Abbrivations

AUDIT: Alcohol Use Disorders Identification Test; KDCA: Korea Disease Control and Prevention Agency; LR: Likelihood Ratio; SGIS: Statistical Geographic Information Service

Introduction

South Korean society still considers alcohol-related problems to be a predominantly male-centric issue rather than one related to women. However, recently, drinking among women has increased, especially among younger age groups. In particular, the proportion of high-risk drinking among Female College Students (FCS; 17.2%) is almost two times higher than that among other women in the same age group (9.6%) [1]. If students acquire improper drinking habits during their college days when they are legally allowed to drink these habits might persist after entering society upon graduation and become normalized within the community [2].

Women have a lower body water content and fewer alcohol metabolizing enzymes than men. Due to changes in their hormone levels during their menstrual cycle, they tend to have a more robust physical response to alcohol than men [3]. Harmful drinking has differing impacts on women than on men [4,5]. Regarding drinking motives, women are more likely to drink to escape negative emotions such as depression or anxiety [4]. Further, women tend to be more intimately affected by the drinking habits of those close to them. Female drinkers are more prone than males to endure emotional trauma, which leads to alcohol consumption [6].

An appropriate ecological model

According to the ecological model suggested by McLeroy, et al. humans continue to evolve and adjust to the environment while interacting with it; thus, as various factors can influence health problems, individual strategies should be prepared for each disease [7]. This ecological approach is primarily implemented in the smoking prevention industry; it emphasizes the influence of local communities’ participation and highlights the necessity of maximizing local resources for communities health [8,9]. Likewise, drinking is influenced by the environmental characteristics of the local community’s institutions and cultures [10].

Existing ecological model-based research on health problems related to harmful drinking has revealed that personal factors include socio-demographic and psychosocial characteristics [11-13]. Intrapersonal factors include drinking-related traits, alternatives to drinking, and social activities involving parents, family, and friends; institutional and local society factors include school characteristics as well as the social and physical features of the local community [11,14]. Few studies comprehensively approaching the problem of harmful drinking have focused on women. Furthermore, research on drinking has tended to focus on individual changes with regard to drinking habits. Thus, developing a macro perspective that considers environmental changes is critical.

Study aims and objectives

We aimed to identify the current drinking situation among FCS to facilitate the early detection of harmful drinkers before their symptoms become severe, determine the factors that affect their drinking, and prepare a basis for developing prevention and health promotion programs through pre-emptive interventions. The study objectives involved comprehensively investigating personal and environmental factors affecting harmful drinking among FCS grounded in the ecological model of McLeroy, et al. The specific objectives were:

1. To scrutinize the state of harmful drinking among FCS.
2. To explore the ecological factors affecting harmful drinking among FCS.

Materials and Methods

Study design

Using the ecological model of McLeroy, et al. we comprehensively identified personal and environmental factors affecting harmful drinking among FCS.

Setting

The sample included FCS from eight universities in Seoul, South Korea, who were recruited through convenience sampling. We posted notices to recruit the participants, and an explanation of the study was made available on the universities’ official/unofficial homepages. If a female student wished to participate, she was provided a link to an online survey.

Ethical considerations

This study was approved by the Institutional Review Board of Yonsei University (Y-2018-0048) and was conducted in accordance with the 1964 declaration of Helsinki and its later amendments or comparable ethical standards. The participants expressed their intention to voluntarily take part through the recruitment announcement and received a description of the study. The description explained the study’s purpose and potential risks and benefits. They were also informed that participation was voluntary and that they could withdraw at any time.

voluntarily take part through the recruitment announcement and received a description of the study. The description explained the study’s purpose and potential risks and benefits. They were also informed that participation was voluntary and that they could withdraw at any time.

Participants

The participants were FCS aged 18-29 years who understood the purpose of the research and voluntarily agreed to participate.

Study size

The total number of respondents was 229. To identify the minimum necessary sample size, we used the G-Power 3.1 program. To perform logistic regression analysis, we calculated the odds value of the estimated H₀ value by checking the ratio of harmful drinkers among FCS in previous studies [13,15]. When we calculated input values as Pr (Y=1|X=1) H0=0.45, α=0.05, power=0.80, odds ratio=1.55, the number was 182. Thus, the sample size was appropriate.

Variables and measurement

Harmful drinking: The Alcohol Use Disorders Identification Test (AUDIT) was developed by Babor to detect early harmful drinking and alcohol dependence [16]. To measure harmful drinking, we used the Korean translation of the test: AUDIT-K [17]. It contains 10 questions and is measured on a 5-point Likert scale. A higher total score indicates a greater risk of drinking. We set 8 points as the threshold for harmful drinking, as recommended by the World Health Organization (WHO) and the developer of the AUDIT [7]. In Lee, et al. study, Cronbach’s α=0.92; the value in this study was 861 [17].

Intrapersonal factors: We gauged drinking motivation using an instrument developed by Shin and Han. The instrument consists of four motivation sub-domains: Enhancement, social, coping, and conformity. Each domain consisted of four items (16 questions total), each measured on a 5-point Likert scale. A higher score indicates a greater related motivation in each domain. Cronbach’s α in Shin and Han’s study was 0.76-0.89. In this study, it was 0.879. Cronbach’s α for each domain was 0.784, 0.878, 0.902, and 0.790, respectively [18].

The instrument for measuring childhood trauma was initially created by Bernstein and Fink and translated by Yu, et al. who confirmed its validity [19,20]. It contains 28 questions in five sub-categories-emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect-as well as three questions for testing validity. Responses are given on a 5-point scale (1=not at all to 5=definitely yes).

The questionnaire also includes reverse questions (1=definitely yes to 5=not at all). For childhood trauma, Cronbach’s α was 0.925. For each sub-category, Cronbach’s α was 0.781, 0.812, 0.699, 0.847, and 0.630, respectively. For the validity scale, Cronbach’s α=0.837.

We measured stress perception levels and ways to reduce them using a question from the 2017 Korean Youth Health Behavior Survey, conducted directly by the Korea Centers for Disease Control and Prevention [21]. Respondents rated stress levels on a 4-point rating scale (1=rarely or none to 4=mostly) to the item, “How much stress do you feel in daily life?” Higher scores imply higher stress levels.

We formulated a question on how to relieve stress. The respondents chose from the fol-lowing items in response: Do nothing, sleep, sing songs, exercise, eat food, or drink.

Interpersonal factors: To determine family drinking history, we measured respondents’ perceptions of their parents’ drinking. Specific questions were developed based on the AUDIT. The respondents answered three items about their parents’ drinking frequency, the amount consumed in a single drinking session, and the frequency of heavy drinking on a 5-point Likert scale. Higher scores signal deeper levels of parents’ harmful drinking (Cronbach’s α=0.758).

We measured the respondents’ social activity levels using a question on regular exercise from the Adolescent Health Behavior Survey. For the present study, we created some questions regarding intra-mural club activities and extramural social activities. Whether the respondent used college facilities (e.g., performance and cultural facilities, gyms) was also enquired by revising the question formed by Kim, et al. [22].

We also gauged the drinking behavior of the respondents’ best friends. We created four questions on friends’ drinking habits relating to drinking frequency, amount of alcohol consumed when drinking, the frequency of heavy drinking, and dependent drinking, respectively. Furthermore, the respondents answered questions using a 5-point Likert scale. Higher scores indicate greater harmful drinking levels among one’s friends (Cronbach’s α=0.845).

Institutional and community factors

Institutional factors:
a) Variables regarding the general characteristics of each selected university concerned whether it was public or private, and whether it was a girls or a co-educational university. We obtained this information from the relevant university’s homepage.

b) Drinking-related characteristics of the selected universities, as perceived by the respondents.

To examine the drinking-related characteristics of the universities, we revised questions from Kim, et al [22]. We used the mean value of scores of respondents belonging to a certain university as the university value. We measured the degree to which a university was strict or tolerant toward drinking using the item, “What are the rules or practices for alcohol prevention in your current school?” rated on a 7-point scale. (0=none at all to 6=applied very strictly). Higher scores indicate more strictness from the relevant university.

Liquor accessibility refers to the degree to which liquor can be accessed on campus under diverse conditions. The variable consists of six questions, and respondents answered on a 4-point Likert scale. Higher scores signal greater degrees of strictness. Accessibility to in-tra-mural facilities refers to the extent to which numerous facilities (e.g., performance halls, video and audio facilities, coffee shops, gyms, Swimming pools, and tennis courts) are accessible to students; the respondents answered on a 4-point Likert scale. Higher scores imply greater degrees of ease.

Community factors

Community characteristics: We obtained data for judging the local community level based on administrative data from homepages for cities/counties/districts and the Statistical Geographic Information Service (SGIS; provided by Statistics Korea) [23]. Specifically, we collected local data on the degree of social instability (including rates of crime, divorce, and suicide); rates of harmful driving, obesity, smoking, and stress perception; the usage rate of facilities meant for cultural activities; and the number of coffee shops, Internet cafes, and bars in the community.

Subjective local environment: To examine the respondents’ subjective perceptions of the environment near their campus, we revised the questions used in the 2017 Community Health Survey conducted by the Korea Disease Control and Prevention Agency (KDCA). We measured the items on a 4-point Likert scale; they involved the ease of finding places to drink, exercise facilities, and cultural activities. Higher scores indicate greater degrees of ease.

Statistical Methods

We used the SPSS 23.0 program to analyze the descriptive statistics and factors affecting harmful drinking.

• We employed STATA 13.0 to check the significance between the models in the logistic regression analysis.
• We analysed harmful drinking and ecological factors using descriptive statistics such as frequency and mean.
• We identified ecological factors that influence harmful drinking using hierarchical logistic regression analysis. We identified the models’ goodness-of-fit based on the work of Hosmer and Lemeshow test, and pinpointed the difference in significance between models via the Likelihood Ratio (LR) test.

Results

General characteristics of FCS

Table 1 outlines the respondent’s general characteristics.

Table 1: General characteristics of FCS (N=229).

Drinking status and harmful drinking among FCS

Table 2 portrays the drinking status and harmful drinking levels among the respondents. Among the 229 respondents, an overwhelming majority (n=202, 88.2%) drank, while 27 (11.8%) did not.

Table 2: Drinking status and harmful drinking of FCS (N=229).

We administered the AUDIT to 202 students who reported that they engaged in drinking alcohol. The average AUDIT score was 11.49 (SD=7.11; range: 2-36). A total of 69 (34.2%) scored less than 8 points, while 133 (65.8%) scored 8 points or more.

Additional analysis of the total AUDIT score shows that skewness was 0.91 and right-skewed, and kurtosis was 0.30 and leptokurtic. The largest share of respondents had a critical score of 8, and 31 (14.5%) had AUDIT scores that were 20 points or above (indicating possible alcohol dependence).

Ecological factors influencing harmful drinking

To identify factors that influence harmful drinking among FCS, we established a hierar-community chical model by inputting variables related to interpersonal, intrapersonal, institutional, and the factors; the variables and results are depicted in Table 3. Each of the variables incorporated into this model was significant in the univariate analysis.

Table 3: Logistic regression analysis of ecological factors on harmful drinking (N=202).

The test of model significance demonstrated χ²=126.014 (p<0.001), which means that it is significant. The goodness-of-fit test, using the work of Hosmer and Lemeshow, revealed that χ²=10.423, p=0.237, which is greater than the significance probability of 0.05. Thus, the model was good. Further, Nagelkerke i²=0.642, meaning the model’s explanatory power was 64.2%. As for interpersonal factors, we found that when social drinking motivation rises by 1 point on the Likert scale, the odds of harmful drinking increase by 1.522 times (95% Confidence Interval (CI): 1.230-1.883), and when enhancement drinking motivation goes up by 1 point, the odds increase by 1.345 times (1.102-1.643). Meanwhile, when conformity drinking motivation increases by 1 point, the odds of harmful drinking fall by 0.765 times (0.631-0.928). If one relieves stress with drinking, the odds increase by 5.380 times (0.835-34/641).

As an intrapersonal factor, the use of performance and cultural facilities increased the odds of harmful drinking by 4.523 times (1.585-12.905). Those with friends who drank tended to be 1.170 times more likely to become harmful drinkers, but this was not statistically significant.

Regarding institutional and community factors, the accessibility of intra-mural gyms de-creases the odds of harmful drinking. We found that a 1-point rise in such accessibility reduces the odds of drinking by 0.411 times (0.218-0.777). Further, an increase in one drinking place in the neighbourhood increases the odds of harmful drinking by 1.003 times, and a 1-point increase in the ease of using exercise facilities in the community decreases the odds of harmful drinking by 0.505 times. However, these differences were not statistically significant.

There were some changes in the models’ explanatory power. The Nagelkerke R² value of Model 1 was 49.2%, and the Nagelkerke R² of Model 2-where the intrapersonal factor was added to personal factors-was 55%. The 6% increase in explanatory power from Models 1 to 2 remained significant through the LR test (χ²=12.87). The Nagelkerke R² of Model 3 (and the final model) was 0.642. The explanatory power of Model 3, which included interpersonal, intrapersonal, and institutional and local factors, was 64.2%. This represented a 9% increase from Model 2; the LR test of the increase in explanatory power was also significant (χ²=16.03, p=.0136).

Discussion

Regarding the relationship between drinking motives and harmful drinking, social motivation was effective in raising the probability of harmful drinking. Female students tended to drink as a means of maintaining the closeness of interpersonal ties. According to Cox and Klinger, enhanced motivation has a stronger effect on harmful drinking [24]. According to Suh, et al. FCS are more likely to drink based on a social reason compared to male college students, and those who drink because of conformity are not likely to become harmful drinkers; this is consistent with our findings.

A study on male college students reported that enhancement motivation, a drinking motive, had the highest predictive power, indicating that motivation prioritization among male college students may differ from that of female students [25]. Moreover, alcohol drinking among female students may serve to boost their sociality. However, it is necessary to educate them to recognize how excessive drinking could lead to harmful drinking and, ultimately, to alcoholic dependence, which could result in social isolation.

Respondents who mentioned drinking as their way to relieve stress had significantly higher scores for harmful drinking than those who preferred other ways of relieving stress. A previous study suggests that stress relief is behind FCS becoming harmful drinkers [26]. Do, et al. reported that two-thirds of FCS start drinking before graduating from high school [27]. According to Kang and Kim, the earlier one starts to drink, the higher the likelihood of becoming a harmful drinker, as indicated by our results [28]. Thus, educational institutions should intervene in high school and educate students about drinking to postpone the onset of their alcohol use.

The interpersonal factor shows that the ratio of harmful drinkers among those who used performance and cultural facilities was higher than among their peers, which contradicts the results of a previous study. However, we considered general cases where students tended to drink after using such facilities and found that their odds of becoming harmful drinkers were 4.52 times higher than those who did not use the facilities. This outcome is consistent with the finding that social motivation is the most significant factor for drinking among FCS. Social gatherings held after performances and cultural events may include drinking, as indicated by the results of another study, which found that students who emphasized on-campus events and were active in social activities were 2.22 times more likely to become harmful drinkers [26].

Furthermore, family drinking history did not have a significant effect on the respondents’ harmful drinking. Many other studies besides ours have reported on the impact of family drinking history Suh, et al. however, most of their findings are rooted in a model where only family factors are included [14,29]. The model that we used includes respondents’ inner motivations, as well as family factors, and with the increase in personal inner factors, the effect of family drinking history became insignificant. Thus, it is essential to identify the impact of family drinking history by considering diverse variables according to the analytical model.

As for college and community factors, we measured the accessibility of intramural facilities based on respondents’ subjective perceptions, and we used the average value of such perceptions as the school value. We found that the odds of harmful drinking among students in universities that scored higher in terms of accessibility to such facilities were 0.411 times lower than that of students in universities that scored lower points for accessibility. This outcome is similar to that of a study dealing with drinking-related damage affected by environmental characteristics, which examined university employees and found that better access to other facilities that were available to members led to lower drinking-related harm [26]. However, future research needs to include both school-related objective environmental attributes and subjective personal variables to determine the causes of harmful drinking and damage from drinking.

We investigated university rules regarding drinking and the possibility of introducing liquor on campus, but they were not related to harmful drinking. Likewise, an educational program on drinking did not impact harmful drinking. These findings could be attributable to the fact that there was little difference in terms of rules on drinking and the possibility of introducing liquor on campus among the universities included in this study. Furthermore, even when there was such a rule, the rule was not obeyed, and most universities did not provide educational programs on drinking.

As Hoe and Jang suggested, Korean universities must include guidelines on drinking in school regulations to allow students to receive education on the assessment of their drinking problems and behaviors [30]. Moreover, it is necessary to deal with drinking problems by linking them to a course and placing professional medical personnel at the student consultation center to detect those at high risk of becoming harmful drinkers so that they can intervene actively in students’ problems through treatment programs [30]. In addition, it is necessary to pursue policies to adopt new physical environments, such as exercise facilities, to gradually replace the current drinking culture.

Many studies have identified that drinking problems are greatly influenced by social in-situations and cultures where people can easily buy and drink liquor [31-33]. Research on some areas in the US shows that the higher the accessibility to liquor and the greater the density of drinking places, the more problems there are tied to drinking and related, complicated issues such as sex-mediated infections, violent accidents, and car accidents [33]. In this study, while the density of drinking places had some effect on harmful drinking, it was not statistically significant. This result may have been derived because liquor stores do not exist separately in South Korea; people can buy liquor at any store, implying a lack of regulations and policies regarding drinking in the country. Hence, it is necessary to change the environment to lower accessibility to liquor in South Korea by adopting regulations on liquor sales and revising tax policies on liquor.

Our findings verify that the ecological model is appropriate for explaining harmful drinking among FCS. Other studies dealing with health behavior have adopted an ecological model similar to the one used in this study [9,34]. These investigations have commonly pointed out the effects of the organizations and local communities to which people belong and suggest that maximizing local resources is cost-effective for inducing behavioral changes.

We identified factors and variables affecting harmful drinking among FCS through an ecological approach and proved the necessity of comprehensively considering both personal and environmental factors. Thus, we assert that different factors should be considered when related policies are established or when programs to prevent and intervene in drinking problems are developed. In particular, we propose that gyms, among diverse intramural facilities, should be enlarged, improved and made accessible, and accessibility should be raised to enhance the physical and mental health of FCS.

Conclusion

This study provides a knowledge base for explaining harmful drinking among FCS using an ecological model; that is, we identified environmental elements, such as family and or-ganizations, as well as personal traits. This study offers a basis for developing and operating an intervention program to prevent harmful drinking among FCS. Specifically, the study establishes a foundation for requiring school authorities to strengthen drinking-related education and rules and recommends that universities and local leaders create policies to build alternative facilities, such as exercise facilities, on campuses and in nearby local communities. In particular, many activities and facilities inside and outside of the university should be expanded so that students can maintain healthy social relationships without drinking

Limitations

As this study was cross-sectional, it has limitations in determining causal relations among the variables. Future longitudinal research could determine the relationship between the variables. Furthermore, as we chose universities in Seoul for convenience sampling, we encountered the problem of the representativeness of the population. Thus, a large-scale study utilizing data from all universities across the country should be conducted to identify variances among universities and to allow for multi-level model and cross-level interaction analyses. Additionally, this study has limitations in determining institutional characteristics; we only depended on the subjective views of each student. Lastly, although the data collection period was 2018, before COVID-19, the results can be compared with the current situation in which the social environment, including university culture, has shifted due to the pandemic.

Declarations

Conflicts of interest

The authors declare no conflicts of interest.

Sources of funding

The study had no external funding.

Author contributions

All authors listed meet the authorship criteria according to the latest guidelines of the International Committee of Medical Journal Editors (ICMJE) and are in agreement with the manuscript. J.P., S.K., and C.P. were responsible for the study’s concept and design. J.P. performed data collection. J.P., S.K., and C.P. were responsible for data analysis and interpretation. J.P. wrote the manuscript under the supervision of S.K and C.P. All authors contributed to and approved the final manuscript.

Data availability

The data that support the findings are available from the corresponding author upon reasonable request.

References

1. Ju YJ, Oh SS, Park SI, Lee HJ, Yoo MG, et al. College alcohol study for alcohol-related behaviors and problems. Health Policy and Management. 2019;29(1):58-67.

   [Google Scholar]

2. Presley CA, Meilman PW, Leichliter JS. College factors that influence drinking. Journal of Studies on Alcohol, supplement. 2002;14:82-90.

   [Crossref] [Google scholar] [Indexed]

3. Fama R, Le Berre AP, Sullivan EV. Alcohol's unique effects on cognition in women: A 2020 (re) view to envision future research and treatment. Alcohol Research. 2020.

   [Crossref][Google Scholar][Indexed]

4. Lee SY, Lee E, Park SJ, Namkoong K. Clinical characteristics of female inpatients with alcohol dependence. Journal of the Korean Academic of Addiction Psychiatry. 2007;11:23-28.

   [Crossref]

5. Schulte MT, Ramo D, Brown SA. Gender differences in factors influencing alcohol use and drinking progression among adolescents. Clinical Psychology Review. 2009;29(6):535-547.

   [Crossref] [Google Scholar] [Indexed]

6. Ha JS. Study on factors that affect drinking behavior for female employees: Focused on gender role, workplace drinking subculture. 2009.
7. McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Education Quarterly. 1988;15(4):351-377.

   [Crossref][Google Scholar][Indexed]

8. Kim HK. Development of health communication strategies for health behavior change: Application of social ecological models to smoking cessation intervention. Korean Journal of Health Education and Promotion. 2010;27(4):177-188.

   [Google Scholar]

9. King JL, Merten JW, Wong TJ, Pomeranz JL. Applying a social-ecological framework to factors related to nicotine replacement therapy for adolescent smoking cessation. American Journal of Health Promotion. 2018;32(5):1291-1303.

   [Crossref][Google Scholar] [Indexed]

10. Kim KK, JeKarl J, Kwon YM, Park MS. Effects of the community environmental characteristics on drinking problems of adults: A multilevel analysis. Korean Journal of Health Education and Promotion. 2012;29(5):103-113.

    [Google Scholar]

11. Cho HC. A meta-analysis on the factors related to adolescents' alcohol use: From 1990 to 2012. Korean Journal of Youth Studies. 2014;21(6):191-221.

    [Google Scholar]

12. Kong SS. Factors associated with Multi-impulsivity in Women Patients with Eating Disorders. Journal of Korean Academy of Psychiatric and Mental Health Nursing. 2012;21(2):137-148.

    [Crossref][Google Scholar]

13. Park JY, Kim SA, Lee S. Factors affecting problem drinking in Korean college women. Journal of Korean Academy of Psychiatric and Mental Health Nursing. 2017;26(3):291-300.

    [Crossref] [Google scholar]

14. Lee ES, Bong EJ. Impact of childhood trauma and parents' problem drinking on problem drinking of their university student children: Mediating effects of experiential avoidance. Journal of Korean Academy of Psychiatric and Mental Health Nursing. 2015;24(3):207-215.

    [Crossref] [Google Scholar] [Indexed]

15. Kang SY, Kim HW. Comparison of awareness about alcohol use during pregnancy among female college students by drinking levels: Using the theory of planned behavior. Journal of Korean Academy of Nursing. 2019;49(6):690-700.

    [Crossref][Google Scholar][Indexed]

16. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The alcohol use disorders identification test. World Health Organization. 2001.

    [Crossref][Google Scholar]

17. Lee BO. Development of Korean version of alcohol use disorders identification test (AUDIT-K): Its reliability and validity. Journal of Korean Academy of Addiction Psychiatry. 2000;4:83.

    [Google Scholar]

18. Shin HW, Han SY. A preliminary study for developing drinking motives scale. Korean Journal of Psychology. 1999;5.

    [Google scholar] [Indexed]

19. Berstein DP, Fink L. Childhood trauma questionnaire: A retrospective self-report. The Psychological Corporation: San Antonio. 1998.

    [Crossref][Google Scholar]

20. Yu JH, Park JS, Park DH, Ryu SH, Ha JH. Validation of the korean childhood trauma questionnaire: The practical use in counselling and therapeutic intervention. Korean Journal of Health and Psychology. 2009;14(3):563-578.

    [Crossref] [Google scholar]

21. Korea Centers for Disease Control and Prevention. Korea Youth Risk Behavior Web-Based Survey: Survey Summaries. 2017.
22.  
23. Kim KK, Jang SO, JeKarl J. Effects of environmental correlates on alcohol-related problems among colleges. Korean Journal of Health Education and Promotion. 2006;23(3):65-83.

    [Google Scholar]

24. Cox WM, Klinger E. A motivational model of alcohol use. Journal of Abnormal Psychology. 1988;97(2):168.

    [Crossref][Google Scholar][Indexed]

25. Statistical Geographic Information System (SGIS). Current drinking and smoking rates.
26. Lee EK, Park JH. The effects of drinking motives, refusal self-efficacy, and outcome expectancy on high risk drinking. Journal of the Korean Data and Information Science Society. 2016;27(4):1047-1057.

    [Crossref][Google Scholar]

27. Do EY, Kim YH, Kim JH. Comparison of problem drinking and alcohol expectancy among female college students, workers and housewives. Journal of Korean Alcohol Science. 2009;10(1):93-106.

    [Google Scholar]

28. Ryu HS, Baek MJ. Factors influencing the drinking behavior in female university students. Journal of Korean Academy of Community Health Nursing. 2012;23(3):307-315.

    [Crossref] [Google scholar]

29. Kang MK, Kim IK. Drinking motivation, daily stress, and problem drinking behavior of female university students. Journal of the Korea Academia-Industrial Cooperation Society. 2014;15(8):5053-5061.

    [Crossref][Google Scholar]

30. Sudhinaraset M, Wigglesworth C, Takeuchi DT. Social and cultural contexts of alcohol use: Influences in a social-ecological framework. Alcohol Research: Current Reviews. 2016;35-45.

    [Crossref] [Google scholar] [Indexed]

31. Hoe M, Jang SO. Trends in alcohol problem among young adult cohorts: Effects of gender and depression. Social Science Research. 2010;26(4):327-350.

    [Google Scholar]

32. Connor JL, Kypri K, Bell ML, Cousins K. Alcohol outlet density, levels of drinking and alcohol-related harm in New Zealand: A national study. Journal of Epidemiology and Community Health. 2011;65(10):841-816.

    [Crossref][Google Scholar][Indexed]

33. Karriker-Jaffe KJ. Areas of disadvantage: A systematic review of effects of area level socioeconomic status on substance use outcomes. Drug and Alcohol Review. 2011;30(1):84-95.

    [Crossref][Google Scholar][Indexed]

34. Nielsen AL, Hill TD, French MT, Hernandez MN. Racial/ethnic composition, social disorganization, and offsite alcohol availability in San Diego County, California. Social Science Research. 2010;39(1):165-175.

    [Crossref][Google Scholar][Indexed]

35. Kim S, Kim GS. Ecological factors affecting obesity among middle school students in South Korea. Journal of School Health. 2019;89(3):181-190.

    [Crossref][Google Scholar][Indexed]

Citation: Individual and Ecological Factors as Predictors of Harmful Drinking Among Female College Students: An ecological approach ASEAN Journal of Psychiatry, Vol. 24(4) April, 2023; 1-12