S

i

FLa

b

a

ARR1A

KCCETCAFi

1

(wWpcah

tanCs

i

0d

Resuscitation 82 (2011) 436–441

Contents lists available at ScienceDirect

Resuscitation

journa l homepage: www.e lsev ier .com/ locate / resusc i ta t ion

imulation and education

CPR: A new application of high-quality cardiopulmonary resuscitation training�

ederico Semeraroa,∗, Floriana Taggia, Gaetano Tammaroa, Guglielmo Imbriacoa,uca Marchettib, Erga L. Cerchiari a

Department of Anaesthesia and Intensive Care, Ospedale Maggiore, Bologna, Italyd-Sign, Bologna, Italy

r t i c l e i n f o

rticle history:eceived 31 May 2010eceived in revised form0 September 2010ccepted 22 November 2010

eywords:ardiopulmonary resuscitationhest compressionducation

a b s t r a c t

Objectives: The present study evaluates a new CPR feedback application for the iPhone (iCPR) designed toimprove chest compression performance tested in a cardiac arrest simulation to evaluate performanceand acceptance by healthcare professionals and lay people.Methods: We built an application specifically dedicated to self-directed CPR training through a tutorial thatincludes a simple feedback module to guide training in order to improve the quality of chest compressions.We tested it in a sample of 50 users to evaluate the effect of iCPR on performance and it is acceptance.The participants were randomly assigned to one of the study groups and were asked to perform a trialof 2 min of chest compressions (CC), to answer a predefined set of questions and then to perform twomore minutes of CC. The first group performing the sequence of CC with iCPR – questions – CC withoutfeedback, and the second the sequence CC without feedback – questions CC with iCPR.

−1

rainingardiac arrestccelerometereedback devicePhone

Results: The mean compression rate was 101 ± 2.8 min when CC were performed with iCPR and107.8 ± 20.5 min−1 when performed without iCPR (p < 0.01). Overall, the participants considered iCPRuseful to maintain CC at the desired rate of 100 compressions per minute.Conclusions: The iCPR feedback tool was able to significantly improve the performance of chest com-pressions in terms of the compression rate in a simulated cardiac arrest scenario. The participants alsobelieved that iCPR helped them to achieve the correct chest compression rate and most users found this

device easy to use.

. Introduction

The consolidated importance of cardiopulmonary resuscitationCPR) for the survival of cardiac arrest patients has been integratedith the role of high-quality CPR to significantly improve outcomes.hen tested on mannequins, the quality of CPR performed by lay

eople and healthcare professionals tends to deteriorate signifi-antly within a few months after training.1,2 The quality of CPR islso often poor in the clinical setting, though the survival benefit ofigh-quality CPR is well documented.3,4

The European Resuscitation Council (ERC) guidelines emphasisehe importance of high-quality, uninterrupted CPR and also specifytarget compression depth of 4–5 cm and a rate of 100 min−1.5 A

umber of devices have been developed to provide guidance duringPR,3,6–8 and these have been used in both training and clinicalettings.

� A Spanish translated version of the abstract of this article appears as Appendixn the final online version at doi:10.1016/j.resuscitation.2010.11.023.∗ Corresponding author. Tel.: +39 3336592670; fax: +39 0516478867.

E-mail address: fsemeraro2008@gmail.com (F. Semeraro).

300-9572/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.resuscitation.2010.11.023

© 2010 Elsevier Ireland Ltd. All rights reserved.

The devices range in complexity from a simple metronome,which guides the compression rate, to more complex devices thatmonitor and provide combined audiovisual feedback on the actualCPR performance. The present study evaluates a new CPR feed-back application for the iPhone and iPod (iCPR – www.icpr.it) ina cardiac arrest simulation based on improvements in the exter-nal chest compression performance and acceptance by healthcareprofessionals and lay people.

2. Materials and methods

We built an application specifically dedicated to self-directedCPR training through a tutorial that includes a simple feedbackmodule dedicated to guide the training in relation to the qual-ity of chest compressions. The software can be used with iPodsand iPhones on armbands commonly used for running (Fig. 1). Theaccelerometer mounted inside of iPods and iPhones measures the

rate of compressions and gives audiovisual feedback to the opera-tor on the performance in order to reach the guideline target (100compressions per minute). The App Store is a service for the iPhone,iPod Touch and iPad created by Apple Inc. which allows users tobrowse and download applications from the iTunes Store that were

F. Semeraro et al. / Resuscitation 82 (2011) 436–441 437

on for

dDc

bAg

q

bbi

3tTitim0mtd

Fig. 1. iCPR applicati

eveloped with the iPhone SDK and published through Apple.epending on the application, they are available either free, or at aost.

The iCPR application is used wearing the device with an arm-and and keeping it with the display clearly visible for the user.built-in metronome, producing a clearly audible sound sample,

ives the user the right chest compression rate.The display shows two indicators: the chest compression fre-

uency and count.A visual feedback of the quality of chest compressions is shown

y green/red color on the frequency display. The green zone rangesetween 95 and 105 chest compressions per second. The red zone

s under and over this range.The software detects the gravity vector and subtracts it from the

-axis acceleration detected by the accelerometer and the magni-ude of the resulting acceleration vector is used for computation.he sampling frequency used to quantize the instant accelerations 50 Hz. The frequency of the chest compressions is computed byracking substantial changes in the direction of the acceleration ands displayed as the average of the three last chest compressions. A

agnitude threshold of 0.28 g during the compression phase and.20 g during release phase is used to filter out unwanted move-ents. No minimal depth of the chest compression is necessary to

rigger the frequency calculation and the algorithm used does notepend on the device orientation.

the iPhone and iPod.

The iCPR tutorial is based on ILCOR (American Heart Associationand European Resuscitation Council guidelines) and periodicallycan be used to easily refresh the CPR algorithm through a slideshow.Our study was carried out in the intensive care unit (ICU) at theSurgical and Obstetric Department of the Maggiore Hospital and atthe d-Sign Office in November of 2009. The participants comprisednurses, doctors, healthcare professionals and administrative per-sonnel. We used a prospective, randomized crossover design andasked the participants not to inform the other participants aboutthe scenario set-up. In Fig. 2 is described the CONSORT flow diagramfor the study.

All subjects were informed about the study through a stan-dardised leaflet and were told that their performance would beevaluated. Each participant provided written consent for the dataacquisition and analysis. All subjects received a standardised expla-nation of the feedback features of iCPR using a demonstration ofthe device and gained familiarisation without further instruction.Afterwards, participants were randomly assigned to one of thestudy groups and were asked to perform a trial of 2 min of con-tinuous chest compressions (CC), to answer a predefined set of

questions (about 10 min of pause) and then to perform two moreminutes of CC. The first group performing the sequence of CC withiCPR feedback, then questions and then CC without feedback, andthe second performing the sequence of CC without feedback, thenquestions and then CC with iCPR feedback.

438 F. Semeraro et al. / Resuscitation 82 (2011) 436–441

diagr

q

dfcaac

Ctrb

rtsc1s

3

i3d

Fig. 2. CONSORT flow

At the end, each participant was then asked to respond to auestionnaire about iCPR.

The questionnaire, based on a seven-point Likert scale (LS), wasesigned to explore the trainee’s perception in the areas of userriendliness, wearability and audibility of the device. We used aommercially available monitor/defibrillator with CPR-quality andudio recording capabilities (MRx-QCPR, Philips Medical Systems)nd a Resusci Anne mannequin that was placed on Emergency Res-ue bed with a thin mattress.

The data regarding the quality of CPR were collected with Q-PR Review (Laerdal Medical). According to the default settings ofhe device, a compression depth of 38–51 mm was assessed as cor-ect. A compression rate between 90 and 120 min−1 was acceptedecause the device was programmed according to 2005 guidelines.

For each subject, the following demographic parameters wereecorded: rescuer gender, age, weight, height and profession (doc-or, nurse or other). Each participant was asked to rate the followingtatements using a 10-cm visual analogue scale (VAS): “The chestompressions I performed were effective” (scale: 0 = not effective,0 = very effective); “It was exhausting to deliver chest compres-ions” (scale: 0 = no exhaustion at all, 10 = extreme exhaustion).

. Results

We analysed data from 50 iCPR participants. The samplencluded 58% males and 42% females, with an average age of4.4 ± 7.3 years and a BMI of 24.23 ± 4.2. The sample included 8%octors, 44% nurses and 48% lay people.

am of randomization.

The mean compression rate was 101 ± 2.8 min−1 when chestcompressions were performed with iCPR and 107.8 ± 20.5 min−1

when performed without iCPR (p < 0.01).The mean compression depth was 37.2 ± 12 mm with iCPR and

41.1 ± 13.1 mm without iCPR guidance (p = 0.57) (Table 1).All participants performed CPR scenarios with and without iCPR

guidance, and the opinions regarding CPR were assessed using theVAS. Performing chest compressions without iCPR guidance wasreported to be less efficient (p = 0.0059), and iCPR guidance did notaffect fatigue (p = 0.13) (Table 1).

The evaluations of the user-friendliness, wearability and audi-bility are reported in Table 2. The difficulty in wearing the armbandwith an iPod (Q1: 2.3 ± 1.9) was judged to be low.

The visibility of the numbers on the monitor (Q2: 6.3 ± 1.2) andthe audibility of the metronome during the CPR (Q3: 6.4 ± 1.2) wereevaluated as good.

The CPR manoeuvres with the armband and iPhone were judgedto be non-hampered (Q4: 1.7 ± 1.8), and finally, the participantsconsidered it useful to use iCPR to perform chest compressions ata rate of 100 compressions per minute (Q5: 6.3 ± 1) (Table 2).

4. Discussion

There is general agreement on the need to improve outcomes in

resuscitation, and it is recognised that changes in practise may berequired to make this possible.

However, concerns remain about the ability of lay people andhealthcare professionals to perform or recognise good chest com-pressions even shortly after training.

F. Semeraro et al. / Resuscitation 82 (2011) 436–441 439

Table 1Participants’ chest compression data without and with iCPR feedback during 2 min. Participants’ opinions regarding fatigue and chest compression efficacy without and withiCPR guidance using visual analogue scale (VAS).

Mean SD *p < 0.01

Compression data without iCPR feedbackAverage compression rate [min−1] [90–120] 107.8 20.5 *

Average compression depth [mm] [38–51] 41.1 13.1Total number of compressions 215.6 42.4 *

Average count per minute 107.8 21.1 *

Compression data with iCPR feedbackAverage compression rate [min−1] [90–120] 101.1 2.8 *

Average compression depth [mm] [38–51] 37.2 12Total number of compressions 202.4 6 *

Average count per minute 101.1 2.8 *

The compression data are presented as the mean and standard deviation and were analysed with the Student’s t-test

Without iCPR With iCPR p value

T xon t

qnwcad

a

wtr

od

co

wo

4

taaap

ifa

TUp

VAS efficacy 6 (1–9)VAS fatigue 6 (1–10)

he results are presented as the medians (range) and were analysed with the Wilco

This study demonstrated that iCPR guidance did improve theuality of the chest compression rate during a simulated CPR sce-ario. When iCPR guidance was used, the chest compression rateas within the current guidelines, and the number of performed

hest compressions per minute was correct. In addition, iCPR guid-nce had no effect on the depth of compressions, and there was noifference in rescuer fatigue.

The European Resuscitation Council (ERC) guidelines state thatrescuer should compress the chest at a rate of about 100 min−1.5

In our study, the mean chest compression rate (107.8 ± 20.5)ithout the use of iCPR was much higher than the recommenda-

ions. The use of the iCPR completely corrected the rate to thatecommended in the guidelines (101 ± 2.8).

Recently published data demonstrate the weak performancef healthcare professionals, especially regarding the rate and theepth of compressions.

For in-hospital resuscitation, Abella et al. detected a correcthest compression rate of 90–110 min−1 during only 31.4% of thebserved time during resuscitation attempts.

The survivors had a mean compression rate of 90 min−1,hereas the non-survivors received compressions at a mean rate

f 79 min−1.3

.1. CPR feedback

Many different methods have been employed to improvehe resuscitation skills, including video-recording evaluations,9,10

udio-feedback systems during training12 or integration inton automated external defibrillator13. Most methods have beenpplied in the training environment rather than in actual clinical

ractice.14

The first example of a chest compression feedback system camen 1992, when Kern et al. demonstrated that chest compression per-ormance was improved by employing a simple metronome,15 andfew years later Milander et al. concluded that audible tone guid-

able 2ser responses to iCPR. Participants rated the following statements using a seven-oint Likert scale (1 = completely disagree, 7 = completely agree).

Mean SD

Q1. It was difficult to wear the armband with an iPod 2.3 1.9Q2. The numbers on the monitor were easily visible 6.3 1.2Q3. The metronome was easily audible during the CPR 6.4 1.2Q4. The armband with the iPhone hampered the chestcompressions

1.7 1.8

Q5. iCPR helped you to perform chest compressions ata rate of 100 bpm

6.3 1

6.25 (0–10) 0.0059*

5 (3–9) 0.13

est.

ance could lead to higher compression rates.11 The CPR-Plus wasthe first stand-alone support developed to provide feedback onlyon the compression depth and incomplete release, but it demon-strated a reduction in the number of incorrect compressions andfatigue effects.16 The CPREzyTM was evaluated by Boyle et al. ina test with non-medical hospital staff, and they reported signifi-cant improvements in the compression rate and depth, especiallyduring the last minute of CPR. These results were validated in arandomized controlled study with 202 laypersons, with improve-ments in the same parameters and a learning effect regarding thecompression depth.17

Yeung et al.14 in a recent systematic review concluded thatthere is good evidence supporting the use of CPR feedback/promptsystems during CPR training to improve CPR skill acquisition andretention. The author suggests that their use in clinical practice aspart of an overall strategy might be beneficial to improve the qual-ity of CPR. Yeung et al. underlines that the accuracy of the devicesto measure compression depth should be calibrated to take intoconsideration the stiffness of the support surface upon which CPRis being performed (e.g. floor/mattress).

The authors also declared that further studies are required toestablish if the improvements in CPR quality can bring improve-ments in patient’s outcomes.

4.2. CPR on Apple Store

In App Store there is a large number of active applica-tions. The latest report describes approximately 250,000 activeapplications.18 The first three categories are: books 43,809 applica-tions (17.32%); games 36,719 applications (14.52%); entertainment28,771 applications (11.38%). The medical category is the seven-teenth with 3820 applications (1.51%). In the US App Store usingthe iTunes search engine with the term “CPR” there are 77 applica-tions. iCPR is the seventh application among CPR applications withan assessment rating of 3.5 out of 5 (average of 310 ratings).

The CPR top ten contains the following applications (update on1st September 2010): “CPR & Choking”,19 a series of video demon-strations on CPR and choking in adults and children; “Pocket FirstAid & CPR from the American Heart Association”20 with illustra-tions, including topics such as CPR, choking, bites, bruises, burns,seizures, diabetic emergencies and detailed videos showing how

to respond in critical first aid situations; “CPR-App”21 with voiceprompting to the persons trained in CPR through rescue proceduresfor adult, child and baby victims; “ResQr First Aid & CPR Coach”22

with first aid scenarios, including topics such as CPR, choking,bites, burns, seizures, diabetic emergencies, severe wounds, head

440 F. Semeraro et al. / Resuscitation 82 (2011) 436–441

Table 3Apple iTunes – top ten CPR applications comparison (update on 1st September 2010).

Name App Referenceguidelines

AudioVisualVideoTutorial

AudioVisualTactileFeedback

DepthRateFeedback

AverageRatings(1–5)Number ratings

Cost

CPR and choking AHAILCOR

Video – – 3.51698

Free

Pocket first aid and CPR AHA VisualVideo

– – 3.51025

$3.99$3.99

CPR-App AHA Audio Audio – 00

ResQr first aid & CPR coach AHARed cross

Visual Audio (Metronome) – 2.520

$3.99

CPR video instruction Red cross Video – – 2.550

$0.99

Pocket CPR AHAILCOR

AudioVisual

Audio(Metronome)Visual

Depth (inches)Rate

31053

$3.99

Hand-only CPR AHA VisualVideo

– – 3.5247

Free

iCPR AHAERC

Visual Audio(Metronome)Visual

Rate 3.5310

Free

CPR baby Red cross Visual – – 2.58

$0.99

Audio(MetrTactil

ipICodAilyt

S3ttrdtACcavspcptwtpiiC

tta

CPR buddy – Slide

njuries, to GPS assistance nearest ER location and proper CPRacing metronome and adult instructions; “CPR Video

nstruction”23 includes a series of video demonstrations onPR; “Pocket CPR”24 gives real-time feedback and instructionsn CPR. The technology is the same as that used in medicalevice Pocket CPR25; “Hand-Only CPR”26 from the American Heartssociation teaches how to perform Hand-Only CPR through video

nstructions; “CPR Baby – Know CPR for babies one year old oress”27 with a tutorial teaching how to perform CPR to any oneear old baby or less in just 5 steps; “CPR Buddy”28 includes a CPRutorial and a CPR chest compression metronome.

Since 18th November 2009, date of publication of iCPR on Apptore, the total downloads were 16,023: 69% in the United States,% in Canada, 3% in Italy, 23% in other countries. We comparedhe differences in between top ten CPR applications regardinghe reference guidelines, the presence of audio/visual/video tuto-ial, the presence of audio/visual/tactile feedback, the presence ofepth/rate feedback, the average rating and number of ratings andhe cost of application (see Table 3). The only two applications onpp Store which give feedback on chest compressions are PocketPR and iCPR. Pocket CPR measures the actual rate and depth ofhest compressions and give voice prompts to push faster or slower,nd whether you should push harder or softer. In addition to theoice prompts you can always hear the sound of the metronomeet at 100 beats per minute. iCPR measures the actual chest com-ressions rate and gives visual prompts about rate and number ofompressions with a sound metronome prompt set at 100 beatser minute. In Pocket CPR when chest compressions are started,he iPhone, when placed between or on top of the rescuer’s hands,ill measure the compression depth and rate. Whereas iCPR using

he iPhone placed on the armband is leaving your hands free toerform chest compressions with no difficulty. Another difference

s the cost of application: Pocket CPR actually costs $3.99 and iCPRs free. Pocket CPR and iCPR are both intended to be used only for

PR training.

They cannot be used in both cases if the iPhone is placed underrainer’s hands on patient’s chest; the screen will be obstructed andhe phone may be broken. Recently, iSkin produces a soft cases thatllows Pocket CPR and iCPR to be worn on the knuckles.29

onome)e-visual

– 231

$0.99

There were some limitations in our study. The use of a simulatedcardiac arrest model with a mannequin differs from the clinicalreality. It is not clear whether the quality of CPR given to man-nequins, especially the chest compression depth, is equivalent tothe quality of CPR given to patients. However, the parameters ofthe compression rate and total number of chest compressions givenare probably very close in the mannequin scenario and in a similarclinical setting.

The Q-CPR and other accelerometer devices include the amountof mattress compression when reporting chest compression depth.In this study we used an Emergency Rescue bed with a thin mattress(despite this probable underestimate depth by 1–1.5 cm). The studydesign that includes 2 min of chest compressions, about 10 minof pause and two more minutes of chest compressions does notexclude an order and carry over effect: it is possible that the orderin which performances are made may affect both the result and the“carry-over” between treatments, which confound the estimatesof the treatment effects. In practice, “carry-over” effects can beavoided with a sufficiently long “wash-out” period between per-formances and most probably 10 min of pause is not enough. Thereare three previous studies that used mobile phones to prompt CPR.

Zanner et al.30 implied that the action of activating the videoclip and following cellular phone instructions does take up precioustime and can delay important actions. Choa et al.31,32 instead haveshown that animation-assisted CPR group reached better checklistscore, including chest compression rate, depth and hand position-ing. Animation-assisted CPR could be used as a reminder tool inachieving effective one person-CPR performance.

In our study, however, the time to confirming arrest, calling forhelp and starting first chest compression were not calculated andcompared. This is a limitation that does not allow comparison withprevious studies. Nevertheless, the opening time of the applicationand starting iCPR takes about 5 s in experienced users of the iPhone.Participants also declare that the CPR manoeuvres with the arm-

band and iPhone were judged to be non-hampering (Q4: 1.7 ± 1.8).Therefore, further research and technology development is neededto simplify the operation of cellular phones, as well as, simplifyinginstruction method for the bystander to be able to perform essentialactions without unnecessary delays.

scitat

oTaci

fw

5

ifitru

C

mraw

A

mstFt

R

1

1

1

1

1

1

1

1

1122222222223

F. Semeraro et al. / Resu

In future studies, it would be interesting to examine the usef iCPR for different clinical scenarios and with depth feedback.he new iPhone 4 features an accelerometer motion detector andgyroscope that can be used to improve the measurement of chestompression rate and depth. Further studies are required to assessf the iPhone 4 can improve iCPR performances.

Furthermore iCPR needs to be tested comparing healthcare pro-essionals and lay people with previous CPR training versus peopleho had no training whatever.

. Conclusion

The iCPR feedback described here was able to significantlymprove the performance of chest compressions by healthcare pro-essionals and lay people in a simulated cardiac arrest scenarion terms of the compression rate. The participants also believedhat iCPR helped them to achieve the correct chest compressionate, and overall, most users found this device to be simple andsable.

onflict of interest statement

No relationship exists between any of the authors and any com-ercial entity or product mentioned in this manuscript that might

epresent a conflict of interest. No inducements have been made byny commercial entity to submit the manuscript for publication. Allithin 3 years of beginning the work submitted.

cknowledgments

We thank all of the Intensive Care, Surgery and Obstetric Depart-ent healthcare professionals for volunteering participate in the

tudy. We would also like to thank the d-Sign groups for work onhe application and for volunteering to participate in the study.inally, we thank Laerdal Italy for providing the equipment and forechnical support.

eferences

1. Stross JK. Maintaining competency in advanced cardiac life support skills. J AmMed Assoc 1983;249:3339–41.

2. Su E, Schmidt TA, Mann NC, Zechnich AD. A randomized controlled trial toassess decay in acquired knowledge among paramedics completing a pediatricresuscitation course. Acad Emerg Med 2000;7:779–86, 7.

3. Abella BS, Sandbo N, Vassilatos P, et al. Chest compression rates duringcardiopulmonary resuscitation are suboptimal: a prospective study during in-hospital cardiac arrest. Circulation 2005;111:428–34.

3

3

ion 82 (2011) 436–441 441

4. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopul-monary resuscitation during out-of-hospital cardiac arrest. J Am Med Assoc2005;293:299–304.

5. Handley AJ, Koster R, Monsieurs K, et al. European resuscitation council guide-lines for resuscitation 2005 Section 2. Adult basic life support and use ofautomated external defibrillators. Resuscitation 2005;67:S7–23.

6. Perkins GD, Augre C, Rogers H, Allan M, Thickett DR. CPREzyTM: an eval-uation during simulated cardiac arrest on a hospital bed. Resuscitation2005;64:103–8, 1.

7. Handley AJ, Handley SA. Improving CPR performance using an audible feed-back system suitable for incorporation into an automated external defibrillator.Resuscitation 2003;57:57–62.

8. Chiang WC, Chen WJ, Chen SY, Ko PC, Lin CH, Tsai MS, et al. Better adherence tothe guidelines during cardiopulmonary resuscitation through the provision ofaudio-prompts. Resuscitation 2005;64:297–301.

9. Donelly PD, Lester CA, Morgan CL, Assar D. Evaluating CPR performance in basiclife support: the VIDRAP protocol. Resuscitation 1998;36:51–7.

0. Lester CA, Morgan CL, Donnely PD, Assar D. Assessing with CARE: an innovativemethod of testing the approach and casualty assessment components of basiclife support, using video recording. Resuscitation 1997;34:43–9.

1. Milander MM, Hiscok PS, Sanders AB, Kern KB, Berg RA, Ewy GA. Chest compres-sion and ventilation rates during cardiopulmonary resuscitation: the effects ofaudible tone guidance. Acad Emerg Med 1995;2:708–13.

2. Wik L, Thowsen P, Steen PA. An automated voice advisory manikin system fortraining in basic life support without an instructor. A novel approach to CPRtraining. Resuscitation 2001;50:167–72.

3. Anthony JH, Simon AJH. Improving CPR performance using an audible feed-back system suitable for incorporation into an automated external defibrillator.Resuscitation 2003;57:57–62.

4. Yeung J, Meeks R, Edelson D, Gao F, Soar J, Perkins GD. The use of CPR feed-back/prompt devices during training and CPR performance: a systematic review.Resuscitation 2009;80:743–51.

5. Kern KB, Sanders AB, Raife J, Milander MM, Otto CW, Ewy GA. A study of chestcompression rates during cardiopulmonary resuscitation in humans. The impor-tance of rate-directed chest compressions. Arch Int Med 1992;152:145–9.

6. Elding C, Baskett P, Hughes A. The study of the effectiveness of chest compres-sions using the CPR-plus. Resuscitation 1998;36:169–73.

7. Beckers SK, Skorning MH, Fries M, et al. CPREzy improves performanceof external chest compressions in simulated cardiac arrest. Resuscitation2007;72:100–7.

8. http://148apps.biz/app-store-metrics/.9. http://learncpr.org.0. http://jive.me/firstaid/.1. http://www.cpr-app.com/.2. http://resqrsoftware.com/.3. http://redivideos.com/.4. http://www.pocketcpr.com/iphone.html.5. http://www.pocketcpr.com.6. http://handsonlycpr.org/.7. http://www.dashyapps.com/.8. http://dividebyzerosoftware.com/.9. http://www.iskin.com/.0. Zanner R, Wilhelm D, Feussner H, Schneider G. Evaluation of M-AID, a first aid

application for mobile phones. Resuscitation 2007;74:487–94.

1. Choa M, Park I, Chung HS, Yoo SK, Shim H, Kim S. The effectiveness of car-

diopulmonary resuscitation instruction: animation versus dispatcher througha cellular phone. Resuscitation 2008;77:87–94.

2. Choa M, Cho J, Choi YH, Kim S, Sung JM, Chung HS. Animation-assisted CPRIIprogram as a reminder tool in achieving effective one-person-CPR performance.Resuscitation 2009;80:680–4.