The role of alcohol as a major causal factor in trauma has been highlighted in several studies.1–5 In recent years concern has spread to drugs other than alcohol,6–13, including prescribed or nonprescribed psychotropic drugs.14,15 Studies to date conclude that exposure to alcohol and drugs are associated with a higher risk of injury regardless of the mechanism of injury (traffic crashes, falls, etc.).

Although there appears to be agreement that alcohol and other drugs are important risk factors for injury, there is no consensus on the role of these substances in severity. Some studies found that the presence of these substances increases trauma severity,16–24 whereas others found no such association.14,25–27 However, studies that evaluated the effect of illicit or psychotropic drugs on trauma rarely focused on the use of these substances separately,14,24,27 because the main focus of study was usually alcohol consumption. This leads to important methodological deficiencies in assessing the association between substances other than alcohol and injury severity. For example, drugs with different effects, including opposite effects, on the central nervous system (CNS) are usually pooled for statistical analysis.21–23,26 Moreover, some studies systematically excluded patients with positive tests for benzodiazepines and opiates,21,24,26 because these drugs are frequently used in prehospital care for trauma patients. This makes it impossible to analyze the role of these substances in trauma severity. To these limitations must be added the fact that these studies were frequently based on data obtained from national, regional or hospital-based trauma registries.16,19,21,22,27 These registries record only patients whose medical care provider decided to request an alcohol and/or drug test at the time the patient was admitted to the emergency room.

The factors related to injury severity merit more in-depth analysis. Greater injury severity is strongly associated with increased mortality, more complications, longer hospital stays, and ultimately greater economic costs.28 We believe it is worthwhile to investigate the specific roles of drugs in trauma severity, based on the hypothesis that a separate analysis of the effect of each substance on the CNS (depressants, stimulants and hallucinogens) will indicate a greater magnitude of association for the group of depressant substances, due to the reduced awareness and impaired cognitive and motor function that these agents cause.29 The aims of the present study were thus to analyze the effect of stimulant, hallucinogenic and depressant drugs other than alcohol on injury severity in hospital patients admitted for trauma, and to determine the possible role of alcohol as a modifier of the effect of these drugs.

Of all 1818 patients aged 16 to 70 years who were hospitalized for trauma during the study periods, 1187 (65.3%) could be screened for alcohol and drug use; 609 patients were not screened and 22 rejected it. Of all patients screened, 371 (31.3%) were positive for drugs other than alcohol: 32 (2.7%) for CNS stimulant drugs, 186 (15.7%) for CNS depressants, 78 (6.6%) for hallucinogenics and 75 (5.6%) for polydrug. Alcohol was the substance detected most frequently (295 patients; 24.9%), followed by THC (144; 12.2%), benzodiazepines (199; 16.8%), cocaine (81; 6.8%), opiates (26; 2.2%), methadone (25; 2.1%); barbiturates (7; 0.6%), amphetamines (9; 0.8%) and PCP (2; 0.2%).

When we compared the demographic and clinical characteristics of the five groups according to type of drug consumed (Table 1), we found that age was oldest for consumers of depressant substances, with a median age (interquartile range) of 51 years (42-60), and much younger in patients who tested positive for hallucinogenics (median: 26 years [22-34]). The proportion of women was greater in the subgroup of individuals who tested positive for depressant drugs (44.6%) than in the other four groups. A fall on the same level was the most frequent mechanism of injury in this group (46.2%), in contrast to the stimulant drugs group, in which the most frequent mechanism was traffic crashes (43.5%). In the subgroup of polydrug, alcohol consumption was especially frequent (66.7%). There were no relevant differences between groups in the duration of hospital stay (in days) or mortality.

In the multivariate analysis (Table 2), the estimates of strength of association between each drug group and the frequency of moderate and severe injuries in model 1 showed that alcohol modified the association between depressant drugs and injury severity. This association was nonexistent in the group who had not been exposed to alcohol, whereas alcohol consumption was associated with an increase in injury severity in the group of patients exposed to CNS depressants, with an adjusted aOR of 4.63 (95%CI: 1.37-15.60) for moderate injuries and 7.83 (95%CI: 2.53-24.21) for severe injuries. This modifying effect did not appear in the other groups.

In model 2 (Table 3), when pooled data were analyzed by grouping drugs with opposite effects, no significant associations were found in the adjusted estimates.

A separate analysis of drug subgroups defined according to their effect on the CNS allowed us to identify the specific influence of substances with CNS-depressant effects on injury severity, especially in patients who had consumed alcohol. The results of this study suggest that in the presence of alcohol, substances with a CNS-depressant effect have a strong influence on trauma severity.

We were able to distinguish between cases in which the substance was administered as part of the trauma patient's medical care and cases in which the patient had consumed the substance before the injury. This allowed us to quantify the effect of CNS-depressant substances on severity without the need to systematically exclude patients who screened positive for opiates or benzodiazepines, in contrast to similar studies that excluded these patients.21,24,26

There was a high prevalence of benzodiazepine exposure in our population of injured patients. The influence of benzodiazepines on injury prevention is an area of growing interest. Although the exact frequency of use of these nonprescribed substances is not known, it appears evident that the consumption of benzodiazepine and other psychotropic drugs is increasing.14,37

Particular attention should be given to motor vehicle crashes, which can cause serious injury. In Spain, the percentage of traffic deaths in which psychoactive drug use was detected alone or in combination with alcohol and/or other drugs has doubled in the last five years, and is now 33.3%.33 In this connection, drugs other than alcohol have begun to acquire particular relevance. The conclusions of the European DRUID (Driving Under the Influence of Drugs, Alcohol and Medicines) project38 highlight the need for increased monitoring of substances other than alcohol because their frequency of use is clearly increasing. According to the DRUID study, in some parts of Europe, drivers are more likely to be detected under the influence of drugs than under the influence of alcohol. Moreover, a systematic review by Smink et al.15 found sufficient evidence that exposure to benzodiazepines is related to a greater risk of accidents; although these authors recognize that more research is needed to clarify the relationship between the use of these drugs and injury severity. We found that the effect of benzodiazepines along with other depressants on trauma severity was very high, but only in patients who had also consumed alcohol. The biological interpretation of these results is likely to be related with the known synergistic interaction of combinations of these drugs in CNS depression. This effect impairs driving performance and the ability to react to a potential hazard by causing sedation and increasing risk-taking behavior. Driving under the combined influence of drugs and alcohol has been found to increase the risk of accidents in various ways due to their influence on the driver's vision, awareness, perception, reaction time and concentration.29

Statistical analysis with all patients who were positive for substances in a single group (model 2) allowed us to confirm that pooled data analysis by grouping drugs with opposite effects on the CNS is a potentially misleading strategy. Therefore, some associations between depressant drugs and injury severity may have been masked in previous studies.21,22,26

The profile of patients who tested positive for CNS depressant substances was different from other profiles: female sex was more frequent, age was substantially older, and falls were the most common mechanism of injury. These findings reveal the need for further analysis of the differences in these characteristics compared to the classic profile of individuals with trauma associated with alcohol and other drugs, i.e. young, male, and with injuries related to violence or traffic accidents.3,18,26,39–43 In any case, the results after adjustment for age, sex and mechanism of injury showed that these differences cannot be the cause of the heightened injury severity observed in the subgroup of patients exposed to CNS-depressant substances.

Our systematic analysis was based on all patients between 16 and 70 years of age who were admitted to a trauma hospital, with no other previous selection criteria. This sampling approach decreased the possibility of selection bias. Some previous studies were based on data obtained from national, regional or hospital trauma registries from which the authors selected records that included available information on the exposure to drugs and alcohol.16,19,21,22,27 If patient screening is not used consistently and systematically, the availability of data on substance use will depend on the physician's decision to use or omit the screening protocol at the time the patient is admitted to the emergency department. This uncertainly introduces an important source of selection bias, because physicians may be more likely to request substance use testing for more severely injured patients or for those with signs of alcohol or drug use.

Our study shares a common limitation with previous reports,14,17 i.e., the possible presence of false negative results for alcohol use in patients who were not tested for this substance at the time of admission and who subsequently denied alcohol consumption when interviewed. Fortunately, this was not the case for the other substances analyzed here, because all patients were screened with appropriate laboratory tests. On the other hand, our study could be affected by selection bias: we have not included those traumatized patients who died at the scene of the accident due to the extreme severity of their injuries; frequency of drug detection may be different in these subjects. However, statistics of substance distribution among all drivers killed in traffic accidents in Spain show that alcohol and/or psychotropic drugs are substances frequently found (29.1% of drivers killed in 2015).33 Finally, another selection bias could arise because not all hospitalized trauma patients could be screened. Nonetheless, we have no reasons to suspect an independent association between the presence of drugs and the probability of being screened.

We detected an association between the presence of sedative drugs and injury severity in the subgroup of trauma patients who screened positive for alcohol consumption. We believe further knowledge is needed regarding the effect of drugs other than alcohol on trauma severity. Our results support the additional benefits of systematic screening programs for psychoactive drug use in all trauma patients. The information gained from testing would not only provide important data during clinical assessment, but would also enable the healthcare provider to offer appropriate education, counseling, or other types of intervention.

The corresponding author on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

Cristina Linares Gil.

All authors contributed to the design of the study, writing and final approval of the version to be submitted. S. Cordovilla-Guardia contributed to the literature search, data analysis, data collection, data interpretation and editing. P. Lardelli-Claret contributed to the data analysis and data interpretation. R. Vilar-López contributed to the data analysis and data interpretation. F. López-Espuela contributed in the data interpretation. F. Guerrero-López contributed in the data collection and data interpretation. E. Fernández-Mondéjar contributed to the literature search, data analysis, data interpretation and editing.

This work was supported by the Dirección General de Tráfico, Spain [No. 0100DGT22389] and the Consejería de Salud, Junta de Andalucía, Spain [No. PI-0691-2013].

None.