Assessment of binge drinking of alcohol in highly educated employees

Assessment of binge drinking of alcohol in highly educated employees

Addictive Behaviors 28 (2003) 1299 – 1310 Assessment of binge drinking of alcohol in highly educated employees Robert A. Matano*, Cheryl Koopman, Sta...

139KB Sizes 1 Downloads 65 Views

Addictive Behaviors 28 (2003) 1299 – 1310

Assessment of binge drinking of alcohol in highly educated employees Robert A. Matano*, Cheryl Koopman, Stanley F. Wanat, Shelly D. Whitsell, Anne Borggrefe, Darrah Westrup Department of Psychiatry and Behavioral Sciences, School of Medicine, Stanford University, Stanford, CA 94305-5724, USA Accepted 19 March 2002

Abstract This study evaluated the usefulness of the Alcohol Use Disorders Identification Test (AUDIT) and CAGE, a standardized screening instrument for detecting alcohol dependence in identifying binge drinking among highly educated employees. Brochures were mailed to an entire workforce inviting employees to learn about their coping strategies, stress levels, and risk for alcohol-related problems, with 228 employees providing complete data. Binge drinking in the previous 3 months was reported by 29% of the employees, with greater binge drinking reported by White employees, of mixed/other ethnic background, or younger. The AUDIT achieved a sensitivity of 35% in identifying respondents who reported binge drinking and a specificity of 98% in accurately identifying respondents who did not report binge drinking. Sensitivity using the cut-off of scoring one or more positive hits on the CAGE was 67%, and specificity was 84%. Therefore, neither the AUDIT nor the CAGE achieved adequate sensitivity, as well as specificity, as screening tools for assessing binge drinking. A more accurate method for assessing binge drinking appears to be by directly asking for the largest number of drinks consumed in a single drinking session. D 2002 Elsevier Science Ltd. All rights reserved. Keywords: Binge drinking; Alcohol; AUDIT; CAGE; Assessment

* Corresponding author. Tel.: +1-650-725-5716; fax: +1-650-498-5760. E-mail address: [email protected] (R.A. Matano). 0306-4603/02/$ – see front matter D 2002 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0306-4603(02)00248-4

1300

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1. Introduction Problems with alcohol are varied. Most of our attention has been on the problems of alcohol abuse and alcohol dependence; however, over the last decade, particularly in the past 5 years, a considerable focus of attention has been on a relatively understudied construct, a phenomenon that has been called binge drinking. For the purpose of this study, we conceptualize binge drinking as drinking behavior in which a person consumes a high number of drinks in one session. However, the definition of binge drinking has not been established with consistency (Epstein, Kahler, McCrady, Lewis, & Lewis, 1995). The research literature over the past 5 years has been divided as to what constitutes binge drinking. This construct assumes that consuming a certain quantity of alcohol in a limited time frame could result in one or more negative consequences, such as drunk driving, date rape, physical violence, and/or other behavioral problems. What constitutes a ‘‘high number of drinks’’ differs in definitions, in part because this area of research is relatively new and is beginning to clarify the issues involved. Certain assessment tools assume a cut-off level of six (or more) drinks represents binge drinking. For example, the Alcohol Use Disorders Identification Test (AUDIT), a widely used tool that was developed by the World Health Organization to screen for problem drinking, includes an item that uses the six or more drink cut-off (Conigrave, Hall, & Saunders, 1995). Different definitions have emerged even from within the National Institute of Alcohol Abuse and Alcoholism. For example, one definition that has been used to operationalize binge drinking in national surveys of college students (Hingson, 1998; Wechsler, Molnar, Davenport, & Baer, 1999) suggests that five drinks for males and four for females constitutes problematic drinking. A cut-off of eight or more drinks within the same day has also been used to operationalize binge drinking (Nadeau, Guyon, & Bourgault, 1998). Epstein et al. (1995) not only questions what constitutes a high number of drinks, but how to operationalize a ‘‘drinking session.’’ They propose a detailed operationalization of binge drinking as a certain number of drinks over a certain number of consecutive days, following and being followed by a certain number of abstinent or light drinking days. Specifically, they define binge drinking within a 175-day period as 5–54 drinking days, with a minimum of 3 heavy drinking days and a maximum of 14 heavy and moderate successive drinking days, followed by a minimum of 14 abstinent or light drinking days, with a maximum of four deviations from this pattern. Epstein et al. have proposed such a rigorous model to allow more precise categorizations of problem drinking such as binge drinking as distinct from other problematic patterns of alcohol use. Despite the differences in these specific definitions by various researchers, the general point they have in common is to acknowledge that certain levels of drinking per occasion are problematic. The reason why there is no clear-cut consensus about what constitutes binge drinking is due in part because the quantity of alcohol consumption that would constitute binge drinking that is likely to result in negative consequences cannot be identified with certainty across individuals and settings. Individual variations in responsiveness to alcohol are partly attributable to variations in blood volume as a function of weight, age, gender, metabolic differences, and psychological factors such as expectancies about alcohol (Chris-

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1301

tiansen, Goldman, & Brown, 1985; Froehlich, 1997; Greeley & Oei, 1999; McCarthy, Wall, Brown, & Carr, 2000). Because of these individual variations, it would be impossible to identify an absolute criterion for a simple quantity of alcohol that would be valid and reliable in consistently reflecting a level of danger. However, the literature to date has established a problem with alcohol that is not at the level of abuse or dependence but that does involve heavy, extended, and dangerous drinking. Binge drinking represents a major cause of violent crimes and injuries (Shepherd, 1994). Presley, Meilman, and Cashin (1997) found that for male college students, carrying a weapon was more prevalent among those who reported binge drinking compared to those who reported no binge drinking. In addition, a high proportion of fatally injured drivers have been found to have elevated levels of alcohol (Gjerde & Morland, 1993), suggesting that episodes of binge drinking could elevate the risk of traffic accidents. Interestingly, in a National Health and Nutrition Examination Survey follow-up study (Anda, Williamson, & Remington, 1988), the number of drinks per occasion was an important risk factor for death from injury, whereas the frequency of drinking was not. There is some evidence of serious health problems being more likely among those who engage in binge drinking. It increases the risk of stroke and sudden death (Altura & Altura, 1999), cardiovascular problems (Puddey, Rakic, Dimmitt, & Beilin, 1999), renal failure (Wen, Parthasarathy, Iliopoulos, & Oberley, 1992), and possible fetal damage (Allebeck & Olsen, 1998). Furthermore, the abandonment of safe-sex techniques has been reported among young persons who engage in binge drinking, putting them at risk for sexually transmitted diseases (Meilman, 1993). Furthermore, binge drinkers have been found to have more disturbed interpersonal relationships (Jacob, Dunn, & Leonard, 1983). Given such problems, the need for early intervention requires a brief and reliable method for assessing binge drinking. Both the AUDIT (Babor, de la Fuente, Saunders, & Grant, 1989; Saunders, Aasland, Amundsen, & Grant, 1993, Saunders, Aasland, Babor, de la Fuente, & Grant, 1993) and the CAGE (Ewing, 1984) are widely used assessment tools in screening for alcohol-related problems. The AUDIT was specifically developed to assess problem drinking (Saunders, Aasland, Babor, et al., 1993) as distinguished from alcohol dependence. Since the CAGE was developed with a greater emphasis on assessing alcohol dependence, some researchers have used a different cut-off on the CAGE to identify problem drinking as contrasted with alcohol dependence (Fertig, Allen, & Cross, 1993). To evaluate the utility of a screening tool such as the AUDIT or CAGE in detecting binge drinking, two criteria are of particular importance, sensitivity and specificity. Sensitivity is the percentage of persons accurately identified as having a specific condition (Cherpitel, 1997), and in this context, as those who have been independently identified as binge drinking. Specificity is the percentage of persons who do not have a condition who have been accurately assessed as not having the condition (Cherpitel, 1997). Sensitivity is important because it indicates how well the instrument performs in identifying binge drinking. Specificity is important because it indicates how well the instrument performs in identifying those who should be excluded from being considered to have the condition. Several studies have specifically examined the utility of the AUDIT in identifying binge drinking in different populations. Shakeshaft, Bowman, and Sanson-Fisher (1998), in a study

1302

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

of 586 drug and alcohol counseling clients, compared the utility of a retrospective drinking diary, the AUDIT, and quantity/frequency measure in detecting binge drinking. The AUDIT detected the highest proportion of binge drinkers, followed by a quantity/frequency measure and the drinking diary. Previous research found that the AUDIT shows some success in identifying binge drinking among male patients in a primary medical care setting (Bradley et al., 1998). However, other research has found that the AUDIT fails to identify binge drinking. For example, Lennings et al. (1997) found that 79% of male mine workers who scored below the standard cut-off (of 8) on the AUDIT reported that they binge drank at least sometimes. See the review by Fiellin, Reid, and O’Connor. (2000). Several studies have specifically examined the utility of the CAGE in identifying binge drinking using a cut-off of one positive on the CAGE. The sensitivity of the CAGE in detecting binge drinking has ranged considerably across different populations, including a low of 59% in a study of older primary care patients (Adams, Barry, & Fleming, 1996), to 68% in African American women (Russell et al., 1994), 72% in active duty army personnel (Fertig et al., 1993), 86% in emergency room patients (Cherpitel, 1995), and as high as 88% in an elderly sample (Jones, Lindsey, Yount, Soltys, & Farani-Enayat, 1993). Specificity in detecting binge drinking has also ranged in these studies from a low of 66% (Cherpitel, 1995) to a high of 89% (Adams et al., 1996). Our previous research, which used the cut-off of 6 or more drinks in a single session to define binge drinking, found that it was 15% in a workforce sample (Matano et al., 2002). This suggests that it is important to identify tools to allow rapid and accurate assessment of binge drinking in this population. Given the widespread use of the AUDIT and CAGE as screening tools, our goal is to assess the usefulness of these tools in identifying binge drinking in the workforce. Therefore, we conducted a descriptive study to evaluate the sensitivity and specificity of these two screening tools in identifying employees who report binge drinking.

2. Methods 2.1. Research participants The institutional review board at Stanford University prior to recruiting any individuals into this study approved the ethics of human experimentation for this study. Research participants in the present study were predominately highly educated employees at a Northern California work site. Further information describing the site from which these employees were drawn cannot be provided, as we had to promise the organization that we would not name or otherwise identify the organization in order to be able to conduct our research with the employees. Brochures were mailed to the entire workforce (approximately 8570 employees) inviting them to learn about their coping strategies, stress levels, and risk for alcoholrelated problems. There were 299 employees who responded to the brochures by contacting the project staff either by telephone or e-mail. These employees were subsequently mailed a packet of printed materials containing a consent form, presurvey, and two self-addressed, stamped envelopes. Employees were asked to return the completed presurvey and their

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1303

consent form in separate envelopes to ensure confidentiality. Three hundred sixteen employees returned completed presurveys and were then mailed logon numbers to access the CopingMatters website at hhttp://copingmatters.stanford.edui. However, this study required completion not only of the presurvey but also of the next stage of the study, the website intervention, which was completed by 229 employees. Of the 229 employees, 228 provided complete data on the AUDIT and CAGE, as well as on the presurvey assessment of their binge drinking. Therefore, this study was based on the data from these 228 employees. Because this sample size constituted only about 3.5% of the employees, this is not large enough for reliably assessing the prevalence of binge drinking among these employees. Furthermore, the sample of the respondents was comprised of a higher percentage of females (76%) than was employed within the total workforce (55% female). Therefore, this sample was comprised of a substantially greater number of women than was representative of the entire workforce. We are unable to report percentages of the overall workforce by age, ethnicity, education level, and occupational status. However, we are able to describe our sample on these characteristics. Their mean age was 40.5 years (S.D. = 11.2). Their ethnic backgrounds were: 79% Caucasian, 8.4% Asian/Pacific Islander, 4% African American, 3.1% Hispanic Native American/Alaskan Native (0.9%), Middle Eastern (0.4%), mixed (1.8%) or other (2.29%). Education levels were predominantly high: 9% had a doctoral degree, 24% had a master’s degree, 43% had a bachelor’s degree, 22% had some college, and only 2% had not attended college at all. The distribution of occupational status included: 39% professional, 39% administrative support, 18% management, and 4% other. Because we do not have the data available to us on the distribution of these characteristics in the total workforce, we are unable to evaluate how representative our sample is of the workforce from which it was drawn. This raises questions about the generalizability of our sample. However, because we were focused not on prevalence of binge drinking and scores on the screening measures but on the relationships among these measures, we were able to evaluate the sample size of 228 as adequate to provide an indication of the sensitivity and specificity of the AUDIT and CAGE for identifying those who reported binge drinking. 2.2. Measures 2.2.1. Assessment of binge drinking For this study, we defined binge drinking using the criteria suggested by Wechsler et al. (1999). This definition uses a cut-off of five or more drinks for males and four or more for females. We operationalized our assessment of binge drinking by asking in the presurvey questionnaire for this study: ‘‘Please estimate total number of times during past 3 months that you had 4 or more drinks on a single occasion (if female), 5 or more drinks on a single occasion (if male).’’ Any respondent who indicated that she/he had engaged in this behavior at least once was identified as binge drinking for the purposes of this study. 2.2.2. CAGE The CAGE is a standardized screening instrument developed by Bush, Shaw, Cleary, et al. (1987) and is used to assess participants’ alcohol use. The CAGE screens for both current and

1304

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

lifetime alcohol dependence and consists of four questions: (1) Have you ever felt you should cut down on your drinking? (2) Have people annoyed you by criticizing your drinking? (3) Have you ever felt bad or guilty about your drinking? (4) Have you ever had a drink first thing in the morning to steady your nerves or to get rid of a hangover (eye opener)? The usefulness of the CAGE is in the instrument’s brevity and rapid screening capability, and it is also highly valued for its strong clinical validity. When two or more items are endorsed on the CAGE, sensitivity ranges from 72% to 91% and specificity ranges from 77% to 96% (Beresford, Blow, Hill, et al., 1990; Bush et al., 1987). When using a cut-off point of one endorsed item on the CAGE, sensitivity is reported at approximately 79% and specificity at 67% (Cherpitel 1995; Fertig et al., 1993). When screening for heavy drinking, Bradley et al. (1998) found that the CAGE’s sensitivity ranged between 49% and 69% and the specificity ranged between 75% and 95%. 2.2.3. Alcohol Use Disorders Identification Test We used the AUDIT, a standardized screening instrument developed by the World Health Organization (WHO) (Babor, de la Fuente, Saunders, et al., 1992) to assess current problem drinking. The AUDIT consists of the following 10 questions pertaining to average quantity of alcohol consumed, frequency of consumption, binge drinking patterns, and signs of alcohol dependence (Reid, Fiellin, & O’Connor, 1999). (1) How often do you have a drink containing alcohol? (2) How many drinks containing alcohol do you have on a typical day when you are drinking? (3) How often do you have six or more drinks on one occasion? (4) How often during the last year have you found that you were not able to stop drinking once you had started? (5) How often during the last year have you failed to do what was normally expected from you because of drinking? (6) How often during the last year have you needed a first drink in the morning to get yourself going after a heavy drinking session? (7) How often during the last year have you had a feeling of guilt or remorse after drinking? (8) How often during the last year have you been unable to remember what happened the night before because you had been drinking? (9) Have you or someone else been injured as a result of your drinking? (10) Has a relative or friend, or a doctor or other health worker been concerned about your drinking or suggested you cut down? Participants’ responses were weighted according to the AUDIT’s scoring procedure, with a score of 8 or more indicating a strong likelihood of hazardous or harmful alcohol consumption (Babor et al., 1992). The AUDIT is a widely respected standard screening instrument that has been extensively evaluated with different cut-off points and criterion standards in a variety of clinical settings (Fleming, Barry, & MacDonald, 1991; McQuade, Levy, Yanek, Davis, & Liepman, 2000; Steinbauer, Cantor, Holzer, & Volk, 1998; Volk, Steinbauer, Cantor, & Holzer, 1997). Using a cut-off point of 8 or more endorsed items, Babor et al. (1992) reported that the instrument’s overall sensitivity value at 92% and a specificity value of 93%. Saunders, Aasland, Amundsen, et al. (1993) and Saunders, Aasland, Babor, et al. (1993) found that when using a cut-off point of 8, the sensitivity and specificity of the AUDIT were 97% and 78%, respectively, for hazardous alcohol use, as compared to a sensitivity of 95% and a specificity of 85% for harmful alcohol use. Steinbauer et al. (1998) reported that the AUDIT and CAGE both successfully detected alcohol abuse or depend-

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1305

ence in an ethnically diverse population. In that study, the sensitivity of the AUDIT ranged between 70% and 92%, with a specificity of between 73% and 94% using a cut-off point of 8 (Steinbauer et al., 1998). 2.3. Data analysis procedures We computed a 2  2 crosstabulation of the frequencies analyzing respondents’ reports as binge drinkers or not (using the criteria described above in Section 2.2) by whether or not they met criteria on the AUDIT. This yielded the percentages of respondents who reported binge drinking (or not) analyzed by whether or not they met criteria on the AUDIT as problem drinkers. We then repeated this analysis for the CAGE to evaluate the percentages of respondents who reported binge drinking (or not) analyzed by whether or not they met criteria on the CAGE. We also used chi-square analysis to evaluate the statistical significance of gender, ethnic, and educational differences in binge drinking, meeting criteria on the AUDIT and the CAGE. t Tests for independent samples were conducted to test the statistical significance of mean age differences in these three indices.

3. Results 3.1. Prevalence of binge drinking and relationships to demographic characteristics Among all of the respondents, 28.9% reported that they had engaged in binge drinking on at least one occasion in the past 3 months. The prevalence of binge drinking did not differ significantly by gender, with 27.6% of the women reporting binge drinking, compared with 33.3% of the men, [c2(1) = 0.66, P = n.s.]. Binge drinking did differ significantly between ethnic groups, reported by only to 5.3% of Asian Americans, compared to 16.7% of African Americans, 32.2% of Caucasians, and 40% of mixed/other ethnic backgrounds [c2(3) = 8.01, P < .05]. Younger respondents were significantly more likely to report binge drinking compared to older respondents [Spearman’s rho (226) =  0.14, P < .05]. The prevalence of binge drinking did not significantly differ by level of education completed [c2(3) = 1.59, P = n.s.] or by type of job [c2(3) = 1.62, P = n.s.). 3.2. Relationship of binge drinking to AUDIT screening tool Using the AUDIT cut-off of scoring 8 or more ‘‘hits,’’ 11.8% of the respondents met criteria on this measure for problem drinking. Men were significantly more likely than women to meet criteria on the AUDIT, 22.2% versus 8.6% [c2(3) = 7.39, P < .01]. However, no significant differences in meeting criteria were found by age, ethnicity, or education. Table 1 shows the numbers of respondents reporting binge drinking analyzed by whether or not they scored at or above this cut-off on the AUDIT. These results show that the AUDIT achieved a sensitivity of only 34.8% (23/66) in identifying respondents who reported binge drinking but achieved a specificity of 97.5% (158/162) in accurately identifying respondents who did not

1306

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

Table 1 Analysis of employees analyzed by whether they did or did not report binge drinking in relation to their score on the AUDIT (N = 228) Did not report binge drinking (4+ for women and 5+ for men) Did not meet criteria on AUDIT (scored 0 – 7 ‘‘positives’’) Count N = 158 (Specificity = 97.5% of those who do not report binge drinking)

Reported binge drinking (4+ for women and 5+ for men)

Total

N = 43 (Inaccurately categorizes 65.2% of those who do report binge drinking)

Met criteria on AUDIT (scored 8 or more ‘‘positives’’) Count N = 4 (Inaccurately categorizes 2.5% of N = 23 (Sensitivity = 34.8% of those those who do not report binge drinking) who do report binge drinking) Total count 162 66 Percentage of total

228 100.0%

c2(1) = 47.10, P < .001.

report binge drinking. Using the single item on the AUDIT that assessed the frequency of respondents’ having consumed six or more drinks in a single session, we found that 72.7% of the binge drinkers were correctly identified (48/66). Specificity was 92.6% (150/162) using this single item to identify binge drinking. 3.3. Relationship of binge drinking to CAGE screening tool Using the cut-off of scoring one or more positive hits on the CAGE, 39.3% of the respondents met criteria. Men were significantly more likely than women to meet criteria on the CAGE, 51.9% versus 35.4% [c2(3) = 4.67, P < .05]. However, no significant differences in meeting criteria were found by age, ethnicity, or education. Sensitivity of the CAGE for identifying binge drinkers was 66.7% (44/66). Furthermore, the CAGE showed a specificity of 84.1% (116/138) in accurately identifying respondents who did not report binge drinking on the CAGE.

4. Discussion In this employee sample, 29% reported that they had engaged in binge drinking on at least one occasion in the past 3 months. The prevalence of binge drinking did not differ significantly by gender, education, or type of job; however, it did differ significantly among ethnic groups. Asian Americans reported the lowest prevalence of binge drinking, with African Americans the next lowest. Nearly a third of the Caucasians and 40% of mixed/other ethnic backgrounds reported binge drinking. The age of respondents was significantly related to binge drinking, with younger respondents reporting significantly more binge drinking compared to older respondents. These findings suggest demographic characteristics that may help to identify those employees who are more likely to engage in binge drinking, namely, employees who are younger and/or Caucasian.

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1307

Neither the AUDIT nor the CAGE achieved adequate sensitivity, as well as specificity, to support their utility as screening tools for assessing binge drinking. Although the AUDIT showed excellent specificity, accurately identifying 98% of those who did not report binge drinking, it achieved very low sensitivity in identifying binge drinking, with only 35% of binge drinkers scoring at or above the cut-off on the AUDIT. Therefore, the low sensitivity undermines the utility of the AUDIT as a screening tool for binge drinking. Although sensitivity using the cut-off of scoring one or more positive hits on the CAGE was somewhat better than on the AUDIT total score, a third of the respondents who reported binge drinking did not score above the cut-off of one or more hits on the CAGE. Therefore, the CAGE also does not appear to adequately assess binge drinking in the workforce sample. The most accurate method for assessing binge drinking of the three possibilities examined in this study was to use the single item on the AUDIT asking about the frequency of respondents’ having consumed six or more drinks in a single session. However, using that single item on the AUDIT also failed to detect a fairly high percentage, 27%, of those who reported binge drinking. Although specificity was excellent at 93%, the mediocre sensitivity of this item as a possible screening tool would diminish its utility in screening for binge drinking as recently defined. Given these findings, we would recommend that a more accurate method for assessing binge drinking is to inquire about it directly, specifically asking for the largest number of drinks consumed in a single drinking session. This is similar in format to using the single item on the AUDIT assessing the frequency of respondents’ having consumed six or more drinks in a single session. Given that the definition of binge drinking is a controversial subject, our study was constrained by having to choose a cut-off to define binge drinking. Our assessment of binge drinking was quite conservative in the sense that it might overestimate the problem. In contrast, using the AUDIT item may underestimate binge drinking because it uses six or more drinks in a single session as the cut-off, which is greater than the cut-offs of four or more drinks for women and five or more drinks for men that have become more normative since the AUDIT was developed. However, the concept of binge drinking rests upon quantity per occasion and using this concept and the newly emerging standards for binge drinking (of four or more for females and five or more for men), we can see that a significant proportion of the respondents, 29%, consumed that quantity in a single drinking episode. Detecting this group therefore becomes problematic if we were to rely upon the traditional assessment measures, namely the AUDIT and the CAGE. It is important to acknowledge several conceptual and methodological limitations of the study. In this study, we are unable to resolve the problematic issues of definition, reliability and validity of the binge drinking construct itself. There is a need for considerably greater attention to these matters than we devoted to them in this particular study. The self-report assessment of binge drinking is likely to lead to underreporting of the amount of drinking given that this is stigmatized behavior and that such behavior is often underreported (Duffy & Waterton, 1984). Given this likelihood, the prevalence found in this study of 29% of the sample reporting binge drinking is remarkably high. However, the composition of this sample may not be representative of other groups of employees, particularly because this sample was drawn from employees who were willing to participate in an Internet-based intervention on

1308

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

drinking, stress, and coping. Therefore, this self-selected sample is likely to include a high proportion of persons who would be likely to report binge drinking. In addition, this sample may also be somewhat special in being motivated to learn more about this topic. Furthermore, these findings are based on a sample that may have restricted generalizability due to being comprised of a high proportion of women, highly educated persons, and persons of Caucasian or Asian American ethnicity. The high level of education tends to restrict generalizability because it tends to be associated with predominantly economically stable and higher functioning than the patients who tend to be seen in substance abuse treatment (Humphreys & Weisner, 2000). Nevertheless, given that the focus of this study is not to assess the prevalence of binge drinking but rather to evaluate whether two highly respected measures detect this behavior, the problems with generalizability of the sample should not be considered fatal flaws in this particular context. Asking employees to reveal the largest quantity of drinks they consume in a single session must be done with full consideration for privacy and respect for the individual that such questions necessitate, for ethical, as well as for practical, reasons. In addition, as this study suggests, it is important for scientific reasons to ask the right questions in the first place to assess binge drinking among employees. Research is needed to further develop and refine the tools used to ask these questions. Acknowledgements This study was supported by a grant from the Center for Substance Abuse Prevention, no. 5 U1K SPO8101-02. Dr. Anne Borggrefe is supported by the Swiss National Science Foundation and the Novartis-Stifung (previously Ciba-Geigy Jubila¨umsstiftung). We wish to acknowledge the contributions to this study of the following persons: Dr. Deborah Galvin, Project Officer, Center for Substance Abuse Prevention. References Adams, W. L., Barry, K. L., & Fleming, M. F. (1996). Screening for problem drinking in older primary care patients. JAMA, the Journal of the American Medical Association, 276, 1964 – 1967. Allebeck, P., & Olsen, J. (1998). Alcohol and fetal damage. Alcoholism: Clinical and Experimental Research, 22, 329S – 332S. Altura, B. M., & Altura, B. T. (1999). Association of alcohol in brain injury, headaches, and stroke with braintissue and serum levels of ionized magnesium: A review of recent findings and mechanisms of action. Alcohol, 19, 119 – 130. Anda, R. F., Williamson, D. F., & Remington, P. L. (1988). Alcohol and fatal injuries among US adults: Findings from the NHANES I epidemiologic follow-up study. JAMA, the Journal of the American Medical Association, 260, 2529 – 2532. Babor, T. F., de la Fuente, J. R., Saunders, J., & Grant, M. (1992). AUDIT The alcohol use disorders identification test: Guidelines for use in primary health care. Geneva: World Health Organization. Babor, T. F., de la Fuente, J. R., Saunders, J., & Grant, M. (1989). The alcohol use disorders identification test: Guidelines for use in the primary health care. Geneva: World Health Organization.

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

1309

Beresford, T. P., Blow, F. C., Hill, E., Singer, K., & Lucey, M. R. (1990). Comparison of CAGE questionnaire and computer-assisted laboratory profiles in screening for covert alcoholism. Lancet, 336, 482 – 485. Bradley, K. A., McDonell, M. B., Bush, K., Kivlahan, D. R., Diehr, P., & Fihn, S. D. (1998). The AUDIT Alcohol consumption questions: Reliability, validity, and responsiveness to change in older male primary care patients. Alcoholism: Clinical and Experimental Research, 22, 1842 – 1849. Bush, B., Shaw, S., Cleary, P., Delbanco, T. L., & Aronson, M. D. (1987). Screening for alcohol abuse using the CAGE questionnaire. American Journal of Medicine, 82, 231 – 235. Cherpitel, C. J. (1995). Screening for alcohol problems in the emergency room: A rapid alcohol problems screen. Drug and Alcohol Dependence, 40, 133 – 137. Cherpitel, C. J. (1997). Brief screening instruments for alcoholism. Alcohol Health and Research World, 21, 348 – 351. Christiansen, B. A., Goldman, M. S., & Brown, S. A. (1985). Differential development of adolescent alcohol expectancies may predict adult alcoholism. Addictive Behaviors, 10, 299 – 306. Conigrave, K. M., Hall, W. D., & Saunders, J. B. (1995). The AUDIT questionnaire: Choosing a cut-off score. Addiction, 90, 1349 – 1356. Duffy, J. C., & Waterton, J. J. (1984). Under-reporting of alcohol consumption in sample surveys: The effect of computer interviewing in fieldwork. British Journal of Addiction, 79, 303 – 308. Epstein, E. E., Kahler, C. W., McCrady, B. S., Lewis, K. D., & Lewis, S. (1995). An empirical classification of drinking patterns among alcoholics: Binge, episodic, sporadic, and steady. Addictive Behaviors, 20(1), 23 – 41. Ewing, J. A. (1984). Detecting alcoholism. The CAGE questionnaire. JAMA, the Journal of the American Medical Association, 252, 1905 – 1907. Fertig, J. B., Allen, J. P., & Cross, G. M. (1993). CAGE as a predictor of hazardous alcohol consumption in U.S. Army personnel. Alcoholism: Clinical and Experimental Research, 17, 1184 – 1187. Fiellin, D. A., Reid, M. C., & O’Connor, P. (2000). Screening for alcohol problems in primary care: A systematic review. Archives of Internal Medicine, 160, 1977 – 1989. Fleming, M. F., Barry, K. L., & MacDonald, R. (1991). The Alcohol Use Disorders Identification Test (AUDIT) in a college sample. International Journal of Addiction, 26, 1173 – 1185. Froehlich, J. C. (1997). Opioid peptides. Alcohol Health and Research World, 21, 132 – 136. Gjerde, H., & Morland, J. (1993). Incidence of alcohol and drugs in fatally injured car drivers in Norway. Accident Analysis and Prevention, 4, 479 – 483. Greeley, J., & Oei, T. (1999). Alcohol and tension reduction. In K. E. Leonard, & H. T. Blane (Eds.), Psychological theories of drinking and alcoholism (2nd ed.) (pp. 14 – 53). New York, NY: Guilford Press. Hingson, R. W. (1998). College-age drinking problems. Public Health Reports, 113, 52 – 54. Humphreys, K., & Weisner, C. (2000). Use of exclusion criteria in selecting research subjects and its effect on the generalizability of alcohol treatment outcome studies. American Journal of Psychiatry, 157, 588 – 594. Jacob, T., Dunn, N. J., & Leonard, K. (1983). Patterns of alcohol abuse and family stability. Alcoholism: Clinical and Experimental Research, 7, 382 – 385. Jones, T. V., Lindsey, B. A., Yount, P., Soltys, R., & Farani-Enayat, B. (1993). Alcoholism Screening Questionnaires: Are they valid in elderly medical outpatients? Journal of General Internal Medicine, 8, 674 – 678. Lennings, C. J., Feeney, G. F., Sheehan, M., Young, R. McD., McPherson, A., & Tucker, J. (1997). Work-place screening of mine employees using the Alcohol Use Disorders Identification Test (AUDIT) and alcohol breathalyzation. Drug and Alcohol Review, 16, 357 – 363. Matano, R. A., Wanat, S. F., Westrup, D., Koopman, C., & Whitsell, S. (2002). Prevalence of alcohol and drug use in a highly educated workforce. Journal of Behavioral Health Services & Research, 29, 30 – 43. McCarthy, D. M., Wall, T. L., Brown, S. A., & Carr, L. G. (2000). Integrating biological and behavioral factors in alcohol use risk: Role of ALDH2 status and alcohol expectancies in a sample of Asian Americans. Experimental and Clinical Psychopharmacology, 8, 168 – 175. McQuade, W. H., Levy, S. M., Yanek, L. R., Davis, S. W., & Liepman, M. R. (2000). Detecting symptoms of alcohol abuse in primary care settings. Archives of Family Medicine, 9, 814 – 821.

1310

R.A. Matano et al. / Addictive Behaviors 28 (2003) 1299–1310

Meilman, P. W. (1993). Alcohol-induced sexual behavior on campus. Journal of American College Health, 42(1), 27 – 31. Nadeau, L., Guyon, L., & Bourgault, C. (1998). Heavy drinkers in the general population: Comparison of two measures. Addiction Research, 6(2), 165 – 187. Presley, C. A., Meilman, P. W., & Cashin, J. R. (1997). Weapon carrying and substance abuse among college students. Journal of American College Health, 46, 3 – 8. Puddey, I. B., Rakic, V., Dimmitt, S. B., & Beilin, L. J. (1999). Influence of pattern of drinking on cardiovascular disease and cardiovascular risk factors—a review. Addiction, 94, 649 – 663. Reid, M. C., Fiellin, D. A., & O’Connor, P. G. (1999). Hazardous and harmful drinking in primary care. Archives of Internal Medicine, 159, 1681 – 1689. Russell, M., Martier, S. S., Sokol, R. J., Mudar, P., Bottoms, S., Jacobson, S., & Jacobson, J. (1994). Screening for pregnancy risk-drinking. Alcoholism: Clinical and Experimental Research, 18, 1156 – 1161. Saunders, J. B., Aasland, O. G., Amundsen, A., & Grant, M. (1993). Alcohol consumption and related problems among primary health care patients: WHO collaborative project on early detection of persons with harmful alcohol consumption—I. Addiction, 88, 349 – 362. Saunders, J. B., Aasland, O. G., Babor, T. F., de la Fuente, J. R., & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption—II. Addiction, 88, 791 – 804. Shakeshaft, A. P., Bowman, J. A., & Sanson-Fisher, R. W. (1998). Comparison of three methods to assess binge consumption: One-week retrospective drinking diary, AUDIT, and quantity/frequency. Substance Abuse, 19, 191 – 203. Shepherd, J. (1994). Violent crime: The role of alcohol and new approaches to the prevention of injury. Alcohol and Alcoholism, 29, 5 – 10. Steinbauer, J. R., Cantor, S. B., Holzer, C. E., & Volk, R. J. (1998). Ethnic and sex bias in primary care screening tests for alcohol use disorders. Annals of Internal Medicine, 129, 353 – 362. Volk, R. J., Steinbauer, J. R., Cantor, S. B., & Holzer, C. E., III (1997). The Alcohol Use Disorders Identification Test (AUDIT) as a screen for at-risk drinking in primary care patients of different racial/ethnic backgrounds. Addiction, 92, 197 – 206. Wechsler, H., Molnar, B. E., Davenport, A. E., & Baer, J. S. (1999). College alcohol use: A full or empty glass? Journal of American College Health, 47, 247 – 252. Wen, S. F., Parthasarathy, R., Iliopoulos, O., & Oberley, T. D. (1992). Acute renal failure following binge drinking and nonsteroidal antiinflammatory drugs. American Journal of Kidney Diseases, 20, 281 – 285.