22  Alcohol, AUDIT-C, and the fourteen-unit line

Alcohol is in a dental guideline for one reason: it causes oral cancer, and combined with tobacco it does so far more than either alone. The recommendations that follow are about a dentist asking you how much you drink.

This chapter is short, because the pattern is one the book has already established twice, and because I have less to add than in Part II.

The advice

Three Strong recommendations, structured like the tobacco pathway:

Ask. Use the AUDIT-C tool (or similar) to assess a patient’s level of risk of alcohol harm by completing 3 consumption questions.

Strength of recommendation: Strong

Advise. If AUDIT C score is 4 or below, give positive feedback and encourage your patient to keep their drinking at lower risk levels. If score is 5 to 10, give brief advice to encourage a reduction in alcohol consumption and reduce the risk of alcohol harm.

Strength of recommendation: Strong

Act. Feed back to the patient that their level of drinking is putting them at risk of developing a range of health problems (including cancers of the mouth, throat and breast) and this increases the more you drink and the more frequently you drink. Highlight ‘low risk’ guidelines for alcohol consumption from UK Chief Medical Officers: • to keep health risks from alcohol to a low level, it is safest not to drink more than 14 units a week on a regular basis […] Give a leaflet

Strength of recommendation: Strong

— Delivering Better Oral Health, chapter 2, table 3c (1)

The Act row runs to three bulleted sub-points; I have kept the first, which is the one everybody quotes, and marked the cut. The two I left out are about spreading drinking over three or more days and having drink-free days.

Notice the wording of that first bullet, because it is not the wording the fourteen-unit rule is usually given in. It is not “the limit is 14 units.” It is that to keep health risks to a low level, it is safest not to exceed it. That is a statement about a risk gradient with no threshold in it, and an earlier draft of this chapter quoted it as though it were a limit, which is precisely the transformation this chapter goes on to complain about.

What the guideline says its evidence is

Recommendation based on moderate certainty evidence from a systematic review showed that brief interventions can reduce alcohol consumption in those drinking hazardous or harmful amounts of alcohol when compared to minimal or no interventions.

— Delivering Better Oral Health, chapter 13 (1)

That evidence statement attaches to the Ask row. The Advise and Act rows carry no evidence statement at all in chapter 13.

The same structure as tobacco, one step weaker

Chapter 21 set out three conditions under which a component with no direct evidence can legitimately inherit a Strong label: it is logically necessary to a pathway whose downstream steps work, the burden is trivial, and the harm is nil. Alcohol brief intervention meets the first two comfortably. Screening questions cost nothing and you cannot advise without asking.

But there is a real difference from tobacco, and it is worth being precise about because it is the only thing this chapter adds.

The evidence is about brief interventions in general, not in dental settings. The tobacco chapter had a Cochrane review of cessation delivered by dental professionals, 20 trials, in dental clinics. Here the moderate-certainty evidence is for brief alcohol interventions as a class, largely from primary care. Whether a conversation about drinking works the same way when it happens with a dental nurse holding a suction tube is an open question, and it is precisely the kind of setting mismatch that Chapter 12 flagged for supervised school programs.

DBOH does not disclose this one the way it disclosed the school-program gap. Chapter 13 says “a systematic review showed that brief interventions can reduce alcohol consumption” without noting where those interventions were delivered. Given how well the same chapter handles the equivalent gap for mouth rinses, this looks like an inconsistency in drafting rather than a considered judgment.

The fourteen units

The number in the Act row is not DBOH’s. It comes from the UK Chief Medical Officers’ guidelines, and the guideline correctly presents it as such: “Highlight ‘low risk’ guidelines for alcohol consumption from UK Chief Medical Officers.”

I am not going to audit the CMO guidance. It rests on a large body of epidemiology about all-cause mortality and a wide range of cancers, it was developed through its own process, and doing it justice would be a different book by a different author. Relaying another body’s guidance, clearly attributed, is a reasonable thing for a dental guideline to do, and it is the same legitimate borrowing as Chapter 14.

What I will note is that the oral cancer risk is the part relevant to a dental setting, and the 14-unit figure was not derived from it specifically. Nor is it a threshold, whatever the phrase “the fourteen-unit line” in this chapter’s title suggests. The CMO wording is a statement about keeping risk low, not about a point at which risk begins, and the difference is the same one this book keeps finding: a graded, hedged statement in the source becomes a bright line by the time it reaches the person it is aimed at.

The pregnancy row

For those who are pregnant or think they could become pregnant, the safest approach is not to drink alcohol at all, to remove the risk of alcohol-related harm to the baby.

Strength of recommendation: Good practice

A well conducted systematic review highlights the paucity of evidence evaluating light drinking in pregnancy compared with abstinence. Based on this research, the CMO’s guidance supports a ‘precautionary’ approach.

— Delivering Better Oral Health, chapter 13 (1)

This is a model row and I want to hold it up as one. It states the evidence gap, names the reasoning as precautionary rather than empirical, attributes the recommendation to its source, and labels it Good practice rather than Strong.

Every criticism in this book is a request for other rows to be written like this one.

PubMed, 21 August 2026. I did not obtain the systematic review of brief alcohol interventions that DBOH cites; it was not among the supplied documents, and this chapter relies on DBOH’s characterization of it.

I have not audited the UK CMO alcohol guidelines, which are the source of the 14-unit figure. That is outside this book’s scope and I say so rather than implying coverage.

Full record in appraisals/searches/ch22-alcohol.md.

Verdict

Certainty of evidence
Moderate that brief interventions reduce alcohol consumption in hazardous or harmful drinkers, per DBOH’s account of a review I have not read, in settings that are mostly not dental. None stated for the Advise and Act steps. Explicitly paucity for light drinking in pregnancy, and the guideline says so.
Directness to the advice as worded
Good on intervention and outcome, unassessed on setting. The gap between brief interventions in general and brief interventions in a dental chair is real and undisclosed, which is the one thing this chapter holds against the guideline.
Is the strength label defensible?

Yes, on the tobacco reasoning. Asking about alcohol costs nothing, harms nobody, and is a necessary entry to advice that works. The pathway structure justifies the label even where individual steps lack their own evidence.

The improvement wanted is one clause: name the setting the evidence came from, as the same chapter does for supervised school mouth rinse programs.

What would change my mind
A trial of alcohol brief intervention delivered in dental practice, which would close the setting gap. And, for the dental purpose specifically, evidence on whether reducing consumption from hazardous levels reduces oral cancer incidence, which is the outcome the recommendation is ultimately aimed at and which nobody has demonstrated within this evidence base.

What this means for you

If a dentist asks how much you drink, they are not being nosy. Alcohol and tobacco together are the main modifiable causes of oral cancer, and the mouth is where the consequence shows up first.

The brief conversation that follows has moderate-certainty evidence behind it, from other healthcare settings. Fourteen units a week is the UK Chief Medical Officers’ threshold, not a dental one, and it is about your health generally rather than your mouth specifically.