For people with the chronic liver infection hepatitis C, heavy drinking is an obvious no-no, but a new study links even modest alcohol consumption with an increased risk of death - and not just from liver disease.
"What this study shows is... truly, even what might be considered a moderate and safe amount of alcohol use in people without hepatitis C is dangerous to your health if you have hepatitis C," said Rae Jean Proeschold-Bell, a hepatitis C researcher at Duke University in Durham, North Carolina, who was not involved in the study.
The findings support what liver specialists typically recommend - that people with hepatitis C should limit their alcohol use, said Dr. Zobair Younossi, the study's lead author and chair of medicine at Inova Fairfax Hospital in Falls Church, VA.
"Patients with hepatitis C should not really drink," he said.
But the reality is that people with hepatitis C have higher rates of alcohol use than people without the liver disease, said Proeschold-Bell, who studies interventions to reduce drinking among people with the disease.
Doctors have known that excessive drinking can exacerbate liver disease caused by hepatitis C, but there's some debate about whether less frequent drinking would have a similar effect.
Younossi and his colleagues looked to a large national survey on health and lifestyle that tracked people for several years.
They compared 8,767 people without hepatitis C to 218 people with the disease.
Hepatitis C is a virus spread through blood. Some 3.2 million people in the U.S. have a chronic hepatitis C infection, according to the Centers for Disease Control and Prevention.
The disease can cause serious liver damage, and while some people are treated with medications, others will go on to require a liver transplant.
The survey tracked the participants for 13 to 14 years. During that period, 19 percent of those with hepatitis C and 11 percent of those without the infection died.
Younossi's team found that people with hepatitis C who drank excessively - three or more drinks a day - were five times more likely to die than heavy drinkers who were not infected.
That result was not surprising, "We've known heavy drinking is particularly bad if you have hepatitis C," Proeschold-Bell told Reuters Health.
But people infected with hepatitis C who had up to two drinks a day were also twice as likely to die during the study than those with similar drinking habits who were not infected.
For the purposes of the study, a drink was equivalent to 10 grams of alcohol, which is roughly the amount in four ounces of wine, 12 ounces of beer or one ounce of hard liquor.
Younossi said the increased risk of death from liver disease is driving the numbers.
"What is incredibly striking is liver-related death in patients with hepatitis C who even drink moderately," said Younossi.
For instance, the risk of liver-related death among people with hepatitis C who averaged two or fewer drinks a day was 74 times that of similar people without hepatitis C.
Those moderate drinkers with the virus were also nearly three times more likely to die of "all causes," the researchers report in the medical journal Alimentary Pharmacology & Therapeutics.
"A drink a day is not OK," Younossi told Reuters Health. "Even a moderate amount of alcohol use in the setting of hepatitis C can increase the risk of death and liver-related mortality specifically."
Proeschold-Bell said there is a great opportunity for intervening with people's alcohol use given that they are already interacting with the medical system if they have a chronic hepatitis C infection.
"This is potentially very powerful, because if the person with hepatitis C is already going in for medical care, they have some relationship with that clinic. They have some degree of trust, so (perhaps) you can provide alcohol treatment right then and there," she said.
Younossi said there's some evidence that if heavy drinkers without hepatitis C abstain from alcohol, their liver disease can improve.
He said he suspects the same might be true for patients with the infection, but that future studies will have to confirm that hunch.