Prevalence of Alcohol Use and Associated Factors among Dilla University Students, Dilla Town, Southern Ethiopia: A Cross-Sectional Study

Introduction Globally, alcohol is one of the most prevalent forms of substance use that is particularly high among young age groups. Despite the adverse health and social challenges associated with alcohol use, it is one of the most common risky behaviours among university students. Objective This study aimed to assess the prevalence of alcohol use and associated factors among Dilla University students in Southern Ethiopia. Methods An institution-based cross-sectional study was conducted at Dilla University among undergraduate regular students from January to February 2018. A systematic random sampling technique was used to get a total sample of 803 students each year from the department of the university. The collected data were coded, entered into Epi Info version 7.1, and analyzed with SPSS version 20. Results A total of 803 participants were successfully interviewed with a response rate of 91.7%. Alcohol use prevalence was 41.8% (n = 336) among participants. Being in fourth year (AOR = 2.66, 95% CI: 1.64, 4.31), having friends who use the substance (AOR = 1.53, 95% CI: 1.09, 2.1), being a khat user (AOR = 1.48, 95% CI: 1.05, 2.09), and being a cigarette smoker (AOR = 1.76, 95% CI: 1.09, 2.84) were found to be significantly associated with alcohol use. Conclusion The current study revealed that fourth-year students, having friends who use the substance, being khat user, and being cigarette smoker had higher odds of alcohol use among the students. So the findings suggest that effective campus-based counselling and peer education should be implemented for early prevention, detection, and alleviation of alcohol use among students in the university.


Introduction
Substance use among adolescents can lead to a variety of detrimental consequences. Cigarette smoking, alcohol drinking, and cannabis use can increase accidental or intentional injuries, commission of crimes, mood disorders, and mortality and can complicate normal psychosocial development [1][2][3]. e use of substances is one of the rising major public health and socioeconomic problems worldwide, which has dramatically increased particularly in developing countries [4][5][6][7][8]. Substance use in Africa is increasing rapidly, especially for tobacco, alcohol, and khat [9].
Alcohol is one psychoactive substance that has been widely used in many cultures for centuries [10,11]. It is often consumed before, along with, or after other psychoactive substances, and the comorbidity of alcohol and tobacco dependence is strong [12]. Alcohol use causes large disease and social and economic burden in many societies [13].
Harmful use of alcohol accounts for nearly 6.5% of all deaths in Europe, where alcohol-related harm is highest, and even in the world, but the burden is much higher in certain countries and for certain groups within countries [14]. About 3.3 million deaths, or 5.9% of all global deaths, were attributable to alcohol use [11,13].
ere are significant gender differences in the proportion of global deaths attributable to alcohol, for example, 7.6% of deaths among males and 4.0% of deaths among females were attributable to alcohol [13].
Younger people are more affected by alcohol than older people. Possibly 13.5% of all deaths among those who are 20-39 years of age may be attributed to alcohol [12]. An estimated 24.6 million Americans aged 12 or older had used an illegal drug, approximately 9.4% of the population [15]. In most parts of Africa, health and social problems related to alcohol use are increasing. In sub-Saharan Africa, an estimated 1.8% of the disease burden is attributable to alcohol use [16,17].
In a study done in western Kenya among college students, the lifetime prevalence rate of alcohol use was 51.9% [18]. Being male, ease of getting an adult to buy alcohol, cigarette use, and having multiple sexual partners were risk factors [11].
About 5.9% of all global deaths were attributable to alcohol consumption. Global deaths were attributable to alcohol among males (7.6%) and females (4.0%) [14].
A study done in western Kenya among college students showed that 75.1% were introduced to substance use by a friend while 23.5% were introduced by a relative. Majority of users wanted to relax (62.2%) or relieve stress (60.8%) [18].
Over 20% of out-of-school youth had unprotected sex during the 12 months before the interview compared to 1.4% of in-school youth. Daily alcohol intake and khat intake were also associated with unprotected sex [19].
One Ethiopian study of Haramaya University students found the prevalence of substance use was high, and about two-thirds (62.4%) of the participants used at least one substance, most commonly alcohol (50.2%) [20].
Another study at Axum University found that the lifetime prevalence of alcohol drinking was 34.5% and the current prevalence was 32.8% [21]. Similarly, in Debre Markos University, the prevalence of alcohol drinking was 33.8% [22].
Another study done in Hawassa found that 40.8% used alcohol, 20.3% chewed khat, 11.9% smoked cigarettes, and 0.9% used marijuana [23]. In another study among RVUC students, the lifetime prevalence of alcohol drinking was 40.2%. Similarly, the current prevalence of alcohol drinking was 35.6% [24]. e Hawassa study found that having family members who used psychoactive substances (PAS), peer influence, being male, and living alone during school age were positively associated with overall PAS use in the past 12 months [23]. e most common reasons for use among university students were for relaxation with friends (84 (53.8%)), peer pressure (72 (46.2%)), and to get relief from stress (56 (35.9%)), respectively [23].
As in many developing countries, alcohol is the most frequently used substance in Ethiopia [6,25]. erefore, the present study aimed at assessing the prevalence of alcohol use and associated factors among Dilla University students. e findings are expected to inform university students about the possible dangers of alcohol use and to contribute significantly to national guidelines, targets, policies, and strategies that pertain to the management and evaluation of youth health programs in universities.

Study Design, Period, and Study Area.
e institutionalbased cross-sectional study was conducted from January to February 2018 among Dilla University students in the Gedeo Zone, Southern Ethiopia, which is found on the main road from Addis Ababa to Kenya 360 km south of Addis Ababa and 90 km south of Hawassa (capital of SNNPR). e Gedeo Zone has a longitude and latitude of 6°24′30″N 38°18′30″E with an elevation of 1570 meters above sea level [26,27].
Dilla University is one of the public universities located in the fertile coffee belt of the Gedeo Zone. Its history can be traced back to the year 1996 when it was called Dilla College of Teacher Education and Health Sciences and was included in the then newly emerging University, Debub University in 2001 EC, and became Dilla University in 2004 EC. It has three campuses, six colleges, three schools, and two institutes offering undergraduate and postgraduate programmes. Currently, more than 31,000 students take part in regular, summer, weekend, and distance programmes.

Sample Size and Sampling Technique.
e sample size was determined using a single population proportion formula by an assumption of p� 45.7%, based on a previous study from Southern Ethiopia (Hawassa University) [23] with 95% confidence interval and 5% margin of error. Assuming a 15% nonresponse rate and a design effect of two for multistage sampling, the total sample size was computed to be 876.
All regular undergraduate students attending at Dilla University were eligible to participate. A multistage sampling technique was used to select the study participants. First, students were divided into their colleges and departments. en, further stratification was used based on the year of study. Students from each year of study were allocated proportionally to the class size of their departments. "K" was calculated as 12138 ÷ 876 � 13.86∼14 for each year separately. erefore, the study subjects were selected every 14 from the list of students' names.

Data Collection Procedures and Instruments.
A structured questionnaire adapted from the previous literature on alcohol use was used to collect the data. e questionnaire had different components including sociodemographic, clinical, and alcohol use characteristics. Sociodemographic characteristics included age, sex, religion, ethnicity, monthly income, place of residency, types of high school attended, current living, year of study, friends' psychoactive substance use, and family's psychoactive substance use.
e clinical factors and psychoactive substances included the diagnosis of mental illness, family history of mental illness, history of chronic illness, khat use, and cigarette smoking.
Alcohol use was measured by a 9-item semistructured questionnaire which was prepared by reviewing the WHO students' drug use survey questionnaire, the ASSIST guideline [28], and the relevant previous literature [6,8,18,22,24,[29][30][31][32]. Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) guideline specifically asked about the following: (1) all substances a participant had ever used in his/her lifetime, (2) frequency of use in the past three months preceding the survey, (3) frequency of experiencing a strong desire to use a substance in the past three months preceding the survey, (4) frequency of health, social, legal, or economic problems related to substance use in the past three months preceding the survey, (5) frequency of substance use interfering with students' roles and responsibilities in the past three months preceding the survey, (6) if she/he has a friend or a relative or anyone else ever expressed concern about your use of the substance, and (7) if she/he has ever tried and failed to control, cut down, or stop using the substance. e questions are in increasing order of specificity.
e questionnaire explicitly asked questions about alcohol, cigarette, khat, and cannabis use as these are recognized as the most frequently tried from other studies [28]. Participants indicated their responses to the questionnaire with pen and paper.

Measurement.
e outcome variable (dependent variable) was "alcohol use" among Dilla University students.
us, the independent variables used for multiple logistic regression analysis were age, sex, religion, ethnicity, monthly income, place of residency, types of high school attended, current living, year of study, friends' psychoactive substance use, family's psychoactive substance use, diagnosis of mental illness, family history of mental illness, history of chronic illness, khat use, and cigarette smoking.

Data Quality Assurance.
e pretest was done on students with 5% of the sample size in Hawassa University. Data collectors and supervisors were trained on the data collection tool and sampling techniques. Supervision was held regularly during the data collection period by both the researcher and supervisors. e wholeness and consistency of the data were crosschecked and cleaned manually.

Statistical Analysis and
Processing. Data were entered into Epi Info version 7 computer programs and transported to SPSS version 20 computer program for further analysis and cleaning. Descriptive statistics were used to determine alcohol use and associated factors of participants. Tables were used for the data presentation. Associations of alcohol use variables and demographic characteristics were analyzed using the chi-square test, Fisher's exact test, and binary logistic regression with odds ratio and 95% CI.
Multivariate logistic regression analysis was carried out to examine the associations between each independent variable and the outcome variable. e model was checked for fitness with R-squared value. An R-squared value greater than 50% was considered as good. Hosmer and Lemeshow goodness-of-fit test was also used to check the model fitness. All variables with a p value of ≤0.25 in the bivariable analysis were considered as the candidate for multivariable regression to control possible confounders. Finally, variables with a p value of <0.05 were having a statistically significant association with alcohol use at corresponding 95% CI.

Ethical Clearance.
Ethical approval was obtained from the Institutional Review Board of Dilla University. e purpose and importance of the study were explained to each participant before they proceed into actual activities. Confidentiality was maintained by anonymous questionnaires, and informed consent was obtained from each participant.

Results
A total of 876 participants were recruited, and 803 participants completed the questionnaires sufficiently for analysis, giving the overall response rate of 91.7%.

Habit of Alcohol Drinking in Respondents.
Participants who have a habit of alcohol use were 41.8% (n � 336) and with current use in the last 30 days were 24.4% (n � 196). e most common reasons for alcohol use were for pleasure and to socialize (18.7% (n � 150) and 10.7% (n � 86), respectively). Participants' main reasons for avoiding alcohol were for the religious and economic reasons (19.3% (n � 155) and 18.7% (n � 150), respectively). However, the study participants' initiation of alcohol use was by friends and family (19.1% (n � 153) and 8.7% (n � 70), respectively) ( Table 2).

Factors Associated with Alcohol Use among Students.
In bivariate analyses, age, sex, religion, ethnicity, year of study, friends' use of psychoactive substances, family's psychoactive substance use, diagnosis of mental illness, family history of mental illness, history of chronic illness, khat use, and cigarette smoking were analyzed.
Multivariate logistic regression was also used to analyze associations between variables which have a p value of <0.2 in bivariate logistic regression. After adjusting for possible covariates, year of study, being friends with psychoactive substance user, khat use, and cigarette smoking were significantly associated with alcohol use among students with a p value<0.05 and with R-squared value of 0.573 (57.30 in percent).
Fourth-year students were about 2.66 times more likely to have alcohol use as compared to first-year students (AOR � 2.66, 95% CI: 1.64, 4.31).
Students having friends who were psychoactive substance users were 1.53 times more likely to use alcohol on the campus as compared with others (AOR � 1.53, 95% CI: 1.09, 2.1). ose who had used khat were about 1.48 times more likely to have used alcohol compared to nonkhat users (AOR � 1.48, 95% CI: 1.05, 2.09). ose who smoked cigarettes were about 1.76 times more likely to use alcohol compared to nonsmokers (AOR � 1.76, 95% CI: 1.09, 2.84) ( Table 3).

Discussion
e study has tried to determine the prevalence of alcohol use among 803 students and associated factors using a crosssectional study design and a semistructured questionnaire. e study participants were predominantly males (66.9%; n � 537), which is similar to previous studies [11,18]. Of the participants, 41.8% were lifetime users of alcohol, which is comparable to other Ethiopian student surveys (40.8% in [23] and 40.2% in [24]). Alcohol prevalence was significantly lower than a similar study done in Kenya (51.9%) [18] and Ethiopia (50.2%) [29]. However, prevalence here was significantly higher than a similar study done in Axum University (34.5%) [6] and Debre Markos University (33.8%) in Ethiopia [22]. e possible explanation for the observed difference could be due to the difference in alcohol accessibility, knowledge, and the study area. e prevalence of current alcohol use in the last 30 days was 24.4%. e finding was in line with a study done in Addis Ababa University among medical students (22%) (25% males vs. 14% females) [6], which is lower compared to a study done among Rift Valley University College students (35.6%) in Ethiopia [24]. e possible reason might be the university setup and knowledge of the students. e main reasons for alcohol use given by participants were for pleasure and to socialize (18.7% (n � 150) and 10.7% (n � 86), respectively). Similar studies done in Western Kenya among college students reported 62.2% or 60.8% to relax or relieve stress, respectively [18], a study in Ethiopia, among Debre Markos University students, reported 79.0% for relaxation [22], and a study among Axum University students reported 65.5% for relaxation or 37.7% to relieve stress, respectively [21]. Also, a large proportion of the study participants were introduced to alcohol use by friends and family (19.1% (n � 153) and 70 (8.7% (n � 70), respectively), which is similar to studies done in Western Kenya among college students showing 75.1% by friends and 23.5% by relatives [18] and a study among Debre Markos Universtiy students is showing that 63% of students were introduced to alcohol use by friends in Ethiopia [22]. Here, alcohol use prevalence increased with the seniority of the students. e fourth-year students were about 3 times more likely to have alcohol use as compared to first-year students. Similar findings were reported in Hawassa and Haramaya Universities of Ethiopia [20,23], Sudan [33], and Nigeria [34].
Students having friends who use psychoactive substances were 1.53 times more likely to use alcohol on the campuses as compared with their counterparts. e study is in line with a study done in Axum University among students; having family members and peer friends who drink alcohol were strongly associated with alcohol use [21]. Also, the finding was similar to a study done in Ethiopia among students whose friends consume alcohol were more likely to consume alcohol [35].
In this study, khat chewers had a strong association with alcohol use being about 1.48 times more likely to use alcohol as compared to counterparts.
is study supported by studies of emergency hospital cases has found that a wide variety of controlled psychoactive substances are commonly taken before, along with, or after alcohol, and both substances are affecting the body [15]. e reason might require joint effect or augment to diminish or counteract the effect of the first substance.
Cigarette smoking had a strong positive association with alcohol use with smokers being twice as likely to use alcohol compared to their counterparts. Similarly, a study done in Ethiopia among students showed that the use of cigarette was also significantly associated with alcohol consumption [21,36,37]; the two tend to co-occur. e other common reason is that many cigarette smokers use alcohol to reduce the effect of smoking.
Chewing khat was also associated with alcohol use, with chewers being 1.48 times more likely to use alcohol Journal of Addiction 5 than nonchewers. Studies of emergency hospital cases have also found that a wide variety of psychoactive substances are commonly taken before, along with, or after alcohol [15]. However, in contrast to other studies, here sex, age, family history of substance use, and student's origin of residence were not associated with alcohol use [6,37,38].

Conclusion
In conclusion, the prevalence of alcohol use among students at Dilla University was high. Being a 4th-year student, having friends who use substances, being khat user, and being cigarette smoker were associated significantly with alcohol use. In Ethiopia, being a university student appears to increase alcohol use. e findings suggest that effective campus-based counselling and peer education should be implemented for early prevention, detection, and alleviation of alcohol use among students. Moreover, longitudinal studies should be showed to check the reproducibility of alcohol use among university students in Ethiopia.

Data Availability
e data used to support the finding of this study are available from the author upon request.

Additional Points
is study did not do a detailed validation of the alcohol screening tool. e collected data were measured using a self-reported questionnaire, but it was safeguarded anonymity to reduce subjectivity bias. e cross-sectional nature of the study design might not show the causal relationships between sociodemographic and risk factors and alcohol use.t

Ethical Approval
Ethical clearance was obtained from the Institutional Review Board (IRB) of Dilla University.

Consent
Formal permission was obtained from students, and finally, written consent was obtained from each participant during data collection. All participants were well informed about the aims and purpose of the study and its contribution to the future development of the health system in the universities and the country. ose participants were informed that the right is given to the study participants to refuse and stop or withdraw from participation at any time during data collection without loss of any entitlement.

Conflicts of Interest
e author declares that he has no conflicts of interest.