found that 9% (168 of 1,878) involved
pursuits of any type (foot, motor vehicle or a combination thereof), though most involved foot

6
pursuits. Regression analyses revealed that pursuits increased the odds of officer use of physical
force by 345% (Kaminski, DiGiovanni & Downs, 2004). Other research found that 12-14% of
incidents in which officers were injured involved foot pursuits (Brandl, 1996; Brandl &
Stroshine, 2003, 2012).
In a study of foot pursuits in the Richland County (SC) Sheriff’s Department, 79 of 224
deputies who engaged in foot pursuits reported being intentionally injured in 10% and
accidentally injured in 14% of foot pursuits over a 6-month period (Kaminski, 2007). Career-
based injury estimates were much higher in this study, with 33% of deputies reporting having
been injured intentionally and 43% accidentally one or more times. The cumulative toll on
productivity also appeared high, with intentionally injured deputies reporting missing 273 days
of work and working in a reduced capacity for 358 days. Deputies injured accidentally reported
missing 496 days of work and working in a reduced capacity for 575 days.
A review of nearly 300 foot pursuits in the LASD found that one or more deputies sustained
injuries in 17% of the incidents and that one or more suspects were injured in 60%, a seemingly
high rate of suspect injury. This suggests that suspects who flee on foot substantially increase
their risk of injury. Multiple regression analysis of these data also found that the use of
conducted energy devices (CEDs) was associated with a significant increase in the risk of
suspect injury, a finding is contrary to most prior studies (for a review, see Kaminski, Engel,
Rojek, Smith & Alpert, 2013). The researchers speculated that the use of CEDs on persons
actively fleeing on foot may increase both their risk of injury and the severity of injuries
sustained (Kaminski et al., 2012).
In summary, despite significant concerns raised over the safety of foot pursuits in recent
years, there have been few studies of the hazards associated with this tactic. However, as

7
reviewed above, the limited research to date indicates that 1) substantial numbers of officer-
involved shootings involve foot pursuits, 2) that foot pursuits are associated with a high
likelihood of the use of force generally, 3) that they are associated with substantial productivity
losses due to accidental and assault-related injuries, and that 4) certain use-of-force tactics (e.g.,
CEDs) may substantially increase the odds of injury among suspects actively fleeing police on
foot. It makes sense for police administrators to consider adopting policies or practices to better
manage foot pursuits and for officers to be trained in the use of safer foot-pursuit tactics.
Despite the promotion of safer foot pursuit policies and tactics by the International
Association of Chiefs of Police (2003) and others, however, the extent to which law enforcement
agencies in the U.S. have adopted them is unknown. The present study, therefore, was designed
to help fill this gap in the literature. Specifically, we surveyed agencies that employed 100 or
more sworn officers to determine 1) how many had a written foot-pursuit policy, 2) how
restrictive written policies were, 3) how many agencies practiced recommended tactics, and how
many provided training dedicated to foot pursuits. Further, using crosstab analyses, we explore
whether or not there were significant differences by agency type, size, and geographic location.
The methodological details are presented below.
METHODS
The research team developed a series of questions pertaining to foot-pursuit policies,
practices and training based largely on the IACP model policy and other sources noted earlier. A
survey packet containing the questionnaire, a cover letter and a letter of support from a law
enforcement source was mailed to all agencies in mid-February, 2011. A reminder letter was
mailed two weeks later, and a full survey packet was sent to non-respondents in mid-March. All

8
agencies that did not respond to this second mailing were subsequently mailed another survey
packet a few weeks later.
The sampling frame was based on the Bureau of Justice’s 2003 Law Enforcement
Administrative and Statistics (LEMAS) survey (the 2007 LEMAS survey data were not yet
available), which consisted of 793 agencies with 100 or more full-time sworn personnel
(http://bjs.ojp.usdoj.gov/index.cfm?ty=dcdetail&iid=248).
1
Of the agencies mailed packets, 512
returned a survey for a response rate of 65%. An additional 9 agencies with no regular patrol
function and that did not respond to calls for service were eliminated, leaving 503 agencies for
analysis. Law enforcement agencies from 47 of the 50 states and the District of Columbia were
represented (no surveys were received from agencies located in Vermont, West Virginia, or
Wyoming).
For the analysis, we describe survey responses using basic descriptive statistics. We then
examine relationships between the policy, practice and training variables and agency type, size
and regional location using basic crosstab analyses. (To conserve space, only partial information
from the crosstabs is presented). One consideration for the analysis concerned the relationship
between policies and practices in agencies with all one-officer patrol units and agencies that have
some or most of their sworn personnel assigned to two-officer patrol units. For example,
agencies with all one-officer units may be less likely to prohibit partner-splitting in policy or in
practice. To assess these potential relationships, we regressed the various outcomes on an
indicator variable coded 1 if agencies deployed two-officer patrol units (n = 196 or 40.5%) and
zero if they did not (n = 288 or 59.5%). A series of logistic regression models (not shown)
1
Note that the 2007 LEMAS data became available prior to the submission of this paper for publication. Thus, we
used the more recent data for the number of full-time sworn employees.

9
indicated no significant differences (no p < .10) and we therefore do not consider this issue
further.
Lastly, prior to answering the survey questions, respondents were instructed to read the
following definitions
2
:
Foot pursuit - A foot pursuit is an attempt by an officer to follow or track, on foot, a fleeing person who is
attempting to avoid arrest, detention or observation.
Partner Splitting - "Partner splitting" during a foot pursuit occurs when loss of visual contact, distance or
obstacles separates partners to a degree that they cannot immediately assist each other should a
confrontation take place. For the purposes of this survey, partner splitting does not pertain to lone officers
assigned to static containment positions.
Containment - The establishment of a perimeter to keep a suspect within a specified area and prevent
escape.
RESULTS
Table 1 presents descriptive statistics for the variables used in the analysis. Of the responding
agencies, 65% (325) were municipal police departments and 35% (178) had county-wide
jurisdictions (of the latter, 30% or 152 were county sheriffs’ offices, and 5% or 26 were county
police departments). Regarding agency size, 32% (163) employed between 81 and 150 full-time
sworn, 33% (166) employed between 151 and 300, and 35% (174) employed between 3-1 and
13,336 full-time sworn.
3
Regionally, 14% (72) of responding agencies were located in the
Northeast, 23% (116) were located in the West, 18% (90) in the Midwest, and 45% (225) were
located in the South.
2
We acknowledge the LASD as the source for these definitions.
3
Our initial sample frame was based on the LEMAS 2003 census of large law enforcement agencies (i.e., those with
100 or more full-time sworn). That a small number of agencies linked to the 2007 LEMAS reported having fewer
than 100 sworn suggests that some agencies experienced reductions in manpower between the two waves.

10
Table 1. Descriptive statistics for variables used in the analysis
Variable
Description
Code
N
%
Type
Agency type
0 – county
1 – municipal
178
325
35.4
64.6
Size
Number full-time sworn officers/deputies
1 – small (81-150)
2 – medium (151-300)
3 – large (301-13,336)
163
166
174
32.4
33.0
34.6
Region
U.S. regions
1 – Northeast
2 – West
3 – Midwest
4 – South
72
116
90
225
14.3
23.1
17.9
44.7
Policy
Agency has written policy
0 – no
1 – yes
414
70
85.5
14.5
Restrictive
Restrictiveness of written policy
1 – prohibition
2 – discouragement
3 – restrictive
4 – judgmental
5 – other
0
0
16
54
0
0.0
0.0
22.9
77.1
0.0
Apprehend
Lone officer may individually apprehend
fleeing suspects
0 – no
1 – yes
16
437
3.5
96.5
Containment
Lone officer may pursue but apprehend using
containment only
0 – no
1 – yes
302
62
83.0
17.0
Radio
Officers are required to radio in information
before or within first few seconds of pursuit
0 – no
1 – yes
63
388
14.0
86.0
Communication
Lone officer must cease pursuit if
communication with dispatch is lost
0 – no
1 – yes
298
106
73.8
26.2
Sight
Lone officer must cease pursuit after losing
sight of suspect
0 – no
1 – yes
356
69
83.8
16.2
Structures
Lone officer may pursue suspects into buildings
and other structures
0 – no
1 – yes
67
371
15.3
84.7
Splitting
Officers may engage in partner splitting
0 – no
1 – yes
32
342
8.6
91.4
CED
Officers may deploy conducted energy device
on suspects actively running away
0 – no
1 – yes
130
256
33.7
66.3
Recruits
New recruits provided foot pursuit training
0 – no
1 – yes
246
218
53.0
47.0
In-service
In-service training on foot pursuits provided
0 – no
1 – yes
394
104
79.0
21.0
When asked if their agency had a written policy or procedural directive (hereafter referred to
as a policy) on foot pursuits, only 15% (70) of respondents indicated they did, whereas the vast
majority (86% or 414) indicated they did not. However, 15 agencies without a policy indicated
they were in the process of developing one (not shown). Among respondents reporting that their
agency had a policy, no agency reported that it prohibited all foot pursuits or discouraged all foot

11
pursuits; 23% (16), however, reported that it restricted foot pursuits to specific criteria, while a
majority (77% or 54) indicated it left the decision to pursue to officer/deputy discretion.
The next series of questions in the survey asked about tactical foot pursuit practices,
excluding emergency exceptions, such as imminent danger to officers or civilians. When asked if
lone officers may individually apprehend fleeing suspects, the vast majority indicated a positive
response (97% or 437). When asked if lone officers may individually apprehend suspects using
containment only, relatively few responded in the affirmative (17% or 62). Most agencies,
however, required officers/deputies to radio in relevant information to dispatch before or within
the first few seconds of initiating a foot pursuit (86% or 388) but only about one-fourth of the
respondents indicated that lone officers/deputies must cease a foot pursuit if communication with
dispatch is lost (26% or 106). Few agencies required lone officers/deputies to discontinue a
pursuit after losing sight of a suspect (16% or 69), and the vast majority allowed them to pursue
suspects into buildings or other structures (85% or 371). The vast majority of agencies also
allowed partner splitting
4
during foot pursuits (91% or 342), and most allowed the use of a
conducted energy device on a suspect actively running away (66% or 256).
A final set of questions queried respondents about foot pursuit training. When asked if new
recruits received training from field-training officers (FTOs), frontline sergeants, or via other
means during their shifts, 47% (218) indicated they did. In-service training dedicated to foot
pursuits, however, was less common (21% or 105 agencies). Although not presented in Table 1,
when asked how often training was provided, only 16% (22) indicated annually and 8% (11)
indicated semiannually, while the vast majority (76% or 107) indicated training was provided
4
We suspected that responses to the question on partner splitting would vary by whether or not agencies did and did
not deploy two-officer patrol units. However, additional analysis (not shown) indicated that this was not the case;
92.5% of agencies that deployed two-officer patrols allowed partner splitting while 90.1% of agencies that deployed
only one-officer patrols did so.

12
based on some “other” schedule (e.g., as part of other training, during squad meetings, or most
commonly “as needed”).
Of the agencies that provided training, nearly all (96% or 95) indicated that they provided
classroom-based training and nearly two-thirds indicated they provided physical-based training
(67% or 52). Fifty-one or 66% of the agencies indicated they provided both classroom and
physical in-service training to their officers/deputies. As part of agency in-service defensive
tactics and firearm training, some agencies required officers/deputies to run prior to simulating a
physical struggle with another officer/deputy and/or prior to target practice with a firearm to help
prepare them for resistive and deadly encounters in the field. Specifically, of 409 respondents,
51% (208) required officers/deputies to run prior to simulating a physical struggle and 49% (201)
required officers to run prior to target practice.
Crosstab Analysis
We next examine the relationship between agency type, size, and regional location and
agency policy, practice and training variables using crosstab analyses. The Pearson chi-square
statistic is used to test for independence and Phi (for 2x2 tables) and Cramer’s V (for higher
order tables) are used as measures of the strength of relationships. A few statistically significant
relationships were observed between agency type and agency practices (Table 2) as well as
between agency region and the policy, practice and training variables (Table 4). No statistically
significant associations, however, were found between agency size and any of the outcome
variables (Table 3). Specific findings are discussed below.
Table 2 presents the results of the crosstab analysis between agency type and the variables of
interest. As can be seen, only two statistically significant relationships are observed for these
comparisons. There is a modest but significant association (p = .01) between agency type and

13
Containment (χ² = 7.12; ϕ = .172). Specifically, sheriffs’ agencies were nearly twice as likely as
municipal agencies to report allowing lone officers to pursue suspects fleeing on foot but
apprehend them using only the tactic of containment (24.4% versus 13.3%, respectively).
Sheriffs’ agencies (75.6%) were also significantly (p = .01) more likely than municipal agencies
(61.4%) to authorize the use of CEDs on suspects actively running away from officers (χ² = 7.23;
ϕ = .145).
Table 2. Crosstab analysis of policy, practice and training variables by agency type
Variable
Agency Type
Statistics
Policy
Sheriff
Municipal
% Yes
n
% Yes
n
χ²
ϕ
12.0
20
15.8
50
0.692
.018
Apprehend
96.9
156
96.2
281
0.005
.003
Containment
24.4
30
13.3
32
7.116**
.173
Radio
87.3
137
85.4
251
2.889
.080
Communication
30.5
43
24.0
63
1.863
.068
Sight
19.0
28
14.7
41
1.463
.059
Structures
81.0
124
86.7
247
2.665
.078
Splitting
89.3
109
92.5
233
1.103
.054
CED
75.6
102
61.4
154
7.926**
.145
In-service
18.1
32
22.4
72
1.762
.059
Post-academy
43.4
69
48.9
149
0.316
.026
* = p = .05, ** = p = .01, *** = p ≤ .001; χ² = Pearson chi-square; ϕ = Phi
No statistically significant relationships were observed between the variables and agency size
in Table 3. However, to test for possible linear and nonlinear relationships between Agency Size
(the number of full-time sworn officers) and the various outcomes, we estimated a series of
logistic regression models with and without the number of sworn officers squared using robust
standard errors (not shown). Only one statistically significant relationship emerged at the p = .05

14
level. For each 100 additional sworn officers, the odds that agencies provided new recruits post-
academy foot-pursuit training increased by 2.4% (OR = 1.024, p = .049).
5
Table 3. Crosstab analysis of policy, practice and training variables by agency size
Variable
Agency Size
Statistics
Policy
81-150
151-300
300-13,336
% Yes
n
% Yes
N
% Yes
n
χ²
V
17.3
27
13.8
22
12.4
21
1.638
.058
Apprehend
96.6
142
94.6
141
98.1
154
2.694
.077
Containment
22.4
26
13.0
15
15.8
21
3.817
.102
Radio
87.7
128
88.4
130
82.3
130
2.885
.080
Communication
26.9
35
29.5
38
22.8
33
1.629
.064
Sight
18.7
26
15.2
21
14.9
22
0.933
.047
Structures
81.6
115
86.0
123
86.4
133
1.592
.060
Splitting
87.5
105
94.0
110
92.7
127
3.652
.099
CED
64.7
77
65.9
83
68.1
96
0.347
.030
In-service
16.0
26
21.2
35
25.1
43
4.183
.092
Post-academy
43.7
66
49.7
79
47.4
73
1.127
.049
* = p = .05, ** = p = .01, *** = p ≤ .001; χ² = Pearson chi-square; V = Cramer’s V
Results from the crosstab analyses between agency regional location and the policy, practice
and training variables (presented in Table 4) indicate several significant relationships. Agencies
located in the West were much more likely (27.6%) than those located in other regions to
indicate they had a written foot-pursuit policy (V=.216, p ≤ .000). With respect to agency
practices, agencies in the Northeast were more likely than those located in the West to allow lone
officers to apprehend a fleeing suspect (98.4% versus 91.7%, respectively), though only
somewhat more likely than those located in the Midwest (97.5%) and the South (98.0%) to do so
(V=.145, p =.024). Agencies in the Northeast were the most likely to allow partner splitting
(90.0%) with those in the West (83.0%) the least likely to do so (V=.179, p =.008). Agencies in
the South were substantially more likely (75.6%) to allow the use of CEDs on actively fleeing
5
Other relationships were observed that would be considered statistically significant only at the p = .10 level. The
odds that agencies required officers to radio in information before or within first few seconds of pursuit increased
1.6% with each 100 additional officers (OR= 1.016, p = .078). In addition, each 100 additional officers was
associated with a 12.4% increase in the odds that agencies allowed partner splitting (OR = 1.124; p = .079).

15
suspects than those located in the other regions, with those in the Northeast the least likely to do
so (50.0%) (V = .194, p = .01).
Table 4. Crosstab analysis of policy, practice and training variables by agency region
Variable
Agency Region
Statistics
Northeast
West
Midwest
South
% Yes
n
% Yes
n
% Yes
n
% Yes
n
χ²
V
Policy
11.1
8
27.6
32
13.6
12
8.7
18
22.525***
.216
Apprehend
98.4
60
91.7
100
97.5
79
98.0
198
9.479*
.145
Containment
16.7
9
16.1
15
14.3
10
19.0
28
0.855
.048
Radio
87.5
56
79.2
84
86.3
69
89.1
179
5.709
.113
Communication
16.7
9
26.3
26
24.7
18
29.8
53
3.801
.097
Sight
8.6
5
18.4
19
14.5
11
18.1
34
3.490
.091
Structures
91.5
54
81.1
86
84.6
66
84.6
165
3.164
.085
Splitting
90.0
45
83.0
73
95.5
64
94.7
160
11.918**
.179
CED
50.0
20
58.3
56
63.5
47
75.6
133
14.512**
.194
In-service
9.9
7
31.3
36
19.1
17
19.7
44
13.132**
.162
Post-academy
25.0
17
66.0
70
37.5
30
48.1
101
31.636***
.261
* = p = .05, ** = p = .01, *** = p ≤ .001; χ² = Pearson chi-square; V = Cramer’s V
(V=.194, p=.002). Regarding training, agencies in the West were the most likely to provide in-
service training dedicated specifically to foot pursuits (31.3%) while those in the Northeast
(9.9%) were the least likely to do so (V =.162, p=.004). The pattern regarding post-academy
training dedicated to foot pursuits is different, however, with agencies in the South (48.1%)
being the most likely to report that their officers receive additional training while agencies in the
Northeast (25.0%) were the least likely to do so (V=.261, p=.000).
CONCLUSION
Despite efforts by the IACP (International Association of Chiefs of Police, 2003) and others
(Bobb, 2003, 2004; Bohrer et al., 2000; Pinizzotto et al., 2002) to promote safer foot-pursuit
policies and practices, the results from the national survey indicate little progress has been made.
Most responding agencies did not have a written foot pursuit policy and no agency surveyed had

16
a policy that discouraged foot pursuits. Some reported that they restricted foot pursuit to specific
criteria, but the vast majority of agencies left the decision to pursue to officer/deputy discretion.
Regarding tactics, excluding emergency exceptions (e.g., imminent danger to officers or
civilians), the vast majority of agencies allowed lone officers/deputies to individually close and
apprehend fleeing suspects, a tactic that the IACP model policy does not recommend. Similarly,
few agencies required lone officers/deputies to use containment as a tactic, which the IACP
model policy promotes. On a positive note, most agencies required officers/deputies to radio in
relevant information to dispatch before or within the first few seconds of initiating a foot pursuit,
but relatively few agencies required officers/deputies to terminate foot pursuits when
communication with dispatch is lost, as suggested by the IACP. Further, few agencies required
lone officers/deputies to discontinue a pursuit after losing sight of a suspect, most allowed them
to pursue suspects into buildings or other structures, and the vast majority allowed partner
splitting, all risky tactics according to the IACP and other sources. Although not addressed in the
IACP model policy, most agencies allowed the use of a conducted energy device on suspects
actively running away. Though more research is needed, forward momentum combined with a
lack of body control among “tased” suspects actively fleeing may contribute to higher injury
rates and/or more serious injuries (Kaminski et al., 2012).
In terms of training specific to foot pursuits, nearly half of responding agencies indicated
new recruits received some form of training from FTOs, frontline sergeants or via other means
during their shifts. However, it was our sense that this may often be ad hoc and not as systematic
as it perhaps could be. Of concern was that relatively few agencies provided in-service training
dedicated to foot pursuits, and fewer still provided it on an annual or semiannual basis. Most
typically, training was provided on an “as needed” basis, during squad meetings, line ups, etc.

17
Among agencies that provided in-service training, nearly all provided classroom-based training,
with substantially fewer providing physical-based training. Some promising findings were that
more than half of the agencies required officers/deputies to run prior to simulating a physical
struggle (as part of defensive tactics training) and to run prior to target practice (as part of
firearms training). Such “reality-based” training may be particularly important for officer and
public safety.
Finally, we were interested to see how responses varied by agency type (county vs.
municipal), size (small, medium and large) and regional location. Although in some cases there
was substantial variability across these factors percentage-wise, few of the differences were
statistically significant. Based on the crosstab analyses, no significant differences were observed
between agency size and the various outcomes, though a logistic regression model indicated that
larger agencies were more likely to provide post-academy foot-pursuit training for new recruits.
Additionally, there were few significant differences by agency type. County agencies, however,
were more likely than municipal agencies to report using containment as a tactic (perhaps
because of geographic structural differences) that may reduce the risk of injury to officers, but
they were significantly more likely than municipal agencies to allow the use of CEDs on actively
fleeing suspects, which may increase the risk of injury to suspects (Kaminski et al., 2012).
Unlike agency size and type, a number of significant regional differences emerged, especially
between the West and the other regions. While there are some exceptions (noted below),
agencies in the West tended to be more progressive than those located in other regions.
Specifically, Western agencies were significantly: 1) more likely than agencies in other regions
to have a written foot pursuit policy, 2) more likely to provide post-academy foot pursuit training
to new recruits, 3) more likely to provide in-service training on foot pursuits, 4) less likely to

18
allow partner splitting, and 5) less likely to allow lone officers to close and apprehend a fleeing
suspect. Agencies in the Northeast, however were the least likely to allow the use of CEDs on
actively fleeing suspects.
In conclusion, while we do not necessarily agree with highly restrictive foot-pursuit policies,
given police and public safety concerns regarding foot pursuits, we believe law enforcement
agencies should adopt at least some of the elements of the IACP’s model foot pursuit policy and
safer practices as suggested by others (Bohrer et al., 2000; Pinizzotto et al., 2002). Furthermore,
we recommend that law enforcement agencies provide training specific to foot pursuits as part of
regular in-service training rather than on an ad-hoc bases. Though additional study is needed, this
could lead to reductions in officer-involved shootings and fewer injuries and fatalities among law
enforcement officers as well as citizens.

19
REFERENCES
Adcox, S. (2009, September 25). Welford mayor revokes no-foot-chase call. Associated Press,
Retrieved September 26, 2009, from http://www.thestate.com/breaking/story/957112.html.
Alpert, G. P., 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.
Alpert, G. P., Dennis K., R. Dunham & Smith, W. (2000). Police pursuits: What we know.
Washington, DC: Police Executive Research Forum.
Bohrer, S., Davis, E., & Garrity, T. (2000). Establishing a foot pursuit policy: Running into
danger. FBI Law Enforcement Bulletin, 69(5),10-15.
Brandl, S. G. (1996). In the line of duty: A descriptive analysis of police assaults and
accidents. Journal of Criminal Justice, 24(3), 255-64.
Brandl, S. G., & Stroshine, M. S. (2003). Toward an understanding of the physical
hazards of police work. Police Quarterly, 6(2),172-91.
Brandl, S. G. & Stroshine, M. S. (2012). The physical hazards of police work revisited. Police
Quarterly, published online before print May 29, 2012, doi: 10.1177/1098611112447757.
Charles, M. T., D. N. Falcone, &Wells, E. 1992. Police Pursuits in Pursuit of Policy: The
Pursuit Issue, Legal and Literature Review, and Empirical Study. Washington D.C.:
AAA Foundation for Traffic Safety.
Crew, R. E. 1992. An Effective Strategy for Hot Pursuit: Some Evidence from Houston.
American Journal of Police, 11, 89-95.

20
Crew, R. E. & Hart, R. A. (1999). Assessing the Value of Police Pursuit. Policing: An
International Journal of Police Strategies and Management, 22, 58-73.
Glasser, E. (2102). FHP Trooper on Tasing handcuffed woman Danielle Maudsley: ‘I needed to
stop her.’ 10 News, 17, Feb. Retrieved July 1, 2012 from
http://www.wtsp.com/news/article/239247/8/FHP-Trooper-on-Tasing-handcuffed-
woman-Danielle-Maudsley-I-needed-to-stop-her.
Graham, T. (2009). Lawsuit shines spotlight on police foot pursuits. Philadelphia Inquirer, 16,
Nov. Retrieved December 31, 2009 from
http://www.philly.com/philly/news/homepage/20091116_Lawsuit_shines_spotlight_on_p
olice_foot_pursuits.html?viewAll=y.
Joyner, C. (2010). Chases called most dangerous part of police job. ClarionLedger.com, 15, Aug.
Retrieved August 30, 2010 from
http://www.clarionledger.com/article/20100815/NEWS/8150360.
International Association of Chiefs of Police. (2003). Foot Pursuit Model Policy. Retrieved Feb.
26, 2011 from http://www.tacp.org/getdoc/0b850b0f-7570-4a25-a460-
c37996807d8b/Foot_Pursuit_Policy.
Kaminski, R. J. (2010). A descriptive analysis of foot pursuits in the Los Angeles County Sheriff's
Department. Retrieved from the University of South Carolina, Department of
Criminology & Criminal Justice Web site: http://www.cas.sc.edu/crju/research.html.
Kaminski, R. J. (2006). An exploratory study of police foot pursuits in the Richland County
Sheriff’s Department. Retrieved from the University of South Carolina, Department of
Criminology & Criminal Justice Web site: http://www.cas.sc.edu/crju/footsurvey.html.

21
Kaminski, R. J. Police foot pursuits and officer safety. (2007). Law Enforcement Executive
Forum, 7(3), 59-72.
Kaminski, R. J. & Alpert, G.P. (2013). Recent findings on police foot pursuits. Research in brief.
The Police Chief, 80(1), 14.
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.
Kaminski, R. J., Engel, R.S., Rojek, J., Smith, M. R., & Alpert, G.P. (2013). A quantum of force:
The consequences of measuring routine conducted energy device punctures as injuries.
Justice Quarterly, DOI:10.1080/07418825.2013.788729.
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.
Lum, C. & Fachner, G. (2008). Police pursuits in an age of innovation and reform: The IACP
police pursuit database. Alexandria, VA: International Association of Chiefs of Police.
Nugent, H. (1990). Restrictive policies for high-speed police pursuits. Washington, D.C.: U.S.
Department of Justice, National Institute of Justice.
Oechsli, S. (1992). Kentucky State Police pursuit study: 1989-1992. Frankfort, KY: Kentucky
State Police.
Pinizzotto, A. J., Davis, E. F., & Miller, C. E., III. (2002). Escape from the killing zone. The FBI
Law Enforcement Bulletin, 71(3), 1-13.
Plohetski, T. (2013). Dangerous pursuit: Austin police foot chases can lead to injuries, shootings.
Austin American-Statesman, 8 Sept. Retrieved Sept. 15, 2013 from
http://kokefm.com/tony-plohetsky-from-the-american-statesman-joined-us-for-face-off/.

22
Pfeifer, S. (2007). Deputies lament limits on foot chases. Los Angeles Times, 1 Jan. Retrieved
Feb. 2, 2007 from http://www.latimes.com/news/local/la-me-
chase21jan21,1,5092198.story?page=3&coll=la-default-underdog.
Russ, V. (2010). Overkill in park shooting? Cops say they're investigating. Philadelphia Daily
News, 6, May. Retrieved May 6, 2010 from
http://www.philly.com/philly/hp/news_update/92943419.html?viewAll=y.