Blog

How does one identify something no one has a test for? Designer drugs replicate the effects of known, illicit drugs but evade law enforcement. The chemical structure variations that help these compounds avoid detection also make them unpredictable in the body – a quality that poses serious health consequences.

Now, a research team has used computer modeling to create a database of predicted chemical structures for improved detection of designer drugs.

Jason Liang, a rising senior in the Science, Mathematics and Computer Science Magnet Program at Montgomery Blair High School, will present the team’s results at the fall meeting of the American Chemical Society (ACS). ACS Fall 2025 is being held Aug. 17-21; it features about 9,000 presentations on a range of science topics.

This library of computationally generated metabolic signatures and mass spectra, which we’re calling the Drugs of Abuse Metabolite Database [DAMD], could lead to more thorough detection of new psychoactive substances and more accurate surveillance of designer drug usage.”

Jason Liang, Senior, Science, Mathematics and Computer Science Magnet Program, Montgomery Blair High School

An illicit drug that can be misused is usually identified by its chemical “fingerprint” called a mass spectrum. This fingerprint is a pattern created by the shape, weight and makeup of the drug molecule. When a person’s urine is screened for drugs, a technician uses a technique called mass spectrometry to acquire and compare spectra from molecules in the sample to catalogues of spectra for known drugs and their metabolites (small molecules created when the body breaks down a drug). However, new psychoactive substances and their metabolites don’t usually have matches in existing databases.

“It’s a bit of a chicken and the egg problem,” says Liang’s mentor Tytus Mak, a statistician and data scientist with the mass spectrometry center at the National Institute of Standards and Technology (NIST). “How do you know what this new drug is if you’ve never measured it, and how do you measure it if you don’t know what you’re looking for? Using a computational prediction methodology could help us find a solution.”

The idea to develop DAMD started with Mak and Hani Habra, a former postdoc at NIST and a current bioinformatician at Michigan State University. They thought that computer modeling could keep up with the seemingly endless iterations of unknown synthetic compounds burdening health care systems and challenging drug surveillance initiatives. Then, in the summer of 2024, Mak and Habra approached Liang about working with them.

“Building a predicted mass-spectral library requires strong programming skills and a solid foundation in chemistry – both of which align well with my background,” says Liang. “After learning about the devastating number of overdose deaths, including cases within the local community, I was eager to work on this project that could potentially help people.”

As a starting point, the researchers used the mass-spectral database maintained by the U.S. Drug Enforcement Administration-chaired Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG). This database provides reliable mass spectra for the identification of more than 2,000 drugs confiscated by law enforcement. Then, using computational approaches, Habra, Liang and Mak predicted nearly 20,000 chemical structures and corresponding mass-spectral fingerprints for possible metabolites of SWGDRUG substances and their metabolites.

The team is currently validating their predicted mass spectra by matching them to real spectra from datasets of human urine analyses. These datasets are catalogs of spectra from all detectable substances found in human urine samples. Finding a match, or something close to a match, in these datasets “tells us if the chemical structures and spectra our algorithms are producing are plausible,” says Habra. Subsequently, the team will compare already-collected, real-world data to DAMD, showing a proof-of concept for forensic toxicology.

DAMD could someday be a publicly available supplement to the current illicit drug mass-spectral databases, making it easier to detect and identify evidence of drug use in human urine samples. One of its primary applications is to develop a system to help people get the medical interventions they need.

“Someone could have ingested a substance that, unbeknownst to them, was cut with a fentanyl derivative,” says Mak. “Using DAMD, a doctor could see metabolites from the person’s toxicological report that are strong candidates for a fentanyl-like drug and inform their treatment plan.”

Demographic profiles of respondents

The demographic characteristics of the study participants are demonstrated in Table 1.

All food handlers surveyed were aged between 20 and 55 years old; however, most of them (55%) were of middle age (30 and 49 years old). Most of the respondents (94.2%) were females. Approximately a third of food handlers (34%) interviewed had education at primary level. Over a half of food handlers (52%) were literate at the secondary education level. Among food handlers surveyed, 8.5% reported having vocational training and only 6% of respondents possess university degrees. Of these 358 food handlers, 17% performed duties as cooks, 30% were involved as kitchen assistants, 31% were food servers, 14% were part of the cleaning staff, and 8% were managers. A large proportion of participants (82.7%) had no or expired medical examination cards. No significant differences in food safety knowledge and attitudes were identified based on the gender, age, educational level, and work experience of the food handlers. This aligns with a study conducted by Abdul-Mutalib, Abdul-Rashid [18], which similarly found no significant association between the knowledge level of respondents and their socio-demographic characteristics. In contrast, McIntyre, Vallaster [19] reported that socio-demographic factors, such as the length of work experience in the food industry and the educational level of food handlers, were significantly linked to enhanced food safety knowledge. Further studies are necessary to gain a more comprehensive understanding of the impact of socio-demographic factors on the food safety knowledge and attitudes of food handlers. The majority of participants in the present study (94%) were females. In the Kazakhstani context, the prevalence of women working in restaurants is not unexpected, given that this demographic is traditionally seen as primarily responsible for household food preparation activities. In line with this cultural norm, the female labor force tends to dominate the food service sector. The substantial representation of female handlers in our study aligns with findings from similar previous studies [20,21,22].

More than two-thirds of respondents (69%) had 6 months to 2 years of experience in the food industry. In this study, the cumulative experience of food handlers in the food service industry was as follows: less than 1 year (44%), followed by 1–2 years (28%), 2–4 years (23%), and more than 4 years (5%). The limited diversity in food handling experience raises the concern of an elevated risk of food contamination, particularly due to the potential lack of adherence to good manufacturing practices. Approximately 34% of the food handlers had not advanced beyond primary education, and of this group, over 60% were female. Considering that women engage in diverse food handling activities both at work and at home, it becomes crucial to consider their educational background and literacy level when developing and implementing food safety training programs. Previous studies have demonstrated that longer job tenure is often associated with improved food safety practices and reduced risk of contamination. For example, food handlers with over 10 years of experience exhibited significantly lower rates of bacterial contamination compared to less experienced workers [23]. However, job experience alone does not ensure safe practices unless reinforced by structured training. Evidence suggests that formal education in food safety substantially enhances compliance with hygiene protocols and reduces critical violations during inspections [24, 25]. Therefore, both accumulated practical experience and regular food safety training are essential to ensuring safe food handling and minimizing contamination risks.

Food safety knowledge of respondents

Evaluation of food handlers’ food safety knowledge is illustrated in Table 2. Over half (57%) of the surveyed food handlers had not participated in any food safety training programs. This gap is concerning, as previous studies have demonstrated that formal training significantly enhances food safety knowledge and practices [26, 27]. The lack of training may explain the mixed levels of understanding observed in this study. For instance, while 80% of respondents were aware that improper food handling could cause foodborne illnesses, the majority (74%) relied on personal experience as their primary source of knowledge. This reliance on informal sources could lead to incomplete or inaccurate understanding of food safety principles. Only 19% of the food handlers reported receiving food safety knowledge through formal training, 9% from public health officials, and a mere 1% from media. The disparity in knowledge sources suggests that more structured and accessible training programs are needed to ensure comprehensive food safety education. Research by Seaman and Eves [28] highlights that well-designed training programs can effectively improve food handlers’knowledge and compliance with safety practices. Moreover, other studies have also demonstrated that theoretical and practical training contributes to improving hygiene conditions in establishments and changing the behavior of food handlers [25, 29,30,31]. The study also highlighted specific gaps in knowledge. For example, 40% of participants recognized expired food as a primary source of foodborne infection, while nearly half (48%) acknowledged that multiple factors, including uncooked, expired, and improperly stored foods, contribute to foodborne illnesses. This indicates a partial understanding that could lead to inconsistent application of food safety measures. Moreover, while 70% of respondents correctly identified the importance of hygiene, sanitizing food contact surfaces, and rinsing raw foods and utensils in reducing contamination risks, a significant portion (69%) failed to recognize all contributing factors to microorganism proliferation, such as lack of cleaning in the cooking area, unsanitized utensils, not wearing protective clothing, and poor food handling. This limited knowledge is consistent with findings from other studies, such as those by Green and Selman [32], which suggest that food handlers often have limited awareness of all critical control points in food safety. Regarding microbial proliferation, only one-third (33%) of participants correctly identified that germs proliferate better in warmed foods. This lack of knowledge about temperature control is particularly concerning, as proper temperature management is crucial for preventing foodborne illnesses [33].

Overall, the study underscores the necessity for comprehensive and regular food safety training for food handlers. Structured training programs, supported by public health initiatives and accessible through various media, can bridge the knowledge gaps identified. By improving education on all aspects of food safety, including cross-contamination, proper storage, and temperature control, we can significantly reduce the risk of foodborne illnesses. Enhanced training and continuous education are vital for ensuring that food handlers are well-equipped with the necessary knowledge and skills to maintain high standards of food safety.

Food safety self-reported attitudes of respondents

It is considered that food safety attitudes are an essential factor of food safety, specifically in preventing hazards of foodborne related diseases [34]. Data obtained from food handlers responses related to their food safety attitudes are demonstrated in Table 3. Of the study population, a large portion of respondents (71%) reported that ready-to-eat food must be maintained at 60℃ and above. Seventy-four percent reported that prepared food should be kept in refrigerator below 4 ℃. More than two-thirds of respondents (68%) were certain about possible germ proliferation in foods prepared in advance. The present study indicates that nearly 30% of the food handlers’ attitudes toward food contamination was found to be fair. Similar findings have been reported in prior studies, highlighting a deficiency in the knowledge of food handlers regarding cross-contamination and sanitization [35, 36]. These results suggest that food handlers may not be fully acquainted with the essential procedures required for proper sanitation and the prevention of cross-contamination. Osaili, Obeidat [35] assert that”food handling mistakes during food preparation have been associated with most incidences of foodborne diseases.”Given the incorrect execution of these operations, the likelihood of foodborne diseases increases. This may also indicate inadequate or a lack of training specifically addressing cross contamination.

Our findings demonstrated that overall knowledge among study respondents is consistent with those studies described by Lee, Halim [21], where most respondents were aware that they should wear protective clothing and not come to work if feeling sick. An overwhelming majority of participants (87%) reported that raw and cooked food should be kept separately to avoid cross-contamination. Almost all respondents (99%) stated that cleaning solutions and toxic chemicals should not be stored in close proximity to the food preparation area. Most of food handlers (61%) accurately believed that it is necessary to monitor the temperatures of refrigerators and freezers on a regular basis. Regarding self-hygiene, most participants (74%) knew the importance of hand washing after using the toilet. This aligns with the findings of Ansari-Lari, Soodbakhsh [37], which demonstrated an adequate level of knowledge regarding general sanitary measures among food handlers. However, one-third (33%) of the respondents provided incorrect responses concerning the procedures for reheating and preparation of food. It has become a common practice in food service establishments, particularly in self-service restaurants, to prepare food in advance. While some establishments employ water baths to maintain food temperature and inhibit the proliferation of harmful microorganisms, others leave dishes on shelves or tables for extended periods. This practice can result in food contamination and the proliferation of microorganisms [18].

Food safety self-reported practices of respondents

The responses of food handlers towards Food Safety Practices are outlined in Table 4. A majority of the participants, comprising 60%, indicated that they engaged in hand hygiene practices prior to food preparation or service. Among food handlers, 52% specifically reported the consistent use of antibacterial soap for handwashing. In terms of food preparation practices, 61% of respondents emphasized the importance of cleaning utensils and equipment before commencing food preparation. Moreover, a significant proportion of food handlers, totaling 72%, adhered to the practice of employing separate utensils for handling raw and cooked food items. Approximately 84% of food handlers affirmed their practice of inspecting the expiration dates on food items before their utilization. In contrast, a notable 74% of food handlers were unaware of established guidelines for food thawing procedures, with a propensity to thaw frozen food items at room temperatures.

Vast majority of food handlers, constituting 73%, demonstrated a concerning trend of reporting to their workplaces despite being ill, which consequently resulted in unfavorable scores in this regard. In terms of the thawing process for frozen food items, a mere 3% of food handlers followed the recommended food safety practice of refraining from thawing foods at room temperature. Conversely, approximately 82% of workers correctly observed the practice of wearing uniforms in food preparation areas. However, only 34% of food handlers demonstrated the habit of regularly laundering their uniforms. Lastly, approximately 37% of food handlers adhered to the appropriate food safety protocol of employing tissues when coughing or sneezing, followed by immediate hand washing. This outcome aligns with findings from previous studies by Parry-Hanson Kunadu, Ofosu [38] and Sneed, Strohbehn [39], which also reported suboptimal food safety practices among food handlers.

In our study, the scores related to personal hygiene aspects, such as the washing of hands and utensils before engaging in food handling activities were lower than that in the previous studies [18, 34, 37]. Although the majority of food handlers in this study reported using soap when washing their hands, the effectiveness of hand hygiene is highly dependent on proper technique. According to the Centers for Disease Control and Prevention, effective handwashing involves five essential steps: wetting, lathering, scrubbing for at least 20 s, rinsing, and drying [40]. Simply using soap without adhering to these steps may leave harmful microorganisms on the hands. Several studies have shown that inadequate handwashing practices contribute significantly to cross-contamination in food service settings [41, 42]. Therefore, it is imperative that all food handlers consistently adhere to the practice of handwashing at each stage of food production, especially before directly handling the food.

Knowledge, attitudes, and practices scores among respondents

Food handlers achieved scores ranging from a minimum of 1 to a maximum of 8 out of 8 on the food safety knowledge section of the questionnaire. The average score for food safety knowledge was 4.76 ± 1.44 (52%), indicating a low level of knowledge, as it fell within the 50–80% threshold. The scores for food safety attitudes among respondents ranged from 1 to 7 out of a possible 7. The average score for food safety attitudes was 6.11 ± 2.12 (87%), which is considered a good score. The food handlers achieved food safety practices scores ranging from a minimum of 1 to a maximum of 8 out of 10. The mean score for food safety practices was 7.15 ± 2.21, which is classified as a low score (40–80%).

In a similar study, Dudeja, Singh [43]. found that food handlers interviewed had good scores in food safety knowledge, attitudes, and practices. Likewise, Al-Shabib, Mosilhey [34] reported satisfactory scores in their evaluation of the KAP attributes of food handlers in Saudi Arabia. A study conducted in Malaysia among food handlers also demonstrated satisfactory scores in food safety knowledge, attitudes, and practices [18]. In contrast, Nee SiowOi and Norrakiah Abdullah Sani [44] reported that while food handlers in Malaysia demonstrated good food safety knowledge and a solid understanding of hygiene importance, their knowledge scores regarding food storage practices were poor.

Association among attributes related to knowledge, attitudes, and practices of food handlers

In accordance with Akoglu [45], a correlation is considered moderate to strong when Spearman’s rho exceeds 0.700, represented as r_s > 0.700 (Table 5). Notably, there was a moderate to strong correlation identified between food safety knowledge and the food safety practices of food handlers, yielding an r_s value of 0.775. Furthermore, this correlation was statistically significant (p < 0.05). A Spearman correlation analysis was also conducted to examine the relationship between food safety attitude and food safety practices scores among food handlers. The findings revealed a strong correlation, with an r_s value of 0.733, and once again, this correlation was statistically significant (p < 0.05). This significant positive correlation implies that both the food safety knowledge and attitude of food handlers exert a discernible influence on their food safety practices. Similarly Ansari-Lari, Soodbakhsh [37] also reported a significant positive correlation between the food safety knowledge and food safety attitude of food handlers. Nevertheless, it is essential to note that imparting knowledge of personal hygiene alone does not suffice to ensure the accurate application of food safety knowledge, attitude and practice principles. Practical, hands-on training and the implementation of regular training programs are imperative in enhancing the overall state of food safety [18, 46].

The correlation between demographic factors and the attributes of food safety knowledge, attitudes, and practices

Chi-square test was performed to evaluate association between the demographic features of food handlers and their food safety KAP attributes. Notably, no statistically significant association was found between”age”and the food safety knowledge of food handlers (p = 0.061). However, it is important to highlight that all the other demographic characteristics demonstrated a statistically significant association (p < 0.05) with the food safety knowledge of food handlers, as indicated in Table 6. A chi-square test was also conducted to examine the association between the demographic factors of the participants and their food safety attitude. The findings indicated that”age”and”gender”exhibited no significant correlation with food safety attitude, as evidenced by p-values of p = 0.055 and p = 0.178, respectively.

In contrast, all other demographic characteristics, including level of education, professional category and current job tenure, displayed a statistically significant association (p < 0.05) with food safety attitude. Furthermore, the chi-square test revealed that, except for the”gender”of food handlers, all the other demographic attributes exhibited a significant association with their food safety practices.

The research findings revealed that there was no statistically significant link between age and the food safety knowledge of food handlers. Likewise, food safety attitude did not exhibit any notable correlation with the age of the food handlers. Among the various attributes related to KAP in food safety, only food safety practices displayed a significant connection with the age of the study participants.

Most of the food handlers (55%), within the age range of 20–33 years, demonstrated a significant association with food safety practices. Similarly, in Ireland, younger food handlers particularly those aged 26–35 years, exhibited a higher level of knowledge regarding standard food safety practices. This trend may be attributed to the increased exposure and dedication to learning food practices among younger food handlers [47].

Previously conducted evaluation of the KAP attributes of food handlers in Turkey, showed a significant difference among the food safety knowledge, food safety attitude, and food safety practices between male and female respondents. In contrast, Patil, Cates [48] reported that men demonstrated inadequate food safety practices.

Similarly, McIntyre, Vallaster [19] conducted a KAP attribute study in Canada, which concurred with the present research by demonstrating no association between the food safety knowledge of food handlers and their gender. Since current study predominantly comprised of male food handlers, it was not feasible to draw definitive conclusions regarding the impact of gender on food safety KAP attributes among food handlers in Kazakhstan.

Food safety knowledge exhibited the most robust correlation with the educational attainment of food handlers, followed by food safety attitude and food safety practices. A study conducted in Vietnam by Vo, Le [49] defined that a level of education positively influenced both food safety knowledge and food safety practices among food handlers. Additionally, Souza, Azevedo [50] found a significant correlation between the level of education and the food safety knowledge of food handlers. In another study, carried out in Brazil, it was observed that education had a significant correlation with the food safety knowledge of participants [51]. Similar findings were reported by McIntyre, Vallaster [19] in a KAP-based survey involving food handlers in Canada. Additionally, concerning the KAP scores of food handlers, Yarrow, Remig [52] found that a higher level of education was associated with improved KAP scores. Consequently, the present study concludes that the educational level of food handlers exhibits a significant association with food safety attributes based on KAP.

Despite the strengths of this study, several limitations should be acknowledged. One important limitation is that the research focused solely on individual-level knowledge, attitudes, and practices of food handlers, without assessing the implementation of broader food safety management systems such as Hazard Analysis and Critical Control Points (HACCP). Future research should consider assessing the presence and impact of HACCP systems to provide a more comprehensive understanding of food safety practices at both individual and organizational levels. Second, the findings are limited to a specific region and may not fully reflect the diversity of food safety practices across other parts of Kazakhstan. Consequently, future research efforts, with a focus on investigating the precise nature of the correlation between these food safety attributes and involving larger sample sizes, will contribute significantly to enhancing food safety standards in developing nations. Additionally, there is a need for future research to explore the impact of the COVID-19 pandemic on the food safety knowledge, practices, and attitudes of food.

Most people recover from Dengue fever, which is caused by a virus transmitted from the Asian tiger mosquito, but some infected individuals experience serious bleeding, a sudden drop in blood pressure, and even death.

Although Dengue fever has long been recognized as a disease of the tropics and sub-tropics, a study published in Global Change Biology reveals that it is likely to spread increasingly northward and through Western Europe as climate change expands the Asian tiger mosquito’s habitat. 

The mosquito lays its eggs in water where, with sufficient temperature, larvae develop and give rise to flying bloodsucking adults. After arriving in Albania, in Southeast Europe, in 1979, it spread through Southwest Europe and now is expanding northward in France. Using modeling and observations of recent spread, investigators determined that with climate change conditions, major Western European cities-including London, Vienna, Strasburg, and Frankfurt-would be well suited to the Asian tiger mosquito. 

The team notes that although the mosquito is not in these cities yet, its rate of northward spread in France has been accelerating from about 6 km per year in 2006 to 20 km per year in 2024. 

Extrapolating from the results, it is estimated that the mosquito could establish itself in northern France within a decade, from where it could easily reach London-which is already climatically suitable for hosting this vector.”

Andrea Radici, PhD, Study Corresponding Author and Postdoctoral Researcher, Université de Montpellier

In addition to Dengue fever, the Asian tiger mosquito also transmits Zika and chikungunya. 

The global burden of strokes

Between 1990 and 2021, the global burden of stroke in women of reproductive age generally declined. Specifically, the prevalence of ischemic stroke increased, the YLDs rate remained stable, and the DALYs rate increased. In contrast, the burden of ICH and SAH showed an overall downward trend (Supplementary Fig. 1 and 2).

In 2021, the prevalence of stroke was 527.51 (95%UI: 486.94 to 564.65), with an AAPC of −0.22 (95%CI: −0.23 to −0.21). The YLDs rate was 83.89 (95%UI: 59.70 to 107.25), with an AAPC of −0.33 (95%CI: −0.37 to −0.30), and the DALYs rate was 432.28 (95%UI: 392.06 to 472.61), with an AAPC of −1.27 (95%CI: −1.30 to −1.23) (see Table 1).

For ischemic stroke, the prevalence rate was 330.12 (95%UI: 294.61–364.87), with an AAPC of 0.08 (95%CI: 0.07 to 0.09). The YLDs rate was 51.79 (95%UI: 36.79 to 67.04), with an AAPC of 0.00 (95%CI: −0.02 to 0.03), and the DALYs rate was 103.77 (95%UI: 86.24 to 121.68), with an AAPC of −0.49 (95%CI: −0.52 to −0.45) (see Supplementary Table 1).

For ICH, the prevalence rate was 128.83 (95%UI: 115.55 to 143.89), with an AAPC of −0.83 (95%CI: −0.86 to −0.81). The YLDs rate was 21.00 (95%UI: 14.84 to 27.98), with an AAPC of −0.93 (95%CI: −0.99 to −0.85), and the DALYs rate was 258.64 (95%UI: 228.30 to 286.74), with an AAPC of −1.37 (95%CI: −1.42 to −1.33) (see Supplementary Table 2).

For SAH, the prevalence rate was 69.60 (95%UI: 62.39 to 78.02), with an AAPC of −0.41 (95%CI: −0.47 to −0.38). The YLDs rate was 11.09 (95%UI: 7.65 to 14.46), with an AAPC of −0.53 (95%CI: −0.59 to −0.47), and the DALYs rate was 69.86 (95%UI: 62.26 to 82.45), with an AAPC of −1.86 (95%CI: −1.89 to −1.83) (see Supplementary Table 3).

Global trends by regions

From 1990 to 2021, the stroke burden in women of reproductive age across the five SDI regions generally exhibited a declining trend. However, significant regional variations in stroke and related indicators were observed in 2021. Overall, the burden of stroke was the highest in the High-middle SDI regions and the lowest in the Low SDI regions, with the latter showing a pronounced decrease. Among the different stroke subtypes, ischemic stroke presented the highest burden in the High-middle SDI regions, whereas the most notable reductions in burden were observed for ICH and SAH in the Low SDI regions. The stroke-related DALYs rates across all SDI regions displayed a downward trend, with the High–middle SDI regions showing the most substantial reduction. Notably, the Low SDI regions experienced the most marked decrease in both prevalence and YLDs rates for ischemic stroke, whereas the other regions exhibited a general upward trend (see Supplementary Fig. 1 and Supplementary Fig. 2).

In 2021, stroke indicators showed significant disparities between different SDI regions. The prevalence of stroke was highest in High-middle SDI regions (580.59, 95%UI: 532.15 to 624.36) and lowest in Low SDI regions (467.34, 95%UI: 438.79 to 497.32), with a notable downward trend observed in Low SDI regions (AAPC: −0.45, 95%CI: −0.46 to −0.45). The YLDs rate peaked in the High-middle SDI regions (106.57, 95%UI: 74.27 to 136.43) and was at its lowest in the Low SDI regions (63.82, 95UI: 46.04 to 81.09), with the most pronounced decline in the Low SDI regions (AAPC: −0.45, 95%CI: −0.50 to −0.42). The DALYs rate was highest in the Low-middle SDI regions (500.18, 95%UI: 446.86 to 559.40) and lowest in the High SDI regions (221.88, 95%UI: 199.22 to 246.92), with the most notable reduction occurring in High SDI regions (AAPC: −1.61, 95%CI: −1.63 to −1.57) (see Table 1).

Please refer to the Supplementary Materials and Table 1 for detailed results on the burden of stroke subtypes across the five SDI and global regions.

Supplementary Fig. 3 illustrates the distribution of prevalence, YLDs, and DALYs rates for ischemic stroke, ICH, and SAH across the 27 super-regions. In these regions, ischemic stroke accounted for the largest proportion of both the prevalence and YLDs rates. ICH contributed to the highest proportion of DALYs rates across most regions, highlighting a distinct trend from other stroke indicators.

Global trends by countries and SDI

We conducted an analysis of the burden of stroke, ischemic stroke, ICH, and SAH, focusing on prevalence, YLDs, and DALYs rates across 204 countries and regions globally in 2021 (see Fig. 1).

In 2021, we observed considerable regional variations in stroke burden indicators. Nauru had the highest prevalence, YLDs, and DALYs rates, with values of 1,223.56 (95% UI: 1,180.50 to 1,271.92), 241.44 (95% UI: 173.40 to 311.30), and 2,646.68 (95% UI: 1,977.34 to 3,663.05), respectively. Conversely, Switzerland exhibited the lowest prevalence and DALYs rates, at 287.78 (95% UI: 267.64–308.64) and 70.69 (95% UI: 57.65 to 83.30), respectively. Paraguay, on the other hand, had the lowest YLDs rate of 33.11 (95% UI: 23.89 to 43.01) (see Supplementary Table 4).

Please refer to Supplementary Materials for detailed results on the burden of stroke subtypes across 204 countries and territories.

We also explored the relationship between the SDI values and stroke burden indicators in 204 countries and regions from 1990 to 2021. These findings revealed that the burden of ischemic stroke and ICH generally decreased over time as well as with increasing SDI values. The burden of SAH exhibited a gradual increase with time and SDI values until it reached an SDI value of approximately 0.60, at which point it peaked. Following this peak, the burden began to decline as both the time and SDI values continued to increase (see Fig. 2).

Variation in stroke burden across different age groups

Comprehensive age-specific burden estimates for stroke and its subtypes are presented in Supplementary Materials. From 1990 to 2021, the overall burden of stroke in women of reproductive age has shown a clear downward trend. More specifically, the burden of ischemic stroke, ICH, and SAH demonstrated declines across various indicators, with the most pronounced reductions occurring in the DALYs rates (see Supplementary Fig. 4).

The 15–24 age group experienced the most substantial reduction in stroke prevalence, with an AAPC of −0.59 (95%CI: −0.61 to −0.58). The 20–24 age group showed the most significant decline in the YLDs rate, with an AAPC of −0.87 (95%CI: −0.91 to −0.83). The 35–39 age group demonstrated the largest reduction in the DALYs rate, with an AAPC of −1.82 (95%CI: −1.85 to −1.79) (see Table 1).

For ischemic stroke, the 20–24 age group showed the most notable decrease in both prevalence and YLDs rates, with AAPCs of −0.34 (95%CI: −0.36 to −0.33) and −0.57 (95%CI: −0.63 to −0.52), respectively. In comparison, the 45–49 age group exhibited the most significant decline in DALYs rate, with an AAPC of −1.03 (95%CI: −1.06 to −0.99) (see Supplementary Table 1).

In the analysis of ICH, the 45–49 age group had the most substantial decrease in prevalence rate, with an AAPC of −1.31 (95%CI: −1.35 to −1.29), while the 20–24 age group showed the greatest reduction in YLDs rate, with an AAPC of −1.44 (95%CI: −1.56 to −1.32). The 35–39 age group showed the most significant drop in DALYs rate, with an AAPC of −2.00 (95%CI: −2.03 to −1.97) (see Supplementary Table 2).

For SAH, the 45–49 age group displayed the most pronounced decrease in prevalence rate, with an AAPC of −0.82 (95%CI: −0.84 to −0.81), while the 20–24 age group experienced the largest reduction in YLDs rate, with an AAPC of −1.06 (95%CI: −1.20 to −0.88). The 35–39 age group showed the most significant drop in the DALYs rate, with an AAPC of −2.35 (95%CI: −2.40 to −2.31) (see Supplementary Table 3).

We found that, as age increased, the burden of various types of strokes also tended to increase. Among all stroke types, ischemic stroke exhibited the highest prevalence and YLDs rates in the same age group in 1990 and 2021, whereas SAH had the lowest prevalence and YLDs rates. Intriguingly, ICH consistently had the highest DALYs rate in the same age group across the same period, highlighting a distinct trend compared with the prevalence and YLDs rates (see Fig. 3).

Strokes-related DALYs rate in different regions and age groups associated with risk factors

In 2021, 23 risk factors were related to the DALY rates of stroke and ischemic stroke, 17 risk factors were associated with the DALY rates of ICH, and 14 risk factors were linked to the DALY rates of SAH (see Fig. 4). As age increased, the association between these risk factors and the DALY rates of stroke and its subtypes became progressively more pronounced, and the number of relevant risk factors also increased (see Fig. 4).

Risk factors affecting stroke-related DALYs rates in women of reproductive age can be categorized into three main types: metabolic, environmental/occupational, and behavioral risks. Compared with 1990, among the three categories of risk factors, stroke-related DALYs associated with metabolic risks exhibited a continued upward trend, whereas those associated with environmental/occupational and behavioral risks showed a declining trajectory across 27 super-regions. This trend was also evident in the DALYs related to ischemic stroke, ICH, and SAH (Fig. 5).

We analyzed the risk factors associated with stroke-related DALY rates in women of reproductive age from 1990 to 2021, along with trends in their attributable burden. The analysis revealed a significant upward trend in DALY rates linked to four specific risk factors over the period: “Diet high in sugar-sweetened beverages” (AAPC = 2.33), “High body-mass index”(AAPC = 1.81), “High temperature” (AAPC = 0.77), and “High fasting plasma glucose” (AAPC = 0.51). However, “ambient particulate matter pollution” did not have a statistically significant effect on the DALY rate. Seventeen other risk factors showed downward trends. Because of the inability of the Joinpoint software to handle negative values, analysis of the impact of “diet high in red meat” on the DALY rate was not conducted in this study (see Supplementary Table 5).

In women of reproductive age, “high systolic blood pressure” demonstrated the strongest association with DALY rates among all assessed risk factors for stroke, ICH, and SAH, with contributions of 177.67 (95% UI: 120.80 to 234.60), 108.93 (95% UI: 75.44 to 146.96), and 26.97 (95% UI: 18.16 to 36.75), respectively. “Diet high in red meat” was associated with the lowest stroke burden, with values of −28.66 (95% UI: −121.24 to 36.86), −21.56 (95% UI: −88.88 to 27.57), and −7.77 (95% UI: −32.18 to 9.97). “High LDL cholesterol” demonstrated the strongest association with DALY rates of ischemic stroke, with an estimated value of 51.58 (95% UI: 25.68 to 73.64), whereas”diet low in polyunsaturated fatty acids”was associated with the lowest stroke burden, with a value of 0.05 (95% UI: 0.02 to 0.08) (see Supplementary Table 6).

Similar patterns were observed in the 45–49 age group. The DALY rates for stroke, ICH, and SAH were strongly correlated with”high systolic blood pressure,”with values of 665.65 (95% UI: 882.12–465.50), 427.90 (95% UI: 573.32–301.72), and 87.78 (95% UI: 118.37 to 59.28), respectively.”Diet high in red meat”was associated with a relatively minor association, with values of −76.82 (95% UI: 103.01 to −327.12), −61.77 (95% UI: 80.40 to −262.75), and −17.20 (95% UI: 23.48 to −70.54).”High LDL cholesterol”exhibited the strongest association with the burden of ischemic stroke–related DALYs, at 154.88 (95% UI: 74.22 to 221.02).

We observed substantial uncertainty in the DALY rates of different stroke subtypes associated with several risk factors, including the”diet high in red meat,””diet low in whole grains,””diet low in fiber,”and”lead exposure”. The 95% UI for these risk factors encompassed zero, indicating that their effect on stroke DALY rates was ambiguous and likely lacked statistical significance.