Screening and brief alcohol intervention in primary health care: extending the existing evidence base via the UK SIPS trial

There have been many trials of screening and brief alcohol intervention in primary care. Most

have reported positive effects of brief intervention, in terms of reduced consumption in heavy

drinkers. Despite this considerable evidence-base, key questions remain including: the

applicability of the evidence to routine practice; the most efficient screening strategy; and the

required intensity of brief intervention. This trial aimed to evaluate the effectiveness and costeffectiveness

of different models of screening and different intensities of brief alcohol

intervention in routine primary care. GPs and nurses from 24 practices across the North East,

London and South East of England were recruited. Practices were randomly allocated to 1 of

3 intervention conditions: a leaflet-only control group; brief structured advice; and brief lifestyle

counselling. Practices were also randomised to either universal or targeted screening and to

use the modified single item (M-SASQ) or FAST screening tool. To inform the screening

strategy, 190 practitioners were surveyed (prior to trial commencement) to ascertain their

views on which health conditions and/or contexts were most suitable for identifying heavy

drinking. This presentation focuses on screening outcomes including the efficiency of targeted

or universal screening. Practitioners (n = 111) preferred targeted to universal screening (55%

vs. 30%) although 15% felt that a combined approach should be used. The top 4 targeting

conditions were mental health, gastrointestinal problems, hypertension and accidents (rated

important by 87%, 77%, 72% and 63% of practitioners respectively). Of 2291 eligible patients,

30% (n = 900) screened positively for heavy drinking and 84% (n = 754) consented to

participate in the trial. The screen conversion rate (proportion of eligible patients who

screened positively) was significantly higher for targeted screening (36%) compared to

universal screening (26%). Targeting also identified more heavy drinkers who were older

(mean age 52 vs. 49), male (56% vs. 41%), Caucasian (83% vs. 81%) and less well

educated compared to universal screening. The presentation will further examine the

characteristics of patients identified by each screening strategy and the relative efficiency of

the screening tools. Whilst targeted screening was preferred by practitioners and appeared to

be more efficient than universal screening, key groups of patients may be missed by this

strategy.