The feasibility and effectiveness of a web-based personalised feedback and social norms alcohol intervention in UK university students: A randomised control trial

Abstract

Objective

Alcohol misuse amongst University students is a serious concern, and research has started to investigate the feasibility of using e-health interventions. This study aimed to establish the effectiveness of an electronic web-based personalised feedback intervention through the use of a randomised control trial (RCT).

Methods

506 participants were stratified by gender, age group, year of study, self-reported weekly consumption of alcohol and randomly assigned to either a control or intervention condition. Intervention participants received electronic personalised feedback and social norms information on their drinking behaviour which they could access by logging onto the website at any time during the 12-week period. CAGE score, average number of alcoholic drinks consumed per drinking occasion, and alcohol consumption over the last week were collected from participants at pre- and post-survey.

Results

A significant difference in pre- to post-survey mean difference of alcohol consumed per occasion was found, with those in the intervention condition displaying a larger mean decrease when compared to controls. No intervention effect was found for units of alcohol consumed per week or for CAGE scores. Sixty-three percent of intervention participants agreed that the feedback provided was useful. Those intervention participants who were above the CAGE cut off were more likely to report that the website would make them think more about the amount they drank.

Conclusions

Delivering an electronic personalised feedback intervention to students via the World Wide Web is a feasible and potentially effective method of reducing student alcohol intake. Further research is needed to replicate this outcome, evaluate maintenance of any changes, and investigate the process of interaction with web-based interventions.

Introduction

The prevalence of alcohol misuse amongst the student population continues to be highlighted (Royal College of Psychiatrists, 2003). Of particular concern are the high levels of heavy episodic or binge drinking within the student population (e.g. White et al., 2006), and the negative effect that this has on academic performance, physical and psychological health, and anti-social and risky behaviour (Wechsler et al., 1994, Wechsler et al., 2002, Ham and Hope, 2003). Levels of heavy drinking during higher education have been well documented in the US (e.g. Wechsler et al., 1994). Although research in the United Kingdom (UK) has not been as wide spread or consistent, surveys of undergraduates also find high levels of alcohol consumption, with 52% of male and 43% of female students drinking above the recommended limits (Gill, 2002).

One intervention approach attracting increasing interest is that of providing personalised feedback. Brief personalised feedback intervention (PFI) programmes focus on an individual's alcohol consumption and provide personalised risk level and alcohol related information (Larimer et al., 2004/2005, Larimer et al., 2004/2005, Saunders et al., 2004, Walters and Neighbors, 2005, White, 2006). Interventions aim to tackle problems early, and there is evidence to suggest that these interventions, that emphasise reduction rather than abstinence, are efficacious with high-risk drinkers (Murphy et al., 2001). The effectiveness of PFI methods has been established using written (White et al., 2006, Murphy et al., 2004), face-to-face (Borsari and Carey, 2001, White et al., 2006) mailed (Collins et al., 2002, Walters et al., 2000), and computer feedback (Neighbors, Larimer, & Lewis, 2004), although Walters (2000) did not find a significant reduction using the postal method at 6-week follow-up.

Extending this, the social norms approach incorporates the use of personalised normative feedback (often alongside other intervention components) to provide corrective information about actual levels of peer alcohol consumption (Lewis and Neighbors, 2006, Neal and Carey, 2004). The tendency for overestimation of peer alcohol intake amongst the student population has been widely found (Perkins & Berkowitz, 1986), and in light of this, personalised normative feedback interventions that focus on the discrepancy between the perceived and actual levels of consumption have been developed (Agostinelli et al., 1995, Neighbors et al., 2004). These personalised interventions tackle the apparent discrepancy by providing information about an individual's alcohol consumption in relation to an analysis of the group norms, and have successfully facilitated behavioural change by targeting and engaging with students who drink at high levels (Walters, 2000, Lewis and Neighbors, 2006).

However, traditional methods of providing personal feedback and motivational interventions are limited in a number of important ways. For example, interventions are limited by their reach and ability to screen a wide range of at risk groups, as recruitment can be difficult in university settings, and may be restricted to one department or school. In addition, interventions can prove costly and highly labour intensive. In response to these issues, research has started to develop web-based interventions that can be completed at a time convenient for the participant, and without the need for staff to actually administer the intervention, or manually collect the data.

Accordingly, studies conducted mainly in North America and New Zealand have begun to investigate the feasibility and effectiveness of delivering personalised feedback and social norms interventions via the internet (e.g. Bendsten et al., 2006, Chiauzzi et al., 2005, Cunningham et al., 2000, Kypri et al., 2004, Linke et al., 2004, Moore et al., 2005, Neighbors et al., 2006). A recent review of web-based alcohol interventions conducted by Bewick, Trusler, Barkham, Hill, Cahill & Mulhern (in press) concluded that current evidence of the effectiveness of e-SBI is promising but inconsistent and that further controlled trials are needed to investigate their efficacy.

The current study therefore aimed to investigate the feasibility and effectiveness of an electronic feedback and social norms alcohol intervention administered to UK students in a randomised controlled trial (RCT). It was hypothesised that this intervention would decrease the units of alcohol consumed over the last week and would also decrease the number of units consumed on an average occasion. The study also aimed to examine gender differences and investigate the effectiveness of the intervention in heavy drinkers.