Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Evaluation of factors influencing Los Angeles Tiered-Dispatch System’s improvement on bystander CPR rate and inter reliability between electronic patient care report (ePCR) and 911 call review on...
(USC Thesis Other)
Evaluation of factors influencing Los Angeles Tiered-Dispatch System’s improvement on bystander CPR rate and inter reliability between electronic patient care report (ePCR) and 911 call review on...
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
1
Evaluation of Factors Influencing Los Angeles Tiered-Dispatch System’s
Improvement on Bystander CPR Rate and Inter Reliability between
Electronic Patient Care Report (ePCR) and 911 Call Review on Bystander
CPR Rate
By
Huihui Zhang
A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)
August 2016
Copyright 2016 Huihui Zhang
2
ACKNOWLEDGEMENTS
I would like to thanks all the people who helped and supported me in the completion of this
research project.
First, I would like to express my deepest appreciation to my committee chair, Dr. Christianne
Lane, who has been guiding me through this project, this educational journey. Without her
guidance this research project would not have been possible.
Second, I would like to thank Dr. Wendy J. Mack for helping me proof reading through this
research project and Dr. Stephen G. Sanko for the use of his data and guiding me through the
research question.
Third, I would like to thank the entire faculty from the Department of Preventive Medicine
and all employees at the Soto Building for creating an amazing educational experience at the
University of Southern California.
Last but not least, a special thanks also goes to my parents for supporting me spiritually
throughout my life.
3
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................ 2
LIST OF TABLES ......................................................................................................................... 4
ABSTRACT ................................................................................................................................... 5
CHAPTER 1: INTRODUCTION .................................................................................................. 7
CHAPTER 2: MATERIALS AND METHODS .......................................................................... 10
2.1. Participants and study design .......................................................................................... 10
2.2. Variables .......................................................................................................................... 10
2.3. Statistical Analyses .......................................................................................................... 12
CHAPTER 3: RESULTS ............................................................................................................. 14
3.1. Characteristics of the Patients ......................................................................................... 14
3.2. Concordance of ePCR and 911-call review report of bystander CPR ............................. 15
3.3. Bystander CPR performance and associations of case and call characteristics with
performance of bystander CPR ................................................................................................ 16
CHAPTER 4: DISCUSSION ....................................................................................................... 20
CHAPTER 5: CONCLUSION .................................................................................................... 22
REFERENCES ............................................................................................................................. 23
APPENDIX ................................................................................................................................... 25
4
LIST OF TABLES
Table 1. Baseline Characteristics of the Cases ............................................................................. 15
Table 2. Bystander CPR Rate Calculated both from ePCR and 911 Call Review and Inter
Reliability between ePCR and Call Review ................................................................................ 17
Table 3. Associations of Case and Call Characteristics with Performance of Bystander CPR ... 19
5
ABSTRACT
Background – Previous studies that have examined bystander CPR performance using data from
the electronic patient care report (ePCR) or 911 call review. Few studies have shown the inter-
reliability of the two methods of data collection. In this study we evaluated agreement between
the two data sources and the influence of measure used on whether the new dispatch system (Los
Angeles Tiered-Dispatch System) implemented in Los Angeles in December 1, 2014 improved
bystander CPR performance over the old system, or showed different relationship factors that
were associated with bystander CPR performance.
Methods – We conducted a retrospective case-series study. We calculated the CPR rate based on
the records from ePCR and 911 call review separately and used Cohen’s Kappa test to estimate
and test the agreement between the two data sources. We used logistic regression to estimate and
test odds ratios for association between call characteristics and whether or not bystander CPR
was performed for each measure.
Results – The records as to whether or not bystander CPR was performed from 911 call review
and ePCR only fairly agreed (Kappa=0.37, 95%CI: 0.29, 0.44). This fair level of agreement was
consistent across subgroups of patient age, ethnicity, dispatch system, caller’s emotional status,
caller’s English proficiency, incidents area and whether incidents occurred at home. Compared to
911 call review, ePCR underreported the bystander CPR in both the Los Angeles Tiered-
Dispatch System (LATDS) and the Medical Priority Dispatch System (MPDS). Regardless of
measure used, LATDS significantly improved the bystander CPR performance with either ePCR
or 911 call measures. Under LATDS, the CPR rate was 66.4%, and was 51.9% under MPDS.
After adjusting for dispatch system, the odds of receiving bystander CPR in a residential setting
were 2.65(95%CI: 1.38,5.08) times higher than the odds for incidents occurring in public places.
6
Under MPDS, the odds of receiving bystander CPR for callers who were not significantly
emotional were 3.85(1.49,9.90) times higher than the odds for emotional callers. Patient age,
race, caller’s English proficiency and incidents area were not significantly associated with
performance of bystander CPR.
Conclusions – Among the cardiac arrests occurring out of hospital in Los Angeles, ePCR
underreports the bystander CPR rate compared to 911 call review and they only moderately or
fairly agree with each other on the performance of bystander CPR. The new system LATDS
significantly improved the bystander CPR performance above the MPDS. Callers in significant
emotional status and cases occurring in public places tended to have a lower likelihood of
bystander CPR performance. More studies are needed to analyze other factors related to
improved bystander CPR performance.
7
CHAPTER 1: INTRODUCTION
Out-of-hospital cardiac arrest is a major public health concern in modern society, given that
there are approximately 420,000 cases in the United States annually.
1,2
Approximately 250,000
Americans die each year from sudden cardiac arrest, the overwhelming majority of which occurs
outside of the hospital, in people’s homes as well as in public places.
3
For decades, cardiac arrest
has been the leading cause of death among adults. Decreasing the time to treatment is crucial for
improving the chance of survival in cases of cardiac arrest.
4,5
The most important response
measures that currently can be taken outside a hospital setting are: (1) recognizing early that a
cardiac arrest is occurring, (2) placing an alarm call, (3) performing cardiopulmonary
resuscitation (CPR), and (4) performing defibrillation.
6,7
It is widely accepted that early initiation of bystander CPR before arrival of EMS personnel
and before initiation of advanced life support is an important factor that can improve
neurological outcome and long-term survival.
8
CPR slows the dying process, buying time for the
defibrillator to arrive and shock the heart into a normal rhythm. CPR delays the onset of
irreversible damage to the brain and the heart by keeping a flow of oxygenated blood circulating
to the vital organs.
9,10
Even with the help of CPR, however, the flow of blood is far from normal
and probably achieves only 10 to 30 percent of normal cardiac output. However, this trickle of
oxygenated blood is sufficient to delay the death of cells, especially the sensitive cells of the
brain.
11
Modern CPR was introduced just 50 years ago and traditionally has consisted of chest
compression – decompression interspersed with rescue breathing.
12
When bystanders, usually laypeople, take action and attempt CPR, the chances of survival
improve substantially. CPR performed by bystanders doubles the probability of surviving a
cardiac arrest,
13,14
and yet, less than half of cardiac arrest victims receive bystander CPR.
15
8
Taken together, these circumstances necessitate ongoing efforts to achieve more comprehensive
implementation of bystander CPR.
Previous studies identify a host of barriers preventing bystander CPR, which include, fear of
injuring the patient, concern for incorrect or poor CPR performance, physical inability of the
bystander to perform CPR, legal liability, and the risk of infectious disease transmission.
16
In
studies that have investigated bystanders who actually witnessed an out-of-hospital cardiac arrest,
common themes are the challenge of identifying the arrest victim and the lack of confidence and
composure to act.
17
Given the host of barriers, a successful strategy to improve bystander CPR
will not be any single intervention, but instead needs to incorporate a variety of modalities that
would appeal to the spectrum of bystanders and address the range of potential barriers.
18
One
modality is for 911 dispatchers to recruit bystanders to perform CPR while emergency services
are en route.
Since 1989, the Los Angeles Fire Department (LAFD) has used the Medical Priority Dispatch
System (MPDS) to guide 911-dispatchers to ask a series of scripted questions to determine the
presence of an emergency medical condition. Although MPDS prompts dispatchers to provide
CPR instruction, decades of local experience with this system has shown a delay in call-
processing, with subsequently longer response times due to its structure requiring a lengthy
query of the caller prior to dispatching resources.
On December 1, 2014, LAFD developed a new card system, called the Los Angeles Tiered-
Dispatch System (LA-TDS), which applied a new series of scripted questions that drastically
decreases the number of questions needed to identify potential victims of cardiac arrest and
lowers the threshold to provide dispatcher-assisted CPR.
The new series of scripted questions applied by LA-TDS reduces delays related to gathering
9
information such as phone number and address, and ensures that the most appropriate Los
Angeles Fire Department (LAFD) resource is sent to the patients immediately. Despite the 14%
increase in call loads since 2014, there have been improvements in response time — including a
16-second improvement in call processing times, and at least five seconds of improvement in
turnout time.
19
In order to estimate whether there is improvement in performance of bystander CPR, it is
important to accurately measure this outcome. Many of the studies on bystander CPR
performance rely on data collected from the field report filled by EMS personnel, which can be
inaccurate, since it is recorded primarily after the incident and CPR is done. This may lead to
underreporting of bystander CPR if it is not witnessed by emergency responders or if they forget
to note it down. Using data from a retrospective interrupted time series study examining 911
calls under the two systems, we compared prevalence of bystander CPR from field notes and
reviews of the calls. This study examines (1) whether there is underreporting in bystander CPR
by examining electronic measures versus review of call recordings in the two call systems
(LATDS vs. MPDS); (2) whether the effect of the change of call system on prevalence of
bystander CPR is similar in both electronic records and call reviews; and (3) what call
characteristics are associated with performance of bystander CPR.
10
CHAPTER 2: MATERIALS AND METHODS
2.1. Participants and study design
The study was conducted in the City of Los Angeles, CA with a population of 4.1 million and
whose 911 center is staffed by uniformed firefighters certified as emergency medical
dispatchers.
20
Dispatch and emergency medical service (EMS) are provided by the Los Angeles
Fire Department (LAFD). Each call made to the system is captured by an electronic patient care
report (ePCR) detailing the call and follow through by EMS at the scene. Calls from January to
March 2014 using the MPDS system were compared to calls from January to March 2015 using
the new LA-TDS system. The LA-TDS was implemented in December 2014, so that there was a
lead-in of one month before data collection began on the new system to allow for training effects.
During the study periods under review here, there were no significant changes in dispatch
personnel, population served, number of daily EMS incidents, or field care protocols for the
emergencies under study.
Cases included here were selected by retrospectively reviewing LAFD ePCR for patients
diagnosed in the field with non-traumatic cardiac arrest, and filtering based on the Utstein
Template, which excluded traumatic cardiac arrest and cardiac arrest patients without
resuscitation attempted.
21
Study inclusion criteria were: all non-traumatic cardiac arrests handled
by LAFD with attempted resuscitation, and not witnessed by EMS providers or trained medical
staff (e.g., cases occurring in a health facility) before the ambulance arrived. The cases’ weight
was measured in pounds and recorded on ePCR. Disease history of asthma, cancer, cardiac
disease, chronic respiratory failure or emphysema, end-stage renal disease (ESRD) or chronic
kidney disease (CKD), cerebrovascular accident (CVA), diabetes mellitus type 2 (DM2), high
blood pressure (HTN), seizure disorder, psychiatric problems, substance abuse and HIV was
11
indicated on the ePCR.
After identifying cases of cardiac arrest confirmed by paramedics in the field, the 911 call
associated with each incident was obtained by LAFD dispatch personnel, de-identified, and
reviewed by trained abstractors. Four trained non-LAFD abstractors listened to all recorded calls
meeting the inclusion criteria, and recorded if dispatcher-assisted-CPR was initiated, as well as
the time elapsed from the start of the call until key events identified in the call (delivery of CPR
instructions, and delivery of first chest compression).
22
High inter reliability was established for
the reviewers (>.9).
2.2. Variables
The primary outcome of interest was performance of bystander CPR. Both the ePCR and the
call review included a question about whether bystander CPR occurred. The dispatch system was
clearly defined in the dates when it was applied: January - March 2014 calls were MPDS while
January – March 2015 calls were LA-TDS. In addition, characteristics about the case, the call,
and the caller were obtained from the ePCR and call reviews. Patient age was dichotomized at
age 65. This cut point was used in accordance with previous studies that concluded the average
sudden cardiac arrest age was about 65 and the incidents of cardiac arrests increased with
increasing age.
23
Patient ethnicity was dichotomized as non-white (Hispanic/Black/Asian/Pacific
Islander) and White. These large groupings are due to the fact that most of the race/ethnicity data
were not recorded. Due to the large amount of missing data, models were run with and without
the ethnicity variable. From the call review, the caller’s emotional state was categorized as
whether the caller was significantly emotional or not. If during the call, the caller were too
emotional, resulting in difficulty of the 911 dispatcher obtaining information, then the incident
12
will be categorized as ‘ Yes’. Otherwise, it will be defined as ‘No’. Call reviewers also noted if
the caller appeared to have limited English proficiency, dichotomized as ‘Yes’ or ‘No’. The
location of arrest was categorized as residential or non-residential. Finally, the responding units
were divided into geographic response bureaus: Central, West, Valley, and South Bureaus based
on the dispatch battalion. (See Appendix)
2.3. Statistical analyses
Cohen’s Kappa statistics were calculated to determine to what degree ePCR and call review
report agree. This was performed for the overall calls, and stratified by the factors (type of
system, and call characteristics). Kappa statistic< 0 is interpreted as less than chance agreement,
Kappa statistic between 0.01 and 0.20 is interpreted as slight agreement, Kappa between 0.21
and 0.40 is interpreted as fair agreement and Kappa between 0.41 and 0.60 is interpreted as
moderate agreement.
24
Logistic regression models tested whether the above call characteristics were associated with
the dichotomous outcome bystander CPR performance. Bystander CPR performance identified
from 911 call review and from ePCR were run in separate models. Models were created for each
of the seven variables mentioned above, and odds ratios with 95% confidence intervals reported.
Except for dispatch system, all other variables were adjusted for dispatch system. Interactions
with dispatch system were also tested to evaluate whether the relationship between these
variables and bystander CPR performance differed according to the two dispatch systems. For
variables that had significant interaction with dispatch systems, models were rerun under MPDS
and LATDS separately without the dispatch system and interaction term. The models tested
whether the case and call characteristic, dispatch system and their interaction was associated with
13
bystander CPR performance using records from ePCR and 911 call review .
All Kappa statistical tests were two-sided, and a P value of less than 0.05 was considered to
indicate statistical significance. SAS software, version 9.4 (SAS Institute), was used for all
statistical analyses.
14
CHAPTER 3: RESULTS
There were 1027 cases of non-traumatic cardiac arrest in Los Angeles during January-March
2014 and January-March 2015. Thirteen (1.3%) records were excluded from the analyses due to
inability to retrieve the calls for review, 170 (16.6%) cases were excluded because they were
witnessed by EMS and 243 (23.7%) incidents were excluded because the arrests occurred in an
emergency department, medical clinic, skilled nursing facility, or nursing home. Of all the cased
collected, 642 (62.5%) observations were included in the final analysis.
3.1. Case Characteristics
Table 1 reports characteristics of the 642 cases included in this study. For ethnicity, 431 (67%)
of the values were missing and 6 (1%) values about the weight were missing. Sixty-three percent
of the patients were male and 51% of the patients were below age 65. The average weight of
patients was 173 lb (95% CI: 168, 177). Twenty-four percent of the cases were White non-
Hispanic, 76% were Hispanic, Black, Asian or Pacific Islander. Six percent of the patients were
known to have asthma, 4% were known to have cancer, 29% to have cardiac disease, 5% to have
chronic respiratory failure or emphysema, 4% to have end-stage renal disease (ESRD) or chronic
kidney disease (CKD), 4% to have cerebrovascular accident (CVA), 23% to have diabetes
mellitus type 2 (DM2), 30% to have high blood pressure (HTN), 3% to have Seizure disorder, 2%
to have psychiatric problems, 3% to have substance abuse and only one case of HIV. Overall the
demographic patterns of the patients between the two systems were very similar.
15
Table 1 - Baseline Characteristics of the Cases.
Variable Overall
(N=642)
MPDS
(N=310)
LATDS
(N=332)
Weight (lb, M (95% CI)) 173(168, 177) 174(167, 182) 171(165, 177)
Age (N (%))
>65
<65
316(49%) 155(50%) 161(48%)
326(51%) 155(50%) 171(52%)
Race (N (%))
White
Non-White
Unknown
50(8%) 18(6%) 32(10%)
161(25%) 95(31%) 66(20%)
431(67%) 197(64%) 234(70%)
Gender (N (%))
Female
Male
Unknown
235(37%) 116(37%) 119(36%)
405(63%) 192(62%) 213(64%)
2(0%) 2(1%) 0(0%)
Disease (N (%))
Asthma 39 (6%) 23 (7%) 16 (5%)
Cancer 25 (4%) 15 (5%) 10 (3%)
Cardiac Disease 185 (29%) 92 (30%) 93 (28%)
CRF/Emphyzema 30 (5%) 15 (5% ) 15 (5%)
CKD/ESRD 28 (4%) 13 (4%) 15 (4%)
CVA 28 (4%) 18 (6%) 10 (3%)
DM2 146 (23%) 73 (24%) 73 (22%)
HTN 192 (30%) 99 (32%) 93 (28%)
HIV 1 (0%) 1 (0%) 0 (0%)
Seizure disorder 17 (3%) 5 (2%) 12 (4%)
Psychiatric Problems 12 (2%) 4 (1%) 8 (2%)
Substance Abuse 16 (3%) 8 (3%) 8 (2%)
Note. Non-White Race includes: Hispanic, Black, Asian/Pacific Islander;
Disease categories are not exclusive.
3.2. Concordance of ePCR and 911-call review report of bystander CPR
For both systems combined, the Kappa coefficient of 0.37 (95%CI: 0.29, 0.44) indicated only
fair agreement between the two data sources. From the call reviews, the bystander CPR rate was
59.3%, while the rate was 52.1% from ePCR (i.e., the ePCR underreported the overall bystander
CPR rate, relative to 911 call review). These rates were similar for each system (Table 2) and
indicated an underreporting of bystander CPR in the ePCR. Among the incidents under the
16
MPDS system, the bystander CPR rate was 47.7% from the ePCR records while the rate was
51.9% from the 911 call review. The Kappa coefficient was 0.44 (95%CI: 0.34, 0.54), indicating
moderate agreement between ePCR and call review. Among incidents under the LATDS system,
the bystander CPR rate was 56.3% from the ePCR records while the rate was 66.4% from the
911 call review, indicating only fair agreement (Kappa=0.31, 95% CI: 0.10, 0.53). The ePCR
underreported the CPR rate regardless of the system, though both systems demonstrate a higher
prevalence of bystander CPR in LA-TDS.
When examining the concordance of ePCR and call review across call characteristics,
Kappa coefficients ranged from 0.30 to 0.43, indicating only fair agreement or moderate
agreement. ePCR underreported the bystander CPR prevalence relative to 911 call review,
regardless of the age groups, English proficiency and callers’ emotion. For cases occurring in
areas belonging to Central Battalion, ePCR slightly over-reported the rate from 48.2% to 49.1%,
while for other battalions, ePCR under-reported the rate. For non-residential cases, the bystander
CPR rate was 40.9% from 911 call review versus 60.0% from the ePCR, which was over-
reported. Also this rate was 60.0% among Hispanic, Black, Asian and Pacific Islanders from 911
call review versus 64.4% from ePCR for White patients, slightly over-reported by ePCR. (Table2)
3.3. Bystander CPR performance and associations of case and call characteristics with
performance of bystander CPR
The new dispatch system LATDS significantly (p<0.05) improved the bystander CPR
performance, regardless of whether it was measured by ePCR or call review (Table 3). The odds
ratio for LATDS was 1.8, meaning that the odds of receiving bystander CPR under LATDS were
1.83 (95%CI: 1.31, 2.56) times higher than odds under MPDS measured by call review. Based
17
Table 2 - Bystander CPR Rate Calculated both from ePCR and 911 Call Review and Inter
Reliability between ePCR and Call Review
Variable Bystander CPR Rate Kappa coefficient
(95%CI)
911 Call Review
(%)
ePCR
(%)
Dispatch System
MPDS (1/14-3/14) 51.9 47.7 0.44(95%CI: 0.34, 0.54)
LATDS (1/15-3/15) 66.4 56.3 0.31(95%CI: 0.10, 0.53)
Age
<65 62.0 54.9 0.30(95%CI: 0.19, 0.41)
>=65 56.5 49.1 0.43(95%CI: 0.33, 0.54)
Residential
Yes 63.2 50.5 0.40(95%CI: 0.29, 0.50)
No 40.9 60.0 0.39(95%CI: 0.23, 0.54)
Ethnicity
Non-White 57.9 46.2 0.33(95%CI: 0.18, 0.48)
White 60.0 64.4 0.52(95%CI: 0.27, 0.78)
Incident Area
Central 48.2 49.1 0.39(95%CI: 0.21, 0.56)
South 62.4 47.0 0.30(95%CI: 0.17, 0.43)
Valley 61.8 56.5 0.40(95%CI: 0.26, 0.53)
West 61.0 56.0 0.40(95%CI: 0.22, 0.58)
Emotional
Yes 56.9 40.0 0.31(95%CI: 0.10, 0.53)
No 59.7 53.5 0.38(95%CI: 0.30, 0.46)
Limited English Proficiency
Yes 51.7 39.7 0.35(95%CI: 0.11, 0.58)
No 60.2 53.5 0.37(95%CI: 0.29, 0.45)
Overall 59.3 52.1 0.37(95%CI: 0.29, 0.44)
on the records from 911 call review to define the dependent variable of bystander CPR, the
location of arrests was significantly associated with the bystander CPR performance after
adjusting for dispatch system. After adjusted for dispatch system, the odds of receiving bystander
CPR for callers who were able to speak fluent English were 3.85(95%CI: 1.49,9.90) times higher
than the odds for callers with limited English proficiency. Adjusting for dispatch system, the
odds of receiving bystander CPR in a residential setting were 2.65(95%CI: 1.38,5.08) times
18
higher than odds for incidents occurring in public places. Location was not statistically
significantly different in the ePCR report of bystander CPR, and the OR of 0.65 indicated an
effect in the opposite direction.
Caller’s emotion was significantly (p<0.05) associated with bystander CPR performance after
adjusting for dispatch system and also had a significant interaction (p<0.05) with dispatch
system. Except for emotional state, there were no significant interactions of call and case
characteristics with dispatch system, indicating that under different dispatch systems the
relationship of emotional state and bystander CPR performance differed. From call review under
LATDS, emotional state was not significantly associated with bystander CPR performance (P-
value=0.11). However under MPDS the association of bystander performance and emotional
state was significant, the odds of receiving bystander CPR for callers who were not significantly
emotional were 3.85 (95%CI: 1.49, 9.90) times higher than odds for emotional callers.
Patients’ age, ethnicity and incident area were not significantly associated with bystander CPR
performance, after adjusting for dispatch system.
19
Table 3 - Associations of Case and Call Characteristics with Performance of Bystander CPR
911 call review ePCR
Odds Ratio
(95%CI)
p
Interaction
(p-value)
Odds Ratio
(95%CI)
p
Interaction
(p-value)
LATDS 1.83(1.31,2.56) <0.01 1.46(1.07,1.99) 0.02
Age<65 1.47(0.93,2.33) 0.11 0.35 1.30(0.83,2.03) 0.25 0.86
Residential 2.65(1.38,5.08) <0.01 0.88 0.65(0.36,1.18) 0.15 0.93
Non-white 1.17(0.40,3.38) 0.78 0.80 0.51(0.15,1.82) 0.19 0.92
Battalion
Central
South
Valley
1.63(0.74,3.59)
0.68(0.34,1.35)
1.19(0.59,2.40)
0.07
0.23
0.27
0.62
0.16
0.79
0.20
0.54
1.34(0.63,2.86)
1.52(0.79,2.94)
1.28(0.65,2.50)
0.67
0.45
0.22
0.47
0.68
0.79
0.68
0.27
Non-emotional Callers
LATDS
MPDS
0.53(0.24,1.17)
3.85(1.49,9.90)
<0.01
0.12
<0.01
<0.01 2.98(1.16,7.69) 0.02 0.11
Fluent English Callers 2.65(1.11,6.31) 0.03 0.07 1.71(0.74,3.94) 0.21 0.79
Note. Age, Residential, Race, Battalion, Emotional and LEP are all adjusted for Dispatch System;
Sample size for Race (White vs Non-white) is 211.
20
CHAPTER 4: DISCUSSION
We found that the prevalence of CPR was underreported on field reports relative to the 911
call, regardless of call system used. This is likely due to the requirements of EMS actually
witnessing CPR, which may underreport in some cases, for instance if the individual performing
CPR on the patient had to stop to answer the door when EMS arrived. Such an instance was
recorded as no CPR performed. We did see that there was underreporting of CPR in residences.
However recruiting people to listen to the 911 call is time-consuming and expensive, and so it is
reassuring to see similar effects of the call system on the prevalence of bystander CPR, even if
underreported. Bystander CPR rate in Los Angeles under the new dispatch system LA-TDS is
66.4% which is significantly higher than 51.9% under the MPDS system and in line with other
cities, such as Seattle with an overall prevalence of bystander CPR of 54% in cardiac arrest
events of cardiac origin.
25
A previous study from Bergner et al. reported bystander CPR in 34%,
of which the contribution of lay CPR was 29%.
26
There are other factors that may influence under- or over reporting. Bystander CPR
performance tended to be over reported among White patients and cases occurring in public
places. For cases occurring in public places, bystander CPR rate was higher from ePCR than
from call review, which is inconsistent with the overall findings. This gives us a clue that
bystanders in public places may exaggerate CPR reporting. Most other factors had similar
agreement when examined by call characteristics. From the Kappa statistical test, almost all of
the tests showed only moderate or fair agreement between 911 call review and ePCR.
Knowing factors that influence whether or not bystander CPR is performed has implications
for targeting of community-wide interventions. Previous studies showed that CPR-trained
bystanders were 6.6 times more likely to preform CPR.
27
Patients with higher socioeconomic
21
status are more likely to receive bystander CPR. For cases happened residentially, each $100,000
increment in the dwelling’s value, the likelihood of receiving bystander CPR increased (OR =
1.07; 95% CI 1.01–1.14).
28
In our study, using 911 call review to define the bystander CPR
outcome, location of arrests (residential/non-residential) and emotional state of the caller were
significantly associated with performance of bystander CPR. Emotional callers tended to not
perform CPR on patients. In the future, the dispatchers should be trained to calm down the caller.
Or if the dispatchers notice that the callers are extremely emotional, she/he might determine if
there is someone else at the scene to speak with.
We calculated the bystander CPR rate using data from 911 call review, which is more accurate.
In this study, we conducted Cohan’s Kappa test controlled for dispatch system and many call
characteristics. All the logistic models were adjusted for dispatch system. Cases witnessed by
EMS and occurring in nursing facilities were excluded in this study, making the estimation of
bystander CPR performance more accurate. Our study also has some limitations. First, for some
variables, a large proportion of data was missing and it was hard to retrieve the reason about the
missing data, which can impair the power of the study to some extent. For ethnicity, 67% of the
data was missing. Location of arrest was missing in 34% of the cases. Second, this study
analyzed only a limited number of variables that might be associated with bystander CPR and
agreement of two data sources.
22
CHAPTER 5: CONCLUSION
911 call review and ePCR only fairly agreed with each other, which may impact prevalence
reports of bystander CPR. Cohan’s Kappa statistics calculated under each level of patients’ age,
ethnicity, dispatch system, caller’s emotional status, caller’s English proficiency, incidents area
and whether incidents happened at home suggested that the two data sources still only fairly or
moderately agree with each other. Under LATDS, the bystander CPR rate was 66.4%, and was
51.9% under MPDS. After adjusting for dispatch system, arrests happened at home was 2.60
times as much as arrests happened in public places to receive bystander CPR. Cases of which
callers were not emotional were 3.85 times to have bystander CPR than cases with emotional
callers. Patients’ age, ethnicity, callers’ English proficiency and incidents area were not
significantly associated with the performance of bystander CPR.
23
References:
1. Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics — 2014 update: a
report from the American Heart Association. Circulation 2014;129(3): e28-e292.
2. Berdowski J, Berg RA, Tijssen JG, Koster RW. Global incidences of out-of- hospital cardiac
arrest and survival rates: systematic review of 67 prospective studies. Resuscitation
2010;81:1479-87.
3. American Heart Association. 2002 Heart and Stroke Statistical Update. Dallas, TX: American
Heart Association; 2002.
4. Weaver WD, Cobb LA, Hallstrom AP, Fahrenbruch C, Copass MK, Ray R. Factors
influencing survival after out-of-hospital cardiac arrest. J Am Coll Cardiol 1986;7: 752-7.
5. Cummins RO, Eisenberg MS, Hall- strom AP, Litwin PE. Survival of out-of- hospital cardiac
arrest with early initiation of cardiopulmonary resuscitation. Am J Emerg Med 1985;3:114-9.
6. Field JM, Hazinski MF, Sayre MR, et al. Part 1: executive summary: 2010 Ameri- can Heart
Association guidelines for car- diopulmonary resuscitation and emergen- cy cardiovascular care.
Circulation 2010; 122:Suppl 3:S640-S656.
7. Nolan JP, Soar J, Zideman DA, et al. European Resuscitation Council Guide- lines for
Resuscitation 2010 Section 1. Executive summary. Resuscitation 2010; 81:1219-76.
8. Bossaert L, Van Hoeyweghen R, Cerebral Resuscitation Study Group. Bystander
cardiopulmonary resuscitation (CPR) in out-of-hospital cardiac arrest[J]. Resuscitation, 1989, 17:
S55-S69.
9. Ritter G, Wolfe R A, Goldstein S, et al. The effect of bystander CPR on survival of out-of-
hospital cardiac arrest victims[J]. American heart journal, 1985, 110(5): 932-937.
10. Roth R, Stewart R D, Rogers K, et al. Out-of-hospital cardiac arrest: factors associated with
survival[J]. Annals of emergency medicine, 1984, 13(4): 237-243.
11. Mikey SE. Resuscitate; 2009.
12. Steven MB, Thomas DR. Improving bystander cardiopulmonary resuscitation. Current
Opinion in Critical Care 2011, 17:219 – 224.
13. Sasson C, Rogers M A M, Dahl J, et al. Predictors of survival from out-of-hospital cardiac
arrest a systematic review and meta-analysis[J]. Circulation: Cardiovascular Quality and
Outcomes, 2010, 3(1): 63-81.
14. Bardy G H. A critic's assessment of our approach to cardiac arrest[J]. New England Journal
of Medicine, 2011, 364(4): 374-375.
24
15. Nichol G, Thomas E, Callaway CW, et al. Regional variation in out-of-hospital cardiac arrest
incidence and outcome. JAMA 2008; 300:1423–1431.
16. Coons SJ, Guy MC. Performing bystander CPR for sudden cardiac arrest: behavioral
intentions among the general adult population in Arizona. Resus- citation 2009; 80:334–340.
17. Swor R, Khan I, Domeier R, et al. CPR training and CPR performance: do CPR-trained
bystanders perform CPR? Acad Emerg Med 2006; 13:596– 601.
18. Abella BS, Aufderheide TP, Eigel B, et al. Reducing barriers for implementa- tion of
bystander-initiated cardiopulmonary resuscitation: a scientific state- ment from the American
Heart Association for healthcare providers, policymakers, and community leaders regarding the
effectiveness of cardio- pulmonary resuscitation. Circulation 2008; 117:704–709.
19. http://www.lamayor.org
20. About the LAFD. Los Angeles Fire Department. Retrieved February 20, 2007.
21. Perkins G D, Jacobs I G, Nadkarni V M, et al. Cardiac arrest and cardiopulmonary
resuscitation outcome reports: Update of the utstein resuscitation registry templates for out-of-
hospital cardiac arrest: A statement for healthcare professionals from a task force of the
international liaison committee on resuscitation (american heart association, european
resuscitation council, australian and new zealand council on resuscitation, heart and stroke
foundation of canada, interamerican heart foundation, resuscitation council of southern
africa ...[J]. Resuscitation, 2015, 96: 328-340.
22. Stephen S. Dispatcher-Assisted Cardiopulmonary Resuscitation Before and After Institution
of Los Angeles Fire Department Tiered-Dispatch Card System; 2015
23. Herlitz J, Eek M, Engdahl J, et al. Factors at resuscitation and outcome among patients
suffering from out of hospital cardiac arrest in relation to age[J]. Resuscitation, 2003, 58(3): 309-
317.
24. Viera A J, Garrett J M. Understanding interobserver agreement: the kappa statistic[J]. Fam
Med, 2005, 37(5): 360-363.
25. Litwin P E, Eisenberg M S, Hallstrom A P, et al. The location of collapse and its effect on
survival from cardiac arrest[J]. Annals of emergency medicine, 1987, 16(7): 787-791.
26. Bergner L, Eisenberg M S. CBR: Who Should Be Trained?[J]. Emergency health services
quarterly, 1982, 1(3): 41-48.
27. Swor R, Khan I, Domeier R, et al. CPR Training and CPR Performance: Do CPR‐trained
Bystanders Perform CPR?[J]. Academic Emergency Medicine, 2006, 13(6): 596-601.
28. Vaillancourt C, Lui A, De Maio V J, et al. Socioeconomic status influences bystander CPR
and survival rates for out-of-hospital cardiac arrest victims[J]. Resuscitation, 2008, 79(3): 417-
423.
25
Appendix:
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The effect of renal function on toxicity of E7389 (eribulin) among patients with bladder cancer
Asset Metadata
Creator
Zhang, Huihui
(author)
Core Title
Evaluation of factors influencing Los Angeles Tiered-Dispatch System’s improvement on bystander CPR rate and inter reliability between electronic patient care report (ePCR) and 911 call review on...
School
Keck School of Medicine
Degree
Master of Science
Degree Program
Applied Biostatistics and Epidemiology
Publication Date
08/01/2016
Defense Date
07/27/2016
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
911 call review,bystander CPR,Cohen’s Kappa test,dispatch system,field report,inter reliability,OAI-PMH Harvest
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Lane, Christianne Joy (
committee chair
), Mack, Wendy (
committee member
), Sanko, Stephen (
committee member
)
Creator Email
huihuiz@usc.edu,zhanghuihui0122@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-288460
Unique identifier
UC11280683
Identifier
etd-ZhangHuihu-4694.pdf (filename),usctheses-c40-288460 (legacy record id)
Legacy Identifier
etd-ZhangHuihu-4694-2.pdf
Dmrecord
288460
Document Type
Thesis
Format
application/pdf (imt)
Rights
Zhang, Huihui
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
911 call review
bystander CPR
Cohen’s Kappa test
dispatch system
field report
inter reliability