Please leave your comments for the authors below Please see the attached Word file for a better formatted version of the below comments.

Thank you for the opportunity to review this revision. I congratulate the authors for making use of the feedback and improving their manuscript. I am largely satisfied with the changes and I feel that the majority of the issues have been addressed well. I have only minor issues that I would like the authors to address in their final version of the paper. Well done.

Minor issues Page 4, Line 28 – “Red Book” Since the review, the 9th edition of the Red Book has been released. The authors may want to update the reference.

Page 15, Lines 53-57 – Injuries and absenteeism I have commented on this in the previous review, and I note the author’s response. I would counter that injuries as seen in emergency departments do not translate to injuries seen in general practice. The majority of injuries seen in general practice would not be due to alcohol misuse. The context is very different – ED is not general practice. Also, although individuals with alcohol use disorders may have frequent work absenteeism, the reverse logic does not hold. The majority of individuals coming to see a GP for an illness certificate (repeat or otherwise) do so not because of alcohol use problems, but because of chronic illness (both physical and mental). The questionnaire did not ask GPs whether alcohol use disorders are associated with work absenteeism. Rather, it asked whether requests for frequent illness certificates would prompt questioning about alcohol, amongst a list of other scenarios. The updated language in this section of the paper has been improved, but in my view, the rationale underlying the statement “Injuries and work absenteeism are very common outcomes of harmful drinking,[23, 24] yet ‘suspicious or frequent injuries’ and ‘frequent requests for sickness certificates’ were ranked in the top three presentations by 20% or fewer of GPs in our survey” is weak. I agree with the overarching thesis of this section of the paper, that GPs use clinical judgements of individuals to identity risky drinking (and so they should!) and tend not to use more systematic or health screening approaches. The use of the results on injuries and work absenteeism, in my view, do not especially strengthen or support this idea. I do not consider this to be a critical issue, and the existing text is publishable. However, I feel this section would be improved and streamlined by simply removing the interpretation of results on injuries and work absenteeism as a noteworthy point.

Page 17, Lines 41-43 – AUDIT-C and WHO I’m not sure it would be accurate to describe the AUDIT-C as a WHO developed modification of the AUDIT. My understanding is that the consumption items of the AUDIT was first tested as a brief screening tool by Bush and colleagues (who labelled it the “AUDIT-C”) in a Veteran Affairs Population: http://archinte.jamanetwork.com/article.aspx?articleid=208954 Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA, for the Ambulatory Care Quality Improvement Project (ACQUIP). The AUDIT Alcohol Consumption Questions (AUDIT-C): An Effective Brief Screening Test for Problem Drinking. Arch Intern Med. 1998;158(16):1789-1795. doi:10.1001/archinte.158.16.1789.