Whether or not it's safe to drink alcohol is a complicated question for people who have rheumatoid arthritis.

Alcohol can interact with the powerful medications that help to control the disease, increasing the likelihood of liver damage. Alcohol can also affect balance, increasing the risk of falls and broken bones, and may disrupt sleep. Alcohol can also contribute to depression, a condition that's frequently a close companion to chronic pain.

On the other hand, having an occasional cocktail with friends is an important way to maintain the social connections that studies have shown are just as integral to good health as exercise and a healthy diet. 

And last week, alcohol seemed to get a big endorsement from a study published in the journal Rheumatology, which concluded that people with rheumatoid arthritis who said they frequently drank alcoholic beverages had disease that was significantly less severe than people with RA who said that they rarely or never drank.

What’s more, when people with rheumatoid arthritis were compared to a group of people without rheumatoid arthritis, non-drinkers appeared to have four times the odds of developing RA as frequent drinkers

According to Google News, the paper generated at least 322 articles with headlines that included “Drinking May Lessen the Effects of Rheumatoid Arthritis” and “Alcohol Helps Rheumatoid Arthritis Symptoms.”

But experts say you shouldn’t raise a glass to this study's conclusions just yet.

“I think this study has a lot of flaws,” says Timothy S. Naimi, MD, a medical epidemiologist at Boston University School of Medicine who studies the effects alcohol on health. “This should not be a reason for people to start drinking or for people to drink more frequently.”

The study compared 873 RA patients to 1004 people without the chronic condition. All of the participants were asked to complete a questionnaire explaining how often they’d had alcohol the month before joining the study. X-rays and blood tests were also taken, and a nurse examined patients.

Both men and women who drank most often, classified as more than ten days a month, had 20 to 30 percent less severe disease – including less pain, swelling, joint damage and disability and lower levels of inflammation – than teetotalers.

This isn't the first study to find that alcohol may lessen the risk of developing rheumatoid arthritis, but it is the first to suggest that alcohol can ease the severity of the disease.

But experts, including the study’s authors, say that there are important caveats that muddy those conclusions.

“I would urge caution until these results have been replicated in other studies. I tell my patients that a small amount of alcohol, consumed responsibly on two to three days per week is unlikely to be harmful to them, but on no account should they try and use alcohol as a treatment for their symptoms,” says James Maxwell, MD, the lead author of the study and a rheumatologist at Rotherham Hospital in the U.K.

One flaw involves the makeup of the two groups that researchers used for comparison, the group of people with RA (the cases) and the group without the disease (the controls).

In this type of study, researchers generally take great pains to gather a control group that is as similar to the case group as possible. The greater the degree of similarity between the two groups, the more powerful the study is.

In this investigation, however, the members of the control group were younger, were less likely to smoke and were more likely to be male than the group with RA.

Those differences are important because each is independently associated with RA risk. Women are more likely to get rheumatoid arthritis than men, smokers are at greater risk of developing the disease and the risk of RA goes up with age.

So the group with rheumatoid arthritis already had a higher risk, before researchers started to consider alcohol intake.

“They ended up controlling for those things statistically, but it raises the likelihood that their cases and controls are different in many, many profound ways that would influence their alcohol consumption, their arthritis or both of those things,” Dr. Naimi explains.

Dr. Naimi also thinks another phenomenon, something called reverse causation, could be at work:  Instead of drinking reducing the risk of arthritis, it may be that having arthritis reduces the likelihood of drinking.

“People who have arthritis or those with physical disabilities or with poor health status generally will often reduce the frequency of their drinking,” he says, because they’re taking medications that don’t mix with alcohol, like methotrexate, or because they’re just in pain and don’t feel like it.

Additionally, the study only asked how often, not how much, people drank. That makes it impossible to characterize exposure. A person who binges once a week, for example, might be drinking as much or more than someone who has a glass of wine with dinner on three nights each week.

“I would say exercise caution all around,” says Bruce Richardson, MD, PhD, a professor of internal medicine at the University of Michigan Medical School in Ann Arbor. “The headline on this says ‘increased alcohol consumption protects’, which is scary in my mind and sets off alarms. “

Dr. Richardson says this study may be valid. But he says for now it must be compared to the larger body of evidence that warns of the dangers of alcohol consumption beyond recommended rates.

“For women, it’s one drink a day, usually wine. For men - one to two, may actually improve overall survival. But if you drink more than that, death rates and complication rates increase exponentially. So this study on the effect of alcohol and RA needs to be taken in perspective,” Dr. Richardson says.

“It’s not safe to say I can drink as much as I want because I’m helping myself,” he says. “Absolutely don’t change your drinking habits because of this study. Not until we get more data. It’s not safe."