Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Jonathan Rosand

Neurocritical Care · Broad Institute

View

Dr. Jinsy Andrews

ALS · NYU Langone

View

Dr. Maisha Robinson

Hospice & Palliative Care · Mayo Clinic Florida

View

Dr. Michael Kentris

Epilepsy · St. Elizabeth Medical Center

View

Dr. Daniel Correa

Epilepsy · AECOM

View

Dr. Aaron Berkowitz

General Neurology · UCSF

View

Dr. Jose Hernan Posas

Sports Neurology · UVA

View

Dr. Michael Okun

Movement Disorders · UF

View

Dr. Joseph Sirven

Epilepsy · Mayo Clinic College of Medicine

View

Dr. Rachel Marie Salas

Sleep Medicine · Johns Hopkins

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Association of Component Strategies of the Target Stroke Phase 3 Nationwide Quality Improvement Program With Accelerated Door-to-Puncture and Door-In-Door-Out Times for Ischemic Stroke Endovascular Thrombectomy in the United States.

    Oct 20, 2025DOI: 10.1161/circoutcomes.125.012456Journal ArticleAHA

    Written by: Brian Mac GroryKaiz S AsifHaolin XuBrooke AlhantiJay B LuskDavid HasanSoojin ParkAmelia K BoehmeKori S ZachrisonMayank GoyalAndrew M SoutherlandAshutosh JadhavSantiago Ortega GutierrezAmeer HassanKyle FargenKevin N ShethEdward C JauchYing XianEric D PetersonEric E SmithSteven R MesseLee H SchwammPeter PanagosCharles WiraJeffrey L SaverGregg C Fonarow

    Find these authors on Roon

    AJ

    Dr. Ashutosh Jadhav

    Interventional Neurology · Barrow Neurological Institute

    JS

    Dr. Jeffrey Saver

    Vascular Neurology · RRUCLA

    INTRODUCTION

    The benefit of acute stroke treatment is time dependent.1, 2 With respect to endovascular thrombectomy (EVT),3 a major determinant of treatment efficacy is the time elapsed between stroke onset and reperfusion through the occluded blood vessel. Improving stroke onset to treatment time is a major goal of stroke centers and regional systems of stroke care and can be accomplished by improving pre-hospital workflows, emergency department (ED) triage and care delivery protocols, the efficiency of stroke protocol execution, and the rapidity of inter-facility transfer.The American Heart Association (AHA) has implemented a series of nationwide quality improvement programs to promote optimal stroke treatment – the Target Stroke (TS) initiative. TS Phase I,4 launched in 2010, and TS Phase II,5 launched in 2014, targeted rapidity of intravenous thrombolysis delivery and were associated with substantial improvements at participating hospitals.5–9 TS Phase III, launched in 2017, for the first time added as targets two care processes related to improved timeliness of EVT: door-to-first pass with a thrombectomy device (DTFP) times at thrombectomy-capable hospitals and door-in/door-out (DIDO) times at non-thrombectomy-capable hospitals transferring patients to thrombectomy-capable hospitals.To support the achievement of benchmarks for DTFP and DIDO, TS Phase III disseminated a suite of strategies including pre-hospital and ED strategies to optimize the speed of EVT delivery. The strategies focused on door-to-arterial puncture [DTP] time (the modifiable component of the primary metric in Target Stroke Phase III – DTFP) and efficiency of inter-facility transfer initiation (DIDO time at the transferring facility). The objective of the present study is assess the strength of the association of each of these TS Phase III strategies with faster DTP and DIDO times.

    METHODS

    An observational cohort study was performed using prospectively collected data from the American Heart Association’s (AHA) GWTG-Stroke registry in the United States (US) between January 2017 and March 2022. We included institutions that participated in the GWTG-Stroke registry and in the AHA’s Target Stroke Phase III initiative. Given this was a study of anonymized patient data, the Institutional Review Board of the Duke University School of Medicine, Durham, North Carolina, USA waived the requirement for informed consent. As data were collected for quality improvement purposes, rather than primarily for research, data sharing agreements require an application process for other researchers to access the data.
    1) The Get With The Guidelines-Stroke (GWTG-Stroke) program encompasses a nationwide registry and quality improvement initiative under the stewardship of the AHA. Details of registry organization have been previously published in-depth.10 Briefly, trained site personnel collect data (by means of a web-based patient management tool) on consecutive patients with suspected stroke admitted to each participating hospital. Data elements captured include patient demographics, medical history, medication use, diagnostic testing, acute stroke management, and hospital complications. There is a high degree of concordance between data within the registry and medical record review.11 Each participating site receives either A) human subjects research approval to enroll patients with a waiver of written, informed consent under the Common Rule (45 CFR §46) or B) a waiver of authorization and exemption from subsequent review by their institutional review board. IQVIA (Parsippany, New Jersey) serves as the data collection and coordination center. The Duke Clinical Research Institute (DCRI; Durham, NC) served as the data analysis center for this analysis and has an agreement to analyze the aggregate deidentified data for research purposes.
    2) AHA Target Stroke Phase III was a nationwide initiative that set the first stringent benchmarks for key quality metrics pertaining to endovascular reperfusion therapy. Target Stroke Phase III disseminated 12 classes of strategy to thrombectomy-capable hospitals to promote faster DTFP times and 9 classes of strategy to non-thrombectomy hospitals to promote faster DIDO times in those patients undergoing inter-facility transfer. These strategies are outlined in the Supplementary Material.
    The primary benchmark set by Target Stroke Phase III was a door-to-device (i.e. ED arrival to first deployment of a thrombectomy device) time of ≤90 minutes for patients presenting directly to a thrombectomy-capable hospital or ≤60 minutes for patients who were transferred in to a thrombectomy-capable hospital in ≥50% cases. As DTP time is ascertained with greater consistency within the GWTG-Stroke database, is strongly collinear with door-to-first pass time, and was the actual focus of the preponderance of Target Stroke Phase III-recommended strategies, the DTP time metric was used in the present study as a surrogate for door to first deployment of a thrombectomy device. In addition, DTP values were converted to estimates of door-to-first pass time values by the addition of the median interval from puncture for first pass among patients with documented times for both procedures.
    For DIDO times, the Target Stroke Phase III initiative recommended as a strategy establishing a target time of tracking tools and strategies but not specific target metrics. However, target metrics have been established by other consensus groups and have ranged from within 120 to within 90 to within 60 minutes.12–15
    Patients with ischemic stroke who presented to a participating GWTG-Stroke site that self-identified as a Target Stroke Phase III site and who received EVT or IVT were included. We excluded patients who had a stroke while in-hospital, with symptoms that had resolved on hospital presentation, with an onset-to-arrival time that was missing or documented as greater than 24 hours from time last known well, and patients who received EVT as part of an investigational protocol. A cohort was derived after applying these criteria and then 3 overlapping cohorts were derived from this larger cohort:
    We excluded those with DTP time that was missing or documented as ≥12 hours (to exclude patients being treated for post-arrival worsening than for their presenting deficit). We also excluded patients with DTN time of ≥4.5 hours in patients who received “bridging” IV thrombolysis (i.e. IV thrombolysis prior to EVT), and patients presenting to sites with <5 total EVT procedures performed during the study period (as these sites were felt to be non-representative of practice patterns generally).
    We included patients who underwent EVT at a GWTG-Stroke and TSIII-participating hospital and who had been transferred in from another acute care hospital. We implemented the same additional exclusion criteria as for Cohort 1.
    We included patients who presented to a GWTG-Stroke and TSIII-participating hospital, received IVT, had an NIHSS of ≥10, and who were transferred out to another acute care hospital. The rationale for choosing an NIHSS of 10 was to enrich for a population that were potentially EVT eligible because the odds of having a large vessel occlusion increase in proportion to stroke severity. We did not require a documented large vessel occlusion because of limitations in how this is ascertained within the GWTG-Stroke database. We excluded patients with missing DIDO time and those whose record indicated an EVT procedure was performed prior to transfer. Because GWTG-Stroke is organized on the encounter level (as opposed to patient-level), it is conceivable that some patients in this cohort were also represented in Cohort 2.
    The exposures within this study were responses to individual questions on a survey distributed to all participating programs at the time of their entry into the Target Stroke Phase III program between June and August 2022. Representatives from participating sites were invited to complete this survey online regarding which of the strategies with regards to EVT and inter-facility transfer their hospital had been deploying in the 6 months before the survey. These questions referenced emergency medical services (EMS) strategies, telestroke strategies, and pre-hospital workflow strategies, all with the objective of improving hospital performance with respect to door-to-first pass time and DIDO time. The survey was structured to invite binary (yes/no) responses or responses on a continuous scale (i.e. the percentage of time that a strategy was used [ranging from 0% to 100%]). In addition to strategies, data on performance measures were collected from each survey respondent (as continuous values and as proportion of patients with DTP ≤ 30, ≤60, and ≤90 minutes) and similarly used as independent variables within models. AHA staff were available to all sites to address questions related to the completion of survey questions. The survey is included in the Supplementary eMaterial.
    The primary end point for analysis of EVT sites was DTP time. This was constructed as a continuous variable (denominated in minutes) and computed as the interval between arrival time at the ED of the treating hospital and arterial puncture time. The primary end point for non-thrombectomy hospitals was DIDO time, computed as the interval between the arrival time at the ED of the transferring hospital and the departure time at the ED of the transferring hospital.
    This analysis was performed in accordance with a pre-specified statistical analysis plan. Hospital characteristics were described for hospitals that participated in both GWTG-Stroke and Target Stroke Phase III (defined as returning a survey corresponding to Target Stroke Phase III). Patient-level characteristics were described including key demographic, clinical, and treatment variables (Supplementary Material). Patients were separated in to three separate cohorts: the “Directly-Arriving EVT” cohort, the “Transfer-In EVT” cohort, and the “Transfer Out” cohort. Counts and percentages for categorical variables and medians (with first [Q1] and third [Q3] quartiles) for continuous variables were presented. Missing rates of key variables were tabulated.
    Survey responses to Target Stroke Phase III were collated and presented in tabular form. These were presented at the site level and, as above, included strategies pertaining to ED workflow, telestroke strategies, and pre-hospital strategies. Responses were described as counts (with percentages) for survey questions that invited a categorical response and medians (with Q1 and Q3) for survey questions that invited a continuous response (i.e. percentage of time that a strategy was observed, ranging from 0% to 100%).
    To evaluate the association between Target Stroke Phase III strategies and DTP or DIDO times, a series of generalized linear models were fit. For each end point, a separate linear regression model was constructed with the end point as the dependent variable and the survey response as the independent variable. Generalized estimating equations were used to account for clustering of patients within each site. First, unadjusted estimates (with their corresponding 95% confidence interval [CI]) for each strategy were calculated. Then, adjusted estimates (with their corresponding 95% CI) were calculated from a multiple linear regression model with covariates including demographic, clinical, and site-level variables (full details of all covariates are included in eMethods). Race and ethnicity were included as model covariates because of known disparities in stroke outcomes across these categories.16
    A status of “missing” for variables pertaining to past medical history and anti-thrombotic exposure was imputed to “No”. Those who were ≥65 and with missing insurance status were assumed to be Medicare beneficiaries. For other model covariates, missing data were imputed using multiple imputation via chained equations. End points and hospital characteristics were not imputed.
    A two-sided alpha of 0.05 was used to connote significance for all statistical tests. The widths of the 95% confidence intervals were not adjusted for multiplicity. Analyses were performed using SAS® software (version 9.4., SAS Institute, Cary, North Carolina). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)17 checklist is included in the Supplementary eMaterial to this manuscript.

    RESULTS

    Study entry criteria were met by 37,373 directly-arriving patients receiving EVT (median age 72 years [Q1, Q3: 61, 82]; 50.0% women) at 450 centers, 28,665 transfer-in patients receiving EVT (median age 71 [Q1, Q3: 60, 80]; 49.9% women) at 395 centers, and 12,703 transfer-out patients who were potentially eligible for EVT (median age 71 years [Q1, Q3: 60, 82]; 50.8% women) at 1,107 centers. Figure 1 depicts the application of eligibility criteria for this study. Baseline demographic and clinical characteristics are presented in Table 1 and site-level characteristics are presented in Table S1.
    Figure 2 shows the distribution of DTP and DIDO time at participating hospitals on the patient level and the hospital level. The median site-level DTP time at EVT sites was 106.5 minutes [Q1, Q3: 93, 127] for direct-arriving patients and 53 minutes [Q1, Q3: 36, 75] for transfer-in patients. The median puncture-to-first pass interval was 20 minutes [Q1, Q3: 13, 30] for direct-arriving EVT patients and 19 minutes [Q1, Q3: 13, 29] for transfer-in EVT patients, Adding these median values to the patient-level DTP times (Figure 2) yields projected door-to-first pass times of median 122 minutes ([Q1, Q3: 90, 165] for direct-arriving patients and 71 minutes [Q1, Q3: 44, 113] for interfacility transfer patients. On the hospital-level, the median percentage of directly-arriving patients achieving DTP ≤90 minutes was 31% (Q1, Q3: 15, 47) and the median percentage achieving DTP ≤60 minutes was 6% (Q1, Q3: 1, 14). On the hospital-level, the median percentage of transfer-in patients achieving DTP ≤90 minutes was 81% (Q1, Q3: 65, 92) and the median percentage achieving DTP ≤60 minutes was 59% (Q1, Q3: 27, 77). The median site-level DIDO was 127.5 minutes [Q1, Q3: 105, 158] for potentially-EVT eligible transfer-out patients. Site-level metrics are outlined in Table S1.
    In total, 2,455 sites were invited to complete Target Stroke Phase III survey of whom 1,469 (59.8%) completed a survey. Key characteristics are compared between Target Stroke Phase III participating and non-participating sites in Table S2. On average, sites that returned a survey were larger than sites that did not return a survey (median size 219 beds [Q1, Q3: 119, 365] vs. 164 beds [Q1, Q3: 80, 300]), less likely to be located in a rural area (17.5% vs. 24.9%), and more likely to be a comprehensive stroke center (10.9% vs. 0.0%). The Joint Commission performed the majority of certifications (65.3%), followed by Det Norske Veritas (16.8%). Of all potentially-eligible sites, 1,305 had patients that met criteria for inclusion in the present study. Table 2 outlines TS III survey responses for the 1,305 sites.
    The frequency of use of each of the Target Stroke Phase III strategies is shown in Table 2. In total, 44.2% of sites endorsed the use of a prehospital stroke screening tool 76–99% of the time while 18.2% of sites endorsed the use of a stroke screening tool “all of the time”. By contrast, the use of a prehospital stroke severity scale, helpful for LVO identification, was endorsed 76–99% of the time by 20.5% of hospitals and “all of the time” by 5.6% of hospitals. Overall, 1,259/1,305 (96.5%) sites had 24/7 neurology coverage (either in-person or via telemedicine) for stroke alerts while 978/1,305 (74.9%) were involved in a telestroke program in some manner. Strategies used as a component of acute stroke care included “stroke tools” (including stroke-specific order sets or hospital-specific algorithms; 92.7% of sites), IV thrombolytic agents stored in the ED (79.5% of sites), a protocol for rapid imaging performance and interpretation (93.2% of sites), and CT and/or MRI scanners physically located in the ED (46.7% of sites).
    Figure 3 displays the association between use of individual strategies and DTP time (and the same strategies are outlined in tabular form in Table S3). A strategy of alerting the neurointerventional team based on emergency medical services (EMS) pre-notification of LVO stroke was associated with a significantly shorter adjusted DTP time (−21.9 minutes [95% CI: −42.5 to −1.3 minutes]). In addition, performance of a brain CT and CTA in all patients presenting ≤24 hours from time last known well was associated with significantly shorter DTP time (−6.6 minutes [95% CI: −11.8 to −1.5 minutes]). No change in ER workflow was associated with a shorter adjusted DTP time (−8.4 minutes [95% CI: 13.9 to −2.9). By contrast, adding a CTA or CTP that would otherwise would not have been performed concurrently to a non-contrast CT was associated with longer DTP time (7.1 minutes [95% CI: 1.1 to 14.4 minutes]) as was an EMS policy of bypass/re-routing was associated with a longer DTP time of 8.3 minutes (95% CI: 0.3 to 16.3).
    Figure 4 displays the association between use of individual strategies and DTP time (and the same strategies are outlined in tabular form in Table S4). A strategy of increased use of stroke screening tools was associated with a significantly shorter DTP time (–3.5 minutes [95% CI: −6.4 to −0.6 minutes] per 25% increase in use of the screening tool). Additionally, increased use of a camera during telestroke consultations was associated with a significantly shorter DTP time (−5.8 minutes [95% CI: −10.7 to −0.9 minutes] per 25% increase in camera use).
    Figure 5 displays the relationship between use of individual strategies and DIDO time. In total, 8 Target Stroke Phase III strategies were independently associated with shorter DIDO time. One was related to emergency medical services (EMS) interactions: patient-specific data feedback to EMS providers (−2.5 minutes [95% CI: −4.2 to −0.8 minutes] per 25% increase). Two were related to imaging: i) simultaneous performance of vascular imaging (CTA or MRA) on all acute stroke patients within 24 hours from last known well, regardless of the severity of examination findings (−5.7 minutes [95% CI: −10.1 to −1.2 minutes]) and ii) and use of automated neuroimaging software (−7.7 minutes [95% CI:−12.2 to −3.1]). One was related to IVT care processes: IVT commenced in the imaging suite (−2.8 minutes [95% CI: −5.1 to −0.5 minutes] per 25% increase in proportion of time this strategy was used). Two were related to telestroke processes: i) involvement in telestroke as a hub site (i.e. the hospital that is initiating the transfer is itself a hub within a telestroke network; −14.5 minutes [95% CI: −25.2 to −3.9]) and ii) telestroke providers based at a hub facility within the same health system where most patients are transferred for intervention (−11.4 minutes [95% CI: 16.2 to −6.6]). Two were related to speed of IVT delivery: i) achievement of a door-to-needle time of ≤60 minutes (−3.8 minutes [95% CI: −6.5 to −1.1 minutes] per 25% increase in proportion of time this is achieved) and ii) achievement of a door-to-needle time of ≤45 minutes (−2.4 minutes [95% CI: −4.4 to −0.4]). These strategies are outlined in Table S5.

    DISCUSSION

    In this nationwide survey of GWTG-Stroke hospitals joining the Target Stroke Phase III program in the US, the frequency of use of individual recommended strategies designed to reduce DTP time and DIDO time ranged widely, from 0% to 100%. Among 37,373 patients receiving EVT in the surveyed sites, the median site-level DTP time was 106.5 minutes [Q1, Q3: 93, 127] for direct-arriving patients. This indicates that, on average, patients undergoing EVT are outside of the Target Stroke Phase III benchmark of ≤90 minutes in directly-arriving patients. The median site-level DTP time in transferred patients was 53 minutes [Q1, Q3: 36, 75], indicating that, on average, patients transferred for EVT are outside the Target Stroke Phase III benchmark of DTFP ≤60 minutes for transfer-in patients (estimated as 71 minutes [Q1, Q3: 44, 113] in the present study).
    For optimizing DTP time among direct-arriving patients, Target Stroke Phase III strategies associated with faster performance included alerting the neurointerventional team based on EMS pre-notification of a possible LVO, and performing a non-contrast brain CT and CTA on all patients presenting within 24 hours of time last known well. Some successful strategies are endorsed in part in the most recent AHA/ASA Guidelines for the Early Management of Patients with Acute Ischemic Stroke.18 These include use of a stroke assessment tool by EMS (certainty of recommendation and level of evidence (LOE): B-NR), and pre-hospital notification (COR: 1, LOE: B-NR). A strategy of performing a brain CTA/CTP selectively in only some patients alongside the initial non-contrast brain CT was associated with longer DTP time of, on average, 7.7 minutes. This may reflect that sites that perform CTA or CTP selectively are less efficient in imaging conduct than sites that perform them in all patients. In contrast, a response of “no change in ER workflow” as a result of the Target Stroke Phase III program signified a lower adjusted DTP (−8.4 minutes [95% CI: −13.9 to −2.9]). It is plausible that such a response indicates a site that had a well-performing ED workflow prior to the implementation of the program, that no changes were deemed necessary, and that underlying difference in patients arriving to these sites were not fully captured in measured covariates.
    For optimizing DTP time among transfer-in patients, Target Stroke Phase III strategies associated with faster performance included the use of stroke screening tools and increased use of a camera during telestroke consultations. It is counter intuitive that the use of stroke screening tool was associated with shorter DTP times in transfer-in patients (−3.5 minutes [95% CI: −6.4 to −0.6]) because stroke must have already been diagnosed at the transferring site prior to transfer. It is likely this finding is a sign of generally better stroke systems at those hospitals. The use of a camera during telestroke consultations may facilitate a more accurate or rapid assessment by the remote neurologist. Alternatively, it may act as an indicator for a site that is more well resourced and can therefore deploy more advance equipment. There were no other reported strategies that were associated with faster DTP times. This may be because DTP times were strong among transfer-in EVT patients – the median site-level DTP was 53 minutes compared with 106.5 minutes in directly-arriving EVT patients – and thus the scope for average improvement is more limited (i.e. a greater magnitude of association corresponding to each strategy would have to be substantially greater to yield an improvement).
    For non-thrombectomy centers, multiple care processes were associated with faster DIDO time indicating robust opportunities for sites to optimize their workflow. The care processes were components of several different segments of care. Providing feedback to EMS providers on case course and outcome indicates the benefits of close communication between the receiving stroke center and the prehospital system. Routine use of more advanced imaging, CTA and automated imaging processing software permits faster identification of patients with LVO who are EVT-eligible. Faster delivery of intravenous thrombolysis, including by start of drug in the imaging suite, enables faster readiness for transfer. Telestroke provider expertise (reflected in being a provider to other hospitals) and greater system-specific knowledge (from working at a hub within the same health system as the spoke) would improves speed of telestroke decision-making. In the present study, while only 5% of transferring sites were also telestroke hubs, this designation was associated with shorter DIDO times, an intuitive observation because a site with sufficient expertise to act as a telestroke hub would be expected to similarly perform well in other areas such as having an infrastructure to facilitate rapid transfers. This comports with pilot work suggesting that telestroke care delivered by an external vendor tends to be associated with slower treatment times than that provided by teams embedded within health systems.19 This finding is also broadly concordant with AHA/ASA guideline recommendations that the use of telemedicine should be employed to provide 24/7 coverage (COR 1, LOE C-EO), that telestroke care can be effective for IV thrombolysis decision-making (COR 2A, LOE B-R), and that telestroke may be reasonable to facilitate rapid inter-facility transfer (COR 2B, B-NR).
    An important finding from this study is that a majority of interventions associated with faster treatment time are likely to be cost neutral. For instance, the administration of IV thrombolysis while a patient is in the imaging suite – while certainly requiring collaboration and participation from radiology staff – does not require any added equipment or personnel beyond that present in a conventional ED (although it could hamper overall ED workflow by slowing down CT throughput). Similarly, alerting the neurointerventional team based on EMS pre-notification does not directly add cost. However, the risk of “false alarms” could place a strain on neurointerventional staff. The observation that pre-notification is associated with shorter DTP time is in keeping with a rich literature suggesting that EMS pre-notification improves subsequent ED workflows20–23 and that improved pre-hospital stroke recognition is associated with improved stroke care.24–29
    Several observations from the present study warrant further consideration. The finding that – based on EMS pre-notification – ordering a CTA/CTP (that otherwise would not have been done concurrently) along with a non-contrast brain CT was associated with longer DTP time raises concerns about selective rather than uniform use of advanced imaging. The finding that when EMS agencies follow a bypass/re-route protocol, DTP times are longer at the surveyed hospitals was somewhat counterintuitive. However, this could reflect that the resulting higher volume of EVT cases could occasionally result in delays when more patients present concurrently than the site can rapidly handle. The utility of re-route/bypass protocols in the treatment of potentially-EVT eligible patients is not known and remains an area of active inquiry.
    This study has inherent limitations. First, our principal exposures in this study are survey responses, and not measurement of adherence with Target Stroke Phase III strategies in individual patients. Additionally, survey responses are vulnerable to social desirability bias and recall bias. Second, findings are associative in nature, and do not establish that a strategy itself is causative of a change in DTP time or DIDO time. It may be that the highest performing hospitals are more likely to observe Target Stroke Phase III strategies (a form of “inverse causation”). This may be exemplified by the observation that increased use of cameras during telestroke consultations is associated with reduce DTP time in transfer-in EVT patients – as the use of newer technology may be a proxy for a more well resourced, higher-performing hospital. Establishing causation would necessitate a prospective, pragmatic clinical trial that employs randomization (either on the patient or hospital level). Such pragmatic trials may even be conducted using an electronically-facilitated performance dashboard with iterative improvement plans tailored to each site.30 Third, it is not known if the findings may be impacted by the increasing use of tenecteplase rather than alteplase, which may interact in important ways with stroke systems of care, especially those pertaining to inter-hospital transfer and emergency drug preparation. Fourth, the interplay of race, ethnicity, socioeconomic status or the intersection of these factors with Target Stroke Phase III strategies is not known. Fifth, it is likely that such strategies may differ in their effectiveness depending on whether a patient presents during “on hours” (during business hours, Monday to Friday) or “off hours” (overnight, weekends, or federal holidays) and a future analysis will be needed to explore this. Finally, approximately 40% of sites did not respond to the survey and responding sites were generally larger, more likely to be comprehensive stroke centers, and more likely to be in an urban location. Thus, the findings may not be generalizable outside of this setting.

    CONCLUSION

    Among 1,305 hospitals participating in Target Stroke Phase III, there was moderate uptake of strategies designed to reduce DTP and DIDO time. Many of these strategies were independently associated with more rapid stroke care; however, opportunities for further implementation and quality improvement remain.
    AJDr. Ashutosh Jadhav

    Dr. Ashutosh Jadhav

    Neurology · Interventional Neurology
    Barrow Neurological Institute
    ashutosh