Health providers should screen all adults and pregnant women for risky drinking habits, a government-backed expert panel said Monday in new draft recommendations.

The U.S. Preventive Services Task Force guidelines were released along with a summary of past research showing a few brief counseling sessions can help cut excessive drinking in people who aren't full-blown alcoholics.

Researchers found that with such interventions, at least one in every ten risky drinkers can be brought back down to what is considered a safe drinking level.

For men, that means no more than four drinks on any one occasion and at most 14 a week. For women, the numbers are three and seven, respectively.

"The really good interventions incorporate self-help materials, they incorporate an action plan made by the doctor and the patient - they're not just a prescription," said Dr. Daniel Jonas of the University of North Carolina at Chapel Hill, who led the research review but is not part of the USPSTF.

It's estimated that about one in three Americans drink too much. Most of the misuse is known as risky or hazardous drinking, which doesn't involve physical dependence or diseases, but nonetheless has been linked to accidents and premature deaths.

The new draft recommendations repeat the USPSTF's previous position from 2004, noting that the evidence is still insufficient to give advice regarding adolescents.

They are based on the research update from Jonas and colleagues, which included 23 trials of alcohol counseling; each trial had anywhere from 72 to more than 1,500 people and lasted from six to 48 months.

One of the most effective screening techniques is simply asking a person if he or she has had five or more drinks on any one occasion over the past year, said Jonas.

If the answer is yes, the health provider should gauge the level of misuse. If it's limited to risky use, not yet causing direct physical or mental harm, then a few counseling sessions with the doctor can be helpful.

"The poster-child for these interventions was two visits with the primary care doctor, each one about 10 to 15 minutes and about a month apart and each one with a follow-up call from a nurse to reinforce the behavior," Jonas told Reuters Health.

He and his colleagues found such interventions cut the average weekly alcohol use by 3.6 drinks on average compared with no intervention. They also reduced binge drinking by 12 percentage points and risky drinking by 11 percentage points.

That means nine people would need to be counseled for one to cut back to safe drinking levels. And that might even be an underestimate of the true benefit of counseling, said Jonas. That's because even people who didn't get counseling had been screened, making them aware they drank too much and possibly leading them to cut back.

Screening and counseling has little potential to do harm, according to the researcher. Still, there is no proof that drinking less actually leads to fewer accidents or deaths, which is why the task force gave the recommendation a grade "B" instead of an "A."

Jonas acknowledged that with about a fifth of Americans being "risky drinkers," a universal screening program could be pricey. While Medicare reimburses doctors for it, it's time out of the calendar for both patients and health providers.

"That is really the biggest challenge with implementing this," Jonas said, adding that UNC is in the process of training its doctors to do the screening.

"Training is a big piece of it too, because a lot of providers need to learn how to do it," the researcher said.

The USPSTF's draft guidelines are open for public comments until October 22.