Content area
Despite its relatively high cost and time-consuming analysis, scanning electron microscopy coupled with energy dispersive X-ray spectroscopy (SEM-EDS) has been the “gold standard” technique for analysing inorganic gunshot residue (GSR) particles for more than 40 years. Conversely, research has largely focused on developing new techniques and innovations that have struggled to gain acceptance in routine GSR examination.This study aims to explore the persistent gap perceived between research and practice in the field of GSR, specifically why research continues to propose novel instrumental analyses that are not adopted in practice. This question was examined through a comprehensive literature review, a survey of GSR examination, and a round table organised at the annual meeting of the European Network of Forensic Science Institutes (ENFSI) Expert Working Group (EWG) Firearms / GSR in Brussels in 2023.The literature review highlighted that publications on GSR have steadily increased over the past 20 years, followed by a slight decrease after 2020 to reach approximately 40 publications in 2022. In that year, 42 % of the relevant publications focused on the development of novel methods, while 26 % addressed issues related to GSR persistence, prevalence, and interpretation.Survey responses from 45 GSR experts confirmed that residues are mainly collected from hands with carbon stubs and analysed by SEM-EDS. Among respondents, 90 % work in accredited laboratories, and 95 % have little time for research beyond routine duties. While practical innovations have largely concentrated on the optimisation of current approaches, experts strongly support collecting additional forensic data on persistence, prevalence, and secondary transfer. However, such data often suffer from a lack of harmonisation and are only indirectly useful for the interpretation, which is gradually shifting from source to activity inferences.Both practitioners and academics advocate for increased collaboration to define and conduct more impactful GSR research. However, addressing the identified barriers may require systemic changes that go beyond simply increasing resources.
1 Introduction
When a firearm is discharged, gunshot residues (GSR) are produced and expelled from the weapon through its various openings. GSR consists of an inorganic component (mainly originating from the primer, but also the cartridge case and projectile) and an organic component (primarily from the unburnt and burnt propellant powder). The resulting particles and vapours are transferred onto the shooter's hands, face and clothing, as well as on nearby objects and people. Detecting these GSR particles to infer the activities of persons of interest (POI) in relation to a discharge event is one of the main areas of GSR expertise.
Current knowledge and advances in the field of GSR are regularly reviewed from the scientific literature
[1–9]. However, a recent paper suggested a lack of impactful research in forensic science, using GSR as an example
[10]. The barriers identified were attributed to how research and practice operate. Published research is mainly led by academics and generally follows the implicit rule of “publish or perish”. Thus, research that is easy to conduct and publish is favoured over research with actual impact on forensic practice, revealing divergent goals between research and practice
[11]. This results in a multitude of proof-of-concept articles that focus on novel techniques rather than trace
1
In this contribution, the term “trace” will be preferred to “evidence”. It is not used to designate a small amount, but all types of GSR transferred during the event of interest, as vectors of information on presence and activity [2,3].
relevance [12–14] and are increasingly published in non-forensic journals.On the other hand, practice is subject to specialisation, standardisation, and backlogs resulting in less versatility and time to meet new challenges [15,16]. For example, scanning electron microscopy coupled with energy dispersive X-ray spectroscopy (SEM-EDS) has been the “gold standard” for inorganic GSR particles analysis for over 30 years [17,18]. While the technique is considered fit for purpose, other techniques that may be cheaper or faster, but require different skills, have struggled to gain acceptance in practice.
Based on these observations, this study aimed to quantify the disconnect between GSR research topics and the needs of practitioners, identify systemic barriers to adopting new methods, and propose collaborative frameworks to bridge this gap. To achieve this, scientific papers and reviews on GSR published in the last 20 years (2003–2022) were searched using the Scopus database, and the topics of the papers published in 2022 were analysed in more detail. In parallel, a survey was sent to GSR experts in 2023 through the European Forensic Science Institute Network (ENFSI) Expert Working Group (EWG) Firearms / GSR, asking them about their practices and needs for research and development (R&D). Finally, the responses were discussed between experts and researchers during a round table organised at the 29th meeting of the ENFSI EWG Firearms / GSR.
2 Methods
The assessment of the literature was based on a Scopus search carried out in September 2023 and using the keyword “gunshot residue”. All articles published between 2003 and 2022 (inclusive) and referenced in Scopus were extracted into a Microsoft Excel spreadsheet with the following main information (see supporting information SI-1: Scopus literature search): authors, title, year, journal, doi, access link, document type and whether it is accessible in open access.
The online survey consisted of 31 questions, designed to cover most - if not all - stages and tasks performed by GSR experts, from crime scene investigation to court reporting. The questions were divided into five sections: 1) type and number of cases, 2) collection and analysis, 3) reporting, 4) research and development and 5) final questions. Some items were multiple-choice, while others required free-text responses. The questionnaire was designed by a group of academic scientists, reviewed by one independent forensic specialist and tested by one GSR expert to ensure that questions were sufficiently clear.
The survey was distributed in July 2023 via email to all members of the ENFSI EWG Firearms/GSR, as well as to participants of the 29th meeting of the ENFSI EWG Firearms / GSR (see supporting information SI-2: Survey Overview). Responses were collected entirely anonymously in accordance with ethical guidelines, meaning it was not possible to determine the countries of origin or the number of experts per laboratory who responded. While anonymity encouraged participation, it precluded analysis of regional or institutional biases (i.e., responses may be skewed toward certain countries or laboratory types). Participants were informed that, by taking part, they consented to the aggregated and reformulated survey responses being used for presentation and publication purposes.
Finally, a round table was organised during a workshop held at the 29th meeting of the ENFSI EWG Firearms / GSR. The aim was to discuss the answers to the survey and suggest a path forward for research and development in the field. The round table included one academic scientist and three GSR experts
2
Two additional academics and one GSR expert organised and animated the round table but did not actively participate to the discussions. It was not possible to have more academics as very few registered to participate to the meeting.
and was structured around three main topics: 1) the barriers between practice and (academic) research, 2) the need and utility of forensic case data, and 3) the other relevant GSR research priorities. Notes were taken by three organisers of the workshop during the preparation and the proceedings of the round table. Several workshop participants also contributed to the discussion. No names will be associated with the reported observations and statements. The free version of DeepL 3
3 Results and discussion
3.1 GSR literature overview
It is globally acknowledged that scientific literature has continuously increased in the last 20 years and this was also observed in forensic science
[10] and more specifically gunshot residue (
The number of publications on GSR doubled between 2003 and 2012 and 2013–2022 (see
Title translated from Czech using DeepL (
While Scopus is an interesting database to search for scientific literature, not all journals are indexed at a given time (e.g.,
WIREs forensic science and the
AFTE journal - the Association of Firearm and Tool Mark Examiners were not included at the time of this search) and not all the publications found are necessarily relevant. For example, searching for “gunshot residue” publications results in a significant number of papers about health and environmental issues due to the toxicity of gunshot residue. Therefore, a closer analysis of the publications published in 2022 was conducted (see SI-3: Scopus 2022). Among the 40 listed publications, nine were excluded from further analysis. Four papers were deemed off-topic, as they reviewed the potential of different techniques in forensic science, GSR only being taken as an example among other traces (hyperspectral imaging, atomic spectrometry and surface analysis techniques). Three additional papers concerned health and environmental rather than forensic issues, and one review and one case report were published in an Indian and in a Russian medico-legal journals,
5
It was not possible to further evaluate the relevance of those two papers from the data available on the internet.
respectively. The remaining 31 papers were classified according to the topics they covered (Some papers presented forensic data as well as interpretation models. Thus, those two classes were pooled together in Fig. 3.
These proportions are indicative as slightly different tendencies can be detected from one year to another. For example, there was no article published on luminescent markers in 2021, while four were published in 2022. This can be explained by the variable delays in the publication process or by a particular project or novel trend resulting in several publications close in time. However, several observations remain similar over the years such as the higher focus on method development and optimisation compared to trace transfer, persistence and prevalence. A previous review already suggested that ca. 50 % of the papers published on GSR focused on the method, either through novel method development or validation of current procedures, while ca. 30 % focused on the GSR trace and its interpretation [10]. This trend is also confirmed by funding attributed to research carried out on forensic science in the UK with much more funding attributed to research focussed on developing technological outputs (69.5 %) compared to fundamental forensic research (19.2 %) [14].
While new techniques are struggling to find their way in GSR examination, the exploration of new technologies is an important part of the Action Plan for the European Forensic Science Area 2.0 [19,20]. The development of interpretation models and quality assurance of implemented procedures were also described as important objectives. The action plan is divided into three pillars to shape the future of forensic science until 2030 [19,20]:
• The first pillar aims at meeting the future for example by developing new technologies for the crime scene and exploring the potential of artificial intelligence (AI).
• The second pillar aims at strengthening the impact of forensic results for example by developing forensic interpretation. Data sharing and multidisciplinary approaches are also considered important.
• The third pillar aims at demonstrating the reliability in forensic results by developing forensic science fundamentals such as transfer, persistence and prevalence studies, understanding human factors and assuring quality and competence.
The focus on the development of technologies and tools is particularly emphasised in the first pillar including a specific section (i.e.,
new tools and emerging technologies). It is also mentioned several times in the second pillar (i.e.,
tools for evaluative reporting, tools to share data, forensic data science tools) and third pilar (i.e.,
review technology solutions for intelligence and multidisciplinary approaches and
develop technologies to optimize the recovery of forensic evidence). In the last pilar, it is also noted that
quality and competence is essential to implement new technologies
7
The citations in italics are excerpts, please refer to the action plan for the full sentences and contexts [19,20].
. Surprisingly, the published research on GSR does not seem to diverge so much from the expressed R&D needs of forensic science, at least on paper.Stamouli and Walters have previously discussed the R&D needs for gunshot examination [16]. The proposed innovation requirements were:
• The development of suitable reference databases to implement Bayesian interpretation of results. This requires harmonisation of the applied method.
• The need to explore new technologies for the analysis of other types of GSR such as organic residue.
• The further investigation of existing technologies to study bullet holes and shooting distances.
Quality assurance, proficiency testing and training of new forensic scientists were mentioned as particular challenges of ENFSI laboratories. This confirms that interpretation, OGSR analysis and quality of current and future technologies are seen by different stakeholders as crucial future developments. However, Stamouli and Walters identified the main barriers to innovation in practice as a lack of resources rather than unwillingness or capability of ENFSI gunshot examiners [16]. They also suggested that the high demands of routine work hamper practitioners to publish the results of their research projects. These aspects will be further explored in the next sections.
3.2 GSR practice survey
45 completed responses were received in reply to the survey (see SI-2 for a visual summary of the answers). General trends were identified and used as a basis for the round table discussions. However, it is important to note that the results of this survey have several limitations, particularly because no personal information were collected from the participants in order to respect their anonymity. Among other factors, participants’ education and training,
8
While some participants may be technicians, others may have a forensic, material science or chemistry degree (Bsc, Msc, PhD). A person specifically trained in surface analysis may be less receptive of novel approaches focused on bulk analysis [17].
their dedication to the topic 9Some participants indicated that they were not only working on GSR, but also on other areas of expertise such as soil.
Ability refers to the opportunities (availability, time) as well as the capability, as academic literature may not be adapted reading material for practitioners.
The majority of respondents (n = 30) handle 10–100 cases per year, but 10 experts indicated that their laboratory receives more than 100 cases per year (
SI-1: Fig. 1). Their main tasks are laboratory work and reporting, while most experts rarely attend the crime scene or give testimony for court (
One third of the respondents had not yet encountered heavy metal free (HMF) ammunition in their practice, while ca. 10 % encounter it regularly (in 20–50 % of their cases). As this was a free text question, the answers varied somewhat between participants. Thus, one respondent indicated that the encountered HMF ammunition (ca. 20–30 %) come mainly from police officers’ firearms or secondary transfer from police officers, while one respondent indicated that tagged police ammunition was excluded from the given percentage (≤5 %). Finally, one respondent stated that she/he had occasionally come across HMF ammunition as reference material.
As expected, respondents indicated that they mainly work with stubs and SEM-EDS (
see SI-I: Figs. 5 and 11). The specimens are mostly collected from the right and left hands, followed by clothes and face (see
Questions were asked about the delay after which GSR were no longer searched for. Most respondents indicated that there is no maximum delay, and the relevance to collect GSR is considered on a case-by-case basis (SI-1:
Fig. 7). In practice, it is usual to collect GSR up to 8 h (see
The delay between sample collection and analysis is typically within one month, although it can vary significantly, even within a single laboratory. Notably, nine respondents indicated that delays could exceed three months. One respondent attributed these prolonged delays to backlogs.
Responses from the survey informed that reports are generally addressed to investigators and prosecutors, rarely for intelligence purposes and almost never to private parties. This may indicate that the respondents work mainly for national or regional forensic laboratories, which can be expected as the survey was sent through the ENFSI communication channels.
The results are often reported as a number of particles, complemented by an interpretation (
Statements were reformulated using the free version of DeepL Write to avoid linguistic recognition. The original meaning was preserved.
:− For example, if more than 1 characteristic particle is detected: this indicates that the person has discharged a firearm, been in the vicinity of another discharging firearm, or had physical contact with a surface to which characteristic GSR particles have adhered.
− Possible explanations are formulated for the presence or absence of GSR. A positive result is defined as a population of 7 characteristic and consistent particles (at least one of which must be characteristic
12
This statement is actually very close to the first but give precisions about the type of particle (at least one characteristic particle among others).
).− High levels of GSR may be found in environments frequented by firearms users or where firearms are stored. Therefore, information on this should be obtained.
16 respondents (36 %) indicated that they use a likelihood ratio (LR) approach or at least formulate two alternative hypotheses. Among these, eight use both a threshold and a LR approach, and only two report their results as a calculated LR value.
Interestingly, almost all respondents consider additional information when interpreting their results from ammunition type to contextual information about the case, as well as GSR persistence and contamination risks (see
The respondents agree on the fact that their results are generally well understood (
SI-Figure 17), but 16 stated that their results are sometimes contested. Among the 45 respondents, 7 never (and 12 rarely) reported their results in court. The experts presenting their results in court noted that case-specific issues were often raised, as well as questions about trace transfer, persistence, and interpretation at the activity level (
While 37 respondents regularly attend conferences and 16 have already published in a scientific journal (
SI-1: Figures 21, 23 and 24), 38 have no or little time for R&D (
SI-1: Figure 25). 36 respondents work in an accredited laboratory. Accreditation is generally seen as having both advantages and disadvantages for R&D (
Regarding R&D, 29 experts indicated that transfer and persistence studies are a priority, and 23 also considered prevalence and contamination as very important (see
To identify perceived needs for innovation, open-ended questions were posed to the experts. First, the experts were asked to specify GSR-related innovations implemented in their laboratory in the past decade (see
The experts were then asked what innovation they would like to see developed / implemented in their laboratory (see
These results highlight that research and perceived needs may not be as far apart as originally thought. However, the use of the term innovation and the way in which the written information was pooled together may have skewed the results towards technologies (and methodological improvements) rather than forensic data collection and interpretation. This is supported by the multiple answers about R&D priorities in which transfer, persistence, prevalence and contamination data were considered essential by the respondents (see Fig. 11). It may also be linked to the fact that the R&D question focused on research in general, whereas the two innovation questions were about the respondents’ laboratory. Thus, while GSR experts would like to see more research on forensic data, their laboratories may not have the resources for such research. This again points to similar difficulties faced by academics and practitioners as both struggle to launch more complex and less immediately rewarding research.
Finally, open-ended questions were asked to the experts about the main difficulties they encounter in their practice (
17 respondents consider that HMF ammunition and/or OGSR analysis are future challenges of GSR examination. 15 experts also consider interpretation issues to be an important challenge. This confirms that experts expect HMF ammunition to increase in the future and that interpretation is and will remain an important challenge of their forensic practice. Method development (including AI tools to automatise some repetitive tasks) and quality, as well as resources, were also mentioned by some experts (see
Figure SI-30). Few new elements were brought up by the 10 participants that answered the question “Do you have other comments about GSR examination?” (see
Figure SI-31). The analysis and interpretation of free text answers have a subjective component, and summaries should be considered with caution. To show the difficulties of interpretation, a few examples are given below.
13
The sentences were modified using DeepL write to avoid linguistic recognition.
One respondent wrote: “GSR examination should be able to identify the shooter unequivocally. That's what the law requires.” This was classified as an interpretation challenge ( Fig. 14). It may also be important to point out at the fact that forensic scientists do not have to provide unequivocal answers to the questions of the legal system if they scientifically cannot [22]. Indeed, they must consider forensic science possibilities and limitations in parallel to legal requirements [12]. The respondent may have intended to convey that while GSR plays an important role in shooter identification, its use should be accompanied by strict quality controls to minimize uncertainty. The answer may have then be classified in method quality instead. Fortunately, the answer included other elements that allowed to confirm a classification in both interpretation and method quality challenges. The answer of another respondent included “Therefore, all new colleagues need to be trained to a high level to be aware of possible errors in methods and interpretation of results.” can also be classified as an interpretation and method quality challenges, or it could alternatively be seen as a training issue (i.e., staff issue in Fig. 14).Therefore, the qualitative aspects are more significant than the quantitative interpretation and will be further addressed and developed in the roundtable discussion.
3.3 Round table
The round table discussions were divided in three main topics followed by interactions with the workshop participants.
3.3.1 Barriers between practice and academic research
The barriers between practice and academic research were first discussed, particularly the discrepancies between the published GSR research ( Fig. 3) and the R&D priorities suggested by the GSR experts ( Fig. 11). The importance of collaboration between practitioners and academics was emphasized, with a recommendation for experts to better articulate their needs and engage in cooperative efforts with the academic community. However, several challenges to such collaboration were noted, including physical distance, limited resources, and time constraints. Additionally, academic projects often have a short-term focus and prioritise fundamental innovations over practical needs, as already acknowledged in the literature [11]. Limited access to casework instrumentation (i.e., SEM-EDS) and strict regulations on firearm use were highlighted as significant limitations. Efforts to bridge the gap between practice and academia must yield mutual benefits. For instance, it was questioned whether proposed methods offer practical advantages over current practices and whether they are economically viable. Similarly, the applicability of research findings to practice was scrutinized, particularly regarding the adoption of validated procedures. Conversely, academics might also challenge established practices, which are often inflexible and may rely on outdated methods, including those used for data interpretation.
In current systems, it seems easier for practitioners to let academics do their own research and subsequently select what is relevant (which is usually very little). Alternatively, practitioners may conduct their own research, although this is rarely published. It is also more convenient for academic scientists to carry out research without additional constraints, as fully validated methodologies are not required in the early stages. Moreover, proof-of-concept results are easier to publish in scientific journals because they do not yet need to address trace uncertainty and interpretation. The siloed nature of disciplines was also mentioned to explain the gap between practice and research. This is due to certain research activities requiring instruments and expertise that fall outside the remit of their working group (e.g., liquid chromatography / mass spectrometry instruments useful to analyse OGSR are generally found in the drug or explosives sections). To address this disconnect, developing networks and fostering increased collaboration within and between laboratories and universities were suggested as particularly important. However, it seems that such collaboration can only succeed if the participants make significant efforts to establish it or if systematic changes are introduced to facilitate constructive cooperation [11,15,23].
3.3.2 Need and utility of forensic case data
The
need and utility of forensic case data (also sometimes named “ground truth data” or “forensic fundamentals”) was then discussed. It was seen as essential by all round table participants as data are generally missing to support interpretation. However, several difficulties were identified. First, the fact that each case circumstances are specific complicates the use of published data in real cases. It is indeed impossible to cover all possible scenarios and parameters (e.g., firearm, ammunition, number of shots, activity). This is a constraint of forensic science practice as traces are produced in uncontrolled conditions and environment
[12]. As sharing real case data is restricted by legal regulation, collaborative work was seen as essential to collect realistic but unrestricted data. However, it is difficult to gather willing participants and coordinate the results quality (for example in “Monopoly” projects carried out by ENFSI members
14
Finally, the complexity of such studies was briefly touched upon, as it generally is impossible to conduct GSR research in an environment such as those encountered in practice. Indeed, in addition to the authorisation to use firearms, such research also requires access to a shooting range.
15
On the rare occasions when a firearm can be discharged in an open-air environment, the location is usually associated with army or police activities (and it is therefore difficult to guarantee a contamination-free environment).
A shooting range is usually subjected to a large number of discharges, resulting in highly contaminated environments. This was particularly discussed during the workshop, but no alternative solution was found even for research carried out by the ENFSI laboratories. 16As regulations vary from country to country, some forensic laboratories have more possibilities than others. Similarly, academic research on GSR is facilitated in countries with more permissive regulations on the use of firearms and for those rare academic institutions that have a shooting range within their laboratories. Access to a white room may also not be granted to all researchers.
It is important to differentiate background noise from negative controls. Negative controls ensure the laboratory in which the specimens are extracted and analysed is clean (i.e., it is a quality control), while background noise cannot be entirely erased as it is part of the environment in which the research takes place. Cleaning offices and clothes has generally proved insufficient to remove all contamination (similarly to observations made in real cases). Thus, background noise controls aim at characterizing the contamination to include it in the data interpretation.
3.3.3 GSR research priorities
Finally, other relevant GSR research priorities were also addressed. While there is a recognised need to explore more HMF ammunition and OGSR approaches, it was remarked upon that the importance of OGSR has been discussed for over 20 years with little advancement. Although it was suggested that the evolution of the ammunition composition should be closely monitored, significant changes are likely to occur only when deemed necessary (i.e., adoption of a reactive approach due to current time and resource constraints). However, a workshop participant warned about the issue as the occurrence of HFM ammunition raised so quickly in their cases that they were unprepared to deal with such cases. Alternatives to analyse IGSR particles were also discussed, but only a gain in rapidity or portability without losing in reliability and sensitivity was considered interesting. Until now, the benefit of portable instruments (e.g. XRF) is still generally considered insufficient. A debate reported in the literature emphasised the importance of anticipating and proactively participating in changes to avoid the risk of exclusion from such developments [26]. However, achieving this requires flexible management to dismantle silos within forensic laboratory sections and foster greater collaboration with crime scene investigators (whether scientists or technicians) who may need to operate new technologies [23,27]. Current research on luminescent markers was cited as an example of irrelevant research if not conducted in collaboration with the ammunition manufacturers, given the existing challenges associated with adding tagging elements to police ammunition.
The results of the survey and roundtable discussions expressed relatively similar needs to those expressed in the literature section ( Table 1). Indeed, keeping up with trends and new technologies was also mentioned by the experts though at a lower priority than in the literature (e.g., the implementation of AI tools to improve the data processing or the development of methods to exploit other GSR types). The need for more data to improve the interpretation of the results is expressed as a priority ( Figs. 11 and 13) and quality assurance, optimisation and harmonisation of current (and future) methods remain a preoccupation of GSR experts. However, during the round table, experts agreed that published research had relatively little impact on their practice either because it did not respond to practical needs (e.g., development of new but complex methods for IGSR analysis that bring no new information), are not implementable (e.g., development of luminescent markers that could be added to ammunition powder without collaborating with ammunition manufacturers) or are too disconnected to GSR practice (e.g., collection of forensic science data on the transfer, persistence and prevalence of GSR with other methods than those currently used in practice). This also questions about the durability of the collected data when novel methods are implemented, or current methods are improved.
4 Bridging the gap between research and practice
A lack of resources (i.e., time, budget and expertise) was consistently mentioned in the literature, survey and round table discussions to explain the lack of relevant research conducted in both academic and forensic practice settings. However, previous studies also indicated that the available resources are often not allocated to forensically relevant research [9,10,14,28]. This may be explained by systemic, rather than purely economic, issues [23,28,29].
On the one hand, practitioners would like to conduct their own research (e.g., acquiring data useful for case interpretation using realistic settings and current instrumentation). For example, the study of the prevalence of GSR particles on POIs arrested by police - but not involved in shooting incidents - is of particular interest to GSR experts in order to assess the risk of contamination in real-case contexts [30,31]. However, their primary tasks and priorities typically involve responding to the continuous demands of the legal system. Caseloads can be heavy, and efforts are focused on keeping the system operational (e.g., repairing and maintaining instrumentation, managinging case backlogs, ensuring quality and accreditation, covering staff absences, or training new personnel). Pooling resources between ENFSI laboratories has been mentioned as a potential way forward, exemplified by the recent study on the prevalence of GSR in different population categories involving numerous forensic laboratories [32]. However, such projects often face challenges (e.g., laboratories only contributing marginally, lack of comparability of the acquired data). While solutions were formulated, they are difficult to implement in the current systems, which are well-oiled machines that fear jamming and backlogs.
4.1 Example of systemic issue hampering relevant development in forensic practice
The examination of gunshot residue has been strongly fixed on inorganic particles analysis using SEM-EDS instruments for more than 40 years [33–35] . While this is rarely questioned by GSR experts, well versed in the potential of the technique in their routine practice, it is often surprising to external observers [36] . Several explanations can be formulated for this lack of apparent innovation the GSR field. The first is that the instrument is fit for purpose, as it enables comprehensive analysis of the composition, morphology, size and number of GSR particles collected. From this perspective, innovation should primarily focus on enhancing the efficiency of the current method (e.g. faster analysis, automation, etc.). [17] . However, SEM-EDS currently do not entirely avoid backlogs (see Section 2 and SI-Fig. 9 ). While this issue can be solved, as often suggested, by increasing the resources (i.e., number of instruments and staff operating them). It can also be addressed by improving the technology (i.e., enhancing operating speed) or by changing the systems (i.e., implementing forensic triage) [15,23] .
A second explanation relates to the current structure of most forensic laboratories, promoted by quality management and forensic specialisation
[15]
. Indeed, the examination of traces produced by a firearm discharge are often siloed in different accredited sections (e.g. ballistics, firearm identification and GSR). This is due to the different specialities considered important for the examination (e.g., physics, statistics, chemistry, material science), but also to avoid contamination between firearms, ammunitions and the specimens collected on POIs
18
The very small quantities expected on the hands of a POI should ideally be examined in a different section of the laboratory than the reference material and highly contaminated items.
. However, it has the disadvantage to hamper the development of new techniques outside the laboratory or the expertise of current GSR examiners. This is well exemplified by the organic GSR component, that remains largely unstudied because the ”bulk” analysis instruments are found in another section. This is due to both the physical distance between the sections, and the lack of time and knowledge to implement a completely different technique. As previously suggested by Charles et al. [17] “only a new technique offering substantial benefits in terms of analytical performances will gain the favour of these experts and change their analytical paradigm”.This indicates that only a change in the trace submitted to the laboratory (e.g., GSR from HMF ammunition), a novel technique offering the same evidential value for decreased costs or a modification of the current structure and management of forensic laboratories will promote significant development in the GSR field [10,15,23,28] .
On the other hand, research can also be carried out in universities. Though academic systems are also submitted to routine tasks (i.e., teaching) and rules (i.e., metrics achievement, priorities of funding bodies) that hampers out of the box thinking and research. While some research must remain fundamental or technically innovative (as required by the current systems), more impactful development for our society would benefit from a closer cooperation with practitioners and other stakeholders. Not just cooperation in the theoretical sense of multidisciplinarity, as promoted by funding schemes, but real interest from the participants at interconnecting their different disciplines and points of views in order to solve real-life problems. Some examples of relevant academic research were proposed in the literature section (see, for example [37,38]).
4.2 Example of systemic issue hampering relevant research in the academia
The same kind of systemic issues were observed in the academia, where scientists mainly work in isolated disciplinary silos and are increasingly evaluated by biased metrics
[9, 10, 28, 39–41]
. The “publish or perish” motto driving the academia has been increasingly criticised, but it is difficult to move away from it in the current research models and systems
[42–45]
. This drives academic researchers to focus on easily publishable topics in order to achieve quantity, rather than risk wasting effort on relevant topics that do not meet publication criteria
[39]
. Just as forensic practitioners protect themselves from critics through quality systems and by distancing themselves from the context and police services, forensic researchers working in universities tend to distance themselves from practical and political considerations in order to avoid being instrumentalised. In addition to these difficulties a straightforward metrics evaluation system (e.g., number of publications, number of citations, number of views, number of likes, …) ensures that scientists have little room to spend time and effort outside their laboratories to rethink their practice. A change in the way research is evaluated was more broadly proposed in science through the DORA declaration
19
One difficulty of forensic scientists, whether they are experts on GSR, fibres or DNA, is that each case is unique [15,46]. Moreover, secondary transfer questions increasingly come up with the improvement of the instrumentation’s detection limits. Thus, producing knowledge about the prevalence of GSR characteristics in the environment and about the risk of subsequent transfer on the scene and during collection (as it is mainly performed by first respondents) is essential to interpret the meaning of traces found in very small amounts [13]. However, as previously stated, such data will be useful only if acquired with instruments measuring the same characteristics with the same limit of detections (i.e., harmonisation of the results need to be achieved). It should be noted that the introduction of artificial samples as part of interlaboratory tests can play a major role in harmonising detection limits between different laboratories. On the other hand, switching to real samples still brings its share of challenges, as shown by a study in which the analytical performance between the different laboratories was still quite high [47]. This was globally acknowledged as one of the main barriers to collaborative projects as it requires partners to openly exchange about their methods and adapt those for the project (an endeavour that can be limited by accreditation and case load). For fruitful collaboration, GSR experts also need to remain open to novelties and actively participate to research designs. As for academics, they need to get out of their laboratories to understand routine work issues and adapt as much as possible to the needs of practitioners in order to produce more impactful research for society.
4.3 Examples of successful collaborations between forensic scientists and academics
An example of such a collaborative project, designed by GSR experts working in different European laboratories and reviewed in collaboration with an academic scientist, is the recent GSR project on transfer and persistence studies exploitable for activity level reporting. This project stems from the ENFSI EWG Firearms / GSR and is open to all participants to the annual meeting, whether there are working in a forensic laboratory or in academia. Another such example is the Organization of Scientific Area Committees for Forencic Science (OSAC)
20
However this last solution has been deemed by many as impossible to implement in their own context [29]. This suggests institutional and cultural blockages. Thus, breaking the disciplines and systemic silos probably can be achieved only by rethinking the systems and forensic science culture in which we work [15, 28, 54–56]. Moving from personal and institutional performance metrics towards common goals may be the only path toward more impactful research and development in forensic science [11,57].
4.4 Example of a successful systemic change in a forensic laboratory
A recent publication presents how the reorganisation of an operational laboratory structure, functioning and delivery allowed to reduce backlogs and increase positively collaboration between the different stakeholders [23] . The Australian Federal Police (AFP) forensic laboratory adopted a transdisciplinary forensic operating model in addition to the consultancy model already in place since 2013. The new approaches aimed at breaking the silos between the specialities by promoting a integrated and collaborative consideration of submitted items. By shifting the focus from the sections to the whole laboratory, all specialities are considered when a specific case is submitted to the forensic laboratory. This transcends the silo effects, promotes collaborative problem-solving approaches and increases the flexibility and responsivity of the forensic experts, and the lab as a whole. Such a change necessitated structural and physical changes to the facility as well as a change in culture. It was supported by a reformed training programme that not only focused on discipline specific technical skills, but also on transversal forensic science knowledge and communication skills. The latter aimed to provide greater flexibility in staff deployment and improved coordination.
5 Conclusion
This work aimed to evaluate the perceived gap between GSR research and practice, and to collaboratively assess how it might be bridged. The literature review and GSR practice survey, although limited in scope, highlighted that needs of research and practice were not as far apart as originally thought. The main needs expressed by the European Forensic Science Area (for forensic science in general) and by the forensic experts consulted in this study (for GSR in particular) focused on three priorities: improving technologies, developing interpretation frameworks and studying forensic science fundamentals (e.g., trace transfer, persistence and prevalence). All three topics are well represented in the literature, although both forensic academics and practitioners agree that there is relatively little specific research that has produced directly useful data or developments.
It was interesting to note that both academics and practitioners struggle with the same lack of time and resources to carry out more relevant research. Routine workloads and the constant evaluation of their work generally promote more straightforward developments (e.g., technical improvements), to the detriment of research that requires greater efforts in data acquisition and interpretation (e.g., persistence studies).
During the round table discussions, the importance of collaboration was particularly emphasised as a mean to increase the relevance and utility of forensic research in the field of GSR. While the annual meeting of the ENFSI EWG Firearms / GSR, which resulted in this publication, is a good example of such collaboration, it tends to be attended by GSR experts rather than academic researchers. With a few exceptions, the current systems only superficially rewards practitioners-academics collaboration. The amount of work and expertise required for true collaboration and interdisciplinary projects is largely underestimated and undervalued by both the forensic laboratories and universities.
Thus, it appears that increasing collaboration will only be successful if significant efforts are made by the participants to build trusting relationships, or if systemic changes are introduced to facilitate more constructive cooperation. This may require a shift involving the forensic community at large, transcending the current disciplinary approach adopted by forensic laboratories and academic research. A change in culture, moving from the techniques and sub-disciplines towards the concepts of trace and crime investigation, has been suggested as a path forward to achieve more relevant and coordinated forensic endeavours. To achieve this goal, following the examples of AFP
[23] and DORA,
21
CRediT authorship contribution statement
Céline Weyermann: Writing – original draft, Data curation, Formal analysis, Conceptualization. Virginie Redouté Minzière: Writing – review & editing, Formal analysis, Conceptualization. Thomas Tilborg: Writing – review & editing, Data curation. Kal Chana: Writing – review & editing, Formal analysis. Hervé Ménard: Writing – review & editing, Formal analysis. Bart Nys: Writing – review & editing, Formal analysis. Ruediger Schumacher: Writing – review & editing, Formal analysis. Sébastien Charles: Writing – review & editing, Formal analysis, Conceptualization
Declaration of Competing Interest
None
Acknowledgments
The authors would like to thank all the GSR experts who responded to the survey and participated in the discussion. Special thanks also go to Sheila Willis, Olivier Ribaux, Claude Roux and Pierre Margot for discussions on this project. Céline Weyermann, Virginie Redouté Minzière and Thomas Tilborg also wish to acknowledge the
Appendix A Supporting information
Supplementary data associated with this article can be found in the online version at
Appendix A Supplementary material
Supplementary material
Supplementary material
Supplementary material
Table 1 –
| 2003–2012 | 2013–2022 |
TOTAL
(20 years) | |
| # articles | 159 | 362 | 521 |
| # sources | 53 | 114 | 148 |
| % forensic source | 34 % | 25 % | 29 % |
© 2025 The Authors