A systematic literature review was conducted by one of the authors (A.H.K) of websites, books and journal articles for strategies about how to help someone who may be developing, or may have, a drinking problem. This involved a comprehensive internet search using Google search engines (http://www.google.com, http://www.google.co.uk and http://www.google.com.au). The following search terms were entered into each: alcohol or alcoholic and intoxication, alcohol poisoning, binge drinking, alcohol abuse, alcohol dependence. The first 50 sites for each set of search terms were examined for strategies about how to help someone with a drinking problem. This technique yielded 250 sites per search engine. Any links that appeared on these web pages that were thought may contain useful information were followed. Relevant journal articles published between January 1997 and December 2007 were sought from PsycINFO and PubMed. This yielded 997 and 1572 articles respectively, which were then scanned for any relevance to first aid. The 50 most popular books on the Amazon website published from 1980 onwards were also selected and reviewed. Following this extensive review of the literature, suggestions for first aid actions were obtained from approximately 45 websites, 3 books and 7 journal articles. The majority of first aid actions came from websites, as few books and journal articles focused on pre-clinical interventions.
The information gathered from these sources was analysed by one of the authors (A.H.K) and written into first aider action statements that could be presented to the panels for rating. These statements were first presented to a working group, who screened the items to ensure they fitted the definition of MHFA for problem drinking, were comprehensible and had a consistent format (with the aim of remaining as faithful as possible to the original meaning and wording of the information). After several draft surveys, the group identified 285 items that formed the Round 1 survey. The Round 1 survey was organized around five main sections. (1) The problem drinking section included items about recognizing and understanding problem drinking, approaching the person, managing the person's unwillingness to change, and facilitating and managing resistance to seeking professional help. (2) The low-risk drinking section included items about understanding low-risk drinking, encouraging the person to drink at lower levels, providing practical tips on doing so, encouraging other supports, and dealing with social pressure to drink. (3) The alcohol intoxication section included items about recognizing and understanding alcohol intoxication, helping an intoxicated person, talking to them, getting them home, and managing aggression. (4) The emergencies related to alcohol intoxication section included items about general principles of assisting in an emergency, seeking medical help, and managing vomiting, drowsiness, alcohol poisoning and other alcohol-related emergencies. (5) The alcohol withdrawal section included items about severe alcohol withdrawal. Comment boxes were included in the Round 1 survey, which allowed panel members to comment and give feedback after each section.
Consumers, carers and clinicians with expertise or experience in problem drinking were recruited from Australia, Canada, Ireland, New Zealand, the United Kingdom, and the United States. Clinical experts (panel one) approached were international authorities on problem drinking, as well as experienced senior clinicians working within alcohol and other drug treatment settings. Clinical experts were recruited through direct email contact with members of the international editorial boards of the top seven peer-reviewed substance use journals, addiction specialist colleges and societies, and major addiction treatment centres in each country. Consumers (people with a past history of problem drinking) and carers (people with experience caring for someone with problem drinking) were integrated into a second panel as there were not sufficient numbers to divide them into separate panels (Delphi convention recommends a minimum of 15 members per panel ). Consumers and carers were recruited by distributing information about the study to consumer and carer organizations associated with alcohol and drug and/or mental health problems in each country. Consumers and carers with experience in an advocacy role were targeted, to ensure that participants had an understanding of problem drinking beyond their own personal experience. Consumers and carers who had authored books about their experience with problem drinking were also invited to participate. No attempt was made to make panels representative. The Delphi method does not require representative sampling; it requires panel members who are information- and experience-rich.
Ninety-nine panel members were recruited from Australia (14 consumers/carers, 39 clinicians), Canada (6 consumers/carers, 6 clinicians), Ireland (1 clinician), New Zealand (1 consumer, 2 clinicians), the United Kingdom (8 consumers/carers, 9 clinicians) and the United States (5 consumers/carers, 8 clinicians). Fifty-three participants were female (68% of the consumers and carers, 46% of the clinicians). The age of consumers and carers ranged from 18-60+ years (median age category was 50-59 years), while the age of clinicians ranged from 30-60+ years (median age category was 40-49 years).
Once participants agreed to participate in the study, they were given the option of completing the surveys online (using SurveyMonkey, http://www.surveymonkey.com) or via postal mail. The study was approved by the Human Research Ethics Committee at the University of Melbourne.
The Delphi process
The Delphi process was used to survey expert opinion. This was achieved by asking panel members to rate the importance of potential first aid strategies, bearing in mind that a first aider was a member of the general public and therefore did not necessarily have a medical or clinical background. The rating scale used was essential, important, don't know/depends, unimportant and should not be included. Not qualified to answer was included in the rating scale in section 4 of the survey. On completion of each round (there was a total of three rounds), the survey responses were analysed by obtaining percentages for the consumer/carer and clinician panels for each item. The following cut-off points were used:
Criteria for accepting an item
Criteria for re-rating an item
If 80% or more of the panel members in only one group rated an item as essential or important as a MHFA guideline for problem drinking, we asked all panel members to re-rate that item in the next round.
If 70%-79% of panel members from both groups rated an item as essential or important, we asked all panel members to re-rate that item.
Items were re-rated once only. If an item was not endorsed after two rounds it was excluded from the guidelines.
Criteria for rejecting an item
After each Round, each panel member was sent a report describing how the results had been analysed and listing all items endorsed in that Round as MHFA guidelines. The report also contained items that required re-rating, accompanied by a summary (as a percentage) of each panel's ratings and the panel member's previous rating for each item. In light of this feedback, panel members were asked to maintain or modify their original ratings in the next Round. In addition, the Round 1 report also contained new items generated through panel members' comments to be rated for the first time.
To analyse the comments that panel members had written during the Round 1 survey, one of the authors (A.H.K) reviewed the comments and wrote them up as first aid strategies. The working group evaluated the suggested strategies to determine whether they were original ideas that had not been included in the Round 1 survey, and whether they met the criteria for a MHFA item. Any strategy that was judged by the group to be an original idea was included as a new item to be rated in the Round 2 survey.