Caffeine consumption, intoxication, and stress among female university students: a cross-sectional study

University students use caffeine to cope with stress in spite of its adverse effects. The purpose of this study is to explore caffeine consumption among university students in Saudi Arabia, as well as its correlation with stress and caffeine intoxication. This cross-sectional study examined a convenience sample of 547 students at Princess Nourah Bint Abdulrahman University (PNU). A self-administrated questionnaire was used to assess caffeine consumption in milligrams per day, stress was assessed by the perceived stress scale (PSS), and caffeine intoxication was assessed using the DSM-5 criteria. The mean total caffeine consumption was 424.69 ± 385.31 mg/day. High levels of caffeine consumption were found among students of non-health colleges and students who were undiagnosed with psychiatric disorders (p values <0.040 and 0.027, respectively). A significant positive correlation was found between caffeine consumption and perceived stress (p<0.045). Only 13.26% of all participants fulfilled the DSM-5 criteria for caffeine use disorder. The majority of participants showed moderate and high stress levels (69.9% and 18.7%). This study revealed high caffeine consumption and perceived stress levels among female undergraduate students with a significant positive association between them. The results emphasize the importance of educational campaigns about caffeine consumption and intoxication. They also encourage the development of stress management programs. Longitudinal studies need to be designed for evidence-based intervention.


Background
Caffeine is a stimulant of the central nervous system and metabolism that is used for recreational and for medical reasons, such as decreasing physical exhaustion and increasing mental alertness [1]. Caffeine intake has positive and negative effects. The positive effects are enhanced mood and readiness, improved ability to stay conscious and alert, and strengthened exercise performance [2]. On the other hand, negative effects may occur when caffeine intake exceeds 250 mg, it can result in a condition called caffeine intoxication. Symptoms include fidgeting, excitement, insomnia, increased urination, gastrointestinal disturbance, muscle twitching, irregular or rapid heartbeat, and psychomotor agitation according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) [3].
Students experience stressful times in college due to classes, homework, exams, projects, and extracurricular activities [4]. Studies done on college students in Puerto Rico, Saudi Arabia, and Turkey showed that 49%, 49.5%, and 58.99% of them use caffeine as a coping mechanism, respectively [2,5,6]. In order to deal with this stress and fulfill academic requirements, college students may consume caffeine in the belief that it can aid their academic performance [4].
In Bahrain, the mean of daily caffeine consumption was assessed among college students. Females were consuming less than males (246 and 306 mg/day respectively) [7]. Another study assessed caffeine consumption and sleep habits among a sample of 228 students at Princess Nourah Bint Abdulrahman University (PNU) and found that most of them had high caffeine consumption, and a need for future studies concerning caffeine intoxication was suggested [8]. In this study, we aimed to estimate the level of caffeine consumption among students at health colleges and non-health colleges in PNU and to explore the correlation of caffeine consumption (including all types of caffeinated beverages) with caffeine intoxication and perceived stress.

Study design and population
This cross-sectional study examined a convenience sample of students who were Arabic speakers at the age of 18 years or above at health and non-health colleges at PNU, Riyadh, Saudi Arabia. PNU is the first women's university in the Middle East and also the largest one. It accommodates 33,825 students and includes 18 colleges.
A self-administrated questionnaire was distributed conveniently in October 2019 to 547 students from different colleges. Equal numbers of questionnaires were distributed to health and non-health colleges (humanities, community, and science). The comparison was chosen to be between health and non-health colleges based on the significant differences in the studying years, academic system, training requirements, and healthrelated knowledge. Health college students have an average of 6 years of studying and non-health college students have an average of 4 years. These factors may affect the level of their perceived stress and caffeine consumption.
The consenting process started by going to the students on the campus and explaining to them the study and asks them if they are willing to participate, mentioning that their identity is confidential as it does not require a name or an ID. If they agree to participate, a copy of the questionnaire will be given to them, and they will be encouraged to read the front page thoroughly which has a statement saying "Filling this questionnaire means you agree to be part of this study." It also has the information and purpose of the study and the contact information.
Previous literatures showed that the average daily caffeine consumptions vary from 10 to 11.2 mg/day, and the standard deviation was around 5 mg/day. Gpower software was used to calculate the minimal size required for this study, considering alpha =0.05 (level of significance of 95%) and beta to be 0.20 (power of the study is 80%), the minimal sample size required is 548.

Data collection tool
The questionnaire included four main divisions. The first one covered demographic characteristics, academicrelated characteristics, and personal clinical history. The second and third divisions covered caffeine consumption and caffeine intoxication. The fourth division assessed perceived stress level by PSS-10. The first three division of the survey were first written in Arabic then translated to English by a professional translator, then reviewed and translated into Arabic by bilingual speakers. Pilot test was done for the Arabic version, and then face validation was done by sending the survey to four mental health professionals. Furthermore, the fourth division involved the Arabic version of PSS-10, which has been validated previously by Chaaya et al. The PSS-10 was translated from English into Arabic, and then it was reviewed by a bilingual psychiatrist. After that, it was back translated into English by the psychiatrist and a comparison with the original one was done [9].
Demographic characteristics, academic-related characteristics, and personal clinical history Demographic characteristics included age, nationality, marital status, number of children, and income, which was estimated among students as having enough income, enough income with saving, not enough income, or debt. Academic-related characteristics included college, academic level (first and second-years were counted as "juniors," whereas third, fourth, and fifth years were accounted as "seniors"), GPA, and academic satisfaction. Personal clinical history included questions about their history of diagnosed chronic diseases or psychiatric disorders, received psychiatric help, and smoking.

Caffeine consumption
Caffeine intake per day was measured in milligrams. A table was included in the questionnaire ( Table 1). The common caffeinated drinks had been enlisted in the table. All common caffeinated drinks were investigated, including drinks such as coffee, decaffeinated coffee, tea, cola, citrus, and energy drinks. The table included the size and the number of cup/cans per day for each drink. A reference image was attached to illustrate the size, and the amount of fluid in ounces. The participants were asked to fill out only the size and number of the drinks they regularly consume. The amount of caffeine in each drink size was calculated (Table 2), which were later multiplied by the number of cups/cans consumed daily. Then, the total numbers of caffeine milligrams per day were summed. Caffeine intake was examined as "low" and "high." Low intake was considered less than 250 mg per day, while high intake was considered more than 250 mg per day. The validity and reliability of measurements have been confirmed in similar studies [10,11].

Caffeine intoxication
Intoxication was assessed using the criteria of DSM-5, which includes 12 symptoms. Participants were asked if they developed symptoms during or shortly after caffeine consumption. Any participant with five or more symptoms was diagnosed with caffeine intoxication according to the criteria.

Perceived stress level
Stress level was measured using the Arabic version of the PSS, which is a 10-question tool that is used to measure perception of stress over the past 30 days. The scale was developed in 1983 [12] and was modified in 1988 by Cohen [13]. It is a validated stress questionnaire with established acceptable psychometric properties [14,15]. A Likert-type scale was used to capture responses to the PSS ("never," "almost never," "sometimes," "fairly often," and "very often"). A score of 0-13 is considered as low stress, 14-26 is considered moderate, and 27-40 is considered high perceived stress.

Statistical analysis
Data were analyzed using SPSS 23. We described the variables as means ± the standard deviation (SD) or percentages as appropriate. A t test was used to determine the difference between quantitative variables, while the chi-squared test was used to determine the association between qualitative variables.

Demographic characteristics, academic-related characteristics, and personal clinical history
The total number of participants in the study was 547, and the average age was 20.30 ± 1.91 years. The majority were Saudi (98.40%) and single (96%). More than half of the participants (59.20%) reported that their income was enough, and 29.20% reported that their income was enough with saving. Almost half of the participants were from health colleges (50.10%), whereas the other half were from nonhealth colleges (49.90%). More than half of the participants were junior students (61.80%). The GPA 4.50-5.00 for 42.50% of the participants, and only 9.40% of them had a GPA less than 3.50. More than half of them were either satisfied or very satisfied with their academic achievement (46.60% and 22.80%, respectively).
A minority of the participants reported that they were diagnosed with chronic diseases and psychiatric disorders (6.80% and 10.10%, respectively). Half of those who had been diagnosed with a psychiatric disorder received psychiatric help (56.36%). In addition, the majority of participants were non-smokers (93.90; Table 3).

Caffeine consumption
The mean total caffeine consumption per day was 424.69±385.31 mg. Specialty coffee was the most consumed caffeine source with a mean of 93.06±126.99 mg, followed by regular brewed coffee, capsule coffee, and black tea with means of 62.74±114.30 mg, 55.39±114.62 mg, and 51.60±83.98 mg, respectively ( Table 4). The mean of low caffeine consumption group (< 250 mg/ day) was 126.6±68.01 mg/day, while it was 628.28±381.6 mg/day for the high caffeine consumption group (> 250 mg/day).
A high level of caffeine consumption was significantly more evident among students of non-health colleges than health college students (53.50% versus 46.50%, respectively; p<0.040). In addition, a high level of caffeine consumption was significantly more evident among students undiagnosed with psychiatric disorders than diagnosed students (87.60% versus 12.40%, respectively; p<0.027). Moreover, a high level of caffeine consumption was significantly more evident among students who experienced caffeine intoxication symptoms than asymptomatic students (75% versus 57.1%; p< 0.005). Finally, students who had a high level of caffeine consumption had significantly higher mean scores of perceived stress than students with a low level of caffeine consumption (21.40±6.38 and 20.27±6.31, respectively; p<0.045; Table 5).

Perceived stress
More than two-thirds of the participants showed moderate stress levels (69.9%), whereas 18.7% reported high stress levels (Table 6). There were significant associations between the level of perceived stress and academic satisfaction. A high level of stress was also more evident In addition, the level of perceived stress was significantly associated with students' income; a higher level of stress was more evident among students who expressed that they were in debt or their income is not enough those who had enough income or enough income with savings (24.11±6.95 and 23.24±7.90 versus 20.99±5.75 and 20.16±6.69, respectively; p<0.008). Another significant association was found between the level of perceived stress and some personal clinical histories. A high level of stress was more evident among students who were diagnosed with psychiatric disorders than undiagnosed students (25.20±6.31 versus 20.52±6.20, respectively; p<0.001). The level of stress was also significantly higher among students diagnosed with chronic disease than undiagnosed students (23.69±6.17 versus 20.80± 6.33, respectively; p<0.008). Furthermore, a high level of stress was more evident among smokers than nonsmokers (24.36±7.44 versus 20.79±0.02, respectively; p< 0.002). Finally, a high level of stress was more evident among students who experienced caffeine intoxication symptoms than asymptomatic students (24.12±5.92 than 20.51±6.28, respectively; p<0.001; Table 7).

Discussion
This study represents the first Saudi university-based survey of caffeine consumption including all types of    [10,11,[19][20][21][22]. This could be due to the higher level of perceived stress that was found among university students in this study as caffeine may relieve stress [23].
In addition, other numerous factors for caffeine intake among undergraduate university students were reported in the USA including improving alertness, concentration, mood, energy, and enjoying the taste [24]. Furthermore, the level of caffeine consumption was significantly lower among students diagnosed with psychiatric disorders, which could be attributed to their awareness or previous experience of the effects of excessive caffeine consumption, which increases the risk of anxiety, panic attacks, and psychotic symptoms [25,26].  And those who are suffering from anxiety conditions may have more caffeine sensitivity, which contribute in caffeine avoidance due to the undesirable effects [27][28][29]. It could also be attributed to their awareness or previous experience with the potential interaction of caffeine with psychotropic drugs that are used for their psychiatric conditions, which is due to the metabolism of caffeine by CYP1A2 enzyme. Caffeine can inhibit this enzyme and cause side effects that may affect their treatment plan [30]. In addition, a high level of caffeine consumption was significantly more evident among students who experienced caffeine intoxication because the more caffeine they consume, the more symptoms they experience. A related study done in the USA showed that excessive caffeine consumption can lead to caffeine intoxication [4]. And it was found that only 13% of participants experienced caffeine intoxication according to the DSM-5 criteria. This is similar to the prevalence of intoxication that was found among psychiatric patients in Italy (10.3%), which was significantly higher compared to healthy participants (2.9%). However, comparing our results with the Italian results was limited by the samples differences as the Italian study had wider age range and more severe psychiatric cases compared to our study [22]. Perceived stress was prevalent in this study. This is not surprising as similar results were found in previous studies that were conducted among university students in Saudi Arabia, Iran, and Malaysia [5,[31][32][33]. In addition, a significant positive relationship was found between the level of caffeine consumption and the level of perceived stress. This is supported by a previous study that found a significant positive relationship between the consumption of energy drinks and stress [34,35]. This might be due to the beneficial effects of caffeine in maintaining cognitive function under conditions of stress and improving work performance [23].
Moreover, smoker students reported significantly higher stress levels. There are several theories on the   [36]. Perceived stress was found to be significantly higher among students who were diagnosed with psychiatric disorder or chronic disease. This is not surprising as it is evident that stress is a risk factor for various psychiatric and medical conditions [37][38][39][40]. Research shows that almost every system in the body can be influenced by chronic stress. When chronic stress goes unreleased, it suppresses the body's immune system and ultimately manifests as illness. If stress continues and the body is unable to cope, there is likely to be a breakdown of bodily resources [41].

Limitations
This is the first Saudi university-based survey of caffeine consumption among students from both health and non-health colleges that included all types of caffeinated beverages. The results provided valuable information about caffeine consumption, caffeine intoxication, and stress. However, the convenience sampling and female participants limit the generalizability of the study. Although all common caffeinated drinks were investigated in this study, other possible sources of caffeine such as caffeine pills and chocolate were not included. In addition, even if intoxication symptoms listed in the survey were developed during or shortly after caffeine intake, it was difficult to differentiate between caffeine intoxication and symptoms of other medical or psychiatric conditions. Furthermore, a cross-sectional study cannot identify causality relationships.

Conclusion
Caffeine is highly consumed by female undergraduate students, mostly specially coffee, and the level is significantly higher among students of non-health colleges. In addition, caffeine consumption levels are positively and significantly correlated with perceived stress levels, which were prevalent among the students. However, only 13.26% of all participants fulfilled DSM-5 criteria for caffeine use disorder which was associated with high level of stress. This emphasizes the importance of educational campaigns about caffeine consumption and intoxication. Furthermore, this study could be useful for future university education and stress management planning. It could also be used as a primary resource for future investigations. However, longitudinal studies need to be designed for evidence-based intervention. Further studies also need to involve both sexes and postgraduate students.