Who this is for
Every adult, and again from time to time. USPSTF (2018) recommends screening all adults 18 or older in primary care, including pregnant women, and it does not exempt people who drink lightly. The WHO AUDIT manual (2001) gives the reason for screening everyone in its own words: "If health workers screen only those they consider most likely to have a 'drinking problem', the majority of patients who drink excessively will be missed." NIAAA's Core Resource (2025) suggests clinicians make it routine and reassure patients that "we ask everyone."
What counts as too much depends on whose limits you read, and this page keeps them apart. The USPSTF 2018 statement quotes NIAAA's 2005 limits: more than 4 drinks per day or 14 per week for healthy men aged 21 to 64, and more than 3 per day or 7 per week for all adult women and for men 65 or older. NIAAA's own current page (updated January 2026) defines heavy drinking as 5 or more drinks on any day or 15 or more per week for men, and 4 or more on any day or 8 or more per week for women, with no age split at 65. Whichever set you sit under, the screening questions are the same; the limits are what the clinician measures your answers against.
If you are pregnant or trying to become pregnant, the threshold is zero. USPSTF (2018) states that any alcohol use is considered unhealthy in pregnant women, and ACOG's Committee Opinion 711 (2017, reaffirmed 2026) says screening should happen at the first prenatal visit and be universal. For adolescents aged 12 to 17, USPSTF says the evidence is insufficient either way.
Where the recommendations diverge
- USPSTF 18 and older
- USPSTF 12 to 17 12 to 17
- ACOG Not age-bounded Not age-bounded
- ACOG Not age-bounded Not age-bounded
- VA / DOD Not age-bounded Not age-bounded
- WHO Not age-bounded Not age-bounded
The USPSTF recommends screening for unhealthy alcohol use in primary care settings in adults 18 years or older, including pregnant women, and providing persons engaged in risky or hazardous drinking with brief behavioral counseling interventions to reduce unhealthy alcohol use.
Our note: The Grade B is for screening plus brief counseling of people who screen positive, not for the questions alone. Interval, from the same page's Clinical Considerations: "The USPSTF did not find adequate evidence to recommend an optimal screening interval for unhealthy alcohol use in adults." No upper age is stated, so stop_age is left empty. The statement replaced the 2013 recommendation on "alcohol misuse". A draft update posted August 5, 2025 carries the same population and grade and states it is consistent with this statement; no final had been posted as of 2026-09-10.
The USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening and brief behavioral counseling interventions for alcohol use in primary care settings in adolescents aged 12 to 17 years.
Our note: An I statement is not a recommendation against screening; it means the USPSTF could not weigh benefits against harms from the evidence it reviewed. The same page states that any alcohol use is considered unhealthy in adolescents and names the CRAFFT instrument, recommended by NIAAA and the American Academy of Pediatrics, for risky substance use in this age group. The August 5, 2025 draft repeats this I statement unchanged.
Before pregnancy and in early pregnancy, all women should be routinely asked about their use of alcohol and drugs, including prescription opioids and other medications used for nonmedical reasons.
Our note: Committee Opinion 711, Obstet Gynecol 2017;130:e81-94, reaffirmed 2026. The document prints August 2017 with no day given, so we record the first of that month. Visit timing from the same opinion: "Screening for substance use should be a part of comprehensive obstetric care and should be done at the first prenatal visit in partnership with the pregnant woman." ACOG attaches no grade, names no alcohol instrument and sets no threshold. Its alcohol-specific opinion, Committee Opinion 496 (2011), the document USPSTF cites, has been withdrawn by ACOG.
All patients should be routinely asked about their use of alcohol, nicotine products, and drugs, including prescription opioids and other medications used for nonmedical reasons.
Our note: Committee Opinion 762, joint with the American Society for Reproductive Medicine, published online December 20, 2018, Obstet Gynecol 2019;133:e78-89, reaffirmed 2024. No grade, no instrument, no ages and no threshold are stated. Being "routinely asked" is a universal-inquiry position rather than an instrument-based screening recommendation, which is why this row carries no AUDIT-C or SASQ cutoff.
For patients in general medical and mental healthcare settings, we recommend screening for unhealthy alcohol use periodically using the three-item Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) or Single Item Alcohol Screening Questionnaire (SASQ).
Our note: VA/DoD Clinical Practice Guideline for the Management of Substance Use Disorders, Version 5.0. The page footers print August 2021 with no day, so the first of the month is recorded, the same convention this page uses for ACOG 711. The recommendation is categorized "Not reviewed, Amended"; the 2015 edition said "annually". VA/DoD scores a positive AUDIT-C at 5 or higher for both sexes, chosen to cut false positives.
All patients should be screened for alcohol use, preferably annually.
Our note: AUDIT: The Alcohol Use Disorders Identification Test, Guidelines for Use in Primary Care, second edition (WHO/MSD/MSB/01.6a). The manual carries the full ten-item AUDIT and assumes a 10 gram standard drink, smaller than the US 14 gram drink. No later edition exists on WHO IRIS; the year is the document's own and no month is printed, so the last day of 2001 is recorded.
Why they differ
Every body that has published on this says adults should be screened. They part ways on how often, on what score counts as positive, and on which drinking limits define the problem. Each of those changes whether a given person is told they drink too much, so this page keeps every position under its own name and date.
How often. Three bodies, three answers:
- USPSTF (2018): "The USPSTF did not find adequate evidence to recommend an optimal screening interval for unhealthy alcohol use in adults." The August 2025 draft repeats the finding.
- VA/DoD (2021) recommends screening "periodically" with the AUDIT-C or the single question, graded "Strong for". The word is deliberate: the 2015 edition said "annually", and the 2021 edition carried the recommendation forward with that one change, categorized as "Not reviewed, Amended". The USPSTF page still reports that the VA "recommends annual screening" because it cites the 2015 edition.
- WHO (2001), in the AUDIT manual: "All patients should be screened for alcohol use, preferably annually."
What counts as positive on the AUDIT-C. The three questions score 0 to 12. USPSTF (2018) reports the instrument's accuracy "at the standard cutoffs of 3 or higher (female) and 4 or higher (male)". Those cutoffs were validated by Bradley and colleagues in 2007, in 392 men and 927 women at a family practice clinic, where 4 or higher in men and 3 or higher in women simultaneously maximized sensitivity and specificity, the same thresholds as previously published VA studies. VA/DoD (2021) sets its positive at 5 or higher for men and women alike, and says why: "This higher AUDIT-C score was selected to minimize the false-positive rate and to target implementation efforts." Follow-up of scores under 5 is left to the provider. NIAAA's Core Resource (2025) prints the questions and no cutoff at all. So a man scoring 4, or a woman scoring 3 or 4, is positive under the cutoffs USPSTF reports and not positive under VA/DoD.
Which drinking limits. USPSTF's 2018 statement defines risky use by quoting NIAAA's 2005 Clinician's Guide: 4 drinks per day or 14 per week for men 21 to 64, 3 per day or 7 per week for women and for men 65 or older. NIAAA's current drinking-patterns page (updated January 2026) defines heavy drinking as 5 or more on any day or 15 or more per week for men and 4 or more on any day or 8 or more per week for women, with no age split. The USPSTF draft posted August 5, 2025 drops the 2005 limits entirely and instead cites CDC's binge definition, 5 or more drinks for men or 4 or more for women during an occasion, and the USDA and HHS definition of moderation, no more than 2 drinks in a day for men and 1 for women. The single screening question drifted the same way: the 2018 statement applies the 4-drink threshold to women and all adults older than 65, while the 2025 draft and NIAAA's Core Resource say 4 for women and 5 for men with no age qualifier. The draft is not final. The 2018 statement is the operative recommendation, and the draft states that it "updates and is consistent with the 2018 USPSTF recommendation".
Pregnancy. USPSTF (2018) includes pregnant women in its Grade B population and states that "Any alcohol use is considered unhealthy in pregnant women and adolescents." ACOG's current positions come from its opioid and prepregnancy committee opinions rather than from an alcohol-specific document, because Committee Opinion 496 (2011), the alcohol opinion USPSTF cites, has been withdrawn. Committee Opinion 711 (2017, reaffirmed 2026) puts screening at the first prenatal visit and says "it is essential that screening be universal"; Committee Opinion 762 (2018, reaffirmed 2024) says all patients should be routinely asked about alcohol; and the Committee on Ethics (Opinion 633, 2015, reaffirmed 2021) says routine screening for substance use disorder "should be applied equally to all people, regardless of age, sex, race, ethnicity, or socioeconomic status." None of the three names an instrument or a number, which is why the ACOG rows carry no grade and no threshold.
What the test involves
Your clinician asks the questions, in person or on an intake form. USPSTF (2018) says the AUDIT-C takes 1 to 2 minutes and the single question takes less than 1 minute. The sequence USPSTF describes is a brief, sensitive screen first and a longer, more specific instrument to confirm: "When patients screen positive on a brief screening instrument (eg, SASQ or AUDIT-C), clinicians should ensure follow-up with a more in-depth risk assessment to confirm unhealthy alcohol use and determine the next steps of care." The longer instrument it names is the full ten-question AUDIT, which adds seven questions on signs of dependence and problems from drinking and takes approximately 2 to 5 minutes.
You can take the AUDIT self-test on this site before the appointment. It is the WHO instrument, scored on the WHO manual's zones, where a score of 8 or more is the level at which the manual moves from alcohol education to simple advice. Two things to know about it. USPSTF positions the full AUDIT as the confirmation step after a positive brief screen, and says a lower cutoff of 3, 4, or 5 may be used if it is the first screen. And the WHO instrument assumes a 10 gram drink, so counting 14 gram US drinks against it runs toward over-flagging rather than under-flagging.
The CAGE questionnaire is the one to skip. USPSTF (2018): "The Cut down, Annoyed, Guilty, Eye-opener (CAGE) tool is well known but only detects alcohol dependence rather than the full spectrum of unhealthy alcohol use." NIAAA's Core Resource (2025) calls it outdated for the same reason. For pregnancy, USPSTF names four validated tools by acronym, TWEAK, T-ACE, 4P's Plus and NET; ACOG's Committee on Ethics (2015, reaffirmed 2021) says routine screening can be done by validated questionnaires or conversations and that routine laboratory testing of biologic samples is not required.
If the result is abnormal
A positive screen is not a diagnosis. VA/DoD (2021) puts it in one sentence: "Most screen-positive patients will not have AUD and should not be given a diagnosis solely based on screening results". The follow-up assessment separates risky drinking, which is drinking above the limits without meeting criteria for alcohol use disorder, from the disorder itself.
For risky drinking, the graded response is brief counseling. USPSTF's Grade B (2018) covers "providing persons engaged in risky or hazardous drinking with brief behavioral counseling interventions to reduce unhealthy alcohol use." VA/DoD (2021) suggests "a single initial brief intervention regarding alcohol-related risks and advising to abstain or drink within established limits for daily and weekly consumption", a recommendation it downgraded from "Strong for" in 2015 to "Weak for" in 2021. The WHO manual (2001) scales the response to the AUDIT score: alcohol education at 0 to 7, simple advice at 8 to 15, simple advice plus brief counseling and continued monitoring at 16 to 19, and referral to a specialist for diagnostic evaluation and treatment at 20 to 40.
The screening recommendations stop there. The USPSTF draft (2025) states that it does not apply to people already diagnosed with unhealthy alcohol use or seeking evaluation or treatment for it. If the assessment finds alcohol use disorder, what follows is diagnosis and treatment, which is separate guidance and outside this page.
More on this screening
Frequently asked questions
I have one drink most evenings. Will I screen positive?
It depends on whose cutoff your clinician uses, which is the point of this page. Scored on the VA/DoD (2021) AUDIT-C key, drinking 4 or more times per week scores 4 on the first question, 1 or 2 drinks on a typical day scores 0 on the second, and never having 6 or more (men) or 4 or more (women) on one occasion scores 0 on the third, for a total of 4 (our arithmetic on their key). That is positive at the standard cutoffs USPSTF (2018) reports, 4 for men and 3 for women, and not positive at VA/DoD's 5. On the single question, VA/DoD counts any occasion of 5 or more drinks (men) or 4 or more (women) in the past year as positive, so a pattern of one drink a night with no heavy days is negative there.
How often should I be screened?
No single answer exists, and the bodies say so. USPSTF (2018) did not find adequate evidence to recommend an optimal interval. VA/DoD (2021) says periodically, having dropped the word annually from its 2015 edition. The WHO AUDIT manual (2001) says preferably annually. The practical reading is that a routine visit is a reasonable time to be asked, and no body sets a longer gap than a year.
Is there a blood test for this?
None of the recommendations on this page is a blood test; all of them are questions. ACOG's Committee on Ethics (Opinion 633, 2015, reaffirmed 2021) says routine screening for substance use disorder can be done by validated questionnaires or conversations and that routine laboratory testing of biologic samples is not required. The related ALT page covers what a liver enzyme result can and cannot tell you; it is not a screening test for alcohol use under any of these bodies.
I am pregnant or planning to be. What changes?
The limit becomes zero and the timing becomes specific. USPSTF (2018) states that any alcohol use is considered unhealthy in pregnant women, and its Grade B recommendation includes them. ACOG's Committee Opinion 711 (2017, reaffirmed 2026) says screening for substance use should be done at the first prenatal visit, and Committee Opinion 762 (2018, reaffirmed 2024) says all patients should be routinely asked about alcohol at prepregnancy counseling. Neither ACOG opinion names an instrument; USPSTF names TWEAK, T-ACE, 4P's Plus and NET as pregnancy-specific tools.
Why does the AUDIT self-test on this site ask about six drinks when the screening question asks about five or four?
Because they are different instruments calibrated to different drinks. The WHO AUDIT (2001) asks about six or more drinks on one occasion and assumes a 10 gram drink. NIAAA's Core Resource (2025) notes that the original, internationally validated AUDIT-C sets a threshold of 6 drinks for both men and women, while the USAUDIT-C and the single question use 5 for men and 4 for women to match the 14 gram US standard drink. VA/DoD (2021) prints its own variant, 6 or more for men and 4 or more for women. The self-test uses the WHO wording and zones, so counting US drinks against it errs toward flagging more people, not fewer.