Evidence
Surveillance decision
We will not update the guideline on behaviour change: individual approaches.
Reasons for the decision
Assessing the evidence
The purpose of this exceptional review was to examine any impact on NICE's guideline on behaviour change following the publication of a Cochrane review on the Effectiveness of brief alcohol interventions in primary care populations (Kaner et al. 2018). This is an update of a 2007 Cochrane review (Kaner et al. 2007). Apart from considering the findings of the previous surveillance review, no additional evidence published since the launch of the NICE guideline in January 2014 was considered by the exceptional review.
Cochrane review summary
Methods
The Cochrane review assessed the effectiveness of brief alcohol interventions compared to no or minimal intervention in reducing alcohol consumption in hazardous or harmful drinkers in general practice or emergency care settings. Randomised control trials (RCTs) and cluster RCTs were included. Trials could include people of all ages who were hazardous or harmful drinkers. People who were alcohol dependent or seeking treatment specifically for an alcohol problem were excluded.
A brief intervention was defined as comprising of a single session and up to a maximum of 5 sessions of verbally-delivered information, advice or counselling that was designed to achieve a reduction in risky alcohol consumption, alcohol-related problems, or both. Control conditions included screening or assessment only, usual care for the presenting condition or written information such as a health or alcohol education leaflet (described as minimal intervention). Extended interventions were also included, defined as those consisting of more than 5 sessions or a total combined session duration of more than 60 minutes. Digital interventions were excluded.
Interventions must have taken place within a primary care or emergency care settings.
The primary outcome measures of interest included self-reported drinking quantity, binge-drinking frequency, drinking frequency and intensity; plus proportion of heavy drinkers and proportion of binge drinkers.
The Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, and 12 other bibliographic databases were searched up to September 2017. Alcohol and Alcohol Problems Science Database (up to December 2003, after which the database was discontinued), trials registries, and websites were also searched. Hand-searching and reference list checks of included studies and relevant reviews were also undertaken.
Standard methodological procedures expected by Cochrane were employed to assess risk of bias, perform meta-analysis and assess quality of evidence (see Cochrane Handbook for Systematic Reviews of Interventions [Higgins and Green 2011]).
Results
Sixty-nine RCTs/cluster RCTs were included, of which 27 studies were identified in the original review. The primary meta-analysis which assessed alcohol consumption at 12 months follow-up (34 trials; n=15,197 participants) found moderate-quality evidence that a brief intervention results in a small but significant reduction in alcohol consumption when compared with minimal or no intervention which was the equivalent of a decrease of 2 to 3 UK standard units of alcohol per week. There was substantial heterogeneity among studies, which was due to the different settings (general practice or emergency care settings), populations, screening instruments used, baseline consumption of alcohol, and the content of interventions and control conditions. Subgroup analyses indicated that both men and women reduced alcohol consumption after receiving a brief intervention, and there was no significant treatment effect by setting or intervention modality (advice-based versus counselling-based intervention content) after adjusting for year of publication in meta-regression analysis.
Additional meta-analyses found moderate-quality evidence that at 12 months' follow-up a brief intervention results in a very small but significant reduction in days of drinking and binge days per week, which was the equivalent of a reduction of 1 day per 2 months when compared with minimal or no intervention, but there was no difference in drinking intensity per day.
A meta-analysis assessing the impact of extended interventions found moderate-quality evidence that, compared with a control condition, there was a significant reduction in frequency of drinking (number of days drinking per week; 2 trials; n=319 participants), but no significant difference in quantity of drinking per week (6 trials; n=1,296 participants) or binges per week (2 trials; n=456 participants) at 12 months' follow-up. There was low-quality evidence that there was no significant difference in alcohol outcomes at 12 months' follow-up in extended compared to brief interventions.
Guideline development
The effectiveness review for the NICE guideline involved a systematic search for RCTs with a sample size greater than 100 which assessed the effectiveness of individual-level behaviour change interventions compared to a control condition in changing either alcohol use, diet, physical activity, sexual behaviour and/or smoking, published from 2003 up to September 2012. It included 26 RCTs assessing the effectiveness of individual-level behaviour change interventions compared to a control condition in changing alcohol use in people who consume a hazardous or harmful amount of alcohol. A meta-analysis indicated that individual-level behaviour change interventions had a very small but significant overall effect on alcohol consumption (and there was low heterogeneity between studies). However when results were assessed according to sub-groups, it was found that compared with usual care, interventions were only effective at changing the alcohol consumption behaviour of pregnant and postpartum women. There was no significant effect on the alcohol consumption behaviours of:
-
Individuals with or at risk for cardiovascular conditions.
-
Emergency Department or hospitalised patients.
-
Primary care patients.
-
University students.
Overall, studies found that brief alcohol interventions (single sessions lasting less than 30 minutes) were no more effective than usual care at altering behaviour (usual care varied from minimal intervention to fairly robust alcohol interventions). Few studies assessed extended alcohol interventions. These were mainly provided to heavy/risky drinking university students with the aim of reducing alcohol consumption. However, across the 5 behaviours of interest, multi-session interventions were more likely to be used and more likely to be effective than the other types of intervention.
Previous surveillance
A surveillance review of the NICE guideline was undertaken in November 2017. Sixty new studies, 2 reports and 10 pieces of ongoing research were identified. While none of the new evidence was assessed as having a substantial effect on the guideline recommendations, it did indicate that there was evidence that addressed a recommendation for research on remotely delivered behaviour change interventions. This led to the commissioning of the in-development NICE guideline on behaviour change: digital and mobile health interventions. No evidence directly relevant to the Cochrane review findings concerning brief and extended behaviour change interventions for people who are hazardous or harmful drinkers was identified in the surveillance review.
Views of topic experts
In this exceptional review we engaged with topic experts who were members of the NICE Centre for Guidelines Expert Advisers Panel to represent their specialty. We received feedback from 3 topic experts, all of whom felt the guideline should not be updated.
The responses from the topic experts suggested that the recommendation on delivering very brief, brief, extended brief and high intensity behaviour change interventions and programmes in the NICE guideline did not need to be updated. Topic experts noted that the findings of the Cochrane review reinforce the existing recommendation concerning delivering brief behaviour change interventions for people who engage in risky behaviours including hazardous or harmful drinking. They did not think that the evidence questioning the effectiveness of extended interventions had an impact on existing recommendations in the NICE guideline due to the very small sample size of studies measuring brief against extended interventions and the low quality of those studies. Topic experts also noted that while dose-response is an important area of research, other issues also impact on effectiveness such as the content of the intervention (behaviour change techniques, psychological constructs addressed) context, person characteristics and characteristics of the intervention provider (skills, attitude, empathic approach, and so on), fidelity of intervention and how effectiveness is evaluated. These issues are addressed within recommendations in the NICE guideline.
Impact
The Cochrane review findings are directly relevant to recommendation 9 in the NICE guideline on delivering very brief, brief, extended brief and high intensity behaviour change interventions and programmes. This states that adults involved in risky behaviours, including alcohol misuse, should be offered interventions to support them in reducing their alcohol intake. Depending on the health and social care staff's background, staff's level of contact with people, and an individual's needs, a very brief intervention, brief intervention, extended brief intervention or a high intensity intervention should be offered.
The new evidence identified showed that when compared with no or minimal intervention, brief alcohol interventions result in a small but significant reduction in alcohol consumption per week, number of days of drinking and binge days per week at 12 months follow-up in people who consume a hazardous and harmful amount of alcohol. The evidence indicates that more intensive interventions may not provide any additional benefit over and above that which is achieved following a brief alcohol intervention (as defined within the Cochrane review). These findings appear to support the recommendation to provide brief interventions, but question whether extended interventions are appropriate given that they appear to be no more effective than lower intensity interventions at changing alcohol consumption.
It should be noted that the definition of a brief alcohol intervention within the Cochrane review would encompass a very brief intervention, brief intervention or extended brief intervention as defined within the NICE guideline; and that an extended alcohol intervention as defined within the Cochrane review would most closely represent a high intensity intervention within the NICE guideline: typically these last more than 30 minutes and are delivered over a number of sessions. The current recommendation in the NICE guideline states that behaviour change service providers and practitioners should be encouraged to provide high intensity interventions for people they regularly work with under specific circumstance: for those who have been assessed as being at high risk of causing harm to their health and wellbeing and/or have a serious medical condition that needs specialist advice and monitoring and/or have not benefited from lower intensity interventions.
The conclusions in the Cochrane review concerning the (lack of) impact of extended (high intensity) versus brief alcohol interventions was made on low-quality evidence from a small number of studies that assessed extended interventions. In addition, the remit of the NICE guideline was to provide recommendations for a range of behaviours, not just alcohol use, and to consider more broadly what the common components are of effective behaviour change interventions to ensure that interventions are appropriately developed and delivered.
After taking into account the new evidence and views of topic experts, we have concluded that the new evidence does not have an impact on the recommendations within the NICE guideline. For this reason, we will not update the guideline at this time.
Overall decision
After considering the impact of the evidence on current recommendations, we decided that no update is necessary at this time.
See how we made the decision for further information.
This page was last updated: