Skip to main content
NCBI home page
Alcohol and Alcoholism (Oxford, Oxfordshire) logoLink to Alcohol and Alcoholism (Oxford, Oxfordshire)
. 2015 Mar 2;50(3):302–309. doi: 10.1093/alcalc/agv007

Screening and Brief Intervention for Substance Misuse: Does It Reduce Aggression and HIV-Related Risk Behaviours?

Catherine L Ward 1,*, Jennifer R Mertens 2, Graham F Bresick 3, Francesca Little 4, Constance M Weisner 5,6
PMCID: PMC4398989  PMID: 25731180

Abstract

Purpose: To explore whether reducing substance misuse through a brief motivational intervention also reduces aggression and HIV risk behaviours.

Methods: Participants were enrolled in a randomized controlled trial in primary care if they screened positive for substance misuse. Substance misuse was assessed using the Alcohol, Smoking and Substance Involvement Screening Test; aggression, using a modified version of the Explicit Aggression Scale; and HIV risk, through a count of common risk behaviours. The intervention was received on the day of the baseline interview, with a 3-month follow-up.

Results: Participants who received the intervention were significantly more likely to reduce their alcohol use than those who did not; no effect was identified for other substances. In addition, participants who reduced substance misuse (whether as an effect of the intervention or not) also reduced aggression but not HIV risk behaviours.

Conclusions: Reducing substance misuse through any means reduces aggression; other interventions are needed for HIV risk reduction.

INTRODUCTION

Substance misuse is a key risk for adolescents, particularly since it is also associated with aggression and sexual risk behaviours. Binge drinking, underage drinking and marijuana use are all associated with sexual victimization (Chanpion et al., 2004); correspondingly, substance misuse has been found to be associated with a range of sexual risk behaviours, such as early sexual debut, multiple partners and no or inconsistent condom use, and with the consequences of risk behaviours, such as sexually transmitted infections (including HIV) and unintended pregnancies (Fergusson and Lynskey, 1996; Kotchick et al., 2001;Tapert et al., 2001). Similarly, alcohol and illegal drug use are associated with aggression in young people (Valois et al., 1995; Resnick et al., 2004; Baxendale et al., 2012). In both the case of aggression and sexual risk behaviours, the disinhibition and expectations of disinhibition associated with intoxication appear to be the mechanisms by which substance use is causally associated with risk behaviours (Fergusson and Lynskey, 1996; Anderson and Huesmann, 2003).

Three studies of interventions aimed at reducing alcohol-related aggression have been identified in the literature (McMurran, 2012; Goodall, 2015). The Control of Violence for Angry Impulsive Drinkers (COVAID) program is a 10-session cognitive behavioural therapy program that addresses both alcohol use and aggression, and aggression-related expectancies associated with alcohol (McCulloch and McMurran, 2008). A small (n = 10) pre–post trial with long-term follow-up found a reduction in alcohol-related aggression that persisted (as measured by convictions for violent offences) to a mean follow-up time of 29 weeks (McCulloch and McMurran, 2008). The SafERteens programme, by contrast, is an emergency-room-based brief intervention aimed at adolescents who are admitted for emergency medical treatment and who report both alcohol use and aggression (Cunningham et al., 2010). SafERteens takes advantage of the ‘teachable moment’ provided by an emergency room admission and used motivational interviewing combined with resetting norms and skills-training role-plays for conflict management and alcohol refusal (Cunningham et al., 2009). A large randomized controlled trial (n = 726) compared three arms: receiving the intervention from a computer, receiving the intervention from a therapist and no intervention; this trial found that those who had received the intervention from a therapist or from a computer reported reductions in alcohol consequences at 6 months, and those in the therapist group also reported reductions in peer aggression (Cunningham et al., 2010). Finally, a large randomized controlled trial comparing a brief intervention for alcohol misuse with a single-session version of COVAID (focused on both alcohol and aggression) in facial trauma patients found reduction in drinking but none in aggression in both arms at 12-month follow-up (Goodall et al., 2012). Studies of substance abuse treatment and aggression thus typically examine only interventions that include a specific focus on aggression and—with one exception (Goodall et al., 2012)—do not examine whether simply reducing substance misuse is sufficient also to reduce aggression. This is despite a robust set of studies finding that many brief interventions in trauma settings for alcohol use alone do reduce trauma and injury recidivism, implying that there may be some effect on aggression (Nilsen et al., 2008).

The literature on sexual risk behaviour reduction through treatment for substance misuse also tends not to examine whether reducing substance misuse is, on its own, sufficient to reduce sexual risk behaviours. A review of drug abuse treatments for HIV prevention found that studies focused primarily on methadone maintenance treatments; these were found to be effective in reducing both HIV risk behaviours (mainly needle-sharing) and HIV infection (Sorenson and Copeland, 2000). Alternatively, the reduction in sexual risk behaviours is achieved through an intervention targeted specifically at these risk behaviours and integrated into substance abuse treatment (Prendergast et al., 2001; Lewis et al., 2014). However, (1) a 16-week culturally tailored cognitive behavioural treatment for methamphetamine dependence also reduced sexual risk behaviours (Shoptaw et al., 2005), and (2) there are some studies (of treatment for alcohol and for cocaine abuse) that found that sexual risk behaviours did reduce once substance misuse reduced, without any treatment specific to the sexual risk behaviours (Metzger et al., 1998; Gossop et al., 2002).

If substance misuse increases the likelihood of aggression and of sexual risk behaviours via intoxication, it follows that reductions in substance misuse should also reduce these risk behaviours, without necessitating an additional intervention targeting aggressive or sexual risk behaviours. Surprisingly, this hypothesis appears to have received very little attention in studies of substance abuse treatment. Further, screening and brief interventions for substance misuse are gaining in popularity because of their low cost, effectiveness and potential for integration into other health systems (Bertholet et al., 2005; Babor et al., 2006; Institute of Medicine, 2006; Moyer, 2013); it would be enormously helpful if this mode of delivery of substance abuse treatment also reduced other risk behaviours.

We had conducted a randomized controlled trial, which found that a single brief motivational interview did reduce alcohol misuse in young South Africans (Mertens et al., 2014); we took advantage of this and reanalysed data from the trial to investigate further whether reductions in substance misuse would also be associated with reductions in aggression and sexual risk behaviours.

METHODS

Sample

Patients aged 18–24 presenting to a community health centre for primary care in Cape Town, South Africa, were screened for substance misuse (N = 2047). Because evidence for screening tools in young adults is not strong (Jonas et al., 2012; Patton et al., 2014) and because of time pressure on services in the health centre, we used single-item screening questions for alcohol and for drugs (Smith et al., 2009), an approach now recommended by the National Institute on Drug Abuse (NIDA) and used in health plans (NIDA, 2011). To screen for alcohol use in women, we asked: ‘In the past year, how many times have you had 3 or more drinks on one occasion?’, and in men we asked: ‘In the past year, how many times have you had 5 or more drinks on one occasion?’. To screen for use of other drugs, we asked: ‘In the past three months, how many times have you used drugs? When I talk about a drug here, I don't mean cigarettes or snuff. I mean drugs like dagga [cannabis], mandrax [methaqualone], tik [crystal methamphetamine], cocaine, heroin, LSD, or sleeping pills or other medication that you have used in a way that it was not prescribed, or not prescribed for you’. In a previous study in the same population, we found positive answers to these questions to have 83% sensitivity and 93% specificity for identifying alcohol use assessed by the AUDIT and 92% sensitivity and 99% specificity for risky drug use assessed by the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) (Mertens et al., 2008).

Of those screened, 1478 were excluded, because they screened negative (n = 1370), were too ill to participate (n = 22) or had no telephone and therefore could not be followed (n = 86). A further 88 patients who screened positive refused the opportunity to participate in the study and were given an information sheet listing other treatment agencies in Cape Town. A total of 57 left the clinic before they could be approached for recruitment, and 21 were excluded from the analysis because of baseline data loss due to mobile device failure. Figure 1 provides the CONSORT diagram for the study.

Fig. 1.

Fig. 1.

Open in a new tab

CONSORT flowchart.

Measures

After screening positive, substance misuse was assessed using the WHO ASSIST 3.0 (WHO ASSIST Working Group, 2002), which provides scores for each substance that can be categorized into low/zero- and medium/high-risk use.

Aggression was assessed using a slightly modified form of the Explicit Aggression Scale (Barnwell et al., 2006). The original scale assessed aggression while drinking; we were also interested in the effects of illicit substances and so adapted the scale so that questions read (for instance) ‘How many times in the past 3 months have you been involved in a physical fight while drinking or using drugs?’ Because we were concerned about subject burden in an already lengthy questionnaire, we only asked the 13 questions from the Explicit Aggression Scale that dealt with aggression after substance use. The Explicit Aggression Scale is based on the questions asked in the Revised Conflict Tactics Scale (Straus et al., 1996) but deals only with the less serious and therefore more frequent forms of aggression (slapping, yelling, threatening to hurt, pushing, breaking things, pulling hair, twisting someone's arm, arguing, throwing things in anger, saying hurtful things to someone or about someone). It does not deal with the more serious (and less frequent) forms such as beating, stabbing or rape.

HIV risk behaviour was assessed as a count of positive answers to six questions (Avalos et al., 2009), which asked whether the following had occurred in the last 3 months: having a partner who had engaged in transactional sex, who was a man who had had sex with a man, who used injection drugs and who had a sexually transmitted infection; having multiple partners; or failure to use a condom at last intercourse.

Procedure

The remaining 403 patients who screened positive and who agreed to participate in the study were interviewed in a private setting about their substance misuse, substance-related aggression and HIV risk behaviours. The study was approved by the institutional review boards of the University of Cape Town, South Africa; the Human Sciences Research Council in South Africa; and the Division of Research, Kaiser Permanente, USA. Participants gave informed consent before being enrolled in the study.

After screening, half of the participants were randomly assigned (using block randomization) to receive a brief motivational intervention plus a referral resource list (n = 206); the other half (n = 197) received usual care plus a referral resource list (minimally enhanced usual care or mEUC; Freedland et al., 2011). Trained nurse practitioners used the procedures outlined in Health Behaviour Change: A Guide for Practitioners (Rollnick et al., 1999) to carry out brief motivational interventions addressing alcohol and drug use and were supervised weekly in order to ensure fidelity of the intervention. Both the groups were interviewed again (in person) 3 months later. Incentives (in the form of a grocery store or cell phone voucher, to the value of ZAR50—approximately USD4.50) were offered to participants who completed follow-up interviews. Participants who came to the clinic for their follow-up interviews were reimbursed for their transport costs; otherwise, we visited their homes to complete the follow-up interviews.

For further details about the study, please see Mertens et al. (2014); here, we present an analysis of the data specifically concerning the aggressive and HIV-related risk behaviour outcomes.

Data analysis

We summarized the prevalence of substance misuse, aggression and HIV risk behaviour using frequencies and percentage frequencies. We summarized the association between substance use reduction and aggression at different time points using frequencies and percentage frequencies for the intervention and control groups separately and overall. We summarized the association between substance use reduction and HIV risk behaviour at different time points using means and standard errors for the intervention and control groups separately and overall. For each patient, we determined the change in HIV risk behaviour at follow-up compared with baseline by calculating the difference in the summed counts. A negative difference indicated a decrease in HIV risk behaviour and vice versa. We illustrated the distribution of these changes using a percentage histogram and compared these differences in the change in HIV risk behaviour between those patients who reduced substance use and those who did not reduce substance use using a Kruskal–Wallis test.

A large number of participants (50%) reported zero aggression at either baseline or follow-up. Because of this large percentage of zero scores, a zero-inflated negative binomial regression model (Long, 1997; Long and Freese, 2001) was used to investigate the relationship between reductions in total substance misuse scores and aggression scores in the full sample, and again in the group that used alcohol at risky levels. This model assumes that some of the zeroes are structural zeroes in that some respondents have no propensity for violence (33% of all respondents recorded a zero score for aggression at both time points). It thus models both the relative risk of a zero score (using a logistic model) and the relative increase in the mean score (using a negative binomial model) as a function of the predictor variables. As predictor variables we included an indicator variable for substance use reduction (defined as any reduction in ASSIST scores), time (baseline and follow-up) and their interaction (to test whether the change in aggression scores over time is the same for those who reduced substance use and those who did not), as well as an intervention group indicator. The model, therefore, allows us to explore the following: (1) whether there was an effect of intervention versus mEUC group on aggression, and (2) whether any reduction in substance misuse (regardless of intervention or mEUC group membership) was associated with a reduction in aggression over time. We used robust standard errors to take into account the repeated measures at the two time points within subjects.

We used this modelling approach to explore the role of participants' total substance misuse scores, as our hypothesis was that any reduction in substance misuse would lead to a concomitant reduction in risk behaviour scores. We used the same modelling approach in the specific sub-group who used alcohol at risky levels, because of alcohol's strong association with aggression (Anderson, 2015), and because our previous work demonstrated that significantly more people in the intervention group compared with the mEUC group reduced their alcohol use and risk (38 versus 21%, respectively; F = 4.79, P = 0.0293), and that this was the only substance for which the intervention appeared to have an effect (Mertens et al., 2014).

In the current paper, we therefore examined whether less overall substance misuse had an impact on aggressive and HIV risk behaviours, and the same question in the sub-group that used alcohol at risky levels.

RESULTS

A total of 363 patients completed both baseline and follow-up interviews, with a follow-up rate of 90% (please see Fig. 1 for further details). Those lost to follow-up were more likely to fall in the low-risk groups (Mertens et al., 2014). No participant reported use of heroin or inhalants; alcohol was the substance most used at risky levels, with half of the sample that was followed up (51.5%) using it at risky levels at baseline (see Table 1 for a report of substance use by category and of aggression and HIV risk behaviours by substance and whether that substance was used at risky levels).

Table 1.

Prevalence of substance misuse and risk behaviours by substance misuse category (N = 363)

Substance Risk category Substance misuse at baseline (n, % of sample) Substance misuse at follow-up (n, % of sample) n reporting any aggression at baseline (n, % of substance risk group)a n reporting any aggression at follow-up (n, % of substance risk group) n reporting any sexual risk at baseline (n, % of substance risk group) n reporting any sexual risk at follow-up (n, % of substance risk group)
Alcohol No/low-risk use 176 (48.5) 245 (67.5) 93 (52.8) 80 (32.6) 123 (69.9) 171 (69.8)
Risky use 187 (51.5) 118 (32.5) 130 (69.5) 118 (63.1) 164 (87.7.2) 95 (80.5)
Cannabis No/low-risk use 288 (79.3) 316 (87.1) 161 (55.9) 109 (34.5) 240 (83.3) 234 (74.1)
Risky use 74 (20.4) 47 (12.9) 62 (83.8) 33 (70.2) 47 (63.5) 32 (68.1)
Methaqualone No/low-risk use 352 (97.0) 359 (98.9) 213 (60.5) 138 (38.4) 282 (80.1) 263 (73.3)
Risky use 10 (2.8) 4 (1.1) 10 (100.0) 4 (100.0) 5 (50.0) 3 (75.0)
Cocaine No/low-risk use 362 (99.7) 363 (100) 223 (61.6) 142 (39.1) 287 (79.3) 266 (73.3)
Risky use 1 (0.3) 0 0 0 0 0
Methamphetamine No/low-risk use 329 (90.6) 347 (95.6) 192 (58.4) 126 (36.3) 263 (79.9) 255 (73.4)
Risky use 33 (9.1) 16 (4.4) 31 (93.9) 16 (100.0) 24 (72.3) 11 (68.8)
Sedatives No/low-risk use 357 (98.3) 363 (100) 218 (61.1) 142 (39.1) 283 (79.3) 266 (73.3)
Risky use 5 (1.4) 0 5 (100.0) 0 4 (80.0) 0
Open in a new tab

a‘% of substance risk group’—i.e. these are row percentages: the percentage of the low-/no-risk group or the risky use group that reported either aggression or sexual risk at these time points.

Participants reported the full range of possible aggression scores (0–39) at baseline, with a median of 2 (interquartile range 0–9). The majority of the sample (223; 61.43%) reported zero aggression at baseline, which accounts for the low mean value. For the HIV risk behaviours, total scores ranging from 0 to 6 were possible, actual scores ranged from 0 (n = 76; 20.94%) to 4 (n = 2, 0.55%) at baseline and at follow-up from 0 (n = 97, 26.72%) to 5 (n = 1, 0.28%). At baseline, the majority (268; 73.83%) reported one or two HIV risk behaviours in the prior 3 months and similarly in the past 3 months at follow-up (n = 253; 69.70%).

Table 2 describes the proportions of the sample who had zero aggression and breaks this down by use of any substance at risky levels and use of alcohol at risky levels, intervention versus mEUC group and within those, into sub-groups of those who reduced or did not reduce their substance misuse. The table also provides the mean and standard error for the HIV risk behaviours for each group. In both the intervention and mEUC groups, whether in the group that used any substance or just the sub-group that used alcohol at risky levels, there is an increase in zero aggression scores from baseline to follow-up. Those who reduced their substance misuse also consistently show an increase in zero aggression scores from baseline to follow-up. However, there is very little change or variation by group in the HIV risk behaviour scores.

Table 2.

Aggression and HIV risk scores by intervention and mEUC groups and substance use reduction status

Group Substance use reduced? Percentagea with zero aggression score, % (n)
 HIV risk score, mean (S.E.)
Baseline Follow-up Baseline Follow-up
Aggression scores in the full sample
 Intervention group (n = 173) Non-reducer (n = 52) 48 (25) 42 (22) 1.10 (0.11) 1.29 (0.12)
Reducer (n = 121) 33 (40) 60 (73) 1.20 (0.08) 1.13 (0.08)
Total 37 (64) 55 (95) 1.23 (0.06) 1.18 (0.07)
 mEUC group (n = 190) Non-reducer (n = 39) 49 (19) 54 (21) 1.33 (0.13) 1.23 (0.12)
Reducer (n = 151) 37 (56) 69 (105) 1.21 (0.07) 1.11 (0.08)
Total 39 (75 of 190) 66 (126 of 190) 1.24 (0.06) 1.14 (0.07)
Aggression scores in the subgroup of participants who used alcohol at risky levels
 Intervention group (n = 84) Non-reducer (n = 14) 57 (8) 45 (15 of 33) 1.21 (0.19) 1.24 (0.14)
Reducer (n = 70) 27 (19) 39 (9) 1.42 (0.10) 1.22 (0.18)
Total 32 (27) 43 (24) 1.39 (0.09) 1.23 (0.11)
 mEUC group (n = 103) Non-reducer (n = 18) 44 (8) 58 (15) 1.44 (0.18) 1.27 (0.14)
Reducer (n = 85) 26 (22) 47 (17) 1.36 (0.09) 1.31 (0.18)
Total 29 (30) 52 (32) 1.37 (0.08) 1.29 (0.12)
Open in a new tab

aPercentages here are row percentages: percentages of reducers (or non-reducers) in the intervention or mEUC groups.

The relationship between reductions in total substance misuse scores and aggression scores in the full sample is reported in Table 3 and further explored in the sub-group that used alcohol at risky levels (see Table 4). The exponentiated coefficients of the two components of the zero-inflated models can be interpreted as estimates of (1) the relative odds of a zero versus nonzero score and (2) the ratio of the mean scores in two groups. Among those in the full sample who reduced their overall substance use, there was a significant reduction in aggressive behaviour, indicated by the reduction in means of scores (ratio of means, follow-up versus baseline = 0.79, P = 0.014) and by the increased odds of a zero score at follow-up compared with baseline (OR = 3.71, P < 0.0001). In the sub-group with risky levels of alcohol use, we observed a similar reduction in terms of the increased odds of a zero score at follow-up that was significant (OR = 2.32, P = 0.010); the reduction in mean aggression scores (OR = 0.86, P = 0.320) was not significantly different between those who reduced their substance use at follow-up and those who did not. In the participants in both the full and alcohol risk sub-samples who did not reduce substance use, there was no significant change in aggression scores, whether means or the odds of a zero aggression score were compared. In addition, there was no effect of intervention group on these findings, indicating that the change in substance use, and not the intervention itself, was the factor associated with change in aggressive behaviour.

Table 3.

Relationship between substance use and aggression scores in the full sample

Nonzero scores Ratio of mean aggression scores P 95% CI
Time (follow-up versus baseline)
 Among substance use reducers 0.79 0.014 (0.66; 0.95)
 Among non-reducers 1.11 0.595 (0.75; 1.65)
 Intervention group versus control group 1.08 0.406 (0.89; 1.34)
Relative odds of a zero aggression score Odds ratio P 95% CI
Time (follow-up versus baseline)
 Among substance use reducers 3.71 <0.000 (2.72; 5.10)
 Among non-reducers 0.98 0.949 (0.58; 1.66)
 Intervention group versus control group 0.75 0.155 (0.50; 1.12)
Open in a new tab

Estimated from a zero-inflated negative binomial models for aggression scores that included time (follow-up versus baseline), substance use reduction (yes versus no), the time*reduction interaction and the intervention group versus control group indicator.

Table 4.

Relationship between substance use and aggression scores among alcohol users only

Nonzero scores Ratio of mean aggression scores P 95% CI
Time (follow-up versus baseline)
 Among substance use reducers 0.86 0.320 (0.66; 1.15)
 Among non-reducers 1.96 0.879 (0.53; 1.71)
 Intervention group versus control group 1.03 0.801 (0.81; 1.32)
Relative odds of a zero aggression score Odds ratio P 95% CI
Time (follow-up versus baseline)
 Among substance use reducers 2.32 0.010 (1.22; 4.42)
 Among non-reducers 1.04 0.930 (0.41; 2.63)
 Intervention group versus control group 0.91 0.747 (0.51; 1.62)
Open in a new tab

Estimated from a zero-inflated negative binomial models for aggression scores that included time (follow-up versus baseline), substance use reduction (yes versus no), the time*reduction interaction and the intervention group versus control group indicator.

Figure 2 illustrates the frequency distributions of the different levels of HIV risk behaviour change among patients who did and who did not reduce substance use, respectively. There is an indication of higher percentages of decreased HIV risk scores (i.e. negative changes) among those who did not reduce their substance use and higher percentages of increased HIV risk scores (i.e. positive changes) among those who did reduce their substance use. However, the median change was zero for both groups, and the groups did not differ significantly, Kruskal–Wallis P-value = 0.3086.

Fig. 2.

Fig. 2.

Open in a new tab

Change in six-item HIV Risk score among those who did not and who did reduce substance use.

DISCUSSION

Our findings suggest that reducing substance misuse may be associated with a reduction in aggressive behaviour but that this is not so for HIV risk behaviours.

In contrast to the only other trial we could identify that examined a brief intervention for substance misuse and its effect on aggression (Goodall et al., 2012), we found that reducing substance misuse did reduce aggression. However, one key similarity between our results and those of Goodall et al. (2012) was that there was no effect of intervention group on the aggression outcome, despite the fact that the intervention group did reduce their substance use more than the control group. In essence, we have thus found that our intervention did reduce substance misuse, and reduction in substance misuse was related to reduction in aggression—but reduction in aggression was not greater in the intervention group. This is possibly because both the intervention and control groups reduced their substance misuse, and the difference between the two groups was small—as might be expected from a very light-touch intervention. The reduction in substance misuse in both groups may have been partly due to participant reactivity—that is, the baseline assessment itself performed as an intervention (Walters et al., 2009). In addition, of course, our intervention did not directly address aggression and its relationship to substance misuse.

Given this, our results thus imply that reduction in substance misuse for any reason, with or without a substance abuse intervention, is what leads to reductions in aggression and that therefore public health efforts to reduce aggression should be broadly focused on attempts to reduce substance use—through any means available. Such means might indeed include interventions such as screening and brief motivational interviewing but might also include a wide range of initiatives directed at controlling the supply and demand sides of substance availability, as well as efforts to improve safety in drinking establishments (Anderson, 2015).

HIV risk behaviours, on the other hand, were not reduced, whether or not participants reduced their substance use. While studies do suggest that sexual risk behaviours are associated with intoxication—and therefore might reasonably reduce if intoxication is reduced (Fergusson and Lynskey, 1996)—other literature suggests that sexual risk behaviours occur in a more complex context than the simple relationship between intoxication and alcohol. For instance, the nature of the relationship, its length and partners' experiences of trust and intimacy (Halpern-Felsher et al., 1996) and patterns of substance misuse (such as binge-drinking and marijuana use; Guo et al., 2002) may be more important in sexual risk behaviour than simply use of substances. Interventions that focus on social norms and views of peers may also be effective in reducing HIV risk and are brief to implement and so could practically be integrated into brief substance abuse treatments. Norms feedback interventions have proven useful in reducing both substance misuse (Neville, 2015) and HIV risk behaviours (Kelly et al., 1991) separately, and so there is reason to believe that a combined intervention would be helpful. This may be a fruitful area for future research.

Our results suggest, therefore, that reducing substance misuse (through any means) can reduce aggression, at least in minor forms such as yelling and slapping that may precede more serious forms, but they do not speak to more serious forms of aggression. Future research should explore whether serious aggression is also reduced through reducing substance misuse. Given the small number of participants who did respond to our intervention by reducing their substance use, future research might also explore whether a more intensive intervention, such as several sessions of motivational interviewing, instead of just one, might be more effective in reducing both substance misuse and aggression—and possibly even HIV risk behaviours. Future research should also explore whether the effects of such interventions are maintained. While some trials show effects at several months after one brief intervention (see, for instance, WHO Brief Intervention Study Group, 1996, which showed effects at 9 months post-intervention), many preventive interventions show a drop-off of effect over time (Nation et al., 2003).

The possibilities of other interventions, such as brief motivational interviewing combined with problem-solving therapy (Sorsdahl et al., 2014), should also be explored; as should the question of whether interventions to reduce aggression and HIV risk behaviours must be integrated into substance abuse treatments in order to achieve reductions in those related risk behaviours as well as in substance misuse (Sorenson and Copeland, 2000; Prendergast et al., 2001; McCulloch and McMurran, 2008; Cunningham et al., 2010).

In the South African context (as in many other low- and middle-income countries), there are twin problems with regard to intervening in risk behaviours: substance abuse, aggression and HIV risk behaviours all occur at high levels (Coovadia et al., 2009), yet there also is a dearth of professionals available to deliver interventions (Hanson et al., 2010). Substance misuse also plays a key role in both aggression and HIV risk. It is crucial, therefore, that interventions that address substance misuse in such a way that both aggression and HIV risk are also reduced are developed, which are sufficiently simple for delivery by paraprofessionals and which can easily be integrated into existing service systems (Weisner and Schmidt, 2001). Brief motivational interventions are precisely the kind of intervention that appear to meet these criteria: their simple delivery format means that paraprofessionals can probably be trained to deliver them, and their brevity makes them easier to integrate into service systems than longer, more complex interventions. For these reasons, in contexts such as South Africa, they are very attractive. Yet our results suggest that perhaps the simplest of such interventions—screening and a single-session brief motivational intervention—while it can reduce alcohol use (Mertens et al., 2014), is not sufficient to achieve the goal of reducing substance use as well as aggression and HIV risk behaviour. It may be that multiple session brief motivational interventions are needed to decrease substance misuse sufficiently to have an impact on HIV risk behaviours. Alternatively, other, only slightly more complex, interventions are being developed for precisely this context (Sorsdahl et al., 2014). Their promise to achieve reductions in substance misuse, and in other risk behaviours, must urgently be explored further.

FUNDING

This work was funded by the National Institute on Drug Abuse (R21DA022557 to JRM); writing this paper was supported in part by a grant to CLW from the University of Cape Town's University Research Committee.

CONFLICT OF INTEREST STATEMENT

None declared.

ACKNOWLEDGEMENTS

We dedicate this manuscript with gratitude and sorrow to Dr Alan Flisher, whose untimely death prevented him from contributing as an author. This study could have not been accomplished without his leadership, wisdom and scientific contributions. We thank the patients and staff of the Metro District Health Services in Cape Town and our research assistants Lesley Arendse, Wendy Booy, Vuyisa Dumile, Nokuthula Kulati and Rodney Stoffberg, and project coordinator Lynn Hendricks for patient recruitment, participant tracking and follow-up interviewing and coordination. We thank the nurse practitioners Sr Yvette MacDonald, Sr Winifred Mgudlwa and Sr Zoliswa Ngqiniso for their excellent work in carrying out the interventions.

REFERENCES

  1. Anderson P. (2015) Alcohol and violence. In: Donnelly PD, Ward CL. (eds). The Oxford Textbook of Violence Prevention: Epidemiology, Evidence and Policy. Oxford, UK: Oxford University Press, 291–6. [Google Scholar]
  2. Anderson CA, Huesmann LR. (2003) Human aggression: A social-cognitive view. In: Hogg MA, Cooper J. (eds). The Sage Handbook of Social Psychology. Thousand Oaks, CA: Sage Publications, Inc., 296–323. [Google Scholar]
  3. Avalos LA, Mertens JR, Ward CL, et al. (2009) Stress, substance use and sexual risk behaviours among primary care patients in Cape Town, South Africa. AIDS Behav 14:359–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babor TF, Higgins-Biddle JC, Dauser D, et al. (2006) Brief interventions for at-risk drinking: Patient outcomes and cost-effectiveness in managed care organizations. Alcohol Alcohol 41:624–31. [DOI] [PubMed] [Google Scholar]
  5. Barnwell SS, Borders A, Earleywine M. (2006) Alcohol-aggression expectancies and dispositional aggression moderate the relationship between alcohol consumption and alcohol-related violence. Aggress Behav 32:517–27. [DOI] [PubMed] [Google Scholar]
  6. Baxendale S, Cross D, Johnston R. (2012) A review of the evidence on the relationship between gender and adolescents’ involvement in violent behavior. Aggress Violent Behav 17:297–310. [Google Scholar]
  7. Bertholet N, Daeppen JB, Wietlisbach V, et al. (2005) Reduction of alcohol consumption by brief alcohol intervention in primary care: Systematic review and meta-analysis. Arch Intern Med 165:986–95. [DOI] [PubMed] [Google Scholar]
  8. Chanpion HLO, Foley KL, Durant RH, et al. (2004) Adolescent sexual victimization, use of alcohol and other substances, and other health risk behaviors. J Adolesc Health 35:321–8. [DOI] [PubMed] [Google Scholar]
  9. Coovadia H, Jewkes R, Barron P, et al. (2009) The health and health system of South Africa: Historical roots of current public health challenges. Lancet 374:817–34. [DOI] [PubMed] [Google Scholar]
  10. Cunningham RM, Walton MA, Goldstein A, et al. (2009) Three-month follow-up of brief computerized and therapist interventions for alcohol and violence among teens. Acad Emerg Med 16:1193–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cunningham RM, Chermack ST, Shope JT, et al. (2010) Effects of a brief intervention for reducing violence and alcohol use among adolescents: A randomized trial. JAMA 304:527–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fergusson D, Lynskey MT. (1996) Alcohol misuse and adolescent sexual behaviors and risk taking. Pediatrics 98:91–6. [PubMed] [Google Scholar]
  13. Freedland KE, Mohr DC, Davidson KW, et al. (2011) Usual and unusual care: Existing practice groups in randomized controlled trials of behavioral interventions. Psychosom Med 73:323–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Goodall C. (2015) Preventing violence through intervention in the health system. In: Donnelly PD, Ward CL. (eds). The Oxford Textbook of Violence Prevention: Epidemiology, Evidence and Policy. Oxford, UK: Oxford University Press, 225–30. [Google Scholar]
  15. Goodall C, Bowman A, Smith I, et al. (2012) A randomized trial of brief intervention strategies in patients with alcohol-related facial trauma as a result of interpersonal violence. Addict Sci Clin Pract 7:A66. [Google Scholar]
  16. Gossop M, Marsden J, Stewart D, et al. (2002) Reduced injection risk and sexual risk behaviours after drug misuse treatment: Results from the National Treatment Outcome Research Study. AIDS Care 14:77–93. [DOI] [PubMed] [Google Scholar]
  17. Guo J, Ick-Joong C, Hill KG, et al. (2002) Developmental relationships between adolescent substance use and risky sexual behavior in young adulthood. J Adolesc Health 31:354–62. [DOI] [PubMed] [Google Scholar]
  18. Halpern-Felsher BL, Millstein SG, Ellen JM. (1996) Relationship of alcohol use and risky sexual behavior. J Adolesc Health 19:331–6. [DOI] [PubMed] [Google Scholar]
  19. Hanson K, Cleary S, Schneider H, et al. (2010) Scaling up health policies and services in low- and middle-income settings. BMC Health Serv Res 10:I1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Institute of Medicine. (2006) Improving the Quality of Health Care for Mental and Substance-use Conditions: Quality Chasm Series. Washington, DC: National Academies Press. [PubMed] [Google Scholar]
  21. Jonas DE, Garbutt JC, Brown JM, et al. (July 10, 2012) Screening, behavioral counseling, and referral in primary care to reduce alcohol misuse. Comparative Effectiveness Review No. 64. Available at: http://www.effectivehealthcare.ahrq.gov/search-for-guides-reviews-and-reports/?pageaction=displayproduct&productid=1135 (19 December 2013, date last accessed). [PubMed]
  22. Kelly JA, St Lawrence JS, Diaz YE, et al. (1991) HIV risk behaviour reduction following intervention with key opinion leaders of population: An experimental analysis. Am J Public Health 81:168–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kotchick BA, Shaffer A, Forehand R. (2001) Adolescent sexual risk behavior: A multi-system perspective. Clin Psychol Rev 21:493–519. [DOI] [PubMed] [Google Scholar]
  24. Lewis MA, Patrick ME, Litt DM, et al. (2014) Randomized controlled trial of a web-delivered personalized normative feedback intervention to reduce alcohol-related risky sexual behaviour among college students. J Consult Clin Psychol 82:429–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Long JS. (1997) Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage. [Google Scholar]
  26. Long JS, Freese J. (2001) Predicted probabilities for count models. Stata Journal 1:51–7. [Google Scholar]
  27. McCulloch A, McMurran M. (2008) Evaluation of a treatment programme for alcohol-related aggression. Criminal Behav Mental Health 18:224–31. [DOI] [PubMed] [Google Scholar]
  28. McMurran M. (2012) Individual-level interventions for alcohol-related violence: A rapid evidence assessment. Criminal Behav Mental Health 22:14–28. [DOI] [PubMed] [Google Scholar]
  29. Mertens JR, Ward CL, Flisher AJ, et al. (2008) Development of a brief alcohol and drug screener in public sector Cape Town primary health care clinics. 10th annual meeting of the International Society of Addiction Medicine Cape Town, South Africa. [Google Scholar]
  30. Mertens JR, Ward CL, Bresick GF, et al. (2014) Effectiveness of nurse-practitioner-delivered brief motivational intervention for young adult alcohol and drug use in primary care in South Africa: A randomized clinical trial. Alcohol Alcohol 49:430–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Metzger DS, Navaline H, Woody GE. (1998) Drug abuse treatment as AIDS prevention. Public Health Rep 113:97–106. [PMC free article] [PubMed] [Google Scholar]
  32. Moyer VA. (2013) Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. preventive services task force recommendation statement. Ann Intern Med 159:210–8. [DOI] [PubMed] [Google Scholar]
  33. National Institute on Drug Abuse. (2011) Screening for Drug use in General Medical Settings: Quick Reference Guide. Bethesda, MD: National Institute on Drug Abuse. [Google Scholar]
  34. Nation M, Crusto C, Wandersman A, et al. (2003) What works in prevention: Principles of effective prevention programs. Am Psychol 58:449–56. [DOI] [PubMed] [Google Scholar]
  35. Neville FG. (2015) Preventing violence through changing social norms. In: Donnelly PD, Ward CL. (eds). The Oxford Textbook of Violence Prevention: Epidemiology, Evidence and Policy. Oxford, UK: Oxford University Press, 239–44. [Google Scholar]
  36. Nilsen P, Baird J, Mello MJ, et al. (2008) A systematic review of emergency care brief alcohol interventions for injury patients. J Subst Abuse Treat 35:184–201. [DOI] [PubMed] [Google Scholar]
  37. Patton R, Deluca P, Kaner E, et al. (2014) Alcohol Screening and Brief Intervention for adolescents: The how, what and where of reducing alcohol consumption and related harm among young people. Alcohol Alcohol 49:207–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Prendergast ML, Urada D, Podus D. (2001) Meta-analysis of HIV risk-reduction interventions within drug abuse treatment programs. J Consult Clin Psychol 69:389–405. [DOI] [PubMed] [Google Scholar]
  39. Resnick MD, Ireland M, Borowsky IW. (2004) Youth violence perpetration: What protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. J Adolesc Health 35:424.e1–424.e10. [DOI] [PubMed] [Google Scholar]
  40. Rollnick S, Mason P, Butler C. (1999) Health Behavior Change: A Guide for Practitioners. New York: Churchill Livingstone. [Google Scholar]
  41. Shoptaw S, Reback CJ, Peck JA, et al. (2005) Behavioral treatment approaches for methamphetamine dependence and HIV-related sexual risk behaviors among urban gay and bisexual men. Drug Alcohol Depend 78:125–34. [DOI] [PubMed] [Google Scholar]
  42. Smith PC, Schmidt SM, Allensworth-Davies D, et al. (2009) Primary care validation of a single-question alcohol screening test. J Gen Intern Med 24:783–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sorenson JL, Copeland AL. (2000) Drug abuse treatment as an HIV prevention strategy: A review. Drug Alcohol Depend 59:17–31. [DOI] [PubMed] [Google Scholar]
  44. Sorsdahl K, Myers B, Ward CL, et al. (2014) Adapting a blended motivational interviewing and problem-solving intervention to address risky substance use among South Africans. Psychother Res. doi:10.1080/10503307.2014.897770. [DOI] [PubMed] [Google Scholar]
  45. Straus MA, Hamby SL, Boney-McCoy S, et al. (1996) The revised conflict tactics scale (CTS2): Development and preliminary psychometric data. J Fam Issues 17:283–316. [Google Scholar]
  46. Tapert SF, Aarons GA, Sedlar GR, et al. (2001) Adolescent substance use and sexual risk-taking behavior. J Adolesc Health 28:181–9. [DOI] [PubMed] [Google Scholar]
  47. Valois RF, McKeown RE, Garrison CZ, et al. (1995) Correlates of aggressive and violent behaviors among public high school adolescents. J Adolesc Health 16:26–34. [DOI] [PubMed] [Google Scholar]
  48. Walters ST, Vader AM, Harris TR, et al. (2009) Reactivity to alcohol assessment measures: An experimental test. Addiction 104:1305–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Weisner CM, Schmidt LA. (2001) Rethinking access to alcohol treatment. Recent Dev Alcohol 15:107–36. [DOI] [PubMed] [Google Scholar]
  50. WHO ASSIST Working Group. (2002) The alcohol, smoking and substance involvement screening test (ASSIST): Development, reliability and feasibility. Addiction 97:1183–94. [DOI] [PubMed] [Google Scholar]
  51. WHO Brief Intervention Study Group. (1996) A cross-national trial of brief interventions with heavy drinkers. Am J Public Health 86:948–55. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Alcohol and Alcoholism (Oxford, Oxfordshire) are provided here courtesy of Oxford University Press

RESOURCES