bs_bs_banner
R E V I E W
Drug and Alcohol Review (March 2014), 33, 144–151 DOI: 10.1111/dar.12109
Which alcohol control strategies do young people think are effective? RICHARD O. DE VISSER1, ANGIE HART2, CHARLES ABRAHAM3, ANJUM MEMON4, REBECCA GRABER1 & TOM SCANLON5 1
School of Psychology, University of Sussex, Falmer, UK, 2School of Nursing and Midwifery, University of Brighton, Falmer, UK, 3University of Exeter Medical School, Exeter, UK, 4Brighton & Sussex Medical School, Falmer, UK, 5NHS Brighton & Hove, Prestamex House, Brighton, UK
Abstract Introduction and Aims. The aims of this study were to examine young people’s belief in the effectiveness of various alcohol control strategies and to identify demographic, attitudinal and behavioural correlates of perceived effectiveness. Design and Methods. An online questionnaire hosted on a secure server was completed by 1418 men and women aged 16–21 years living in South-East England. It assessed the perceived effectiveness of various alcohol control strategies. Key correlates included sensation seeking, impulsivity, conscientiousness, alcohol outcome expectancies, drink refusal self-efficacy, perceived peer alcohol use and Alcohol Use Disorders Identification Test scores. Results. The most effective strategies were perceived to be enforcing responsible service legislation, strictly monitoring late-night licensed premises and teaching alcohol refusal skills. Greater belief in the effectiveness of alcohol control strategies was expressed by older participants, those who consumed less alcohol and those who expected more negative outcomes from alcohol consumption. Discussion and Conclusions. The data suggest that in order to increase the perceived effectiveness of alcohol control strategies, we may need to address young people’s beliefs about the negative outcomes of alcohol use. Strategies that young people believe are effective may be easier to implement, but this does not imply that unpopular but effective strategies should not be tried. [de Visser RO, Hart A, Abraham C, Memon A, Graber R, Scanlon T. Which alcohol control strategies do young people think are effective?Drug Alcohol Rev 2014;33:144–151] Key words: alcohol, policy, attitude, youth.
Introduction There is widespread concern about health and social consequences of excessive alcohol consumption among young people. Drinking is common among young people (defined here as 16- to 21-year-olds), many of whom engage in heavy episodic drinking (often called ‘binge drinking’) [1], which is associated with an increased risk of accidents, injuries and violence [2,3]. Alcohol policies and strategies (i.e. measures that affect the market in alcohol, alcohol consumption, or alcohol-related problems) are an important aspect of efforts to reduce harm from heavy episodic drinking [4]. In some countries, alcohol control policy consists of outright prohibition. In the UK, like most other ‘western’ countries, alcohol control strategies seek to
minimise harm while upholding adults’ rights to determine their own alcohol consumption [5–7]. A systematic review of alcohol control policies revealed that several strategies appear to be effective: regulating the marketing, availability and service of alcohol; regulating advertising; enforcing minimum purchase age; and raising prices [8]. There is some evidence that information and education campaigns directed at all members of the general public can be effective [8]. Focusing more specifically on young people, there is currently a lack of evidence for the efficacy of ‘whole-school’ interventions, but focused, one-to-one interventions with risky students appear to be effective [8]. Research in various countries has revealed tepid public support for strategies restricting or controlling
Richard de Visser PhD, Senior Lecturer, Angie Hart PhD, Professor of Child, Family & Community Health, Charles Abraham PhD, Professor of Behaviour Change, Anjum Memon PhD, Senior Lecturer, Rebecca Graber PhD, Research Fellow, Tom Scanlon MRCGP, Director of Public Health, Brighton & Hove. Correspondence to Dr Richard de Visser, School of Psychology, University of Sussex, Falmer BN1 9QH, UK. Tel: +44 (0)1273 876 585; Fax: +44(0)1273 678 058; E-mail: [email protected] Received 30 August 2013; accepted for publication 4 December 2013. © 2014 Australasian Professional Society on Alcohol and other Drugs
Alcohol policies
alcohol availability and/or increasing drink prices for all drinkers [9–12]. Policies directed toward curbing excessive consumption and irresponsible service have greater public support than do policies that affect all drinkers [12–14]. It is perhaps not surprising that support for strategies that restrict availability tends to be lower among people who drink more [10,13–18]. Greater support for alcohol control strategies is expressed by women and older people [9,10,16–18]. Few studies have examined in detail young people’s perceptions of alcohol control strategies [15,19]. One such study compared Dutch 16- to 22-year-olds to older adults: adolescents and young adults expressed significantly lower levels of support for all restrictive policies and all educational strategies than did older adults [15]. Furthermore, those aged 16–18 expressed significantly less support for educational strategies than did 19- to 22-year-olds. Echoing the results of studies of older adults, it was found that among young people, greater support for alcohol policies was expressed by older respondents and those who consumed less alcohol. The vast majority of studies referred to above focused on support for policies, but it is also important to examine people’s beliefs about the effectiveness of different alcohol control strategies. Governments are often concerned about the perceived legitimacy and acceptability of health policies, and strategies that young people believe to be effective may be easier to implement [7,12]. However, it is also important to note that in addition to being affected by health concerns, young people’s alcohol use may be affected by concerns about sociality and pleasure [20]. In addition to a lack of information about beliefs about different alcohol control strategies, there is a lack of information about how perceived effectiveness of strategies is related to other attitudes and beliefs, or aspects of personality. The aims of this study were to examine opinions about various alcohol control strategies among British young people and to identify demographic, attitudinal and behavioural correlates of such support. Eight strategies were chosen on the basis of variations in evidence of their effectiveness [8] and variations in their focus on individuals or the environment. Particular attention was given to two approaches. The first was educational interventions among young people, given that younger people are less supportive of educational interventions [15] and because education may not lead to behaviour change if individuals do not also possess appropriate motivation and the requisite behavioural skills [21]. The second was alcohol pricing, given ongoing political and media debates in the UK about the introduction of minimum unit pricing which would rule out cheap sales of alcohol that are appealing to young people with limited disposable incomes [5,17].
145
Methods The study received ethical approval from the host university. Data were collected between February and June 2012. Links to the online survey were distributed through: emails to students in two universities; advertisements on the websites of 17 secondary schools; and promotional cards distributed at four young people’s services. All participants were aged over 16 and gave individual consent to take part. The study was presented to participants as ‘a study of drinking and social life’: the focus on alcohol control strategies was not apparent in recruitment materials, so it is unlikely that participants could have self-selected on the basis of their beliefs about alcohol control strategies. Sample Online questionnaires were completed by 642 men and 776 women aged 16–21 years (mean = 18.1, standard deviation = 1.4) living in South-East England. This age range was chosen to include young people with a range of experiences with alcohol: in the UK, the legal age for purchase of alcohol is 18, but younger people can drink under adult supervision [22]. Measures Respondents used 7-point scales (anchors: ‘not at all effective’, ‘extremely effective’) to indicate their belief in eight alcohol control strategies. The introductory statement ‘Please indicate how well each approach would address alcohol-related problems’ was followed by eight strategies chosen on the basis of variations in evidence of their effectiveness [8] and variations in their focus on individuals or the environment (see Table 1). Questions focused on perceived strategy effectiveness rather than on beliefs about how strategies would change respondents’ own drinking [17]. Principal components factor analysis identified one factor (Eigenvalue = 4.23; 52.81% of total variance explained), and a scale computed as the mean of the eight items had good internal consistency (α = 0.87). Sensation seeking was assessed using 10 items such as ‘I like doing things just for the thrill of it’ (α = 0.90) [23]. Three other aspects of personality were assessed. The root phrase ‘Typically I . . .’ was followed by 10 items in each of three domains: impulsivity (α = 0.80; e.g. ‘. . . easily resist temptations’), extraversion (α = 0.82; ‘. . . am the life of the party’) and conscientiousness (α = 0.90; ‘. . . am always prepared’) [24]. For all scales above, respondents used 7-point scales (anchors: ‘strongly disagree’, ‘strongly agree’). Two scales assessed alcohol outcome expectancies [25], with responses made using 7-point scales © 2014 Australasian Professional Society on Alcohol and other Drugs
146
R. de Visser et al.
Table 1. Perceived efficacy of various alcohol control strategies
Enforce the law against serving drunk people Strictly monitor late-night licensed premises Teach people skills for resisting pressure to drink Restrict late-night alcohol sales Increase the price of alcohol Reduce the number of outlets that sell alcohol Ban alcohol sponsorship of sporting events Raise the legal drinking age Mean
Men (n = 642)
Women (n = 776)
3.96 (2.04)a 3.70 (1.92)b 3.76 (2.05)b 3.47 (1.99)c 3.33 (2.13)c 3.11 (1.97)d 2.85 (2.03)e 2.20 (1.81)f 3.30 (1.46)
4.41 (1.93)a 4.05 (1.81)b 3.94 (1.94)b 3.76 (1.99)c 3.36 (1.98)d 3.26 (1.94)d 3.15 (1.97)d 2.10 (1.66)e 3.50 (1.35)
F(1,1416) = 18.57, P < 0.01 F(1,1416) = 12.88, P < 0.01 F(1,1416) = 2.72, P = 0.10 F(1,1416) = 7.37, P < 0.01 F(1,1416) = 0.09, P = 0.77 F(1,1416) = 2.04, P = 0.15 F(1,1416) = 7.96, P < 0.01 F(1,1416) = 1.14, P = 0.29 F(1,1416) = 2.75, P < 0.01
Within-columns values with different superscripts are significantly different.
(anchors: ‘strongly disagree’, ‘strongly agree’). Ten items assessed the likelihood of positive outcomes when people drink (e.g. ‘They enjoy the buzz’, α = 0.67). Ten items assessed the likelihood of negative outcomes (e.g. ‘They become aggressive’, α = 0.88). Drink refusal self-efficacy (DRSE)—an individual’s perception of his/her capacity to refuse alcohol—was assessed using 7-point scales (anchors: ‘very difficult’, ‘very easy’) [26]. The three subscales—social pressure (e.g. ‘When someone offers me a drink’), emotional relief (e.g. ‘When I am worried’) and opportunistic drinking (e.g. ‘When I am watching TV’)—were correlated at r > 0.65, so a single 19-item scale was used (α = 0.94). Peer alcohol use was assessed as the mean of four novel items (α = 0.86). Participants used 7-point scales (anchors: ‘none of them’, ‘all of them’) to indicate the proportion of (friends/other people their age) who regularly (drink alcohol/get drunk). The 10-item Alcohol Use Disorders Identification Test (AUDIT) assesses consumption frequency and volume, dependence and alcohol-related problems [27]. Scores were summed: scores above 8 for men and 7 for women are commonly considered to indicate hazardous alcohol use. Similar proportions of male and female respondents were above these thresholds (58% vs. 56%, χ2(1) = 0.15, P = 0.70). Participants used 7-point scales (anchors: ‘strongly disagree’, ‘strongly disagree’) to respond to ‘concerns about my health exert a strong influence over my use of alcohol’. Analysis Pearson correlations were calculated to identify bivariate correlates of perceived strategy efficacy to include in linear regression. The skewness and kurtosis of the outcome variables indicated no major violation of assumptions of normality: ‘increase prices’: skew© 2014 Australasian Professional Society on Alcohol and other Drugs
ness = 0.32, kurtosis = −1.23; ‘teach alcohol refusal skills’: skewness = 0.04, kurtosis = −1.21; strategy scale: skewness = 0.21, kurtosis = −0.32 [28]. Results Table 2 displays the demographic profile of the sample and provides comparisons to relevant population data. The sample over-represented students—66% were secondary school students, 22% were in further education, 11% were employed and 1% were unemployed—but this may reflect the sampling strategies [32].The prevalence of excessive alcohol use in the sample was comparable with that reported in population-representative surveys and in other surveys of young people in the UK [33]. With the exception of religion, the demographic profiles of male and female respondents were not significantly different. Table 2 shows men’s and women’s belief in the efficacy of various alcohol strategies. Only two strategies received support scores above the mid-point of the 7-point scale: enforcing the law against serving drunk people, and strictly monitoring late-night licensed premises. Belief in the efficacy of teaching people skills for resisting pressure to drink was not significantly lower than that for monitoring late-night premises. Raising the legal drinking age was perceived to be the least effective strategy. Women expressed significantly greater belief in the efficacy of four strategies: enforcing the law against serving drunk people, strictly monitoring late-night licensed premises, restricting late-night alcohol sales and banning alcohol sponsorship of sporting events. The mean rating of perceived strategy effectiveness was significantly higher for women. Within-subjects repeated measures anova identified differences in the perceived efficacy of the eight strategies. Of the 28 pair-wise comparisons of support, 26 were significant among men [F(1,641) = 40.74, P < 0.01] and 25 were significant among women [F(1,771) = 98.16,
Alcohol policies
147
Table 2. Comparisons of sample with relevant section of general population and sex differences with sample Sample
Population (%) Ethnicitya White n Mixed Black Other Religionb None Christian Muslim Other Employmentc Working and/or studying Not in employment, education or training Parental education Both university educated One university educated Neither university educated Alcohol consumptiond Maximum daily intake > double daily guideline Maximum daily intake > weekly guideline
Men (n = 642) (%)
Women (n = 776) (%)
Within-sample sex difference χ2(4) = 4.14, P = 0.39
90.7 5.2 1.9 1.6 0.6
89.3 4.8 0.5 2.6 0.4
89.9 3.9 0.4 1.7 0.4
35.5 47.6 6.2 10.7
70.6 22.9 3.1 3.4
64.8 29.8 2.7 2.7
90.3 9.7
89.6 0.3
89.6 0.6
23.4 19.3 57.3
21.1 23.3 55.5
40 27
36 26
− − − Men 45 19
Women 40 21
χ2(3) = 8.75, P = 0.03
χ2(1) = 0.79, P = 0.37 χ2(2) = 3.61, P = 0.16
χ2(1) = 2.60, P = 0.11 χ2(1) = 0.65, P = 0.42
a Comparison population is whole population in South-East England [29]. bComparison population is 16- to 24-year-olds in UK [29]. cComparison population is 16- to 24-year-olds in UK [30]. dComparison population is 16- to 24-year-olds in England [31].
Table 3. Correlates of perceived efficacy of different alcohol control strategies (n = 1418)
Variable
Eight strategy scale
Teach alcohol refusal skills
Increase prices
Age Sensation seeking Impulsivity Extraversion Conscientiousness AOE: positive AOE: negative DRSE Peer alcohol use AUDIT Health concerns affect drinking
r = 0.11, P < 0.01 r = −0.17, P < 0.01 r = −0.10, P < 0.01 r = −0.10, P < 0.01 r = 0.20, P < 0.01 r = 0.03, P = 0.33 r = 0.20, P < 0.01 r = 0.16, P < 0.01 r = −0.07, P < 0.01. r = −0.24, P < 0.01 r = 0.11, P < 0.01
r = 0.07, P < 0.01 r = −0.08, P < 0.01 r = −0.07, P < 0.01 r = −0.01, P = 0.67 r = 0.14, P < 0.01 r = 0.04, P = 0.14 r = 0.16, P < 0.01 r = 0.11, P < 0.01 r = −0.01, P = 0.68 r = −0.12, P < 0.01 r = 0.07, P = 0.01
r = 0.04, P = 0.18 r = −0.14, P < 0.01 r = −0.10, P < 0.01 r = −0.05, P = 0.05 r = 0.14, P < 0.01 r = 0.01, P = 0.71 r = 0.13, P < 0.01 r = 0.16, P < 0.01 r = −0.07, P = 0.01 r = −0.22, P < 0.01 r = 0.09, P < 0.01
AOE, alcohol outcome expectancies; AUDIT, alcohol use disorders identification test; DRSE, drink refusal self efficacy.
P < 0.01]. Thus, although the eight strategies could be combined in a scale, each received discrete ratings of efficacy. Table 3 shows that greater perceived efficacy of alcohol control strategies was expressed by older respondents and respondents who were lower on sensation seeking, impulsivity and extraversion, but higher on conscientiousness. Greater efficacy was also per-
ceived by respondents with more negative outcome expectancies, greater DRSE, perception of less peer drinking/drunkenness, lower AUDIT scores and greater health concerns. Linear regression identified six significant multivariate correlates of greater belief in the effectiveness of the eight strategies [F(6,1410) = 38.72, P < 0.01; adjusted R2 = 0.138]. Table 4 shows that greater belief in the © 2014 Australasian Professional Society on Alcohol and other Drugs
148
R. de Visser et al.
Table 4. Multivariate correlates of perceived efficacy of strategies (n = 1418)
AOE: negative AUDIT Age Conscientiousness Health concerns affect drinking Peer alcohol use
B
Standard error
Beta
T
P
0.35 −0.04 0.14 0.14 0.05 −0.08
0.04 0.01 0.03 0.04 0.02 0.03
0.22 −0.21 0.14 0.11 0.07 −0.07
8.40 −7.51 5.35 4.10 2.65 −2.42
R E V I E W
Drug and Alcohol Review (March 2014), 33, 144–151 DOI: 10.1111/dar.12109
Which alcohol control strategies do young people think are effective? RICHARD O. DE VISSER1, ANGIE HART2, CHARLES ABRAHAM3, ANJUM MEMON4, REBECCA GRABER1 & TOM SCANLON5 1
School of Psychology, University of Sussex, Falmer, UK, 2School of Nursing and Midwifery, University of Brighton, Falmer, UK, 3University of Exeter Medical School, Exeter, UK, 4Brighton & Sussex Medical School, Falmer, UK, 5NHS Brighton & Hove, Prestamex House, Brighton, UK
Abstract Introduction and Aims. The aims of this study were to examine young people’s belief in the effectiveness of various alcohol control strategies and to identify demographic, attitudinal and behavioural correlates of perceived effectiveness. Design and Methods. An online questionnaire hosted on a secure server was completed by 1418 men and women aged 16–21 years living in South-East England. It assessed the perceived effectiveness of various alcohol control strategies. Key correlates included sensation seeking, impulsivity, conscientiousness, alcohol outcome expectancies, drink refusal self-efficacy, perceived peer alcohol use and Alcohol Use Disorders Identification Test scores. Results. The most effective strategies were perceived to be enforcing responsible service legislation, strictly monitoring late-night licensed premises and teaching alcohol refusal skills. Greater belief in the effectiveness of alcohol control strategies was expressed by older participants, those who consumed less alcohol and those who expected more negative outcomes from alcohol consumption. Discussion and Conclusions. The data suggest that in order to increase the perceived effectiveness of alcohol control strategies, we may need to address young people’s beliefs about the negative outcomes of alcohol use. Strategies that young people believe are effective may be easier to implement, but this does not imply that unpopular but effective strategies should not be tried. [de Visser RO, Hart A, Abraham C, Memon A, Graber R, Scanlon T. Which alcohol control strategies do young people think are effective?Drug Alcohol Rev 2014;33:144–151] Key words: alcohol, policy, attitude, youth.
Introduction There is widespread concern about health and social consequences of excessive alcohol consumption among young people. Drinking is common among young people (defined here as 16- to 21-year-olds), many of whom engage in heavy episodic drinking (often called ‘binge drinking’) [1], which is associated with an increased risk of accidents, injuries and violence [2,3]. Alcohol policies and strategies (i.e. measures that affect the market in alcohol, alcohol consumption, or alcohol-related problems) are an important aspect of efforts to reduce harm from heavy episodic drinking [4]. In some countries, alcohol control policy consists of outright prohibition. In the UK, like most other ‘western’ countries, alcohol control strategies seek to
minimise harm while upholding adults’ rights to determine their own alcohol consumption [5–7]. A systematic review of alcohol control policies revealed that several strategies appear to be effective: regulating the marketing, availability and service of alcohol; regulating advertising; enforcing minimum purchase age; and raising prices [8]. There is some evidence that information and education campaigns directed at all members of the general public can be effective [8]. Focusing more specifically on young people, there is currently a lack of evidence for the efficacy of ‘whole-school’ interventions, but focused, one-to-one interventions with risky students appear to be effective [8]. Research in various countries has revealed tepid public support for strategies restricting or controlling
Richard de Visser PhD, Senior Lecturer, Angie Hart PhD, Professor of Child, Family & Community Health, Charles Abraham PhD, Professor of Behaviour Change, Anjum Memon PhD, Senior Lecturer, Rebecca Graber PhD, Research Fellow, Tom Scanlon MRCGP, Director of Public Health, Brighton & Hove. Correspondence to Dr Richard de Visser, School of Psychology, University of Sussex, Falmer BN1 9QH, UK. Tel: +44 (0)1273 876 585; Fax: +44(0)1273 678 058; E-mail: [email protected] Received 30 August 2013; accepted for publication 4 December 2013. © 2014 Australasian Professional Society on Alcohol and other Drugs
Alcohol policies
alcohol availability and/or increasing drink prices for all drinkers [9–12]. Policies directed toward curbing excessive consumption and irresponsible service have greater public support than do policies that affect all drinkers [12–14]. It is perhaps not surprising that support for strategies that restrict availability tends to be lower among people who drink more [10,13–18]. Greater support for alcohol control strategies is expressed by women and older people [9,10,16–18]. Few studies have examined in detail young people’s perceptions of alcohol control strategies [15,19]. One such study compared Dutch 16- to 22-year-olds to older adults: adolescents and young adults expressed significantly lower levels of support for all restrictive policies and all educational strategies than did older adults [15]. Furthermore, those aged 16–18 expressed significantly less support for educational strategies than did 19- to 22-year-olds. Echoing the results of studies of older adults, it was found that among young people, greater support for alcohol policies was expressed by older respondents and those who consumed less alcohol. The vast majority of studies referred to above focused on support for policies, but it is also important to examine people’s beliefs about the effectiveness of different alcohol control strategies. Governments are often concerned about the perceived legitimacy and acceptability of health policies, and strategies that young people believe to be effective may be easier to implement [7,12]. However, it is also important to note that in addition to being affected by health concerns, young people’s alcohol use may be affected by concerns about sociality and pleasure [20]. In addition to a lack of information about beliefs about different alcohol control strategies, there is a lack of information about how perceived effectiveness of strategies is related to other attitudes and beliefs, or aspects of personality. The aims of this study were to examine opinions about various alcohol control strategies among British young people and to identify demographic, attitudinal and behavioural correlates of such support. Eight strategies were chosen on the basis of variations in evidence of their effectiveness [8] and variations in their focus on individuals or the environment. Particular attention was given to two approaches. The first was educational interventions among young people, given that younger people are less supportive of educational interventions [15] and because education may not lead to behaviour change if individuals do not also possess appropriate motivation and the requisite behavioural skills [21]. The second was alcohol pricing, given ongoing political and media debates in the UK about the introduction of minimum unit pricing which would rule out cheap sales of alcohol that are appealing to young people with limited disposable incomes [5,17].
145
Methods The study received ethical approval from the host university. Data were collected between February and June 2012. Links to the online survey were distributed through: emails to students in two universities; advertisements on the websites of 17 secondary schools; and promotional cards distributed at four young people’s services. All participants were aged over 16 and gave individual consent to take part. The study was presented to participants as ‘a study of drinking and social life’: the focus on alcohol control strategies was not apparent in recruitment materials, so it is unlikely that participants could have self-selected on the basis of their beliefs about alcohol control strategies. Sample Online questionnaires were completed by 642 men and 776 women aged 16–21 years (mean = 18.1, standard deviation = 1.4) living in South-East England. This age range was chosen to include young people with a range of experiences with alcohol: in the UK, the legal age for purchase of alcohol is 18, but younger people can drink under adult supervision [22]. Measures Respondents used 7-point scales (anchors: ‘not at all effective’, ‘extremely effective’) to indicate their belief in eight alcohol control strategies. The introductory statement ‘Please indicate how well each approach would address alcohol-related problems’ was followed by eight strategies chosen on the basis of variations in evidence of their effectiveness [8] and variations in their focus on individuals or the environment (see Table 1). Questions focused on perceived strategy effectiveness rather than on beliefs about how strategies would change respondents’ own drinking [17]. Principal components factor analysis identified one factor (Eigenvalue = 4.23; 52.81% of total variance explained), and a scale computed as the mean of the eight items had good internal consistency (α = 0.87). Sensation seeking was assessed using 10 items such as ‘I like doing things just for the thrill of it’ (α = 0.90) [23]. Three other aspects of personality were assessed. The root phrase ‘Typically I . . .’ was followed by 10 items in each of three domains: impulsivity (α = 0.80; e.g. ‘. . . easily resist temptations’), extraversion (α = 0.82; ‘. . . am the life of the party’) and conscientiousness (α = 0.90; ‘. . . am always prepared’) [24]. For all scales above, respondents used 7-point scales (anchors: ‘strongly disagree’, ‘strongly agree’). Two scales assessed alcohol outcome expectancies [25], with responses made using 7-point scales © 2014 Australasian Professional Society on Alcohol and other Drugs
146
R. de Visser et al.
Table 1. Perceived efficacy of various alcohol control strategies
Enforce the law against serving drunk people Strictly monitor late-night licensed premises Teach people skills for resisting pressure to drink Restrict late-night alcohol sales Increase the price of alcohol Reduce the number of outlets that sell alcohol Ban alcohol sponsorship of sporting events Raise the legal drinking age Mean
Men (n = 642)
Women (n = 776)
3.96 (2.04)a 3.70 (1.92)b 3.76 (2.05)b 3.47 (1.99)c 3.33 (2.13)c 3.11 (1.97)d 2.85 (2.03)e 2.20 (1.81)f 3.30 (1.46)
4.41 (1.93)a 4.05 (1.81)b 3.94 (1.94)b 3.76 (1.99)c 3.36 (1.98)d 3.26 (1.94)d 3.15 (1.97)d 2.10 (1.66)e 3.50 (1.35)
F(1,1416) = 18.57, P < 0.01 F(1,1416) = 12.88, P < 0.01 F(1,1416) = 2.72, P = 0.10 F(1,1416) = 7.37, P < 0.01 F(1,1416) = 0.09, P = 0.77 F(1,1416) = 2.04, P = 0.15 F(1,1416) = 7.96, P < 0.01 F(1,1416) = 1.14, P = 0.29 F(1,1416) = 2.75, P < 0.01
Within-columns values with different superscripts are significantly different.
(anchors: ‘strongly disagree’, ‘strongly agree’). Ten items assessed the likelihood of positive outcomes when people drink (e.g. ‘They enjoy the buzz’, α = 0.67). Ten items assessed the likelihood of negative outcomes (e.g. ‘They become aggressive’, α = 0.88). Drink refusal self-efficacy (DRSE)—an individual’s perception of his/her capacity to refuse alcohol—was assessed using 7-point scales (anchors: ‘very difficult’, ‘very easy’) [26]. The three subscales—social pressure (e.g. ‘When someone offers me a drink’), emotional relief (e.g. ‘When I am worried’) and opportunistic drinking (e.g. ‘When I am watching TV’)—were correlated at r > 0.65, so a single 19-item scale was used (α = 0.94). Peer alcohol use was assessed as the mean of four novel items (α = 0.86). Participants used 7-point scales (anchors: ‘none of them’, ‘all of them’) to indicate the proportion of (friends/other people their age) who regularly (drink alcohol/get drunk). The 10-item Alcohol Use Disorders Identification Test (AUDIT) assesses consumption frequency and volume, dependence and alcohol-related problems [27]. Scores were summed: scores above 8 for men and 7 for women are commonly considered to indicate hazardous alcohol use. Similar proportions of male and female respondents were above these thresholds (58% vs. 56%, χ2(1) = 0.15, P = 0.70). Participants used 7-point scales (anchors: ‘strongly disagree’, ‘strongly disagree’) to respond to ‘concerns about my health exert a strong influence over my use of alcohol’. Analysis Pearson correlations were calculated to identify bivariate correlates of perceived strategy efficacy to include in linear regression. The skewness and kurtosis of the outcome variables indicated no major violation of assumptions of normality: ‘increase prices’: skew© 2014 Australasian Professional Society on Alcohol and other Drugs
ness = 0.32, kurtosis = −1.23; ‘teach alcohol refusal skills’: skewness = 0.04, kurtosis = −1.21; strategy scale: skewness = 0.21, kurtosis = −0.32 [28]. Results Table 2 displays the demographic profile of the sample and provides comparisons to relevant population data. The sample over-represented students—66% were secondary school students, 22% were in further education, 11% were employed and 1% were unemployed—but this may reflect the sampling strategies [32].The prevalence of excessive alcohol use in the sample was comparable with that reported in population-representative surveys and in other surveys of young people in the UK [33]. With the exception of religion, the demographic profiles of male and female respondents were not significantly different. Table 2 shows men’s and women’s belief in the efficacy of various alcohol strategies. Only two strategies received support scores above the mid-point of the 7-point scale: enforcing the law against serving drunk people, and strictly monitoring late-night licensed premises. Belief in the efficacy of teaching people skills for resisting pressure to drink was not significantly lower than that for monitoring late-night premises. Raising the legal drinking age was perceived to be the least effective strategy. Women expressed significantly greater belief in the efficacy of four strategies: enforcing the law against serving drunk people, strictly monitoring late-night licensed premises, restricting late-night alcohol sales and banning alcohol sponsorship of sporting events. The mean rating of perceived strategy effectiveness was significantly higher for women. Within-subjects repeated measures anova identified differences in the perceived efficacy of the eight strategies. Of the 28 pair-wise comparisons of support, 26 were significant among men [F(1,641) = 40.74, P < 0.01] and 25 were significant among women [F(1,771) = 98.16,
Alcohol policies
147
Table 2. Comparisons of sample with relevant section of general population and sex differences with sample Sample
Population (%) Ethnicitya White n Mixed Black Other Religionb None Christian Muslim Other Employmentc Working and/or studying Not in employment, education or training Parental education Both university educated One university educated Neither university educated Alcohol consumptiond Maximum daily intake > double daily guideline Maximum daily intake > weekly guideline
Men (n = 642) (%)
Women (n = 776) (%)
Within-sample sex difference χ2(4) = 4.14, P = 0.39
90.7 5.2 1.9 1.6 0.6
89.3 4.8 0.5 2.6 0.4
89.9 3.9 0.4 1.7 0.4
35.5 47.6 6.2 10.7
70.6 22.9 3.1 3.4
64.8 29.8 2.7 2.7
90.3 9.7
89.6 0.3
89.6 0.6
23.4 19.3 57.3
21.1 23.3 55.5
40 27
36 26
− − − Men 45 19
Women 40 21
χ2(3) = 8.75, P = 0.03
χ2(1) = 0.79, P = 0.37 χ2(2) = 3.61, P = 0.16
χ2(1) = 2.60, P = 0.11 χ2(1) = 0.65, P = 0.42
a Comparison population is whole population in South-East England [29]. bComparison population is 16- to 24-year-olds in UK [29]. cComparison population is 16- to 24-year-olds in UK [30]. dComparison population is 16- to 24-year-olds in England [31].
Table 3. Correlates of perceived efficacy of different alcohol control strategies (n = 1418)
Variable
Eight strategy scale
Teach alcohol refusal skills
Increase prices
Age Sensation seeking Impulsivity Extraversion Conscientiousness AOE: positive AOE: negative DRSE Peer alcohol use AUDIT Health concerns affect drinking
r = 0.11, P < 0.01 r = −0.17, P < 0.01 r = −0.10, P < 0.01 r = −0.10, P < 0.01 r = 0.20, P < 0.01 r = 0.03, P = 0.33 r = 0.20, P < 0.01 r = 0.16, P < 0.01 r = −0.07, P < 0.01. r = −0.24, P < 0.01 r = 0.11, P < 0.01
r = 0.07, P < 0.01 r = −0.08, P < 0.01 r = −0.07, P < 0.01 r = −0.01, P = 0.67 r = 0.14, P < 0.01 r = 0.04, P = 0.14 r = 0.16, P < 0.01 r = 0.11, P < 0.01 r = −0.01, P = 0.68 r = −0.12, P < 0.01 r = 0.07, P = 0.01
r = 0.04, P = 0.18 r = −0.14, P < 0.01 r = −0.10, P < 0.01 r = −0.05, P = 0.05 r = 0.14, P < 0.01 r = 0.01, P = 0.71 r = 0.13, P < 0.01 r = 0.16, P < 0.01 r = −0.07, P = 0.01 r = −0.22, P < 0.01 r = 0.09, P < 0.01
AOE, alcohol outcome expectancies; AUDIT, alcohol use disorders identification test; DRSE, drink refusal self efficacy.
P < 0.01]. Thus, although the eight strategies could be combined in a scale, each received discrete ratings of efficacy. Table 3 shows that greater perceived efficacy of alcohol control strategies was expressed by older respondents and respondents who were lower on sensation seeking, impulsivity and extraversion, but higher on conscientiousness. Greater efficacy was also per-
ceived by respondents with more negative outcome expectancies, greater DRSE, perception of less peer drinking/drunkenness, lower AUDIT scores and greater health concerns. Linear regression identified six significant multivariate correlates of greater belief in the effectiveness of the eight strategies [F(6,1410) = 38.72, P < 0.01; adjusted R2 = 0.138]. Table 4 shows that greater belief in the © 2014 Australasian Professional Society on Alcohol and other Drugs
148
R. de Visser et al.
Table 4. Multivariate correlates of perceived efficacy of strategies (n = 1418)
AOE: negative AUDIT Age Conscientiousness Health concerns affect drinking Peer alcohol use
B
Standard error
Beta
T
P
0.35 −0.04 0.14 0.14 0.05 −0.08
0.04 0.01 0.03 0.04 0.02 0.03
0.22 −0.21 0.14 0.11 0.07 −0.07
8.40 −7.51 5.35 4.10 2.65 −2.42
