1. Introduction
Road traffic injuries remain a major public health and safety challenge worldwide. Among the numerous factors contributing to road traffic crashes, alcohol consumption is recognized as one of the leading preventable causes of fatalities and serious injuries [
1,
2,
3]. Despite continuous improvements in vehicle safety and road infrastructure, driving under the influence of alcohol continues to impose a substantial burden on road safety, where alcohol-related traffic mortality remains relatively high [
4,
5].
To address alcohol-impaired driving, many countries have introduced stricter blood alcohol concentration (BAC) limits, enhanced police enforcement, public awareness campaigns, and technological preventive measures. Serbia has followed this trend through successive amendments to the Law on Road Traffic Safety, including the reduction in the legal BAC limit for non-professional drivers and the strengthening of enforcement measures [
6,
7,
8,
9,
10,
11,
12,
13].
Alcohol impairs the cognitive and psychomotor functions essential for safe driving, including reaction time, attention, decision-making, risk perception, and coordination. As the blood alcohol concentration increases, drivers become more likely to underestimate risks, overestimate their driving ability, and engage in unsafe behaviors such as speeding, unsafe overtaking, or failure to comply with traffic regulations [
14,
15,
16,
17]. Consequently, alcohol consumption not only increases the likelihood of road traffic crashes but also contributes to more severe injuries and higher fatality rates among all categories of road users [
18].
In addition to alcohol, psychoactive substances represent another factor affecting road safety, as they may impair perception, attention, reaction time, and decision-making, thereby increasing the risk of traffic crashes [
19,
20]. In Serbia, testing for psychoactive substances among drivers has become increasingly integrated into routine traffic controls in recent years.
Although numerous studies have investigated the relationship between alcohol consumption and road traffic safety, evidence regarding changes in the characteristics of alcohol-impaired drivers following reductions in legal BAC limits remains limited in Serbia. In 2018, Serbia amended its road traffic legislation by reducing the legally permissible BAC limit for non-professional drivers from 0.3 mg/mL to 0.2 mg/mL. However, differences in the characteristics of alcohol-impaired drivers involved in road traffic accidents before and after this legislative amendment have not been comprehensively evaluated.
Therefore, this study aimed to compare the characteristics of alcohol-impaired drivers involved in alcohol-related road traffic accidents before and after the 2018 reduction in the legal BAC limit from 0.3 mg/mL to 0.2 mg/mL, with particular emphasis on differences in demographic characteristics, BAC distribution, temporal and accident-related characteristics, and injury outcomes between the two legislative periods.
2. Materials and Methods
2.1. Study Design
This research was conducted as an analytical, observational, retrospective, longitudinal study encompassing two distinct time periods. The first period, from January 2010 to March 2018, corresponded to the implementation of the Traffic Safety Law, which permitted a blood alcohol concentration (BAC) of up to 0.3 mg/mL for motor vehicle drivers. The second period, from April 2018 to December 2024, followed amendments to the legislation that reduced the legally permissible BAC limit from 0.3 mg/mL to 0.2 mg/mL.
The research was based on forensic medical documentation obtained from judicial authorities with territorial jurisdiction over the Raška Administrative District. The available documentation also included medical records from healthcare institutions where drivers involved in road traffic accidents were hospitalized, as well as forensic autopsy reports in cases with fatal outcomes. The study included motor vehicle drivers involved in road traffic accidents between 2010 and 2024 who underwent forensic medical examination and had a confirmed positive BAC.
2.2. Study Population
The study population consisted of motor vehicle drivers with a confirmed positive BAC who were involved in road traffic accidents and underwent forensic medical examination between 2010 and 2024. Only drivers with verified BAC values documented in forensic medical records obtained from judicial authorities within the Raška Administrative District were included in the study. Therefore, the study population represents a specific forensic sample of alcohol-impaired drivers involved in road traffic accidents and should not be interpreted as representative of all drivers involved in road traffic accidents or of the general driving population.
For injured individuals who required hospitalization following a road traffic accident, relevant data were collected from the available clinical documentation obtained during hospitalization and subsequent medical treatment. These records included medical histories, clinical examination findings, diagnostic procedures, therapeutic interventions, and hospital discharge summaries. In cases resulting in death, data were derived from forensic autopsy reports, which provided information on the cause and mechanism of death, the type and extent of injuries sustained, and other relevant postmortem findings documented during the forensic examination.
The following variables were observed and analyzed: BAC; sex; age (<21 years, 22–34 years, 35–44 years, 45–54 years, 55–64 years, and ≥65 years); month of the year; day of the week; time of the road traffic accident (morning, noon, evening, or night); type of road (highway, main road, local road, or unclassified road); injury outcome (no injury, minor bodily injury, severe bodily injury, or fatal outcome); and accident location (within or outside a populated area).
Drivers aged <21 years were analyzed as a separate age category because, under Serbian traffic legislation, novice drivers are subject to specific restrictions associated with probationary driving licenses until the age of 21 years. The legal drinking age in Serbia is 18 years; therefore, the <21-year age category was defined according to the traffic legislation relevant to novice drivers rather than according to the legal drinking age.
2.3. Blood Alcohol Concentration Assessment
Alcohol concentration data were obtained retrospectively from the available forensic medical and judicial documentation. Depending on the individual case and the procedure applied at the time of examination, alcohol concentration was documented either by evidential breath alcohol testing using a Dräger device or by laboratory analysis of a blood sample. Blood sampling was not routinely performed in all drivers following breath alcohol testing. Measurements obtained by the two methods were recorded in harmonized units to ensure comparability within the study database. Both types of documented alcohol measurements were included in the analysis. The procedures used for alcohol testing remained unchanged during the study period; the legislative amendment introduced in 2018 concerned the reduction of the legally permissible BAC limit from 0.3 mg/mL to 0.2 mg/mL.
2.4. Statistical Analysis
Statistical analysis was performed using the commercial statistical software package SPSS, version 22.0 (Chicago, IL, USA). Categorical variables were summarized as frequencies and percentages, whereas continuous variables were presented as means and standard deviations. The normality of continuous variables was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Differences in the distribution of categorical variables were evaluated using the chi-square (χ2) test. The Kruskal–Wallis test was used to compare continuous variables across more than two independent groups.
To identify predictors of injury outcome, multinomial logistic regression analysis was performed separately for the two study periods. Injury outcome was classified into four categories: no injury, minor bodily injury, severe bodily injury, and fatal outcome. Independent variables were entered into the regression models using a stepwise procedure. The regression analysis presented in this study focused on severe bodily injury in relation to the other injury outcomes. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported. Statistical significance was defined as a p-value ≤ 0.05.
3. Results
Between 2010 and 2024, a total of 18,076 RTAs were recorded, of which 1743 involved drivers with a positive BAC. The annual distribution of the results is presented in
Table 1.
A total of 1743 motor vehicle drivers with positive BAC who were involved in RTAs were included in the study. Of these, 877 drivers were identified from January 2010 to March 2018 (when the legal BAC limit was ≤0.3 mg/mL), and 866 drivers were identified from April 2018 to 2024 (when the legal BAC limit was ≤0.2 mg/mL).
During the first analysis period, among 877 subjects, 827 (94.3%) were male, and 50 (5.7%) were female, with a mean age of 40.15 ± 14.46 years (range: 15–89 years). During the second analysis period, among 866 subjects, 827 (95.5%) were male, and 39 (4.5%) were female, with a mean age of 42.77 ± 15.78 years (range: 15–84 years). A comparison of subjects’ ages between the two time intervals revealed a significantly greater mean age in the second period than in the first period (42.77 ± 15.78 vs. 40.17 ± 14.48; Kruskal–Wallis test, p = 0.001).
Within the first analyzed interval, the proportion of alcohol-impaired drivers involved in RTAs was highest in the 22–34 age group (27.0%), followed by the 35–44 age group (21.9%). In the second analyzed interval, the highest proportions were observed in the 22–34 and 45–54 age groups (both 20.1%). The lowest proportion of such drivers was consistently found in the oldest age group (≥65 years) in both periods, accounting for 5.7% of all drivers in January 2010–March 2018 and 9.4% in April 2018–2024.
The distribution of RTA subjects by age across the two time intervals is presented in
Table 2. A significant difference in age category distribution between the two periods was observed (χ
2 = 23.479,
p < 0.001).
Analysis of the distribution by month, day, and time of occurrence of RTAs revealed that the highest number of accidents involving alcohol-impaired drivers in the overall study period (2010–2024) occurred in August, September, and October, during weekends, and between 17:00 and 18:00 h. The lowest number of cases was recorded in February, on Wednesdays, and during the morning hours. However, statistical analysis of the collected data did not reveal significant differences in the distribution of study participants with respect to the month (χ
2 = 11.062,
p = 0.438), day (χ
2 = 9.455,
p = 0.150), or time of road traffic accident occurrence (χ
2 = 31.256,
p = 0.117), which are presented in
Table 3,
Table 4 and
Table 5.
There was a significant difference in the distribution of RTAs from January 2010 to March 2018 according to the hour of the day and the day of the week on which the accident occurred (χ2 = 213.370, p < 0.001). On weekends, the greatest proportion of road traffic accidents involving alcohol-impaired motor vehicle drivers occurred during “party hours”. Specifically, the highest percentage of accidents occurred on Fridays at 11:00 p.m. (11.6%), Saturdays at 11:00 p.m. and 1:00 a.m. (10.5% each), and Sundays at 2:00 a.m. (11.8%). During the second study period, there was no significant difference in the distribution of road traffic accidents according to the hour of the day in which the accident occurred (χ2 = 154.283, p = 0.163).
A significant difference in the degree of alcohol intoxication among subjects between the two time periods was identified (χ
2 = 28.935,
p < 0.001), as shown in
Table 6.
A notably greater percentage of fully intoxicated motor vehicle drivers involved in traffic accidents was observed before the legislative amendment (28.4% vs. 20.9%). In contrast, the proportion of moderately intoxicated drivers was almost twice as high after the amendment of the Road Traffic Safety Law (9.0% vs. 4.6%).
From an outcome perspective, the analysis of the relationship between the two subgroups indicated a significant difference in the frequency of possible outcomes across the two time intervals (χ
2 = 13.582,
p = 0.004) (
Table 7). The proportion of fatalities was significantly lower in the second study period than in the first study period (4.0% vs. 7.0%). Conversely, after March 2018, the percentage of motor vehicle drivers involved in traffic accidents who experienced no consequences was significantly greater than in the period before the legislative amendments (24.8% vs. 19.2%) (
Table 7).
Regression analysis of the association between the examined variables and RTA outcomes in terms of the occurrence of severe bodily injuries during the first study period indicated statistically significant associations with age category and BAC. The odds of sustaining severe bodily injuries in RTAs involving alcohol-impaired motor vehicle drivers were significantly higher among individuals aged 22–34 years (OR = 4.50, p = 0.001), 35–44 years (OR = 4.71, p = 0.001), 45–54 years (OR = 5.33, p = 0.001), and 55–64 years (OR = 5.46, p = 0.001) than among individuals aged ≥65 years. The highest odds of sustaining severe bodily injuries were observed among drivers aged 55–64 years (OR = 5.46), followed by those aged 45–54 years (OR = 5.33).
Complete intoxication (OR = 12.83, p < 0.001) and severe intoxication (OR = 5.26, p < 0.001) were associated with the highest odds of sustaining severe bodily injuries in RTAs. BAC levels above mild intoxication, except very severe intoxication, were significantly associated with higher odds of sustaining severe bodily injuries in RTAs, with ORs of 12.83 for complete intoxication, 5.26 for severe intoxication, 3.78 for high intoxication, and 4.60 for moderate intoxication, compared with mild intoxication.
Regression analysis of the impact of variables on RTA outcomes in terms of the occurrence of severe bodily injuries during the second study period (April 2018 to 2024) indicated a statistically significant association with age category, the day of the week on which the RTA occurred, and BAC. The odds of sustaining severe bodily injuries in RTAs involving alcohol-impaired motor vehicle drivers were significantly higher among individuals aged 35–44 years (OR = 2.95, p = 0.003), 45–54 years (OR = 3.15, p = 0.001), and 55–64 years (OR = 2.12, p = 0.034) than among individuals aged ≥65 years. The highest odds of sustaining severe bodily injuries were observed among drivers aged 45–54 years (OR = 3.15).
Complete intoxication (OR = 9.32, p < 0.001) and very severe intoxication (OR = 7.80, p = 0.001) were associated with the highest odds of sustaining severe bodily injuries in RTAs. BAC levels above mild intoxication, except medium and moderate intoxication, were significantly associated with higher odds of sustaining severe bodily injuries in RTAs, with ORs of 9.32 for complete intoxication, 7.80 for very severe intoxication, 4.31 for severe intoxication, and 2.85 for high intoxication, compared with mild intoxication.
The odds of sustaining severe bodily injuries were lower in RTAs occurring on Tuesdays (OR = 0.51, p = 0.048), Wednesdays (OR = 0.38, p = 0.008), and Thursdays (OR = 0.51, p = 0.046) compared with Sundays. The lowest odds were observed on Wednesdays (OR = 0.38), corresponding to approximately 62% lower odds of severe bodily injury compared with Sundays.
4. Discussion
This study examined demographic characteristics, temporal patterns, and injury outcomes among alcohol-impaired drivers involved in road traffic accidents across two legislative periods with different legal BAC limits. The findings revealed a consistent demographic profile, characterized by a pronounced male predominance and a modest but statistically significant shift toward older drivers in the second study period.
During the study period, the proportion of road traffic accidents involving alcohol-impaired drivers ranged from 5.4% to 16.0%, with the lowest value recorded in 2012 and the highest in 2021. During the peak of COVID-19, which occurred in Serbia in 2020, with two months of lockdown, we observed a decline in road traffic accidents involving alcohol-impaired drivers, which is consistent with the findings of other studies [
21].
In both time periods, the majority of alcohol-impaired drivers were male (94.3% vs. 95.5%), which is consistent with previous epidemiological studies reporting that men represent the predominant group involved in alcohol-related driving offenses and crashes [
22,
23,
24,
25]. This gender difference may be associated with higher levels of heavy episodic drinking, risk-taking behavior, and greater exposure to driving under the influence among men.
Drivers in middle adulthood were the most frequently represented age group (40.15 ± 14.46 vs. 42.77 ± 15.78 years), in agreement with previous studies identifying individuals in their fourth and fifth decades of life as a high-risk group for alcohol-related traffic injuries [
14,
26,
27,
28,
29]. These findings may reflect greater driving exposure, occupational mobility, and frequent participation in daily traffic among economically active individuals.
A statistically significant increase in mean age was observed in the second study period. Although the difference was modest, these findings may indicate a change in the demographic profile of alcohol-positive drivers involved in crashes. Previous studies suggest that younger drivers may be more responsive to enforcement and preventive measures, whereas older drivers may demonstrate more persistent patterns of alcohol consumption and driving behavior. Evidence from meta-analyses indicates that the effects of regulatory interventions may vary across demographic subgroups, including age-related differences in response to deterrence measures [
30,
31,
32].
Despite the reduction in the legal BAC threshold, the gender distribution remained unchanged between the two periods. This finding is consistent with the persistence of established gender-related patterns of alcohol-impaired driving across the two study periods, which may be influenced by sociocultural and behavioral factors.
Age-stratified analysis revealed that alcohol-impaired driving was most common among younger and middle-aged adults, with the highest proportion observed in the 22–34 age group in both periods. A secondary peak shifted from the 35–44 to the 45–54 age group in the second period, while drivers aged ≥65 years remained the least represented group, despite a slight increase.
These findings are consistent with international evidence indicating that alcohol-related crashes are concentrated among younger and middle-aged drivers, likely because of greater driving exposure, episodic heavy drinking, and risk-taking behavior [
3,
33]. The persistence of the youngest age peak and the shift toward older age groups may reflect gradual changes in the demographic profile of high-risk drivers between the two study periods.
The temporal distribution of alcohol-impaired road traffic accidents showed a non-uniform pattern, with the highest number of cases occurring during late summer and early autumn (August–October). Following the introduction of the new Road Traffic Safety Law and the reduction in the legal BAC limit, the previously observed weekend late-night pattern associated with “party hours” was no longer evident in the descriptive distribution of crashes The only exception was Saturday at midnight, when the highest proportion of Saturday accidents was recorded, whereas on Fridays and Sundays the highest frequencies occurred in the afternoon (17.4% at 5:00 p.m. and 10.5% at 6:00 p.m., respectively).
Although these differences were not significant, the observed temporal patterns are consistent with international evidence describing clustering of alcohol-related crashes during periods of increased social activity, particularly weekends, holidays, and warmer months, when alcohol consumption and driving exposure are elevated [
33]. The late afternoon peak may be associated with after-work alcohol consumption and commute-related traffic exposure, as previously reported [
22,
34,
35]. The observed differences between the two study periods may indicate a potential shift in the temporal profile of alcohol-related crashes.
The present study demonstrated a significant difference in the degree of alcohol intoxication among drivers involved in traffic accidents between the two observed periods. A greater proportion of fully intoxicated drivers was observed before the amendment of the Road Traffic Safety Law, whereas moderately intoxicated drivers represented a greater proportion after the legislative changes. These findings are consistent with a shift toward lower levels of alcohol intoxication among alcohol-impaired drivers in the second study period, although alcohol-impaired driving remains an important public health and traffic safety issue.
The lower proportion of fully intoxicated drivers observed in the second study period is consistent with the findings of previous international studies showing that stricter legislation, lower BAC limits, enhanced enforcement, and stronger penalties have been associated with reductions in high-risk alcohol-related driving behaviors. Studies by Fell and Voas demonstrated that stricter BAC regulations were associated with reductions in alcohol-related crashes and fatalities, whereas European studies reported similar effects following strengthened legal frameworks and intensified police controls [
6,
7,
9].
However, the higher proportion of moderately intoxicated drivers observed in the second study period suggests that alcohol-impaired driving remains a persistent problem despite legislative measures. One possible explanation is behavioral adaptation, whereby drivers adjust their alcohol consumption to remain below thresholds associated with more severe penalties. Previous studies have described similar patterns, indicating that the effectiveness of alcohol policies depends on consistent enforcement, public education, and social attitudes toward drinking and driving [
36].
Despite lower levels of intoxication, moderately impaired drivers remain a significant safety concern, as even low-to-moderate BAC levels can negatively affect reaction time, attention, visual perception, and decision-making. The risk of crashes increases progressively even at BACs below traditional legal limits [
37,
38].
The analysis of outcomes among drivers involved in traffic accidents revealed a significant difference between the two observed periods. The proportion of fatalities was significantly lower in the second study period than in the first study period (4.0% vs. 7.0%), whereas the proportion of drivers without immediate consequences was higher (24.8% vs. 19.2%). These findings describe a more favorable distribution of accident outcomes in the second study period, although alcohol-related crashes with minor or no immediate consequences remained frequent.
Previous studies from the United States and Europe have shown that lower BAC limits, stricter penalties, and enhanced enforcement are associated with reductions in alcohol-related fatalities and severe injuries [
7,
39]. These findings are consistent with previous reports describing associations between lower BAC limits, stricter enforcement, and reductions in alcohol-related fatalities and severe injuries.
The higher proportion of accidents without immediate consequences in the second study period may reflect changes in driver behavior rather than an actual increase in accident occurrence. Drivers may adjust their alcohol consumption to remain below perceived legal thresholds, resulting in lower levels of impairment but continued driving under the influence. Similar behavioral adaptations have been reported following BAC legislation changes, with reductions in severe intoxication accompanied by persistent lower-level impairment [
40].
Although the lower proportion of fatalities observed in the second study period is encouraging, minor accidents may still result in considerable economic, psychological, and societal consequences. These observations support the continued implementation of comprehensive prevention strategies, including legislative measures, public awareness campaigns, targeted enforcement, and education [
33].
Several studies have shown that middle-aged adults, particularly those aged 35–64 years, are disproportionately represented among drivers sustaining severe injuries in alcohol-related crashes, which is consistent with our findings [
41,
42]. This pattern may reflect differences in driving exposure, risk-taking behavior, and individual susceptibility to injury. Younger and middle-aged drivers involved in alcohol-related crashes have also been reported to experience more severe outcomes than older drivers do, possibly because of differences in risk perception and self-regulation [
43].
The association between higher BAC levels and increased injury severity observed in our study is consistent with previous evidence. Severe intoxication has been shown to substantially increase the risk of serious and fatal traffic injuries [
7,
44]. Bédard et al. reported that drivers with BAC levels > 0.10 g/dL had a higher risk of severe injury than those with lower BAC levels [
41]. Furthermore, meta-analytic studies have demonstrated that even moderate alcohol consumption impairs psychomotor performance, reaction time, and hazard perception, contributing to increased crash-related injury risk [
45,
46]. Our findings further support an association between higher BAC levels and greater odds of severe bodily injury.
The regression analysis for the second study period (2018–2024) revealed statistically significant associations between severe bodily injury and age category, BAC level, and day of the week. Compared with drivers aged ≥65 years, those aged 35–44, 45–54, and 55–64 years had significantly higher odds of sustaining severe bodily injuries, with the highest odds observed among drivers aged 45–54 years. In addition, complete and very severe intoxication were associated with the highest odds of severe bodily injury. The odds of sustaining severe bodily injuries were significantly lower in accidents occurring on Tuesdays, Wednesdays, and Thursdays compared with Sundays. The regression analyses for both study periods are presented in
Tables S1 and S2.
These findings are consistent with previous research demonstrating that increasing BAC levels are associated with progressively higher risks of severe and fatal crash outcomes [
7,
44,
47]. Studies have also shown that although younger drivers are generally considered high-risk groups, severe injury outcomes are frequently observed among intoxicated middle-aged drivers [
48].
The lower odds of severe bodily injuries observed on Tuesdays, Wednesdays, and Thursdays compared with Sundays are consistent with previous studies reporting higher rates of alcohol-related crashes during weekends, particularly during periods associated with social gatherings and increased alcohol consumption [
24,
25]. This pattern may reflect reduced recreational alcohol use and lower levels of nighttime social activity during weekdays.
Limitations of the Study
This study has several limitations that should be considered when interpreting the findings. First, the retrospective observational design does not allow causal relationships to be established between the reduction in the legal BAC limit and the observed differences between the two study periods. Given the extended observation period, several time-dependent factors, including changes in traffic law enforcement, vehicle safety, road infrastructure, emergency medical care, public awareness, traffic volume, and alcohol consumption patterns, may have influenced the observed findings. In addition, the post-law period included the COVID-19 pandemic, during which mobility restrictions and changes in traffic exposure may have affected road traffic accident patterns, particularly in 2020.
Comprehensive exposure data, including traffic volume, vehicle kilometers traveled, the number of registered vehicles, and the number of alcohol tests performed, were not consistently available throughout the entire study period and therefore could not be incorporated into the analyses. Furthermore, information on several potentially relevant confounding factors, including vehicle type, vehicle speed, crash type, use of safety equipment, weather conditions, roadway characteristics, comorbidities, and concurrent psychoactive substance use, was not consistently available for analysis. The study population consisted of a specific forensic sample of alcohol-impaired drivers involved in road traffic accidents and did not include a true population-based comparison group. In addition, the study was conducted in a single administrative region of Serbia; therefore, regional differences in traffic characteristics, road infrastructure, enforcement practices, and population structure may limit the generalizability of the findings to other regions and to the overall driving population in Serbia. Finally, the relatively small number of drivers aged ≥65 years, who served as the reference group in the regression analyses, may have contributed to wider confidence intervals and should be considered when interpreting the age-related odds ratios.
Despite these limitations, the study also has important strengths. The relatively large sample size, extended observation period, detailed forensic data, and comparison of two legislative periods allowed a comprehensive assessment of demographic characteristics, intoxication severity, temporal patterns, and injury outcomes among alcohol-impaired drivers. The inclusion of annual data on total and alcohol-related road traffic accidents further provides important temporal context for interpreting the observed findings. Overall, the study contributes to the limited regional evidence regarding changes in alcohol-related road traffic accidents observed in the context of a reduction in the legal BAC limit and provides a basis for further multicenter and population-based research incorporating more comprehensive exposure indicators.