ARTICLE IN PRESS

Weekend/Holiday Versus Weekday Hospital Discharge and

Guideline Adherence (from the American Heart Association’s Get
With the Guidelines – Coronary Artery Disease Database)
Sumit Tickoo, MDa, Gregg C. Fonarow, MDb, Adrian F. Hernandez, MDc, Li Liang, PhDc,
and Christopher P. Cannon, MDd,*
Most hospitals have reduced medical staff on weekends. Furthermore, a recent study on
acute myocardial infarction suggested that weekend admissions were associated with
higher mortality compared with weekday admissions. We sought to determine if compli-
ance with guideline recommendations for acute coronary syndrome performance measures
would be worse on weekends/holidays compared with weekdays. We utilized the American
Heart Association’s Get with the Guidelines (GWTG) – Coronary Artery Disease database.
This study included 154,910 patients admitted to 515 various hospitals from January 14,
2000 to April 30, 2007 with acute coronary syndrome (ACS). Patients discharged on
weekdays were older and were more likely to be women, have a history of atrial fibrillation,
cerebral vascular accident/transient ischemic attack and chronic renal insufficiency, and
present with unstable angina. Although patients discharged on the weekends/holidays were
slightly less likely to receive angiotensin-converting enzyme inhibitors than those dis-
charged on weekdays (68.3% vs 69.5%, p <0.0001), all other measures were similar, and a
composite performance measure for 100% compliance was equal in both groups (81.5% vs
81.4%, p ⴝ 0.77). In conclusion, within GWTG participating hospitals, weekend/holiday
staffing provides the same quality of care in ACS for discharge medications and counseling
compared with full weekday staffing. However, there remain further opportunities to
improve utilization of guideline-recommended therapies irrespective of discharge
day. © 2008 Elsevier Inc. All rights reserved. (Am J Cardiol 2008;xx:xxx)

The aim of this study was to analyze whether conformity
with guideline recommendations at hospital discharge are
enhanced when acute coronary syndrome (ACS) patients
are discharged on weekdays compared with the weekends/
holidays.
Methods
We used the American Heart Association’s Get With The
Guidelines (GWTG) – Coronary Artery Disease database.
GWTG is a continuous quality improvement program that
collects concurrent and retrospective patient level adherence
data within participating U.S. hospitals.1,2 The components
of the GWTG program, previously described, include orga-
nizational stakeholder and opinion leader meetings, hospital
recruitment, collaborative learning sessions, hospital tool
kits, local clinical champions, and hospital recognition.1,2
Data collection, decision support and hospital data feedback
via multiple on-demand reports of performance on all key
measures are computed with an Internet-based patient man-
a
Cardiology Division, Danbury Hospital, Danbury, Connecticut; bUni-
versity of California, Los Angeles Medical Center, Los Angeles, Califor-
nia; cDuke Clinical Research Institute, Durham, North Carolina; and dTIMI
Study Group, Cardiovascular Division, Brigham and Women’s Hospital,
Boston, Massachusetts. Manuscript received October 27, 2007; revised
manuscript received and accepted April 29, 2008.
*Corresponding author: Tel: 617-278-0146; fax: 617-734-7329.
E-mail address: cpcannon@partners.org (C.P. Cannon).
agement tool. The registry captured data on important char-
acteristics, treatments, and in-hospital outcomes of patients
hospitalized with cardiovascular disease through use of the
web-based case report form. All participating institutions
were required to comply with local regulatory and privacy
guidelines with regard to submitting the GWTG protocol for
review and approval by their institutional review board.
Only sites and variables with a high degree of completeness
were used in this analysis. Any patients transferred in or out
were excluded from length of stay analyses.
The GWTG database measures hospitals’ adherence to
secondary prevention guidelines (pharmacological and life-
style interventions) for coronary artery disease, heart fail-
ure, and stroke. This study included 154,910 patients ad-
mitted to 515 hospitals from January 14, 2000 to April 30,
2007 with ACS. The registry-coordinating center is Out-
come Sciences, Inc. (Cambridge, Massachusetts).
Weekends were defined as the hours between 6 P.M.
Friday and 7 A.M. Monday. Holidays were defined as New
Year’s (December 31 and January 1), Christmas (December
24, December 25), Thanksgiving (last Thursday of Novem-
ber), Labor Day (first Monday of September), Independence
Day (July 4), and Memorial Day (last Monday of May).
Measures assessed in GWTG – Coronary Artery Disease
have been previously described.1 Indicator-specific inclu-
sion and exclusion criteria were applied so that only eligible
patients without contraindications or documented intoler-
ance for a specific indicator remained in the denominators.
Performance measures were evaluated based on time of

0002-9149/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. www.AJConline.org
doi:10.1016/j.amjcard.2008.04.053
 ARTICLE IN PRESS
2 The American Journal of Cardiology (www.AJConline.org)

Table 1
Baseline patient characteristics: weekdays versus weekends/holiday discharges
Measure Weekdays Weekends/Holidays p Value
(n ⫽ 120,081) (n ⫽ 34,829)

Age (yrs) 66.2 ⫾ 14.2 65.2 ⫾ 14.0 ⬍0.0001

Women 36.6% 34.6% ⬍0.0001
Body mass index, kg/m2 28.6 ⫾ 6.5 28.6 ⫾ 6.5 0.09
Race ⬍0.0001
White 70.9% 70.8%
Black or African American 6.7% 6.2%
Hispanic 6.2% 6.7%
Asian or Pacific Islander 3.1% 3.3%
Insurance ⬍0.0001
Medicare 40.3% 38.1%
Medicaid 3.3% 3.1%
Other 48.8% 51.0%
No insurance 7.6% 7.9%
Chronic or recurrent atrial fibrillation* 6.6% 5.8% ⬍0.0001
Chronic obstructive pulmonary disease or asthma* 11.6% 11.3% 0.09
Diabetes mellitus* 27.9% 27.0% 0.0008
Hyperlipidemia* 37.2% 37.0% 0.39
Hypertension* 58.0% 57.0% ⬍0.0001
Peripheral vascular disease* 7.6% 7.4% 0.24
Coronary artery disease* 5.3% 5.4% 0.38
Previous myocardial infarction* 18.3% 17.7% 0.003
Cerebrovascular accident/transient ischemic attack* 6.8% 6.1% ⬍0.0001
Heart failure* 12.2% 11.1% ⬍0.0001
Anemia* 0.7% 0.6% 0.13
Dialysis (chronic) 1.6% 1.7% 0.07
Renal insufficiency – chronic (serum creatinine ⬎2.0) 8.0% 7.3% ⬍0.0001
Smoker* 30.0% 31.6% ⬍0.0001
Diagnosis ⬍0.0001
Unstable angina pectoris 8.3% 6.3%
Confirmed ST-segment elevation myocardial infarction 67.8% 68.7%
Confirmed Non–ST-segment elevation myocardial Infarction 24.0% 25.0%
Hospital characteristics
Bed size 401 ⫾ 241 400 ⫾ 236 0.99
Academic 59.5% 57.7% ⬍0.0001
Percutaneous coronary intervention 90.2% 91.2% ⬍0.0001
Heart transplant 15.6% 16.6% ⬍0.0001
Surgery 82.2% 83.1% 0.009
Region ⬍0.0001
Northeast 18.2% 16.7%
Midwest 23.6% 23.6%
South 27.6% 28.0%
West 29.3% 30.4%

* A medical history as documented in the medical records.

discharge. Performance measure definitions for use of aspi-
rin, ␤ blockers, angiotensin-converting enzyme (ACE) inhib-
itor or angiotensin receptor blocker (ARB) in patients with left
ventricular systolic dysfunction as well as smoking cessation
counseling were based on Joint Commission specifications. An
additional performance measure, ACE inhibitor or ARB use in
all patients with ACS at all levels of left ventricular function,
was evaluated using Joint Commission criteria, except for
ejection fraction percent. New lipid-lowering therapy was de-
fined as the percentage of patients who had a low-density
lipoprotein cholesterol ⬎100 mg/dL without previous treat-
ment discharged on lipid-lowering therapy. Conformity to
blood pressure expectations was assessed as the last recorded
hospital blood pressure ⬍140/90 mm Hg. Referral to cardiac
rehabilitation or instructions regarding formal exercise before
discharge were discussed. Additional measures evaluated
were warfarin on discharge for patients with atrial fibrilla-
tion, diet instructions for patients with acute myocardial
infarction or percutaneous intervention who were dis-
charged on clopidogrel and patients with left ventricular
systolic dysfunction discharged on aldosterone antagonist,
and warfarin education for those discharged on anticoagu-
lation. Performance measures beyond the Joint Commission
expectations were developed by an expert steering commit-
tee. Composite performance measure for 100% compliance
was defined as patients discharged on all of the following:
aspirin, ␤ blockers, ACE inhibitor or ARB in patients with
left ventricular systolic dysfunction, smoking cessation
counseling, and lipid-lowering therapy for patients with
low-density lipoprotein cholesterol ⬎100 mg/dL.
 ARTICLE IN PRESS
Coronary Artery Disease/Weekend Discharge and Guideline Adherence 3

Table 2
Discharge performance measures: weekdays versus weekends/holidays
Measure Weekdays Weekends/Holidays p Value Adjusted Odds Ratio
(95% confidence interval)

Aspirin 95.6% 95.7% 0.52 0.99 (0.94–1.04)

Yes 106,434 30,969
Total eligible 111,310 32,360
␤ blocker 92.9% 93.1% 0.26 0.99 (0.95–1.03)
Yes 101,563 29,627
Total eligible 109,363 31,839
Smoking cessation counseling for current smokers 86.6% 86.2% 0.32 0.98 (0.92–1.03)
Yes 30,935 9,442
Total eligible 35,719 10,949
Patients with low-density lipoprotein ⬎100, receive lipid-lowering drugs 86.2% 86.5% 0.46 0.99 (0.93–1.06)
Yes 26,813 8,255
Total eligible 31,093 9,540
Documented left ventricular systolic 0.95 (0.89–1.01)
Dysfunction on angiotensin-converting enzyme inhibitor or angiotensin 77.5% 77.1% 0.51
Receptor blocker
Yes 17,528 4,742
Total eligible 22,609 6,148
Composite performance measure for 100% compliance 81.5% 81.4% 0.77 0.98 (0.95–1.01)
Yes 96,574 28,018
Total eligible 118,480 34,403
Diet Instruction 15.6% 16.0% 0.16 1.04 (0.98–1.09)
Yes 16,176 5,111
Total eligible 103,535 32,042
Warfarin Education 2.3% 2.1% 0.10 0.97 (0.90–1.04)
Yes 2,354 678
Total eligible 103,535 32,042
Warfarin use for patients with atrial fibrillation 36.4% 36.8% 0.74 1.04 (0.94–1.15)
Yes 2,818 732
Total eligible 7,736 1,988
Clopidogrel for those with acute myocardial infarction or percutaneous 77.4% 77.8% 0.26 1.04 (1.00–1.08)
coronary intervention
Yes 50,573 15,243
Total eligible 65,364 19,603
Lipid-lowering medications or statins 80.3% 80.9% 0.02 1.01 (0.97–1.04)
Yes 94,130 27,537
Total eligible 117,216 34,054
Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker 68.1% 67.3% 0.01 0.95 (0.92, 0.97)
for patients with myocardial infarction
Yes 71,631 21,036
Total eligible 105,122 31,239
Angiotensin-converting enzyme inhibitor for all patients 69.5% 68.3% ⬍0.0001 0.93 (0.91, 0.9)
Yes 72,201 20,767
Total eligible 103,836 30,416
Rehab or physical activity recommendations 72.5% 72.0% 0.04 0.97 (0.94, 0.99)
Yes 86,549 9,704
Total eligible 119,325 34,621
Blood pressure ⬍ 140/90 53.8% 52.5% 0.13 0.97 (0.90, 1.04)
Yes 8,705 2,499
Total eligible 16,188 4,758

All statistical analyses were performed independently at the
Duke Clinical Research Institute (Durham, North Carolina).
Baseline characteristics were compared between patients dis-
charged on weekends/holidays versus weekdays using chi-
square test for categorical variables and Wilcoxon rank-sum
test for continuous variables. The mean value with standard
deviation and percentage is reported to describe the distribution
of continuous and categorical variables, respectively.
Multivariable regression analyses were performed to de-
termine if patient discharge on weekends/holidays indepen-
dently influences adherence to aforementioned GWTG per-
formance measures. Generalized estimating equations
analysis was employed to consider clustering within hospi-
tals.3 Variables in the model include demographic charac-
teristics (age, gender, race, body mass index at admission,
admission systolic blood pressure, and insurance status),
cardiac diagnosis (ST-segment elevation myocardial infarc-
tion, non–ST-segment elevation myocardial infarction or
unstable angina), medical histories (anemia, previous myo-
cardial infarction, stroke, chronic obstructive pulmonary
 ARTICLE IN PRESS
4 The American Journal of Cardiology (www.AJConline.org)

100 96 96 95 95
94 94
92 92

90 87 86 87 87 87 88
85 85
82 82 81 81

80 77 77 78 77

Eligiable Patients Treated (%)

70

60 Teaching Weekday
Teaching Weekend
50
Non Teaching Weekday
40 Non Teaching Weekend

30

20

10

0
ACE/ARB ASA B-Blocker Smoking Lipid 100%
Drug

Figure 1. Discharge performance measures in teaching versus nonteaching hospitals. Teaching hospitals, weekday discharges (purple); teaching hospitals,
weekend discharges (magenta); nonteaching hospitals, weekday discharges (off white); nonteaching hospitals, weekend discharges (light blue).

disease, diabetes, hypertension, hyperlipidemia, heart fail- older (66.2 vs 65.2, p ⬍0.001); more likely women (36.6%
ure, atrial fibrillation, peripheral vascular disease, renal in- vs 34.6%, p ⬍0.0001); had greater history of chronic or
sufficiency, and smoker), and hospital characteristics (hos- recurrent atrial fibrillation (6.6% vs 5.8%, p ⬍0.0001),
pital region, number of beds, academic status, and capacity hypertension (58.0% vs 57.0%, p ⬍0.0001), or diabetes
of performing heart transplant, surgery, and percutaneous (27.9% vs 27.0%, p ⫽ 0.0008); had a history of cerebro-
coronary intervention). Length of stay data were continuous vascular accident/transient ischemic attack (6.8% vs 6.1%,
and skewed to the right and were log transformed with the p ⬍0.0001); had a history of chronic renal insufficiency
ratio of adjusted length of stay reported. Additional analyses (8.0% vs 7.3%, p ⬍0.0001); and presented with more un-
were performed comparing conformity with performance stable angina (8.3% vs 6.3%, p ⬍0.0001). Conversely,
measures among teaching and nonteaching hospitals on smokers (31.6% vs 30.0%, p ⬍0.0001) were more likely to
weekends/holidays compared with weekdays. A teaching present on the weekends/holidays than weekdays.
hospital was defined as a hospital having a residency pro- Patients discharged on weekends/holidays were less
gram approved by the Accreditation Council for Graduate likely to receive ACE inhibitors than those discharged on
Medical Education or belonging to the Council of Teaching the weekdays (68.3% vs 69.5%, p ⬍ 0.0001). There were no
Hospitals. significant differences on discharge aspirin, ␤ blocker,
A value of p ⬍0.05 was considered statistically signifi- smoking cessation counseling for current smokers or pa-
cant for all tests. All analyses were performed using SAS tients receiving lipid-lowering drugs with a low-density
software (version 8.2, SAS Institute, Cary, North Carolina). lipoprotein ⬎100 mg/dL. Composite performance measure
with 100% compliance was similar in both groups (81.5 vs
Results 81.4, p ⫽ 0.77; Table 2). Patients discharged on the week-
In the study period from January 14, 2000 to April 30, 2007, ends/holidays had a shorter length of stay than those who were
there were 282,791 patients enrolled into the GWTG-Cor- discharged on the weekdays (5.11 ⴞ 5.8 days vs 5.71 ⴞ 6.6
onary Artery Disease registry. Of these, 101,206 patients days, adjusted ratio and 95% confidence interval 0.92 to 0.95),
were excluded from analysis because they did not meet p ⬍0.001.
status as having an ACS. Another 26,675 patients were In multivariable analyses that adjusted for patient base-
excluded because of missing information on discharge date/ line and hospital characteristics, there were no significant
time (n ⫽ 11), discharge status expiration (n ⫽ 9,400), differences in conformity with performance measures for
transfer to hospice (n ⫽ 1,516), left against medical advice discharge use of aspirin, ␤ blockers, lipid-lowering therapy,
(n ⫽ 1,148), or transfer to another acute care facility (n ⫽ smoking cessation, and blood pressure control. Further,
14,600). The remaining 154,910 patients were analyzed in there was no significant difference in the composite perfor-
this study. mance measure for weekend/holiday discharges compared
Of the ACS patients meeting study criteria, 120,081 with weekday discharges. In contrast, ACE inhibitors/ARB
(77.5%) were discharged on weekdays and 34,829 (22.5%) were less likely to be given to those eligible that were
were discharged on weekends/holidays. Baseline character- discharged on the weekends/holidays with an odds ratio of
istics for the 2 groups are listed in Table 1. Patients dis- 0.95 (95% confidence interval 0.93 to 0.98). Also, when
charged on weekdays had some modest but statistically restricting this measure to only those who were diagnosed
significant differences in characteristics compared with with acute myocardial infarction, patients were still less
weekend/holiday discharges. Weekday discharges were likely to be discharged with ACE inhibitor/ARB on the
 ARTICLE IN PRESS
Coronary Artery Disease/Weekend Discharge and Guideline Adherence
weekends/holidays (odds ratio 0.95, 95% confidence inter-
val 0.92 to 0.97; Table 2).
We examined also for differences in conformity to per-
formance measures by individual day of discharge (e.g.,
Monday to Friday) and observed no differences among
individual weekdays. Data were further analyzed to deter-
mine if hospital teaching status influenced findings. There
were no statistically significant differences between week-
end/holiday versus weekday discharges for the 5 perfor-
mance measures or composite for complete care among
teaching and nonteaching hospitals (Figure 1).
Discussion
Hospitals in the United States generally provide routine care
services with full medical, nursing, and auxiliary staff on
the weekdays and work with a limited or reduced staff on
weekends/holidays.4 Differences exist also in physician
coverage of patients on weekends/holidays. In patients pre-
senting with myocardial infarction, Kostis et al5 found that
weekend admissions were associated with higher mortality
compared with weekday admissions. This study highlighted
the potential adverse consequences of reduced hospital and
physician staffing on patients being admitted on weekends,
but few data are available to determine the influence of day
of hospital discharge on conformity with guideline recom-
mendations.
We observed that for patients hospitalized with ACS,
adherence to guideline recommendations was generally
similar for patients discharged on a weekend/holiday com-
pared with on a weekday. Thus, among GWTG-Coronary
Artery Disease participating hospitals, hospital staffing and
physician coverage on the day of discharge does not signif-
icantly influence the overall provision of guideline-recom-
mended therapies and patient education as judged by ex-
plicit performance measures.
Some but not all previous studies have suggested differ-
ences in the quality of care such as timeliness of reperfusion
and revascularization as well as outcomes for patients hos-
pitalized with ACS by day of admission, including the
GWTG database.2,6 – 8 In contrast, in this study we did not
detect differences in quality of care as a function of dis-
charge day of the week for the vast majority of measures.
The rate of admission and the severity of ACS may vary
according to the day of the week.9 In our study, however, it
is unlikely that differences in patient characteristics can
explain the similar conformity to quality measures among
patients discharged on the weekend versus weekday be-
cause the lack of difference persisted after adjustment for
multiple patient and hospital characteristics. We are not
aware of previous studies that have assessed the quality of
care by day of discharge. One previous study found that day
of the week had an important influence on duration of stay
for patients hospitalized with acute myocardial infarction.
Discharge occurred most often on a Friday and least often
over the weekend, extending length of stay for patients who
otherwise would have been discharged over the weekend.10
One performance measure had lower adherence on a
weekends/holidays: ACE inhibitors/ARB use irrespective of
left ventricular function. However, ACEI/ARB use among
5
ACS patients with documented left ventricular dysfunction
was not different. The latter group has a Class I recommen-
dation, wheras the use in all ACS patients is a Class IIa
recommendation.11 Thus, it appears that in patients with a
less clear indication, there may be some variability in the
use of this class of agents by discharge timing. Unfortu-
nately, at either time of discharge, despite the strong recom-
mendations, many patients were still being discharged without
ACE inhibitors/ARB, especially on weekends/holidays.
The limitations of this study include the lack of data
regarding use of recommended therapies and clinical out-
comes after hospital discharge. Contraindications and intol-
erance were as documented in the medical record, but a
proportion of patients reported to be eligible for treatment
but not treated may have had contraindications or intoler-
ance that were present but not documented. In addition,
unmeasured confounders may have contributed to the re-
ported differences and lack of differences in conformity
with performance measures between patients discharged on
weekends and those discharged on weekdays/holidays.
There were no direct measures of socioeconomic status in
GWTG. Because of the large number of patients in the
dataset, small differences in patient characteristics (Table 1)
may be of little clinical relevance.
1. Smaha LA. The American Heart Association Get With The Guidelines
program. Am Heart J 2004;148(5 suppl):S46 –S48.
2. LaBresh KA, Fonarow GC, Smith SC, Bonow RO, Smaha LC, Tyler
PA, Hong Y, Albright D, Ellrodt AG. improved treatment of hospi-
talized coronary artery disease patients with the Get With The Guide-
lines Program. Critical Pathways in Cardiology 2007;6:98 –105.
3. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a
generalized estimating equation approach. Biometrics 1988;44:
1049 –1060.
4. Angus DC, Shorr AF, White A, Dremsizov TT, Schmitz RJ, Kelley
MA, on behalf of the Committee on Manpower for Pulmonary and
Critical Care Societies (COMPACCS). Critical care delivery in the
United States: distribution of services and compliance with Leapfrog
recommendations. Crit Care Med 2006;34:1016 –1024.
5. Kostis WJ, Demissie K, Marcella SW, Shao YH, Wilson AC, Moreyra
AE. Myocardial Infarction Data Acquisition System (MIDAS 10)
Study Group. Weekend versus weekday admission and mortality from
myocardial infarction. N Engl J Med 2007;356(11):1099 –1109.
6. Peters RW, Brooks MM, Zoble RG, Liebson PR, Seals AA. Chrono-
biology of acute myocardial infarction: Cardiac Arrhythmia Suppres-
sion Trial (CAST) Experience. Am J Cardiol 1996;78:1198 –1201.
7. Magid DJ, Wang Y, Herrin J, McNamara RL, Bradley EH, Curtis JP,
Pollack CV Jr, French WJ, Blaney ME, Krumholz HM. Relationship
between time of day, day of week, timeliness of reperfusion, and
in-hospital mortality for patients with acute ST-segment elevation
myocardial infarction. JAMA 2005;294:803– 812.
8. Jneid H, Fonarow GC, Cannon CP, Palacios IF, Kilic T, Moukarbel
GV, Maree AO, Labresh KA, Liang L, Newby LK, et al. Impact of
time of presentation on the care and outcomes of acute myocardial
infarction. Circulation 2008;117:2502–2509.
9. van der Palen J, Doggen CJ, Beaglehole R. Variation in the time and
day of onset of myocardial infarction and sudden death. N Z Med J
1995;108:332–334.
10. Varnava AM, Sedgwick JEC, Deaner A, Ranjadayalan K, Timmis,
AD. Restricted weekend service inappropriately delays discharge after
acute myocardial infarction. Heart 2002;87:216 –219.
11. Smith SC Jr., Allen J, Blair SN, Bonow RO, Brass LM, Fonarow GC,
Grundy SM, Hiratzka L, Jones D, Krumholz HM, et al. AHA/ACC
guidelines for secondary prevention for patients with coronary and
other atherosclerotic vascular disease: 2006 update endorsed by the
National Heart, Lung, and Blood Institute. J Am Coll Cardiol 2006;
47:2130 –2139.