Sommers 2013.

Methods	Parallel RCT: brief intervention group versus contact control group versus non‐contact control group. ITT: yes.
Participants	Setting: USA, emergency department. Participants: Aged 18 to 44 years admitted to emergency department between 8am and 2am on randomly selected days, drove a motor vehicle two days per week, screened positive on two risky driving and two hazardous drinking items in the three minute screen (at least two of the following in the past month: four standard alcoholic drinks on a typical drinking day, five drinks per occasion for women and six for men, 11 drinks per week for women and 14 for men); excluded if no identifiable residence or contact phone number, or were under arrest at the time of emergency department visit, or were unable to participate in an English language screen due to language or cognitive barriers, or were admitted to the hospital as an inpatient, or had a history of daily alcohol use suggesting alcohol dependence on AUDIT. Number randomised = 476; 70% male; age range = 23 to 34.5 years; 62% non‐White; 57% employed full‐time, 17% employed part‐time, 26% unemployed; 17% less than high school, 38% high school graduate, 45% more than high school. Baseline data not reported for total sample.
Interventions	BI group (N = 150) received (i) a 20 minute SBIRT intervention from a nurse clinician employed for the study, based on FRAMES and reflective listening, on both reducing risky driving practices and reducing alcohol (a drinking typology ‐ dependent, heavy, and low‐risk drinkers), the early consequences of heavy drinking, and tips to cut down on drinking. They received written feedback adapted from the TrEAT protocol on their risky driving and hazardous alcohol behaviours, and the interventionist discussed pros and cons of behaviours and negotiated a contract for change; (ii) a 20 minute booster telephone call 10 to 14 days after discharge the interventionist discussed and renegotiated the contract. CC group (N = 162) received same 20 minute assessment as BIG group but no intervention. NC group (N = 164) received no assessment or intervention.
Outcomes	Number of participants drinking ≥ 5 drinks per drinking day in previous three months, median maximum number of drinks within six hour period, median number of drinks in a typical week, median number of drinks in a typical day, number of participants having ≥ 1 binge per month, number of participants with AUDIT > 10; also driving‐related outcomes. Assessed at 3, 6, 9 and 12 months.
Funding source	This work was funded by a grant (R49/CCR‐523225) from the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention (PI:MSS).
Declaration of interests	Not reported.
Notes	An SBIRT study. USD 20 at enrolment; USD 20 for baseline, three, six, nine month interviews, USD 40 for 12 month interview. Loss to follow‐up: BIG group: 53/150 (35%). CCG group: 53/162 (33%). NCG group: 43/164 (26%). We contacted the authors, who supplied missing data.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation by computer software (p. 1754).
Allocation concealment (selection bias)	Low risk	"Enrolment packets of the same size and thickness were placed in opaque envelopes, sealed and locked in a drawer in consecutive order by participant number" (p. 1754).
Blinding of treatment providers	High risk	The assessment, brief intervention, and booster intervention were delivered by 4 nurse clinicians employed by the study team; i.e. treatment providers had contact with all participants.
Blinding of participants	High risk	No reported attempt to blind participants.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Interviewers were blinded to condition during follow‐up interviews" (p. 1755)
Incomplete outcome data (attrition bias) All outcomes	High risk	Loss to follow‐up > 30% overall.
Selective reporting (reporting bias)	Low risk	Outcomes specified in methods were reported.