Table 2.

Details of Included Interventions

Study ID, country of origin and setting	Details of Sample	Allocation of Participants	Methods	Details of Intervention(s)	Mode of Delivery	Interventionist	Assessment periods	Outcome Measures and findings
Gordon et al. (2003) USA Primary care offices	n = 45 Age > 66 years old Convenience sample gained from waiting rooms of primary care centres. All were hazardous drinkers Randomly allocated to one of three groups: Motivational Enhancement (ME) n = 18 Brief Advice (BA) n = 12 Control group - care as usual (SC) n = 12	Allocated randomly to receive one of two interventions, or standard care as usual.	Enrolled and completed questionnaires at baseline.	ME – Included feedback, goal setting and consequences. First session lasted around 60 min, with two booster sessions of around 10–15 min. BA – one 10–15 min sessions focussing on advice Control group (usual care)- were given care as usual, no discouragement against talking about alcohol.	Face to face	Trained research interventionists	Baseline, 1, 3, 6, 9 and 12 months	Time Line Follow Back (TLFB) questionnaire used measure quantity and frequency of alcohol consumption. Also includes a tool to measure amount of drinks consumed in the last month, amongst other frequency measures. All measures included were self-reports. Each group showed decreases in alcohol consumption measures over time, however these were not statistically different to those who received standard care.
Hansen et al. (2011) Denmark No information on specific setting of intervention.	n = 772 Age > 50 years old Used AUDIT tool to identify only those who were heavy drinkers. Intervention Group n = 391 (n = 316 at 12 months) Control Group n = 381 (n = 300 at 12 months)	No information given as to how participants were randomised into groups	Information gathered on alcohol use at baseline.	BMI – Consisted of a conversation based on the principles of MI, designed to motivate individuals to change behaviour through open ended questions. Also were given an information sheet with information about local alcohol treatment and a brief telephone booster 4 weeks later. Control group received same leaflets about alcohol and local treatment. A pure control group was not included.	Face to Face, telephone booster	Nurses and Postgraduate Students	Baseline, 6 and 12 months.	Outcome measure was drinks per week. This particular study did not find any significant intervention effect on drinks per week.
Kuerbis et al. (2015) USA Primary care clinics	n = 86 Used CARET to ensure only selected at risk drinkers. > 50 years, mean age of just under 65 years old Intervention group n = 44 Control group n = 42	Participants were randomly assigned to the intervention or control group.	Participants completed assessment tool at baseline.	Intervention Group – received personalised mailed feedback outlining the risks specific to their alcohol use. Also received the NIH Rethinking Drinking: Alcohol and Your Health booklet Control group – did not receive anything	Mailed to Participants	n/a	Baseline and 3 months	CARET was used to measure alcohol risk score at 3 months. Significant reductions were found in the intervention group for binge drinking, alcohol use with a medical or psychiatric condition and alcohol with symptoms of a medical or psychiatric condition. Intervention groups were 72% less likely to be an at risk binge drinker and 92% less likely to be at risk due to a medical or psychiatric condition.
Fleming et al. (1999) USA Community based primary care practices	n = 146 for full 12 months Age > 65 years old Used Health Screening Survey (HSS) to include only those who were problem drinkers Intervention group n = 87 Control group (usual care) n = 71	Participants were randomly allocated to receive either the brief intervention (BI) or care as usual.	Participants completed assessment tool at baseline.	BI group – received booklet on general health and were also scheduled to see their personal physicians. Used BI protocol including a workbook containing feedback on individual’s behaviours and other educational resources. Had 2 × 15 min appointments, one month apart consisting of the intervention and then a reinforcement session. Control group (usual care) – only received a general health booklet.	Face to Face and telephone	Physician trained in internal or family medicine	Baseline, 3, 6 and 12 months. Family members contacted at 12 months for back up on results self-reports	Used drinks per week, levels of binge drinking and excessive alcohol use. The intervention group maintained lower levels of drinking throughout and reduced their weekly alcohol use. They also self-reported reduced amounts of binge drinking and excessive levels of drinking. These were statistically significant.
Ettner et al. (2013) USA Community based practice, 7 clinics	n = 1186 Age > 60 years old Used CARET to ensure only selected at risk drinkers. Intervention group n = 546 Control group (usual care) n = 640	Patients were randomly assigned to a group dependant on which group the physician they saw was assigned to.	Participants completed CARET and Alcohol- Related Problems Survey at baseline to determine risk and frequency of alcohol use, whether they had discussed their alcohol use with their physician and self-reported health care use.	Intervention group – used project SHARE which included personalised reports, education material, telephone counselling and physician advice. Control group (usual care) – received care as usual. Alcohol discussions were not discouraged	Face to Face and telephone	Physician for intervention and health educator for telephone counselling.	Baseline, 3, 6 and 12 months.	Alcohol- Related Problems Survey, CARET Primary outcome measure - Whether individual continued to be at risk. Secondary outcome measure – score on CARET, drinks per week, amount of discussions with physician and use of health care including trips to emergency care etc. All measures were self-reports Greater declines and fewer at risk statistically in intervention group compared to control.
Moore et al. (2011) USA Primary care sites	n = 631 Age > 55 years old All participants were at risk, as identified by the CARET tool. Intervention group n = 310 Control group n = 321	Participants were randomly assigned to either the intervention group or the control group.	Patients were at risk of alcohol misuse as determined by CARET at baseline.	Intervention group = received a multi-faceted intervention that included a personalised report, a booklet on alcohol and ageing, a diary to log levels of drinking, advice and telephone counselling Control group = only received a booklet on healthy behaviours	Face to face and telephone	Research assistant and primary care provider	Baseline, 3, 12 months	CARET, number of drinks consumed in the past 7 days and heavy drinking in the past 7 days. All measures were self-reports. 3 At 12 months, intervention group did not have lower levels of at risk drinking, however did reduce levels of alcohol consumption. Results were statistically significant, but researchers suggest that they may not provide clinical significance.
Fink et al. (2005) USA Community primary care practices	n = 665 Age > 65 years old Participants were eligible to take part if they had consumed more than one alcoholic drink in the last three months. Split into three groups, two experimental and one control. Experimental 1 (both physicians and patients received reports) n = 212 Experimental 2 (only patients received reports, physicians did not) n = 245 Control group (usual care) n = 208	Participants were randomly assigned to intervention or control groups.	Participants took part in either one of three groups, two being experimental and one being control.	Experimental group 1 – physician and patients received reports on the patients’ alcohol use, risks and problems. They also received personalised educational tools. Experimental group 2 – Only the patients received their reports, the physicians did not. They also received the personalised educational tools. Control group (usual care) – were not informed on their individual risks, nor did they receive any educational tools	Computerised report	Physician	Baseline and 12 months	Found reductions in hazardous drinking, reductions in harmful drinking and maintenance of non-hazardous drinking in both experimental groups when measured with CARPS – baseline, 12 months later. Measures were self-reports. Both experimental groups had lower risk drinking compared to control. Patient only reports led to reduced harmful drinking and less hazardous drinking. Combined reports only decreased total consumption.