Journal of Atherosclerosis and Thrombosis
Online ISSN : 1880-3873
Print ISSN : 1340-3478
ISSN-L : 1340-3478
Original Article
Effect of Hospital Arrival Time on Functional Prognosis of Stroke Patients: Japan Stroke Data Bank Over 20 Years
Tomoya Omae, Michikazu Nakai, Sohei Yoshimura, Kazunori Toyoda, Toshiharu Yanagisawa, Shotai Kobayashi, Masatoshi Koga, Japan Stroke Data Bank investigators
Author information
Keywords: Off-hour effect, Weekend effect, Ischemic stroke, Intracerebral hemorrhage, Subarachnoid hemorrhage
JOURNAL OPEN ACCESS FULL-TEXT HTML

2025 Volume 32 Issue 1 Pages 70-87

Browse “Advance online publication” version
Details
Download PDF (1266K)
Download citation RIS

(compatible with EndNote, Reference Manager, ProCite, RefWorks)

BIB TEX

(compatible with BibDesk, LaTeX)

Text
How to download citation
Contact us
Abstract

Aims: The impact of weekend/holiday and nighttime hospitalization on functional outcomes and long-term trends in stroke patients is unclear. We examined functional and life outcomes and changes over time.

Methods: We analyzed the clinical data of 203,176patients for hospital arrival day of week and 76,442patients for arrival times using Japan Stroke Data Bank. The endpoints were favorable outcome (Modified Rankin Scale[mRS]0-2), unfavorable outcome(mRS 5-6), and in-hospital mortality. We calculated odds ratios(OR) and 95% confidence interval(CI) of weekends/holidays and off-hours versus weekdays and on-hours for 2000-2009 and 2010-2020 using a mixed-effect multivariate model adjusted for confounding factors and evaluated interactions. Thereafter, we performed to check for year trends.

Results: All endpoints were worse in weekend/holiday admissions for all stroke and in off-hours hospitalization for total stroke(TS), ischemic stroke(IS), and intracerebral hemorrhage(ICH). The adjusted ORs for favorable outcomes of weekend/holiday admissions were TS, 0.90(0.87-0.93); IS, 0.89(0.86-0.93); ICH, 0.91(0.84-0.98) and unfavorable outcome TS, 1.04(1.002-1.08) IS, 1.06(1.01-1.11). Off-hour hospitalization had adjusted ORs for favorable outcome(TS, 0.86 [95% CI: 0.82-0.91]; IS, 0.90 [0.84-0.95]; ICH, 0.85 [0.75-0.96]), unfavorable outcome(TS, 1.14 [1.07-1.22]; IS, 1.13 [1.04-1.23]; ICH, 1.15 [1.01-1.31]), and mortality (TS, 1.15 [1.05-1.26]; IS, 1.17 [1.04-1.32]). For IS, the incidence of unfavorable outcomes during off-hours was significantly lower in 2010-2020 than in 2000-2009; after adjusting for reperfusion therapy, it was no longer significant.

Conclusion: Stroke patients admitted on weekends/holidays and off-hours had worse functional and life outcomes. Functional outcomes for off-hour admission for IS improved at 10-year intervals, possibly due to improvements in stroke care systems.

1.Introduction

Stroke is the second leading cause of death worldwide and the third leading cause of disability-adjusted life years1-3), thus having a significant societal impact. To overcome these problems, various epidemiological studies have been conducted to establish effective stroke care systems4, 5). Since the 2001 report on the effect of hospital arrival time on prognosis of hemorrhagic stroke6), a higher mortality of stroke is associated with admission during weekends/holidays and off-hours, as reported by studies in 2009 7, 8). Subsequently, many investigations have been conducted, mainly in western countries, to identify causes of this phenomenon. The severity of illness, low utilization of the emergency system at night, and delays in the initiation of ward nursing and physical assessment have been identified as potential causes and are considered important issues in the establishment of emergency care and ward response systems9-11). Therefore, a previous study conducted a detailed examination of daily and weekly variations in medical nursing care according to the time of admission9), and the outcomes were evaluated in terms of inpatient deaths. However, to the best of our knowledge, detailed assessments of favorable functional outcomes have not yet been reported.

Furthermore, despite ongoing discussions, there have been no reports regarding specific changes, if any, in weekend and off-hour effects resulting from efforts made to improve stroke care systems for early treatment, particularly for ischemic stroke (IS); these efforts have involved the approval and expansion of indications for intravenous thrombolysis and mechanical thrombectomy over the past two decades. However, there have been reports on this subject within individual treatment groups12, 13).

2.Aim

This study aimed to conduct a detailed examination of functional and life outcomes and the changes of these effects over time using a large, uniformly structured nationwide registry dataset covering 21 years.

3.Methods

3.1 Study Design

Data were collected from the Japan Stroke Data Bank (JSDB), a hospital-based multicenter stroke registry, from January 2000 to December 2020. The data included standardized information on hospitalized patients, detailed diagnoses, and acute management by stroke specialists for patients with acute stroke or transient ischemic attacks. The study protocol was approved by the Institutional Ethical Board of the National Cerebral and Cardiovascular Center (approval number: M27-090-14), as well as by the ethical boards of the collaborative research centers that collected the data. Patient consent for inclusion in the registry was obtained using an opt-out consent method. The data of stroke patients were prospectively collected by stroke physicians or clinical coordinators using an online standardized database form. This registry complied with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

3.2 Participants

This study included stroke patients aged ≥ 18 years who were hospitalized within 7 days of stroke onset. The stroke types, including IS subtypes according to the criteria of the Trial of Org 10172 in Acute Stroke Treatment, and history of any stroke were obtained from the database14). We extracted data on time of hospital arrival for the following groups of stroke patients: IS composed of small vessel occlusion, large artery atherosclerosis, cardioembolism, and other causes, including arterial dissection, moyamoya disease, aortic origin, venous thrombosis, branch atheromatous disease, and unknown origin; intracerebral hemorrhage (ICH) composed of hypertensive hemorrhage, amyloid angiopathy, moyamoya disease, cerebral arteriovenous malformation, cerebral venous thrombosis, dural arteriovenous fistula, cerebral arterial dissection and unknown origin; and subarachnoid hemorrhage (SAH) composed of cerebral aneurysm, moyamoya disease, cerebral arteriovenous malformation, dural arteriovenous fistula, infectious aneurysm, and unknown origin. Total stroke (TS) referred to IS, ICH, SAH. This database includes both first and recurrent stroke cases.

3.3 Patient Characteristics

Patient characteristics, including age, sex, National Institutes of Health Stroke Scale (NIHSS) score15) on admission, World Federation of Neurosurgical Societies (WFNS) grading for SAH16), pre-stroke modified Rankin Scale (mRS) score, hypertension, diabetes, dyslipidemia, smoking history, atrial fibrillation, admission on holidays and national holidays, length of hospital stay, presence of reperfusion therapy comprising intravenous thrombolysis and mechanical thrombectomy, and arrival at the hospital within 4.5 hours of stroke onset, were recorded. Hypertension was defined as present or absent based on a history-related interview at admission, medications administered at admission, and medications administered at discharge. Diabetes mellitus and dyslipidemia were determined to be present or absent based on a history-related interview at admission, pre-stroke medications, medications administered at discharge, and laboratory data at admission. Medical histories were determined based on medications taken prior to admission, history of medical visits, and laboratory data. Current and previous smokers were defined as having a history of smoking. The presence or absence of atrial fibrillation was determined based on a history-related interview at admission and electrocardiograms both at admission and during hospitalization. The NIHSS score and WFNS grading for SAH on admission was extracted as an indicator of neurological severity at admission.

3.4 Outcomes

In this study, outcomes were assessed using mRS at hospital discharge, with scores of: 0–2, indicating favorable outcomes; 5–6, indicating unfavorable outcomes, which have been shown to decrease over the past two decades4); and 6, indicating in-hospital mortality.

3.5 Statistical Analysis

Data are expressed as mean (standard deviation) or as median (interquartile range, IQR) for continuous variables and analyzed using the Mann–Whitney U test or Wilcoxon rank-sum test, as appropriate. Categorical data are expressed as number (%) and analyzed using χ2 tests. Univariate and multivariate multilevel mixed-effects logistic models were constructed, with institution included as a random variable. The data from this registry varied by facility in terms of number of cases, outcomes, and mortality rates; to account for these differences, we employed a mixed-effects logistic regression analysis, using facility factors as intercepts and to explore the hypothesized causal relationships between outcomes and these factors using odds ratio. The models were adjusted for age, sex, pre-stroke mRS score, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission (WFNS grading for SAH), hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking history to create Model 1. For IS patients only, reperfusion therapy was additionally adjusted for in Model 2. The interaction P-value was calculated between year classes. Admission days and hours were categorized as follows: weekdays (Monday through Friday) and weekends (Saturday and Sunday)/national holidays and on-hours (7 am–7 pm) and off-hours (7 pm–7 am). Missing values were handled using the complete case method. The year was considered as an independent variable to perform for year trends. All statistical analyses were performed using Stata version 16.1 (College Station, TX, United States of America).

4.Results

4.1 Patient Characteristics

The baseline characteristics of the study participants and the number of valid data are presented in Tables 1 and 2. Of the 218,858 patients registered in the JSDB from January 2000 to December 2020, valid data for weekdays and weekends/holidays were extracted from 203,176 patients, of whom 85,613 were women (42.1%) with a mean age of 71.8 years. Of these, 150,089 were IS patients, of whom 59,829 were women (39.8%) with a mean age of 73.3 years. The study also included 40,175 ICH patients, of whom 17,108 were women (42.5%) with a mean age of 68.7 years, and 12,912 SAH patients, of whom 8,676 were women (67.1%) with a mean age of 63.9 years. Among the TS patients, 57,970 were hospitalized on weekends and national holidays, including 42,049 IS patients, 12,015 ICH patients, and 3,906 SAH patients. Data on arrival time were available for 76,442 patients (63,691 who arrived during off-hours and 21,604 during on-hours). Of these, 56,769, 14,718, and 4,955 patients had IS, ICH, and SAH, respectively. The study included the records of 82,743 patients enrolled from 2000 to 2009, including 61,113 IS patients, 15,662 ICH patients, and 5,968 SAH patients. From 2010 to 2020, the study included a total of 120,433 stroke patients, including 88,976 IS patients, 24,513 ICH patients, and 6,944 SAH patients (Tables 1, 2 and Supplemental Fig.1).

Table 1.Patient characteristics by hospital admission day or arrival time (total stroke and ischemic stroke)

variable TS IS
weekday weekend/ holiday on-hour off-hour weekday weekend/ holiday on-hour off-hour
No. of patients 145206 57970 57197 19245 108040 42049 43975 12794
Age, mean (SD) 71.7 (13.2) 72.0 (13.3) 72.1 (12.9) 70.40 (13.6) 73.1 (12.4) 73.6 (12.4) 73.3 (12.2) 72.3 (12.5)
Sex (female) No. (%) 60640 (41.7) 24973 (43.0) 23998 (41.9) 8108 (42.1) 42658 (39.4) 17171 (40.8) 17638 (40.1) 4877 (38.1)
Hypertension No. (%) 98025 (67.5) 39055 (67.3) 37601 (65.7) 12452 (64.7) 73798 (68.3) 28570 (67.9) 29164 (66.3) 8360 (65.3)
Diabetes Mellitus No. (%) 34394 (23.6) 12991 (22.4) 13578 (23.7) 4149 (21.5) 29077 (26.9) 10768 (25.6) 11688 (26.5) 3245 (25.3)
Dyslipidemia No. (%) 41382 (28.5) 15564 (26.8) 15697 (27.4) 4613 (23.9) 34703 (32.1) 12809 (30.4) 13382 (30.4) 3598 (28.1)
Atrial fibrillation No. (%) 25193 (17.3) 10778 (18.5) 9845 (17.2) 3696 (19.2) 22663 (20.9) 9717 (23.1) 8978 (20.4) 3314 (25.9)
Smoking No. (%) 37153 (25.5) 14333 (24.7) 17396 (30.4) 5730 (29.7) 28715 (26.5) 10753 (25.5) 13716 (31.1) 4033 (31.5)
Reperfusion therapy No. (%) N.A N.A N.A N.A 10089 (9.3) 4483 (10.6) 3824 (8.7) 1618 (12.6)
Prestroke mRS,
median (IQR) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1)
No 132796 53239 49342 16882 98659 38497 37427 11058
Prestroke mRS0-2 No. (%)

113369

(85.3)

45489

(85.4)

41832

(84.7)

 14858 (88.0) 83478 (77.2) 32535 (77.3) 31512 (71.6) 9669 (75.5)
NIHSS score
median (IQR) 4 (2-12) 5 (2-15) 4 (2-12) 6 (2-17) 4 (2-9) 4 (2-11) 4 (2-9) 5 (2-12)
No 120963 48039 47133 15528 92797 36005 37498 10914
Days of hospitalization, median (IQR) 18 (11-33) 19 (11-34) 17 (11-31) 19 (11-35) 17 (10-29) 18 (11-31) 16 (10-28) 17 (11-30)
Discharge mRS, median (IQR) 2 (1-4) 3 (1-4) 2 (1-4) 3 (1-4) 2 (1-4) 2 (1-4) 2 (1-4) 2 (1-4)
History of stroke No. (%) 35884 (24.7) 13802 (23.8) 14535 (25.4) 4371 (22.7) 28573 (26.4) 10791 (25.6) 11951 (27.1) 3246 (25.3)

mRS: modified Rankin Scale; IQR: interquartile range; SD: standard deviation; NIHSS: National Institutes of Health Stroke Scale

Table 2.Patient characteristics by hospital admission day or arrival time (intracerebral hemorrhage and subarachnoid hemorrhage)

variable ICH SAH
weekday weekend/ holiday on-hour off-hour weekday weekend/ holiday on-hour off-hour
No. of patients 28160 12015 9943 4775 9006 3906 3279 1676
Age, mean (SD) 68.6 (14.3) 68.8 (14.2) 69.3 (14.0) 67.8 (14.3) 63.8 (15.0) 64.2 (14.8) 64.5 (14.5) 62.4 (15.3)
Sex (female) No. (%) 11993 (42.5) 5115 (42.5) 4186 (42.1) 2061 (43.1) 5989 (66.5) 2687 (68.7) 2174 (66.3) 1170 (69.8)
Hypertension No. (%) 20006 (71.0) 8576 (71.3) 6939 (69.7) 3323 (69.5) 4221 (46.8) 1909 (48.8) 1498 (45.6) 769 (45.8)
Diabetes Mellitus No. (%) 4652 (16.5) 1922 (16.0) 1650 (16.5) 792 (16.5) 665 (7.3) 301 (7.7) 240 (7.3) 112 (6.6)
Dyslipidemia No. (%) 5376 (19.0) 2226 (18.5) 1892 (19.0) 831(17.4) 1303 (14.4) 529 (13.5) 423 (12.9) 184 (10.9)
Atrial fibrillation No. (%) 2296 (8.1) 957 (7.9) 808 (8.1) 348 (7.2) 234 (2.6) 104 (2.6) 59 (1.8) 34 (2.0)
Smoking No. (%) 6359 (22.5) 2699 (22.4) 2812 (28.2) 1255 (26.2) 2079 (23.0) 881 (22.5) 868 (26.4) 442 (26.3)
Prestroke mRS, median (IQR) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-0) 0 (0-0) 0 (0-0) 0 (0-0)
No 26042 11164 9080 4346 8095 3578 2835 1478
Prestroke mRS0-2 No. (%) 22236 (78.9) 9560 (79.5) 7651 (76.9) 3793 (79.4) 7655 (85.0) 3394 (86.8) 2669 (81.4) 1396 (83.2)
NIHSS score or WFNS grading
median (IQR) 11 (4-23) 12 (5-24) 11 (4-22) 13 (6-25) 2 (1-4) 2 (1-5) 2 (1-5) 2 (1-5)
No 23479 10029 8235 3933 8506 3732 3137 1594
Days of hospitalization,
median (IQR) 23 (12-41) 24 (12-42) 22 (12-40) 23 (12-41) 30 (18-52) 31 (18-53) 30 (17-54) 30 (17-54)
Discharge mRS, median (IQR) 4 (2-5) 4 (2-5) 4 (2-5) 4 (2-5) 2 (0-5) 3 (1-5) 2 (0-5) 3 (0-5)
History of stroke No. (%) 6486 (23.0) 2658 (22.1) 2279 (22.9) 975 (20.4) 825 (9.1) 353 (9.0) 305 (9.3) 150 (8.9)

mRS: modified Rankin Scale; IQR: interquartile range; SD: standard deviation; NIHSS: National Institutes of Health Stroke Scale; WFNS: World Federation of Neurosurgical Societies

Supplemental Fig. 1. Study flow-chart

JSDB: Japan Stroke Data Bank; TIA: transient ischemic attack IS: ischemic stroke ICH: intracerebral hemorrhage SAH: subarachnoid hemorrhage; mRS: modified Rankin Scale;

4.2 Proportions of Functional Outcomes and Mortality at Discharge

The ratios of favorable outcomes during weekends/holidays (vs. weekdays) and off-hours (vs. on-hours) were lower, whereas the ratios of unfavorable outcome and mortality were higher across all types of strokes. The ratios of favorable outcomes were lower and those of unfavorable outcomes were higher for off-hour (vs. on-hours) for IS in 2000–2009, but the differences were not significant in 2010–2020 (Fig.1 and Supplemental Tables 1, 2, 3).

Fig.1. The proportions of functional outcomes and mortality at discharge between weekdays and weekends/holidays and on-hours and off-hours in 2000–2020, 2000–2009, and 2010–2020

The outcomes at discharge (modified Rankin Scale [mRS] score of 0–2: favorable outcome; score of 5–6: unfavorable outcome; score of 6: mortality) Categorical data are expressed as percentage analyzed using χ2 tests. ＊＊: ＜0.05； ＊: ＜0.001

TS: total stroke; IS: ischemic stroke; ICH: intracerebral hemorrhage; SAH: subarachnoid hemorrhage; mRS: modified Rankin Scale

Supplemental Table 1.Study endpoints by hospital admission day or arrival time in 2000-2020

variable weekday weekend/ holiday P value on-hour off-hour P value weekday weekend/ holiday P value on-hour off-hour P value
TS IS
Favorable outcome (mRS 0-2) 73870 26912 ＜0.001 28676 8488 ＜0.001 60612 21775 ＜0.001 23979 6473 ＜0.001
No. (%) (51.7) (47.1) (50.8) (44.8) (56.9) (52.6) (55.1) (51.3)
Unfavorable outcome (mRS 5-6) 26993 12296 ＜0.001 10754 4487 ＜0.001 15576 6988 ＜0.001 6647 2273 ＜0.001
No. (%) (18.8) (21.5) (19.0) (23.7) (14.6) (16.9) (15.3) (18.0)
Mortality 10238 4802 ＜0.001 3969 1931 ＜0.001 4580 2100 ＜0.001 1908 727 ＜0.001
No. (%) (7.1) (8.4) (7.0) (10.2) (4.3) (5.1) (4.4) (5.8)
ICH SAH
Favorable outcome (mRS 0-2) 8764 3309 ＜0.001 3103 1234 ＜0.001 4494 1828 ＜0.001 1594 781 0.37
No. (%) (31.7) (28.0) (31.8) (26.3) (51.4) (47.9) (50.2) (48.8)
Unfavorable outcome (mRS 5-6) 8708 3982 ＜0.001 3040 1661 ＜0.001 2709 1326 ＜0.001 1067 553 0.51
No. (%) (31.5) (33.7) (31.2) (35.5) (31.0) (34.7) (33.6) (34.5)
Mortality 3835 1833 ＜0.001 1354 814 ＜0.001 1823 869 0.016 707 390 0.1
No. (%) (13.9) (15.5) (13.9) (17.4) (20.8) (22.8) (22.2) (24.4)

Categorical data are expressed as number (%) with χ2 tests. mRS: modified Rankin Scale; NIHSS: National Institutes of Health Stroke Scale; WFNS: World Federation of Neurosurgical Societies;

Supplemental Table 2.Study endpoints by hospital admission day or arrival time in 2000-2009

weekday weekend/ holiday p-value on-hour off-hour p-value weekday weekend/ holiday p-value on-hour off-hour p-value
TS IS
Favorable outcome (mRS 0-2) 30632 10983 ＜0.001 16467 4715 ＜0.001 24848 8786 ＜0.001 13723 3555 ＜0.001
No. (%) (53.0) (48.1) (53.8) (46.3) (56.3) (51.6) (55.9) (49.8)
Unfavorable outcome (mRS 5-6) 11317 5109 ＜0.001 5633 2457 ＜0.001 6637 2958 ＜0.001 3643 1352 ＜0.001
No. (%) (19.5) (22.4) (18.4) (24.1) (15.0) (17.3) (14.8) (18.9)
Mortality 4712 2155 ＜0.001 2285 1104 ＜0.001 2176 976 ＜0.001 1178 457 ＜0.001
No. (%) (8.1) (9.4) (7.4) (10.8) (4.9) (5.7) (4.8) (6.4)
ICH SAH
Favorable outcome (mRS 0-2) 3677 1388 ＜0.001 1702 678 ＜0.001 2107 809 0.011 1042 482 0.15
No. (%) (33.3) (29.8) (35.3) (28.7) (49.8) (46.5) (51.4) (47.5)
Unfavorable outcome (mRS 5-6) 3375 1545 0.002 1374 786 ＜0.001 1305 606 0.004 616 319 0.29
No. (%) (30.6) (33.2) (28.5) (33.3) (30.8) (34.8) (30.4) (31.4)
Mortality 1593 738 0.03 664 407 ＜0.001 943 441 0.014 443 240 0.14
No. (%) (14.4) (15.8) (13.8) (17.2) (22.2) (25.3) (21.8) (23.6)

Categorical data are expressed as number (%) with χ2 tests. mRS: modified Rankin Scale; NIHSS: National Institutes of Health Stroke Scale; WFNS: World Federation of Neurosurgical Societies;

Supplemental Table 3.Study endpoints by hospital admission day or arrival time in 2010-2020

weekday weekend/ holiday p-value on-hour off-hour p-value weekday weekend/ holiday p-value on-hour off-hour p-value
TS IS
Favorable outcome (mRS 0-2) 43238 15929 ＜0.001 12209 3773 ＜0.001 35764 12989 ＜0.001 10256 2918 0.14
No. (%) (50.8) (46.4) (47.2) (43.2) (55.9) (51.8) (52.6) (51.5)
Unfavorable outcome (mRS 5-6) 15676 7187 ＜0.001 5121 2030 ＜0.001 8939 4030 ＜0.001 3004 921 0.12
No. (%) (18.4) (20.9) (19.8) (23.2) (13.9) (16.1) (15.4) (16.2)
Mortality 5526 2647 ＜0.001 1684 827 ＜0.001 2404 1124 ＜0.001 730 270 ＜0.001
No. (%) (6.5) (7.7) (6.5) (9.4) (3.7) (4.4) (3.7) (4.7)
ICH SAH
Favorable outcome (mRS 0-2) 5087 1921 ＜0.001 1401 556 ＜0.001 2387 1019 0.012 552 299 0.63
No. (%) (29.6) (26.0) (27.2) (23.0) (49.9) (46.9) (43.9) (45.1)
Unfavorable outcome (mRS 5-6) 5333 2437 0.002 1666 875 0.001 1404 720 0.002 451 234 0.79
No. (%) (31.1) (33.0) (32.4) (36.2) (29.4) (33.2) (35.9) (35.3)
Mortality 2242 1095 ＜0.001 690 407 ＜0.001 880 428 0.23 264 150 0.42
No. (%) (13.0) (14.8) (13.4) (16.8) (18.4) (19.7) (21.0) (22.6)

Categorical data are expressed as number (%) with χ2 tests. mRS: modified Rankin Scale; NIHSS: National Institutes of Health Stroke Scale; WFNS: World Federation of Neurosurgical Societies;

4.3 Functional and Life Outcomes during Weekends/Holidays vs. Weekdays

For TS patients, the adjusted odds ratio (OR) on weekends/holidays vs. weekdays for favorable outcomes was 0.90 (95% confidence interval [CI]: 0.87–0.93), and the adjusted OR for unfavorable outcomes was 1.04 (1.002–1.08) (Model 1). For IS, the adjusted OR for favorable outcomes was 0.89 (0.86–0.93), and the adjusted OR for unfavorable outcomes was 1.06 (1.01–1.11) (Model 2). For ICH patients, the OR for favorable outcomes was 0.91 (0.84–0.98) (Model 1). For SAH, no significant difference was seen in Model 1 (Fig.2a and Supplemental Table 4).

Fig.2. Functional and life outcomes during weekends/holidays vs. weekdays and during off-hours vs. on-hours

The odds ratios represent the proportion of outcomes at discharge (modified Rankin Scale [mRS] score of 0–2: favorable outcome; score of 5–6: unfavorable outcome; score of 6: mortality) between (a) weekend/holiday arrival patients and weekday arrival patients and (b) between off-hour and on-hour arrival patients. The following univariate and multivariate multilevel mixed-effects logistic models were constructed: Model 1, adjusted for age, sex, mRS score on admission, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission (WFNS grading for SAH), hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking; Model 2, with reperfusion therapy additionally adjusted for in Model 1 only for ischemic stroke.

Supplemental Table 4.Functional and life outcomes of weekends/holidays vs. weekdays

Odds ratio (95% confidence interval)
crude Model1 Model2
favorable outcome (0-2)
TS 0.84 (0.83, 0.86) 0.90 (0.87, 0.93)
IS 0.85 (0.83, 0.87) 0.89 (0.86, 0.93) 0.89 (0.86, 0.93)
ICH 0.85 (0.81, 0.89) 0.91 (0.84, 0.98)
SAH 0.88 (0.81, 0.95) 0.99 (0.88, 1.12)
unfavorable outcome (5-6)
TS 1.15 (1.13, 1.18) 1.04 (1.002, 1.08)
IS 1.17 (1.13, 1.20) 1.06 (1.01, 1.11) 1.06 (1.01, 1.11)
ICH 1.09 (1.04, 1.14) 1.00 (0.92, 1.08)
SAH 1.17 (1.08, 1.27) 1.10 (0.97, 1.25)
mortality
TS 1.16 (1.12, 1.20) 1.02 (0.96, 1.07)
IS 1.17 (1.11, 1.23) 1.00 (0.93, 1.08) 1.00 (0.93, 1.08)
ICH 1.12 (1.06, 1.20) 1.06 (0.96, 1.17)
SAH 1.11 (1.01, 1.22) 0.95 (0.83, 1.09)

The odds ratios represent the proportion of outcomes at discharge (modified Rankin Scale [mRS] score of 0-2: favorable outcome, score of 5-6: unfavorable outcome, score of 6: mortality) between weekend/holiday arrival patients and weekday arrival patients. The following univariate and multivariate multilevel mixed-effects logistic models were constructed: Model 1: adjustment was performed for age, sex, mRS score on admission, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission, hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking. Model 2: reperfusion therapy was additionally adjusted for in Model 1 only for IS.

4.4 Functional and Life Outcomes during Off-Hours vs. On-Hours

For TS patients, the adjusted OR for favorable outcomes during off-hours vs. on-hours was 0.86 (95% CI: 0.82–0.91), the adjusted OR for unfavorable outcomes was 1.14 (1.07–1.22), and the adjusted OR for mortality was 1.15 (1.05–1.26) in Model 1. For IS, the adjusted OR for favorable outcomes was 0.90 (0.84–0.95), the adjusted OR for unfavorable outcomes was 1.13 (1.04–1.23), and the adjusted OR for mortality was 1.17 (1.04–1.32) in Model 1. In Model 2, the adjusted OR for favorable outcomes was 0.89 (0.84–0.95), the adjusted OR for unfavorable outcomes was 1.13 (1.04–1.23), and the adjusted OR for mortality was 1.17 (1.04–1.23). For ICH, the adjusted OR for favorable outcomes was 0.85 (0.75–0.96), and the adjusted OR for unfavorable outcomes was 1.15 (1.01–1.31) in Model 1. For SAH patients, no significant difference was seen in Model 1 (Fig.2b and Supplemental Table 5).

Supplemental Table 5.Functional and life outcomes of off-hour vs. on-hour

Odds ratio (95% confidence interval)
crude Model1 Model2
favorable outcome (0-2)
TS 0.81 (0.78, 0.84) 0.86 (0.82, 0.91)
IS 0.88 (0.84, 0.91) 0.90 (0.84, 0.95) 0.89 (0.84, 0.95)
ICH 0.78 (0.72, 0.84) 0.85 (0.75, 0.96)
SAH 0.96 (0.85, 1.09) 0.86 (0.71, 1.05)
unfavorable outcome (5-6)
TS 1.27 (1.22, 1.32) 1.14 (1.07, 1.22)
IS 1.18 (1.12, 1.25) 1.13 (1.04, 1.23) 1.13 (1.04, 1.23)
ICH 1.20 (1.12, 1.30) 1.15 (1.01, 1.31)
SAH 1.02 (0.90, 1.16) 1.06 (0.87, 1.30)
mortality
TS 1.43 (1.35, 1.52) 1.15 (1.05, 1.26)
IS 1.31 (1.19, 1.43) 1.17 (1.04, 1.32) 1.17 (1.04, 1.32)
ICH 1.29 (1.17, 1.42) 1.14 (0.97, 1.33)
SAH 1.11 (0.96, 1.28) 1.11 (0.90, 1.36)

The odds ratios represent the proportion of outcomes at discharge (modified Rankin Scale [mRS] score of 0-2: favorable outcome, score of 5-6: unfavorable outcome, score of 6: mortality) between off-hour and on-hour arrival patients. The following univariate and multivariate multilevel mixed-effects logistic models were constructed: Model 1: adjustment was performed for age, sex, mRS score on admission, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission, hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking. Model 2: reperfusion therapy was additionally adjusted for in Model 1 only for IS.

4.5 Functional and life outcomes during Weekends/Holidays vs. Weekdays by Decade

No significant interaction was observed between the effect of weekends/holidays vs. weekdays and the periods of 2000–2009 vs. 2010–2020 for any of the endpoint measures (Fig.3 and Supplemental Table 6).

Fig.3. Functional and life outcomes during weekends/holidays vs. weekdays by decade

The odds ratios represent the proportion of outcomes at discharge between weekend/holiday and weekday arrival patients for the years 2000–2009 and 2010–2020. The multivariate analysis of Model 1.2 is the same as that described in Figure 2. Interaction between years is represented, with p＜0.05 indicating a significant difference.

Supplemental Table 6.Functional and life outcomes of weekends/holidays vs. weekdays by decade

crude interaction Model1 interaction Model2 interaction
favorable outcome (0-2)
TS 2000-2009 0.84 (0.81, 0.86) 0.473 0.89 (0.84, 0.94) 0.789
2010-2020 0.85 (0.83, 0.87) 0.90 (0.86, 0.93)
IS 2000-2009 0.84 (0.81, 0.87) 0.325 0.89 (0.83, 0.94) 0.623 0.89 (0.83, 0.94) 0.703
2010-2020 0.86 (0.83, 0.88) 0.89 (0.85, 0.93) 0.89 (0.85, 0.94)
ICH 2000-2009 0.87 (0.80, 0.93) 0.547 0.94 (0.83, 1.08) 0.5
2010-2020 0.84 (0.79, 0.89) 0.89 (0.80, 0.98)
SAH 2000-2009 0.87 (0.78, 0.98) 0.823 0.98 (0.80, 1.21) 0.714
2010-2020 0.89 (0.80, 0.98) 0.99 (0.85, 1.16)
unfavorable outcome (5-6)
TS 2000-2009 1.15 (1.11, 1.20) 0.934 1.02 (0.95, 1.09) 0.576
2010-2020 1.16 (1.12, 1.20) 1.06 (1.01, 1.11)
IS 2000-2009 1.16 (1.11, 1.22) 0.933 1.02 (0.94, 1.10) 0.549 1.01 (0.94, 1.10) 0.489
2010-2020 1.17 (1.13, 1.22) 1.08 (1.02, 1.15) 1.08 (1.02, 1.15)
ICH 2000-2009 1.08 (1.00, 1.17) 0.99 1.01 (0.87, 1.16) 0.981
2010-2020 1.09 (1.03, 1.16) 1.00 (0.91, 1.11)
SAH 2000-2009 1.18 (1.04, 1.33) 0.971 1.07 (0.86, 1.32) 0.873
2010-2020 1.17 (1.05, 1.31) 1.14 (0.97, 1.34)
mortality
TS 2000-2009 1.14 (1.08, 1.20) 0.416 0.95 (0.87, 1.04) 0.062
2010-2020 1.18 (1.12, 1.24) 1.06 (0.99, 1.14)
IS 2000-2009 1.15 (1.06, 1.24) 0.547 0.94 (0.84, 1.06) 0.203 0.94 (0.84, 1.06) 0.201
2010-2020 1.19 (1.11, 1.28) 1.05 (0.96, 1.15) 1.05 (0.96, 1.15)
ICH 2000-2009 1.09 (0.9, 1.20) 0.481 0.95 (0.80, 1.13) 0.08
2010-2020 1.15 (1.07, 1.25) 1.13 (1.00, 1.28)
SAH 2000-2009 1.17 (1.02, 1.33) 0.347 1.06 (0.85, 1.33) 0.709
2010-2020 1.07 (0.94, 1.22) 0.90 (0.75, 1.07)

The odds ratios (95% confidence interval) represent the proportion of outcomes at discharge (modified Rankin Scale [mRS] score of 0-2: favorable outcome, score of 5-6: unfavorable outcome, score of 6: mortality) between weekend/holiday arrival patients and weekday arrival patients. The following univariate and multivariate multilevel mixed-effects logistic models were constructed: Model 1: adjustment was performed for age, sex, mRS score on admission, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission, hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking. Model 2: reperfusion therapy was additionally adjusted for in Model 1 only for IS.

Interaction between years was represented, with p＜0.05 indicating a significant difference.

4.6 Functional and Life Outcomes during Off-Hours vs. On-Hours by Decade and Years

For total stroke, ICH, and SAH patients, no significant interaction was found for favorable outcomes, unfavorable outcomes, and mortality rates during off-hours (vs. on-hours) between the periods of 2000–2009 and 2010–2020 in Model 1. For IS, the incidence of unfavorable outcomes during off-hours was significantly lower in 2010–2020 than in 2000–2009 in Model 1 (p for interaction =0.034). However, this interaction was no longer statistically significant after additionally adjusting for reperfusion therapy (p for interaction =0.06). No significant interaction was seen for favorable outcomes and mortality rates during off-hours between the two decades in IS patients (Fig.4 and Supplemental Table 7). Also, the year-to-year trend analysis showed that, in crude, there were significant differences in TS and IS for favorable outcomes, unfavorable outcomes, and mortality, whereas there was no significant difference in SAH for favorable and ICH and SAH for unfavorable outcomes, without dissociation from the interaction. The adjusted OR for unfavorable outcomes was 0.967 (0.959–0.975) in Model 1 for IS (Table 3).

Fig.4. Functional and life outcomes during off-hours vs. on-hours by decade

The odds ratios represent the proportion of outcomes at discharge between off-hour and on-hour arrival patients for 2000–2009 and 2010–2020. The multivariate analysis of Model 1.2 is the same as that described in Figure 2. Interaction between years is represented, with p＜0.05 indicating a significant difference.

Supplemental Table 7.Functional and life outcomes of off-hour vs. on-hour by decade

crude interaction Model1 interaction Model2 interaction
favorable outcome(0-2)
TS 2000-2009 0.76 (0.72, 0.79) ＜0.001 0.86 (0.79, 0.93) 0.585
2010-2020 0.87 (0.83, 0.91) 0.87 (0.81, 0.93)
IS 2000-2009 0.81 (0.76, 0.85) ＜0.001 0.90 (0.81, 0.99) 0.286 0.90 (0.81, 0.99) 0.502
2010-2020 0.97 (0.91, 1.03) 0.90 (0.83, 0.98) 0.90 (0.82, 0.98)
ICH 2000-2009 0.74 (0.67, 0.83) 0.333 0.83 (0.69, 1.00) 0.611
2010-2020 0.81 (0.72, 0.91) 0.86 (0.74, 1.02)
SAH 2000-2009 0.90 (0.77, 1.06) 0.231 0.90 (0.67, 1.20) 0.415
2010-2020 1.05 (0.87, 1.28) 0.87 (0.67, 1.12)
unfavorable outcome(5-6)
TS 2000-2009 1.34 (1.27, 1.42) 0.002 1.16 (1.05, 1.28) 0.304
2010-2020 1.19 (1.12, 1.26) 1.13 (1.03, 1.24)
IS 2000-2009 1.29 (1.20, 1.38) ＜0.001 1.20 (1.06, 1.36) 0.034 1.20 (1.06, 1.36) 0.06
2010-2020 1.05 (0.97, 1.14) 1.08 (0.97, 1.21) 1.09 (0.97, 1.22)
ICH 2000-2009 1.24 (1.12, 1.39) 0.428 1.09 (0.88, 1.35) 0.765
2010-2020 1.17 (1.05, 1.29) 1.17 (0.99.1.38)
SAH 2000-2009 1.07 (0.90, 1.26) 0.397 0.99 (0.73, 1.35) 0.454
2010-2020 0.95 (0.78, 1.16) 1.07 (0.82, 1.39)
mortality
TS 2000-2009 1.44 (1.33, 1.56) 0.746 1.12 (0.98, 1.27) 0.666
2010-2020 1.41 (1.30, 1.55) 1.19 (1.05, 1.35)
IS 2000-2009 1.34 (1.20, 1.50) 0.543 1.17 (0.98, 1.39) 0.745 1.17 (0.98, 1.39) 0.833
2010-2020 1.25 (1.08, 1.45) 1.16 (0.98, 1.37) 1.16 (0.98, 1.37)
ICH 2000-2009 1.31 (1.15, 1.51) 0.709 1.06 (0.82, 1.37) 0.614
2010-2020 1.26 (1.10, 1.44) 1.19 (0.98, 1.45)
SAH 2000-2009 1.12 (0.93, 1.35) 0.766 0.97 (0.70, 1.34) 0.702
2010-2020 1.07 (0.85, 1.35) 1.19 (0.90, 1.57)

The odds ratios (95% confidence interval) represent the proportion of outcomes at discharge (modified Rankin Scale [mRS] score of 0-2: favorable outcome, score of 5-6: unfavorable outcome, score of 6: mortality) between off-hour and on-hour arrival patients. The following univariate and multivariate multilevel mixed-effects logistic models were constructed: Model 1: adjustment was performed for age, sex, mRS score on admission, history of stroke, admission within 4.5 hours of onset, NIHSS score on admission, hypertension, dyslipidemia, diabetes mellitus, atrial fibrillation, and smoking. Model 2: reperfusion therapy was additionally adjusted for in Model 1 only for IS.

Interaction between years was represented, with p＜0.05 indicating a significant difference.

Table 3.Secular Changes in functional and life outcomes during off-hours vs on-hours in 2000–2020

crude model1 model2
OR 95%CI OR 95%CI OR 95%CI
favorable outcome TS 0.990 0.987–0.994 1.006 1.001–1.012
(0–2) IS 0.993 0.990–0.997 1.013 1.008-1.019 1.009 1.003-1.015
ICH 0.975 0.968–0.982 0.973 0.961-0.984
SAH 0.992 0.980–1.004 0.983 0.967-1.000
unfavorable outcome TS 0.986 0.981-0.990 0.970 0.963–0.976
(5–6) IS 0.985 0.981–0.990 0.967 0.959–0.975 0.972 0.964–0.980
ICH 0.995 0.988–1.003 0.986 0.972–0.999
SAH 0.988 0.975–1.001 0.987 0.969–1.005
mortality TS 0.969 0.963-0.976 0.980 0.971–0.989
IS 0.970 0.962–0.979 0.979 0.968–0.989 0.984 0.973–0.995
ICH 0.986 0.977–0.996 0.997 0.982–1.013
SAH 0.978 0.964–0.992 0.971 0.953–0.989

The odds ratios represent the proportion of outcomes at discharge between off-hour and on-hour arrival patients. The multivariate analysis of Model 1.2 is the same as that described in Figure 2. Bold: p＜0.05 indicating a significant difference.

CI: confidence interval; OR: odds ratio

4.7 The Proportions of the Onset to Arrival Time and Subtype of Ischemic Stroke between Weekdays vs. Weekends/Holidays and On-Hours vs. Off-Hours

There was a significant higher proportion of patients who arrived at hospital within 4.5 hours, as well as those with cardioembolism, during weekends and off-hours compared to weekday and on-hours (Fig.5).

Fig.5. The proportions of the onset to arrival time and subtype of ischemic stroke between (a) weekdays vs. weekends/holidays and (b) on-hours vs. off-hours in 2000–2020

Categorical data are expressed as percentage analyzed using χ2 tests. ＊: ＜0.001

5.Discussion

This study analyzed the effects of off-hour and weekend admissions on stroke outcomes using data from a nationwide registry in Japan from 2000 to 2020. We evaluated deaths during hospitalization, which have been previously reported, as well as functional outcomes at discharge, including successful conversion to favorable outcomes. Furthermore, data over a 20-year period were analyzed and compared between two decades to identify any changes. Our major findings were that TS, IS, and ICH patients exhibited weekend and off-hour effects on functional outcomes at discharge. However, the disparity between off-hours and on-hours for unfavorable outcomes in IS has improved over the past 20 years. This study is the first large registry report to analyze the effects of weekend and off-hour admissions on both life and functional prognoses over a two-decade period.

In our study, the OR for favorable outcomes was significantly lower during off-hours than during on-hours and on weekends/holidays than on weekdays for TS, IS, and ICH. This study demonstrated the weekend and off-hour effects on TS, IS, and ICH, with favorable outcomes as the endpoint. To the best of our knowledge, there are no reports evaluating functional prognosis during off-hours. Regarding the weekend effect, one small-sample-sized study reported better favorable outcomes at discharge for weekday admissions than for weekend admissions17); however, previous reports were only based on proportions, without any multivariate analysis; thus, no detailed analysis was conducted in these studies17-20). It is therefore difficult to draw conclusions based on current literature, as these previous reports lack details and used different analysis methods compared with that used in our study. Nevertheless, it is interesting to note that our functional analysis clearly demonstrated both weekend and off-hour effects after adjusting for patient factors.

This study demonstrated an off-hour effect on in-hospital mortality, as well as on unfavorable outcomes for TS and IS. Previous reports on this effect have presented conflicting evidence, with some studies noting its existence7, 21) and others observing its absence22-24), reports suggest that stroke treatment advances have mitigated this effect7, 22). In our study, the off-hour effect was evident from 2000 to 2020 for IS, and was present from 2000 to 2009 for unfavorable outcome, but it was not observed from 2010 to 2020. The effect reduction during the latter period could be due to the advancements in stroke care. When an off-hour effect is absent, it likely indicates consistent levels of acute stroke care regardless of the admission timing. Conversely, the presence of an off-hour effect implies that care during off-hours may be not well established. Levels of off-hour acute stroke care vary by country and the target of the study.

We did not observe a significant increase in the OR for in-hospital mortality during weekends/holidays. Several previous studies have shown a weekend effect on in-hospital mortality7, 21, 25-28), while others have refuted this association18, 19, 24, 29-34). Some reports have used administrative coding data, which consist of routine clinical information and have the advantage of allowing a large number of cases to be analyzed, but these records may lack detailed stroke diagnosis data because they are entered by non-physician personnel35).

Some of these data appear to be based on death surveys, which may not include detailed information, such as medical history, pre-stroke activities of daily living, NIHSS score at admission, and mRS score at discharge, all of which are widely used for functional and life prognosis and multivariate analysis. Therefore, differences in diagnosis and adjustment factors may have contributed to the inconsistent results reported in previous studies. Previous studies have reported that minor stroke patients wait for the weekend to see a doctor, which is the cause of the weekend effect34). We consider that this view is not applicable to the present study because there were significant differences after adjustment for admission within 4.5 hours of onset and NIHSS score on admission.

The patient background characteristics show a significantly high proportion of admissions within 4.5 hours during weekends and off-hours, which may be attributed to better hospital accessibility during weekends/holidays and off-hours, as well as the likelihood of patients receiving prompt treatment at the stroke center during these periods, bypassing visits to general clinics. Additionally, a previous JSDB study indicated that cardioembolism is associated with an elevated rate of in-hospital death4). The significant rise in hospitalizations for cardioembolism during weekend/holiday and off-hour in this study could be attributed to the onset and delayed detection of the condition during sleep, potentially leading to poorer outcomes. This was accounted for by adjusting for the NIHSS score on admission, reflecting patient severity.

In this study, after adjusting for patient factors, no effect of weekends/holidays or off-hours was seen for in-hospital mortality of ICH patients. Reports on effects of weekends on hemorrhagic stroke patients are limited8, 36). Even studies that have shown an effect acknowledge that their results may not have been fully adjusted for patient-related factors8). Our study also showed that the effects of admission time were strongly influenced by patient factors. However, our evaluation of favorable outcomes associated with admission time showed a significant effect, clarifying a previously unreported effect on functional outcomes in ICH. The lack of effective acute treatment for ICH, such as reperfusion therapy for IS, may explain why differences in mortality rates during off-hours or weekends were not observed.

We found no significant difference in SAH between off-hour admission and on-hour admission for any endpoints. On weekends and holidays, the variability in prognosis was attributed to patient factors related to SAH. Previous reports have also found no weekend or off-hour effect, even before adjusting for patient factors, in patients with SAH24, 37-39).

The results of our 20-year study suggest that unfavorable outcomes in IS patients improved during off-hours in 2010–2020 compared with that in 2000–2009 after adjusting for confounding factors. However, an additional adjustment for reperfusion therapy attenuated the interaction between the off-hour periods. Recently, we reported that overall mortality and unfavorable outcomes improved over 20-years in both men and women with IS, ICH, or SAH4). This improvement may be related to advancements in both preventive and acute treatments. We believe that our results may have been partly influenced by improvements in the stroke care system, which have enabled reperfusion therapy even during off-hours in Japan. During 2010–2020, several changes were made in stroke care systems, including the approval of mechanical thrombectomy in 2010, expansion of the therapeutic time window for intravenous thrombolysis up to 4.5 hours after onset starting in 2012, and certification of primary stroke centers by the Japan Stroke Society in 2019. Accordingly, attempts have been made to establish a practical system to accommodate these changes. The results of an assessment of the competence of the training institutions of the Japan Stroke Society and the Japan Neurosurgical Society, which are the main stroke care centers in Japan, including endovascular treatment, conducted during the 2010s have shown an improvement in competency40). Although we did not observe a significant interaction between the effect of weekends and the two decades for any outcome in Japan, there have been reports of a decrease in the effect of weekends over time in the United States8, 36). Improvements in the effects of weekend and off-hour admissions have been reported with the implementation of the Stroke Care Network in France and the development of stroke treatment systems in Denmark22, 41). In England, the mortality rate due to the weekend effect has improved with implementation of stroke services throughout the country29). Similarly, our study results demonstrate that the development of a stroke care system improves the imbalance in off-hour admissions.

The strength of this study is that it used data from a standardized stroke registry of 131 local acute stroke centers, covering a 21-year period, with a high likelihood of accurate diagnosis by trained stroke specialists with voluntary participation. This study also demonstrated long-term changes at 10-year intervals. This is the first study to thoroughly examine functional outcomes, in contrast to past evaluations that focused primarily on inpatient mortality.

However, our study has some limitations. In addition to the confounders used for multivariate adjustment in this study, other hospital factors causing these effects, such as medical, nursing, and rehabilitation staff composition, delays in admission to a stroke care unit, and potential bias in assessing swallowing function have been previously reported9). Due to insufficient data on these factors within the registry, we were unable to extract and analyze them for multivariate analysis, thus making detailed causal verification difficult. The JSDB data concerning the hospital arrival within 4.5 hours of stroke onset, serving as our adjustment factor, might not reflect the changes of IS treatment in Japan. Intravenous thrombolysis was indicated within 3 hours before 2012, and endovascular thrombectomy is indicated within 8 hours currently. Our registry collects the outcome data at discharge, which we consider to be acceptable for evaluation because there was almost no difference in length of stay between the respective groups. However, it is considered better to evaluate at a fixed time. Thus, a future report is awaited. Finally, since this study is based on a case registry from a voluntary stroke center and not a population-based registry, there may be a possibility of selection bias.

6.Conclusion

We demonstrated the weekend effect on functional outcomes and the off-hour effect not only on in-hospital mortality but also on functional outcomes in stroke patients. These results were likely influenced by the stroke care system, which standardized care and treatment throughout the week and on individual days. Furthermore, over the 20-year observation period, the difference in unfavorable outcomes between off-hour and on-hour admissions for IS has improved, which is thought to be partly due to the development of stroke care systems over the past decade.

Acknowledgements

This work was supported by JSPS KAKENHI Grant Number JP21K07472.

Notice of Grant Support

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflicts of Interest

Toyoda K reports honoraria from Daiichi-Sankyo, Otsuka, Bayer Yakuhin, Bristol-Myers-Squibb, and Novartis, outside the submitted work.

Koga M reports honoraria from Bayer Yakuhin, Daiichi-Sankyo, Mitsubishi Tanabe Pharma Corporation, BMS/Pfizer, BMS/Janssen Pharmaceuticals, and research support from Daiichi-Sankyo, Nippon Boehringer Ingelheim, all of which are outside of the submitted work.

Ethics Approval

The study protocol was approved by the Institutional Ethical Board of the National Cerebral and Cardiovascular Center (approval number: M27-090-14) as well as by the ethical boards of the collaborative research centers that collected the data.

Consent to Participate/Publish

Patient consent for inclusion in the registry was obtained using an opt-out method. Data of stroke patients were prospectively collected by stroke physicians or clinical coordinators using an online standardized database form. This registry complied with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

Data Availability

The dataset of the JSDB study is open to investigators who participate in this registry.

Author Contributions

Tomoya Omae, Michikazu Nakai, and Masatoshi Koga conceived and designed the analyses, wrote the first draft of the manuscript, and provided financial support. Nakai performed statistical analyses. Kazunori Toyoda, Masatoshi Koga, and Sohei Yoshimura had a central role in managing the registry. Kazunori Toyoda and Sohei Yoshimura revised the first draft of the manuscript. Tomoya Omae, Sohei Yoshimura, Kazunori Toyoda, Toshiharu Yanagisawa, and Masatoshi Koga acquired the data. Shotai Kobayashi supervised the registry. All other authors edited the manuscript and approved the final draft.

References
 

This article is licensed under a Creative Commons [Attribution-NonCommercial-ShareAlike 4.0 International] license.
https://creativecommons.org/licenses/by-nc-sa/4.0/
feedback
 Top