Abstract
Background
High patient volumes are associated with better stroke management and outcomes. However, in sparsely populated regions, patients with stroke frequently rely on low‐volume hospitals for acute care. This study evaluates the performance of high‐performing hospitals among low‐volume hospitals and identifies factors contributing to their high performance despite treating fewer stroke cases.
Methods and Results
We analyzed data from the eighth Acute Stroke Quality Assessment Program in 2018, alongside reimbursement claims and death certificate records, to evaluate stroke care in 248 hospitals across South Korea, including all general and tertiary hospitals. The hospitals treating <100 patients with stroke annually were classified as low‐volume. High performance among these hospitals was defined as achieving a defect‐free care rate of >75%. The outcome measured was the poststroke 1‐year mortality rate. Of 28 572 patients, 2521 (8.8%) were treated in low‐volume hospitals, primarily general hospitals outside the capital area. The 1‐year mortality rate was significantly higher in low‐volume hospitals (24.8%) compared with their high‐volume counterparts (17.6%, adjusted P<0.01). High‐performing low‐volume hospitals exhibited a significantly lower 1‐year mortality rate (21.1%) than low‐performing hospitals (26.9%, adjusted P=0.036), with no significant difference in the number of hospital beds or annual stroke admissions. High‐performing low‐volume hospitals were characterized by their location (P=0.048) and a significantly greater number of physicians, particularly neurologists (P=0.010). The presence of neurologists was also associated with a higher defect‐free care rate in these hospitals.
Conclusions
This study highlights the status of low‐volume hospitals in providing stroke care in areas with limited access to larger medical facilities. It suggests that enhancing acute stroke care performance, particularly through human resource investments, can significantly improve quality and patient outcomes at these hospitals.
Keywords: human resources, low‐volume hospital, neurologist, stroke, stroke care
Subject Categories: Cerebrovascular Disease/Stroke, Disparities, Health Services, Quality and Outcomes
Nonstandard Abbreviations and Acronyms
- ASQAP
Acute Stroke Quality Assessment Program
- EVT
endovascular thrombectomy
- IVT
intravenous thrombolysis
- mRS
modified Rankin Scale
Clinical Perspective.
What Is New?
This nationwide study of 248 hospitals in South Korea demonstrates that certain low‐volume hospitals (treating <100 patients with stroke annually) can achieve quality metrics and outcomes comparable with those of high‐volume hospitals, despite resource limitations.
High‐performing low‐volume hospitals achieved significantly lower 1‐year mortality rates (21.1%) compared with their low‐performing counterparts (26.9%).
The presence of neurologists and the hospital's geographic location, rather than facility size or technological capabilities, were identified as the key factors distinguishing high‐performing from low‐performing low‐volume hospitals.
What Are the Clinical Implications?
Although centralization of stroke care to high‐volume hospitals remains critical, low‐volume hospitals continue to play an essential role in providing access to acute stroke care in underserved regions.
Health care systems should prioritize enhancing staffing with stroke specialists in low‐volume hospitals to improve stroke care quality and patient outcomes in these areas.
Developing regional stroke care networks, with standardized protocols and clear referral systems, is vital to supporting low‐volume hospitals and ensuring consistent care quality.
The beneficial relationship between patient volume and clinical outcomes is well‐documented across various medical interventions, including percutaneous coronary intervention and various surgical procedures. 1 , 2 This association extends to stroke care, where studies have shown a link between the volume of endovascular thrombectomy (EVT) procedures and better outcomes in patients with ischemic stroke. 3 , 4 Several studies have previously reported that higher hospital volume is associated with better outcomes in stroke care. 5 , 6 , 7 Notably, both the annual volume of patients with stroke and the number of procedures such as EVT are consistently associated with reduced stroke mortality. 8
Despite the recognized benefits of high‐volume hospitals, such as enhanced management and improved patient outcomes, low‐volume hospitals are inevitable, especially in underserved areas. This is particularly true in sparsely populated regions, where such facilities are not only prevalent but also vital. 9 Often located in rural areas, these low‐volume hospitals serve an older demographic with multiple comorbidities and lower socioeconomic status. 10 Although telemedicine has mitigated some challenges in these settings, 11 not all patients with stroke benefit from such services. 12
Current literature acknowledges the general relationship between hospital volume and stroke outcomes but leaves a critical gap on the efficacy and performance of low‐volume hospitals in managing acute stroke care. 13 Given the vital role low‐volume hospitals play in providing access to care, it is important to understand the performance characteristics that distinguish high‐performing from low‐performing low‐volume hospitals in acute stroke care. 14 This study aims to characterize factors that differentiate high performance from low performance in low‐volume hospitals. By doing so, we hope to identify strategies to improve patient care quality across all hospital settings, regardless of stroke volume.
METHODS
Study Subjects
This study used data from the eighth assessment of the Acute Stroke Quality Assessment Program (ASQAP), a national audit program evaluating the quality of acute stroke care in South Korea, overseen by the Health Insurance Review and Assessment Service. 9 , 15 The ASQAP, established under the National Health Insurance Act and the Medical Aid Act, was initiated by the Health Insurance Review and Assessment Service in 2005. 16 The eighth assessment spanned 6 months, from July 2018 to December 2018. Annual stroke volumes were estimated by doubling the stroke cases reported during this period. Eligibility criteria for the ASQAP included patients who: (1) were admitted through an emergency room within 7 days of symptom onset, (2) were treated in general or tertiary hospitals reporting >10 acute stroke cases during the audit period, and (3) were primarily diagnosed with subarachnoid hemorrhage (I60), intracerebral hemorrhage (I61), or ischemic stroke (I63) according to the International Classification of Diseases, Tenth Revision (ICD‐10). The total number of hospitals in South Korea was 353 at the time of the eighth ASQAP survey. Hospitals that treated <10 stroke cases over a 6‐month period (105 hospitals) were excluded from the survey.
Data Collection
The study combined and analyzed data from the ASQAP, claims data for reimbursement, and death certificate records. 16 The ASQAP data included information on hospital treatment capacities, staff composition (neurologists, neurosurgeons, and physiatrists), patient demographics, stroke characteristics, quality indicators, and outcomes. Information on acute stroke management not covered in the ASQAP, such as EVT, was extracted from claims data. Mortality data for stroke cases were sourced from death certificate records, ensuring comprehensive mortality information for all cases under review. Additionally, detailed hospital information, such as 24/7 availability of intravenous thrombolysis (IVT) and EVT, was obtained from a previous study. 17
Classification of High‐Volume and Low‐Volume Hospitals
The classification of hospitals based on their annual stroke case volume was derived from data included in the eighth ASQAP assessment, encompassing 248 hospitals nationwide. Hospitals were categorized with those treating <100 stroke cases per year (or <50 cases during the 6‐month evaluation period) defined as low‐volume hospitals. In contrast, hospitals with ≥100 annual stroke cases (or ≥50 cases in the 6‐month period) were designated as high‐volume hospitals. 8 According to this classification, 124 hospitals were identified as low‐volume and the remaining 124 as high‐volume hospitals (Figure 1A).
Figure 1. Annual stroke volume per hospital and hospital location according to patient volume and performance.
A, Annual stroke volume per hospital. B, Hospital location of high‐volume (red) and low‐volume (blue) hospitals. C, Hospital location of high‐performance (green) and low‐performance (orange) low‐volume hospitals.
To clarify the effect of hospital volume on stroke outcomes, we performed additional sensitivity analyses. These analyses categorized hospitals treating the lowest 20% of cumulative patient volumes as low‐volume hospitals (163 hospitals) and those treating the highest 20% as high‐volume hospitals (85 hospitals; Figure 1A).
Classification of High‐Performance and Low‐Performance Hospitals Within Low‐Volume Hospitals
In our analysis of low‐volume hospitals, we defined high‐performance hospitals as those achieving a defect‐free care rate of ≥75%. 18 Conversely, low‐performance hospitals were identified based on a defect‐free care rate <75%. The defect‐free care rate is a composite measure reflecting the proportion of patients who received all applicable evidence‐based treatments correctly and timely. 15 These treatments, derived from the performance measures used in stroke center certification programs by The Joint Commission, 19 include: (1) discharge on antithrombotic therapy, (2) anticoagulation therapy for patients with atrial fibrillation or flutter, (3) administration of IVT within 60 minutes of arrival, (4) prescription of antithrombotic therapy within 48 hours of hospitalization, and (5) performance of in‐hospital lipid testing.
Applying these criteria to the cohort of 124 low‐volume hospitals, we identified 42 hospitals as high‐performance based on their adherence to these specified measures. The remaining 82 hospitals were categorized as low‐performance hospitals (Figure S1). To further assess the quality of acute stroke care across high‐ and low‐ performance hospitals, we incorporated additional stroke performance measures (stroke [STK] and comprehensive stroke [CSTK]) as defined by The Joint Commission.
Statistical Analysis
The primary outcome measure was the 1‐year mortality rate, and secondary outcomes were mortality rates at 1 and 3 months, along with discharge modified Rankin Scale (mRS) scores. 20 , 21 Categorical variables are presented as frequencies (percentages), and continuous variables are shown as means±SDs or medians with interquartile ranges. The patient and hospital characteristics were compared between groups based on hospital volume and performance using Student t test or Mann–Whitney U test for continuous variables, and Pearson χ2 test or Fisher exact test for categorical variables as appropriate. The 1‐year cumulative mortality was determined and compared between the designated groups using log‐rank tests. A Cox proportional hazards model was also applied to compare 1‐year cumulative mortalities between the designated groups adjusting for age, sex, stroke severity (National Institutes of Health Stroke Scale or Glasgow Coma Scale), and stroke type (ischemic stroke, intracerebral hemorrhage, or subarachnoid hemorrhage), drawing on previous research. Distribution of mRS scores at discharge was assessed through ordinal logistic regression, controlling for the same variables.
In terms of hospital‐level analysis, the adjusted 1‐year mortality was additionally calculated for each hospital only focusing on lower‐volume hospital through the method of model‐based risk‐adjustment adjusting for age, sex, stroke severity, and stroke subtype. 22 The generalized linear model was applied to ascertain the efficiency of high performance on the adjusted 1‐year mortality with a β distribution and logit link function. Dichotomized number of neurologist (0 or >0) was considered in the model for adjustment due to a significant association between number of neurologists and defect‐free care (verified by ANOVA).
For variables with <5% missing data, simple imputation strategies were applied. Continuous variables were imputed using the median value, whereas categorical variables were imputed using the mode, with both procedures applied on a per‐hospital basis. For hospitals with >50% missing data, group‐level imputation was used. Due to the absence of Glasgow Coma Scale data in ≈25% of hemorrhagic stroke cases, a categorization approach for both National Institutes of Health Stroke Scale and Glasgow Coma Scale scores was introduced. This allowed us to replace missing Glasgow Coma Scale values with corresponding National Institutes of Health Stroke Scale‐based categories, facilitating effective management of missing data through this streamlined methodology.
All statistical analyses were conducted using SAS software (version 9.4; SAS Institute, Cary, NC), with a P value of <0.05 considered as statistically significant.
Standard Protocol Approvals, Registrations, and Patient Consents
The Seoul National University Bundang Hospital Institutional Review Board determined that participant consent was not required and approved this study, which adheres to the Declaration of Helsinki (institutional review board number X‐2202‐736‐902).
Data Availability
In compliance with the Act on the Protection of Personal Information Maintained by Public Institutions, anonymized data were accessible only to authorized researchers and scheduled for deletion following the conclusion of the study, thus limiting data availability.
RESULTS
The eighth ASQAP assessment included 28 572 patients with an average age of 68.2 years, and 44.1% were women (Table S1). Stroke types were 76.4% ischemic, 15.2% intracerebral hemorrhage, and 8.4% subarachnoid hemorrhage. At 1‐year poststroke, 18.2% (5196) of patients had died.
Low‐Volume Versus High‐Volume Hospitals
During the 6‐month period, 8.8% of patients were admitted to 124 low‐volume hospitals (Table 1). Compared with those admitted to high‐volume hospitals, these patients were older, more often women, had more severe neurological deficits, and were less likely to arrive by ambulance. Admission to stroke units was rare (only 2.7%). Patients with ischemic stroke experienced longer door‐to‐needle times, and there was less use of IVT and EVT. Performance on The Joint Commission's stroke measures was lower, with low‐volume hospitals achieving defect‐free care of 60.8% compared with 80.8% at high‐volume hospitals (Table S2).
Table 1.
Comparison of Patient Characteristics and Stroke Management According to Stroke Patient Volume and Performance
| Characteristics and Management | All hospitals | Low‐volume hospitals | ||||
|---|---|---|---|---|---|---|
| High volume (n=26 051) | Low volume (n=2521) | P value | High performance (n=915) | Low performance (n=1606) | P value | |
| Age, y, mean±SD | 67.7±13.8 | 72.5±13.2 | <0.001 | 71.3±13.3 | 73.2±13.0 | <0.001 |
| Sex, women, n (%) | 11 417 (43.8) | 1176 (46.7) | <0.001 | 399 (43.6) | 777 (48.4) | 0.02 |
| Stroke types, n (%) | ||||||
| Ischemic stroke | 19 748 (75.8) | 2082 (82.6) | <0.001 | 744 (81.3) | 1338 (83.3) | 0.44 |
| ICH | 3985 (15.3) | 359 (14.2) | 140 (15.3) | 219 (13.6) | ||
| SAH | 2318 (8.9) | 80 (3.2) | 31 (3.4) | 49 (3.1) | ||
| Use of ambulance, n (%) | 15 568 (59.8) | 1325 (52.6) | <0.001 | 465 (50.8) | 860 (53.6) | 0.19 |
| Onset to arrival time, min, median (IQR) | 272.0 (76.0–1258) | 310.0 (65.0–1316) | 0.81 | 417.0 (90.0–1440) | 240.0 (60.0–1203) | 0.02 |
| Unclear onset, n (%) | 6210 (23.8) | 403 (16.0) | <0.001 | 135 (14.8) | 268 (16.7) | 0.20 |
| NIHSS score at admission, median (IQR)* | 3.0 (1.0–7.0) | 4.0 (2.0–8.0) | <0.001 | 3.0 (1.0–6.0) | 4.0 (2.0–8.0) | <0.001 |
| GCS score at admission, median (IQR)† | 14.0 (8.0–15.0) | 12.5 (5.0–15.0) | 0.02 | 12.0 (5.0–15.0) | 13.0 (6.0–15.0) | 0.44 |
| Categorized NIHSS or GCS score at admission, n (%)‡ | ||||||
| 0≤NIHSS≤6 or GCS=15 | 16 731 (64.6) | 1559 (66.2) | 0.14 | 604 (69.8) | 955 (64.1) | 0.02 |
| 7≤NIHSS≤10 or GCS=14 | 3009 (11.6) | 281 (11.9) | 88 (10.2) | 193 (13.0) | ||
| 11≤NIHSS≤15 or 12≤GCS≤13 | 2190 (8.5) | 172 (7.3) | 52 (6.0) | 120 (8.1) | ||
| 16≤NIHSS≤42 or 3≤GCS≤11 | 3973 (15.3) | 342 (14.5) | 121 (14.0) | 221 (14.8) | ||
| Door‐to‐image time, min, median (IQR) | 22.0 (14.0–36.0) | 18.0 (10.0–31.0) | <0.001 | 19.0 (10.0–32.0) | 18.0 (10.0–30.0) | 0.051 |
| First imaging tool, n (%)§ | ||||||
| CT | 19 621 (88.4) | 1810 (77.3) | <0.001 | 633 (75.2) | 1177 (78.4) | 0.07 |
| MRI | 2576 (11.6) | 533 (22.8) | 209 (24.8) | 324 (21.6) | ||
| Admission to stroke unit, n (%) | 18 073 (69.4) | 67 (2.7) | <0.001 | 46 (5.0) | 21 (1.3) | <0.001 |
| Ischemic stroke | (n=19 748) | (n=2082) | (n=744) | (n=1338) | ||
| IV tPA, n (%) | 1293 (6.6) | 76 (3.7) | <0.001 | 19 (2.6) | 57 (4.3) | 0.047 |
| Door‐to‐needle time, min, median (IQR) | 47.0 (34.0–57.0) | 58.0 (48.0–82.0) | <0.001 | 48.0 (39.0–78.0) | 59.0 (53.0–82.0) | 0.67 |
| EVT, n (%) | 2095 (10.6) | 47 (2.3) | <0.001 | 13 (1.8) | 34 (2.5) | 0.24 |
| Combined IV tPA and EVT, n (%) | 633 (3.2) | 11 (0.5) | <0.001 | 3 (0.4) | 8 (0.6) | 0.56 |
| Carotid endarterectomy, n (%) | 62 (0.3) | 4 (0.2) | 0.53 | 2 (0.3) | 2 (0.2) | 0.62 |
| Angioplasty or carotid artery stenting, n (%) | 479 (2.4) | 23 (1.1) | <0.001 | 13 (1.8) | 10 (0.8) | 0.048 |
| ICH | (n=3985) | (n=359) | (n=140) | (n=219) | ||
| Decompressive surgery, n (%) | 1025 (25.7) | 62 (17.3) | <0.001 | 25 (17.9) | 37 (16.9) | 0.81 |
| SAH | (n=2318) | (n=80) | (n=31) | (n=49) | ||
| Aneurysm clipping, n (%) | 575 (24.8) | 3 (3.8) | <0.001 | 2 (6.5) | 1 (2.0) | 0.31 |
| Aneurysm coiling, n (%) | 1230 (53.1) | 35 (43.8) | 0.10 | 16 (51.6) | 19 (38.8) | 0.26 |
CT indicates computed tomography; EVT, endovascular thrombectomy; GCS, Glasgow Coma Scale; ICH, intracerebral hemorrhage; IQR, interquartile range; IV tPA, intravenous tissue plasminogen activator; MRI, magnetic resonance imaging; NIHSS, National Institutes of Health Stroke Scale; and SAH, subarachnoid hemorrhage.
All hospitals: Of the total 20 952 ischemic stroke cases, 878 were missing cases. Low‐volume hospitals: Of the total 1911 ischemic stroke cases, 171 were missing cases.
All hospitals: Of the total 5065 SAH and ICH cases, 1677 were missing cases. Low‐volume hospitals: Of the total 166 SAH and ICH cases, 273 were missing cases.
All hospitals: Missing 315 cases (high volume 148 cases, low volume 167 cases). Low‐volume hospitals: Missing 167 cases (high performance 50 cases, low performance 117 cases).
All hospitals: Missing 4032 cases (high volume 3854 cases, low volume 178 cases). Low‐volume hospitals: Missing 178 cases (high performance 73 cases, low performance 105 cases).
Low‐volume hospitals, mostly general hospitals, had fewer beds (average 286.5 versus 676.5) and treated less than one‐tenth of the annual patients with stroke than high‐volume hospitals (2521 versus 26 051; Tables 1 and 2). Primarily located outside the capital area, all rural hospitals (in a eup [town] or myeon [township]) were low‐volume, none were certified stroke centers, and less than half offered 24‐hour IVT, with only 19.4% providing EVT (Figure 1B). Neurology departments lacked an on‐call service provider system, and neurosurgery on‐call services were available in only 3 hospitals.
Table 2.
Comparison of Hospital Characteristics Between High‐Volume Versus Low‐Volume Hospitals
| Characteristics | Total (N=248) | High volume (N=124) | Low volume (N=124) | P value |
|---|---|---|---|---|
| Hospital type, n (%) | ||||
| Tertiary hospital | 44 (17.7) | 44 (35.5) | 0 | <0.001 |
| General hospital | 204 (82.3) | 80 (64.5) | 124 (100.0) | |
| Hospital beds, mean±SD | 481.5±328.7 | 676.5±365.7 | 286.5±81.2 | <0.001 |
| Annual stroke admission volume, mean±SD | 234.5±267.3 | 427.2±260.7 | 41.7±24.0 | <0.001 |
| Hospital location, n (%) | ||||
| Seoul capital (metropolitan) area* | 102 (41.1) | 61 (49.2) | 41 (33.1) | 0.04 |
| Noncapital area | 146 (58.9) | 63 (50.8) | 83 (66.9) | |
| Urban area (dong) | 237 (95.6) | 124 (100.0) | 113 (91.1) | <0.001 |
| Rural area (eup or myeon) | 11 (4.4) | 0 | 11 (8.9) | |
| Presence of certified stroke center, n (%) | 51 (68.0) | 51 (71.8) | 0 | 0.02 |
| 24/7 IVT available center (self‐report) | 176 (71.0) | 115 (92.7) | 61 (49.2) | <0.001 |
| 24/7 EVT available center (self‐report) | 132 (53.2) | 108 (87.1) | 24 (19.4) | <0.001 |
| Presence of neurologist, n (%) | 225 (90.7) | 123 (99.2) | 102 (82.3) | <0.001 |
| Presence of neurologist on‐call system, n (%) | 79 (31.9) | 79 (63.7) | 0 | <0.001 |
| Presence of neurosurgeon, n (%) | 240 (96.8) | 124 (100.0) | 116 (93.6) | 0.002 |
| Presence of neurosurgeon on‐call system, n (%) | 85 (34.3) | 82 (66.1) | 3 (2.4) | <0.001 |
| Presence of physiatrist, n (%) | 187 (75.4) | 122 (98.4) | 65 (52.4) | <0.001 |
| Presence of both neurologist and neurosurgeon, and physiatrist, n (%) | 177 (71.4) | 121 (97.6) | 56 (45.2) | <0.001 |
EVT indicates endovascular thrombectomy; and IVT, intravenous thrombolysis.
Seoul capital (metropolitan) area comprises Seoul, Incheon, and Gyeonggi‐do.
For the primary outcome, 1‐year mortality, the rate was significantly higher at low‐volume hospitals compared with high‐volume hospitals (24.8% versus 17.6%; Figure 2, Tables S3 and S4). This difference remained significant after adjusting for age, sex, stroke severity, and stroke type (adjusted P<0.01). Furthermore, discharge mRS scores were worse in low‐volume hospitals (Table S3, Figure S2). The same pattern of results was confirmed in sensitivity analysis (Tables S5 and S6).
Figure 2. Cumulative stroke mortality rates of high‐volume (red) and low‐volume (blue) hospitals.
A, All patients with stroke. B, Patients with ischemic stroke. C, Patients with hemorrhagic stroke.
High‐Performance Versus Low‐Performance Hospitals Within Low‐Volume Hospitals
When classifying low‐volume hospitals into high‐performance (n=42) and low‐performance (n=82) categories, high‐performance hospitals had slightly younger patients and fewer women (Table 1). Acute stroke management quality indicators, including door‐to‐image time and imaging modalities, showed no significant differences. However, high‐performance hospitals had a higher stroke unit admission rate (5.0% versus 1.3%). For ischemic strokes, these hospitals had a slightly lower IVT rate (2.6% versus 4.3%) but quicker door‐to‐needle times (48 versus 59 minutes) and more frequent use of stenting, although use of such interventions was generally rare in both hospital categories.
High‐performance hospitals achieved a notably higher defect‐free care rate (81.9%) than low‐performance hospitals (49.1%), slightly exceeding the rate of high‐volume hospitals (80.8%) (Table S2). Their performance on STK/CSTK measures was also better, making their overall performance measures on par with high‐volume hospitals, except for the rate of IVT administration (Table S2).
The heightened process quality translated into better patient outcomes at high‐performance hospitals (Figure 3, Figure S3, Table S7, and Table S8). The 1‐year crude mortality rate for patients with acute stroke was significantly lower in high‐performance hospitals (21.1%) compared with low‐performance hospitals (26.9%), with a 5.8% difference, which remained statistically significant after adjustments.
Figure 3. Cumulative stroke mortality rates of high‐performance (green) and low‐performance (orange) hospitals within low‐volume hospitals.
A, All Patients with stroke. B, Patients with ischemic stroke. C, Patients with hemorrhagic stroke.
All high‐ and low‐performance low‐volume hospitals were general hospitals, with no significant differences in bed numbers or annual stroke volumes (Table 3). A higher percentage of high‐performance hospitals were in the Seoul metropolitan area (47.6% versus 25.6%) (Figure 1C). Both groups showed similar capacities for delivering 24‐hour IVT and EVT. A notable difference was the median number of neurologists, 2 in high‐performance hospitals versus 1 in low‐performance hospitals, highlighting the potential impact of specialized stroke care staff on quality outcomes (Table 3, Figures S4 and S5).
Table 3.
Comparison of Hospital Characteristics Between High‐Performance Versus Low‐Performance Within Low‐Volume Hospitals
| Characteristics | High performance (N=42) | Low performance (N=82) | P value |
|---|---|---|---|
| Hospital type, general hospital, n (%) | 42 (100.0) | 82 (100.0) | |
| Hospital beds, mean±SD | 286.3±79.3 | 286.6±82.6 | 0.98 |
| Annual stroke admission volume, mean±SD | 44.7±22.1 | 40.2±24.8 | 0.33 |
| Hospital location, n (%) | |||
| Seoul capital (metropolitan) area | 20 (47.6) | 21 (25.6) | 0.048 |
| Noncapital area | 22 (52.4) | 61 (74.4) | |
| Urban area (dong) | 40 (95.2) | 73 (89.0) | 0.55 |
| Rural area (eup or myeon) | 2 (4.8) | 9 (11.0) | |
| Presence of local emergency medical center, n (%) | 16 (38.1) | 27 (32.9) | 0.85 |
| 24/7 IVT available center (self‐report), n (%) | 21 (50.0) | 40 (48.8) | 0.99 |
| 24/7 EVT available center (self‐report), n (%) | 9 (21.4) | 15 (18.3) | 0.92 |
| Presence of neurologist, n (%) | 39 (92.9) | 63 (76.8) | 0.03 |
| No. of neurologists, n (%) | |||
| 0 | 3 (7.1) | 19 (23.2) | 0.010 |
| 1 | 15 (35.7) | 40 (48.8) | |
| 2 | 20 (47.6) | 19 (23.2) | |
| ≥3 | 4 (9.5) | 4 (4.9) | |
| Presence of neurosurgeon, n (%) | 41 (97.6) | 75 (91.5) | 0.19 |
| No. of neurosurgeons, n (%) | |||
| 0 | 1 (2.4) | 7 (8.5) | 0.18 |
| 1 | 12 (28.6) | 33 (40.2) | |
| 2 | 18 (42.9) | 30 (36.6) | |
| ≥3 | 11 (26.2) | 12 (14.6) | |
| Presence of physiatrist, n (%) | 26 (61.9) | 39 (47.6) | 0.13 |
| No. of physiatrists, n (%) | |||
| 0 | 16 (38.1) | 43 (52.4) | 0.24 |
| 1 | 21 (50.0) | 34 (41.5) | |
| ≥2 | 5 (11.9) | 5 (6.1) | |
| Presence of both neurologist and neurosurgeon and physiatrist, n (%) | 24 (57.1) |
32 (39.0) |
0.055 |
EVT indicates endovascular thrombectomy; and IVT, intravenous thrombolysis.
We investigated the relationship between defect‐free care rate and the number of neurologists in low‐volume hospitals. Even after adjusting for other factors such as hospital volume, location, and number of patients with stroke, it showed that defect‐free care rate significantly increased with the presence of neurologists (Table S9, Figure S6). However, there was no significant relationship between the number of neurologists and the risk‐adjusted 1‐year mortality rate in the low‐volume hospitals.
DISCUSSION
This study underscores the disparity in performance and outcomes between low‐ and high‐volume hospitals treating patients with acute stroke, with high‐volume hospitals generally achieving better results. However, within the category of low‐volume hospitals, some hospitals are high performers, achieving better outcomes compared with their lower‐performing counterparts. Notably, there were no significant differences in the number of hospital beds, annual stroke volumes, or the ability to offer 24‐hour recanalization therapy across these low‐volume hospitals. Rather, we found that high‐ and low‐performing low‐volume hospitals differed in their location and the number of neurologists.
The volume‐outcome relationship for acute stroke care is well‐supported by prior research. 3 , 23 , 24 , 25 However, a significant variability in mortality rates among small‐ to medium‐sized hospitals suggests that some may provide excellent care, despite the general trend of higher 7‐day and discharge mortality rates observed in low‐volume hospitals (ie, <50 stroke cases per year) when compared with high‐volume hospitals. 5
Our study focused on low‐volume hospitals, uncovering that they have the potential to deliver better care and outcomes. Whereas low‐volume hospitals had a 1‐year mortality rate of 24.8%, higher than the 17.6% rate observed in high‐volume hospitals, within the low‐volume category, high‐performance hospitals demonstrated a lower 1‐year mortality rate of 21.1% compared with 26.9% found in low‐performance hospitals. Moreover, high‐performing low‐volume hospitals achieved defect‐free care rates comparable to those of high‐volume hospitals (81.9% versus 80.8%).
The criteria for defect‐free care in this study were based on performance measures established by The Joint Commission for stroke center accreditation. 19 The same quality indicators are used by the Get With The Guidelines‐Stroke program in the United States 26 and the Angels Initiative. 27 Adoption of these quality improvement programs and performance measures has been linked to better stroke outcomes, including reduced mortality and better functional outcomes. 26 , 28 The findings of our study support the belief that better performance translates into improvements in patient outcomes even in low‐volume hospital settings.
Determining the factors that lead to high performance in health care is complex. A systematic review has identified 7 key factors associated with high performance, underscoring the importance of building and sustaining a team of skilled personnel as a critical element. 29 Furthermore, research examining the characteristics of high‐performing hospitals in the context of acute myocardial infarction care has highlighted several distinctive features: active involvement of senior management, broad staff presence and expertise specific to acute myocardial infarction care, effective communication and coordination among different groups within the hospital, and the ability to solve problems and learn from experiences across the organization. 30 In addition, a study focusing on anticoagulation clinics found that adequate staffing and the efficient use of support staff were significant factors in achieving high performance. 31 These studies collectively suggest that a combination of leadership, staff expertise, interteam collaboration, and continuous learning and improvement are essential for delivering high‐quality health care outcomes.
Our study suggests that neurologist availability may be a factor that distinguishes between high‐ and low‐performing low‐volume hospitals, with high performers typically having 2 neurologists compared with the median of 1 in low performers. Other factors like bed size, stroke admission volume, and the ability to offer 24‐hour recanalization therapy showed no significant differences.
Literature on staffing's impact on stroke outcomes is limited. A study based on data from 113 US academic hospitals dating back to between 1997 and 1999 found lower mortality rates for patients with ischemic stroke admitted under neurologists compared with generalists, but this advantage disappeared when accounting for unmeasured confounders using an instrumental variable model. 32 Another study from a similar period (1991) analyzing Medicare data in the United States, found that although care by a neurologist was more costly, it led to better outcomes. 33 However, a more recent analysis (2016–2018) of Medicare data did not link involvement of a neurologist to improvement of outcomes. 34 Despite high availability of neurologists and neurosurgeons in South Korea (>90%), 17 our findings suggest that the presence of neurologists may be associated with improved defect‐free stroke care rates. This association highlights the potential leadership role of stroke neurologists in organizing stroke teams and implementing evidence‐based care protocols. Further research is needed to elucidate the mechanisms behind these observations and to determine the most effective strategies for improving care quality.
Another trait distinguishing high‐performing low‐volume hospitals is their greater likelihood of being geographically located in the Seoul metropolitan area. Urban–rural differences in stroke care and outcomes are recognized. 35 , 36 Beyond patient characteristics, disparities may stem from variation in prehospital care, interhospital transfers, and access to emergency services within hospitals. 37 Broader resources, including available hospitals for transfer, may significantly impact stroke mortality. 38
Low‐volume hospitals, prevalent in rural areas with less dense populations, are essential for bridging service gaps in these areas. Rural hospitals often struggle to maintain on‐call systems due to a scarcity of specialized staff, hindering the delivery of essential stroke treatments like IVT and EVT. A potential remedy observed in the United Kingdom and Denmark involves centralizing services by merging nearby low‐volume hospitals to streamline care. 39 , 40 Additionally, developing and implementing region‐specific prehospital notification and referral systems could enhance stroke care, ensuring timely and appropriate treatment for patients, irrespective of their location. 41
Our findings suggest that some low‐volume hospitals may exhibit high‐reliability organization attributes in delivering stroke care, transcending hospital location and neurologist availability. Previous studies have demonstrated that adopting a high‐reliability hospital model can enhance patient safety by reducing mortality and complication rates. 42 Although our current study did not directly investigate socio‐technical factors such as teamwork, communication, or safety culture, these elements likely contribute to the observed performance of high‐performing low‐volume hospitals. 43 , 44 Future qualitative research is warranted to explore these factors and uncover the mechanisms enabling certain low‐volume hospitals to excel in patient safety.
Our study is subject to several limitations. First, it relies on data from the ASQAP, which included only patients with acute stroke treated in acute care hospitals in South Korea, potentially limiting the generalizability of our findings. Another limitation is that small hospitals were not included in this government‐led survey, which may affect generalizability. Second, the study lacks data on 3‐month mRS scores, a common primary outcome for acute stroke trials and other stroke studies. Instead, we studied 1‐year mortality rates. Although discharge mRS data were incomplete, comprehensive mortality data were available for all study participants. Third, the ASQAP data set did not include door‐to‐puncture times, other specific time metrics related to EVT, or certain patient‐specific factors. In low‐volume hospitals, stroke outcomes may be influenced by both patient characteristics (eg, low socioeconomic status, comorbidities) and hospital factors (eg, limited stroke care resources). We observed that patients in low‐volume hospitals were older, had more severe neurological deficits, and experienced lower rates of ambulance use and stroke unit admission. Although the study could not account for all patient‐specific variables due to data limitations, these findings suggest that socioeconomic factors may contribute to difficulties in accessing timely medical care. Importantly, the observed mortality differences persisted even after adjusting for age, sex, and stroke severity, highlighting the robustness of our results. Despite these limitations, the study's national scope and relatively large numbers are significant strengths, because the study covered nearly all patients with stroke in Korea during the study period and provided complete mortality data.
In conclusion, the research highlights performance disparities in stroke care among low‐volume hospitals and factors associated with successful performance in a low‐volume hospital setting. The study underscores the potential importance of an adequate number of neurologists as a staffing resource in the successful care of patients with acute stroke. Our study suggests the need for further exploration of socio‐technical factors, such as teamwork, communication, and coordination. Future studies are warranted to explore whether enhancing the attributes of low‐volume hospitals could lead to improvement of stroke outcomes.
Sources of Funding
None.
Disclosures
None.
Supporting information
Tables S1–S9
Figures S1–S6
Acknowledgments
This study was performed by the Joint Project (M20220127799) on Quality Assessment Research of the Health Insurance Review and Assessment Service, Republic of Korea.
This article was sent to Imama A. Naqvi, MD, MS, Assistant Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.124.038348
For Sources of Funding and Disclosures, see page 11.
References
- 1. Finks JF, Osborne NH, Birkmeyer JD. Trends in hospital volume and operative mortality for high‐risk surgery. N Engl J Med. 2011;364:2128–2137. doi: 10.1056/NEJMsa1010705 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Jalbert JJ, Gerhard‐Herman MD, Nguyen LL, Jaff MR, Kumamaru H, Williams LA, Chen CY, Liu J, Seeger JD, Rothman AT, et al. Relationship between physician and hospital procedure volume and mortality after carotid artery stenting among Medicare beneficiaries. Circ Cardiovasc Qual Outcomes. 2015;8:S81–S89. doi: 10.1161/circoutcomes.114.001668 [DOI] [PubMed] [Google Scholar]
- 3. Gupta R, Horev A, Nguyen T, Gandhi D, Wisco D, Glenn BA, Tayal AH, Ludwig B, Terry JB, Gershon RY, et al. Higher volume endovascular stroke centers have faster times to treatment, higher reperfusion rates and higher rates of good clinical outcomes. J Neurointerv Surg. 2013;5:294–297. doi: 10.1136/neurintsurg-2011-010245 [DOI] [PubMed] [Google Scholar]
- 4. Kim BM, Baek JH, Heo JH, Kim DJ, Nam HS, Kim YD. Effect of cumulative case volume on procedural and clinical outcomes in endovascular thrombectomy. Stroke. 2019;50:1178–1183. doi: 10.1161/strokeaha.119.024986 [DOI] [PubMed] [Google Scholar]
- 5. Svendsen ML, Ehlers LH, Ingeman A, Johnsen SP. Higher stroke unit volume associated with improved quality of early stroke care and reduced length of stay. Stroke. 2012;43:3041–3045. doi: 10.1161/STROKEAHA.111.645184 [DOI] [PubMed] [Google Scholar]
- 6. Hall RE, Fang J, Hodwitz K, Saposnik G, Bayley MT. Does the volume of ischemic stroke admissions relate to clinical outcomes in the Ontario stroke system? Circ Cardiovasc Qual Outcomes. 2015;8:S141–S147. doi: 10.1161/CIRCOUTCOMES.115.002079 [DOI] [PubMed] [Google Scholar]
- 7. Lee KJ, Kim JY, Kang J, Kim BJ, Kim SE, Oh H, Park HK, Cho YJ, Park JM, Park KY, et al. Hospital volume and mortality in acute ischemic stroke patients: effect of adjustment for stroke severity. J Stroke Cerebrovasc Dis. 2020;29:104753. doi: 10.1016/j.jstrokecerebrovasdis.2020.104753 [DOI] [PubMed] [Google Scholar]
- 8. Saposnik G, Baibergenova A, O'Donnell M, Hill MD, Kapral MK, Hachinski V. Hospital volume and stroke outcome: does it matter? Neurology. 2007;69:1142–1151. doi: 10.1212/01.wnl.0000268485.93349.58 [DOI] [PubMed] [Google Scholar]
- 9. Kim JY, Lee KJ, Kang J, Kim BJ, Kim SE, Oh H, Park HK, Cho YJ, Park JM, Park KY, et al. Acute stroke Care in Korea in 2013‐2014: National Averages and disparities. J Korean Med Sci. 2020;35:e167. doi: 10.3346/jkms.2020.35.e167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Marquaille H, Clément G, Lenne X, Pruvot FR, Truant S, Theis D, El Amrani M. Predictive factors for utilization of a low‐volume center in pancreatic surgery: a nationwide study. J Visc Surg. 2021;158:125–132. doi: 10.1016/j.jviscsurg.2020.06.004 [DOI] [PubMed] [Google Scholar]
- 11. Wechsler LR, Demaerschalk BM, Schwamm LH, Adeoye OM, Audebert HJ, Fanale CV, Hess DC, Majersik JJ, Nystrom KV, Reeves MJ, et al. Telemedicine quality and outcomes in stroke: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2017;48:e3–e25. doi: 10.1161/str.0000000000000114 [DOI] [PubMed] [Google Scholar]
- 12. Zhang D, Wang G, Zhu W, Thapa JR, Switzer JA, Hess DC, Smith ML, Ritchey MD. Expansion of Telestroke services improves quality of care provided in super rural areas. Health Aff (Millwood). 2018;37:2005–2013. doi: 10.1377/hlthaff.2018.05089 [DOI] [PubMed] [Google Scholar]
- 13. Hahn M, Gröschel S, Tanyildizi Y, Brockmann MA, Gröschel K, Uphaus T. The bigger the better? Center volume dependent effects on procedural and functional outcome in established endovascular stroke centers. Front Neurol. 2022;13:828528. doi: 10.3389/fneur.2022.828528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Goodwin K, Tobler L. Improving rural health: state policy options. Nat Congress State Legislatures. 2016. [Google Scholar]
- 15. Park HK, Kim SE, Cho YJ, Kim JY, Oh H, Kim BJ, Kang J, Lee KJ, Jang MU, Park JM, et al. Quality of acute stroke care in Korea (2008‐2014): retrospective analysis of the nationwide and nonselective data for quality of acute stroke care. Eur Stroke J. 2019;4:337–346. doi: 10.1177/2396987319849983 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kim JY, Kang J, Kim BJ, Kim SE, Kim DY, Lee KJ, Park HK, Cho YJ, Park JM, Lee KB, et al. Annual case volume and one‐year mortality for endovascular treatment in acute ischemic stroke. J Korean Med Sci. 2022;37:e270. doi: 10.3346/jkms.2022.37.e270 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lee KB, Lee JS, Lee J‐Y, Kim JY, Jeong H‐Y, Kim S‐E, Kim J, Kim DY, Lee K‐J, Kang J, et al. Quality of acute stroke care within emergency medical service system in Korea: proposal for severe emergency medical center. J Korean Neurol Assoc. 2023;41:18–30. doi: 10.17340/jkna.2023.1.3 [DOI] [Google Scholar]
- 18. Man S, Xian Y, Holmes DN, Matsouaka RA, Saver JL, Smith EE, Bhatt DL, Schwamm LH, Fonarow GC. Target: stroke was associated with faster intravenous thrombolysis and improved one‐year outcomes for acute ischemic stroke in Medicare beneficiaries. Circ Cardiovasc Qual Outcomes. 2020;13:e007150. doi: 10.1161/circoutcomes.120.007150 [DOI] [PubMed] [Google Scholar]
- 19. Wadhwa R, Boehning AP. The joint commission. StatPearls. Treasure Island (FL): StatPearls Publishing. Copyright © 2023, StatPearls Publishing LLC; 2023. [PubMed] [Google Scholar]
- 20. Ganesh A, Luengo‐Fernandez R, Wharton RM, Gutnikov SA, Silver LE, Mehta Z, Rothwell PM. Time course of evolution of disability and cause‐specific mortality after ischemic stroke: implications for trial design. J Am Heart Assoc. 2017;6:e005788. doi: 10.1161/jaha.117.005788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Eriksson M, Norrving B, Terént A, Stegmayr B. Functional outcome 3 months after stroke predicts long‐term survival. Cerebrovasc Dis. 2008;25:423–429. doi: 10.1159/000121343 [DOI] [PubMed] [Google Scholar]
- 22. Rosen A, Wu J, Chang B‐H, Berlowitz D, Rakovski C, Ash A, Moskowitz M. Risk adjustment for measuring health outcomes: An application in VA long term care. Arlene S Ash. 2001;16:118–127. doi: 10.1177/106286060101600403 [DOI] [PubMed] [Google Scholar]
- 23. Solomon RA, Mayer SA, Tarmey JJ. Relationship between the volume of craniotomies for cerebral aneurysm performed at New York state hospitals and in‐hospital mortality. Stroke. 1996;27:13–17. doi: 10.1161/01.str.27.1.13 [DOI] [PubMed] [Google Scholar]
- 24. Hannan EL, Popp AJ, Tranmer B, Fuestel P, Waldman J, Shah D. Relationship between provider volume and mortality for carotid endarterectomies in New York state. Stroke. 1998;29:2292–2297. doi: 10.1161/01.str.29.11.2292 [DOI] [PubMed] [Google Scholar]
- 25. Waqas M, Tutino VM, Cappuzzo JM, Lazarov V, Popoola D, Patel TR, Levy BR, Monteiro A, Mokin M, Rai AT, et al. Stroke thrombectomy volume, rather than stroke center accreditation status of hospitals, is associated with mortality and discharge disposition. J Neurointerv Surg. 2023;15:209–213. doi: 10.1136/neurintsurg-2021-018079 [DOI] [PubMed] [Google Scholar]
- 26. Ormseth CH, Sheth KN, Saver JL, Fonarow GC, Schwamm LH. The American Heart Association's get with the guidelines (GWTG)‐stroke development and impact on stroke care. Stroke Vasc Neurol. 2017;2:94–105. doi: 10.1136/svn-2017-000092 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Caso V, Martins S, Mikulik R, Middleton S, Groppa S, Pandian JD, Thang NH, Danays T, van der Merwe J, Fischer T, et al. Six years of the angels initiative: aims, achievements, and future directions to improve stroke care worldwide. Int J Stroke. 2023;18:898–907. doi: 10.1177/17474930231180067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Song S, Fonarow GC, Olson DM, Liang L, Schulte PJ, Hernandez AF, Peterson ED, Reeves MJ, Smith EE, Schwamm LH, et al. Association of get with the Guidelines‐Stroke Program Participation and Clinical Outcomes for Medicare beneficiaries with ischemic stroke. Stroke. 2016;47:1294–1302. doi: 10.1161/strokeaha.115.011874 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Taylor N, Clay‐Williams R, Hogden E, Braithwaite J, Groene O. High performing hospitals: a qualitative systematic review of associated factors and practical strategies for improvement. BMC Health Serv Res. 2015;15:244. doi: 10.1186/s12913-015-0879-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Curry LA, Spatz E, Cherlin E, Thompson JW, Berg D, Ting HH, Decker C, Krumholz HM, Bradley EH. What distinguishes top‐performing hospitals in acute myocardial infarction mortality rates? A qualitative study. Ann Intern Med. 2011;154:384–390. doi: 10.7326/0003-4819-154-6-201103150-00003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Rose AJ, Petrakis BA, Callahan P, Mambourg S, Patel D, Hylek EM, Bokhour BG. Organizational characteristics of high‐ and low‐performing anticoagulation clinics in the veterans health administration. Health Serv Res. 2012;47:1541–1560. doi: 10.1111/j.1475-6773.2011.01377.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Gillum LA, Johnston SC. Influence of physician specialty on outcomes after acute ischemic stroke. J Hosp Med. 2008;3:184–192. doi: 10.1002/jhm.313 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Mitchell JB, Ballard DJ, Whisnant JP, Ammering CJ, Samsa GP, Matchar DB. What role do neurologists play in determining the costs and outcomes of stroke patients? Stroke. 1996;27:1937–1943. doi: 10.1161/01.str.27.11.1937 [DOI] [PubMed] [Google Scholar]
- 34. Moody KA, Maillie L, Dhamoon MS. National Patterns and outcomes of neurologist Care in Acute Ischemic Stroke. Neurohospitalist. 2023;13:13–21. doi: 10.1177/19418744221129428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Hammond G, Luke AA, Elson L, Towfighi A, Joynt Maddox KE. Urban‐rural inequities in acute stroke care and in‐hospital mortality. Stroke. 2020;51:2131–2138. doi: 10.1161/strokeaha.120.029318 [DOI] [PubMed] [Google Scholar]
- 36. Howard G. Rural‐urban differences in stroke risk. Prev Med. 2021;152:106661. doi: 10.1016/j.ypmed.2021.106661 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Leira EC, Hess DC, Torner JC, Adams HP Jr. Rural‐urban differences in acute stroke management practices: a modifiable disparity. Arch Neurol. 2008;65:887–891. doi: 10.1001/archneur.65.7.887 [DOI] [PubMed] [Google Scholar]
- 38. Adeoye O, Nyström KV, Yavagal DR, Luciano J, Nogueira RG, Zorowitz RD, Khalessi AA, Bushnell C, Barsan WG, Panagos P, et al. Recommendations for the establishment of stroke Systems of Care: a 2019 update: a policy statement from the American Stroke Association. Stroke. 2019;50:e187–e210. doi: 10.1161/str.0000000000000173 [DOI] [PubMed] [Google Scholar]
- 39. Elameer M, Price C, Flynn D, Rodgers H. The impact of acute stroke service centralisation: a time series evaluation. Future Healthc J. 2018;5:181–187. doi: 10.7861/futurehosp.5-3-181 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Hastrup S, Johnsen SP, Terkelsen T, Hundborg HH, von Weitzel‐Mudersbach P, Simonsen CZ, Hjort N, Møller AT, Harbo T, Poulsen MS, et al. Effects of centralizing acute stroke services: a prospective cohort study. Neurology. 2018;91:e236–e248. doi: 10.1212/wnl.0000000000005822 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Jauch EC, Schwamm LH, Panagos PD, Barbazzeni J, Dickson R, Dunne R, Foley J, Fraser JF, Lassers G, Martin‐Gill C, et al. Recommendations for regional stroke destination plans in rural, suburban, and urban communities from the prehospital stroke system of care consensus conference: a consensus statement from the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS physicians, National Association of state EMS officials, society of NeuroInterventional surgery, and Society of Vascular and Interventional Neurology: endorsed by the neurocritical care society. Stroke. 2021;52:e133–e152. doi: 10.1161/strokeaha.120.033228 [DOI] [PubMed] [Google Scholar]
- 42. Sculli GL, Pendley‐Louis R, Neily J, Anderson TM, Isaacks DB, Knowles R, Young‐Xu Y, Gunnar W. A high‐reliability organization framework for health care: a multiyear implementation strategy and associated outcomes. J Patient Saf. 2022;18:64–70. doi: 10.1097/PTS.0000000000000788 [DOI] [PubMed] [Google Scholar]
- 43. Gaw M, Rosinia F, Diller T. Quality and the health system: becoming a high reliability organization. Anesthesiol Clin. 2018;36:217–226. doi: 10.1016/j.anclin.2018.01.010 [DOI] [PubMed] [Google Scholar]
- 44. Cochrane BS, Hagins M Jr, Picciano G, King JA, Marshall DA, Nelson B, Deao C. High reliability in healthcare: creating the culture and mindset for patient safety. Healthc Manage Forum. 2017;30:61–68. doi: 10.1177/0840470416689314 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S9
Figures S1–S6
Data Availability Statement
In compliance with the Act on the Protection of Personal Information Maintained by Public Institutions, anonymized data were accessible only to authorized researchers and scheduled for deletion following the conclusion of the study, thus limiting data availability.