Delay in Presentation and Evaluation for Acute Stroke

Stroke Time Registry for Outcomes Knowledge and

Epidemiology (S.T.R.O.K.E.)
Clifton R. Lacy, MD; Dong-Churl Suh, MBA, PhD; Maureen Bueno, PhD; John B. Kostis, MD; for
the S.T.R.O.K.E. Collaborative Study Group*

Background and Purpose—Early treatment is a critical determinant of successful intervention in acute stroke. The study
was designed to find current patterns of stroke care by determining delays in time from onset of signs or symptoms to
arrival at the emergency department and to initial evaluation by physicians and by identifying factors associated with
these delays.
Methods—Data were prospectively collected by nurses and physicians from patients, patients’ family members, and
medical records from 10 hospitals of the Robert Wood Johnson Health System in New Jersey.
Results—A total of 553 patients who presented with signs or symptoms of acute stroke were studied. Thirty-two percent
of patients arrived at the emergency department within 1.5 hours of stroke onset. Forty-six percent of patients arrived
within 3 hours and 61% within 6 hours. Delays in arrival time were significantly associated with sex, race, transportation
mode, and history of cardiovascular disease. Patients arriving by ambulance were more likely to present earlier (odds
ratio [OR] 3.7 for arrival within 3 hours; OR 4.5 for arrival within 6 hours). Patients arriving by ambulance (OR 2.3
within 15 minutes; OR 1.7 within 30 minutes) and those requiring admission to intensive care units (OR 4.5 within 15
minutes and OR 5.2 within 30 minutes) were examined sooner by physicians.
Conclusions—Despite national efforts to promote prompt stroke evaluation and treatment, significant delays still exist. The
lack of improvement throughout the past decade underscores the need for implementation of effective public health
programs designed to minimize the time to evaluation and treatment of stroke. (Stroke. 2001;32:63-69.)
Key Words: emergency service, hospital 䡲 registries 䡲 stroke, acute
Downloaded from http://ahajournals.org by on June 22, 2023

S troke is one of the leading causes of death and serious,

long-term disability in the United States, annually affect-
ing ⬇600 000 people and causing 160 000 deaths—the third
most common cause of death after heart disease and cancer.1
The economic burden of stroke on society was estimated to
be $45 billion in 1999, with direct costs (ie, hospitals,
physicians, rehabilitation, and pharmaceuticals) amounting to
$29 billion and indirect costs such as lost productivity
totaling $16 billion.1
Early treatment is crucial in maximizing the benefit of
stroke intervention. Effective thrombolytic therapy is depen-
dent on timely intervention,2–5 and guidelines for use of
recombinant tissue plasminogen activator recommend ther-
apy within 3 hours after onset of stroke symptoms.6 Clinical
studies suggest that cerebral ischemia persisting ⬎6 hours
results in permanent neurological damage.7 Thus, early hos-
pital arrival is critical to successful stroke treatment.
Several studies have demonstrated delays in stroke
care,2– 4,8 –22 but only few have attempted to study the deter-
minants of delay. In addition, these studies were conducted in
small geographic areas and yielded conflicting results. This
study, conducted at 10 New Jersey hospitals, prospectively
investigated factors associated with the time delay from onset
of stroke signs or symptoms to arrival at the emergency
department (ED) and time from arrival at the ED to patient
evaluation.
Subjects and Methods
Hospital Setting and Patients
The study was conducted at the 10 hospitals of the Robert Wood
Johnson Health System, located in 5 counties in New Jersey,
spanning the spectrum from tertiary care academic health centers to
general community hospitals, and ranging in size from 134 to 453
beds.

Received May 15, 2000; final revision received September 7, 2000; accepted September 7, 2000.
From the Center for Disease Management and Clinical Outcomes, University of Medicine and Dentistry of New Jersey–Robert Wood Johnson Medical
School, New Brunswick, NJ (C.R.L., M.B., J.B.K.); Robert Wood Johnson University Hospital, New Brunswick, NJ (C.R.L., M.B., J.B.K.); Robert Wood
Johnson Health System, New Brunswick, NJ (C.R.L.); and Rutgers–The State University of New Jersey, College of Pharmacy, Piscataway, NJ (C.R.L.,
D.-C.S.).
* Participating Investigators in the S.T.R.O.K.E. study are listed in the Appendix.
Correspondence to Clifton R. Lacy, MD, Division of Cardiovascular Diseases and Hypertension, UMDNJ–Robert Wood Johnson Medical School, One
Robert Wood Johnson Pl, New Brunswick, NJ 08903-0019. E-mail lacycr@umdnj.edu
© 2001 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org

63
 64 Stroke January 2001

All patients with signs or symptoms of acute stroke who arrived at TABLE 1. Descriptive Characteristics of Stroke Patients
the EDs of study hospitals were included. Patients whose stroke
occurred during inpatient hospitalization were excluded from the Frequency
analysis. Characteristic (n⫽553)
Age (mean⫾SD) (n⫽548) 73.4⫾13.0
Data Collection ⬍55 y 52 (10)
Data were prospectively collected by nurses and physicians from
patients, patients’ family members, and medical records between 55–64 y 61 (11)
September 1, 1996, and March 31, 1997. Time from onset of stroke 65–74 y 147 (27)
signs or symptoms to arrival at the ED was recorded. Stroke onset 75–84 y 188 (34)
was defined as the time a neurological deficit was first noticed by the
patient or an observer. If symptoms were present on awakening, the ⱖ85 y 100 (18)
stroke onset time was considered to be the time the patient fell Sex (n⫽547)
asleep. The time delay to first physician evaluation was defined as Female 292 (53)
the interval between the ED arrival time and physician examination.
Marital status (n⫽502)
Statistical Analyses Married 247 (49)
Statistical analyses were conducted with the use of SAS statistical Single 58 (12)
software.23 Two principal sets of time intervals were analyzed: time
Divorced 15 (3)
from onset of stroke signs or symptoms to arrival at the ED and time
from arrival at the ED to physician evaluation. Time windows of Widowed 182 (36)
arrival at the ED more than 3 and 6 hours as well as time windows Race (n⫽501)
for evaluation at the ED after 15 and 30 minutes were prospectively
White 408 (81)
determined for data analysis.
A panel of 4 senior hospital staff prospectively classified study Black 61 (12)
hospitals into 3 levels of complexity (low, middle, and high) based Asian 14 (3)
on size (number of beds), types of services offered (eg, open heart
surgery, trauma center), case mix index for hospital services, and Hispanic 14 (3)
teaching versus nonteaching status. Other 4 (1)
Univariate odds ratios (ORs) were calculated for each patient’s Insurance (n⫽519)
demographics, mode of transportation to the ED, history of cardio-
vascular disease, and hospital complexity in relation to ED arrival Commercial 52 (10)
and initial evaluation times. The number of past cardiovascular HMO 32 (6)
disease diagnoses was included as a proxy for the degree of Medicare 381 (74)
Downloaded from http://ahajournals.org by on June 22, 2023

cardiovascular health impairment.

Study models were constructed by means of logistic regression to Medicaid 23 (4)
evaluate the effect of patient characteristics and other variables on Uninsured 31 (6)
the likelihood of time delay to arrival and evaluation after onset of Employment (n⫽448)
stroke. A forward stepwise logistic regression model was constructed
at a significance level of 0.2 for variable entry into the model.24 Yes 48 (11)
When any of the variables were selected by the use of the forward No 400 (89)
procedure, then variables associated with characteristics of the Mode of transportation (n⫽499)
variable were included in the model. ORs and 95% confidence
intervals (CI) were calculated from the logistic coefficients of the Self/family 174 (35)
variables. Ambulance 325 (65)
Logistic regression was also used to analyze the time interval from
Type of stroke (n⫽526)
arrival at the ED to the initial physician evaluation by means of the
variables that were assumed to affect this time period. Ischemic 279 (53)
Hemorrhagic 58 (11)
Results Unknown or CT not performed 189 (36)
Patient Characteristics History
Characteristics of the study patients are shown in Table 1. A Stroke (n⫽500) 150 (30)
total of 553 patients were studied. As the result of missing TIA (n⫽489) 76 (16)
information, the number of patients observed in each category Atrial fibrillation (n⫽486) 68 (14)
was less than the total number of patients studied. Each of the
Myocardial infarction (n⫽488) 65 (13)
10 hospitals contributed between 20 and 79 subjects to the
Congestive heart failure (n⫽491) 57 (12)
study. The mean age was 73.4 years, and 53% of patients
were women. Almost half (49%) of the patients were married, Values in parentheses are percent.
36% were widowed, 12% single, and 3% divorced. Eighty-
ischemic (53%). A history of cerebrovascular accident was
one percent of patients were white, 12% were black, 3% were
reported in 30%, transient ischemic attack in 16%, atrial
Hispanic, and 3% were Asian. Eighty-nine percent of patients fibrillation in 14%, myocardial infarction in 13%, and con-
were not employed. Seventy-four percent of patients were gestive heart failure in 12% of patients.
insured by Medicare, 10% by commercial insurance, 6% in
managed care, and 6% were not insured. Time to Emergency Department Arrival
Two-thirds of patients arrived at the hospital by ambu- The Figure illustrates the distribution of arrival time at the ED
lance. The most common type of stroke observed was by age. Thirty-two percent of patients arrived at the ED

Time from onset of stroke to arrival in ED. Top bar graph depicts number of patients arriving within each time interval by age group.
Bottom bar graph depicts percentage of patients arriving within each time interval by age group.
within 1.5 hours of stroke onset. Forty-six percent of patients
arrived within 3 hours after onset of symptoms, and 61%
arrived within 6 hours. Forty-nine percent of patients ⱖ65
years of age and 34% of patients ⬍65 years of age arrived
Downloaded from http://ahajournals.org by on June 22, 2023
within 3 hours of stroke onset. Sixty-three percent of patients
ⱖ65 years of age and 52% of patients ⬍65 years of age
arrived within 6 hours of stroke onset.
Table 2 shows the relation between arrival time at the ED
after the onset of stroke signs or symptoms and study
variables such as age, sex, race, and mode of transportation to
the ED. Marital status, employment status, and insurance type
are not presented because these variables did not significantly
influence arrival time at the ED.
Univariate ORs were calculated for arrival times at the ED
within 3 and 6 hours. In general, patients ⬎55 years of age
were more likely to arrive within 3 hours than were younger
patients. Although analysis of all age groups does not show a
statistically significant difference at 6 hours, patients ⱖ65
years of age were more likely to arrive earlier than were
younger patients. There was no statistically significant sex
difference with respect to arrival within 3 hours, but female
patients were more likely than male patients to arrive within
6 hours (P⫽0.038).
Black patients had a significantly greater likelihood of
arriving later than did white patients within the 3-hour time
window (P⫽0.024) but did not arrive significantly later
within the 6-hour interval. The mode of transportation sig-
nificantly affected time to arrival at the ED. Patients who
arrived by using their own vehicles tended to arrive later than
those who arrived by ambulance at both the 3-hour and
6-hour intervals (P⫽0.0001).
Patients with a history of atrial fibrillation or congestive heart
failure were statistically more likely to arrive earlier than patients
with no history of these diseases at both the 3-hour (P⫽0.002,
Lacy et al S.T.R.O.K.E. Study 65
P⫽0.001, respectively) and the 6-hour intervals (P⫽0.001,
P⫽0.002). Patients with previous cerebrovascular accident,
transient ischemic attack, or myocardial infarction tended to
have a shorter time to arrival than those patients without such a
history at the 3-hour interval, but these trends were not statisti-
cally significant. Patients with a history of ⱖ2 cardiovascular
diagnoses were more likely to arrive within 3 hours than patients
without a history of such disease (P⫽0.004).
The adjusted ORs calculated by multiple logistic regres-
sion models are also presented in Table 2. In general, the
adjusted ORs in the model were similar to the ORs obtained
from the univariate analysis. In this analysis, patients ⬎55
years of age were more likely to arrive within 3 hours than
were younger patients. Patients 65 to 74 years of age were
more likely to arrive within 3 hours, which is significantly
earlier than patients ⱕ55 years of age (P⫽0.039). With the
6-hour cutoff, only patients 65 to 74 years of age were more
likely to arrive earlier than were patients ⬍55 years, but this
trend was not statistically significant when controlling for
other variables in the model. The mode of transportation was
significantly associated with delay (P⫽0.0001 for 3-hour and
6-hour intervals). Patients with history of ⱖ2 cardiovascular
diagnoses were more likely to arrive earlier than patients
without a history of such disease (P⫽0.046 for 3 hours and
P⫽0.009 for 6 hours).
Waiting Time for Initial Examination
by Physicians
The univariate ORs of time interval from arrival at the ED to
initial physician evaluation are presented in Table 3. Because
the waiting time for the initial physician examination was not
significantly influenced by patient age, this variable was not
presented in the Table. Hispanic patients were significantly
more likely to be seen by physicians later than 15 minutes
 66 Stroke January 2001

TABLE 2. Delayed ED Arrival After Stroke Symptoms: Univariate Analyses and Adjusted OR
Delayed ED Arrival After Stroke

Delay ⱖ3 h* Delay ⱖ6 h*

Univariate OR Adjusted OR Univariate OR Adjusted OR

Variables No. (95% CI) (95% CI) No. (95% CI) (95% CI)
Age, y
⬍55 28 1 1 19 1 1
55–64 33 0.83 (0.35–1.96) 0.70 (0.27–1.84) 25 1.16 (0.50–2.65) 1.22 (0.48–3.07)
65–74 57 0.52 (0.24–1.11) 0.41 (0.17–0.96)‡ 42 0.78 (0.37–1.63) 0.72 (0.32–1.63)
75–84 72 0.48 (0.23–1.01) 0.55 (0.23–1.28) 52 0.71 (0.35–1.45) 1.09 (0.48–2.46)
ⱖ85 43 0.52 (0.24–1.16) 0.63 (0.26–1.56) 31 0.76 (0.35–1.64) 1.27 (0.53–3.05)
Sex
Male 112 1 䡠䡠䡠 88 1 䡠䡠䡠
Female 120 0.83 (0.57–1.21) 䡠䡠䡠 80 0.66 (0.45–0.98)‡ 䡠䡠䡠
Race
White 177 1 䡠䡠䡠 128 1 䡠䡠䡠
Black 37 2.00 (1.10–3.64)‡ 䡠䡠䡠 26 1.55 (0.87–2.74) 䡠䡠䡠
Asian 8 1.56 (0.50–4.85) 䡠䡠䡠 5 1.08 (0.35–3.37) 䡠䡠䡠
Hispanic 7 1.13 (0.37–3.44) 䡠䡠䡠 6 1.48 (0.49–4.50) 䡠䡠䡠
Other 2 1.94 (0.18–21.6) 䡠䡠䡠 2 3.45 (0.31–38.6) 䡠䡠䡠
Transportation
Self/family 118 1 1 99 1 1
Ambulance 113 0.27 (0.17–0.41)§ 0.29 (0.18–0.45)§ 68 0.22 (0.14–0.33)§ 0.22 (0.14–0.35)§
History†
CVA 69 0.97 (0.64–1.46) 䡠䡠䡠 50 0.97 (0.64–1.48) 䡠䡠䡠
Downloaded from http://ahajournals.org by on June 22, 2023

TIA 30 0.77 (0.45–1.32) 䡠䡠䡠 20 0.71 (0.40–1.26) 䡠䡠䡠

AF 23 0.43 (0.25–0.74)§ 䡠䡠䡠 13 0.34 (0.18–0.66)§ 䡠䡠䡠
CHF 18 0.37 (0.20–0.67)§ 䡠䡠䡠 11 0.34 (0.17–0.69)§ 䡠䡠䡠
MI 28 0.74 (0.43–1.28) 䡠䡠䡠 16 0.53 (0.29–0.98)§ 䡠䡠䡠
No. of history diagnoses
0 116 1 1 89 1 1
1 75 1.07 (0.69–1.68) 1.08 (0.67–1.76) 54 0.94 (0.60–1.47) 0.87 (0.53–1.43)
ⱖ2 40 0.50 (0.31–0.80)§ 0.59 (0.35–0.99)‡ 24 0.42 (0.25–0.71)§ 0.47 (0.26–0.83)§
CVA indicates cerebrovascular accident; AF, atrial fibrillation; CHF, congestive heart failure; and MI, myocardial infarction.
*Patient arrival time was analyzed with 2 time windows. Delay ⱖ3 hours refers to patients who arrived ⱖ3 hours from onset of
stroke signs and symptoms. Delay ⱖ6 hours refers to patients who arrived ⱖ6 hours from onset of stroke signs and symptoms.
†OR represents ratio of probability of patients who have a history of cardiovascular disease arriving at the ED to the probability of
patients who do not have a history of such disease arriving at the ED for each disease.
‡P⬍0.05, §P⬍0.01.

after arrival than were white patients (P⫽0.004). Medicaid
patients also had a higher probability of being examined by
physicians beyond the 15-minute or 30-minute mark when
compared with Medicare patients (P⫽0.047). Increasing
hospital complexity was significantly associated with delays
in physician examination at both 15- and 30-minute intervals.
Patients who were admitted to the intensive care unit (ICU)
were examined significantly earlier than those who were
admitted to medical or surgical units at both 15- and 30-
minute intervals (P⫽0.0001). In addition, patients who were
transported by ambulance were examined earlier by physi-
cians than were patients who had self-transport or family
transport at both time intervals (P⫽0.0001 for 15- and
30-minute intervals).
Table 3 presents adjusted ORs calculated by the multivariate
logistic regression model including significant variables such as
race, insurance, level of hospital complexity, admission unit, and
transportation mode. Overall, when variables in the model were
controlled at both the 15- and 30-minute time windows, initial
physician examination time was significantly influenced by the
same variables that had a significant influence on physician
examination time at the univariate analysis.
Mode of Transportation to the Hospital
Because patients arriving by ambulance presented earlier and
were examined sooner, factors associated with the use of
ambulance versus other types of transportation were ana-
lyzed. As shown in Table 4, patients ⬎75 years of age used
 Lacy et al S.T.R.O.K.E. Study 67

TABLE 3. Time Interval From Arrival in ED to Initial Attendance by Physician

Time Interval From Arrival in ED to Initial Physician Attendance

Time Interval ⱖ15 min* Time Interval ⱖ30 min*

Univariate OR Adjusted OR Univariate OR Adjusted OR

Variables No. (95% CI) (95% CI) No. (95% CI) (95% CI)
Race
White 149 1 1 96 1 1
Black 27 1.29 (0.75–2.23) 0.70 (0.36–1.36) 16 1.09 (0.59–2.02) 0.76 (0.37–1.58)
Asian 3 0.53 (0.14–1.97) 0.21 (0.04–1.07) 2 0.60 (0.13–2.79) 0.34 (0.06–2.09)
Hispanic 12 9.46 (2.09–42.9)‡ 7.95 (1.51–41.6)† 5 1.67 (0.54–5.09) 1.59 (0.43–5.84)
Insurance
Medicare 138 1 1 89 1 1
Commercial 10 0.52 (0.25–1.11) 0.33 (0.14–0.79)† 7 0.64 (0.27–1.49) 0.47 (0.18–1.20)
HMO 16 2.03 (0.93–4.42) 1.89 (0.74–4.82) 5 0.63 (0.23–1.71) 0.45 (0.15–1.37)
Medicaid 16 4.06 (1.55–10.6)‡ 4.46 (1.45–13.7)‡ 10 2.42 (1.01–5.81)† 2.61 (0.92–7.43)
Uninsured 10 0.72 (0.33–1.59) 0.76 (0.30–1.90) 5 0.56 (0.21–1.50) 0.64 (0.22–1.89)
Hospital complexity
Low 49 1 1 25 1 1
Middle 87 1.97 (1.29–3.01)‡ 1.59 (0.95–2.67) 49 1.96 (1.16–3.32)‡ 1.53 (0.83–2.84)
High 77 5.87 (3.54–9.74)‡ 7.02 (3.88–12.7)‡ 56 6.30 (3.61–11.0)‡ 6.94 (3.71–13.0)‡
Admission unit
Medical/surgical 93 1 1 58 1 1
ICU 16 0.22 (0.12–0.41)‡ 0.32 (0.16–0.66)‡ 7 0.19 (0.08–0.47)‡ 0.33 (0.13–0.81)†
Intermediate/cluster 60 0.71 (0.46–1.10) 0.87 (0.51–1.47) 37 0.76 (0.47–1.23) 1.05 (0.58–1.88)
Other 18 0.84 (0.43–1.68) 1.13 (0.50–2.51) 13 1.09 (0.52–2.26) 1.69 (0.73–3.92)
Downloaded from http://ahajournals.org by on June 22, 2023

Transportation
Self/family 85 1 1 50 1 1
Ambulance 103 0.43 (0.29–0.63)‡ 0.44 (0.27–0.70)‡ 65 0.57 (0.37–0.88)‡ 0.60 (0.30–0.87)†
*Time interval to initial examination was analyzed with 2 time windows. Delay ⱖ15 minutes refers to patients who were initially
examined by physicians ⱖ15 minutes after arrival at the ED. Delay ⱖ30 minutes refers to patients who were initially examined by
physicians ⱖ30 minutes after arrival at the ED.
†P⬍0.05, ‡P⬍0.01.

ambulance transport significantly more than did patients ⬍55 Discussion

years of age (P⫽0.001 for patients 75 to 84 years of age;
P⫽0.004 for patients ⬎85 years of age). Patients who were
not employed were more likely to call for an ambulance than
those employed (P⫽0.002). Nonwhite patients were signifi-
cantly less likely to arrive by ambulance than whites
(P⫽0.005 for blacks; P⫽0.042 for Asians; P⫽0.042 for
Hispanics). Commercial or health maintenance organization
(HMO) insurance patients were significantly less likely to use
ambulance transportation than were Medicare patients
(P⫽0.019, P⫽0.030, respectively). Patients with a history of
atrial fibrillation or congestive heart failure (P⫽0.0008,
P⫽0.005, respectively) and patients with a history of ⬎2
prior cardiovascular disease diagnoses (P⫽0.002) were more
likely to use an ambulance than were those without a history
of heart disease.
Adjusted ORs calculated by a multivariate logistic regres-
sion model demonstrated that blacks and Asians were less
likely to use ambulance transportation than were whites
(P⫽0.032, P⫽0.045, respectively). Patients with a history of
ⱖ2 cardiovascular diagnoses were also identified to be
significant predictors of ambulance use (P⫽0.020).
This study found that 46% of patients arrived within 3 hours
of the onset of stroke signs or symptoms (the critical time for
initiation of thrombolytic therapy25) and 61% within 6 hours.
These delays have not improved from those of earlier national
and international studies of stroke. Those studies found that
25% to 59% of stroke patients arrived at the ED within 3
hours and 35% to 66% of patients arrived within 6
hours.2,3,9 –15,26
Consistent with prior investigations, this study found that
sex did not significantly affect arrival time.2,3,15 Age ⬎70
years was a factor delaying presentation in a previous study,14
whereas our study found that patients 65 to 74 years of age
were likely to arrive sooner than patients ⬍55 years of age.
This may be related to the presence of organized retirement
communities in the geographic areas studied. Another expla-
nation is that older patients are more likely to perceive stroke
symptoms as a serious occurrence based on personal experi-
ence or exposure to stroke patients in their communities. A
significant relation between arrival time and marital status,
employment, or insurance type was not observed in our study
or in a previous study.3
 68 Stroke January 2001

TABLE 4. Mode of Transportation Used by Patients graphic groups were less likely than whites to use an
Likelihood of Using Ambulance
ambulance, a likely explanation for the longer delay in
treatment observed in these demographic groups. The use
Univariate OR Adjusted of an ambulance was associated with earlier arrival at the
Variable No. (95% CI) OR (95% CI) ED than self-transportation or family transportation in this
Age, y and previous studies.15–17,26,29 A study of arrival time in
⬍55 21 1 1 patients with myocardial infarction yielded similar re-
55–64 28 1.06 (0.48–2.32) 0.73 (0.31–1.71) sults.27 In addition, 78% of patients who had a history of
65–74 62 1.03 (0.52–2.06) 0.72 (0.30–1.74)
ⱖ2 cardiac diagnoses used an ambulance compared with
61% of patients without a history of heart disease.
75–84 39 3.21 (1.60–6.44)‡ 2.13 (0.85–5.37)
Signs and symptoms of severe stroke have been shown to
ⱖ85 22 3.04 (1.41–6.54)‡ 1.91 (0.70–5.26)
favor early hospital arrival, whereas those of mild stroke
Employment 䡠䡠䡠 often result in delayed presentation.12 A history of illness
Yes 26 1 䡠䡠䡠 probably contributes to heightened awareness and recognition
No 127 2.70 (1.43–5.02)‡ 䡠䡠䡠 of stroke symptoms, prompting earlier pursuit of medical
Marital status attention. Our data suggest that if patients realize that they are
Married 95 1 1 having a stroke or have symptoms comparable to previous
Single 23 0.95 (0.52–1.71) 0.69 (0.34–1.39)
experience, they seek medical treatment sooner.
In this study, patients who presented to more complex
Divorced 7 0.75 (0.27–2.15) 0.67 (0.20–2.21)
hospitals waited longer to be seen by a physician than those
Widowed 48 1.81 (1.19–2.76)‡ 1.19 (0.73–1.93)
admitted to less complex hospitals. A possible explanation
Race lies within the definition of hospital complexity, which is
White 124 1 1 determined in part by the number of beds, types of services
Black 30 0.45 (0.26–0.79)‡ 0.53 (0.28–1.00)† provided, level of intensity and case mix, and teaching status.
Asian 8 0.34 (0.12–1.00)† 0.27 (0.09–0.88)† Factors associated with each of these determinants may
Hispanic 8 0.34 (0.12–1.00)† 0.38 (0.11–1.29) contribute to delay in physician evaluation. Further study is
Insurance
necessary to identify specific contributing factors.
Patients admitted to ICUs were seen earlier than those
Medicare 115 1 1
admitted to medical/surgical units, a finding that is probably
Downloaded from http://ahajournals.org by on June 22, 2023

Commercial 21 0.46 (0.24–0.88)† 0.80 (0.35–1.86)

related to illness severity. Patients who arrived by ambulance
HMO 15 0.43 (0.20–0.92)† 0.93 (0.37–2.30) were seen earlier by physicians, also probably because of the
Medicaid 6 1.31 (0.50–3.41) 2.13 (0.66–6.95) perception of more severe illness. The results of this study are
Uninsured 11 0.80 (0.37–1.73) 2.23 (0.85–5.86) consistent with prior findings that patients who arrive by
History* ambulance are seen earlier by physicians.26,30
Cerebrovascular accident 47 1.13 (0.75–1.71) Although this study was carefully designed and performed
䡠䡠䡠
TIA 21 1.40 (0.81–2.42)
to ascertain patients’ arrival time at the ED and treatment time
䡠䡠䡠
after arrival, some limitations warrant discussion. This study
Atrial fibrillation 11 3.18 (1.62–6.25)‡ 䡠䡠䡠 had potential sampling and measurement errors for time of
Congestive heart failure 10 2.74 (1.35–5.59)‡ 䡠䡠䡠 onset of stroke to arrival at the ED, especially for those
Myocardial infarction 20 1.28 (0.73–2.25) 䡠䡠䡠 patients who awakened with neurological findings. We in-
Number of history cluded patients who arrived at study hospitals with symptoms
diagnoses of stroke during a specific study period rather than randomly
0 94 1 1 throughout the year, preventing assessment of seasonal vari-
1 55 1.02 (0.67–1.57) 0.83 (0.52–1.34) ation and effect of inclement weather on time intervals.
ⱖ2 25 2.25 (1.35–3.76)‡ 1.88 (1.08–3.30)† However, the arrival times in this study are consistent with
*OR represents ratio of the probability of patients who have a history of previous studies, which also relied on information provided
cardiovascular disease arriving at the ED to the probability of patients who do by patients or patient caregivers. Additional limitations of this
not have a history of such disease arriving at the ED, for each disease. study are that data were not collected on (1) education
†P⬍0.05, ‡P⬍0.01. specific to stroke, (2) patient/family members’ understanding
of signs and symptoms of stroke, and (3) changes in delay
In our study and in another recent study,26 black patients over time. Also, severity was not directly assessed other than
took longer than white patients to arrive at the ED. This through the surrogate of need for intensive care.
was not observed in a previous study published in the early Data were collected from a wide variety of hospitals in
1990s.15 Black patients with acute myocardial infarction New Jersey, but study patients’ demographics including race
have also been found to seek treatment later than their were consistent with general trends documented by the US
white counterparts.27,28 In addition, we found that black Census.31 It is yet to be determined whether the results of this
and Hispanic patients were more likely to present to large study can be generalized nationally.
complex hospitals than white patients, probably because Despite national efforts to promote public awareness of the
such hospitals are located in urban areas. These demo- benefits of prompt stroke treatment, this study found that more than

half of patients with acute stroke fail to present to the ED within the
optimal time period for effective intervention. This may be due to
failure to recognize signs and symptoms or lack of awareness of
potential treatment benefits.29 One prior study found that patients
⬎65 years of age were less likely to know the risk factors for stroke
than were patients ⱕ65 years of age.32 This study did not directly
assess patient education or knowledge. Although previous studies
demonstrated that educational efforts to improve the recognition of
stroke symptoms have been shown to reduce time to hospital arrival
after symptom onset,8,15 additional educational initiatives are still
needed to increase public awareness of warning signs and symp-
toms as well as risk factors for stroke.33
In summary, this study showed that although significant
delays still exist, not only were patients who were transported
by ambulance brought to the ED more rapidly but that
physicians examined them sooner. Use of the emergency
medical system shortens arrival time as well as treatment time
for stroke. The lack of improvement in delay to stroke
evaluation and treatment over the past decade underscores the
need for more effective public health programs. Further
efforts to increase public awareness of stroke signs and
symptoms, to disseminate guidelines and recommendations
for stroke evaluation and treatment, and to develop initiatives
(including those targeting healthcare providers, the emer-
gency medical system, and the public at large) may reduce the
time from stroke onset to treatment.
Appendix
The study group consisted of the following research teams of
Downloaded from http://ahajournals.org by on June 22, 2023
doctors, nurses, and coordinators at each collaborative study hospital
and institution in alphabetical order: Bayshore Community Hospital:
J. Jerome Cohen, MD, Lauren Burke, RNC; CentraState Healthcare
System: Benjamin Weinstein, MD, Mary Marinaro, RN; Jersey
Shore Medical Center: Carl Marchetti, MD, Robert Sweeney, MD;
Medical Center of Ocean County: Morris Feitel, MD, Mary Ellen
Bonczek, RN; Muhlenberg Regional Medical Center: Frances Hulse,
MD, Bob Bayly, MD; Rahway Hospital: Uma Viswanathan, MD,
Linda Coughlin, RN; Raritan Bay Medical Center: John Middleton,
MD, Rose Gavin, RN; Robert Wood Johnson University Hospital
and University of Medicine and Dentistry of New Jersey-Robert
Wood Johnson Medical School: Clifton R. Lacy, MD, John B.
Kostis, MD, Maureen Bueno, RN, PhD, Ellen A. Lacy, PsyD,
Andrew Greene, MHCA, Harvey A. Holzberg, MBA; Robert Wood
Johnson University Hospital at Hamilton: Edward Niewiadomski,
MD, Lisa Breza, RN; Rutgers–The State University of New Jersey
College of Pharmacy: Dong-Churl Suh, PhD, Joseph A. Barone,
PharmD, John L. Colaizzi, PhD, Soung-Kook Shin, PhD; Southern
Ocean County Hospital: William Torecki, MD, Ray Bennett, RN.
Acknowledgments
This study was funded in part by an unrestricted grant from
Janssen Pharmaceutica Inc, Titusville, NJ.
References
1. American Heart Association. Heart and Stroke Statistical Update. Dallas
Tex: American Heart Association; 1999.
2. Harper GD, Haigh RA, Potter JF, Castleden CM. Factors delaying hospital
admission after stroke in Leicestershire. Stroke. 1992;23:835–838.
3. Smith MA, Doliszny KM, Shahar E, McGovern PG, Arnett DK, Luepker
RV. Delayed hospital arrival for acute stroke: the Minnesota stroke
survey. Ann Intern Med. 1998;129:190 –196.
4. Wester P, Radberg J, Lundgren B, Peltonen M. Factors associated with
delayed admission to hospital and in-hospital delays in acute stroke and
TIA. Stroke. 1999;30:40 – 48.
Lacy et al S.T.R.O.K.E. Study 69
5. Menon SC, Pandey DK, Morgenstern LB. Critical factors determining
access to acute stroke care. Neurology. 1998;51:427– 432.
6. National Institute of Neurological Disorders and Stroke rt-PA Stroke
Study Group. Tissue plasminogen activator for acute ischemic stroke.
N Engl J Med. 1995;333:1581–1587.
7. Baron JC, von Kummer R, del Zoppo GJ. Treatment of acute ischemic
stroke. Stroke. 1995;26:2219 –2221.
8. Alberts MJ, Perry A, Dawson DV, Bertels C. Effects of public and
professional education on reducing the delay in presentation and referral
of stroke patients. Stroke. 1992;23:352–356.
9. Alberts MJ, Bertels C, Dawson DV. An analysis of time of presentation
after stroke. JAMA. 1990;260:65– 68.
10. Azzimondi G, Bassein L, Fiorani L, Nonino F, Montaguti U, Celin D, Re
G, D’Alessandro R. Variables associated with hospital arrival time after
stroke. Stroke. 1997;28:537–542.
11. Kay R, Woo J, Poon WS. Hospital arrival time after onset of stroke.
J Neurol Neurosurg Psychiatry. 1992;55:973–974.
12. Jorgensen HS, Nakayama H, Reith J, Raaschou HO, Olsen TS. Factors
delaying hospital admission in acute stroke: the Copenhagen stroke study.
Neurology. 1996;47:383–387.
13. Anderson NE, Broad JB, Bonita R. Delays in hospital admission and
investigation in acute stroke. BMJ. 1995;311:162.
14. Fogelholm R, Murros K, Rissanen A, Ilmavirta M. Factors delaying
hospital admission after acute stroke. Stroke. 1995;27:398 – 400.
15. Barsan WG, Brott TG, Broderick JP, Haley EC, Levy DE, Marler JR.
Time of hospital presentation in patients with acute stroke. Arch Intern
Med. 1993;153:2558 –2561.
16. Rosamond WD, Gorton RA, Hinn AR, Hohenhaus SM, Morris DL. Rapid
response to stroke symptoms: the delay in accessing stroke healthcare
(DASH) study. Acad Emerg Med. 1998;5:45–50.
17. Williams LS, Bruno A, Rouch D, Marriott DJ. Stroke patients’
knowledge of stroke. Stroke. 1997;28:912–915.
18. Davalos A, Castillo J, Martinez-Vila E. Delay in neurological attention
and stroke outcome. Stroke. 1995;26:2233–2237.
19. Weissman JS, Stern R, Fielding SL, Epstein AM. Delayed access to
health care: risk factors, reasons, and consequences. Ann Intern Med.
1991;114:325–331.
20. Feldman E, Gordon N, Brooks JM, Brass LM, Fayad PB, Sawaya KL,
Nazareno F, Levine SR. Factors associated with early presentation of
acute stroke. Stroke. 1993;24:1805–1810.
21. Alberts MJ, Brass LM, Perry A, Webb D, Dawson DV. Evaluation times
for patients with in-hospital strokes. Stroke. 1993;24:1817–1822.
22. Foulkes MA, Wolf PA, Price TR, Mohr JP, Hier DB. The stroke data bank:
Design, methods, and baseline characteristics. Stroke. 1988;19:547–554.
23. SAS Institute Inc. SAS/STAT User’s Guide, Version 6. Cary, NC: SAS
Institute; 1989.
24. Agresti A. Categorical Data Analysis. New York, NY: John Wiley &
Sons; 1990.
25. Donnan GA, Davis SM, Chambers BR, Gates PC, Hankey GJ, McNeil JJ,
Rosen D, Stewart-Wynne EG, Tuck RR. Streptokinase for acute ischemic
stroke with relationship to time of administration. JAMA. 1996;276:961–966.
26. Kothari R, Jauch E, Broderick J, Brott T, Sauerbeck L, Khoury J, Liu T.
Acute stroke: delay to presentation and emergency department evaluation.
Ann Emerg Med. 1999;33:3– 8.
27. Clark LT, Bellam SV, Shah AH, Feldman JG. Analysis of prehospital delay
among inner-city patients with symptoms of myocardial infarction: impli-
cations for therapeutic intervention. J Natl Med Assoc. 1992;84:931–937.
28. Dracup K, Moser D, Eisenberg M, Meischke H, Alonzo A, Braslow A.
Causes of delay in seeking treatment for heart attack symptoms. Soc Sci
Med. 1995;40:379 –392.
29. Daley S, Braimah J, Sailor S, Kongable GL, Barch C, Rapp K, Bratina P,
Spilker J, Donnarumma R. Education to improve stroke awareness and
emergent response: the NINDS rt-PA Stroke Study Group. J Neurosci
Nurs. 1997;29:393–396.
30. Bratina P, Greenberg L, Pasteur W, Grotta J. Current emergency department
management of stroke in Houston, Texas. Stroke. 1995;26:409–414.
31. US Census Bureau. Population Estimates Program. Washington, DC: US
Census Bureau; 2000.
32. Kothari R, Sauerbeck L, Jauch E, Broderick J, Brott T, Khoury J, Liu T,
Patients’ awareness of stroke signs, symptoms, and risk factors. Stroke.
1997;28:1871–1875.
33. Pancioli AM, Broderick J, Kothari R, Brott T, Tuchfarber A, Miller R,
Khoury J, Jauch E. Public perception of stroke warning signs and
knowledge of potential risk factors. JAMA. 1998;279:1288 –1292.