Gender differences in the consumption of alcohol mixed with caffeine and risk of injury

Abstract

Background:

There has been increasing concern over the risks associated with the consumption of caffeinated alcoholic beverages; however, research in this area remains limited. We examined whether sex differences existed in the relationship between the combined use of alcohol and caffeine (Alc+Caff) and risk for injury.

Methods:

This Emergency Department (ED) study utilized case-control and case-crossover analyses to examine situ session specific Alc+Caff use and injury risk for males and females, while controlling for sociodemographic variables, dose of alcohol and caffeine, other substance use, risk-taking propensity, and context. The sample comprised 2804 aged 18-years or older who who presented to three hospital EDs in British Columbia.

Results:

A relationship between Alc+CAff use and injury was confirmed, such that Alc+Caff use was associated with a higher risk of injury over and above alcohol use on its own. Further, women were found to have a higher risk injury propensity following Alc+Caff use compared to men in both case-control and case-crossover analyses. These results remained even after controlling for demographic factors, risk-taking, context, and other drug use.

Conclusions:

Women may be at a higher risk of injury following the consumption of alcohol mixed with caffeine. The findings offer support for differential low-risk drinking guidelines by sex and the restriction and regulation of the sale and availability of caffeinated alcoholic beverages.

Introduction

Over the last decade there has been increasing concern over the risks associated with the consumption of caffeinated alcoholic beverages, especially given recent reports of fatalities linked to this type of use (Terry, 2014; Smith, 2018). The combined use of alcohol and caffeine (Alc+Caff) has been found to be associated with riskier drinking practices, higher alcohol consumption, and increased risk of injury and hospitalizations (Brache & Stockwell, 2011; Peacock, Bruno, Martin, 2012; Roemer et al., 2017; Thombs et al, 2010). In fact, the number of visits to U.S. emergency departments that involved prior use of energy drinks was estimated to have doubled between 2007 and 2011, with 13–16% of all admissions related to the combined use of alcohol and energy drinks (Center for Behavioral Health Statistics and Quality, 2013).

It is theorized that caffeine, the principal stimulant ingredient in energy drinks works to mask the sedative effects of alcohol, creating an “awake drunk state”. This state may result in the consumer underestimating their level of intoxication, which may lead to increased alcohol consumption and risk-taking, more hazardous drinking practices, and poorer risk assessment (Ferreira et al, 2006; Marczinski et al., 2006; Howard, 2011). Both increased alcohol consumption and the behavioral changes associated with the Alc+Caff use results in the consumer being at a higher risk of incurring an injury (Room et al, 2005; WHO, 2009). Despite the increasing concerns with this type of use, there is a lack of controlled studies examining the link between Alc+Caff use and injury. Additionally, the existing research has several weaknesses including: the majority of studies are cross-sectional using only case-control designs, and do not control for level of alcohol use or other potentially important variables such as sex or risk-taking propensity.

The study of Alc+Caff use is relatively new and as such, there is limited knowledge on the various factors that may influence the relationship between Alc+Caff use and the risk for injury, such as sex. Sex differences have been reported in the relationship between alcohol use and injury in some studies (McLeod et al., 1999; Stockwell et al., 2002) as well as in the consumption patterns of caffeinated alcoholic beverages (Friis, Lyng, Lasgaard, & Larsen, 2014). Women may be at a higher risk of injury following alcohol use, as they tend to reach a higher BAC compared to men following the consumption of equal amounts of alcohol (Mumenthaler, Taylor, O’Hara, & Yesavage, 1999). However, some research suggests that this sex difference may only exist at higher levels of alcohol consumption (McLeod et al., 1999; Stockwell, 2002), while other research still indicates sex differences may be a factor associated with the type of methodology used. Emergency department studies using case-crossover methods have reported no significant sex differences, whereas ED studies use case-control methods do report sex differences (Stockwell, 2002; Watt et al., 2004). While firm conclusions regarding sex differences in the risk-relationship between alcohol and injury maybe conflicting, determining whether these differences exist in the relationship between Alc+Caff use could have implications for policy, intervention and prevention practices. For example, sex differences in the risk of injury associated with Alc+Caff use could inform low-risk drinking guidelines. Currently, there is disagreement across different countries in terms of whether low-risk drinking guidelines should be different for men and women, which is largely based on the conflicting evidence reporting on sex differences in the risk relationship between alcohol use and alcohol-related harms (Dawson, 2009). Lastly, exploring the differences between case-control and case-crossover methodologies may help to provide more understanding of the potential mechanisms underlying the conflicting findings in the alcohol-related injury research.

The present study is, to our knowledge, the first controlled Emergency Department Study examining the risk relationship between Alc+Caff use and injury. In this paper we will build on a previous analyses of this study that indicated a synergistic effect of Alc+Caff use on the risk for injury (Roemer et al., under review) by exploring potential sex differences in this relationship. Given the strengths and limitations of different study designs, both case-crossover and case-control analyses will be used to compare findings across study designs. Based on previous research of alcohol use and injury risk, we hypothesize that women will show a higher risk of injury relative to men following the consumption of alcohol mixed with caffeine even after controlling for dose of alcohol and caffeine consumed.

Methods

The University of Victoria and Vancouver Island Health Authority review boards both approved the study proposal.

Study Design

Both case-control study and case-crossover study designs were used to investigate the relative risk (RR) of injury due to alcohol and/or caffeine use. The case control design allows for the examination of between person differences; however, using quasi-controls may not suffice as good controls because non-injured patients are more likely to be drinking heavily or abstaining compared to injured patients (Cherpitel, 1993). The case-crossover design has the benefit of a reduction in potentially confounding variables due to within-person factors (e.g. age, education, marital status). This latter analysis also uses a matched-pair approach; which allows for the adjustment of other potential sources of bias (Baker & Yardley, 2002). The case-control study involved testing whether or not there was a significant difference between the proportions of subjects who drank alcohol and/or caffeine drink within six hours of either presenting with an injury or an illness to the emergency departments. Logistic regression was used to calculate odds ratios for both men and women, as well as to test an interaction effect between Alc+Caff use and sex. The case-crossover analysis was conducted by testing for each injured individual whether or not they were significantly more likely to have drunk alcohol and / or caffeine either within six hours of their injury or exactly one week prior to their ED presentation. Conditional logistic regression was used to calculate hazard ratios for both men and women.

Sampling

Data were collected between 2013 and 2015 from representative samples of ED patients (aged 18 and older) at three hospitals located on the West Coast of British Columbia.

Procedure

In each ED, samples of patents aged 18 and over were drawn from computerized registration available on the ED computer, entered in consecutive order of patient arrival at the ED. An approximately equal number of injured and non-injured control subjects were interviewed on each shift. The sample was achieved at quieter times by seeking to interview every presenting injured attendee and every second or third ED attendee with an illness (Chan et al, 2010). At busier times the sampling ratio was adjusted; e.g., every second or third injured patient and every fourth or sixth non-injured patient.

Sampled patients were approached as soon as possible after registering for care with a request for informed consent to provide a breath sample and to be interviewed. Interviews, lasting about 25 minutes, were completed either in a private area in or near the waiting department or in the treatment department. In the case of those who were severely impaired, every attempt was made to interview the patient at a later time. Those patients who were too seriously ill or injured to be approached or interviewed in the ED were followed into the hospital and interviewed after they have been admitted and their condition stabilized. Patients were offered a $10.00 gift card for completing the interview. This methodology has been used in our prior ED studies in both the U.S. and Canada, and has proven acceptable to both patients and ED staff, and successful in obtaining high completion rates (Stockwell et al., 2002; Cherpitel et al., 2014).

Participants

There were a total of 2804 participants, with 1613 non-injured patients and 1191 injured. The total response rate was 69.16%. Participants ranged between the ages of 18–98 with a mean age of 44.96 (SD=20.08). There was an equal distribution of males (52.3%) and females (47.7%) and participants primarily identified as Caucasian (72.25%). The majority of participants were either married/common-law (39.1%) or single and never married (44.2%) and had completed some form of post-secondary education or training.

A small proportion of the sample (2.5%) reported using energy drinks in the 6-hour period. The small N for this group would likely result in difficulties related to power; therefore, it was decided that all caffeinated beverages be combined into a single group. Given that caffeine is the psychoactive substance in energy drinks, and the stimulant effects of the caffeine are the focus of the paper, combining all caffeinated beverages allowed us to examine the dose-response relationship with caffeine and alcohol use in combination with caffeine (henceforth Alc+Caff) using a larger number of participants.

Given the response rate of 69.16%, analysis was done to compare those with missing data to the rest of the sample. No significant differences were found except participants who refused or were unable to participate were slightly older (M=48.24, SD=21.57) than the rest of the sample (M=44.96, SD=20.08). For the missing data, the mean was used to replace missing values for continuous variables and missed values for categorical variables were classified into the most frequent group.

Measures

Patient interview and injury variables:

Patients were interviewed regarding the cause of injury (including violence) or medical problem which brought them to the ED; alcohol use; energy drink use; other caffeine use; and other substance use, variously within the six hours prior to the event, the same six-hour period the previous day and then previous week (for case-crossover and control-crossover analyses respectively); the amount of alcohol, energy drinks, or caffeine consumed; time elapsed between drinking, caffeine consumption, and other substance use and the event, and demographic characteristics (sex, age, marital status, income level, risk-taking propensity, and education level).

Additionally, data were obtained on the place where the patient was and the specific activity the patient was engaged in at the time of injury (or first awareness of the medical condition bringing the patient to the ED), as well as for the same time the day before and the week before the injury or medical event. Such measures have been utilized in previous studies (e.g. Stockwell et al, 2002; Cherpitel et al., 2014; Korcha et al., 2018) and in the BC preliminary studies to date. The place of injury was categorized as: the respondent’s home, other private residence, workplace (school/trade area/office), recreation or sporting areas, premises licensed for the sale of alcohol, an industrial area, a street, or “other”. Activity at time of injury (or medical problem) was classified as involving: passive activities (reference group), sports, household chores or domestic activities, travel, working to earn money, social activities and “other” activities.

Alcohol and Energy Drink use:

Following suit of previous ED studies (Bond et al, 2010; Cherpitel et al., 2014) self-reported alcohol use was measured by asking participants how many standard drinks they consumed in the 6-hour period prior to the injury/illness event, the same 6-hour period the day before and week before, and alcohol use in the previous 12 months. To facilitate in the accuracy of self-reported alcohol use individuals were provided with a description and visual aid of what constitutes a standard drink.

Since energy drinks contain variable amounts of caffeine, as well as other ingredients, data was obtained, using an open-ended question, on the exact beverage line, brand, and amount of each energy drink consumed, as well as coffee, tea, and caffeinated sodas, during the six hours prior to injury/illness event, and during the two control periods, to more accurately quantify caffeine consumption. Following the completion of the interview, the researcher calculated the total amount of alcohol (in gm) and caffeine (in mg) consumed using caffeine content from a food composition database (United States Department of Agriculture, 2011). At the stage of data analysis alcohol was converted into number of standard drinks (13.45 grams/standard drink) and the unit of change of caffeine was converted to 50mg of caffeine (approximately a small cup of coffee).

Statistical analysis

All statistical analyses conducted using SAS 9.3. We first stated the characteristics of the sample and t-test and Chi-square analysis to examine the differences in demographic and behavior measures of the sample by sex. We further conducted logistic regression analysis in the case-control study to estimate the odds ratio (OR) as estimate of relative risk (RR) of injury for alcohol and/or caffeine use within six hours of injury; the analysis adjusted for potential confounding effects of covariates including age, marital status, education attainment, location at time of injury/illness, substance sue within six hours. A hospital variable was also included into the model to determine whether hospital site confounded with the risk estimates. The coefficient changes were <15%, therefore the effect of the hospital can be ignored (Rothman et al, 2008).

Results

Descriptive statistics for all sociodemographic and substance use variables by sex are illustrated in Table 1. The alcohol use and caffeine use variables were both positively skewed. While the non-normality of variables would be problematic in most GLM analyses, logistic regression is a fairly robust analysis that avoids the issue of the violation of normality (Tabachnick & Fidell, 2013). For the alcohol mixed with caffeine variable a total of 490 participants reported consuming alcohol mixed with caffeine in the 6-hour period prior to their injury/medical event.

Table 1.

Sociodemographic, presenting condition and substance use characteristics of Emergency Department attendees included in study sample (n=2804).

	Men		Women
Age M(SD)	44.13 (19.32)		45.79 (20.77)
Ethnicity N
Caucasian	1058		967
Other	406		373
Education N
< Completed High school	216		116
Completed High school	273		232
College	529		519
Completed University or Higher	444		468
Marital Status N
Married/Common-law	559		538
Widowed/Separated/Divorced	175		285
Single/Never Married	728		512
Risk-taking score M(SD)	1.16(.76)		.84(.67)
Injury N	713		478
Illness/non-injury N	751		862
Location at time of injury/illness
Private Dwelling/Home	738		873
Licensed Premises	120		73
Public Space	220		123
Work	180		118
Transport	51		44
Other	115		104
Substance Use 6hr prior N
Stimulants	39		17
Depressants	48		51
Other Drug use	618		612
Alcohol Use (unit=standard drink)	N	M(SD)	N	M(SD)
Six-hour prior	281	7.12(6.51)	207	5.12(4.93)
One day prior	206	5.95(7.51)	151	3.48(4.00)
One week prior	238	6.46(7.67)	144	4.36(5.36)
Caffeine use (unit=mg)
Energy Drink	46	158.48(141.36)	22	119.77(45.76)
Soda	134	45.27(35.47)	91	50.29(49.15)
Coffee/tea	511	165.56(136.38)	467	128.82(87.28)
All Caffeine 6hr prior	633	154.75(138.87)	545	123.61(88.75)
Caffeine 1 day prior	615	171.66(178.74)	558	135.45(106.84)
Caffeine 1 week prior	631	164.81(158.06)	561	138.00(100.38)

The case-control analyses formally testing for a sex difference in injury risk levels following Alc+Caff use was significant. In the unadjusted model, alcohol use, caffeine use, and Alc+Caff use was independently associated with a higher risk of injury. Additionally, males were at an overall higher risk of injury. The overall adjusted model was significant, X² = 1805.12 p<.01, as was the interaction term of Alc+Caff by Sex, X² = 9.49 p<.05. In other words, the risk of injury associated with Alc+CAff use is different among males and females after controlling for the dose of alcohol and caffeine. In addition, the results showed that those who are younger in age, had lower levels of education, are higher in risk-taking propensity, were at a licensed venue/event, and had consumed stimulants or other drugs were at a higher risk of injury. In order to gain a better understanding of the sex difference, the case-control analysis was rerun for males and females separately. Table 2 illustrates the adjusted case-control analyses according to males and females. Alcohol mixed with caffeine was a significant predictor for both men and women at the 1% level. For men, the OR for risk of injury associated with Alc+Caff use was 1.69, whereas for women it was 3.10. In other words, the risk of injury associated with Alc+Caff use was much higher among women.

Table 2.

The odds ratio of injury (OR) and 95% confidence interval (CI) for alcohol and caffeine use within 6 hours of injury or illness in case-control analysis of patients with injury or other illness attending the emergency departments in British Columbia

	Unadjusted		Partially Adjusted †		Fully Adjusted †
Alcohol/caffeine use by sex	OR (95% CI)	95%CI	OR	95%CI	OR	95%CI
Men
Alcohol only	1.20**	1.16–1.23	1.18**	1.14–1.23	1.11**	1.06–1.17
Caffeine only	1.07**	1.05–1.10	1.08*	1.05–1.11	1.08**	1.05–1.11
Alcohol + caffeine					1.69**	1.30–2.20
Women
Alcohol only	1.36**	1.30–1.69	1.31**	1.10–1.72	1.16**	1.04–1.23
Caffeine only	0.95**	.78–.92	0.93**	.76–.98	0.98**	0.92–0.99
Alcohol + caffeine					3.10**	1.78–5.84

Note:

^†Adjusted for location and other substance use.

^*p<.05,

^**p<.01

In the case-crossover analysis models were run separately for males and females (Table 3). In the unadjusted model, alcohol use was an independent predictor of injury for both males and females. Caffeine use was a significant predictor of injury for males; however, it acted as a protective factor among females. The covariates of location at time of injury/illness and other drug use were included. Similar to the case-control analyses, being at a licensed venue/event and consuming stimulants or other drugs were associated with a higher risk of injury. Both adjusted models for males, X² = 151.02, p<.01, and females X² = 73.17, p<.01, were significant and the patterns of results were similar to the unadjusted and partially adjusted model. Alcohol use was a significant independent risk factor for injury among males and females and caffeine use remained a protective factor among females. Alcohol combined caffeine was significant for both males and females; however, the risk was again higher among females. For females the relative risk of injury was 3.21, whereas the relative risk for males was 1.38. In other words, both males and females are at a higher risk of injury when consuming alcohol mixed with higher doses of caffeine, but this effect is much stronger among females. Figures 1 and 2 illustrate the varying effects of Alc+Caff use relative to alcohol use on its own on the risk of injury for both men and women.

Table 3.

The hazard ratio (HR) and 95% confidence interval (CI) of injury for alcohol and caffeine use within 6 hours of injury and one week prior in case-crossover analysis of patients with injury attending the emergency departments in British Columbia

	Unadjusted		Partially Adjusted †		Fully Adjusted †
Alcohol/caffeine use by sex	HR (95% CI)	95%CI	HR	95%CI	HR	95%CI
Men
Alcohol only	1.16**	1.08–1.24	1.12**	1.06–1.19	1.06*	1.01–1.10
Caffeine only	1.14**	1.04–1.26	1.11*	1.02–1.22	1.00	0.93–1.08
Alcohol + caffeine					1.38*	1.02–1.86
Women
Alcohol only	1.46**	1.24–1.72	1.41**	1.12–1.78	1.32**	1.28–1.40
Caffeine only	0.83**	.75–.92	0.83**	.71–.97	1.00.**	1.00–1.01
Alcohol + caffeine					3.21**	1.69–6.12

Note:

^†Adjusted for location and other substance use.

^*p<.05,

^**p<.01

Figure 1.

Increase in risk of injury for men with alcohol consumption in 6 hours before ER admission, for alcohol only and alcohol plus caffeine drinkers

Figure 2.

Increase in risk of injury for women with alcohol consumption in 6 hours before ER admission, for alcohol only and alcohol plus caffeine drinkers

Discussion

With the increasing popularity in the use of caffeinated alcoholic beverages, there has been a growing concern in the risks associated with this use, including the potential increase for the risk of alcohol-related injuries (Terry, 2014; Smith, 2018). Despite this, limited in situ research exists examining the temporal relationship between Alc+Caff use and injury, and even less on other potentially significant factors that could be underlying or contributing to this risk-relationship. The present study examines sex differences for in situ, session specific Alc+Caff consumption and outcomes, allowing us to speak to the temporal relationship between Alc+Caff use and injury risk.

Our hypothesis was that women would show a higher risk of injury relative to men following the consumption of alcohol mixed with caffeine. The results support this hypothesis; when alcohol was combined with caffeine, the risk for injury was much greater among females relative to males, even after controlling for dose of alcohol and caffeine. These findings are not without precedent, as previous research on alcohol and the risk of injury has found a higher risk for females even after controlling for setting, activity, and other drug use variables (Stockwell et al., 2002; Watts et al, 2004). The results also indicated that while alcohol use on its own was associated with a higher risk of injury, this association was again stronger among woman. Furthermore, the sex differences in the risk relationship between alcohol use, Alc+Caff use, and injury were found in both the case-crossover and case-control analyses, which speaks to the robustness of this effect.

The higher risk of injury associated with alcohol and caffeine is particularly notable when considering the combined use of alcohol and energy drinks, as energy drinks tend to contain high amounts of caffeine. The combined use of alcohol and energy drinks is thought to lead to an increased risk of injury via increased alcohol consumption, increased time spent drinking, and engaging in riskier drinking practices (Ferreira et al, 2006; Marczinski et al., 2006; Howard, 2011; Room et al., 2005). Given that, even after controlling for body weight, women may become more impaired than men after drinking equivalent amounts of alcohol (Mumenthaler et al., 1999; Stockwell et al., 2002), it would follow that women would be at a higher risk of injury following the consumption of caffeinated alcoholic beverages. Currently, only some countries have differential low-risk guidelines in place and there is disagreement as to whether these differential guidelines should exist (Dawson, 2009). A higher injury risk propensity for women than men as a function of both alcohol and Alc+Caff use supports differential low-risk drinking guidelines for men and women and suggests these guidelines may need to expand to include risk statements on the combined use of alcohol and caffeine.

The present study helps to further delineate the relationship between Alc+Caff use and risk of injury; however, we do note a number of limitations. First, participants under the age of 18 were excluded and yet, the practice of Alc+Caff use is common among adolescence (Azagba, Langille, & Ashbridge, 2013). Further in situ, session specific research examining the risk outcomes of Alc+Caff use among youth is needed. Second, the 6-hour time period used to measure Alc+Caff use still allows for uncontrolled variability in the effects of consumption on the user. For example, individuals consuming these beverages at a faster rate or in a shorter time period may be more impaired than those consuming equal amounts but at a slower pace. Future research examining the different patterns of Alc+Caff use in the hours leading up to an injury could help to further delineate the casual and temporal nature of this relationship. Lastly, there were limitations in the analyses that could be run due to concerns with power from small group sizes for some variables of interest. The unfortunate outcome of this was an inability to compare outcomes for different types of injuries, such as violent versus non-violent. While men may be at a higher risk of experiencing a violent-related injury overall (Macdonald, Wells, Giesbrecht, & Cherpitel, 1999), women may be at a higher risk of a violent-related injury following the consumption of alcohol (Wells, Thompson, Cherpitel, Macdonald, Marais, & Borges, 2007); however, the research on this remains limited. Further research focusing on sex differences in the risk-relationship between Alc+Caff use and violent versus non-violent injuries is needed to help inform public health policy and effective prevention strategies. Nonetheless, the present study was able to respond to several gaps existing in the current literature on caffeinated alcoholic beverages. In addition to examining the temporal relationship between Alc+Caff use and injury, the present study controlled for a multitude of potential covariates, controlled for dose of alcohol and caffeine, and utilized two methodologies to compare outcomes.

In summary, the present study provides further support for a synergistic effect of Alc+Caff use on injury risk. Even more, the study highlights important sex differences both for alcohol and caffeine independently, as well as for the two substances combined. The results indicate the propensity for incurring an injury following Alc+Caff use is greater for women than men. This effect was found in both a case-crossover and case-control method, denoting the robustness of this effect. These results offer support for the already existing differential low-risk drinking guidelines for men and women. In addition, the results have implications for further policy and prevention strategies. Many countries are currently working to form suitable policy responses to alcoholic energy drink use, and the present study offers helpful insight into the potential risks associated with this type of use. More targeted prevention and intervention strategies for men and women may be more effective than a universal one; for example, including Alc+Caff use in differential low-risk guidelines for men and women. Lastly, informing the general public on the risks associated with Alc+Caff use could help to reduce the risk of alcohol-related injuries, as could regulating and reducing the availability of these types of beverages. For example, restricting the caffeine and sugar content in these beverages, as well as limiting the number of standard drinks to 1.5 per container could help to reduce the harm associated with this type of consumption. Additionally, placing clear labelling of sugar, caffeine, ethanol, and other stimulant content on containers as well as warning labels of various health risks could help to better inform consumers.