Nursing students, the nurses of the future, comprise a population of great interest to the nursing profession. Any significant substance abuse among nursing students potentially affects the profession's mission and image. Alcohol consumption and its subsequent problems may limit the student's ability to learn, practice safely, and develop as a healthy, productive health professional. Schools of nursing have the opportunity to help prevent alcohol dependency in nursing students and to identify and refer dependent students for treatment.
Substance abuse among all nurses is addressed openly by the nursing profession (American Nurses Association, 1984; 1988; Campbell & Polk, 1992). Without a mandatory or uniform national reporting mechanism, however, the extent of substance abuse among nurses in the United States is unknown. Current estimates of nurses' substance abuse are one in 15 nurses employed in U.S. hospitals (Fiesta, 1990) and 7% to 16% of all nurses currently in practice (Cahill et al., 1992; Lippman, 1992). In one southeastern state, 65% of the disciplinary cases addressed by the Board of Nursing in 1991 were related to substance abuse (Johnson & Loquist, 1992). Rehabilitation and peer assistance programs for impaired nurses are being developed in most states (Sullivan, 1987). However, the process of recovery from alcohol and drug dependence continues to be devastating to the nurses involved (Hutchinson, 1992; Sullivan, 1987). Ross (1980) suggested that the nursing profession's loss of functioning nurses due to chemical dependency is equal to the graduating classes often schools of nursing each year.
Students' drinking behaviors have become an area of concern in programs of nursing. Faculty have been alerted to the behavioral characteristics which might be observed in impaired students (Morrow, 1992; O'QuinnLarson & Pickard, 1989). To complicate nursing education and alcoholism issues further, drinking behavior of nurse educators may be problematic (Haack & Hughes, 1989). Research indicates that nurse educators, who drink to escape pressure, consume significantly larger amounts and experience more problem drinking than the general population of women (Gerace, 1988).
Related to its concern about nurses and nursing students' substance abuse, the nursing profession has become increasingly accountable for the needs of clients with alcoholism. Nursing organizations have collaborated to produce standards of nursing practice for use with addicted clients (American Nurses Association, 1988). Nursing programs have examined and improved substance abuse content in nursing curricula (Campbell & Polk, 1992; Maize, 1992) and now offer academic and continuing education courses specific to alcoholism and other dependencies (Jack, 1989). Nursing students can link content related to self and client assessment of potential or actual impairment due to alcohol abuse.
While there is some indication that alcohol dependency is a problem in the nursing profession, the causal issues are unclear. There is little research to indicate whether dependency is an outcome of the demands of nursing practice, or whether persons likely to have dependency problems are attracted to nursing as a career. To explore these causal issues, this study was designed to determine the extent of problem drinking by baccalaureate students in their first nursing course before they experience the stress of nursing education. A secondary but important purpose was to use the study findings to educate students about their own and their cohort's current level of drinking problems.
Estimates of alcohol abuse among late adolescents and young college and non-college adults are highest among college students, according to periodic surveys from 1981 to 1992 (Johnston, O'Malley, & Bachman, 1993). Collegebound high school seniors are consistently less likely than their peers to report heavy drinking periods (more than 5 drinks per drinking encounter during the 2 weeks prior to the survey). However, once in college, the students' drinking evidently overtakes that of the non-college population. In 1992, 41% of college students admitted to heavy drinking over the past two weeks compared to 35% of non-college peers. Furthermore, college students demonstrated much less decline in heavy drinking over the survey years than either their age peers not in college or high school seniors. Between 1981 and 1992, this measure of heavy drinking dropped by 13.5% for high school seniors, by 10.7% for the non-college 19- to 22-year-olds, but only by 2.2% among college students. The college environment may provide more access to alcoholic beverages and peer approval of heavy drinking.
To detect drinking problems of college students and other drinking populations, the Michigan Alcoholism Screening Test (MAST) (Selzer, 1971) is often used. Studies using the MAST indicate that alcohol consumption is a problem among a significant proportion of college students. Favazza and Cannell (1977) used the MAST to evaluate the drinking patterns of 245 midwestern students from a small private college and from a large public university. The findings were that 29% of the students in the small college and 19% in the large university had scores which indicated possible alcoholism. In a similar study of college students in the United Kingdom (West, Drummond, & Eames, 1990), a shortened form of the MAST (SMAST) identified 6% of males and 1% of females with actual drinking problems and 28% of the males and 17% of the females with potential problems. In the same population, 25.6% of the male and 14.5% of the female students reported drinking excessively. Thombs (1991) found that college students' expectancies for drinking benefits (enhanced sexuality, relaxation, etc.) were better discriminators of problematic alcohol consumption than their demographic variables. In the Thombs' study of 1,148 undergraduate students at a large east coast university, the SMAST identified 13% of the sample as nondrinkers, 62% as non-problem drinkers and probable nonproblem drinkers, and 25% as problem drinkers.
Research to examine the prevalence of substance abuse among nursing students is limited. Use of the MAST to screen nursing students for drinking problems has not been reported. However, a review of related literature revealed several studies describing the drinking patterns of nursing students. Of 1,081 female nursing students at several nursing programs in western New York, 90% consumed alcoholic beverages and 25% had five or more drinks per occasion (Dittmar, Haughey, O'Shea, & Brasure, 1989). Viar and Urey (1988) queried 77 of 101 senior undergraduate nursing students in a southeastern university and found that 89% consumed alcoholic beverages and only 4% of these students reported excessive consumption. In a study of students (n=349) in Boston area schools of nursing (Wechsler & Rohman, 1982), 252 (71%) students reported no or light drinking and 97 (29%) reported intermediate or heavy drinking.
In a longitudinal study (N=283) in a large midwestern university college of nursing, the students' drinking frequency and quantity patterns were significantly lower when they were sophomore students than when they were junior and senior students (Haack, 1988). From the same study, Haack (1984) reported that, of the 103 seniors, 3% abstained, 19% drank less frequently than weekly, and 67% reported drinking in the past week. Overall, 13% reported that alcohol interfered with school, job, or both. Engs & Hanson (1989) compared 1982 and 1984 subsamples (n=291 and 170, respectively) of nursing students from a large sample of students at 72 four-year colleges and universities throughout the U.S. The data indicated stable drinking behavior among nursing students over a 2-year period. This study estimated that about 25% of nursing students may have had problems with alcohol abuse. For example, about 30% had driven a car after knowing they had too much to drink.
Alcohol consumption is a problem among a significant proportion of college students, including those in nursing. Although the research is limited, the available information on nursing students' drinking patterns is alarming, and the phenomenon needs further examination.
Responses to Michigan Alcoholism Screening Test (MAST) Items: Drinkers Only
The sample was drawn from all nursing students enrolled in the first nursing course of a baccalaureate curriculum at a large southeastern state university from Fall 1988 to Spring 1992. Of the 660 students enrolled, 315 students participated in the study. The non-participants came from a pool of absentees, refusals, loss through the mail system (TV course), and late returns. The first year course, "Self-Care Behaviors," focuses on individual health promotion and self-analysis of lifestyle practices. Two classes focus on alcohol and substance usage.
Students enrolled in the course ranged in age from 17 to 57 (mean age=22; median age=20), in academic standing from freshman to senior (freshman, 61%; sophomore, 18%; junior, 8%; and senior, 11%), and were predominantly female (92%). Twenty-three percent of the students were transfers to the university and 7% took the course by telecommunication at an off-campus site. To assure anonymity, demographics were not part of the survey; therefore, the exact demographics of the sample are unknown.
Weights Assigned to MAST Items
After approval by the Ethics Committee of the School of Medicine, student volunteers completed a questionnaire anonymously before attending a two class sequence on alcohol and substance usage. The questionnaire administrator was either a guest lecturer or the class instructor. The data were tabulated in group form and presented to the students for discussion. Students were surveyed with the Michigan Alcoholism Screening Test (MAST), and the Children of Alcoholics Screening Test (CAST). Each instrument took approximately 10 minutes to complete. Only the results of the MAST administration are presented here.
The Michigan Alcoholism Screening Test (MAST) is a widely used and tested instrument which provides a simple, quick, economical, and reliable assessment of an individual's drinking problems. The instrument determines the respondent's self-appraisal of his/her drinking habits, involvement with helping agencies, and the social, vocational, medical, legal, and familial problems frequently associated with excessive drinking rather than the quantity of consumption or psychological constructs. Self-appraisal measures such as the MAST have been found to be more valid than biochemical markers to identify substance abusers (Allen, Eckardt, & Wallen, 1988). When used alone, the MAST is not a diagnostic measure of alcoholism but is a component of the diagnostic process in clinical and other settings. In a summary of alcoholism detection studies, the MAST had a reported sensitivity of 84% to 100% and a specificity of 87% to 95% (U.S. Preventive Services Task Force, 1989).
The original version of the MAST (Selzer, 1971) contains 25 weighted items which directly ask about drinking habits, symptoms, and drinking-related problems throughout the respondents' lifetime. A shortened version of 10 items (Selzer, 1971) was proposed, but had lower reliability (Gibbs, 1983). The version used in this study contains 24 face-valid items and reflects the minor wording changes recommended by Whitfield, Davis, & Barker (1986) (Table 1). Each item was weighted (Selzer, 1971) according to the significance of the drinking problem elicited by the item (Table 2). The significance of the content in each item was determined with discriminate analyses.
The MAST has had extensive psychometric evaluation of its ability to identify alcoholics and non-alcoholics. Several studies have established criterion-related, concurrent validity of the test by comparing scores of known alcoholics with those of non-alcoholics (Selzer, 1971; Skinner & Sheu, 1982). MAST scores have also been compared with indicators from other established measures which identify alcoholics and non-alcoholics (Lechman & Umland, 1984; Magruder-Habib, Fraker, & Peterson, 1983; Ross, Gavin, & Skinner, 1990). Leonard and associates (1983) demonstrated that couples had a high level of agreement on MAST scores when evaluating each other. The MAST has demonstrated internal consistency with Cronbach's alpha values ranging from .70 to upper .80s (Selzer, Vinokur, & van Rooijen, 1975; Skinner, 1979; Skinner & Sheu, 1982; Zung, 1979). Test-retest reliability correlations were .65 to .86 in one study (Skinner & Sheu, 1982) and demonstrated no significant difference in another study (Swett, 1984).
The MAST has some recognized limitations. The scoring system may not be appropriate for all populations, especially younger students who have not yet developed the drinking-related behavior/physical problems reflected in the MAST questions. Also, the MAST does not discriminate between practicing and recovering alcoholics because drinking problems are measured over the lifetime and the recovering alcoholic maintains the high score representative of earlier drinking. Finally, an individual can easily conceal his/her drinking problems because of the face validity of each question (Otto & Hall, 1988). Further research will help to determine the most valid scoring method for the MAST (Ross, Gavin, & Skinner, 1990).
Each item was weighted according to the significance of the drinking problem elicited by the item (Table 2). The sum of the weighted items (Selzer, 1971) for each drinking student was categorized into 4 risk-related categories (Table 3): 0-3=probable normal drinker; 4=borderline score; 5-9=probable alcoholism (80% associated with alcoholism/chemical dependence); and 10 or more=alcoholism (100% associated with alcoholism) (Whitfield, Davis, & Barker, 1986). For the purpose of this study, problem drinking was viewed as being either in the category of alcoholism or in the category of probable alcoholism. When appropriate, a fifth category of no risk was established for the students who reported that they did not drink at all (DND).
The data were analyzed to assess internal consistency and to determine distributional characteristics and assignment to scoring categories. The descriptive statistics for individual students and cohorts were grouped according to year from 1988 to 1992. One-way analysis of variance was used to detect evidence of differences among the means of MAST scores for each yearly cohort.
Based on the weighted items, Cronbach's alpha for the drinking participants' scores (n=262) was .78. For all participants including the DNDs (N=315), the alpha was .79. Both of these coefficiente exceeded the .70 minimum requirement.
Description of the MAST Scores
Of the participants, 53 (16.8%) reported that they did not drink at all (DND). The DNDs were included as a distinct subgroup to analyze the scores of the total sample (N=315), and the DNDs were excluded from the sample to analyze drinkers' scores (n=262). With a possible range of scores from 0 to 53, the range of drinkers' total weighted MAST scores for 24 items was 0 (n=141) to 47 (n=1). The mean score of all participants (N=315) was 3.04 with a standard deviation of 4.97. The mean score of all drinkers (n= 262) was 3.66 with a standard deviation of 5.24.
Drinkers' scores were used to analyze responses to the MAST items. One-fourth or more of the responses to three items (Table 3) indicated areas of problem drinking in the sample. ["Normal drinker?" (Item 1), "Morning after could not remember?" (Item 2), and "Friends think normal?" (Item 6).] A notable proportion (15%) of positive responses to two items indicated other problems with drinking- ["Feel guilty about drinking?" (Item 5), and "Created a problem for spouse, family?" (Item 10).] No students reported a history of liver disease, and few reported other medical concerns which are likely to increase with age and time involved in excessive drinking.
Frequencies of MAST Categories
Risk-related categories were formed according to the sum of weighted scores. Table 3 shows the frequencies of those categories and how the proportions change with different group composition. When analyzing the total group UV=315) with DNDs as a subgroup, the DNDs comprised 16.8%, the Normal Drinker subgroup comprised 47%, the Borderline subgroup was 14.6%, Probable alcoholism subgroup was 15.9%, and the Alcoholism subgroup was 5.7% of the total. When only the drinkers were considered (n=262), 56.5% of the participants belonged to the Normal Drinker subgroup, 17.6% were in the Borderline subgroup, the Probable Alcoholism subgroup was 19.1% of the total, and the Alcoholism subgroup was 6.9% of the total.
When all 3 risk categories (Borderline, Probable Alcoholism, and Alcoholism) were combined, 36.2% of all students and 45.5% of drinking students were at some level of risk. When the two highest risk categories (Probable Alcoholism and Alcoholism) were combined, 21.6% of all students and 26% of drinking students scored in these high-risk categories.
The five MAST categories (4 drinking categories and one non-drinking category) were analyzed according to five yearly cohorts (1988 to 1992) (Table 4). The proportion of DNDs and the Normal Drinkers varied widely (Table 3). Percentages of students in the three risk categories (Borderline, Probable Alcoholism, and Alcoholism) decreased over time with the greatest reduction reported in the 1992 cohort. However, when the total weighted scores were analyzed with a one-way analysis of variance, there was no statistical difference with all students or with the drinkers (p>.05).
Over a 5-year period, approximately one-third of nursing students were "at risk" and more than one-fifth were "at high risk" for problems related to alcohol consumption. These percentages are consistent with other studies on college students in general (Favazza &. Cannell, 1977; West, Drummond, & Eames, 1990) and nursing students in particular (Enge & Hanson, 1989; Haack, 1984; Wechsler & Rohman, 1982). While nursing students did not fit the expectation that nurses are more likely to abuse alcohol than the general student population, these findings indicate that there are significant drinking problems in this sample.
MAST Category Frequencies (and Percentages) by Year
The reliability of the MAST as an instrument to measure drinking problems in nursing students was established. When administered in the classroom as a quick and easy method, the MAST was reliable with different cohorts of nursing students over 5 years.
A limitation of this study is the potential for social desirability response bias. Students may have manipulated their scores to conceal drinking problems. The nursing students may have feared that the instructor would use information on the questionnaires to determine suitability for nursing. However, no identifying data accompanied completed questionnaires and there was no change in reliability associated with whether the test administrator was the class instructor or a guest lecturer. Finally, denial of drinking problems, often associated with alcoholism, may also bias the responses.
Because different versions and scoring systems exist to obtain and interpret MAST scores, designating risk is not clearly standardized. Moreover, factors such as age may affect MAST scores. College students may not have had time to accumulate sufficient drinking experiences to indicate their actual risk. Therefore, the percentage of students who do score in an "at risk" category is particularly alarming.
Another limitation of this study is the heterogeneity of the sample. The participants were taken from the fall and spring semesters and represented all four years of undergraduate academic standing even though they were in their first semester of the nursing curriculum. The maturation which takes place among nursing students over semesters may have affected their responses. Second semester students may know more about alcohol abuse and may therefore reduce their alcohol intake or be more likely to respond to substance abuse surveys in a socially desirable manner.
A final limitation of this study is the potential for selection bias. Participation was voluntary, and students absent from class were not given an opportunity to participate. Some of the absences may have been related to drinking problems. Another unidentified but potentially important group was students who refused to complete the questionnaire. Reported response rates in other studies have ranged from 52% to 98% of the potential pool of participants (Dittmar, Haughey, O'Shea, & Brasure, 1989; Engs & Hanson, 1989; Skinner & Sheu, 1982; Wechsler & Rohman, 1982; West, Drummond, & Eames, 1990), with in-class response rates higher than other types of data collection.
Implications for Nurse Educators
Problem drinking is a significant concern for beginning nursing students. Campbell and Polk (1992) indicate that nurse educators should identify students and colleagues suspected of substance abuse and should educate students about substance abuse. One initial strategy for nurse educators is to increase the students' awareness of the problem. Self-administration of the MAST may play a valuable role in raising student awareness of alcohol problems within the student population. Presentation of cumulative class data invariably promotes self- and peeranalysis. Students can compare their individual drinking patterns with those of their classmates or college students in general. They can also determine the proportion of their classmates in risk categories.
Once students are aware of the presence of problem drinking in themselves and their classmates, they may be ready for curricular content on alcohol and substance abuse. Such a curriculum may include content on the physiologic and psychosocial effects of alcohol, diagnostic measures in health care practice, treatment modalities, referral sources, and professional accountability- Of particular importance to beginning nursing students is the effect of alcohol on cognitive learning and knowledge retention. A logical progression is to begin content in the first nursing courses related to self-assessment and then reinforce and apply substance abuse information to patients in the advanced nursing courses. Model curricular guidelines have been developed for the inclusion of this content (Naegle, 1992).
Nurse educators should be sensitive to the need to identify students with drinking problems early in their careers. Each nurse educator should be aware of available referral resources in the university and community. One notable outcome of classes on alcohol abuse is an increase in student requests for referral to treatment centers for themselves, partners, and friends.
The nurse educator can use MAST data to introduce the issue of accountability as a nursing student ultimately seeking a nursing license. Early in the educational program, the educator can prepare the student for the potential of delayed or denied access to the licensing examination if there is a record (e.g., DUI) of drinking problems. The goal of such information is to help students to understand the hazards of problem drinking not only to their health, but also to their futures as professionals.
While self-analysis of drinking was the major objective of this "Self-Care Behaviors" course, client care is the next logical concern of the nurse educator. The MAST is a useful client assessment tool when the student advances to clinical courses and settings. Nursing interventions, including referral to specialty resources, can be based upon appropriate assessment. Instructors can supervise the students' administration of the MAST to determine a client's risk. Students can compare the client's risk as measured by the MAST with other alcoholism screening tools and the client's history, physical findings, and environment. Students who gain a broader view of the client with drinking problems will be more likely to use treatment resources for clients at risk for alcoholism.
Implications for Research
Implications for future research include identifying correlates and determinants of problem drinking and ultimately developing and testing educational interventions to eliminate the problem. Some important variables to include in future studies are family drinking patterns, other health behaviors, and an array of psychological characteristics. Large, nationally representative samples of nursing students taking the MAST would increase the generalizability of findings. Longitudinal data from time points in the undergraduate program and in professional practice would provide insight into the progression of alcoholism.
While meeting psychometric standards with various samples of students, the MAST should have further psychometric evaluation with larger samples of nursing students. Also, criterion-related, concurrent validity testing would strengthen the case for using the MAST with nursing students. This evaluation should result in standardization of the test format and the scoring method.
Inquiries about MAST scores of students in other health professions could provide information about drinking behaviors of groups who ultimately will comprise the health care team. The data among student groups may be compared and/or merged for self and peer analysis.
In a sample of 315 nursing students in yearly cohorts (1988 to 1992), the MAST indicated that 21.5% had serious drinking problems. While drinking among these nursing students was similar to that of other undergraduate college students, nursing students provide direct care to clients in clinical settings and qualify to take the licensing exam. Drinking probleme can jeopardize not only their clients but also the students' futures as professional nurses. Nurse educators, by administering the MAST, may assist the student in self-assessment and behavioral change. Ultimately, the educator can encourage and supervise the use of the MAST in clinical settings to improve the care of clients with drinking problems.
- Allen, J.P., Eckardt, M. J., & Wallen, J. (1988). Screening for alcoholism: Techniques and issues. Public Health Reports, 103(6), 586-592.
- American Nurses Association. (1984). Addictions and psychological dysfunction in nursing: The profession's response to the problem. Kansas City, MO: Author.
- American Nurses Association. (1988). Standards of addictions nursing practice with selected diagnoses and criteria. Kansas City, MO: Author.
- Cahill, J., Cassidy, K, Daly, S., Deutisch, D., Harby, K, Hodgson, B., Hodgson, J., Johnson, T., & McMahon, E. (Eds.) (1992). Nurses handbook of law and ethics. Springhouse, PA: Springhouse Corp.
- Campbell, A.R., & Polk, E. (1992). Legal and ethical issues of alcohol and other substance abuse in nursing education. Atlanta, GA: Southern Council on Collegiate Education for Nursing.
- Dittmar, S.S., Haughey, B.P., O'Shea, R.M., & Brasure, J. (1989). Health practices of nursing students: A survey. Health Values, 13(2), 24-31.
- Enge, R.C., & Hanson, D.J. (1989). Alcohol knowledge and drinking patterns of nursing students over time. College Student Journal, 23(1), 82-88.
- Favazza, A.R., & Cannell, B. (1977). Screening for alcoholism among college students. American Journal of Psychiatry, 134(12), 1414-1416.
- Fiesta, J. (1990). The impaired nurse: Who is liable? Nursing Management, 21, 20, 22.
- Gerace, L. (1988). Patterns of alcohol use among nurse educators. Issues in Mental Health Nursing, 2, 189-200.
- Gibbs, L.E. (1983). Validity and reliability of the Michigan alcohol screening test: A review. Drug and Alcohol Dependence, 12(3), 279-285.
- Haack, M.R. (1984). Drinking patterns among student nurses. The International Journal of the Addictions, 12, 577-583.
- Haack, M.R. (1988). Stress and impairment among nursing students. Research in Nursing and Health, 11, 125-134.
- Haack, M., & Hughes, T. (1989). Addictions in the nursing profession. New York, NY: Springer.
- Hutchinson, SA. (1992, Summer). Nurees who violate the nurse practice act: Transformation of professional identity. IMAGE: Journal of Nursing Scholarship, 24(2), 133-139.
- Jack, L.W. ( 1989, January). The educational impact of a course about addiction. Journal of Nursing Education, 2S(I), 22-28.
- Johnson, M., & Loquist, R. (1992). The role of nurse administrators in interaction with chemically dependent nurses. National Council of State Boards of Nursing Inc., Annual Meeting Book. Chicago, IL: National Council of State Boards of Nursing.
- Johnston, L.D., 0"Malley, P.M., & Bachman, J.G. (1993). Drug use among American high school seniors, college students and young adults. Vol. II: College students and young adults. (USDHHS Publication No. NIH 93-3598). National Institute on Drug Abuse. Washington, DC: U.S. Government Printing Office.
- Lechman, A.L., & Umland, B.E. (1984). Prevalence of alcoholism in a family practice center. Journal of Family Practice, 28(6), 867-870.
- Leonard, K., Dunn, N.J., & Jacob, T. (1983). Drinking and problems of alcoholics: Correspondence between and self and spouse reports. Addictive Behavior, 8(4), 369-373.
- Lippman, H. (1992). Addicted nurses: Tolerated, tormented, or treated? RN, 55(4), 36-41.
- Magruder-Habib, K., Praker, G.B., & Peterson, C.L. (1983). Correspondence of clinicians' judgments with the Michigan Alcoholism Screening Test in determining alcoholism in Veterans Administration outpatients. Journal of Studies in Alcoholism, 44(5), 872-874.
- Maize, K. (1992). Survey of educational curricula response to chemical impairment in nursing. Unpublished results.
- Morrow, J.H. (1992, Spring). Identifying the impaired nursing student. S.C. Nurse, 7(1), 36.
- Naegle, M. (1992). Substance abuse education in nursing. Vols. 1-3). New York, NY: National League for Nursing.
- O'Quinn-Larson, J., & Pickard, M.R. (1989, March/April). The unpaired nursing student. Nurse Educator, 24(2), 36-39.
- Otto, R.K., & Hall, J.E. (1988). The utility of the Michigan Alcoholism Screening Test in the detection of alcoholics and problem drinkers. Journal of Personality Assessment, 52(3), 499-505.
- Ross, H.E., Gavin, D.R., & Skinner, H.A. (1990). Diagnostic Validity of the MAST and the Alcohol Dependence Scale in the assessment of DSM-III alcohol disorders. Journal of Studies on Alcohol, 51(6), 506-513.
- Ross, S. (1980). Alcohol and other substances: Discussion. In Alcoholism and Alcohol Abuse Among Women: Research Issues. (DHEW Publication No. ADM 80-835). Washington, DC: U.S. Government Printing Office.
- Selzer, M.L. (1971). The Michigan Alcoholism Screening Test: The quest for a new diagnostic instrument. American Journal of Psychiatry, 127(12), 1653-1658.
- Selzer, M.L., Vinokur, A., & van Rooijen, L. (1975). A selfadministered short Michigan Alcoholism Screening Test (SMAST). Journal of Studies on Alcohol, 36(1), 117-126.
- Skinner, HA. (1979). A multivariate evaluation of the MAST. Journal of Studies on Alcohol, 40(9), 831-844.
- Skinner, H.A., & Sheu, Wen-Jenn (1982). Reliability of alcohol use indices: The lifetime drinking history and the MAST. Journal of Studies on Alcohol, 43, 1157-1170.
- Sullivan, E.J. (1987, June). A descriptive study of nurses recovering from chemical dependency. Archives of Psychiatric Nursing, 1(3), 194-200.
- Swett, C. (1984). Use of the Michigan Alcohol Screening Test in a prison hospital. American Journal of Drug and Alcohol Abuse, 10(4), 563-569.
- Thombs, D.L. (1991). Expectancies versus demographics in discriminating between college drinkers: Implications for alcohol abuse prevention. Health Education Research, 6(4), 491-495.
- U.S. Preventive Services Task Force (1989). Screening for alcohol and other drug abuse. American Family Physician, 40(1), 137-146.
- Viar, V., & Urey, J-R. (1988). Survey of health-related behavior of senior baccalaureate nursing students. Journal of Holistic Nursing, 6(1), 46-51.
- Wechsler, H., & Rohman, M. (1982). Future caregivers' views on alcoholism treatment. Journal of Studies on Alcohol, 43(9), 939-955.
- West, R., Drummond, C, & Eames, K. (1990). Alcohol consumption, problem drinking, and anti-social behavior in a sample of college students. British Journal of Addictions, 5(4), 479-486.
- Whitfield, C.L., Davis, J.E., & Barker, L.R. (1986). Alcoholism. In L.R. Barker, J.R. Burton, & P.D. Zieve (Eds), Principles of ambulatory medicine (pp. 245-277).
- Zung, B.J. (1979). Psychometric properties of the MAST and two briefer versions. Journal of Studies on Alcohol, 40(8), 84S-854.
Responses to Michigan Alcoholism Screening Test (MAST) Items: Drinkers Only
Weights Assigned to MAST Items
Frequencies of MAST Categories
MAST Category Frequencies (and Percentages) by Year