APPENDIX
American Heart Association Get With the Guidelines—Resuscitation (formerly National
Registry of Cardiopulmonary Resuscitation) Investigators
In addition to the authors Paul S. Chan, MD, MSc, Robert A. Berg, MD, and Raina Merchant,
MD, MSHP, members of the Get With The Guidelines-Resuscitation Clinical Work Group and
Adult Task Force include:
Mary Mancini, RN, PhD, University of Texas at Arlington; Emilie Allen, MSN, RN, Parkland
Health and Hospital System; Steven Bradley, MD and Comilla Sasson, MD, MS, University of
Colorado; Scott Braithwaite, MD, Yale University School of Medicine; Michael W. Donnino,
MD, Beth Israel Deaconess Medical Center; Dana P. Edelson, MD, University of Chicago; Brian
Eigel, PhD and Lana Gent, PhD, American Heart Association; Robert T. Faillace, MD, St.
Joseph’s Regional Center; Romergryko G. Geocadin, MD, Johns Hopkins School of Medicine;
Elizabeth A. Hunt, MD, MPH, PhD, Johns Hopkins Medicine Simulation Center; Lynda Knight,
RN, Lucile Packard Children’s Hospital at Stanford; Kenneth LaBresh, RTI International;
Vincent N. Mosesso, Jr., MD, University of Pittsburgh School of Medicine; Vinay Nadkarni,
MD, University of Pennsylvania School of Medicine; Graham Nichol, MD, MPH and Samuel
Warren, MD, University of Washington; Joseph P. Ornato, MD and Mary Ann Peberdy, MD,
Virginia Commonwealth University Medical Center; and Mindy Smyth, MSN, RN
1
Supplementary Table 1. Patient Characteristics by Survival to Hospital Discharge
Survived to
Discharge
Died
(n = 10,290)
(n = 38,551)
P Value
Demographics
Age
< 0.001
18 to <50
1749 (17.0%)
6114 (15.9%)
50 to 59
1973 (19.2%)
6335 (16.4%)
60 to 69
2551 (24.8%)
8142 (21.1%)
70 to 79
2343 (22.8%)
9012 (23.4%)
80 to 89
1674 (16.3%)
8948 (23.2%)
5995 (58.3%)
22501 (58.4%)
> 90
Male sex
Race
0.85
< 0.001
White
7545 (73.3%)
26368 (68.4%)
Black
1680 (16.3%)
8286 (21.5%)
Other
384 (3.7%)
1502 (3.9%)
Unknown
681 (6.6%)
2395 (6.2%)
Respiratory insufficiency
3422 (33.3%)
16519 (42.8%)
< 0.001
Renal insufficiency
2625 (25.5%)
13583 (35.2%)
< 0.001
Arrhythmia
3651 (35.5%)
11296 (29.3%)
< 0.001
Pre-Existing Conditions
2
Diabetes mellitus
3324 (32.3%)
11605 (30.1%)
< 0.001
Hypotension
1644 (16.0%)
11077 (28.7%)
< 0.001
Heart failure this admission
1719 (16.7%)
6329 (16.4%)
0.48
Prior heart failure
1933 (18.8%)
7439 (19.3%)
0.24
Myocardial infarction this admission
2103 (20.4%)
5582 (14.5%)
< 0.001
Prior myocardial infarction
1725 (16.8%)
5385 (14.0%)
< 0.001
Metabolic or electrolyte abnormality
967 (9.4%)
6078 (15.8%)
< 0.001
Sepsis
952 (9.3%)
7344 (19.1%)
< 0.001
Pneumonia
1040 (10.1%)
5541 (14.4%)
< 0.001
Metastatic or hematologic malignancy
703 (6.8%)
5340 (13.9%)
< 0.001
Hepatic insufficiency
379 (3.7%)
3270 (8.5%)
< 0.001
Baseline depression in CNS function
896 (8.7%)
4597 (11.9%)
< 0.001
Acute CNS non-stroke event
608 (5.9%)
2781 (7.2%)
< 0.001
Acute stroke
325 (3.2%)
1514 (3.9%)
< 0.001
Major trauma
339 (3.3%)
1728 (4.5%)
< 0.001
Characteristics of arrest
Cardiac arrest rhythm
< 0.001
Asystole
2586 (25.1%)
13902 (36.1%)
Pulseless electrical activity
3813 (37.1%)
19167 (49.7%)
Ventricular fibrillation
2447 (23.8%)
3106 (8.1%)
Pulseless ventricular tachycardia 1444 (14.0%)
2376 (6.2%)
Hospital Location
< 0.001
3
Intensive care unit
4136 (40.2%)
19453 (50.5%)
Monitored unit
1925 (18.7%)
5648 (14.7%)
Non-Monitored unit
1385 (13.5%)
7071 (18.3%)
Emergency room
1334 (13.0%)
3660 (9.5%)
Procedural or surgical area
1218 (11.8%)
1987 (5.2%)
Other
292 (2.8%)
732 (1.9%)
Mechanical ventilation
2268 (22.0%)
13901 (36.1%)
< 0.001
IV Vasopressor
1687 (16.4%)
12662 (32.8%)
< 0.001
Pulmonary artery catheter
343 (3.3%)
868 (2.3%)
< 0.001
Dialysis
212 (2.1%)
1549 (4.0%)
< 0.001
Intra-aortic balloon pump
172 (1.7%)
538 (1.4%)
0.04
Interventions in Place
4
Supplementary Table 2. Definitions of Select Variables in Models
Myocardial infarction this admission—Documented diagnosis of acute coronary syndrome or
myocardial infarction during the index admission.
Prior heart failure—Documented diagnosis of congestive heart failure prior to this admission.
Renal insufficiency—Evidence for any of the following within 24 hours of cardiac arrest:

Requirement for ongoing dialysis or extracorporeal filtration therapies.

Serum creatinine > 2 mg/dL
Hepatic insufficiency—Evidence for any of the following within 24 hours of cardiac arrest:

Total bilirubin > 2 mg/dL and AST > 2x normal

Cirrhosis
Hypotension – Evidence for any of the following within 24 hours of cardiac arrest:

SBP < 90 or MAP < 60 mmHg.

Vasopressor/inotropic requirement after volume expansion (except for dopamine ≤ 3
mcg/kg/min).

Intra-aortic balloon pump
Septicemia—Documented bloodstream infection where antibiotics have not yet been started or
the infection is still being treated with antibiotics.
5
Acute stroke—Documented diagnosis of an intracranial/intraventricular hemorrhage or
thrombosis during the index admission prior to cardiac arrest.
Diabetes mellitus— Documented diagnosis of either Type I or Type II diabetes mellitus.
Metabolic/electrolyte abnormality—Evidence for any of the following within 24 hours of
cardiac arrest:

Sodium < 125 or > 150 mEq/L

Potassium < 2.5 or > 6 mEq/L

pH < 7.3 or > 7.5, arterial

Lactate > 2.5 mmol/L,

Blood glucose < 60 mg/dL
Metastatic or hematologic malignancy—Documentation of any solid tissue malignancy with
evidence of metastasis, or any blood borne malignancy.
Major trauma—Evidence of multi-system injury or single system injury associated with shock
or altered mental status during the index hospitalization prior to cardiac arrest.
Mechanical ventilation—Requirement for assisted ventilation via an endotracheal tube or
tracheostomy within 24 hours of cardiac arrest.
6
Dialysis—Requirement for hemodialysis, peritoneal dialysis, or continuous arteriovenous or
veno-venous hemofiltration/dialysis prior to the time of the cardiac arrest
IV vasopressor—Continuous intravenous infusion of at least one of the following vasoactive
agents at the time of cardiac arrest:

Dobutamine

Dopamine > 3 mcg/kg/min

Epinephrine

Norepinephrine

Phenylephrine

Other vasoactive agent
7
Supplementary Table 3. Formula for Calculating Risk-Standardized Hospital Survival
Rates
1) To calculate the expected survival rate for the patient, first determine the sum of all the
predictor coefficients (see below), which will be designated as ‘Patient Beta’ for each patient.
The expected survival rate for the patient can then be calculated by the following formula:
Exponent (Patient Beta)
(1 + exponent [Patient Beta])
To calculate ‘Patient Beta’, we used coefficients from Table 3 in the manuscript:
Start with the baseline average hospital intercept of 1.380. Then, for each patient:

+ 0 if Age is <50, +0.003 if 50-69, -0.010 if 60-69, -0.256 if 70-79, and -0.656 if >80

+ 0 if Initial Cardiac Arrest Rhythm is asystole, +0.048 if PEA, +1.263 if VF, and +1.129
if pulseless VT

+ 0 if Hospital Location is non-monitored unit, +0.564 if ICU, +0.482 if monitored
telemetry unit, +0.562 if ER, +1.155 if procedural area, and +0.621 if other hospital location

0.475 if Hypotension present

0.488 if Sepsis present

0.735 if Metastatic or Hematologic Malignancy present

0.724 if Hepatic Insufficiency present

0.566 if patient on Mechanical Ventilation prior to cardiac arrest

0.733 if patient already on Intravenous Vasopressors at the time of cardiac arrest
8
2) To calculate the expected survival rate for a given hospital, calculate the expected survival
rate for each patient with an in-hospital cardiac arrest (#1 above) and then determine the average
of the expected survival rates of all patients in a given hospital.
3) To calculate the predicted survival rate for a given hospital, the hospital’s specific random
effect intercept is used (not shown, as it is specific to each participating hospital in a given study
sample), rather than the average hospital intercept of 1.380 from the formula above.
4) The hospital’s risk-standardized survival rate is then calculated as the ratio of the hospital’s
predicted to expected survival rate, multiplied by the unadjusted rate for the entire study sample.
9
Supplementary Table 4. Sensitivity Analysis of Study Findings
To examine the possibility that the risk-standardized survival rates in the top 2 hospital quartiles
were due to higher rates of DNR among all admitted patients at these hospitals, we conducted a
sensitivity analysis wherein the DNR rate at hospitals in the upper and second highest quartiles
were modeled to be 100% and 50% higher, respectively, than the DNR rate at hospitals in the
lower 2 quartiles. We found that only 2 out of the 136 hospitals in the top 2 quartiles of riskstandardized survival changed quartiles with these assumptions, thereby suggesting that our
findings were unlikely to be influenced by higher rates of DNR at hospitals with better survival
outcomes.
Risk-Standardized Rate (Original)
Risk-Standardized Rate
Quartile 1
Quartile 2
Quartile 3
Quartile 4
(Sensitivity Analyses)
(<18.4%)
(18.4%-21.0%)
(21.1%-23.7%)
(>23.7%)
Total
Quartile 1 (<9.3%)
67
0
0
68
Quartile 2 (9.3-10.7%)
1
67
0
0
68
Quartile 3 (10.8-12.2%)
0
0
68
0
68
Quartile 4 (>12.2%)
0
0
0
68
68
Total
68
68
68
68
272
10