Blood alcohol levels in road traffic accidents: Factors associated and the relationship between history of alcohol consumption and blood alcohol level detection.

INTRODUCTION: Alcohol consumption contributes to a significant number of road traffic accidents (RTAs), and data regarding the reliability of history and blood alcohol content (BAC) in RTA victims are scant.
METHODOLOGY: This retrospective study was conducted in the emergency departments (EDs) over 6 weeks. All adult RTAs presenting within 12 h of the incident were included for analysis.
RESULTS: The study cohort included 369 RTA patients, with the mean interval before presentation being 3 h (standard deviation: 2.22). Two-wheeler accidents (77.2%) were the predominant mode of injury. Usage of a helmet and seat belt was documented in a meager (6.4% [17/267] and 8.8% [3/34], respectively). A positive history of alcohol consumption was reported by 19.5% of cases (72/369). However, BAC was detectable in 30.1% of cases (111/369), with an alarming 19.78% (73/369) being above the legal limit for driving. Nearly 77.5% (86/111) of those who tested positive for alcohol consumption were driving the vehicle involved. Positive BAC levels showed a significant association with young age (18-39 years), male gender, two-wheeler usage, and between 5 PM and 12 AM.
CONCLUSION: A history of alcohol consumption leading to an RTA is not reliable in the ED. Hence, measuring BAC levels in all RTA patients provides an objective and reliable form of documentation for medico-legal purposes. Copyright:

INTRODUCTION

India has a road network of 5,903,293 km, which is the second largest in the world. Road traffic accidents (RTAs) result in the largest number of fatalities and injuries worldwide by injuring 50 million and killing 1.2 million people each year.[1234] Drunken driving is a major contributor to traffic deaths in India, which is responsible for 70% of road fatalities.[567] As blood alcohol levels increase, there is progressive loss of driving ability due to an increase in reaction time, false confidence, impaired concentration, and decreased auditory and visual acuity. Das et al. reported that alcohol was incriminated in 2%–33% of injuries and 6%–48% of deaths.[8] Esser et al. found that nearly one in six RTA patients is suspected of or reported alcohol use.[9] Studies in India have relied either on postmortem samples or history and examination, with a few utilizing breath analyzers. Tabin et al. used breath analyzers and blood alcohol levels estimated by gas-liquid chromatography, with 22% of cases testing positive.[10] There is a need for capacity building and reporting for substance use in RTA, as evinced by the discrepancies between hospital-based studies and government reports.

Under Section 185 of the Indian Motor Vehicles Act, India, driving with a Blood alcohol concentration (BAC) level above 30 mg/dL is a punishable offense. Evaluation of an RTA patient for alcohol consumption thus carries important legal implications. The existing studies do not tell us if the alcohol consumption is by the driver, pedestrian, or the pillion rider, and hence, the risk of accidents among drunken drivers cannot be assessed. In India, blood alcohol levels are rarely available in most emergency departments (EDs) or trauma centers. Furthermore, a history of alcohol consumption may be unreliable or not forthcoming. No study has been conducted to assess the reliability of history of alcohol consumption in RTA patients. In the present study, we sought to determine the prevalence of alcohol consumption (including above the permissible limit) and the correlation between history and BAC levels and study the various factors associated with alcohol intake. We also studied whether positive BAC levels were associated with poor patient outcomes.

METHODOLOGY

Design

We conducted a retrospective analysis of adult patients presenting with an RTA to determine the factors associated with positive BAC levels and their correlation with a history of alcohol consumption.

Setting

We conducted this study in the adult ED of Christian Medical College, Vellore, which is a large tertiary care hospital in South India with 2700 inpatient beds. The adult ED has 49 beds with about 75,000 admissions yearly.

Participants

Adults who presented with an alleged history of an RTA presenting within 12 h of the incident, over the 6-week study period from June 15, 2018, to July 30, 2018, were included in the analysis. RTA victims aged below 18 and those presenting after 12 h of the incident were excluded from the study.

Variables

Patient data were obtained through the hospital's electronic database. Details of history and physical examination findings and demographic details were recorded on a standard data collection sheet. The variables included age, sex, mechanism of injury, mode of transport, place of occurrence, usage of helmets and seat belts, triage priority, and the type and severity of injury. Triage priority level was defined as follows:

Triage priority 1: Patients with airway, breathing, or circulation compromise, or head injury with Glasgow Coma Scale (GCS) <8

Triage priority 2: Patients with stable airway, breathing, and circulation with long-bone injuries, dislocations, stable abdomino-thoracic injuries, and head injury with GCS 9 or more

Triage priority 3: Hemodynamically stable patients with minor trauma.

The severity of injury was assessed using the Revised Trauma Score (RTS). The regions of the body involved were documented, and only significant deep injuries were considered for analysis. These included penetrating trauma, fractures, dislocations, head injury, and other internal organ injuries. Minor abrasions, superficial lacerations, and minor soft-tissue injuries were considered to be superficial injuries and were not included in the analysis. If indicated, initial stabilization was followed by referral for further management to other departments, with admission to the respective wards.

Outcome variable

The ED outcome (alive/dead/left against medical advice) and hospital outcome (alive/dead) of the patients were compared between those with positive and negative BAC levels. The history of alcohol consumption was correlated with the BAC levels sent from the ED.

Laboratory test (BAC levels)

BAC levels are sent as a routine protocol for all RTA victims in our ED. Whole blood was collected into 2-ml BD Vacutainer® tubes with gray rubber stoppers containing 3-mg sodium fluoride and 6-mg sodium Ethylenediaminetetracetic acid (EDTA) as additives to prevent coagulation and fermentation as per the Clinical and Laboratory Standards Institute standards. These were submitted to the department of clinical biochemistry, for further analysis. The specimens were centrifuged, and plasma was separated and frozen at −20°C till analysis and then discarded. The analysis was performed using a Roche Cobas 8000 analyser (Roche Diagnostics, Mannheim, Germany), based on the alcohol dehydrogenase principle. In this method, the enzyme alcohol dehydrogenase oxidizes blood alcohol to acetaldehyde. The corresponding increase in absorbance at 340 nm is due to the generation of NADH+ from NAD+ (nicotinamide adenine dinucleotide), which is proportional to the concentration of alcohol present in the sample. Interassay coefficient of variation was <5% during the study period. The method is highly specific with crossreactivity of 0.8% for N-propanol and 2.8% for N-butanol at a concentration at 2000 mg/L.[11]

Statistical analysis

The data were entered into Microsoft Excel (version 15.12.3) and were analyzed using Statistical Package for Social Sciences (IBM Corp. Released 2015. IBM SPSS, Version 23.0, Armonk, NY, USA). Continuous variables were presented as mean (standard deviation [SD]). Categorical and nominal variables were presented as percentages. The Chi-square test was used to compare dichotomous variables. Univariate analysis was done to determine the factors associated with positive BAC levels. Similarly, the outcomes associated with high BAC levels were compared. For all tests, a two-sided P < 0.05 was considered statistically significant.

Ethical considerations

This study was approved by the Institutional Review Board (IRB Min. No. 10559 dated 08.08.2018), and patient confidentiality was maintained using unique identifiers and by password-protected data-entry software with restricted users.

RESULTS

During the study period of 6 weeks, our ED attended to 7044 patients, with 696 being RTA victims. Pediatric and adolescent patients, patients for whom BAC levels were not sent by the ED team, and patients whose blood samples got hemolyzed and could not be processed were excluded from the study. The final study cohort included 369 patients on whom BAC levels were available [Figure 1].

Figure 1

Strobe statement

The mean age of the victims was 39.78 (SD: 15.46) years, and there was a male predominance (85.6%). The baseline characteristics and events related to the incident are summarized in Table 1. Two-third of the incidents (77.2%) were two-wheeler accidents. The mean time taken to present to the ED was 3 (SD = 2.2) h. Majority of the accidents took place within a 420-km radius of our hospital. Physical examination findings of these patients are shown in Table 2. The RTS was used to characterize the severity of the trauma. Only six patients had an RTS <4.

Table 1

Baseline characteristics (n=369)

Characteristics	Number (%)
Mean age (years), mean (SD)	39.78 (15.46)
Mean interval before presentation (h), mean (SD)	3 (2.22)
Male gender, mean (SD)	316 (85.6)
Triage priority, n (%)
Priority 1	76 (20.6)
Priority 2	241 (65.3)
Priority 3	52 (14.1)
Brought to the ED by, n (%)
Self	23 (6.2)
Relatives/friends	324 (87.8)
Strangers	19 (5.2)
Undocumented	3 (0.8)
Time of incidence, n (%)
8 am-5 pm	162 (43.9)
5 pm-12 am	149 (40.4)
12 am-8 am	57 (15.4)
Unknown	1 (0.3)
Time of arrival to ED, n (%)
8 am-5 pm	127 (34.4)
5 pm-12 am	170 (46.1)
12 am-8 am	72 (19.5)
Type of road of the incident, n (%)
Highways	34 (9.2)
Smaller roads	303 (82.1)
Other places	2 (0.5)
Undocumented	30 (8.2)
Mode of incidence, n (%)
Two-wheeler	285 (77.2)
Four-wheeler	34 (9.2)
Pedestrian	32 (8.7)
Auto	9 (2.4)
Other vehicles	5 (1.4)
Lorry or bus	4 (1.1)
Driver or passenger, n (%)
Driver	250 (67.8)
Passenger	81 (22)
Not applicable	20 (5.4)
Unknown	18 (4.8)

SD: Standard deviation, ED: Emergency department

Table 2

Examination findings at presentation to the emergency departments (n=369)

Characteristics	n (%)
Bradycardia <60/min	8 (2.1)
Tachycardia >100/min	89 (24.1)
Systolic blood pressure <90 mmHg	20 (5.4)
Systolic blood pressure >140 mmHg	70 (18.9)
Tachypnea >20/min	293 (79.4)
Glasgow Coma Scale
13-15	331 (89.7)
9-12	13 (3.5)
3-8	25 (6.8)
RTS
≤4	6 (1.6)
>4	363 (98.4)

RTS: Revised Trauma Score

There was very poor compliance to road safety rules. Only 6.4% (17/267) of those driving two-wheelers were documented to have worn helmets and 8.8% (3/34) of those in four-wheelers were documented to have used seat belts. Among the 17 people using helmets, 12 (70.6%) did not sustain any serious head injury and none sustained any facial injury. Among the remaining 250 not using helmets, 57 (22.8%) sustained head injuries and 41 (16.4%) sustained facial injuries.

A positive history of alcohol consumption was recorded in only 19.5% (72/369) of all patients. However, BAC levels were positive in 30% of these patients (111/369). A positive history of alcohol consumption was absent in 44.1% (49/111) of those with positive BAC levels. In fact, 65.8% (73/111) of the patients had BAC levels above the legal limit of 30 mg/dL (as stipulated under Section 185, The Motor Vehicles Act). Among the 53 women tested, 5 (9.4%) had detectable alcohol levels, with 1 being above the legal limit for driving, yet only one of them had a positive history of alcohol consumption. The association of detectable blood alcohol levels with age group, triage priority, and role in the RTA is shown in Table 3.

Table 3

Association of detectable blood alcohol levels with risk factors and outcome (n=369)

Variable	BAC positive (n=111)	BAC negative (n=258)	OR	95% CI	P
Age 18-39 (years)	73 (65.8)	128 (49.6)	1.95	1.229-3.096	0.005
Male sex	106 (95.5)	210 (81.4)	4.85	1.874-12.532	0.001
Two-wheeler	97 (87.4)	192 (77.4)	2.38	1.273-4.455	0.007
Between 1700 and 0000 h (n=368)	64/110 (58.2)	85/258 (32.9)	2.83	1.789-4.482	<0.0001
Triage priority 1	32 (28.8)	44 (17.1)	1.97	1.167-3.325	0.011
Triage priority 2	73 (65.8)	168 (65.1)	1.03	0.6443-1.6438	0.904
Driver	86 (77.5)	164 (63.6)	1.97	1.1811-3.2916	0.009
Passenger/pillion rider	19 (17.1)	62 (24)	0.65	0.3690-1.1551	0.143
Helmet not worn (n=250)	84 (75.6)	166 (64.3)	1.21	0.414-3.561	0.723
Seat belt not worn (n=31)	9 (8.1)	22 (8.5)	2.96	0.1388-62.95	0.276
RTS ≤4	5 (4.5)	1 (0.4)	12.12	1.399-105.004	0.024
Head-and-face injuries	71 (70)	129 (50)	1.78	1.123-2.806	0.014
Admission to wards	38 (34.2)	80 (31)	1.16	0.722-1.858	0.542
Discharged stable	53 (47.7)	155 (60.1)	0.61	0.388-0.9505	0.029

BAC: Blood alcohol concentration, OR: Odds ratio, CI: Confidence interval, RTS: Revised Trauma Score

Two-third (65.8%: 73/111) of the patients with positive BAC levels belonged to the age group of 18–39 years. Among the priority 1 patients evaluated, 42.1% (32/76) had detectable BAC levels. We also evaluated the role of the 111 patients with positive BAC levels in the RTA. A vast majority (77.5%: 86/111) were driving a vehicle, while 17.1% (19/111) were pillion riders/passengers and 4.5% (5/111) were pedestrians. RTAs occurring between 5:00 PM and 12:00 AM were significantly more likely to present with detectable BAC levels when compared to those occurring at other times (odds ratio [OR]: 2.83; 95% confidence interval [CI]: 1.789–4.482, P = 0.007).

The association between deep injuries sustained and positive blood alcohol levels is shown in Table 4. We found that patients with positive BAC levels were almost twice as likely to present with head or facial injuries (OR: 1.78; 95% CI: 1.123–2.806, P = 0.014), when compared to those without detectable blood alcohol levels.

Table 4

Distribution of injuries with respect to positive blood alcohol concentration levels (n=369)

Region	BAC positive (n=111), n (%)	BAC negative (n=258), n (%)	OR (95% CI)	P
Head	34 (30.63)	45 (17.44)	2.2745 (1.3698-3.7766)	0.001
Face	19 (17.12)	30 (11.63)	1.5696 (0.8413-2.9282)	0.156
Neck	2 (1.80)	1 (0.39)	4.7156 (0.4231-52.552)	0.207
Thorax	8 (7.21)	12 (5.65)	1.5922 (0.6322-4.010)	0.323
Abdomen	4 (3.60)	10 (3.88)	0.9271 (0.2845-3.022)	0.900
Extremity	53 (47.5)	109 (42.25)	1.2491 (0.7989-1.953)	0.329
Spine	4 (3.60)	6 (2.39)	1.5701 (0.4343-5.677)	0.491

BAC: Blood alcohol content, OR: Odds ratio, CI: Confidence interval

Overall, more than half (56.4%) were discharged stable from the ED, and 32% were admitted for further management. Comparison of ED and hospital outcomes of those with and without detectable BAC levels is shown in Table 5.

Table 5

Emergency departments’ outcome and hospital outcome (n=369)

	Number overall, n (%)	Patients with positive BAC levels, n (%)	Patients with negative BAC levels, n (%)
ED outcome	n=369	n=111	n=258
Discharged stable	208 (56.4)	53 (47.7)	155 (60.1)
Admitted in hospital	118 (32)	38 (34.2)	80 (31)
Dead	1 (0.3)	1 (0.9)	0
LAMA	42 (11.4)	19 (17.2)	23 (8.9)
Hospital outcome	n=118	n=38	n=80
Discharged stable	115 (97.5)	36 (94.7)	79 (98.7)
Dead	2 (1.7)	1 (2.6)	1 (1.3)
LAMA	1 (0.8)	1 (2.6)	0

ED: Emergency department, BAC: Blood alcohol content, LAMA: Left against medical advice

DISCUSSION

Our study showed the association between BAC level positivity and the various factors associated with it. This is one of the very few studies done on BAC levels in the ED. It also highlights the unreliability of history given during an RTA and stresses the need for an objective and quantitative test like BAC level, which is not available or used in most of the EDs in the developing world.

Alcohol is a major risk factor in all RTAs. In our study, we see a third of the patients who presented with RTAs being tested positive for BAC levels. As driving under the influence of alcohol causes imprecise and irrational behavior, there is an increased chance of meeting with an accident. The WHO reports that risks of an accident increase with blood alcohol concentrations (BAC), significantly above 40 mg/dL. A systematic review conducted by Das et al. estimated that alcohol was responsible for 2%–33% of injuries and 6%–48% of deaths in RTAs.[8]

Most studies in India have reported alcohol consumption based on postmortem studies (Baruah and Chaliha), patient history and case notes (Esser et al. and Mohan et al.), or breath analyzers (Kaushik et al.).[6791112] Roui et al. in Germany demonstrated that although there may be slight differences between breath alcohol concentrations and blood alcohol concentrations, breath analyzers have weaknesses which can be supplemented by blood alcohol levels.[13] Tabin et al. estimated a 22% prevalence using gas-liquid chromatography in addition to breath analyzers in their study.[8] In Vietnam, Nguyen et al. reported detectable blood alcohol levels in 43.2% of patients.[5]

Motorization has enhanced the lives of many individuals and societies, but the benefits have come with a price. RTA injures or disables between 20 million and 50 million people a year, with the most vulnerable road users being pedestrians, cyclists, two-wheeler riders, and passengers on public transport.[2] Our study showed the dismal road safety practices in our country, with only a minor proportion using helmets (6.4%) or seat belts (8.8%). The majority of patients wearing helmets were protected from head and facial injuries, and in spite of awareness campaigns and helmets being mandatory by law, their usage remains poor.

In the present study, nearly a third (30%) of all patients tested had detectable BAC levels, with only 19.5% (72/369) giving a positive history of alcohol consumption. Alarming is the fact that two-thirds (65.8%) of those with detectable BAC levels were in fact above the legal limit of driving in India. With almost half of these patients (44.1%) denying consumption of alcohol prior to the incident, we conclusively proved that history of alcohol consumption is very unreliable in RTA cases. Hence, BAC levels should be made mandatory as this single test could tip the scales in a medico-legal fight between the two parties. Another important factor to consider in our study is the mean time of patient arrival to the ED from the time of incident which was 3 h. This is much longer than what is documented internationally in places with good prehospital facility.[14] The reasons for this time lag are a substandard prehospital service in India and our hospital being a tertiary care referral center. According to Holford, alcohol follows a one-compartment model with concentration-dependent elimination. At a BAC of 80 mg/dL, equivalent to the Km, the elimination rate is half the maximum elimination capacity of 230 mg/L/h.[1115] The implication of this time lag is some BAC levels in those who consumed alcohol may have become negative or would have been within the legal limit of driving.

This study clearly demonstrates that in the ED setting of an RTA, a history of alcohol consumption alone will not suffice. BAC levels provide a rapid, reliable, and objective form of documentation, especially in medico-legal cases where drunken driving constitutes a culpable offense.

CONCLUSION

A history of alcohol consumption leading to an RTA is not reliable in the ED. Hence, measuring BAC levels in all RTA patients provides an objective and reliable form of documentation for medico-legal purposes. We, therefore, recommend that BAC levels should be made mandatory for all RTAs and other medico-legal cases in the ED.

Research quality and ethics statement

The authors of this manuscript declare that this scientific work complies with reporting quality, formatting, and reproducibility guidelines set forth by the EQUATOR Network. The authors also attest that this clinical investigation was determined to require IRB/Ethics Committee review, and the corresponding protocol/approval number is IRB Min. No. 10559 dated 08.08.2018. We also certify that we have not plagiarized the contents in this submission and have done a plagiarism check.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Ethical conduct of research

This study was approved by the Institutional Review Board. The authors followed applicable EQUATOR Network (http://www.equator-network.org/) guidelines during the conduct of this research project.