Differences Between Preemptive and Non-Preemptive Physical Activity Among ‘Drunkorexia’-Positive College Students

Abstract

Objective:

‘Drunkorexia’ is characterized by compensating for alcohol-related calories using physical activity (PA). Drunkorexia is common on college campuses but little is known about the PA behaviors within the drunkorexia paradigm.

Methods:

First-year college students living on campus completed an online survey collecting drunkorexia, PA, and alcohol consumption data. A total of 127 participants reported engaging in drunkorexia behaviors.

Results:

Fifty-three participants were classified as preemptively physically active (e.g., PA and drink on Tuesday) compared to 74 as non-preemptively physically active. Preemptively physically active participants consumed more alcohol on Fridays and Saturdays than those non-preemptively physically active. Preemptively physically active participants engaged in significantly greater amounts PA. Females accounted for all significant differences between groups.

Discussion:

Among drunkorexia-positive participants, many made preemptive efforts to control their calories before consuming alcohol, which may predispose them to higher incidences of adverse outcomes such as alcohol poisoning, unwanted sexual advances, and death.

Introduction

The majority of alcohol consumption occurs while binge drinking^{1, 2} and this is a major contributor to college student morbidity and mortality.^2–4 Most first-year college students, 60–70%, do not report habitual alcohol consumption, however 50% report binge drinking when consuming alcohol.^{1, 5} Drinking behaviors pre-matriculation strongly predict first semester college drinking behaviors.^{5, 6} Due to continued missing national binge drinking goals,^{7, 8} a better understanding of students’ pre-matriculation drinking behaviors is warranted.

‘Drunkorexia’ is a colloquial term describing the co-occurrence of alcohol misuse and inappropriate compensatory behaviors including excessive physical activity (PA) and purging behaviors to avoid weight gain and intensify the effects of alcohol.^{2, 4, 9, 10} Drunkorexia behaviors have a higher prevalence among female students possibly from the pressures to control their weight while concomitantly socially consume alcohol.^{3, 11, 12}

PA benefits physical and mental well-being.¹³ Healthy behaviors, like PA and abstaining from alcohol, tend to cluster;^{14, 15} however, due to the paradoxical nature of drunkorexia, evidence is mixed in the exact association between PA and binge drinking.^{9, 10, 16} Differences in preemptive (preceding alcohol consumption) and non-preemptive (following alcohol consumption) PA among drunkorexia-positive college students have not been investigated. Alcohol consumption before PA has been associated with decreased endurance performance while PA after alcohol consumption may be impaired by a dehydrated state, impaired glycogen repletion, or endocrine imbalances.¹⁷

It is important to examine how PA differs within a drunkorexia paradigm to better understand the paradoxical relation of drunkorexia and better tailor strategies to help those who are drunkorexia-positive. This study aimed to identify differences in PA by intensity between preemptive and non-preemptive physically active participants among a group of drunkorexia-positive first-year college students.

Material and methods

Participants

Following Institutional Review Board approval, students (N=2,699) were contacted for participation from a comprehensive email list encompassing all first-year college students living in residence halls at a public Midwestern university. Pre-matriculation first-year students were selected to examine alcohol consumption behaviors before the influence of college. Study participation was not compensated.

Protocol

An initial email consisting of study description, inclusion criteria, survey link, and investigators’ contact information was distributed one week prior to fall semester commencement. Three reminder emails were distributed over the next six days. Participants provided informed consent before taking the survey. Of the 497 surveys completed (18% response rate), 127 participants met inclusion criteria (≥1 on the Drunkorexia Motives and Behaviors scales and recent alcohol use).

Alcohol

A standard drink was defined for the participants (e.g., one 12–ounce beer, one 1.5–ounce shot of distilled spirits).¹⁸ Participants indicated any alcohol consumption within the past four weeks (yes/no) and answered questions regarding binge drinking, daily alcohol use, and frequency of alcohol-related behaviors in line with the National Institute on Alcohol Abuse and Alcoholism’s recommended quantity and frequency questions.^{19, 20} Binge drinking frequency (How many days did you have 5 or more drinks [male] or 4 or more drinks [female] in a two-hour period?) was measured on a seven-point Likert scale: None, 1 or 2 days, 3–5 days, 6–9 days, 10–19 days, 20–29 days, and all 30 days. Daily alcohol use (How many drinks did you consume on each day of the week?) was measured using a six-point Likert scale: None, one drink, 2 drinks, 3 drinks, 4 drinks, 5 or more drinks (Cronbach’s alpha was 0.73).

Drunkorexia

Drunkorexia was assessed using the Drunkorexia Motives and Behaviors scales measure, Drunkorexia Motives and Behaviors Scales (23-items), Drunkorexia Fails Scale (10-items), Drunkorexia During an Alcohol Consumption Event Scale (10-items), and Post-Drinking Compensation Scale (9-items).²¹ Responses for all scales are measured on a 5-point Likert scale: Never, Almost Never, Sometimes, Almost Always, and Always. The Drunkorexia Motives and Behaviors scales had an overall Cronbach’s alpha of 0.95 (Drunkorexia Motives and Behaviors Scale=0.95, Drunkorexia During an Alcohol Consumption Event Scale=0.86, Post-Drinking Compensation Scale=0.91, and Drunkorexia Fails Scale=0.70). Per instrument protocol, all subscales were aggregated (0–260), where higher scores indicate more severe drunkorexia behaviors.²¹ Students were considered drunkorexia-positive if they scored at least one on the Drunkorexia Motives and Behaviors scales.

Physical Activity

The International Physical Activity Questionnaire–Short Form (IPAQ-S) is a seven-item self-reported recall of PA of the previous week.²² Participants report the frequency and duration spent sitting, walking, moderate PA (MPA), and vigorous PA (VPA). The IPAQ-S has strong test-retest reliability (ρ=0.75) and moderate criterion validity (ρ=0.30).²² We asked participants to indicate which days they engaged in nothing, walking, MPA, and VPA using IPAQ-S definitions. Cronbach’s alpha for all measures of PA was 0.88.

Preemptive PA was defined as a day where a participant engaged in both PA and consumed alcohol (e.g., PA and drink Friday). Days where participants consumed alcohol but did not engage in PA (e.g., drink Saturday and PA Sunday) were classified as non-preemptive physically active. Participants were classified as either preemptively or non-preemptively physically active from the greater frequency of preemptive or non-preemptive days throughout a week.

Statistical Analysis

To test whether PA intensity (sedentary, MPA, and VPA) differed between drunkorexia-positive students by PA timing groups (preemptive versus non-preemptive PA), a one-way ANOVA was implemented stratified by sex. All analyses were conducted using SPSS (Version 25, SPSS, Inc., Chicago, IL) with an alpha level set a priori at 0.05.

Results

Among students who completed the survey, 29.58% were considered drunkorexia-positive (n=127). The current sample was predominately Non-Hispanic or Latino (97.6%), White (94.5%) females (68.5%). A total of 53 participants were classified as preemptively physically active and 74 as non-preemptively physically active (Table 1). Both timing groups were predominately female (n=87) with mean body mass index categorized as normal (23.76±5.07 kg·m⁻²).²³ A majority of preemptively physically active participants reported attempting to maintain current weight, while a majority of non-preemptively physically active participants reported seeking to lose weight. There were no differences between timing groups concerning BMI, drunkorexia scores, or binge drinking days. Most reported alcohol consumed was relegated to Thursday, Friday, and Saturday. Significant differences were observed for reported Friday alcohol consumption prevalence between timing groups (77.4% preemptive and 52.7% non-preemptive) as well as on Saturdays (73.6% preemptive and 60.8% non-preemptive; p<0.01). On Fridays, 14.2% of all participants reported engaging in binge drinking while the occurrence increased to 15.0% on Saturdays. Significant differences were observed for Friday binge drinking between timing groups (30.2% preemptive and 2.7% non-preemptive; p<0.001).

Table 1.

Participant characteristics – total sample and by drunkorexia timing

	Total Sample (N=127)	Preemptively Physically Active (n=53)	Non-preemptively Physically Active (n=74)
Age (years)	18.23±1.19	18.13±0.39	18.31±1.50
Sex
Female	68.5%	60.4%	74.3%
Male	31.5%	39.6%	25.7%
Ethnicity
Hispanic or Latino	2.4%	0%	4.1%
Non-Hispanic or Latino	97.6%	100%	95.9%
Race
American Indian or Alaskan Native	0.8%	0%	1.4%
Asian	0.8%	1.9%	0%
Black or African American	2.4%	0%	4.1%
Native Hawaiian or Other Pacific Islander	0%	0%	0%
White	94.5%	98.1%	91.9%
Mixed	0.8%	0%	1.4%
Body Mass Index (kg·m−2)	23.76±5.07	23.81±5.38	23.73±4.87
Binge Drinking Days	1.83±1.01	2.04±1.12	1.69±0.91
Drunkorexia Score	71.38±22.94	74.70±23.13	69.00±22.67
Weight Status
Lose	47.2%	43.4%	50.0%
Maintain	44.9%	50.9%	40.5%
Gain	7.9%	5.7%	9.5%
Total Physical Activity (minutes·day−1)	300.09±262.06	365.77±277.96	253.05±241.09
Physical Activity Intensity (minutes·day−1)
Vigorous	58.77±72.06	79.53±76.64*	43.91±65.12
Moderate	99.08±103.92	129.92±116.96*	76.99±87.80
Sedentary	345.55±258.35	321.98±229.92	362.43±277.23

Continuous variables are presented as mean±standard deviation. Discrete variables are presented as proportions.

Overall, participants reported VPA 2.84±1.86 days per week averaging 58.77±72.06 minutes per day. Significant differences were observed regarding both reported days per week and minutes per day of VPA between timing groups (preemptive 3.36±1.70 days, 79.52±76.64 minutes and non-preemptive 2.47±1.89 days, 43.91±65.12 minutes; p<0.01). Participants reported 4.11±1.83 days per week averaging 99.08±103.92 minutes per day of MPA with significant differences between timing groups (preemptive 4.85±1.66 days, 129.92±116.96 minutes and non-preemptive 3.58±1.77 days, 76.99±87.80 minutes; p<0.01). Participants reported a mean of 345.55±258.35 minutes of sedentary time with preemptively physically active participants reporting 321.98±229.92 minutes per day and non-preemptively physically active reporting a mean of 362.43±277.23 minutes per day with no differences between groups.

No differences were observed among male timing groups regarding any PA intensity. Female preemptively physically active participants reported a mean of 60.94±53.07 VPA minutes per day compared with 32.00±47.13 minutes per day for non-preemptively physically active participants (Table 2.; p=0.01). Significant differences were observed with preemptively and non-preemptively physically active for MPA (118.44±100.61 and 67.67±83.1 minutes, respectively; p=0.013).

Table 2.

Female daily minutes of physical activity intensity level by temporal classification

Physical Activity Intensity	All Females (n=87)	Preemptively Physically Active (n=32)	Non-preemptively Physically Active (n=55)
Vigorous	42.64±51.06	60.94±53.07*	32.00±47.13
Moderate	86.34±92.69	118.44±100.61*	67.67±83.1
Sedentary	360.8±262.94	335.94±194.24	375.27±296.43
All Physical Activity	282.9±257.64	338.75±247.71	250.4±259.94

Data presented as mean±standard deviation

^*p<0.05

Discussion

The present study sought to further discern the differences in PA intensity between preemptive and non-preemptive PA relative to alcohol consumption among drunkorexia-positive first-year college students. To our knowledge, this is the first study to quantify PA at different intensities in drunkorexia-positive students.

Preemptively physically active participants reported engaging in higher levels of both MPA and VPA. Dinger et al¹⁵ found college students who met moderate-to-vigorous PA (MVPA) recommendations to also partake in elevated levels of binge drinking in the previous two weeks. The lower levels of PA observed among non-preemptively physically active participants could be influenced by the adverse short-term effects of alcohol creating barriers to PA. Lethargy, weakness, and lack of coordination are short-term symptoms experienced after alcohol consumption, creating an unfavorable environment for PA.²⁴ While participants may utilize PA to ameliorate feelings of guilt,⁴ the cognitive and physiological short-term impairments from alcohol may stifle the desire for PA.²⁵

Female participants accounted for all significant differences between timing group PA intensities. Female students may be more frequently or intensely afflicted by guilt to compensate for alcohol-related calories. Researchers have observed drunkorexia-positive females are more likely to engage in PA to compensate for increased caloric intake.² Females engaging in drunkorexia behaviors are potentially motivated by the enhancement effects from alcohol, weight control, body dissatisfaction, and societal pressures.^{3, 12, 26–29} The planned behavior of burning calories before drinking may lead to females consuming more alcohol.²⁶ When examining PA motivations of female college students, drunkorexia-positive students tend to exercise more frequently due to weight concerns and physical appearance.^{26, 29, 30} The present results are not surprising as researchers have previously indicated females with eating disorders incur higher instances of intoxication and score higher on alcohol-related problems scales.²⁶ Leasure et al⁴ identified four total motivational themes, ‘Work hard, play hard’ and ‘celebratory motivations’ lead to PA being used preemptively, whereas ‘guilt’ and ‘body image’ motivations result in PA being non-preemptive. Laghi et al³¹ suggest the inability to properly regulate emotions may play a pivotal role in drunkorexia behaviors in this age group.

This study is limited by the ability of participants to recall and report PA and alcohol behaviors. All participants were underage and regardless of anonymity may be unwilling to divulge drinking behaviors. Recalled behaviors were used to categorize participants as preemptively or non-preemptively physically active. Although drinking is typically regulated to later times of the day, participants may have preemptively engaged in PA the day before drinking (i.e., PA on Thursday and drinking of Friday) to compensate for alcohol calories. The use of a more robust methodology for assessing the temporal sequence of behaviors, such as ecological momentary assessment or direct observation, would provide near real-time assessments of behaviors.^{32, 33} The current sample was predominately White Non-Hispanic female first-year college students, limiting the generalizability of the findings to other populations. We dichotomized individuals as engaging in drunkorexia behaviors or not using the Drunkorexia Behaviors and Motives scales, which ranges from 0–260, removing the ability for nuanced assessment of the gradient of drunkorexia behaviors. However, there are no currently established cut points for at what score an individual is considered drunkorexia positive or to the severity of the drunkorexia behaviors. The current study contains several strengths. First-year college students tend to increase alcohol consumption throughout their first year in college,⁵ therefore the pre-matriculation prevalence of drunkorexia behaviors is an important health concern. The use of the Drunkorexia Behaviors and Motives scales,²¹ is a more comprehensive measure used to assess drunkorexia compared with other research utilizing a single item.³

Conclusions

This study highlights the timing and quantity of PA behaviors within the drunkorexia paradigm. There is a paucity of drunkorexia literature focused on PA. PA has been found at pathological levels among participants experiencing disordered eating.^{34, 35} As PA is an integral component of drunkorexia, the timing of PA may provide a better indication of why the participants are engaging in drunkorexia behaviors. Examination of pre-matriculation drunkorexia behaviors is essential to better inform and implement alcohol misuse, disordered eating, or mental illness treatment or prevention efforts among campus health services and student affairs health initiatives. Identifying and targeting strategies to address all components of drunkorexia in tandem and refine the strategies based upon the temporal relation among components, may provide a greater impact on first-year college students’ health during their college career.