G Model

1 RESUS 6741 1–6 ARTICLE IN PRESS

Resuscitation xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Resuscitation
journal homepage: www.elsevier.com/locate/resuscitation

2 Simulation and education

3 Viewing a brief chest-compression-only CPR video improves

4 bystander CPR performance and responsiveness in high school
5 students: A cluster randomized trial夽
6 Q1 Daniel L. Beskind a,∗ , Uwe Stolz a , Rebecca Thiede b , Riley Hoyer b , Whitney Burns b ,
7 Jeffrey Brown b , Melissa Ludgate b , Timothy Tiutan b , Romy Shane b , Deven McMorrow b ,
8 Michael Pleasants a , Ashish R. Panchal c
a
9 The University of Arizona, Department of Emergency Medicine, USA
10 Q2 b The University of Arizona College of Medicine, USA
c
11 The Center for EMS, The Ohio State University Wexner Medical Center, USA
12

13
31 a r t i c l e i n f o a b s t r a c t
14
15 Article history: Background: CPR training in schools is a public health initiative to improve out of hospital cardiac arrest
16 Received 9 October 2015 (OHCA) survival. It is unclear whether brief video training in students improves CPR quality and respon-
17 Received in revised form 12 February 2016 siveness and skills retention.
18 Accepted 30 March 2016
Objectives: Determine if a brief video is as effective as classroom instruction for chest compression-only
19 (CCO) CPR training in high school students.
20 Keywords:
Methods: This was a prospective cluster-randomized controlled trial with three study arms: control (sham
21 Cardiac arrest
video), brief video (BV), and CCO-CPR class. Students were randomized and clustered based on their
22 Q3 Ultra-brief video
23 Brief video
classrooms and evaluated using a standardized OHCA scenario measuring CPR quality (compression rate,
24 Lay bystander depth, hands-off time) and responsiveness (calling 911, time to calling 911, starting compressions within
25 Bystander CPR 2 min). Data was collected at baseline, post-intervention and 2 months. Generalized linear mixed models
26 CPR teaching were used to analyze outcome data, accounting for repeated measures for each individual and clustering
27 Public by class.
28 CPR responsiveness Results: 179 students (14–18 years) were consented in 7 classrooms (clusters). At post-intervention
29 CPR performance and 2 months, BV and CCO class students called 911 more frequently and sooner, started chest com-
30 High-school
pressions earlier, and had improved chest compression rates and hands-off time compared to baseline.
Chest compression depth improved significantly from baseline in the CCO class, but not in the BV group
post-intervention and at 2 months.
Conclusions: Brief CPR video training resulted in improved CPR quality and responsiveness in high school
students. Compression depth only improved with traditional class training. This suggests brief educa-
tional interventions are beneficial to improve CPR responsiveness but psychomotor training is important
for CPR quality.
© 2016 Elsevier Ireland Ltd. All rights reserved.

32 Introduction in 30–50% of cardiac arrests in the United States, with actual rates 36

Q4 varying greatly by geography.2,5 Significant efforts have been made 37

33 Bystander CPR performance is an important aspect for improv- to increase the rate of bystander CPR performance with Hands-only 38

34 ing neurologically intact survival from out of hospital cardiac CPR training for lay bystanders,3,6 dispatcher assisted CPR perfor- 39

35 arrest (OHCA).1–4 Unfortunately, bystander CPR is only provided mance,7 and even programs to increase CPR performance in high 40

risk communities.8 One approach involves the integration of teach- 41

ing CPR in high schools. Currently, 20 states have passed legislation 42

to teach CPR in schools and made learning CPR a requirement for 43

夽 A Spanish translated version of the abstract of this article appears as Appendix
graduation.9 Further, it is unclear whether training CPR in schools 44
in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2016.03.022.
∗ Corresponding author at: University of Arizona, Arizona Emergency Medicine is associated with good long-term knowledge and skills retention. 45

Research Center, 1501 North Campbell Avenue, Tucson, AZ 85724, USA. Recent studies have demonstrated that following an ultra- 46

E-mail address: dbeskind@aemrc.arizona.edu (D.L. Beskind). brief (60 s) video in chest compression only CPR (CCO-CPR), 47

http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
0300-9572/© 2016 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
 G Model
RESUS 6741 1–6 ARTICLE IN PRESS
2 D.L. Beskind et al. / Resuscitation xxx (2016) xxx–xxx

48 bystanders were more likely to attempt CPR and demonstrated 1. Baseline: for baseline (i.e., pre-intervention) CPR measurements. 108

49 superior CPR skills compared to untrained laypersons.10 Further- 2. Post intervention: immediate post baseline measures following 109

50 more, these individuals were able to retain some of their skills each of the three interventions (control [sham video], BV, or CCO 110

51 at a 2-month follow up. This suggests that CPR videos may allow class). 111

52 for increased exposure of lay bystanders to CPR training in mass 3. 2-month follow up: 2 months post CCO-CPR teaching to assess 112

53 gathering venues. It suggests that brief instructional CPR videos CCO-CPR knowledge retention. 113

54 can be shown to students while in school to enhance CPR training

55 retention. Baseline 114
56 This study investigated the effectiveness of a brief video to teach Following study arm assignment, participants were taken indi- 115
57 CCO-CPR in a public high school setting. Our goals were to deter- vidually to a testing room for baseline measurements of CPR 116
58 mine whether: (1) a brief video is an effective instrument to teach performance and collection of demographic data before the inter- 117
59 high school students CCO-CPR, (2) if there are differences in respon- vention. Upon entering the room, the subjects were presented with 118
60 siveness when a brief video is shown compared to a 20-min training a mannequin on the floor and read a standardized scenario: “You are 119
61 course, and (3) whether there is a difference in retention of skills at Park Place Mall when an approximately 45-year old female collapses 120
62 two months following each training modality. in front of you. You are the nearest person to the collapsed female. Act 121

out what you would do in this scenario. You have two minutes to per- 122

63 Methods form the scenario, and we cannot answer any questions at this time. 123

We will let you know when the two minutes are up.” The evaluators 124

64Q5 Setting and population were instructed not to give any verbal or non-verbal cues to the 125

participants during the 2-min testing period other than stating to 126

65 This was a prospective, cluster-randomized controlled trial of “do what you think is best.” 127

66 chest compression only CPR (CCO-CPR) with three study arms (1) Student’s performance was measured using a Laerdal 128

67 CCO training by brief video (BV), (2) CCO training by a 20 min class- SkillreporterTM mannequin including endpoints for respon- 129

68 room training (CCO class), and (3) a control group who watched siveness and CPR performance. Student demographic information 130

69 a college recruiting video (control). The study was approved by was collected once the 2-min scenario was completed to prevent 131

70 the department Institutional Review Board authority at the Uni- cognitive bias. 132

71 versity of Arizona. Participation was voluntary and no personal

72 incentive was given for participation. We obtained parental consent Post-intervention 133
73 per school policy to allow the students to participate. Immediately following the initial presentation of the scenario 134
74 Students at one Arizona High School where CPR was not taught and the baseline measurements, students were then given the 135
75 were enrolled in the study. Student’s eligibility criteria were they intervention as a group based on the previous randomization. The 136
76 had to be enrolled in a health class during the study period and that control group was shown a 1-min University of Arizona recruitment 137
77 both they and their parents provided informed consent. Two hun- “sham” video. The BV group was shown a brief (1.5 min) CCO-CPR 138
78 dred and six students, ages 14–18 years, enrolled in one of seven video that illustrated the steps to perform CCO-CPR. The video was 139
79 Health classes (clusters) were randomized by class into one of the produced by the University of Arizona Sarver Heart Center and 140
80 three study arms: control, BV, or CCO class. Two classes were ran- funded by the Steven M. Gootter Foundation.11 Finally, the CCO 141
81 domized into the control arm, two to the CCO class arm, and three class group participated in a 20-min CCO-CPR class by a Univer- 142
82 into the BV arm. Attempts were made to best match the class sizes sity of Arizona Sarver Heart Center Emergency Medical Technician 143
83 so that each arm would have approximately the same numbers instructor. The instructional class included a power-point presen- 144
84 of students since the class sizes varied. Exclusion criteria included tation and an in-person demonstration of CCO-CPR with hands-on 145
85 not speaking English, absence from class on days the study was training with the mannequin. 146
86 conducted, not providing written consent, or inability to obtain a Following the intervention for each study arm, the participants 147
87 signed consent from a parent or guardian. were taken back (individually) to the same room, read the same 148
88 The intervention occurred at the cluster (class) level and out- scenario used during pre-testing evaluation, and had 2 min to act 149
89 comes were measured at the individual level. We did not conduct a out what they would do in the emergency situation. The same per- 150
90 formal sample size and power calculation, but determined a priori formance measures on bystander responsiveness and performance 151
91 that our study likely had sufficient power because we assumed an were again recorded on each subject. 152
92 almost negligible intra-class correlation and our final anticipated For the control group only, which received a sham-video inter- 153
93 sample size per arm would be similar to a previous (non-cluster) vention, we provided CCO-CPR training for all control group 154
94 study with a similar design.11 We assumed a negligible intra-class students after their CPR performance testing using the Brief CCO- 155
95 correlation because health classes are mandatory for each high CPR Video and an in-person demonstration. However, we did not 156
96 school student and assignment to a particular class is not based retest the control group immediately following this CCO-CPR train- 157
97 on factors that might lead to bias such as age, class, or academic ing. Each student in the control group had the opportunity to 158
98 rank/performance. Our final analysis did however, account for clus- practice chest compressions on a LaerdalTM mannequin (Laerdal 159
99 tering and we did estimate the intra-class correlation. Unequal Medical Corporation, Stavanger, Norway) 160
100 cluster sizes were assumed because of the unequal number of stu-
101 dents in each health class. Study arm allocation was concealed at
Two-month follow up 161
102 the cluster level and one of the researchers generated the random
Two months after the intervention, participants from each of 162
103 allocation.
the three study arms were re-evaluated on the CCO-CPR perfor- 163

mance measures. Participants were taken to the same room and 164
104 Study design read the same scenario as before, but they were not made aware 165

ahead of time that they would be re-tested. Once the follow-up 166
105 Each of seven classes was assigned to one of three study clusters testing was complete, student questions concerning CCO-CPR were 167
106 (control, BV, or CCO class). Each student was evaluated three times answered. The initial control group, which was tested before and 168
107 in this study process from February to April 2014: after a sham intervention, did eventually receive CCO-CPR training 169

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
 G Model
RESUS 6741 1–6 ARTICLE IN PRESS
D.L. Beskind et al. / Resuscitation xxx (2016) xxx–xxx 3

170 as noted above, and thus this group did not represent a true control
171 group for the two-month follow-up testing.

172 Data collection

173 The primary outcomes measured in this evaluation were

174 bystander responsiveness [calling 911 within 2 min (yes/no), time
175 to calling 911 (s), starting chest compressions within 2 min
176 (yes/no), time to starting chest compressions (s)], asking for an AED
177 within 2 min (yes/no), time to ask for an AED and CPR performance
178 measures [total number of chest compressions performed, mean
179 chest compression depth in millimeters, chest compression rate
180 (CCs/minute), hands-off time (s)]. Demographic data included age,
181 grade, CCO-CPR certified (yes/no), and whether they were AHA-
182 CPR certified (yes/no). All time measurements were assessed by
183 marking time points in the Laerdal SkillreporterTM software by the
184 examiner.
185 Hands-off time was measured as the total amount of time the
186 participant stopped doing compressions once compressions were
187 initiated. This included stopping compressions for rescue breath-
188 ing, fatigue, calling 911, etc. All chest compression performance
189 data were directly measured by the Laerdal SkillreporterTM .

190 Statistical analysis

191 Data analysis was performed by a statistician blinded to both

192 study arm assignment and time of data collection (pre/post/2-
193 month). The statistician was unblinded only after the analysis was
194 complete. Categorical data are presented as percent and 95% con-
195 fidence intervals (CIs), parametric data are presented as means
196 and 95% CIs, and non-parametric data as medians and interquartile
197 ranges (IQRs). Generalized linear mixed models were used to ana-
198 lyze outcome data, accounting for repeated measures across the
199 three time points for each individual participant and clustering
200 by health class. Binary outcomes were assessed using the binary Fig. 1. Study enrollment inclusion/exclusion flowchart (Consort).
201 family with a logit link function, parametric continuous data were
202 assessed using a Gaussian family and link function, and all time There was no difference in CCO training or AHA training prior to 229
203 measurements were assessed using the negative binomial family the study between the groups (Table 1). 230
204 and link function. Individual statistical models were developed for Table 2 shows results for bystander responsiveness. The propor- 231
205 each outcome variable and included health class and individuals tion of participants that called 911 and started compressions within 232
206 nested within health class as random effects. Covariates included 2 min along with overall time to call 911 (s) significantly improved 233
207 time of measurement (baseline; post intervention; 2-month follow from baseline following the BV and CCO class (post-intervention). 234
208 up), study arm/intervention (control; BV, CCO class), as well as an This significant improvement in BV and CCO class responsiveness 235
209 a priori determined time × study arm interaction variable. The pri- was maintained at 2 months compared to baseline. In the con- 236
210 mary goal of the study, to compare baseline to post intervention, trol group, consistent with their post-intervention CCO training, 237
211 baseline to 2-month follow-up, and post-intervention to 2-month 2-month follow-up responsiveness improved for the proportion 238
212 follow-up, was assessed individually for each study arm using a who called 911 and started compressions within 2 min as well as 239
213 Bonferroni adjusted p-value of 0.0056 (accounting for 3 tests for the overall time to calling 911 (s). 240
214 each arm = 9 total tests) for each outcome variable. Table 3 shows the CPR performance metrics for the three groups. 241

Chest compression rate (bpm) improved significantly in the BV 242

215 Results and CCO class from baseline to post-intervention and was main- 243

tained at 2 months. Chest compression depth (mm) improved 244

216 A total of 206 students in 7 health classes were asked to par- significantly in the CCO class from baseline to post-intervention 245

217 ticipate in the study. Students (ages 14–18 years) were consented and was maintained at 2 months but did not in the BV group. 246

218 and the final population was 179 with 27 excluded due to lack of Hands-off time (s) decreased significantly in both the BV and 247

219 parental consent. Students were divided into the three study arms CCO class from baseline to post-intervention. At 2 months, the 248

220 as noted in Fig. 1 with some students lost to evaluation. BV group continued to improve whereas the improvement in 249

221 Some students were absent in classes for the post-testing and 2 hands-off time for the CCO class was maintained at 2 months. For 250

222 month follow-up. the control group, compression rate and depth did not improve 251

223 Subject demographics are shown in Table 1 separated by study from baseline to post-intervention. Hands-off time improved from 252

224 arm: control, brief video (BV), CCO class (chest compression only baseline to post-intervention (baseline = 47.7–27.5), respectively 253

225 class). Average age was not different for the three groups (con- (Table 3). Following the control group CCO training after the con- 254

226 trol = 14.9; BV = 15.1; CCO class = 14.9; p = 0.34). The high school trol intervention (“sham” video), CPR metrics improved for all 255

227 grades for the students were different between treatment arms parameters (compression rate, depth and hands-off time) from 256

228 with the majority of students in each arm being Grade 9 (p = 0.04). post-intervention to 2-month follow-up (Table 3). 257

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
 G Model
RESUS 6741 1–6 ARTICLE IN PRESS
4 D.L. Beskind et al. / Resuscitation xxx (2016) xxx–xxx

Table 1
Demographics for participants in three intervention arms of study (control, BV, CCO class).

Control Brief video CCO class P value

Age (years), mean + SD 14.9 + 0.9 15.1 + 0.9 14.9 + 1.1 0.34
Grade, frequency (%)
Grade 9 38/54 (70%) 45/69 (65%) 47/56 (84%) 0.04
Grade 10 9/54 (17%) 9/69 (13%) 4/56 (7%)
Grade 11 5/54 (9%) 14/69 (20%) 2/56 (4%)
Grade 12 2/54 (4%) 1/69 (1%) 3/56 (5%)
CCO trained, frequency (%) 3/54 (6%) 4/69 (6%) 2/56 (4%) 0.83
AHA certified, frequency (%) 8/54 (15%) 11/69 (16%) 5/56 (9%) 0.49

Abbreviations: AHA, American Heart Association; CCO, chest compression only; SD, standard deviation.

Table 2
Bystander responsiveness (% call 911, time to call 911, and % starting compressions) for the three study arms.

Outcomes Intervention

Control* Brief video CCO class

Call 911 (%)

Baseline, % (95% CI) 33.3 (21.1–47.5)a 49.3 (37.0–61.6)a 39.3 (26.5–53.2)a
Post intervention, % (95% CI) 37.0 (24.3–51.3)a 84.5 (72.6–92.7)b 75.0 (61.6–85.6)b
2-month follow-up, % (95% CI) 56.3 (41.1–70.5)b 78.3 (65.8–87.9)b 74.5 (61.0–85.3)b
Call 911 (s)
Baseline, mean (95% CI) 23.8 (17.1–30.6)a 14.5 (11.0–18.1)a 16.0 (11.7–20.3)a
Post intervention, mean (95% CI) 9.5 (6.5–12.5)b 8.5 (6.6–10.5)b 10.9 (8.3–13.5)b
2-month follow-up, mean (95% CI) 8.4 (5.8–11.0)b 7.3 (5.6–9.1)b 5.5 (4.0–7.0)c
Start compressions with 2 min
Baseline, (%) (95% CI) 74.1 (60.3–85.0)a 72.4 (60.4–82.5)a 69.6 (55.9–81.2)a
Post intervention, (%) (95% CI) 74.1 (60.3–85.0)a 100 (93.8–100)b 98.2 (90.4–100)b
2-month follow-up, (%) (95% CI) 93.8 (82.8–98.7)b 98.3 (91.1–100)b 100 (93.5–100)b

Letters “a”, “b”, and “c” indicate statistically significant differences (p < 0.0056) for metrics between groups within interventions (baseline, post-intervention, 2-month follow-
up) such that groups with different letters are statistically different while groups with the same letter are not. Alpha adjusted for multiple comparisons using the Bonferroni
method. All comparisons were made within interventions.
*
All subjects in the control arm were given CCO training after the post-intervention measurement.

Table 3
CPR performance (chest compression rate, depth and hands-off time) for the three study arms.

Control* Brief video CCO class

Chest compression rate (bpm)

Baseline, bpm (95% CI) 76 (68–85)a 74 (66–82)a 77 (70–84)a
Post intervention, bpm (95% CI) 82 (73–90)a 104 (97–112)b 93 (86–99)b
2-month follow-up bpm (95% CI) 103 (95–112)b 99 (92–107)b 97 (90–103)b
Chest compression depth (mm)
Baseline, mm (95% CI) 29.5 (25.9–33.1)a 28.4 (25.3–31.5)a 27.6 (24.1–31.2)a
Post intervention, mm (95% CI) 28.3 (24.6–31.9)a 32.3 (29.4–35.3)a 34.0 (30.8–37.3)b
2-month follow-up, mm (95% CI) 35.7 (32.2–39.2)b 31.1 (28.2–34.0)a 34.5 (31.3–37.8)b
Hands-off time (s)
Baseline, mean (95% CI) 47.7 (35.4–60.0)a 41.1 (30.1–52.2)a 39.1 (27.3–50.9)a
Post intervention, mean (95% CI) 27.5 (18.3–36.7)b 16.6 (11.7–21.5)b 8.1 (4.7–11.4)b
2-month follow-up, mean (95% CI) 12.5 (7.4–17.6)c 6.7 (4.0–9.4)c 9.0 (5.8–12.3)b

Letters “a”, “b”, and “c” indicate statistically significant differences (p < 0.0056) for metrics between groups within interventions (baseline, post-intervention, 2-month follow-
up) such that groups with different letters are statistically different while groups with the same letter are not. Alpha adjusted for multiple comparisons using the Bonferroni
method. All comparisons were made within interventions.
*
All subjects in the control arm were given CCO training after the post-intervention measurement.

258 Discussion and CPR quality. Bystanders in the BV group were more respon- 272

sive to an OHCA scenario after watching a 90 s video (Table 2). 273

259 Though significant efforts have been made to increase bystander In addition, viewing the BV decreased the hands-off times and 274

260 CPR rates, in many geographic areas the rate of bystander CPR per- improved chest compression rates which results in improved 275

261 formance is low.2,12,13 One intervention to increase provision of CPR CPR quality (Table 3). It has been documented that as hands- 276

262 has been to focus on teaching Chest Compression Only CPR (CCO- off time increases, patient morbidity and mortality increases.14–16 277

263 CPR) for lay bystanders. Studies have demonstrated that CCO-CPR This data suggests that use of a BV intervention increases CPR 278

264 is at least as effective as conventional CPR for adults who suddenly quality and responsiveness and may have an impact on OHCA 279

265 collapse.3 This approach has been integrated into the guidelines outcomes.3 280

266 published by both the American Heart Association and European Second, this study demonstrated long-term retention (>1 281

267 Resuscitation Council with the goal being to increase public aware- month) of lay-bystander CPR skills in high school participants after 282

268 ness and education thereby leading to increased performance of lay CCO training. Participants at 2 months were more likely to call 911, 283

269 bystander CPR for OHCA.1,4 start chest compressions sooner, and perform compressions at a 284

270 In this study, we demonstrated that both a brief video (BV) more effective rate compared to those that did not receive any CPR 285

271 and a CCO class were effective at improving CPR responsiveness education (Table 3). 286

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
 G Model
RESUS 6741 1–6 ARTICLE IN PRESS
D.L. Beskind et al. / Resuscitation xxx (2016) xxx–xxx 5

287 Finally, even though the BV group significantly improved in Conclusions 351

288 responsiveness and some quality measures from baseline to post-

289 intervention, depth of compression did not significantly improve Following a brief CPR video training, high school students had 352

290 from baseline to post intervention given our conservative threshold improved CPR quality (compression rate and hands-off time) and 353

291 for statistical significance to account for multiple testing (Table 3). responsiveness (calling 911, time to calling 911, and starting com- 354

292 At 2-month follow up, the compression depth was maintained but pressions within 2 min). There was concurrent improvement in 355

293 this was not statistically different from baseline. Both the control long-term retention (2 months) of lay-bystander CPR skills after 356

294 and CCO class groups improved at 2 months (Table 3). Further study any CPR training. One exception was that compression depth did 357

295 is needed to demonstrate that BV interventions can adequately not improve significantly from baseline for subjects trained using 358

296 teach appropriate chest compression depth before they are used only the brief CPR video. These results suggest brief educational 359

297 exclusively to teach CCO-CPR to lay bystanders. interventions can be beneficial to improve responsiveness but psy- 360

298 Our results are consistent with other studies showing that chomotor training is important for CPR quality and performance. 361

299 watching a brief CCO video provides improved overall CPR

300 performance and responsiveness as compared to untrained Conflicts of interest statement 362
301 bystanders.10,17 There was an increased compression rate consis-
302 tent with AHA CPR standards shown in these studies. In a previous The Gootter Foundation funded the making of the ultra-brief 363
303 study done in a shopping mall setting, after watching a brief video, CCO-CPR video and the CCO class. The Jim Click Foundation helped 364
304 participants were more likely to call 911, start compressions ear- to fund the purchase of the Laerdal SkillreporterTM mannequin 365
305 lier, and more likely to continue chest compression (decreased which was used in this study to test the effectiveness of the ultra- 366
306 hands-off time) which is consistent with our findings.17 Although brief CCO-CPR video and CCO class. The authors have no conflicts 367
307 Bobrow et al. showed an increase in chest compression depth post of interest. 368
308 video intervention, both our study and the shopping mall study did
309 not.10,17 The significant improvement in compression depth post-
310 class intervention and not post-BV intervention could be attributed Uncited references Q6 369

311 to the hands-on training that participants received during the CCO-
312 CPR class. [20,21]. 370

313 Through a public health lens, the accessibility of a brief video

314 on CCO-CPR ensures an easy delivery of a useful lifesaving skill. Acknowledgements 371
315 Previous to the 2010 guidelines many methods were used to teach
316 school children how to perform CPR including medical students We would like to acknowledge the high school students and 372
317 teaching full BLS classes.18,19 With the new ERC and AHA guide- health teachers Ann Smith and Mike Smith, the CCO Class instructor 373
318 lines, the recommendations for the use of hands only CPR allows Derek Smith, the Gootter Foundation, the Sarver Heart Center, the 374
319 another important way to teach school children. The approach of BV Jim Click Foundation, the Catalina Foothills Unified School District, 375
320 CPR lessons will provide exposure of critical resuscitation skills to and the University of Arizona College of Medicine Resuscitation 376
321 the masses while requiring a lower investment of time and money Education and CPR Training (REACT) Group. 377
322 as compared to CCO class intervention. This idea was studied by
323 Nielsen et al. who found that broadcasting CPR instruction on a
References 378
324 local television station in a rural community was an effective way to
325 teach CPR.18 Other studies have aimed at improving retention rates 1. Berg RA, Hemphill R, Abella BS, et al. Part 5: adult basic life support: 2010 379
326 with innovative ideas including mobile phone reminder videos American Heart Association Guidelines for Cardiopulmonary Resuscitation and 380

327 with some success.19 However, specific methods for BV implemen- Emergency Cardiovascular Care. Circulation 2010;122:S685–705. 381
2. Sasson C, Rogers MA, Dahl J, Kellermann AL. Predictors of survival from out-of- 382
328 tation into mass media events, movie theaters, and k-12 education hospital cardiac arrest: a systematic review and meta-analysis. Circ Cardiovasc 383
329 has yet to be studied. Qual Outcomes 2010;3:63–81. 384
3. Bobrow BJ, Spaite DW, Berg RA, et al. Chest compression-only CPR by lay 385
rescuers and survival from out-of-hospital cardiac arrest. J Am Med Assoc 386
330 Limitations 2010;304:1447–54. 387
4. Koster RW, Baubin MA, Bossaert LL, et al. European Resuscitation Council Guide- 388
331 Our study utilized a simulated situation that may not be appli- lines for Resuscitation 2010 Section 2: adult basic life support and use of 389
automated external defibrillators. Resuscitation 2010;81:1277–92. 390
332 cable to an individual’s performance in a truly emergent scenario. 5. Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics 391
333 Every participant needed parental consent before any testing could – 2014 update: a report from the American Heart Association. Circulation 392
334 be performed; the permission form sent home stated “CPR-CCO 2014;129:e28–92. 393
6. Sayre MR, Berg RA, Cave DM, et al. Hands-only (compression-only) cardiopul- 394
335 video”. Participants could have known the purpose of the study
monary resuscitation: a call to action for bystander response to adults who 395
336 from this form, but participants still performed poorly during the experience out-of-hospital sudden cardiac arrest: a science advisory for the 396
337 pre-test evaluation. Participants were brought back post-testing public from the American Heart Association Emergency Cardiovascular Care 397

338 to their classroom and could have discussed their experiences Committee. Circulation 2008;117:2162–7. 398
7. Dumas F, Rea TD, Fahrenbruch C, et al. Chest compression alone cardiopul- 399
339 with other participants that had not been evaluated. To address monary resuscitation is associated with better long-term survival compared 400
340 this common problem for cluster-randomized trials, participants with standard cardiopulmonary resuscitation. Circulation 2013;127:435–41. 401

341 were prompted not to talk with other participants. In addition, 8. Sasson C, Haukoos JS, Bond C, et al. Barriers and facilitators to learning and per- 402
forming cardiopulmonary resuscitation in neighborhoods with low bystander 403
342 our statistical models were designed to account for clustering by cardiopulmonary resuscitation prevalence and high rates of cardiac arrest in 404
343 Health class. Also, more than one model of mannequin was used Columbus, OH. Circ Cardiovasc Qual Outcomes 2013;6:550–8. 405

344 to test applicants. To reduce variability, participants were evalu- 9. Care AHAEC. CPR in schools; 2015. 406
10. Bobrow BJ, Vadeboncoeur TF, Spaite DW, et al. The effectiveness of ultrabrief and 407
345 ated on the same mannequin at baseline, post-intervention and brief educational videos for training lay responders in hands-only cardiopul- 408
346 2 months. Furthermore, the study population represented a sin- monary resuscitation: implications for the future of citizen cardiopulmonary 409
347 gle public high school and may not be generalizable to other resuscitation training. Circ Cardiovasc Qual Outcomes 2011;4:220–6. 410
11. Panchal AR, Meziab O, Stolz U, et al. The impact of ultra-brief chest compression- 411
348 high schools as differences in race, demographics, and geographic only CPR video training on responsiveness, compression rate, and hands-off time 412
349 location may be associated with differences in performance of interval among bystanders in a shopping mall. Resuscitation 2014;85:1287–90. 413
350 students. 12. Sarver Heart Center UoA. Lifesaver training; 2015. 414

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022
 G Model
RESUS 6741 1–6 ARTICLE IN PRESS
6 D.L. Beskind et al. / Resuscitation xxx (2016) xxx–xxx

415 13. Locke CJ, Berg RA, Sanders AB, et al. Bystander cardiopulmonary resuscita- 18. Breckwoldt J, Beetz D, Schnitzer L, Waskow C, Arntz HR, Weimann J. Medical 427
416 tion: concerns about mouth-to-mouth contact. Arch Intern Med 1995;155: students teaching basic life support to school children as a required element 428
417 938–43. of medical education: a randomised controlled study comparing three different 429
418 14. Nichol G, Thomas E, Callaway CW, et al. Regional variation in out-of-hospital approaches to fifth year medical training in emergency medicine. Resuscitation 430
419 cardiac arrest incidence and outcome. J Am Med Assoc 2008;300:1423–31. 2007;74:158–65. 431
420 15. Vaillancourt C, Stiell IG, Canadian Cardiovascular Outcomes Research Team. 19. Plant N, Taylor K. How best to teach CPR to schoolchildren: a systematic review. 432
421 Cardiac arrest care and emergency medical services in Canada. Can. J. Cardiol Resuscitation 2013;84:415–21. 433
422 2004;20:1081–90. 20. Nielsen AM, Isbye DL, Lippert FK, Rasmussen LS. Can mass education and a tele- 434
423 16. Christenson J, Andrusiek D, Everson-Stewart S, et al. Chest compression frac- vision campaign change the attitudes towards cardiopulmonary resuscitation 435
424 tion determines survival in patients with out-of-hospital ventricular fibrillation. in a rural community? Scand J Trauma Resusc Emerg Med 2013;21:39. 436
425 Circulation 2009;120:1241–7. 21. Ahn JY, Cho GC, Shon YD, Park SM, Kang KH. Effect of a reminder video using a 437
426 17. Idris AH, Guffey D, Pepe PE, et al. Chest compression rates and survival following mobile phone on the retention of CPR and AED skills in lay responders. Resus- 438
out-of-hospital cardiac arrest. Crit Care Med 2015;43:840–8. citation 2011;82:1543–7. 439

Please cite this article in press as: Beskind DL, et al. Viewing a brief chest-compression-only CPR video improves
bystander CPR performance and responsiveness in high school students: A cluster randomized trial. Resuscitation (2016),
http://dx.doi.org/10.1016/j.resuscitation.2016.03.022