Abstract
-
Objective
Despite disparities in the availability and quality of emergency care, the extent to which countries with different incomes participate in research on emergency medicine (EM) remains understudied. This study evaluated productivity in EM research depending on national income.
-
Methods
Research published since 2004 in Scopus-indexed journals dealing with EM was analyzed quantitatively. Publication, citation, journal impact, and national socioeconomic data were compared. Automated topic modeling was applied using a latent Dirichlet allocation model.
-
Results
The analysis included 154,458 publications (89.7% in English) from 177 countries. The studies received 1,817,635 citations. High-income countries (HICs) outperformed upper-middle-income countries (UMICs), lower-middle-income countries (LMICs), and low-income countries (LICs) by 11-, 41-, and 72-fold, respectively, by the weighted number of publications (per million population per country), and by 21-, 54-, and 171-fold, respectively, by the weighted count of citations. The annual number of publications is predicted to rise considerably for HICs, to a lesser extent for UMICs, and to far less for LMICs, but not for LICs. Research productivity was significantly related to national socioeconomic indicators. Topic modeling revealed that HICs pay relatively more attention to advances in resuscitation, whereas LICs focus more on injuries.
-
Conclusion
While global research productivity for EM is rising, LICs lag far behind HICs. Priorities in EM research differ among countries with different incomes. The development of country-specific EM research agendas would help boost national academic productivity and determine context-appropriate interventions for improving outcomes in emergency care.
-
Keywords: Emergencies; Emergency medicine; Global health; Research; Resource-limited settings
Capsule Summary
What is already known
A clear understanding of the cumulative global and national research output in the field of emergency medicine (EM) is critical to identify emerging trends and research gaps, prioritize academic agenda, formulate strategic plans and develop interventions that can advance the field in meaningful ways. The extent to which countries with different income levels participate in research on EM remains underinvestigated.
What is new in the current study
Global academic productivity in EM is growing steadily. High-income countries far outstrip lower-income ones in research output. The volume of academic contributions of lower-income countries is predicted to remain negligible for the foreseeable future. A country’s academic productivity was associated with socioeconomic well-being, and priority trends in key EM research themes differ by national income levels. This suggests that developing country-specific research agendas can enhance and tailor academic productivity in EM at the national level.
INTRODUCTION
A clear understanding of the cumulative global and national research output in a particular field is critical to identifying emerging trends and research gaps, prioritizing academic agenda, formulating strategic plans, and developing interventions that advance the field in meaningful ways [
1,
2]. This is particularly true for the diverse domain of emergency medicine (EM), which is experiencing extensive and rapid growth of its literature pool [
3,
4].
Bibliometric studies have examined the production of EM articles, summarized research hotspots and investigated the impact of the publications [
3–
11]. The findings of these studies on research productivity in EM according to country of origin are mainly limited to the evidence regarding the highest publication and citation counts from developed nations [
6,
10]. However, the extent to which countries with different income levels contribute to EM research remains unclear. While disparities in the burden of critical illness and injury, availability and quality of emergency care, and EM research performance between higher- and lower-income countries vary [
12–
16], no quantitative and comparative studies of the scientific contribution to the EM domain from countries of different income levels have been conducted.
This study comprehensively analyzed global academic productivity in the EM field depending on country income level to explore and compare trends in publication and citation counts, investigate potential associations of the research output with national socioeconomic indicators, and look into differences in the distribution of key EM research themes.
METHODS
Sources and selection of data
Scopus was chosen as a source database because it has broader journal coverage compared with MEDLINE, Embase, or the Web of Science, and offers a user-friendly interface that allows for easy filtering and downloading of literature data.
In October 2024, the Scopus Sources page, which allows users to browse a list of journals available on Scopus, was searched for all sources classified with the “Emergency Medicine” subject area. Book series, conference proceedings, and journals without impact metrics were excluded. For all remaining sources, lists of documents published from 2004 onward were retrieved without publication language restrictions. Filters were applied to remove errata, conferences, and retracted papers to ensure the lists included only articles, reviews, notes, letters, editorials, and short surveys. Case reports, although not classified by Scopus as a distinct document type, were indexed under established categories such as “article.” The resulting journal-specific lists of documents, incorporating document title, publication year, citation count, document link, author affiliations, and language of an original document, were downloaded as .csv files using the export function of Scopus, and subsequently merged into a single data table. Documents with non-English titles only or an empty title field were excluded from the dataset.
For every journal, the latest impact indicators retrieved from Scopus included its CiteScore (the number of citations received in four calendar years to peer-reviewed documents published by a journal in the same 4 years, divided by the number of peer-reviewed documents indexed in Scopus and published in the same 4 years) and the journal’s quartile (a metric that ranks journals based on their CiteScore, where the first quartile represents the top 25% and the fourth quartile corresponds to the bottom 25% of the journals in the subject area).
For all documents, a country of origin was determined based on the primary institutional affiliation of the first author (considered the lead researcher), except for publications for which the country information was unavailable. Countries were assigned to world region and income level consistent with the current classification of the World Bank.
Using the World Bank World Development Indicators database as a source, various country-specific socioeconomic data were collected for analysis, including, but not limited to, population counts, birth rates, death rates, mortality rates from communicable diseases, noncommunicable diseases, injuries, life expectancy at birth, gross domestic product (GDP), adjusted net national income, and current health expenditures.
Automated topic modeling
The study employed a title-based automated topic modeling methodology that has been pilot-tested on a less extensive collection of EM literature [
17]. A corpus of the publication titles was prepared for automated text analysis by manually deleting duplicate titles in languages other than English (provided in Scopus in square brackets next to the English titles) and applying OpenRefine software to automatically detect and remove all punctuation, special characters, and superfluous whitespaces, and to change all text to lowercase [
18].
Next, the refined corpus was split into four documents of randomized publication titles depending on the originating country’s income level (high, upper middle, lower middle, and low) and imported to Voyant Tools software, which is a free, web-based collection of instruments for digital text analysis [
19]. This study used the instrument called Topics, which is based on a latent Dirichlet allocation machine learning algorithm. Latent Dirichlet allocation is a generative model that assumes that every document is represented by a random mixture of latent topics, where each latent topic is characterized by a probability distribution of terms [
20]. The Topics instrument randomly assigns terms to a user-specified number of topics (term clusters) and goes through a number of iterations to refine the model and determine which terms are best suited to which topics based on co-occurrence in the documents. The Topics tool was configured to generate 15 topics, each containing 10 terms.
Before beginning the topic modeling, the program’s list of stop words (those with little to no analytical value) was manually supplemented with additional common, health-related terms (e.g., care, health, medical, patient, and treatment) and terms associated with research methodology (e.g., analysis, cohort, controlled, prospective, systematic). The final stop-word list is provided in
Suppl. 1.
Descriptive statistics summarized key findings, linear regression analysis modeled and predicted temporal changes in publication counts, Kruskal-Wallis tests compared data on research output between country income levels, and the Spearman correlation coefficient measured the relationship between academic productivity, research-impact indicators and national socioeconomic data. For all statistical analyses, a two-sided P<0.05 defined statistical significance. Data were analyzed using IBM SPSS ver. 26.0 (IBM Corp).
RESULTS
The search found 143 journals, of which 109 (listed in
Suppl. 2) had impact indicators (CiteScore and journal quartile) and were used for publication-data retrieval, including 28, 30, 27, and 24 journals from the first, second, third, and fourth quartiles, respectively.
Within the studied period, the journals published 162,743 papers. After excluding 31 document records with empty title fields, 374 records with non-English titles, and 7,880 records without data on the first author’s primary national affiliation, 154,458 publication records were included in the analysis. Of the documents, 85,395 (55.3%), 40,758 (26.4%), 14,274 (9.2%), and 14,031 (9.1%) were published in first-, second-, third-, and fourth-quartile journals respectively.
The papers were originally published in 18 different languages, mostly English (n=138,626, 89.7%). Other languages included German (n=5,496, 3.6%), French (n=2,571, 1.7%), Russian (n=2,870, 1.9%), Chinese (n=2,852, 1.8%), Spanish (n=1,142, 0.7%), Ukrainian (n=433, 0.3%), Turkish (n=324, 0.2%), Czech (n=61, <0.1%), Persian (n=32, <0.1%), Arabic (n=30, <0.1%), Japanese (n=14, <0.1%), Indonesian (n=2, <0.1%), Catalan, Danish, Portuguese, Slovak, and Slovenian (each n=1, <0.1%).
The publications received 1,817,635 citations (median, 3; interquartile range, 0–12; range, 0–3,039), including 1,396,459 (76.8%), 360,338 (19.8%), 49,404 (2.7%), and 11,434 (0.6%) citations of articles published by journals in the first, second, third, and fourth quartiles, respectively. Approximately 72.8% of the publications (n=112,508) had at least one citation.
The articles originated from 177 countries (81.6% of 217 countries in the World Bank list), including 67 of 86 (77.9%) high-income countries (HICs), 48 of 54 (88.9%) upper-middle-income countries (UMICs), 39 of 51 (76.5%) lower-middle-income countries (LMICs), and 23 of 26 (88.5%) low-income countries (LICs).
By world region, the number of publications and citations was highest for North America (37.0% and 44.7%, respectively) and Europe and Central Asia (35.9% and 33.8%, respectively), followed by East Asia and Pacific (18.4% and 15.4%, respectively), Middle East and North Africa (4.4% and 3.1%, respectively), South Asia (2.3% and 1.4%, respectively), Latin America and the Caribbean (1.0% and 0.9%, respectively), and Sub-Saharan Africa (1.0% and 0.7%, respectively). The count of publications by country was the highest for the United States, followed by the United Kingdom, Germany, China, Australia, Turkiye, France, Canada, Italy, and Iran (
Table 1). The top 10 countries by citation count were the United States, the United Kingdom, Australia, Canada, Germany, Italy, China, Turkiye, the Netherlands, and France (
Table 2).
Fig. 1 breaks down publications and citations according to world region, along with the top 10 countries with the highest counts per region.
Linear regression analysis produced an increment of the annual number of publications of 431.9 per year for all countries (95% confidence interval [CI], 378.3–485.4; R
2=0.941; P<0.001), 311.0 per year for HICs (95% CI, 271.8–350.2; R
2=0.939; P<0.001), 93.1 per year for UMICs (95% CI, 76.1–110.2; R
2=0.880; P<0.001), 25.0 per year for LMICs (95% CI, 21.5–28.5; R
2=0.924; P<0.001), and 2.7 per year for LICs (95% CI, 1.9–3.6; R
2=0.712; P<0.001) (
Fig. 2). According to the regression model, the annual count of papers published in the EM journals in the year 2030 will reach approximately 14,386 for all countries, 11,059 for HICs, 2,610 for UMICs, and 655 for LMICs, but would not exceed 62 for LICs.
Table 3 compares data on publications, citations, and potential publication impact (as reflected by cumulative CiteScore and journal quartile) between country income levels (the corresponding data for individual countries are available in the dataset [
21]). HICs consistently outperformed countries of other income levels in all indicators. Measures calculated per country did not show considerable differences between UMICs, LMICs, and LICs. A comparison of counts per million population indicated that LIC publication productivity, citations, and potential publications’ impact were significantly lower than that of UMICs and similar to that of LMICs.
A correlation analysis of national data showed a moderate to very strong positive association between all publication, citation, and potential publication-impact indicators (
Suppl. 3). Absolute country-level counts of publications, citations, publications in first-quartile journals and cumulative CiteScore correlated moderately with total country population. The same indicators calculated per million population additionally had a moderate to strong relationship with the percentage of the population aged 65 years and older (
Suppl. 3). All publication, citation, and potential publication-impact indicators were positively associated with the percentage of urban population and negatively associated with the percentage of rural population. The indicators had no significant relationship or a weak positive correlation with crude death rate and a weak to strong negative correlation with crude birth rate and mortality indicators, except the proportion of deaths from noncommunicable diseases, which exhibited a moderate positive association. All research-output indicators had a significantly positive relationship with life expectancy at birth and major economic indicators, including GDP, adjusted net national income, and current health expenditure (
Table 4,
Suppl. 3).
The title-based topic modeling encompassed 2,036,064 total words (HICs, n=1,621,364; UMICs, n=335,481; LMICs, n=72,920; LICs, n=6,299) and 46,321 unique word forms.
Table 5 lists 15 generated topics and their distribution depending on country income level. Publications by authors from HICs paid considerable attention to cardiac arrest and cardiopulmonary resuscitation (topics 9 and 13; 24.6% of the research output, compared with 5.2%, 3.8%, and 3.4% for UMICs, LMICs, and LICs, respectively). Articles from UMICs commonly dealt with acute injury in animal models (topic 12) and skeletal trauma (topics 10, 14, and 15), whereas LMIC papers focused more on the latter theme (46.0% compared with 11.8%, 27.5%, and 15.8% for HICs, UMICs, and LICs, respectively). For LICs, the research demonstrated a notable shift toward injuries and pain in emergency hospital units (topic 7) and investigations of factors and outcomes of emergencies on a national level (topic 8).
DISCUSSION
In this study, a comparative quantitative evaluation of research productivity in the EM field depending on the originator country’s income level was carried out. To the best of the author’s knowledge, the representativeness of the EM literature pool in the current paper is superior to that of any previous bibliometric investigation of the EM domain. This was achieved by applying one of the most comprehensive literature databases, covering journals of various impact levels and including research published in diverse languages and articles of different types, and less commonly analyzed products, such as notes and commentaries. In addition to ensuring broad coverage, this helped avoid biasing the results toward most frequently cited articles published in high-rating journals that primarily originate from the developed world, which is a common case with bibliometric research in EM [
5–
8].
In line with previous reports [
3,
4,
11], the current investigation found that the global body of EM research literature is growing steadily, indicating general progress and a promising outlook for this domain. Annual counts of the publications increased continuously over the studied period and are predicted to grow by 25% by the year 2030 compared with 2023.
For the most part, this progress is maintained by HICs (primarily the United States). They greatly outstrip countries of other income levels in the total publication and citation counts. First authors from HICs published their work much more often in high-impact journals, including first-quartile journals, which accounted for almost 80% of all citations for the EM publications. The dominance of HICs in the number of EM publications and their citations is consistent with observations made previously [
3,
10,
22].
An analysis of relationships between major national socioeconomic indicators and research output confirmed a positive association between the level of economic development and well-being of the country and its academic productivity in the EM field. A younger population, higher percentages of public living in rural areas, higher birth rates, higher mortality rates from injuries and communicable diseases, lower life expectancies, and lower current health expenditures, GDP and adjusted net national income, all of which are typical of less-developed countries, were significantly correlated with lower weighted national publication, citation, and potential publication-impact indicators. A positive relationship between academic output and GDP has been previously shown by bibliometric studies of other fields of medical research [
23–
25].
This study found that the gap in scientific productivity for EM between HICs and LICs is vast. For example, the number of publications per million population per country for HICs was 11, 41, and 72 times higher than that of UMICs, LMICs, and LICs, respectively. The order of magnitude of the discrepancies was even more pronounced for citations per million population per country that reflects the actual academic impact of the research (21, 54, and 171 times higher for HICs compared with UMICs, LMICs, and LICs, respectively), and for the weighted counts of papers published in first-quartile journals (24, 83, and 166 times, respectively) and CiteScore (20, 55, and 106 times, respectively), which represent potential research impact. As for the forecast, the annual number of publications in the coming years is expected to continue to rise significantly for HICs, less so for UMICs, far less for LMICs, and not at all for LICs.
Compared with HICs, countries with lower incomes have a disproportionately higher burden of acute diseases and injuries but considerably less emergency care use and worse outcomes from the emergencies [
26,
27]. While EM is an integral component of healthcare of any nation, LICs and LMICs commonly lack organized emergency medical services, have insufficient funding for such services, shortages of basic materials and emergency personnel and first responders, and a deficit of specialty training in emergency care [
12,
13,
16,
28,
29]. Despite global progress in EM, translation of emergency care guidelines and practices from HICs to countries with lower incomes may be limited because of the specifics of demography; differences in epidemiology of emergency conditions; and resources, organization, and performance of emergency systems.
Automated topic modeling analysis revealed pronounced diversity in the distribution of research themes for countries of different income levels. In contrast to HICs, which pay relatively large amounts of attention to cardiopulmonary resuscitation, LICs were significantly more focused on skeletal trauma and other injuries, which puts a large burden on healthcare systems in the developing world [
27]. These findings suggest that countries with different income levels have distinct research priorities that explain some of the dissimilarities among countries. Consequently, there is a need to plan and potentiate EM research at the country level to develop cost-effective, context-specific, and locally relevant interventions aimed at reducing morbidity and mortality from acute diseases and injuries. For now, however, the low counts of EM publications and article citations from LMICs and LICs reflect their limited ability to contribute to the field and the relatively poor prospects for improvement.
Limited EM research productivity among LICs can be attributed to a range of factors, including the more recent establishment of EM as a separate discipline and specialty, inadequate research infrastructure and human resources, poor research mentorship opportunities, and inadequate access to training to develop research skills compared with the resources available in developed nations [
22]. Possible avenues for improvement include enhancing the pool of local researchers, mentored research training, regional and global collaboration to support education and mentoring, the establishment of leadership at academic institutions, and introduction of procedures for standardized collection of reproducible data on EM performance and outcomes [
22]. These organizational measures require stable financial support from national stakeholders and HIC donors. Funding allocation, in turn, demands careful research prioritization and the development of sound national research agendas that consider disease burden, system structure, resource availability, and existent research gaps. Measuring research productivity and mapping scientific knowledge at a national level by applying a bibliometric approach with topic modeling analysis of research output can provide essential insights for managing a country’s EM scientific agenda and supporting decision-making aimed at improving outcomes in emergency care.
This study has several limitations. First, the representativeness of the studied sample of literature may be limited due to the exclusion of document records without information on the national affiliation of the first author and records with non-English titles that had no English translation. Second, although the study identified more than 100 journals indexed in one of the most comprehensive and authoritative literature databases, the analysis did not involve Scopus-indexed journals without impact indicators and unindexed journals, leading to the inevitable omission of potentially relevant papers. Third, the journal search was restricted to the EM subject area defined by Scopus. This likely led to the omission of EM articles published in journals classified within other domains such as general medicine, surgery, cardiology, or respiratory medicine. Conversely, the analyzed journals primarily assigned by Scopus to EM may be multidisciplinary, and this could have resulted in the inclusion of publications unrelated to EM. Another limitation arises from the assumption that the first authors, whose affiliation formed a basis for the assignment of publications to specific countries, were also the lead researchers. Although this approach is common in bibliometric research [
30,
31], it may have introduced bias to the distribution of publications by country. Furthermore, the lead author’s institutional affiliation may not be an accurate reflection of the author’s actual country of origin. Finally, it should be noted that the generated research topics may have had partial thematic overlap, which is an unavoidable effect of the random assignment of terms to term clusters using the automated topic modeling methodology based on the latent Dirichlet allocation machine learning algorithm. Despite these limitations, the study offers important new insights into academic productivity in the EM field and differences produced by country income level.
Over the past two decades, global productivity in EM research has shown a constant and rapid rise. Although the publications originated from the majority of world countries, the quantitative academic contribution of LICs was markedly less than that of HICs and is predicted to remain negligible for the least economically developed countries in the foreseeable future. A country’s academic progress in EM was strongly associated with its socioeconomic well-being, and the priority trends of the key EM research themes differ by national income level. This suggests the importance of developing country-specific research agendas that can enhance and tailor productivity in EM research at the national level and develop context-appropriate interventions aimed at improving outcomes of acutely ill and injured patients.
NOTES
-
Conflicts of interest
The author has no conflicts of interest to declare.
-
Funding
The author received no financial support for this study.
-
Data availability
The data analyzed in this study are available from the Mendeley Data repository (https://doi.org/10.17632/ns8nrjdtyp.1).
Supplementary materials
Fig. 1.Emergency medicine (A) publication count and (B) citation count by world region, including names of the most productive countries per region. Each bar represents the total number of publications or citations per world region broken down into segments by country. The length of a segment is proportional to a country’s contribution to the total number of publications or citations in the region.
Fig. 2.Yearly distribution of the publications in emergency medicine journals (total counts and counts by country income level) with linear regression estimates. HIC, high-income country; UMIC, upper-middle-income country; LMIC, lower-middle-income country; LIC, low-income country.
Table 1.Top 10 countries by publication (n=154,458)
Table 1.
|
Rank |
Country |
No. of publications (%) |
|
1 |
USA |
51,498 (33.3) |
|
2 |
UK |
10,286 (6.7) |
|
3 |
Germany |
8,528 (5.5) |
|
4 |
China |
8,496 (5.5) |
|
5 |
Australia |
6,716 (4.3) |
|
6 |
Turkiye |
5,758 (3.7) |
|
7 |
France |
5,705 (3.7) |
|
8 |
Canada |
5,686 (3.7) |
|
9 |
Italy |
4,997 (3.2) |
|
10 |
Iran |
3,374 (2.2) |
Table 2.Top 10 countries by citation count (n=1,817,635)
Table 2.
|
Rank |
Country |
No. of citations (%) |
|
1 |
USA |
730,658 (40.2) |
|
2 |
UK |
155,548 (8.6) |
|
3 |
Australia |
95,056 (5.2) |
|
4 |
Canada |
82,161 (4.5) |
|
5 |
Germany |
77,757 (4.3) |
|
6 |
Italy |
69,415 (3.8) |
|
7 |
China |
58,667 (3.2) |
|
8 |
Turkiye |
45,345 (2.5) |
|
9 |
The Netherlands |
45,107 (2.5) |
|
10 |
France |
30,702 (1.7) |
Table 3.Characteristics of emergency medicine research output by country income level
Table 3.
|
Characteristic |
HIC |
UMIC |
LMIC |
LIC |
|
No. of publications (n=154,458) |
125,656 (81.4) |
23,158 (15.0) |
5,258 (3.4) |
386 (0.2) |
|
Per country |
226.0 (13.0–1,049.0)a)
|
10.0 (3.0–144.5)b)
|
11.0 (3.0–77.0)b)
|
7.0 (2.0–21.0)b)
|
|
Per million population per country |
28.9 (11.2–107.5)a)
|
2.7 (0.7–6.2)b)
|
0.7 (0.3–2.8)b),c)
|
0.4 (0.2–0.8)c)
|
|
No. of citations (n=1,817,635) |
1,606,285 (88.4) |
171,785 (9.5) |
37,506 (2.1) |
2,059 (0.1) |
|
Per country |
1,748.0 (97.0–15,856.0)a)
|
58.5 (10.5–774.2)b)
|
61.0 (6.0–663.0)b)
|
21.0 (4.0–69.0)b)
|
|
Per million population per country |
222.9 (61.0–1,224.7)a)
|
10.4 (1.9–44.8)b)
|
4.1 (0.6–13.7)b),c)
|
1.3 (0.3–2.9)c)
|
|
No. of publications in first-quartile journals (% of publications from countries of the income level) |
73,192 (58.2) |
9,846 (42.5) |
2,245 (42.7) |
112 (29.0) |
|
Per country |
83.0 (8.0–709.0)a)
|
4.0 (1.0–38.5)b)
|
4.0 (0–25.0)b)
|
3.0 (1.0–6.0)b)
|
|
Per million population per country |
16.6 (4.6–61.4)a)
|
0.7 (0.2–3.3)b)
|
0.2 (0–0.9)b),c)
|
0.1 (0.0–0.3)c)
|
|
CiteScore |
567,144.1 |
71,397.8 |
16,459 |
1,194.9 |
|
Per country |
715.5 (64.1–5,162.5)a)
|
29.8 (9.4–443.4)b)
|
33.6 (7.3–242.8)b)
|
26.4 (6.1–53.2)b)
|
|
Per million population per country |
116.3 (33.3–502.7)a)
|
5.8 (2.5–20.3)b)
|
2.1 (0.6–8.9)b),c)
|
1.1 (0.4–3.0)c)
|
Table 4.Heatmap of correlation of country-level statistics
Table 4.
|
Variable |
Spearman rank correlation coefficient |
|
Publications per country |
Citations per country |
Publications in the first-quartile journals per country |
CiteScore total per country |
|
2023 Total population |
0.540 |
0.496 |
0.480 |
0.502 |
|
2023 Population ages ≥65 yr (% of total population) |
0.407 |
0.441 |
0.453 |
0.432 |
|
2023 Rural population (% of total population) |
–0.369 |
–0.401 |
–0.367 |
–0.377 |
|
2023 Urban population (% of total population) |
0.369 |
0.401 |
0.367 |
0.377 |
|
2022 Crude birth rate (per 1,000 people) |
–0.396 |
–0.438 |
–0.453 |
–0.429 |
|
2019 Cause of death, noncommunicable diseases (% of total) |
0.429 |
0.430 |
0.461 |
0.444 |
|
2019 Mortality (%) |
|
|
|
|
|
Road traffic injury (per 100,000 people) |
–0.479 |
–0.475 |
–0.504 |
–0.495 |
|
CVD, cancer, diabetes, or CRD, ages 30–70 yr |
–0.494 |
–0.551 |
–0.561 |
–0.538 |
|
2022 Total life expectancy at birth (yr) |
0.482 |
0.538 |
0.535 |
0.514 |
|
2023 GDP (US$) |
0.851 |
0.850 |
0.820 |
0.840 |
|
2021 Adjusted net national income (US$) |
0.849 |
0.843 |
0.810 |
0.834 |
|
2021 Current health expenditure per capita (US$) |
0.502 |
0.530 |
0.547 |
0.533 |
Table 5.Research topics generated from the emergency medicine publications’ titles, and percentage distributions of the topics across the totalities of publications from countries of different income levels
Table 5.
|
No. |
Term clustersa)
|
Topic interpretation |
Topic distribution across publications (%) |
|
HIC |
UMIC |
LMIC |
LIC |
|
1 |
Outcomes, pediatric, children, mortality, score, low, injuries, syndrome, secondary, intensive |
Pediatric outcomes, mortality, scores, injuries, syndromes, and intensive care |
5.5 |
7.0 |
7.8 |
7.2 |
|
2 |
Cardiopulmonary, pediatric, emergency, trauma, covid, validation, myocardial, life, caused, resuscitation |
Pediatric emergencies, cardiopulmonary resuscitation, trauma, COVID-19, validation, and myocardial involvement |
7.3 |
3.8 |
3.7 |
3.4 |
|
3 |
Injury, burn, trauma, wound, acute, tissue, treated, approach, role, fluid |
Injuries from burns and trauma, wounds, acute tissue involvement, treatment, approaches |
5.3 |
10.0 |
8.0 |
2.5 |
|
4 |
Emergency, trauma, abdominal, CPR, fracture, young, evaluation, presenting, department, outcome |
Emergency trauma, including abdominal injuries and fractures, CPR, young patients, evaluation, and outcomes |
7.7 |
2.7 |
5.3 |
2.6 |
|
5 |
Department, following, emergency, early, critical, trauma, head, hip, training, impact |
Emergency department response to critical conditions, involving head and hip injuries, training, impact |
6.9 |
2.8 |
5.9 |
4.6 |
|
6 |
Cardiac, pandemic, emergency, different, diagnosis, covid, surgery, arrest, air, tool |
Cardiac emergencies during the COVID-19 pandemic, differences, diagnosis, surgery, tools |
6.5 |
4.0 |
4.1 |
1.3 |
|
7 |
Hospital, emergency, unit, pain, support, traumatic, outcomes, adult, injury, triage |
Emergency hospital units, pain, trauma and injuries, outcomes, adult patients, and triage |
7.3 |
6.0 |
5.2 |
20.5 |
|
8 |
Emergency, associated, factors, hospital, practice, outcome, national, characteristics, disaster, burn |
Associated factors, practices, and outcomes of emergencies, national characteristics, disasters, and burns |
6.9 |
3.6 |
4.0 |
24.1 |
|
9 |
Department, emergency, cardiac, injury, resuscitation, prehospital, acute, arrest, model, effects |
Emergency department and prehospital response to cardiac arrest and injuries, resuscitation, models, effects |
10.7 |
2.5 |
1.8 |
1.7 |
|
10 |
Fracture, femoral, rare, syndrome, experience, dislocation, fixation, following, acute, year |
Femoral fractures, dislocations, fixation, rare cases, syndromes, experience |
3.2 |
10.0 |
17.6 |
3.3 |
|
11 |
Emergency, risk, new, acute, cardiac, point, trauma, ultrasound, pediatric, impact |
Risks in emergencies, cardiac issues, point-of-care ultrasound in trauma, pediatric patients, impact |
9.1 |
2.9 |
3.1 |
4.4 |
|
12 |
Acute, effects, rats, injury, effect, induced, sepsis, model, factors, lung |
Acute injury effects induced in animal models, sepsis, lung involvement |
1.1 |
24.4 |
3.1 |
10.1 |
|
13 |
Emergency, arrest, department, cardiac, resuscitation, acute, pediatric, quality, syndrome, prehospital |
Emergency department management of cardiac arrest, resuscitation, pediatric patients, quality, syndromes, prehospital care |
13.9 |
2.7 |
2.0 |
1.7 |
|
14 |
Fractures, burn, trauma, burns, injuries, injury, emergency, outcome, unusual, hospital |
Fractures and burns, outcomes of traumatic injuries, unusual cases |
4.8 |
4.3 |
21.9 |
10.7 |
|
15 |
Acute, severe, injury, effect, traumatic, covid, fractures, infarction, technique, pain |
Severe traumatic injuries, effects, COVID-19 impact, fractures, infarction, techniques, and pain |
3.8 |
13.2 |
6.5 |
1.8 |
REFERENCES
- 1. Graham SM, Brennan C, Laubscher M, et al. Orthopaedic research in low-income countries: a bibliometric analysis of the current literature. SICOT J 2019;5:41.
- 2. Liu T, Li Y, Li J, Fan H, Cao C. Temporal trend and research focus of injury burden from 1998 to 2022: a bibliometric analysis. J Multidiscip Healthc 2023;16:1869-82.
- 3. Cetin M, Long B, Gottlieb M. A 10-year bibliometric analysis of publications in emergency medicine. Am J Emerg Med 2022;58:215-22.
- 4. Golfiruzi S, Nouri M, Sheikhshoaei F, et al. Mapping global research in emergency medicine; a bibliometric analysis of documents indexed in the Web of Science database. Arch Acad Emerg Med 2023;11:e53.
- 5. Tsai YL, Lee CC, Chen SC, Yen ZS. Top-cited articles in emergency medicine. Am J Emerg Med 2006;24:647-54.
- 6. Li Q, Jiang Y, Zhang M. National representation in the emergency medicine literature: a bibliometric analysis of highly cited journals. Am J Emerg Med 2012;30:1530-4.
- 7. Shuaib W, Acevedo JN, Khan MS, Santiago LJ, Gaeta TJ. The top 100 cited articles published in emergency medicine journals. Am J Emerg Med 2015;33:1066-71.
- 8. Shao J, Zheng X. The 100 top-cited articles published in emergency medicine journals: a bibliometric analysis. Hong Kong J Emerg Med 2016;23:329-39.
- 9. Kokulu K, Mutlu H, Sert ET. Scientific publication productivity of emergency physicians: a bibliometric analysis of the last decade. J Emerg Med 2019;57:13-20.
- 10. Xu L, Tang F, Wang Y, et al. Research progress of pre-hospital emergency during 2000-2020: a bibliometric analysis. Am J Transl Res 2021;13:1109-24.
- 11. Can D. Bibliometric analysis of emergency medicine. Era’s J Med Res 2022;9:174-87.
- 12. Suryanto , Plummer V, Boyle M. EMS systems in lower-middle income countries: a literature review. Prehosp Disaster Med 2017;32:64-70.
- 13. Kironji AG, Hodkinson P, de Ramirez SS, et al. Identifying barriers for out of hospital emergency care in low and low-middle income countries: a systematic review. BMC Health Serv Res 2018;18:291.
- 14. Mahajan P, Visclosky T, Bhoi S, Galwankar S, Kuppermann N, Neumar R. The importance of developing global emergency medicine research network. Am J Emerg Med 2019;37:744-5.
- 15. Lecky FE, Reynolds T, Otesile O, et al. Harnessing inter-disciplinary collaboration to improve emergency care in low- and middle-income countries (LMICs): results of research prioritisation setting exercise. BMC Emerg Med 2020;20:68.
- 16. Bhattarai HK, Bhusal S, Barone-Adesi F, Hubloue I. Prehospital emergency care in low- and middle-income countries: a systematic review. Prehosp Disaster Med 2023;38:495-512.
- 17. Birkun AA. Automated topic modelling of emergency medicine research outputs. Am J Emerg Med 2025;91:179-81.
- 18. OpenRefine [Internet]. OpenRefine; 2025 [cited 2025 Jun 28]. Available from: https://openrefine.org
- 19. Voyant Tools [Internet]. Voyant; 2025 [cited 2025 Jun 28]. Available from: http://voyant-tools.org
- 20. Jelodar H, Wang Y, Yuan C, et al. Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey. Multimed Tools Appl 2019;78:15169-211.
- 21. Birkun A. Dataset of global academic productivity in the emergency medicine field (2004-2024) [Internet]. Version 1. Mendeley Data; 2025 Available from: https://doi.org/10.17632/ns8nrjdtyp.1
- 22. Aluisio AR, Waheed S, Cameron P, et al. Clinical emergency care research in low-income and middle-income countries: opportunities and challenges. BMJ Glob Health 2019;4:e001289.
- 23. Clarke A, Gatineau M, Grimaud O, et al. A bibliometric overview of public health research in Europe. Eur J Public Health 2007;17 Suppl 1:43-9.
- 24. Guerin MB, Flynn TH, Brady J, O'Brien CJ. Worldwide geographical distribution of ophthalmology publications. Int Ophthalmol 2009;29:511-6.
- 25. Liang Z, Luo X, Gong F, et al. Worldwide research productivity in the field of arthroscopy: a bibliometric analysis. Arthroscopy 2015;31:1452-7.
- 26. Chang CY, Abujaber S, Reynolds TA, Camargo CA, Obermeyer Z. Burden of emergency conditions and emergency care usage: new estimates from 40 countries. Emerg Med J 2016;33:794-800.
- 27. Delaney PG, Eisner ZJ, Geduld H. The emergency burden in low and middle-income countries. Surgery 2024;176:528-30.
- 28. Obermeyer Z, Abujaber S, Makar M, et al. Emergency care in 59 low- and middle-income countries: a systematic review. Bull World Health Organ 2015;93:577-86.
- 29. Patiño AM, Chen J, DeVos EL, et al. Emergency medicine around the world: analysis of the 2019 American College of Emergency Physicians International Ambassador country reports. J Am Coll Emerg Physicians Open 2022;3:e12681.
- 30. Allen LN, Fox N, Ambrose A. Quantifying research output on poverty and non-communicable disease behavioural risk factors in low-income and lower middle-income countries: a bibliometric analysis. BMJ Open 2017;7:e014715.
- 31. English KM, Pourbohloul B. Increasing health policy and systems research capacity in low- and middle-income countries: results from a bibliometric analysis. Health Res Policy Syst 2017;15:64.
