Individuals presented in clinical forensic cases in 2019

On a national level, the police can request a clinical forensic examination for victims and perpetrators involved in criminal acts. These examinations assist in investigating the case by securing DNA samples, documenting injuries, and obtaining urine and blood samples for toxicological analysis. Clinical examinations are conducted only in cases where the police deem the results critical for the investigation; consequently, they are not performed in all criminal cases or for all individuals implicated. Typically, clinical examinations are reserved for more serious cases.

In 2019, 358 cases involving a total of 455 individuals over the age of 18 included urine sampling. These cases were categorized into six types of crime, as detailed in Table 1.

Table 1 Overview of the case types and individuals included in the study. The case type “Other” includes robbery, road traffic accidents, and suspicious homicide among others. The mean age of individuals for each category is presented in cases of more than one person and the min, max, and median values are presented for the categories with more than three persons to maintain the anonymization

As shown in Table 1, the most frequent case type was sexual assault, followed by blunt force and sharp force cases. Since 2016, there has been an increase in the number of victims and perpetrators examined in sexual assault cases, likely due to the Ministry of Justice’s directive for more thorough investigations in this area [23]. The remaining categories (pyromania, shooting, and others) were less common in the study. The average age of individuals varies across case types, but the majority fall within the 26–35 year age range. In 73 of the 358 cases, both a victim and a perpetrator were involved. Additionally, in nine cases, there were two to five perpetrators associated with a single victim.

Alcohol and substances detected in clinical forensic cases

Results from the analysis of 455 urine samples from clinical forensic examinations of adult victims and perpetrators are summerized in Table 2. For a comprehensive list of the compounds detected in each class, please refer to Supplementary Table 1.

Table 2 Prevalence of alcohol, drugs, medicals, and performance and image-enhancing drugs in the six categories of clinical forensic case types in the study distributed among victims (V) and perpetrators (P). Both percentages and the number of positive findings are listed in the table

Nearly all classes of substances were detected across the six case types (Table 2). Despite this, no substances were identified in 32% of victims and 19% of perpetrators. Alcohol was the most frequently detected compound, found in 34% of victims and 40% of perpetrators, excluding those in the lifestyle category. This finding is consistent with previous studies on both victims and perpetrators [6, 7]. CNS stimulants were detected in 21% of victims and 28% of perpetrators. Blunt and sharp force cases showed the highest percentage of CNS stimulants, a trend aligned with other studies [9, 10]. This is expected, as cocaine, the most commonly detected CNS stimulant, is known for its mood-altering effects and increased propensity for risky behavior, including violence [24]. Similarly, cannabis, which has also been linked to violent behavior [25], is frequently encountered in street-level drug reports by the Danish police [26]. [26].

The presence of opioids in blunt (32% in victims and 20% in perpetrators) and sharp force cases (50% in victims and 33% in perpetrators) is unexpected given their primary effects of analgesia and sedation. Despite their usual sedative effects, opioids have been associated with an increased risk of violence in various studies [27,28,29]. It should be noted that some opioids, such as fentanyl, might be detected due to medical treatment rather than self-administration. Steroids and other doping-related substances were found in approximately 2% of samples. A testosterone-to-epitestosterone (T/E) ratio between 4 and 6 was observed in 2% of urine samples, suggesting possible steroid use, though this was not confirmed by metabolites or other methods. This finding is consistent with a Danish study from 2014 [30], which also included cases involving driving under the influence. However, these figures are relatively low compared to a ten-year Swedish study, which found 33.5% of individuals testing positive for steroids [4].

When comparing drug occurrences between victims and perpetrators across all case types, a significant difference was observed (p < 0.01) in blunt and sharp force cases. However, there were no significant differences in sexual assault cases (p = 0.04). The higher number of detected substances in perpetrators compared to victims in blunt and sharp force cases may explain these differences. Conversely, more substances were detected in victims than in perpetrators in opioid cases. Due to small sample sizes, no tests were performed on other categories. No significant difference was found between perpetrators in blunt force and sharp force cases (p = 0.47), but significant differences were observed between perpetrators in blunt and sharp force cases versus those in sexual assault cases (p < 0.01). Interestingly, substances of abuse were detected in both victims and perpetrators in most case types, with some cases showing a higher frequency of substances in victims, such as CNS stimulants in sharp and blunt force cases. A possible explanation for this could be a higher incidence of criminal activities within drug-abusive environments.

Polydrug use is frequently observed among individuals engaged in criminal activities, which is crucial to recognize due to the escalated risks associated with simultaneous drug consumption. The number of substances detected in each case is summarized in Table 3.

Table 3 The number of substances detected in each case type. The occurrence of nicotine, Paracetamol, and ibuprofen was not included when counting the number of substances detected for each case type

The range of drugs used in sharp force and blunt force cases varied from one to eleven, without a distinct pattern. This finding aligns with the known correlation between polydrug use and increased physical and verbal aggression [31]. In contrast, polydrug use in sexual assault cases was less common, with only a small percentage (5–7%) involving more than three substances among victims and perpetrators. This is consistent with other studies, which found either no drug use or only alcohol in sexual assault cases, with limited occurrences of polydrug use [13, 14, 16].

The interrelationship of drug patterns between victims and perpetrators

Understanding whether both the victim and the perpetrator were under the influence during the incident is crucial for accurately assessing the circumstances and the potential impact of substance use on the event. Figure 1 illustrates the relationship in 71 cases involving a single victim and a single perpetrator.

Fig. 1

An overview of the interrelationship of alcohol and drug abuse between victim and perpetrator in cases where there is only one victim and one perpetrator. The illustration comprises 71 relationships, with 13 involving sharp force (blue), 12 involving blunt force (yellow), and the remaining 46 falling into the category of sexual assault (green). Abbreviation used in this Figure: AL: alcohol, AD: antidepressants, AP: antipsychotic agents, B: benzodiazepines, CA: cannabis, CNS: central nervous system stimulants, O: opioids

As shown in Fig. 1, both the victim and the perpetrator were under the influence of at least one substance in cases of sharp force and blunt force where both individuals were present. Most cases involving sexual assault exhibited a similar pattern between victims and perpetrators. However, in nine cases, no substances of abuse were detected in either the victim or the perpetrator. This unexpected finding contradicts the common assumption that victims use fewer drugs compared to perpetrators. It has been reported that perpetrators often come from environments characterized by high levels of polydrug use, which may increase their likelihood of engaging in criminal activities [1, 3].

The sampling interval for victims ranged from one to 24 h, while for perpetrators it ranged from one to eight hours, except for one case with a 40-hour interval, suggesting that late sampling could contribute to negative results. Additionally, there were nine cases with multiple perpetrators for each victim. Details of these cases are provided in Supplementary Table 2.

Currently, toxicological analyses are requested by the police in less than 10% of clinical forensic cases. This limited analysis provides only a partial understanding of this population segment. Gaining insight into the types and quantities of substances used, as well as the combinations and associated harms, is crucial for developing effective information campaigns and prevention strategies aimed at reducing criminal incidents. Regular toxicological analyses could also help monitor the effectiveness of various interventions in this area.

The importance of the timing of sample collection for results in clinical forensic cases

For optimal accuracy, urine samples should be collected as soon as possible after an incident. Delays in sampling increase the risk of contamination and degradation of drug metabolites, which complicates the determination of substances present and their concentrations. This compromises the reliability of forensic toxicology results in determining whether an individual was under the influence at the time of the crime. Figure 2 illustrates the time interval from the incident to sample collection.

Fig. 2

The time from the incident until the sample was taken by a forensic pathologist. The time distribution was 367 samples ≤ 24 h (yellow line) < 44 samples ≤ 48 h (orange line) < 28 samples ≤ 72 h (red line), 12 > 72 h. The time interval was not registered for four samples

Figure 2 shows that 367 samples, representing 81% of the total, were collected within 24 h of the incident. 10% were collected between 24 and 48 h, 6% between 48 and 72 h, and 3% after 72 h. Most late samplings involved alleged victims in sexual assault cases, with some perpetrators in blunt force cases also falling into this category, possibly due to delayed arrests. Only 1% of the samples lacked a defined time interval. This unknown interval often resulted from inability to recall the timing of the incident or post-registration for cases reported over an extended period.

The detection window for most illegal drugs in urine is approximately 48 h, though some can be detected for longer after a single dose [32]. For individuals with daily drug consumption, the detection window may extend to several days or even months, depending on the drug’s administration and other factors [32]. Thus, the fact that most urine samples were collected within 24 to 48 h enhances the reliability of the results.

Alcohol has a shorter detection window in urine compared to most substances. Detection time can range from approximately one to 24 h, influenced by factors such as intake amount, diet, and genetics. To assess whether the absence of alcohol in urine samples was due to delayed sampling or a true negative result, Fig. 3A shows the collection times for negative alcohol samples. Additionally, Fig. 3B depicts the alcohol levels in urine samples across different case types.

Fig. 3

A) An overview of the timeline for cases where alcohol was not detected. For 209 out of the 286 negative sample, the time interval from the incident to sampling was found below 24 h and for the remaining 77 samples, the time interval was above 24 h. B) An overview of the alcohol level in urine for 169 samples with a confirmed positive detection

As seen in Figs. 3A and 209 corresponding to 73% of the negative results were from samples collected within 24 h. Thus, it is reasonable to conclude that these samples were negative for alcohol at the time of the incident. For the remaining 27% of negative results, delayed sampling could mean that individuals might have been under the influence of alcohol during the crime, but this cannot be conclusively determined.

Alcohol-positive findings in the urine ranged from 0.01‰ to 3.65‰, as seen in Fig. 3B. Positive results included both low and high levels of alcohol in victims and perpetrators across blunt force, sharp force, and sexual assault cases. I should be mentioned that the urine-to-blood alcohol ratio averages 1.3, but can vary between 1.0 and 2.0 depending on whether the urine is in the absorptive or post-absorptive phase [33].

Strengths and limitations of the study

Most published studies have concentrated on either the victim, the perpetrator, or a combination of both categories within a single group. The main advantage of this study is its ability to examine the substances consumed by both the victim and the perpetrator across different case types, and to investigate the correlation between these substances when both are present. Additionally, the use of two analytical methods for sample analysis of substance use enhances the robustness of the results. However, not all detected compounds were included in the targeted LC-MS/MS method, resulting in some compounds being detected by only one analytical method. The substances selected for analysis were based on national statistical data and practical considerations aimed at streamlining processing and conserving time. While some relevant substances might be missing, the TOF-MS data collected enables retrospective reprocessing to identify additional compounds, potentially offering a more comprehensive overview of substance use within our sample groups.

Regarding limitations, 153 individuals in our sample lacked a corresponding case registration. These non-linked victims and perpetrators may not be included in our sample due to being under 18 years of age, a forensic clinical examination not being performed or being performed at a different forensic department, the inability of the police to trace the perpetrator, or the absence of a urine sample. Additionally, in a few cases, victims declined to undergo examination, resulting in the absence of sample collection.

A clinical forensic examination should ideally occur as close to the time of the crime as possible to strengthen the findings. As discussed in Section. “Alcohol and substances detected in clinical forensic cases”, some cases were reported or discovered by the police later than 48 h post-incident. Knowing the time interval between the criminal event and the biological sample collection is crucial for interpreting results. Detecting drugs and their metabolites in urine becomes increasingly difficult if the sample is collected after the detection window for the drug of interest has closed. When the time interval surpasses the detection window of a specific drug, a negative result might be attributed to delayed sampling rather than the absence of drug consumption. Moreover, we cannot determine if some drugs were ingested before the incident or during the interval between the incident and urine sample collection. Lastly, as noted in Section. “Study material and ethics”, individuals undergoing clinical forensic examinations are selected and referred by the Danish police, which means that not all individuals involved in various types of crimes are represented in our sample. Therefore, it is unclear whether our sample is representative of all victims and perpetrators in Denmark. Furthermore, the selection process for clinical forensic examinations may differ across countries and authorities, warranting caution when comparing studies based on individuals who have undergone such examinations.