Printer Friendly

Validation of the alcohol Use Disorders Identification Test in university students: AUDIT and AUDIT-C.

Abstract

The aim of this study was to determine the psychometric properties of the Alcohol Use Disorders Identification Test (AUDIT and AUDIT-C) in order to detect problems related to the consumption of alcohol in the university population. The sample consisted of 1309 students. A Weekly Alcohol Consumption Diary was used as a gold standard; Cronbach's Alpha, the Kappa index, Spearman's correlation coefficient and exploratory factor analysis were applied for diagnostic reliability and validity, with ROC curves used to establish the different cut-off points. Binge Drinking (BD) episodes were found in 3.9% of men and 4.0% of women with otherwise low-risk drinking patterns. AUDIT identified 20.1% as high-risk drinkers and 6.4% as drinkers with physical-psychological problems and probable alcohol dependence. Cronbach's alpha of 0.75 demonstrates good internal consistency. The best cut-off points for high-risk drinking students were 8 for males and 6 for females. As for problem drinkers and probable ADS, 13 was the best cut-off point for both sexes. In relation to AUDIT-C, 5 and 4 were the best cut-off points for males and females with high-risk patterns, respectively. The criterion validity of AUDIT and AUDIT-C to detect binge drinking episodes was found to have a moderate K value. The results obtained show that AUDIT has good psychometric properties to detect early alcohol abuse disorders in university students; however, it is recommended that the cut-off point be reduced to 8 in men. AUDIT-C improves its predictive value by raising the cut-off point by one unit. Items 2 and 3 should be reviewed to increase its predictive value for BD. Keywords: Alcohol-Related Disorders/diagnosis*, Psychometrics/methods*, University students, AUDIT, AUDIT-C.

Resumen

El objetivo de este trabajo fue determinar las propiedades psicometricas de los cuestionarios Alcohol Use Disorders Identification Test (AUDIT y AUDIT-C) para la deteccion de problemas por consumo de alcohol en universitarios. Participaron 1309 estudiantes, utilizando el Diario de consumo semanal como patron de referencia y para el analisis de fiabilidad y validez el alfa de Cronbach, analisis exploratorio, indice kappa, coeficiente de correlacion de Spearman y curvas ROC para los diferentes puntos de corte. Un 3,9% de hombres y un 4% de mujeres con consumo de riesgo bajo presentaba consumo intensivo de alcohol (BD). AUDIT identifico un 20,1% de Bebedor de riesgo y un 6,4% de Bebedor con problemas fisico-psiquicos y probable dependencia alcoholica. El instrumento presento un alfa-Cronbach de 0,75 demostrando buena consistencia interna. Los puntos de corte optimos fueron ocho para Bebedor de riesgo en hombres y seis para mujeres; trece puntos para Bebedor con problemas fisico-psiquicos y probable dependencia alcoholica en ambos sexos; con AUDIT-C, fueron para Consumo de riesgo cinco para hombres y cuatro para mujeres. La validez de criterio para detectar consumo intensivo de alcohol (BD) con ambas versiones presento un valor K moderado. Los resultados indican que AUDIT cuenta con adecuadas propiedades psicometricas para detectar precozmente problemas de consumo de alcohol en universitarios, recomendandose reducir su punto de corte a ocho en varones. El AUDIT-C mejora su poder predictivo aumentando en una unidad el punto de corte. Se recomienda revisar los items 2 y 3 para aumentar el valor predictivo del BD. Palabras clave: Trastornos relacionados con el alcohol, Psicometria, Universitarios, AUDIT, AUDIT-C.

AUDIT (Alcohol Use Disorders Identification Test) is a self-administered questionnaire which is short, easy to use and flexible. It is consistent with CIE-10 definitions of alcohol consumption and dependence and assesses consumption over the previous year, thereby generating information which can be useful in providing feedback for those taking the questionnaire. As one of the most extensively used screening tests worldwide, the WHO recommends its systematic use in health care and other settings (Roson, 2008).

The Spanish version has been validated in health care settings (Martinez 1996; Rubio, Bermejo, Caballero, & Santo-Domingo, 1998) and even in a university student population (Adewuya, 2005; Kokotailo et al., 2004), but as this was in different cultural contexts and with different terminology it is difficult to know whether its psychometric properties make it a suitable instrument for the Spanish university population.

AUDIT consists of ten questions, the first three of which refer to high-risk consumption, with the following three exploring potential dependence symptoms, and the final four focusing on hazardous consumption. Depending on the cutoff points, the total score enables the identification of risky consumption patterns and possible alcohol dependence. In the Spanish validation, (Rubio et al., 1998), the cut-off point for high-risk consumption was set at 9 or above for men and 6 or above for women. A short version, AUDIT-C, which uses the first three questions, is sensitive and specific enough to be a valid tool for identifying high-risk consumption.

Given the changes in alcohol consumption patterns among university students, there is great interest in developing an instrument to detect these; indeed, various studies have confirmed differences in drinking patterns among young people depending on educational background, so for example while non-university students drink more often (Dawson, Grant, Stinson, & Chou, 2004; Kypri, & McAnally, 2005), university students tend to binge-drink more (O'Malley & Johnston, 2002). Binge drinking (BD) is one of the major high-risk behaviors among university students (Cortes, Espejo, & Gimenez, 2008; Kypri et al., 2009; Wicki, Kuntsche, & Gmel, 2010), which can provoke short- and long-term problems. Recent research among young binge drinkers shows physiological changes in parameters such as blood pressure and state anxiety immediately on consumption (Vinader-Caerols, Monleon, & Parra, 2014), and structural changes in white and gray matter, neuronal hyper/hypoactivation, and cognitive impairment in memory, attention and execution associated with a pattern of binge drinking over time (Lopez-Caneda et al., 2014). While some international controversy surrounds the definition of binge drinking, the last report in 2001 by the Spanish Observatory on Drugs and Drug Addiction defines it as drinking which takes place in short sittings of a few hours, mainly at the weekend, in peer groups and with a low perception of risk. The prevalence of BD among young people aged 20 to 29 is 15.2%, according to the EDADES study of 2011/12.

For these reasons, it is important to be able to rely on instruments which allow early detection and point towards preventive measures, hence the interest in discovering the sensitivity, specificity and predictive power of AUDIT and AUDIT-C among the university population, and in determining the utility and validity of these tests.

The hypotheses considered in this study are firstly that the AUDIT questionnaire is a good tool for the early detection of drinking problems and discrimination between low-,moderate- and high-risk alcohol use, and secondly that the short version (AUDIT-C) is of sufficient reliability and diagnostic validity among the population of university students with high-risk consumption.

The aim of the study is to determine the psychometric properties of AUDIT and AUDIT-C for the detection of problems linked to alcohol consumption at university.

Method

Participants

The questionnaire was completed by a total of 1352 students selected by proportional sampling, of which 43 were rejected for not meeting basic quality criteria. This left a sample of 1309 (59.1% women) with an average age of 21.78 (SD [+ or -] 4,45) and an age range of 18 to 65. The expected frequency of consumption in the previous month was considered to be 63%, with an absolute margin of error of 2.5% (Committee for Equality and Social Welfare, Government of Andalusia, 2009)

Instruments

All participants completed an anonymous self-report with socio-demographic variables, the AUDIT questionnaire, and the weekly alcohol consumption diary.

AUDIT is a self-administered questionnaire with a scoring range of 0-40 points. The first eight questions are scored from 0 to 4, and questions nine and ten with 0, 2 and 4 points. The cut-off points proposed by Roson (2008) identify three population subtypes: low-risk drinker (with cut-offs of 0-7 for men, 0-5 for women), high-risk drinker (cut-offs 8-12 and 6-12), and drinker with physical-psychological problems and probable alcohol dependence (ADS) (cut-off 13 for both). The first three questions of this test make up AUDIT-C, in which scores of 4 or higher in men (Bush, Kivlahan, McDonell, Fihn, & Bradley, 1998) and 3 or higher in women (Bradley et al., 2003) are considered to represent high-risk drinking.

The weekly alcohol consumption diary is a record of Standard Drink Units (SDU) consumed at different times or occasions throughout the day (morning, aperitif, lunchtime, afternoon, evening meal, night). Following Rodriguez-Martos (2007a), three categories were considered: low-risk consumption (SDUs [less than or equal to] 21 and [less than or equal to] 14 for men and women respectively), moderate-risk consumption (22-27 and 15-16 SDUs) and high-risk consumption ([greater than or equal to] 28 and 17 SDUs). Special attention was paid to BD as a separate category, considered to be the consumption of five or more SDUs in a single sitting, which could represent problematic drinking behavior (Rodriguez-Martos, 2007b).

Procedures

In order to investigate the validity and internal structure of the AUDIT and AUDIT-C scores in this population, a weekly alcohol consumption diary was used as a 'gold standard'. The diary is not only a record of the quantity, frequency, typology of drinks consumed and of drinking patterns but also enables, through the evaluation of these data, the identification and classification of at-risk populations such as binge drinkers.

Data analysis

For the reliability and validity analyses of AUDIT version 20 of the SPSS and FACTOR 9.2 were used.

Reliability. Cronbach's alpha was used as a measure of internal consistency. To determine the correlation between the different sub-scales and the global score, the Spearman correlation coefficient was applied.

To assess content validity, the mean scores of two students groups with significantly different alcohol consumption (moderate- and high-risk) were compared. Similarly, the scores on each AUDIT item were compared to the mean scores obtained in the weekly consumption diary for each of the three low-, medium- and high-risk groups of students.

Internal structure validity was investigated by means of exploratory factor analysis (EFA) to assess underlying dimensionality with psychological value, bearing in mind that the original version of AUDIT already has three dimensions or factors. FACTOR 9.2 (Lorenzo-Seva & Ferrando, 2006) was used to check whether the structure obtained in this university population matched the theoretical structure. Minimum Average Partial (MAP) (Velicer, 1976) was applied as the method for determining the number of dimensions; Unweighted Least Squares (ULS) with Promin rotation was used for extraction; and the Simplimax rotation was used as the most efficient method for obtaining the most suitable cut-off points.

We carried out a multivariate analysis of variance (MANOVA) of the responses to the AUDIT in the three categories of risk established by the weekly diary of alcohol consumption (in SDUs).

The criterion or diagnostic validity of the AUDIT and AUDIT-C questionnaire scores was assessed on the basis of values obtained with the weekly consumption diary used as a gold standard. The sensitivity and specificity of each item was calculated for the three categories of risk consumption (low, moderate and high) and for BD. ROC curves were obtained and various cut-off points were considered for identifying optimal sensitivity and specificity. The efficiency of these points was evaluated with the Youden index, and the concordance between observations was established with the Kappa index, using Altman criteria for classification. To evaluate the criterion validity of the AUDIT and AUDIT-C scores for the diagnosis of BD, a benchmark was established of 5 or more SDUs consumed in a single sitting.

Results

Alcohol consumption patterns

The weekly consumption diary shows that 68.1% of students drank alcohol during the previous week, with men be ing more frequent drinkers (73.1%) than women (64.6%). Taking only those students who had consumed alcohol in the previous week (n=891), the prevalence of moderate-risk consumption was 23.7%, while 9.7% was high-risk. The average weekly intake was 15.03 SDUs among men, with significantly lower values (p = .000) among women (10.22 SDUs per week) (Table 1).

BD was found independently of whether weekly consumption patterns were low-, moderate- or high-risk; for example 3.9% of men and 4.0% of women had binge drinking episodes despite their consumption being considered low-risk.

AUDIT figures show that a quarter of the sample (26.5%) had alcohol consumption problems, of which 20.1% were high-risk drinkers and 6.4% were drinkers with physical-psychological problems and probable ADS. Among those who had drunk alcohol in the last week, these figures rose to 28.5% and 9.4% respectively (Table 2).

With regard to AUDIT-C, more than half of the drinkers (63.2%) reached the cut-off for high-risk consumption, with a predominance (p = .000) of women (36.4%, compared to 26.8% men).

Reliability study

Internal consistency. The internal consistency (Cronbach's alpha) of AUDIT was 0.75; by domain, the sub-scale high-risk consumption yielded a value of 0.83, with 0.79 obtained for dependence symptoms and 0.75 for hazardous consumption.

The highest values were obtained by questions 1, 2 and 3, which make up the high-risk consumption domain, exploring the frequency and quantity of alcohol consumed. Conversely, items 5, 6, 9 and 10 yielded the lowest scores. These questions deal with information regarding dependence symptoms, injuries suffered as a result of drinking, and the worry of others about the amount of alcohol drunk by the subject (Table 3).

Reliability of AUDIT and weekly alcohol consumption diary. Comparing AUDIT to the gold standard of the weekly consumption diary, consistent results were obtained in 90% of cases. The Spearman correlation coefficient was 0.87, pointing to a good level of reliability, and on the basis of the Altman criteria the Kappa value of 0.85 is very good (Table 4).

Reliability of AUDIT-C and weekly alcohol consumption diary. Comparing the short version, AUDIT-C, to the weekly diary measuring consumption in SDUs, reliability was also moderate to good. Spearman's correlation coefficient was 0.57, indicative of a good level of concordance. The Kappa value of 0.5 is moderate according to the Altman criteria (Table 5).

Content validity

For this analysis, students with moderate- and high-risk consumption were selected. The average AUDIT score in the moderate-risk consumption group was 9.45 among men, and 7.92 for women, while the average AUDIT scores for those with high-risk consumption were 15.19 and 13.09 for men and women respectively. Statistically significant (p [less than or equal to] .001) differences were found in AUDIT for the different types of consumption; thus the test is shown to be valid for discriminating between groups of individuals with different consumption patterns.

Analyzing the item scores in the different risk groups, we can see that they increase gradually from low- through moderate- to high-risk groups depending on the types of consumption pattern, except in item 2. This item states: How many drinks do you have on a day with normal alcohol consumption? with possible responses being: 0 (one or two), 1 (three or four), 2 (five or six), 3 (from seven to nine), or 4 (ten or more). For this question, the average score in the moderate-risk group is higher than in the high-risk group (1.81 vs. 1.64). The response "five or six drinks" in the moderate-risk consumption group represents an average of 24.15 SDUs per week, and 38.74 SDUs per week in the high-risk consumption group, which would mean that given an equal number of drinks (five or six), the intake of alcohol (SDUs) is higher in one group than in the other.

Internal structure validity

On analyzing whether the three original dimensions of AUDIT are reflected in our population, three factors are obtained: the first factor consists of items 4, 5, 6, and 7 with a reliability of 0.87; the second is made up of items 1, 2, 3, and 8 with a reliability of 0.92; the third factor covers items 9 and 10 with a reliability of 0.83. The average size of residual correlations (RMSR) was 0.02, a value which shows that the model is of acceptable fit, although the association of items found differs from the original.

When looking at two dimensions, we find one consisting of the first eight items, risk consumption, and a second one covering the two last items, hazardous consumption, which would explain 66% of the accumulated variance with a reliability of 0.82 and 0.92 respectively. The root mean square correlation (RMSR) was 0.03 (Table 6)

The different risk levels of alcohol consumption patterns are reflected in the responses to the AUDIT questions (especially 3, 5, and 8, according to MANOVA). The variability which was not explained by group differences was 32% in men (Wilks' F42.8 p<0.001) and 25% in women (F75.6 p<0.001). All AUDIT questions were found to be significant in discriminating between different risk patterns of consumption.

Criterion validity

The diagnoses provided by AUDIT and AUDIT-C were compared to the weekly alcohol intake used as the gold standard. The criterion validity of AUDIT for low-risk drinker, high-risk drinker, and drinker with physical-psychological problems and probable ADS was found to have a high K value for both sexes (0.83; 0.83; 0.92 for men and 0.88; 0.88; 0.90 for women); the criterion validity for the three categories of drinker shows strong concordance, with even better results found for women in the sub-group high-risk drinkers.

The sensitivity values in AUDIT-C were 98% for men and women, although the K values obtained by both sexes were moderate (0.49 and 0.51 respectively) with reference to the Altman criteria.

ROC curves

With regard to AUDIT, the best balance between sensitivity and specificity for detecting high-risk drinkers was obtained with a cut-off point of 7.5 for men (sensitivity of 95% and specificity of 93%) and 5.5 for women (sensitivity of 98% and specificity of 95%). For drinkers with physical-psychological problems and probable ADS, the best cut-off was 12.5 for both sexes (sensitivity of 95% for men and 86% for women, and specificity of 99% for both).

In the case of high-risk drinker, the area under the ROC curve showed very good performance for men (0.963) and excellent performance for women (0.976). For drinker with physical-psychological problems and probable ADS, the area under the curve was excellent in both sexes (0.983 for men and 0.973 for women). In terms of validity, comparing the scores with the weekly consumption diary yielded values close to 1 for the areas under the curve of both categories.

For AUDIT-C, the best balance between sensitivity and specificity for high-risk consumption was found at 4.5 for men (sensitivity of 91% and specificity of 84%) and 3.5 for women (sensitivity of 86% and specificity of 88%). The area below the curve according to sex showed the test to be very good (0.941 for men, 0.945 for women). In terms of validity, the results yielded values close to 1 for the categories under investigation when comparing scores to the weekly consumption diary.

Analyses of cut-off points

For the AUDIT categories of high-risk drinker and drinker with physical-psychological problems and probable ADS, the optimal cut-off points were found to be 8 and 13 for men, and 6 and 13 for women respectively. The best cut-offs for high-risk consumption in AUDIT-C were 5 for men and 4 for women (Table 7).

Discussion

The socio-demographic characteristics of the sample are similar to those of other studies into alcohol consumption among university students (Adewuya, 2005; Kokotailo et al., 2004; Londono, Garcia, Valencia, & Vinaccia, 2005; Martin-Montanez et al., 2011; Montano, Morales, Gomez, Vera, & Gantiva, 2011; Salazar, Avila, Perez, & Martinez, 2010; Seguel, Santander, & Alexandre, 2013; Zaldivar, Lopez, Garcia, & Molina, 2011), with women (59.01%) slightly outnumbering men, and an average age of 21.78. A third of these students live away from the family home, and it is this group that presents more frequent moderate- to high-risk consumption patterns, a greater number of high-risk drinkers, and drinkers with physical-psychological problems and probable ADS. These facts highlight the need take into consideration students' living arrangements when planning preventive measures.

Reliability estimates of AUDIT scores using Cronbach's alpha were good, both globally and for each of the subscales, thereby confirming its suitability for screening for alcohol consumption problems in the university context. The reliability values are quite similar to those found in a systematic review which included 10 studies in different contexts (Meneses-Gaya, Waldo, Regina, & Crippa, 2009), in samples of the general population (Alvarado, Garmendia, Acuna, Santis, & Arteaga, 2009; Contel, Gual, & Colom, 1999; Rubio et al., 1998), and in two samples of American (Kokotailo et al., 2004) and Chilean (Seguel et al., 2013) university students, where average values of 0.8 were obtained.

The fact that the highest scores were found in items 1, 2, and 3 (focusing on the quantity and frequency of alcohol consumption) and lower sgcores in items 5, 6, 9 and 10 (measuring the adverse consequences of drinking such as dependence, injuries or the criticism of others) is indicative of alcohol abuse, and is in line with, on the one hand, the changing pattern of alcohol consumption among young people in which binge drinking (BD) plays an important role, and the lower frequency of alcohol dependence in this sector of the population on the other.

The results of the AUDIT subscale for hazardous drinking revealed that the prevalence of students who claimed difficulties in giving up alcohol was 6.4% of the total sample and 9.4% for the drinkers subgroup; this highlights the fact that up to a tenth of the university student drinker population is at risk of ADS. A Chilean study of university drinkers (Baader et al., 2014) reported probable ADS of just 1.5%.

With regard to item 2 in AUDIT, the response option "five or six drinks" reveals that weekly consumption is higher among drinkers with moderate-risk consumption than with high-risk consumption (24.18 SDUs vs. 38.74 SDUs per week). This item reflects the quantity of drinks, not the frequency of consumption, so when analyzing content validity we have to bear in mind that an individual may consume many low-strength drinks every day, while others may have fewer but stronger drinks. Moreover, on validating AUDIT in Catalan and Spanish (Contel et al., 1999), this item performed differently compared to the validation in Spanish by Rubio et al. (1998); a consequence, it was claimed, of the translation. Given these considerations, it would appear advisable to revise item 2 in such a way as to make it possible to analyze frequency and quantity separately. Given the growth of the BD phenomenon, a further improvement would be to specify the frequency parameter more exactly, taking into consideration the number of drinks consumed at a single sitting and not on a whole day; and in terms of quantification, it would be interesting to incorporate the concept of SDUs.

With respect to item 3 in AUDIT: How often do you have six or more alcoholic drinks at a single sitting? Kokotailo et al., (2004) in their validation for American university students recommended that this should be reduced to five or more drinks for men and four or more drinks for women. If we remember that BD consists of drinking a quantity of alcohol equivalent to or greater than 5 SDUs in a single sitting, it should be pointed out that this can happen independently of the consumption pattern. In our study, on analyzing the criterion validity of those with high consumption with AUDIT, a moderate K value is obtained for both sexes (0.54 for men and 0.59 for women), with a positive predictive value of 82.8% for men and 80.2% for women. Given the above, it would be advisable to redraft this item, reducing it to five or more drinks at a single sitting and, better still, incorporating the concept of five SDUs, thereby more accurately reflecting the BD phenomenon.

The original structure of AUDIT features three dimensions: high-risk consumption (items 1-3), dependence symptoms (items 4-6) and hazardous alcohol consumption (items 7-10). The factor analysis of AUDIT, however, can produce different results depending on the prevalence of the problem in the population under consideration. In our case, the three-dimensional factorial model obtained does not exactly match the original structure since items 7 and 8 do not remain in the hazardous drinking domain: item 7 is moved to symptoms of probable dependence, and item 8 to high-risk consumption.

We believe that from the point of view of designing intervention programs for the university population, it would be more suitable to adopt a two-factor design where the AUDIT test scores discriminate with great reliability between high-risk and hazardous consumption.

Indeed, in our study university students displayed a prevalence of problematic alcohol consumption of 26.5% (with 20.1% in the high-risk drinker category, and 6.4% in drinker with physical-psychological problems and probable ADS), and two factors were obtained consisting of items 1-8 and 9-10 respectively.

Other studies of the general population also support a bi-dimensional structure (Contel et al., 1999; Lima et al., 2005; Meneses-Gaya et al., 2009), while in the Chilean university population (Seguel et al., 2013), two components were also established which were linked to consumption itself and to the adverse consequences of excessive consumption.

In terms of internal validity, the sensitivity and specificity results obtained for AUDIT (sensitivity = 32-96%, specificity = 84-96%) are among the highest published (Barry, & Fleming, 1993; Rubio et al., 1998; Saunders, Aasland, Babor, de la Fuente, & Grant, 1993; Schmidt, Barry, & Fleming, 1995), which demonstrates its utility in screening for alcohol related problems in this sector of the population.

Numerous AUDIT validation studies have achieved good sensitivity and specificity when applying different cut-off points (Adewuya, 2005; Dawson, Grant, & Stinson, 2005; Gache et al., 2005; Knight, Sherritt, Harris, Gates, & Chang, 2003; Kokotailo et al., 2004; Rubio et al., 1998), highlighting the need to establish different cut-offs depending on sex. Given that equal consumption is more likely to have repercussions among women than men, it is advisable to reduce the cut-off point for women when planning interventions (Reinert & Allen, 2002, 2007).

In terms of ROC curves, the area under the curve can be seen as an indicator of the quality of a diagnostic test, and for the category high-risk drinker in the case of AUDIT this was 0.96 for men and 0.98 for women, while for drinker with physical-psychological problems and probable ADS, the areas were 0.98 for men and 0.97 for women. These values can be considered extremely good, and better than those obtained in the Spanish validation of AUDIT carried out in primary health care (Rubio et al., 1998), where an area of 0.87 was obtained for alcohol use, and the Chilean version (Alvarado et al., 2009), with areas of 0.93 for high risk drinking, 0.88 for hazardous drinking and 0.91 for dependence. Turning to values obtained in studies of university students, the results yielded by the AUDIT validation carried out in the University of Wisconsin were worse, with areas of 0.87 for alcohol consumption and 0.77 for abuse or dependence.

In this study, the most effective cut-off points in AUDIT for problematic alcohol use (high-risk drinker) among men was 8 (sensitivity = 95% and specificity = 93%) and 6 for women (sensitivity = 98% and specificity = 95%). In the Spanish validation of the instrument (Rubio et al., 1998), the general cut-off point for detecting problems with alcohol use was 8, but when analyzing sensitivity and specificity by sex, differences were encountered and the cut-off point for men was 9 and higher, with 6 or higher among women. The validation for Catalan and Spanish (Contel et al., 1999) also established a cut-off point of 9 for men but did not detect a different point for women. In an AUDIT validation study carried out exclusively with women in two primary care centers and in a drug dependence center (Perula et al., 2005), a cut-off point of 6 or higher was obtained.

Taking the results obtained into account, we recommend that the cut-off point for male university students be reduced by one point below the values found in two validation studies in the general population in primary care. This recommendation is supported by a systematic review of AUDIT (Meneses-Gaya et al., 2009), which affirms that its sensitivity and specificity are lower when applying the standard cut-off points to the university population.

Regarding the detection of drinkers with physical-psychological problems and probable alcohol dependence (ADS), the most efficient cut-off point in our study was 13 or higher for both sexes, identical to that recommended in primary care for the general population (Roson, 2008).

With reference to the criterion validity of AUDIT-C, it may appear contradictory that a test which detects 98% of cases is not useful in detecting high-risk consumption, but it must be pointed out that the low prevalence of consumption among university students generates as many false negatives as true positives, which means that there is only an even probability of detecting positive cases, as reflected in the kappa value.

Despite the reliability estimate for AUDIT-C in this population not being particularly good (Kappa index 0.50), this could be improved for the detection of high-risk consumption if cut-off points are raised by one point above those recommended for men (Bush et al., 1998) and women (Bradley et al., 2003). Raising the cut-off points in this way to 5 and 4 respectively significantly increases their predictive power.

The need to detect BD among young people is stressed in the latest report (2011) by the Spanish Observatory on Drugs and Drug Addiction, which describes a high prevalence of this intermittent consumption pattern. Moreover, according to the data provided in EDADES 2011/12, the figures for binge drinking in the last 30 days have risen slightly; 15.2% of those surveyed have consumed alcohol in this way, and this high-risk pattern of consumption is particularly extended among young adults, aged 20-29, of both sexes. The criterion validity or diagnostic validity of detecting binge drinkers with AUDIT is low, with moderate Kappa values obtained for both sexes. AUDIT-C, however, displays good sensitivity for detecting problems with alcohol use and binge drinking in both men and women (AUDIT: sensitivity of 0.58 for men and 0.64 for women vs. AUDIT-C: sensitivity of 0.81 for men and 0.85 for women)

A possible limitation of this study is the lack of a test-retest analysis, but this was not carried out on organizational grounds and to avoid doubts on the part of the students regarding loss of confidentiality.

In conclusion, its psychometric properties make AUDIT a suitable instrument for the timely detection of problems with alcohol use in the university population, although lowering the cut-off point to 8 for men and revising items 2 and 3 is recommended in order to improve its predictive power in detecting binge drinking. The predictive strength of AUDIT-C would be improved by raising the cut-off points by one for both sexes.

Acknowledgments

The study is part of a series of actions carried out by the University of Cadiz in cooperation with the Government of Andalusia in the field of drug dependence prevention. For this study, an instrument was needed with the ability to discriminate between populations and inform the design of preventive actions based on its results.

We would like to express our gratitude to the University of Cadiz for its support and the facilities offered in carrying out the field work of this study, and to the drug dependency service of the provincial government of Cadiz for its guidance and help with the referral protocol of cases detected.

Conflict of interests

The authors declare no conflict of interests.

References

Adewuya, A. O. (2005). Validation of the Alcohol Use Disorders Identification Test (AUDIT) as a screening tool for alcohol-related problems among Nigerian university students. Alcohol and Alcoholism, 40, 575-577.

Alvarado, M. E., Garmendia, M. L., Acuna, G., Santis, R. & Arteaga, O. (2009). Validez y confiabilidad de la version chilena del Alcohol Use Disorders Identification Test (AUDIT). Revista Medica Chile, 137, 1463-1468.

Baader, T., Rojas, C., Molina, J. L., Gotelli, M., Alamo, C., Fierro, C.,... Dittus, P. (2014). Diagnostico de la prevalencia de trastornos de la salud mental en estudiantes universitarios y los factores de riesgo emocionales asociados. Revista Chilena de Neuro-Psiquiatria, 52, 167-176.

Barry, K. & Fleming, M. (1993). A comparison of the Alcohol Use Disorders Identification Test (AUDIT) to the SMAST-13 in a primary care simple. Alcohol and Alcoholism, 2, 33-42.

Bradley, K. A., Bush, K. R., Epler, A. J., Dobie, D. J., Davis, T. M., Sporleder, J. L.,... Kivlahan, D. R. (2003). Two brief alcohol-screening tests from the Alcohol Use Disorders Identification Test (AUDIT): Validation in a female veterans affairs patient population. Archives of Internal Medicine, 163, 821-829.

Bush, K., Kivlahan, D. R., McDonell, M. S., Fihn, S. D. & Bradley, K. A. (1998). The AUDIT Alcohol Consumption Questions (AUDIT-C): An effective brief screening test for problem drinking. Archives of Internal Medicine, 158, 1789-1795. Consejeria para la Igualdad y Bienestar Social de la Junta de Andalucia (2009). La poblacion andaluza ante las drogas XI. Direccion General para las Drogodependencias y Adicciones de Andalucia. Fundacion andaluza para la atencion a las drogodependencias e incorporacion social.

Contel Guillamon, M., Gual Sole, A. & Colom Farran, J. (1999). Test para la identificacion de trastornos por uso de alcohol (AUDIT): traduccion y validacion del AUDIT al catalan y castellano. Adicciones, 11, 337-347.

Cortes Tomas, M. T., Espejo Tort, B., & Gimenez Costa, J. A. (2008). Aspectos cognitivos relacionados con la practica del botellon. Psicothema, 20, 396-402.

Dawson, D. A., Grant, B. F., Stinson, F. S. & Chou, P. S. (2004). Another look at heavy episodic drinking and alcohol use among college and non college youth. Journal of Studies on Alcohol, 65, 477-489.

Dawson, D. A., Grant, B. F. & Stinson, F. S. (2005). The AUDIT-C: screening for alcohol use disorders and risk drinking in the presence of other psychiatric disorders. Comprehensive Psychiatry, 46, 405-416.

Espana, Observatorio Espanol sobre Drogas. Delegacion del Gobierno para el Plan Nacional sobre Drogas. Ministerio de Sanidad y Consumo. Informe 2011. Retrieved from http://www.pnsd.msc.es/Categoria2/observa/pdf/oed2011.pdf

Gache, P., Michaud, P., Landry, U., Accietto, C., Arfaoui, S., Wenger, O. & Daeppen, J. B. (2005). The Alcohol Use Disorders Identification Test (AUDIT) as a Screening Tool for Excessive Drinking in Primary Care: Reliability and Validity of a French Version. Alcoholism: Clinical and Experimental Research, 29, 2001-2007.

Knight, J. R., Sherritt, L., Harris, S. K., Gates, E. C. & Chang, G. (2003). Validity of brief alcohol screening test among adolescents: A comparison of the AUDIT, POSIT, CAGE and CRAFFT. Alcoholism: Clinical and Experimental Research, 27, 67-73.

Kokotailo, P. K., Egan, J., Gangnon, R., Brown, D., Marlon, M. & Fleming, M. (2004). Validity of the Alcohol Use Disorders Identification Test in College Students. Alcoholism: Clinical and Experimental Research, 28, 914-920.

Kypri, K. & McAnally, H. M. (2005). Randomized controlled trial of a web-based primary care intervention for multiple health risk behaviors. Preventive Medicine, 41, 761-763.

Kypri, K., Paschall, M. J., Langley, L., Baxter, J., CashellSmith, M. & Bourdeau, B. (2009). Drinking and Alcohol-Related Harm Among New Zealand University Students: Findings From a National Web-Based Survey. Alcoholism: Clinical and Experimental Research 33, 307-314.

Lima, C. T., Freire, A. ?., Silva, A. P., Teixeira, R. M., Farrel, M. & Farrel, M. (2005). Concurrent and construct validity of the Audit in urban Brazilian simple. Alcohol and Alcoholism, 40, 584-589.

Lopez-Caneda, E., Mota, N., Greco, A., Velasquez, T., Corral, M., Rodriguez Holguin, S. & Cadaveira, F. (2014). Anomalias neurocognitivas asociadas al consumo intensivo de alcohol (binge drinking) en jovenes y adolescentes: Una revision. Adicciones, 26, 334-359.

Londono Perez, ?., Garcia Hernandez, W., Valencia Lara, S. ?. & Vinaccia Alpi, S. (2005). Expectativas frente al consumo de alcohol en jovenes universitarios colombianos. Anales de Psicologia, 21, 259-287.

Lopez-Caneda, E., Mota, N., Crego, A., Velasquez, T., Corral, M., Rodriguez Holguin, S. & Cadaveira, F. (2014). Anomalias neurocognitivas asociadas al consumo intensivo de alcohol (binge drinking) en jovenes y adolescentes: Una revision. Adicciones, 26, 334-359.

Lorenzo-Seva, U. & Ferrando, P. J. (2006). FACTOR: A computer program to fit the exploratory factor analysis model. Behavior Research Methods, 38, 88-91.

Martin-Montanez, E., Baron-Lopez, F. J., Rubio Lamia, L. O., Pavia Molina, J., Miranda Paez, J. & Santos Amaya, I. M. (2011). Consumo de alcohol, tabaco, cannabis y otras sustancias psicoactivas en estudiantes de la Universidad de Malaga. Trastornos Adictivos, 13, 160-166.

Martinez Delgado, J. M. (1996). Validacion de los cuestionarios breves: AUDIT, CAGE y CBA para la deteccion precoz del sindrome de dependencia de alcohol en Atencion Primaria. Unpublished Doctoral Thesis. Cadiz: University of Cadiz, Spain.

Meneses-Gaya, ?., Waldo Zuardi, A., Regina Loureiro, S. R. & Crippa, J. A. (2009). Alcohol Use Disorders Identification Test (AUDIT): An Updated systematic review of psychometric properties. Psychology and Neuroscience, 2, 83-97.

Montano, M. R., Morales Guarin, A., Gomez Millan, M. A., Vera Maldonado, A. & Gantiva Diaz, ?. A. (2011). Consumo de alcohol en estudiantes universitarios. Psicogente, 14, 27-35.

Observatorio Espanol de la Droga y las Toxicomanias (2011). Encuesta Domiciliaria sobre Alcohol y Drogas en Espana (EDADES) 2011/12. Retrieved from http://www.pnsd.msc.es/Categoria2/observa/pdf/EDADES2011.pdf

Observatorio Espanol de la Droga y las Toxicomanias (2011). Informe 2011. Delegacion del Gobierno para el Plan Nacional sobre Drogas. Retrieved from http://www. pnsd.msc.es/Categoria2/observa/pdf/oed2011.pdf

Observatorio Europeo de las Drogas y las Toxicomanias (2011). El problema de la drogodependencia en Europa. Informe anual 2011. Retrieved from http://www. emcdda.europa.eu/attachements.cfm/att_143743_ES_ EMCDDA_AR2011_ES.pdf

O'Malley, P. & Johnston, L. (2002). Epidemiology of alcohol and other drug use among American college students. Journal of Studies on Alcohol, 14, 23-39.

Perula de Torres, L. A., Fernandez-Garcia, J. A., Arias-Vega, R., Muriel-Palomino, M., Marquez-Rebollo, E. & Ruiz Moral, R. (2005). Validacion del cuestionario AUDIT para la identificacion del consumo de riesgo y de los trastornos por el uso de alcohol en mujeres. Atencion Primaria, 36, 499-506.

Reinert, D. F. & Allen, J. P. (2002). The Alcohol Use Disorders Identification Test (AUDIT): a review of recent research. Alcoholism: Clinical and Experimental Research, 26, 272-279.

Reinert, D. F. & Allen, J. P. (2007). The Alcohol Use Disorders Identification Test: an update of research findings. Alcoholism: Clinical and Experimental Research, 31, 185-199.

Rodriguez-Martos Dauer, A. (2007a). Primera conferencia de Prevencion y Promocion de la salud en la practica clinica en Espana. Prevencion de los problemas derivados del consumo de alcohol. Ministerio de Sanidad y Consumo.

Rodriguez-Martos Dauer, A. (2007b). ?Por que es tan dificil legislar sobre alcohol en Espana? Adicciones, 19, 325-332.

Roson Hernandez, B. (2008). Consumo de riesgo y perjudicial de alcohol. Prevalencia y metodos de deteccion en la practica clinica. Galicia Clinica, 69, 29-44.

Rubio Valladolid, G., Bermejo Vicedo, J., Caballero Sanchez-Serrano, ?. ?. & Santo-Domingo Carrasco, J. (1998). Validacion de la Prueba para la Identificacion de Trastornos por Uso de Alcohol (AUDIT) en Atencion Primaria. Revista Clinica Espanola, 198, 11-14.

Salazar Garza, ?. L., Avila Hernandez, O. P., Perez Castro, J. L. & Martinez Martinez, K. I. (2010). Deteccion temprana de jovenes universitarios en riesgo por su consumo de alcohol. Investigacion y Ciencia de la Universidad Autonoma de Aguascalientes Mexico, 59, 40-46.

Saunders, J., Aasland, O., Babor, T., de la Fuente, J. & Grant, ?. (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.

Schmidt, A., Barry, K. & Fleming, ?. (1995). A new screening test for the detection of problem drinkers: The Alcohol Use Disorders Identification Test (AUDIT). Southern Medical Journal, 88, 52-59.

Seguel Palma, F., Santander Manriquez, G. & Alexandre Barriga, O. (2013). Validez y confiabilidad del test de+ identificacion de los trastornos debidos al consumo de alcohol (AUDIT) en estudiantes de una universidad chilena. Ciencia y Enfermeria XIX, 1, 23-35.

Velicer, W. F. (1976). Determining the number of components from the matrix of partial correlations. Psychometrika, 41, 321-337.

Vinader-Caerols, C., Monleon, S. & Parra, A. (2014). Physiological and psychological effects of a high dose of alcohol in Young men and women. Adicciones, 26, 238-246.

Wicki, M., Kuntsche, E. & Gmel, G. (2010). Drinking at European universities? A review of students' alcohol use. Addictive Behaviors, 35, 913-924.

Zaldivar Basurto, F., Lopez Rios, F., Garcia Montes, J. M. & Molina Moreno, A. (2011). Consumo autoinformado de alcohol y otras drogas en poblacion universitaria espanola. Electronic Journal of Research in Educational Psychology, 9, 113-132.

MIGUEL ANGEL GARCIA CARRETERO (*), JOSE PEDRO NOVALBOS Ruiz (**), Jose MANUEL MARTINEZ DELGADO (***), CRISTINA O'FERRALL GONZALEZ (****).

(*) Grupo PAIDI CTS-391 Grupo Multidisciplinario para el progreso de la Salud Mental, Universidad de Cadiz; (**) Facultad de Medicina, Universidad de Cadiz; (***) Centro de Tratamiento de Adicciones. Diputacion Provincial de Cadiz; (****) Facultad de Enfermeria y Fisioterapia, Universidad de Cadiz.

Received: December 2014; Accepted: July 2015.

Send correspondence to:

CTS/391 Grupo Multidisciplinar para el progreso de la Salud Mental (Responsable: Profa O'Ferrall). Universidad de Cadiz. Facultad de Enfermeria. Miguel Angel Garcia Carretero. Avda. Ana de Viya n[degrees] 52. 11009 Cadiz. E-mail: miguel.garcia.carretero@hotmail.com
Table 1. Alcohol consumption patterns among university students (total
population and drinkers), by sex.

Weekly consumption
of alcohol and
consumption patterns

Total          Consumption         % (n)
population     last week
(N=1309)
               SDUs                Monday to Wednesday
               last week
               Mean (SD)           Thursday
                                   Friday to Sunday
                                   Total weekly
               Weekly consumption  Low-risk consumption (a)
               pattern
               (%)                 Moderate-risk consumption (b)
                                   High-risk consumption (c)
               Binge drinking      [greater than or equal to] 5
                                   SDUs per sitting
Drinkers       Consumption         % (n)
(n=891)        last week
               SDUs                Monday to Wednesday
               last week
               Mean (SD)           Thursday
                                   Friday to Sunday
                                   Total weekly
               Weekly consumption  Low-risk consumption (a)
               pattern
               (%)                 Moderate-risk consumption (b)
                                   High-risk consumption (c)
               Binge drinking      [greater than or equal to] 5
                                   SDUs per sitting

Weekly consumption
of alcohol and
consumption patterns  Men            Women              Total

Total                 73,1 (535)     64,6 (774)         68,1 (1309)
population
(N=1309)
                       1,60 (3,7)     0,52 (1,9) (***)   1,01 (2,9)
                       1,25 (3,0)     0,79 (2,1)         0,98 (2,5)
                       8,14 (9,3)     5,30 (6,3) (**)    6,29 (8,0)
                      10,99 (12,6)    6,60 (8,1) (***)   8,39 (10,4)
                      77,2           77,4               77,3
                      15,0           16,9               16,1
                       7,9            5,7                6,6
                      38,3           32,7 (***)         35,0
Drinkers              73,1 (391)     64,6 (500)         68,1 (891)
(n=891)
                       2,18 (4,1)     0,80 (2,3) (***)   1,41 (3,3)
                       1,71 (3,4)     1,22 (2,6)         1,44 (2,9)
                      11,13 (9,2)     8,20 (6,2) (*)     9,49 (7,8)
                      15,03 (12,5)   10,22 (8,0) (***)  12,33 (10,5)
                      68,8           65,0               66,7
                      20,5           26,2               23,7
                      10,7            8,8                9,7
                      52,4           50,6 (***)         51,4

Note. SDU (Standard Drink Unit); SD (Standard Deviation). (a) [less
than or equal to]21 SDU men and [less than or equal to]14 SDU women.
(b) 22-27 SDU men and 15-16 SDU women.
(c) [greater than or equal to]28 SDU men and [greater than or equal
to]17 SDU women. (*) p = .018; (**) p = .002; (***) p = .000.

Table 2. Problems with alcohol consumption by sex according to
AUDIT (total population and drinkers).

AUDIT                     Men          Women          Total
                          N (%)        N (%)          N (%)

Total       Low-risk      390 (72,9)   572 (73,9)     962 (73,5)
population  drinkera
(N=1309)
            High-risk     101 (18,9)   162 (20,9)     263 (20,1)
            drinkerb
            Drinker with   44 (8,2)     40 (5,2)       84 (6,4)
            probable ADSc
            Total         535 (100,0)  774 (100,0)   1309
                                                     (100,0)
Drinkers    Low-risk      249 (63,7)   304 (60,8)     553 (62,1)
(n=891)     drinkera
            High-risk      98 (25,1)   156 (31,2)     254 (28,5)
            drinkerb
            Drinker with   44 (11,3)    40 (8,0)       84 (9,4)
            probable ADSc
            Total         391 (100,0)  500 (100,0)    891 (100,0)

Note. ADS (Alcohol Dependence Syndrome). (a) [less than or equal to] 7
men and [less than or equal to] 5 women.
(b) 8-12 men and 6-12 women. (c) [greater than or equal to]13
men and women.

Table 3. Correlation of AUDIT item scores with scale total.

AUDIT   Media  SD     Inter-element       Inter-element
                      correlation matrix  co-variance matrix

Item 1  1,68   1,001  0,706               2,960
Item 2  0,64   0,856  0,662               2,372
Item 3  0,66   0,938  0,811               3,187
Item 4  0,29   0,719  0,610               1,836
Item 5  0,18   0,448  0,602               1,130
Item 6  0,12   0,410  0,503               0,864
Item 7  0,30   0,592  0,593               1,471
Item 8  0,37   0,638  0,694               1,855
Item 9  0,10   0,549  0,414               0,951
Item 10 0,11   0,593  0,369               0,918
TOTAL   4,45   4,188  1,000              17,542

AUDIT    Cronbach's alpha
         if item deleted

Item 1   0,709
Item 2   0,718
Item 3   0,700
Item 4   0,726
Item 5   0,736
Item 6   0,741
Item 7   0,731
Item 8   0,723
Item 9   0,742
Item 10  0,744
TOTAL    0,800

Table 4. Reliability of AUDIT compared to Weekly Alcohol
Consumption Diary.

                                              Consumption Diary
                                              Low-risk
                                              consumption (a)

AUDIT    Low-risk drinkerd            n (%)    952 (54,1)
         High-risk drinkere           n (%)     55 (5,4)
         Drinker with probable ADSf   n (%)      5 (0,5)
         Total                        n (%)   1012 (100)

                                     Consumption Diary
                                     Moderate-risk      High-risk
                                     consumption (b)    consumption (c)

AUDIT    Low-risk drinkerd             6 (2,8)           4 (4,7)
         High-risk drinkere          204 (96,7)          4 (4,7)
         Drinker with probable ADSf    1 (0,5)          78 (90,7)
         Total                       211 (100)          86 (100)

                                       Consumption Diary
                                       Total

AUDIT    Low-risk drinkerd             962 (73,5)
         High-risk drinkere            263 (20,1)
         Drinker with probable ADSf     84 (6,4)
         Total                        1309 (100)

(a) [less than or equal to] 21 SDU men and [less than or equal to]
14 SDU women. (b) 22-27 SDU men and 15-16 SDU women. (c) [greater
than or equal to] 28 SDU men and [greater than or equal to] 17 SDU
women. (d) [less than or equal to] 7 men and [less than or equal to]
5 women. (e) 8-12 men and 6-12
women. f [greater than or equal to]13 men and women. * Spearman
correlation (0.87), Kappa (0.85).

Table 5. Reliability of AUDIT compared to Weekly Alcohol
Consumption Diar.

                                          Consumption Diary
                                         Low-risk consumption (a)

AUDIT-C  Low-risk consumptionc   n (%)    705 (69,7)
         High-risk consumptiond  n (%)    307 (30,3)
         Total                   n (%)   1012 (100)

                                             Consumption Diary
                                 Moderate-risk consumption (b)

AUDIT-C  Low-risk consumptionc     4 (1,3)
         High-risk consumptiond  293 (98,7)
         Total                   297 (100)

                                   Consumption Diary
                                   Total

AUDIT-C  Low-risk consumptionc      709 (54,2)
         High-risk consumptiond     600 (45,8)
         Total                     1309 (100)

(a) [less than or equal to] 21 men and [less than or equal to]
14 women. (b) [greater than or equal to] 22 men and [greater than
or equal to] 15 women. (c) [less than or equal to] 3 men and
[less than or equal to] 2 women. (d) [greater than or equal to]
4 men and [greater than or equal to] 3 women. *Spearman correlation
(0.57), Kappa (0.50)

Table 6. Exploratory factor analysis of AUDIT.

AUDIT          Two factors                Three factors
           Factor 1   Factor 2   Factor 1   Factor 2   Factor 3

Item 1                0,957                 0,647
Item 2                0,742                 0,615
Item 3                1,028                 1,081
Item 4                0,560      0,618
Item 5                0,718      1,012
Item 6                0,591      0,717
Item 7                0,554      0,498
Item 8                0,761                 0,584
Item 9     0,832                                       0,937
Item 10    0,957                                       0,693
Estimated  0,825      0,923      0,870      0,925      0,833
reliability

Table 7. Indexes of validity and utility for the AUDIT and AUDIT-C
cut-off points.

                           AUDIT
     High-risk  High-risk   Drinker with  Drinker with
     drinker    drinker     probable ADS  probable ADS
     C.P.: 8    C.P.: 6     C.P.: 13      C.P.: 13
     Men        Women       Men           Women

S     0,95       0,98        0,95           0,86
SP    0,93       0,95        0,99           0,99
PPV   0,8        0,85        0,91           0,95
NPV   0,98       0,99        0,99           0,99
PWC   0,93       0,95        0,99           0,99
PBC   0,06       0,04        0,01           0,01
LR+  13,54      18,88      117,38         315,23
LR-   0,05       0,02        0,05           0,14
K     0,83       0,88        0,92           0,90
SE    0,93       0,95        0,99           0,99
YJ    0,88       0,92        0,94           0,86

            AUDIT-C
     High-risk    High-risk
     consumption  consumption
     C.P.: 5      C.P.: 4
     Men          Women

S    0,91         0,86
SP   0,84         0,88
PPV  0,63         0,69
NPV  0,97         0,96
PWC  0,86         0,88
PBC  0,14         0,12
LR+  5,69         7,49
LR-  0,11         0,15
K    0,65         0,68
SE   0,86         0,88
YJ   0,75         0,75

Note. ADS (Alcohol Dependence Syndrome); C.P. (Cut-off point);
S (Sensitivity); Sp (Specificity); PPV (Positive Predictive Value);
NPV (Negative Predictive Value); PWC
(Percentage Well Classified); PBC (Percentage Badly Classified);
LR+ (Positive Likelihood Ratio); LR- (Negative Likelihood Ratio);
K (Kappa Coefficient); SE (Standard
Efficiency); YJ (Youden's J).
COPYRIGHT 2016 Socidrogalcohol
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:original
Author:Carretero, Miguel Angel Garcia; Ruiz, Jose Pedro Novalbos; Delgado, Jose Manuel Martinez; Gonzalez,
Publication:Adicciones
Date:Dec 1, 2016
Words:8178
Previous Article:Why is it (also) so difficult to legislate gambling in Spain? 'Deja vu' of what occurred with alcohol.
Next Article:Impulsivity in men with prescription of benzodiazepines and methadone in prison.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters