Abstract
Background: Despite decades of educational efforts, patients with acute ischemic stroke (AIS) remain delayed in seeking medical care, which becomes the greatest obstacle to the successful management of the condition. Objective: The objective of this study was to systematically explore the incidence and influencing factors of prehospital care-seeking delay in AIS patients. Methods: We systematically searched the PubMed, Embase, Cochrane Library, Web of Science, and Cumulative Index to Nursing and Allied Health Literature from database inception to September 30, 2023. Meta-analysis was conducted using the Stata 15.0 software package. The pooled incidence was calculated using a random-effects model. The quality of studies reporting incidence data was assessed using Joanna Briggs Institute’s Critical Appraisal Checklist and Newcastle-Ottawa Scale. Subgroup analyses were performed according to study location, country income, recruitment date, and age. Results: Finally, 30 related articles were included, involving a total of 287,102 people. The estimated incidence of prehospital care-seeking delay was 68%, and there were differences in this incidence in different countries (p = 0.035). Meta-analysis results showed that the delay rate was highest in low-income countries (85%) and lowest in high-income countries (62%). Patients who live farther from hospitals, have a lower level of education, diabetes, hyperlipidemia, or a history of stroke are more likely to experience delays (all p < 0.05). Conversely, those who can recognize stroke symptom, perceive the severity of early symptom, understand thrombolysis treatment, atrial fibrillation, consciousness disturbance, visual disturbance, and symptom score at admission, emergency medical service use, and immediate help-seeking have a lower risk of delay (all p < 0.05). Conclusion: Prehospital care-seeking delays are common among patients with AIS, especially in low-income countries. To reduce delays, it is crucial to increase public awareness of stroke symptoms, improve education levels, and optimize healthcare accessibility.
Introduction
Stroke is the second leading cause of death worldwide. According to statistics from the Lancet Neurology Commission, the mortality of stroke is projected to increase by 50%, from 6.6 million (95% uncertainty interval [UI] 6.0 million–7.1 million) in 2020 to 9.7 million (8.0 million–11.6 million) in 2050 [1], imposing a substantial burden on patients, families, healthcare systems, and the broader economy. Acute ischemic stroke (AIS) accounts for more than 85% of all strokes. The key to treating AIS is to quickly restore blood supply to brain tissue [2]. According to clinical guidelines, administering recombinant tissue-type plasminogen activator within 3.0–4.5 h is the most effective treatment for AIS, so as to save the brain tissue of patients from infarction and improve the prognosis and outcome. However, this treatment method is limited by a strict time window [3]. Therefore, effectively shortening the reperfusion time is critical to improve the therapeutic effect and prognosis of AIS patients.
The treatment of AIS can be divided into three stages: prehospital, emergency, and hospitalization. The American Heart Association and the American Stroke Association divided the AIS treatment process into an “8D” survival chain: detection, dispatch, delivery, door, data, decision, drug, and disposition, among which detection, dispatch, and delivery belong to prehospital stages, namely, the time from symptom appearance to hospital arrival (onset to door time) [4]. A study [5] has confirmed that prehospital care-seeking delay is an independent risk factor for the delay in AIS treatment, and any delay in these steps will lead to AIS patients missing the golden treatment time. In recent years, with the generation of concept “time is brain” and the in-depth study on the ultra-early treatment for ischemic cerebrovascular diseases, it is particularly important to shorten the prehospital care-seeking delay in AIS treatment.
Several studies have identified factors contributing to prehospital delays in AIS, including sociodemographic characteristics, initial symptoms (e.g., hemiplegia, headache, and blurred vision), and, more recently, awareness and perception of stroke symptoms [6‒9]. However, several gaps remain in the understanding of these issues. Inconsistencies include findings on sex differences (e.g., Fukuda et al. [10] found male patients more prone to delay, while Hassan et al. [9] reported the opposite) and prior medical history (e.g., Zachrison et al. [11] found that patients with a history of stroke were less prone to delay, while Yuan et al. [12] did not). These conflicting results highlight the necessity for a meta-analysis to better understand the strength of the association between these factors and prehospital delays.
Additionally, cultural, regional, and systemic healthcare disparities can impact prehospital care-seeking behaviors, as evidenced by the variability in delay rates across countries – 42% in the USA, 69% in Iran, and 86% in Indonesia [13‒15]. Despite numerous regional studies, a globally representative estimate of the incidence of prehospital delays remains unavailable, limiting our ability to fully address this public health challenge.
Given these gaps in the literature, our study aimed to conduct a comprehensive meta-analysis with the following objectives (1) to estimate the global incidence of prehospital delay in AIS patients, (2) to provide a summary effect estimate to quantify the association between sociodemographic, clinical, psychological factors and prehospital delays, and (3) to establish an evidence-based foundation for designing targeted interventions aimed at reducing prehospital delays in AIS treatment.
Methods
Retrieval Strategy
From the inception of each database to September 30, 2023, we searched for relevant English-language studies in PubMed, Embase, Cochrane Library, Web of Science, and Cumulative Index to Nursing and Allied Health Literature (CINAHL). These databases were selected for their extensive coverage of international literature and their capability to support the search for interdisciplinary studies. PubMed and Embase provide a wide range of research resources in the medical and life sciences; the Cochrane Library focuses on systematic evaluations and evidence-based research; Web of Science aggregates the diverse natural science, social science, and humanities literature; and CINAHL focuses on nursing and health-related disciplines. This choice ensures comprehensive access to high-quality research relevant to our topic. The retrieval was carried out by combining medical subject headings and free term, and the medical subject headings were as follows: ischemic stroke, stroke, risk factor, and observational study. The specific retrieval strategy is shown in online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000542765). Meanwhile, the references of studies included in this analysis were additionally retrieved to ensure that the retried studies were as comprehensive as possible.
Literature Screening
Inclusion criteria: (1) population: AIS patients aged ≥18 years; (2) study design: observational research; (3) outcomes: the incidence or the influencing factors of prehospital care-seeking delay was reported; (4) studies provided original data that can be used to calculate the incidence or correlation coefficient, or reported the relative risk, odds ratio (OR), and its 95% confidence interval (CIs) for measurement of the correlation strength, or reported the estimated values and standard errors of logistic regression coefficients for data conversion; (5) the years of research or publication was provided, with a clear definition of the sample size; (6) studies clearly and similarly defined prehospital care-seeking delay.
Exclusion criteria: studies were excluded if they were (1) reviews, case reports, study protocols, or conference papers; (2) clinical trials, animal, or in vitro studies; (3) repeated and inaccessible documents; (4) literature that containing outcome indicators which were unable to be extracted. Two reviewers (Y.Z. and Y.X.) independently screened the literature according to the above criteria, and any disputes during the screening process were resolved through discussion or by a third reviewer (S.F.).
Data Extraction and Quality Assessment
Two reviewers (S.Z. and D.M.) independently extracted the data, including the first author, publication year, country, study design, sample size, sex, age, cutoff of prehospital care-seeking delay, and factors of analysis.
The quality of the cross-sectional studies was evaluated using JBI Critical Appraisal Tools by two independent reviewers (D.S. and J.S.) [18]. This tool contains seven criteria: inclusion criteria of samples, description of the theme and environment, effective and reliable exposure measurement, objective and standard condition measurement, identification of confounding factors and response strategies, effective and reliable result measurement, and appropriate statistical analysis. All items have four corresponding options: yes (1 point), no (0 point), unclear (0 point), or not applicable (0 point). According to these items, the articles were classified into high quality (80% and above), medium quality (60–80%), and low quality (<60%). Newcastle-Ottawa Scale (NOS) [19] was used to evaluate the cohort studies or case-control studies, mainly from three aspects including subject selection, comparability, and exposure/outcome.
Data Synthesis and Statistical Analysis
The meta-analysis was conducted using Stata 15.0. OR was used as an effect indicator to estimate the correlation between influencing factors and prehospital care-seeking delay. 95% CI was calculated along with OR. Heterogeneity between included studies was evaluated using Q, χ2, and I2 statistics. The fixed-effects model (Mantel-Haenszel method) was used for meta-analysis if there was no significant heterogeneity between studies (I2 < 50% and p > 0.1). Otherwise, a random-effects model (DerSimonian-Laird method) was used. All statistical tests were two-tailed tests, with p = 0.05 indicating a significance level.
Subgroup analysis and regression analysis were carried out based on study location, country income group, recruitment date, and age to clarify the level and source of heterogeneity among the studies. Sensitivity analysis (required for ≥ 5 studies) was used to evaluate the robustness of the results of meta-analysis, funnel plots were created to assess the presence of publication bias in the included literature, and publication bias was statistically examined using Egger’s or Begg’s test (required for ≥ 5 studies). For results with significant publication bias, the trim-and-fill method was adopted to assess the impact of publication bias on the results.
Results
Results and Flowchart of Literature Screening
A total of 33,250 papers were initially retrieved from the databases, and no additional studies were identified from reference screening. After removing duplicates, 22,532 articles were examined by title and abstract. Among these, 22,489 articles were excluded for not meeting the inclusion criteria, and 3 articles with unavailable full texts were excluded. Then, the full texts of 40 articles were reviewed. Ten studies were excluded after a thorough review of the full text for the following reasons: four studies had populations that did not consist of individuals with AIS, one study was published in a non-English language, four studies were focused on topics that were irrelevant to the research focus, and one study involved participants who were not adults. Finally, 30 studies were included in this meta-analysis [6‒15, 20‒39]. The literature screening process is shown in Figure 1.
Basic Characteristics of Included Studies
The included studies (n = 30) were conducted in 21 countries in six continents, including China (n = 3), USA (n = 3), UK (n = 2), Denmark (n = 1), Egypt (n = 1), Indonesia (n = 1), Iran (n = 1), Ireland (n = 1), Japan (n = 1), Korea (n = 1), Morocco (n = 1), Nepal (n = 1), Peru (n = 1), Saudi Arabia (n = 1), Senegal (n = 1), Singapore (n = 1), Somalia (n = 1), South Africa (n = 1), Spain (n = 1), Thailand (n = 1), and Turkey (n = 1). These studies were published between 1992 and 2023, and the sample size ranged from 50 to 118,683. There were 287,102 participants (154,807 males and 132,295 females), with an average age of 48.2–77.4 years. The prehospital care-seeking delay is defined as the time from symptom onset to hospital arrival ≥2 h (n = 3), ≥3 h (n = 14), ≥3.5 h (n = 4), ≥4 h (n = 2), and ≥4.5 h (n = 7). The basic features of the included studies are displayed in Table 1.
Characteristics of the included studies
No. . | First author . | Year . | Country . | Study design . | Sample size . | Sex (male/female) . | Age . | Cut-off of prehospital delay . | Factors of analysis . | Adjustment crude or adjustment . | Confounders . |
---|---|---|---|---|---|---|---|---|---|---|---|
1 | Kim et al. [40] | 2011 | Korea | Prospective | 500 | 310/190 | 67 | 3 h | NIHSS score, Previous stroke, diabetes, hyperlipidemia, atrial fibrillation, coronary heart disease, living alone, arrival through referral, EMS use, knowledge by patient of thrombolysis, awareness of the patient/bystander that the initial symptom was stroke-related | Adjustment | NM |
2 | Ghadimi et al. [14] | 2021 | Iran | Prospective cohort | 204 | 114/90 | 68.99±13.91 | 4.5 h | Consultation after the onset of symptoms, EMS use, patient’s perception of early symptoms | Crude | - |
3 | Fukuda et al. [10] | 2023 | Japan | Retrospective cohort | 9,651 | 5,342/4,309 | Capital: 76.0±11.5 | 4 h | Sex, EMS use, ischemic stroke subtype, ADI | Adjustment | Age, population density, and stroke hospital access |
Noncapital: 77.4±11.2 | |||||||||||
4 | Mellon et al. [20] | 2016 | Ireland | Cross-sectional | 149 | 89/60 | 69.3±13.05 | 3.5 h | Context of symptom onset, first action upon noticing symptoms, thoughts following onset, response of other people, recall of FAST campaign | Crude | - |
5 | Hong et al. [21] | 2013 | Korea | Prospective cohort | 184 | 98/86 | Non-delay: 61.9±16.5 | 2 h | Aggravated symptom, symptom development, history of atrial fibrillation, ED presentation | Crude | - |
Delay: 69.5±11.1 | |||||||||||
6 | Mkoma et al. [22] | 2022 | Denmark | Cohort | 49,817 | 20,895/28,922 | Danish-born: 73 (63.81) | 4.5 h | Aboriginal, Immigrant | Adjustment | Age, sex, stroke severity, previous stroke or TIA, smoking, myocardial infarction, atrial fibrillation, diabetes, hypertension, income, occupation, education, marital status, and duration of residence |
Immigrants: 69 (59.78) | |||||||||||
7 | Maze et al. [23] | 2004 | USA | Cross-sectional | 50 | 26/24 | 49.2 (22–81) | 3 h | Mode of transportation, income | Crude | - |
8 | Harper et al. [24] | 1992 | UK | Prospective | 374 | 176/198 | 77 (29–98) | 3 h | Route, living alone, nocturnal onset | Adjustment | NM |
9 | Nasreldein et al. [25] | 2023 | Egypt | Prospective cohort | 618 | 351/267 | Rural: 63.57±11.51 | 4.5 h | Initial misdiagnosis, presentation to non-stroke-ready hospitals | Crude | - |
Urban: 63.48±11.96 | |||||||||||
10 | Kharbach et al. [26] | 2021 | Morocco | Prospective cross-sectional | 197 | 90/107 | 68.77±12.28 | 4.5 h | Level of education, patient behavior, knowledge of stroke, direct admission, vertigo and balance disturbance, EMS use, distance to the hospital | Adjustment | - |
11 | Acharya et al. [27] | 2009 | USA | Retrospective cohort | 330 | 143/187 | 65.6±15.6 | 3 h | Distance to the hospital | Adjustment | Age, race, sex, degree, of neurologic deficit (as measured by NIHSS score), insurance status (insured vs. noninsured), day of week (weekday vs. weekend), 22 and arrival during rush hour (6–8 a.m. or 4–6 p.m.) versus non-rush hour |
12 | Madsen et al. [41] | 2015 | Peru | Retrospective cross-sectional | 1,991 | 894/1,097 | Men: 67 (57–79) | 3 h | NIHSS score, presenting symptoms, living alone, EMS use, wake-up stroke, night arrival, mRS score | Adjustment | Age, initial retrospective National Institutes of Health Stroke Scale (rNIHSS), 17 black race, insurance status, marital status, living situation, emergency medical service (EMS) use (yes/no), presenting symptoms, prestroke modified Rankin Scale (mRS) score, wake-up stroke (yes/no), night arrival (i.e., between 6 p.m. and 6 a.m. [yes/no]), and history of prior stroke (yes/no) |
Women: 74 (60–84) | |||||||||||
13 | Soto-Cámara et al. [28] | 2019 | Spain | Cross-sectional | 322 | 181/141 | ≤75: 118 | 3.5 h | Asking for help immediately, patient’s perception of controlling symptoms, outside the home, EMS use, speech/language difficulties, self-managed, knowledge of stroke, time of day, type of day, onset of symptoms | Adjustment | Age, sex |
>75: 204 | |||||||||||
14 | Sim et al. [29] | 2015 | Korea | Cross-sectional | 229 | 138/91 | 68.5±10.57 | 3 h | Symptoms, perceived of patients | Crude | - |
15 | Park et al. [42] | 2016 | Korea | Multicenter | 20,780 | 12,153/8,627 | >65: 7,622 | 2 h | EMS use, transfer, age, sex, level of education, level of urbanization, past medical history, night, weekend, symptom | Adjustment | Age over 65 years, sex, education level, level of urbanization, past medical history (diabetes, cardiovascular disease, cerebrovascular disease), symptom onset hour of day, symptom onset day of week, and presentation symptoms at arrival. OR, |
≤65: 13,158 | odds ratio; 95% CI, 95% confidence interval; EMS, emergency medical services; Direct, arrived final hospital directly; Indirect, arrived final hospital via other hospital. Reference values are shown in parentheses | ||||||||||
16 | Sheikh Hassan et al. [9] | 2022 | Somalia | Cross-Sectional | 212 | 113/99 | 62±10 | 4 h | Sex, level of education, residence, GCS score, distance to the hospital, living alone, stroke onset, lack of knowledge about thrombolytic treatment for acute stroke, non-hemiplegic presentation | Crude | - |
17 | Mark O'Meara et al. [34] | 2022 | South Africa | Retrospective | 381 | 186/195 | 62 (51, 70) | 3 h | EMS use | Crude | - |
18 | Yuan et al. [12] | 2023 | China | Cross-sectional | 78,389 | 49,999/28,390 | 65.63±11.92 | 3 h | EMS use, age | Adjustment | Age, provinces, settings, onset during night, stroke severity, admitted through ED, and arrival via ambulance, which were consistent with logistic regressions and can be meaningful for clinical practice |
19 | Alkhotani et al. [35] | 2022 | Saudi Arabia | Prospective cross-sectional | 98 | 52/46 | 60.4±10.3 | 4.5 h | Level of education, employment, early CT finding | Crude | - |
20 | Damon et al. [31] | 2022 | Senegal | Retrospective cross-sectional | 56 | 30/26 | 48.2±13.2 | 3 h | Sex, marital status, type of day | Adjustment | NM |
21 | Lee et al. [30] | 2021 | Korea | Prospective cross-sectional | 539 | 297/242 | 68.3±13.1 | 4.5 h | Sex, level of education, clear onset, mRS score, NIHSS score | Adjustment | Sex, age, educational status, type of stroke onset time (clear- or unclear-onset strokes), previous mRS score, initial NIHSS score |
22 | Minalloh et al. [15] | 2022 | Indonesia | Prospective, multicenter | 126 | 73/53 | 58.56±10.71 | 4.5 h | Referral patient, wake-up stroke, unaware of stroke symptoms, unaware of the gravity of symptoms | Adjustment | NM |
23 | Sittirat et al. [36] | 2022 | Thailand | Cross-sectional | 120 | 61/59 | 70.65±8.06 | 3 h | Distance to the hospital, perceived severity of stroke symptoms | Crude | - |
24 | Lin et al. [32] | 1999 | China | Prospective | 157 | 85/72 | 68.5±11.5 | 2 h | Referral patient, ataxia | Adjustment | NM |
25 | Banks et al. [37] | 1998 | UK | Prospective | 153 | 63/90 | Male: 74 (42–92), female: 79 (46–95) | 3 h | Patient factors, Gp factors, ambulance factors | NM | NM |
26 | Zhou et al. [33] | 2016 | China | Prospective | 1,835 | 1,149/686 | ≤65 years: 918 | 3 h | Previous stroke, knowing someone who had suffered a stroke, onset location, patients noticed the symptoms first, severity of the symptoms, EMS use, went to hospital directly, contacted relative/acquaintance, referral patient, distance to the hospital, age, residence | Crude | - |
>65 years: 917 | |||||||||||
27 | Nepal et al. [6] | 2019 | Nepal | Cross-sectional | 228 | 121/107 | <60 years: 74 | 3 h | Location, identification of stroke, response of symptoms, awareness of stroke treatment, traffic jam, distance to the hospital, income | Crude | - |
>60 years: 154 | |||||||||||
28 | Onder et al. [39] | 2020 | Turkey | Retrospective cross-sectional | 87 | 47/40 | 69.7±11.7 | 3 h | NM | Crude | - |
29 | Tan et al. [38] | 2014 | Singapore | Observational cross-sectional | 642 | 409/233 | Wake-up stroke: 64 (56, 74) | 3.5 h | Wake-up stroke | Crude | - |
Stroke while awake: 65 (57, 74) |
No. . | First author . | Year . | Country . | Study design . | Sample size . | Sex (male/female) . | Age . | Cut-off of prehospital delay . | Factors of analysis . | Adjustment crude or adjustment . | Confounders . |
---|---|---|---|---|---|---|---|---|---|---|---|
1 | Kim et al. [40] | 2011 | Korea | Prospective | 500 | 310/190 | 67 | 3 h | NIHSS score, Previous stroke, diabetes, hyperlipidemia, atrial fibrillation, coronary heart disease, living alone, arrival through referral, EMS use, knowledge by patient of thrombolysis, awareness of the patient/bystander that the initial symptom was stroke-related | Adjustment | NM |
2 | Ghadimi et al. [14] | 2021 | Iran | Prospective cohort | 204 | 114/90 | 68.99±13.91 | 4.5 h | Consultation after the onset of symptoms, EMS use, patient’s perception of early symptoms | Crude | - |
3 | Fukuda et al. [10] | 2023 | Japan | Retrospective cohort | 9,651 | 5,342/4,309 | Capital: 76.0±11.5 | 4 h | Sex, EMS use, ischemic stroke subtype, ADI | Adjustment | Age, population density, and stroke hospital access |
Noncapital: 77.4±11.2 | |||||||||||
4 | Mellon et al. [20] | 2016 | Ireland | Cross-sectional | 149 | 89/60 | 69.3±13.05 | 3.5 h | Context of symptom onset, first action upon noticing symptoms, thoughts following onset, response of other people, recall of FAST campaign | Crude | - |
5 | Hong et al. [21] | 2013 | Korea | Prospective cohort | 184 | 98/86 | Non-delay: 61.9±16.5 | 2 h | Aggravated symptom, symptom development, history of atrial fibrillation, ED presentation | Crude | - |
Delay: 69.5±11.1 | |||||||||||
6 | Mkoma et al. [22] | 2022 | Denmark | Cohort | 49,817 | 20,895/28,922 | Danish-born: 73 (63.81) | 4.5 h | Aboriginal, Immigrant | Adjustment | Age, sex, stroke severity, previous stroke or TIA, smoking, myocardial infarction, atrial fibrillation, diabetes, hypertension, income, occupation, education, marital status, and duration of residence |
Immigrants: 69 (59.78) | |||||||||||
7 | Maze et al. [23] | 2004 | USA | Cross-sectional | 50 | 26/24 | 49.2 (22–81) | 3 h | Mode of transportation, income | Crude | - |
8 | Harper et al. [24] | 1992 | UK | Prospective | 374 | 176/198 | 77 (29–98) | 3 h | Route, living alone, nocturnal onset | Adjustment | NM |
9 | Nasreldein et al. [25] | 2023 | Egypt | Prospective cohort | 618 | 351/267 | Rural: 63.57±11.51 | 4.5 h | Initial misdiagnosis, presentation to non-stroke-ready hospitals | Crude | - |
Urban: 63.48±11.96 | |||||||||||
10 | Kharbach et al. [26] | 2021 | Morocco | Prospective cross-sectional | 197 | 90/107 | 68.77±12.28 | 4.5 h | Level of education, patient behavior, knowledge of stroke, direct admission, vertigo and balance disturbance, EMS use, distance to the hospital | Adjustment | - |
11 | Acharya et al. [27] | 2009 | USA | Retrospective cohort | 330 | 143/187 | 65.6±15.6 | 3 h | Distance to the hospital | Adjustment | Age, race, sex, degree, of neurologic deficit (as measured by NIHSS score), insurance status (insured vs. noninsured), day of week (weekday vs. weekend), 22 and arrival during rush hour (6–8 a.m. or 4–6 p.m.) versus non-rush hour |
12 | Madsen et al. [41] | 2015 | Peru | Retrospective cross-sectional | 1,991 | 894/1,097 | Men: 67 (57–79) | 3 h | NIHSS score, presenting symptoms, living alone, EMS use, wake-up stroke, night arrival, mRS score | Adjustment | Age, initial retrospective National Institutes of Health Stroke Scale (rNIHSS), 17 black race, insurance status, marital status, living situation, emergency medical service (EMS) use (yes/no), presenting symptoms, prestroke modified Rankin Scale (mRS) score, wake-up stroke (yes/no), night arrival (i.e., between 6 p.m. and 6 a.m. [yes/no]), and history of prior stroke (yes/no) |
Women: 74 (60–84) | |||||||||||
13 | Soto-Cámara et al. [28] | 2019 | Spain | Cross-sectional | 322 | 181/141 | ≤75: 118 | 3.5 h | Asking for help immediately, patient’s perception of controlling symptoms, outside the home, EMS use, speech/language difficulties, self-managed, knowledge of stroke, time of day, type of day, onset of symptoms | Adjustment | Age, sex |
>75: 204 | |||||||||||
14 | Sim et al. [29] | 2015 | Korea | Cross-sectional | 229 | 138/91 | 68.5±10.57 | 3 h | Symptoms, perceived of patients | Crude | - |
15 | Park et al. [42] | 2016 | Korea | Multicenter | 20,780 | 12,153/8,627 | >65: 7,622 | 2 h | EMS use, transfer, age, sex, level of education, level of urbanization, past medical history, night, weekend, symptom | Adjustment | Age over 65 years, sex, education level, level of urbanization, past medical history (diabetes, cardiovascular disease, cerebrovascular disease), symptom onset hour of day, symptom onset day of week, and presentation symptoms at arrival. OR, |
≤65: 13,158 | odds ratio; 95% CI, 95% confidence interval; EMS, emergency medical services; Direct, arrived final hospital directly; Indirect, arrived final hospital via other hospital. Reference values are shown in parentheses | ||||||||||
16 | Sheikh Hassan et al. [9] | 2022 | Somalia | Cross-Sectional | 212 | 113/99 | 62±10 | 4 h | Sex, level of education, residence, GCS score, distance to the hospital, living alone, stroke onset, lack of knowledge about thrombolytic treatment for acute stroke, non-hemiplegic presentation | Crude | - |
17 | Mark O'Meara et al. [34] | 2022 | South Africa | Retrospective | 381 | 186/195 | 62 (51, 70) | 3 h | EMS use | Crude | - |
18 | Yuan et al. [12] | 2023 | China | Cross-sectional | 78,389 | 49,999/28,390 | 65.63±11.92 | 3 h | EMS use, age | Adjustment | Age, provinces, settings, onset during night, stroke severity, admitted through ED, and arrival via ambulance, which were consistent with logistic regressions and can be meaningful for clinical practice |
19 | Alkhotani et al. [35] | 2022 | Saudi Arabia | Prospective cross-sectional | 98 | 52/46 | 60.4±10.3 | 4.5 h | Level of education, employment, early CT finding | Crude | - |
20 | Damon et al. [31] | 2022 | Senegal | Retrospective cross-sectional | 56 | 30/26 | 48.2±13.2 | 3 h | Sex, marital status, type of day | Adjustment | NM |
21 | Lee et al. [30] | 2021 | Korea | Prospective cross-sectional | 539 | 297/242 | 68.3±13.1 | 4.5 h | Sex, level of education, clear onset, mRS score, NIHSS score | Adjustment | Sex, age, educational status, type of stroke onset time (clear- or unclear-onset strokes), previous mRS score, initial NIHSS score |
22 | Minalloh et al. [15] | 2022 | Indonesia | Prospective, multicenter | 126 | 73/53 | 58.56±10.71 | 4.5 h | Referral patient, wake-up stroke, unaware of stroke symptoms, unaware of the gravity of symptoms | Adjustment | NM |
23 | Sittirat et al. [36] | 2022 | Thailand | Cross-sectional | 120 | 61/59 | 70.65±8.06 | 3 h | Distance to the hospital, perceived severity of stroke symptoms | Crude | - |
24 | Lin et al. [32] | 1999 | China | Prospective | 157 | 85/72 | 68.5±11.5 | 2 h | Referral patient, ataxia | Adjustment | NM |
25 | Banks et al. [37] | 1998 | UK | Prospective | 153 | 63/90 | Male: 74 (42–92), female: 79 (46–95) | 3 h | Patient factors, Gp factors, ambulance factors | NM | NM |
26 | Zhou et al. [33] | 2016 | China | Prospective | 1,835 | 1,149/686 | ≤65 years: 918 | 3 h | Previous stroke, knowing someone who had suffered a stroke, onset location, patients noticed the symptoms first, severity of the symptoms, EMS use, went to hospital directly, contacted relative/acquaintance, referral patient, distance to the hospital, age, residence | Crude | - |
>65 years: 917 | |||||||||||
27 | Nepal et al. [6] | 2019 | Nepal | Cross-sectional | 228 | 121/107 | <60 years: 74 | 3 h | Location, identification of stroke, response of symptoms, awareness of stroke treatment, traffic jam, distance to the hospital, income | Crude | - |
>60 years: 154 | |||||||||||
28 | Onder et al. [39] | 2020 | Turkey | Retrospective cross-sectional | 87 | 47/40 | 69.7±11.7 | 3 h | NM | Crude | - |
29 | Tan et al. [38] | 2014 | Singapore | Observational cross-sectional | 642 | 409/233 | Wake-up stroke: 64 (56, 74) | 3.5 h | Wake-up stroke | Crude | - |
Stroke while awake: 65 (57, 74) |
NM, not mentioned; NIHSS, National Institutes of Health Stroke Scale; EMS, emergency medical service; ADI, Areal Deprivation Index; FAST, Face, Arm, Speech, Time; ED, emergency department; GCS, Glasgow Coma Scale; CT, computed tomography; mRS, modified Rankin Scale; Gp, general practitioner.
Quality Assessment
The quality evaluation results of 24 cross-sectional studies using the JBI scale were as follows: 21 studies scored 6 points or higher, and the overall quality of the included studies was high (online suppl. Table S1). In terms of whether to adjust the confounding factors, 12 studies scored 1, and 12 studies scored 0, which may cause biased results due to confounding factors. The quality of six cohort studies was assessed using the NOS scale. The results showed that five studies scored 6 points or higher (online suppl. Table S2). In terms of design or analysis based on important factors, four studies scored 0 and two studies scored 1, which may also lead to biased results due to confounding factors.
Meta-Analysis Results
Prehospital Care-Seeking Delay Incidence
A total of 30 studies reported the incidence of prehospital care-seeking delay, and the random-effects model was used for meta-analysis. The estimated incidence of prehospital care-seeking delay was 68% (95% CI: 59–76%, p < 0.001, I2 = 99.95%).
The results of subgroup analysis are summarized in Table 2. Based on study location, country income group, recruitment date, and age, subgroup analysis was carried out to explore the source of heterogeneity. Among them, there were differences in the incidence of prehospital care-seeking delay in different country income groups (p = 0.035) (low-income countries [85%, 95% CI: 75–89%], upper to middle-income countries [79%, 95% CI: 70–87%], lower to middle-income countries [68%, 95% CI: 60–76%], high-income countries and lower to middle-income countries [62%, 95% CI: 55–68%]). In terms of subgroup analysis by study location, recruitment date, and age, there was no significant difference in the incidence of prehospital care-seeking delay (p > 0.05).
Subgroup analyses of the incidence of prehospital care-seeking delay
Subject . | Reports, n . | Prevalence, % . | 95% CI . | I2, % . | p value for heterogeneity . | p value between groups . |
---|---|---|---|---|---|---|
Study location | 0.681 | |||||
African region | 5 | 74 | 61–85 | 96.11 | <0.001 | |
Region of the Americas | 4 | 57 | 36–78 | 96.67 | <0.001 | |
Southeast Asia region | 3 | 79 | 70–86 | 0 | - | |
European region | 6 | 58 | 44–70 | 97.83 | <0.001 | |
Eastern Mediterranean region | 2 | 61 | 56–67 | 0 | - | |
Western Pacific region | 10 | 73 | 64–82 | 99.85 | <0.001 | |
Country income group | 0.035 | |||||
High | 17 | 62 | 55–68 | 99.8 | <0.001 | |
Upper to middle | 7 | 79 | 70–87 | 99.12 | <0.001 | |
Lower to middle | 5 | 68 | 60–76 | 87.81 | <0.001 | |
Low | 1 | 85 | 79–89 | - | - | |
Recruitment date | 0.423 | |||||
≤2010 | 5 | 65 | 50–79 | 95.73 | <0.001 | |
2011–2020 | 12 | 66 | 55–77 | 99.85 | <0.001 | |
≥2021 | 13 | 70 | 54–85 | 99.96 | <0.001 | |
Age, years | 0.851 | |||||
≤59 (middle-aged) | 3 | 68 | 39–91 | - | - | |
60–74 (young-old) | 21 | 68 | 57–79 | 99.97 | <0.001 | |
75–89 (old-old) | 3 | 70 | 64–76 | - | - | |
Other groups | 3 | 63 | 41–82 | - | - |
Subject . | Reports, n . | Prevalence, % . | 95% CI . | I2, % . | p value for heterogeneity . | p value between groups . |
---|---|---|---|---|---|---|
Study location | 0.681 | |||||
African region | 5 | 74 | 61–85 | 96.11 | <0.001 | |
Region of the Americas | 4 | 57 | 36–78 | 96.67 | <0.001 | |
Southeast Asia region | 3 | 79 | 70–86 | 0 | - | |
European region | 6 | 58 | 44–70 | 97.83 | <0.001 | |
Eastern Mediterranean region | 2 | 61 | 56–67 | 0 | - | |
Western Pacific region | 10 | 73 | 64–82 | 99.85 | <0.001 | |
Country income group | 0.035 | |||||
High | 17 | 62 | 55–68 | 99.8 | <0.001 | |
Upper to middle | 7 | 79 | 70–87 | 99.12 | <0.001 | |
Lower to middle | 5 | 68 | 60–76 | 87.81 | <0.001 | |
Low | 1 | 85 | 79–89 | - | - | |
Recruitment date | 0.423 | |||||
≤2010 | 5 | 65 | 50–79 | 95.73 | <0.001 | |
2011–2020 | 12 | 66 | 55–77 | 99.85 | <0.001 | |
≥2021 | 13 | 70 | 54–85 | 99.96 | <0.001 | |
Age, years | 0.851 | |||||
≤59 (middle-aged) | 3 | 68 | 39–91 | - | - | |
60–74 (young-old) | 21 | 68 | 57–79 | 99.97 | <0.001 | |
75–89 (old-old) | 3 | 70 | 64–76 | - | - | |
Other groups | 3 | 63 | 41–82 | - | - |
Related Factors of Prehospital Care-Seeking Delay
Demographic Characteristics of Patients. The relationship between nine demographic factors and the delay in prehospital care-seeking delay was analyzed, including sex (n = 9) [7‒10, 12, 30‒33], residence (n = 4) [8, 9, 12, 33], living alone (n = 5) [7, 9, 11, 32, 33], marital status (n = 2) [7, 31], distance to the hospital (n = 4) [6, 9, 26, 27], age (n = 6) [7, 8, 12, 30, 32, 33], income (n = 2) [6, 33], medical insurance (n = 2) [7, 33], and level of education (n = 6) [8, 9, 26, 30, 32, 33]. Meta-analysis showed that the risk of prehospital care-seeking delays significantly increased with greater distances from the hospital (odds ratio = 2.47, 95% CI: 1.18–5.16, p = 0.016, I2 = 68.9%) and lower education levels (odds ratio = 1.46, 95% CI: 1.04–2.03, p = 0.027, I2 = 71.5%).
Factors that did not show a significant correlation with the prehospital care-seeking delays include sex (odds ratio = 1.05, 95% CI: 0.95–1.17, p = 0.341, I2 = 66.7%), residence (odds ratio = 1.50, 95% CI: 0.95–2.36, p = 0.081, I2 = 96.5%), living alone (odds ratio = 1.29, 95% CI: 0.80–2.08, p = 0.301, I2 = 66.8%), marital status (odds ratio = 2.19, 95% CI: 0.30–16.31, p = 0.442, I2 = 84.2%), age (odds ratio = 1.04, 95% CI: 0.99–1.09, p = 0.111, I2 = 89.6%), income (odds ratio = 0.47, 95% CI: 0.20–1.09, p = 0.080, I2 = 65.9%), and medical insurance (odds ratio = 0.92, 95% CI: 0.66–1.29, p = 0.727, I2 = 13.8%) (shown in online suppl. Fig. S1).
Risk Factors of Stroke. Six factors, including atrial fibrosis (n = 2), coronary heart disease (n = 2), diabetes (n = 3), prior stroke (n = 5), hyperlipidemia (n = 2), and cerebrovascular disease (n = 2), were reported to have relationship with the delay in prehospital care-seeking [7, 8, 11, 12, 21, 32, 33, 39]. Meta-analysis showed that compared with patients without related diseases, patients with a history of diabetes, hyperlipidemia, and stroke were more likely to delay seeking medical treatment (odds ratio = 1.33, 95% CI: 1.24–1.43, p < 0.001, I2 = 0%; odds ratio = 1.85, 95% CI: 1.02–3.37, p = 0.045, I2 = 0%; odds ratio = 1.17, 95% CI: 1.10–1.25, p < 0.001, I2 = 0%, respectively). Compared with those without atrial fibrosis, patients with a history of atrial fibrosis were less prone to prehospital care-seeking delay (odds ratio = 0.37, 95% CI: 0.18–0.74, p = 0.039, I2 = 43.7%).
Coronary heart disease and cerebrovascular disease had no significant correlation with prehospital care-seeking delays in stroke patients (odds ratio = 0.82, 95% CI: 0.42–1.59, p = 0.572, I2 = 12.9%; odds ratio = 1.24, 95% CI: 0.72–2.15, p = 0.443, I2 = 59.5%, respectively) (shown in online suppl. Fig. S2).
Initial Symptoms. Seven factors, including consciousness disturbance (n = 3), dizziness/vertigo (n = 5), headache (n = 2), hemiparesis (n = 2), language disturbance (n = 4), numbness (n = 2), visual disturbance (n = 3), were reported to have relationship with the delay in prehospital care-seeking [7‒9, 26, 28, 32, 33].
Meta-analysis showed that compared with patients without consciousness disturbance and visual disturbance, patients with both symptoms were less likely to suffer from prehospital care-seeking delays (odds ratio = 0.66, 95% CI: 0.46–0.97, p = 0.034, I2 = 63.7%; odds ratio = 0.78, 95% CI: 0.59–0.96, p = 0.022, I2 = 0%, respectively).
Factors that did not show a significant correlation with prehospital care-seeking delays include dizziness/vertigo (odds ratio = 1.01, 95% CI: 0.93–1.11, p = 0.784, I2 = 47.1%), headache (odds ratio = 1.08, 95% CI: 0.85–1.37, p = 0.531, I2 = 0%), hemiparesis (odds ratio = 0.43, 95% CI: 0.16–1.16, p = 0.095, I2 = 43.7%), language disturbance (odds ratio = 0.78, 95% CI: 0.56–1.10, p = 0.155, I2 = 73%), and numbness (odds ratio = 0.80, 95% CI: 0.45–1.39, p = 0.425, I2 = 66.2%) (shown in online suppl. Fig. S3).
Relevant Circumstances at the Time of Onset. Six factors, including Modified Rankin Scale (mRS) (n = 2), National Institutes of Health Stroke Scale (NIHSS) (n = 3), home (n = 4), stroke day (n = 5), time of day (n = 5), wake-up stroke (n = 3), were reported to have a relationship with the delay in prehospital care-seeking [6‒8, 11, 12, 15, 28, 30‒33].
Meta-analysis showed that individuals with lower mRS scores at admission and higher NIHSS scores were not more prone to prehospital care-seeking delays (odds ratio = 0.67, 95% CI: 0.55–0.81, p < 0.001, I2 = 0%; odds ratio = 0.88, 95% CI: 0.79–0.98, p = 0.025, I2 = 84%, respectively).
There was no significant correlation between prehospital care-seeking delays and various factors including the location of the home (odds ratio = 0.97, 95% CI: 0.28–3.38, p = 0.962, I2 = 90.9%), the day of the stroke (odds ratio = 1.06, 95% CI: 0.93–1.21, p = 0.395, I2 = 68.8%), time of the day (odds ratio = 0.96, 95% CI: 0.75–1.22, p = 0.719, I2 = 93.3%), and wake-up strokes (odds ratio = 9.26, 95% CI: 0.86–99.90, p = 0.067, I2 = 94.7%) (shown in online suppl. Fig. S4).
Prehospital Care-Seeking Behavior of Patients. Three factors, including emergency medical service (EMS) use (n = 11), arrival through referral (n = 5), and asking for help (n = 4), were reported to have a relationship with the delay in prehospital care-seeking [6, 7, 10‒12, 14, 15, 21, 26, 28, 32, 33, 36, 39].
Meta-analysis showed that compared with patients who did not use EMS, patients who used EMS were less likely to delay seeking medical help (odds ratio = 0.36, 95% CI: 0.24–0.54, p < 0.001, I2 = 94.8%). If the patients seek help immediately at disease onset, the risk of prehospital care-seeking delay could be reduced (odds ratio = 0.07, 95% CI: 0.02–0.25, p < 0.001, I2 = 67.5%). There was no significant correlation between arrival through referral and prehospital care-seeking delay in stroke patients (odds ratio = 1.63, 95% CI: 0.31–8.66, p = 0.566, I2 = 94.3%) (shown in online suppl. Fig. S5).
Stroke-Related Knowledge by Patients. Two factors, including the knowledge by patient of thoroughness (n = 4) and previous knowledge of stroke (n = 2), were reported to have a relationship with the delay in prehospital care-seeking [6, 9, 11, 26, 28, 32]. Meta-analysis showed that patients aware of thrombolytic therapy for stroke were less likely to delay seeking prehospital care compared with those who were unaware of the therapy (odds ratio = 0.19, 95% CI: 0.07–0.56, p = 0.002, I2 = 74.3%). There was no significant correlation between the previous knowledge of stroke and the occurrence of the delay in prehospital care-seeking (odds ratio = 0.49, 95% CI: 0.20–1.19, p = 0.114, I2 = 54.7%) (shown in online suppl. Fig. S6).
Perception of Disease by Patients. Two factors, including perceiving the severity of early symptoms (n = 3) and stroke recognition (n = 7), were reported to have a relationship with prehospital care-seeking delay [6, 9, 11, 14, 15, 21, 28, 32, 33]. Meta-analysis results showed that patients who could perceive the severity of early symptoms and had stroke recognition were less likely to suffer from prehospital care-seeking delay (odds ratio = 0.23, 95% CI: 0.07–0.72, p = 0.011, I2 = 69.5%; odds ratio = 0.36, 95% CI: 0.19–0.68, p = 0.002, I2 = 86.2%, respectively) (shown in online suppl. Fig. S7).
Sensitivity Analysis
The sensitivity analysis of the influencing factors was carried out by one-by-one exclusion method. The analysis results indicated that none of the pooled results were significantly influenced by any single study. This showed that the results of this meta-analysis were relatively reliable, as shown in online supplementary Figure S8.
Publication Bias
In order to ensure the validity of meta-analysis results, a funnel plot, Egger’s test, and Begg’s test were employed to identify publication bias. The results showed that the publication bias was not significant regarding sex, living alone, age, level of education, prior stroke, dizziness/vertigo, stroke day, time of day, EMS use, arrival through referral, and stroke recognition (p = 0.076, p = 0.462, p = 0.26, p = 0.065, p = 0.462, p = 0.221, p = 0.086, p = 0.806, p = 0.533, p = 0.806, p = 0.230, respectively), which further confirmed the reliability of our results.
Discussion
The current research shows that the total incidence of prehospital care-seeking delay in AIS patients is 68% (95% CI: 59–76%). These results showed that more attention should be paid to the delay in prehospital care-seeking in AIS patients. In addition, our research also confirmed that AIS patients near the hospital, with a high education level, a history of atrial fibrillation, symptoms of consciousness disturbance and visual disturbance at the time of onset, or realizing that they had a stroke and perceive its severity, have a lower risk of prehospital care-seeking delay. However, a history of diabetes, hyperlipidemia, and stroke, high mRS score and low NIHSS score at admission, not using EMS for prehospital care-seeking, not seeking help in time after onset, and not knowing thrombolytic therapy were risk factors for prehospital care-seeking delay. These factors should be considered when formulating strategies to prevent the delay in prehospital care-seeking.
The subgroup analysis showed that the incidence of prehospital care-seeking delay in AIS treatment was the lowest in high-income countries and increased in upper-middle-income countries and low-income countries sequentially. These differences are likely attributed to many factors, including variances in medical resources, publicity and education, cultural habits, and medical systems [43], which reflect the inequality in the accessibility of medical resources and health services. In high-income countries, although the incidence of prehospital care-seeking delay in treating AIS was 62%, which was the lowest compared with other low-income countries, prehospital care-seeking delay is still prominent. This delay may be due to a variety of factors, including the lack of awareness of early stroke, the complexity of the medical system, and patient’s misunderstanding of symptoms [44, 45]. In low-income countries, the incidence of prehospital care-seeking delay was as high as 85%, which may be partly caused by a lack of medical resources, insufficient coverage of medical resources, and inconvenient transportation [46‒48]. These problems may affect patients’ and their families’ perception of symptoms and their ability to seek medical treatment quickly. However, it is worth noting that there are few related studies from low-income and upper-middle-income countries, which highlights the need for these countries to provide medical services and strengthen relevant research. Therefore, it is necessary to conduct more studies in these countries to further clarify the incidence of prehospital care-seeking delay in AIS patients and its related factors.
Fifteen Factors Have Been Confirmed to Be Related to the Delay in Prehospital Care-Seeking
AIS patients with the disease occurring close to the hospital were less likely to delay in prehospital care-seeking. The influence of the distance from the hospital on the delay in prehospital care-seeking may be related to factors such as traffic conditions, allocation of medical resources, and individual’s perception of symptoms [47]. Usually, patients who are far away from hospitals may encounter more challenges in seeking medical treatment because they need more time and resources to reach medical facilities [49]. Patients with higher education levels are less likely to delay in prehospital care-seeking because they usually know more about health knowledge and stroke symptoms, so it is easier for them to identify symptoms and seek medical treatment in time [50]. This result has important implications for improving community education, strengthening health awareness, and improving the efficiency of prehospital care-seeking for AIS patients. Therefore, it is suggested that measures such as improving the cognition of stroke symptoms, strengthening the construction of community medical resources, and improving traffic conditions may help reduce the delay in prehospital care-seeking and improve the treatment efficiency for patients with AIS.
Compared with those without underlying diseases, AIS patients with a history of diabetes, hyperlipidemia, and stroke were more likely to suffer from prehospital care-seeking delay. This difference may be caused by many factors. For example, patients with diabetes and other diseases may misunderstand or confuse the symptoms of stroke, thus delaying seeing a doctor [51, 52]. In addition, these patients may focus more on the management chronic disease, so they may pay less attention to emerging symptoms [53]. On the contrary, AIS patients without a history of atrial fibrosis are more likely to delay in prehospital care-seeking. Atrial fibrosis is a common arrhythmia, and this discovery may suggest that patients with atrial fibrosis are more sensitive to perceive the symptoms or risks related to stroke and thus will seek medical help faster [54]. Hence, developing customized education and medical programs is necessary for patients with different types of stroke, especially those with other chronic diseases or risk factors. Individualized education and intervention measures may help reduce the delay in prehospital care-seeking, thus improving the outcomes and prognosis of patients with AIS.
AIS patients with consciousness disturbance and visual disturbance are not prone to prehospital care-seeking delay. Disturbance of consciousness is common in AIS patients, which are related to a higher incidence of stroke-related complications and 3-month death/disability, mainly due to patients’ advanced age and large-scale cerebral infarction [55]. These symptoms are usually obvious and easy to be noticed by patients themselves or people around them. Especially when the level of consciousness drops rapidly, patients or their families often feel a sense of urgency to seek medical help quickly. Visual impairment may be manifested as blurred vision and visual field defect [56, 57]. These symptoms directly affect people’s daily lives and activities, such as reading, walking, and driving. Affected patients may quickly perceive these problems and realize the need for immediate medical treatment, thus reducing the incidence of prehospital care-seeking delay.
The higher the mRS score of AIS patients at admission, the lower the NIHSS score, the more likely it is for them to delay in prehospital care-seeking. This finding is very important and practical. First, it is very significant to understand the influence of NIHSS score and mRS score on the delay in prehospital care-seeking. The NIHSS score is used to evaluate the degree of acute stroke, while the mRS score is used to assess the functional status and disability degree of patients after stroke. A high mRS score may indicate that the patient has a certain disability in a previous stroke or other related events, which may lead to a more cautious attitude toward seeking medical treatment when facing new symptoms, and an increase in the possibility of prehospital care-seeking delay [7]. This delay may be because patients or their families are depressed or anxious about seeing a doctor again, or they may be unwilling to see a doctor quickly because of their previous treatment experience [30]. On the other hand, a low NIHSS score may indicate that the symptoms of stroke are relatively mild [58]. Patients or people around them may underestimate the severity of symptoms, leading to the delay in prehospital care-seeking. This finding emphasizes the importance of comprehensive evaluation for stroke patients. In terms of education and publicity, more efforts are needed to improve public awareness of the symptoms of stroke, especially the great significance of prompt treatment to the final rehabilitation.
If AIS patients can seek help and use EMS for medical treatment immediately, the risk of prehospital care-seeking delay can be reduced. The purpose of EMS system is to provide first aid and ensure that patients with acute diseases are treated in time. When AIS occurs, EMS can quickly start first aid and delivery, thus minimizing the possibility of prehospital care-seeking delay [59]. In addition, EMS personnel can usually provide necessary medical support during AIS emergencies, including intravenous thrombolysis. It ensures that patients can get professional first-aid services and delivery in the shortest time, which can greatly affect the follow-up rehabilitation and prognosis of patients [60].
In this study, AIS patients who know about AIS thrombolytic therapy are obviously less likely to delay in prehospital care-seeking. Thrombolytic therapy is very important for AIS patients because it can help dissolve thrombus to restore blood supply, thus minimizing the damage from stroke [61, 62]. For AIS patients, understanding the importance of thrombolytic therapy means that they are more likely to seek medical treatment quickly because they can realize that time is crucial for stroke treatment. This understanding enables them to actively seek medical assistance, or their relatives are more likely to realize the seriousness of the emergency condition and take actions faster. Therefore, compared with those who do not know about thrombolytic therapy, AIS patients who know about this therapy can get first aid and treatment faster and reduce the risk of prehospital care-seeking delay.
Patients with AIS can perceive the severity of early symptoms and recognize stroke, and it is less likely for them to delay in prehospital care-seeking. This is because when AIS patients and people around them identify the symptoms of stroke and understand the importance of timely medical treatment, they are more likely to seek medical assistance quickly [63]. Knowing the severity of symptoms will help patients and others realize the potential emergency situation more quickly, thus reducing the risk of prehospital care-seeking delay [64]. Thus, it is crucial to educate AIS patients and the public on how to identify stroke symptoms and understand the medical treatment behavior in an emergency.
In terms of effect sizes, the most significant variable related to a lower risk of prehospital care-seeking delay was the distance of AIS onset to the hospital, followed by hyperlipidemia, and patient’s experience of receiving a higher education. In practice, all these identified factors should be added to the routine AIS assessment in order to identify the AIS population at high risk of prehospital care-seeking delay. In addition, these factors should be considered when formulating intervention strategies to prevent prehospital care-seeking delay, such as carrying out education and publicity activities for AIS patients in nearby communities, especially those far away from hospitals. Furthermore, it is also essential to emphasize the importance of timely prehospital care-seeking and the harm of delay. It is suggested that countries or regions with relevant resources should establish a system to regularly monitor and evaluate the effectiveness of intervention measures and make adjustments according to feedback, so as to ensure the pertinence and sustainability of these measures.
Limitations
This study reveals the global incidence of AIS prehospital care-seeking delay, which may arouse concerns about this problem and help us to determine its related factors. The results of subgroup analysis and meta-analysis of related factors of prehospital care-seeking delay provide a reference for optimizing the prevention and intervention strategy in this respect. However, there are some limitations to be acknowledged. First, a high degree of heterogeneity was observed. The heterogeneity between studies may be caused by the differences in research design, background, samples, and results evaluation. Second, due to the heterogeneity of the single factor of prehospital care-seeking delay and insufficient information, subgroup analysis was not performed. Finally, this meta-analysis only included studies published in English, and studies in other languages were neglected, which may cause some bias in the results.
Conclusion
Our research indicates that the global incidence of prehospital care-seeking delays in AIS patients is 68%. Factors such as the distance from stroke onset to the hospital, hyperlipidemia, and the patient’s educational background should be incorporated into routine AIS assessments to better identify populations at high risk for prehospital delays. Improving public awareness of stroke symptoms and the critical need for rapid response, enhancing accessibility to EMS in remote or underserved areas might help shortening prehospital delays in AIS.
Acknowledgments
We would like to thank the researchers and study participants for their contributions.
Statement of Ethics
Study approval was not required since this was a meta-analysis of published studies. Written informed consent was not required since this was a meta-analysis of published studies.
Conflict of Interest Statement
The authors declare that they have no competing interests.
Funding Sources
This study was supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region, China (grant 2022D01C440) and Xinjiang Medical University Student Innovation and Entrepreneurship Training Program, China (grant X202310760131).
Author Contributions
Yanjie Zhao and Yuezhen Xu: study concept and design, drafting of the first version of the manuscript, critical revision of the manuscript, data analysis, and interpretation of data. Shuyan Fang, Shengze Zhi, Dongfei Ma, and DongPo Song: study concept and design, data extraction, quality assessment, drafting of the manuscript, interpretation of data, and critical revision of the manuscript. Shizheng Gao and Yifan Wu: literature search, interpretation of data, data analysis, and drafting of the manuscript. Qiqing Zhong, Changxu Jin, and Rui Wang: critical revision of the manuscript. Jiao Sun: supervision and writing – review and editing. All authors reviewed the final draft and were willing to take responsibility for all aspects of the work.
Data Availability Statement
All data generated or analyzed during this study are included in this article.