(NYTines) -Over the past year, I’ve tried to clear up a lot of the misconceptions on food and drink: about salt, artificial sweeteners, among others, even water.

Now let me take on alcohol: wine, beer and cocktails. Although I have written about the dangerous effects of alcohol abuse and misuse, that doesn’t mean it’s always bad. A part of many complex and delicious adult beverages, alcohol is linked to a number of health benefits in medical studies.

That doesn’t mean the studies provide only good news, either, or that the evidence in its favor is a slam dunk. You won’t be surprised to hear that, once again, my watchword — moderation — applies.

Research into how alcohol consumption affects health has been going on for a long time. A 1990 prospective cohort study included results of more than 275,000 men followed since 1959. Compared with those who never drank alcohol, those who consumed one to two drinks a day had a significantly reduced mortality rate from both coronary heart disease and “all causes.” Those who consumed three or more drinks a day still had a lower risk of death from coronary heart disease, but had a higher mortality rate over all.

A 2004 study came to similar conclusions. It followed about 6,600 men and 8,000 women for five years and found that compared with those who drank about one drink a day on average, those who didn’t drink at all and those who drank more than two drinks a day had higher rates of death. Results like these have been consistent across a number of studies in different populations. Even studies published in the journal Alcoholism: Clinical and Experimental Research agree that moderate drinking seems to be associated with a decreased risk of death over all.

However, alcohol seems to have different effects on different diseases. Almost all of the major benefits of drinking are seen in cardiovascular illnesses. In fact, with men, even consumption of a surprisingly large amount can seem protective.

When it comes to cancer, the picture isn’t as rosy. For instance, a 2007 study involving the Women’s Health Study cohort found that increased alcohol consumption was associated with an increased risk of breast cancer. More broadly, a 2014 systematic review of epidemiologic and experimental studies looking at alcohol and breast cancer found that the overall consensus is that each additional drink per day increases the relative risk (comparing the risk in two groups) of breast cancer by a statistically significant, but small, 2 percent — although not the absolute risk.

A meta-analysis of colorectal cancer and alcohol found that heavy drinkers, but not light or moderate drinkers, were at increased risk of the disease. No relationship is seen with respect to bladder cancer or ovarian cancer. A study that included all cancers found that light drinking was protective; moderate drinking had no effect; and heavy drinking was detrimental.

Moderate alcohol consumption has been found to be associated with other benefits, though. A cohort of about 6,000 people followed in Britain found that those who consumed alcohol at least once a week had significantly better cognitive function in middle age than those who did not drink at all. This protective effect on cognition was seen in people who drank up to 30 drinks a week.

A 2004 systematic review found that moderate drinking was associated with up to 56 percent lower rates of diabetes compared with nondrinkers. Heavy drinkers, though, had an increased incidence of diabetes.

This is where savvy readers should be asking: What about randomized controlled trials? After all, epidemiologic evidence and associations only go so far; they cannot get us to causation.

Recently, in Annals of Internal Medicine, such a trial was published. Patients with well-controlled Type 2 diabetes were randomized to drink 150 milliliters of water, white wine or red wine with dinner for two years. The beverages were provided to patients free of charge. They were all placed on a Mediterranean diet with no calorie restrictions.

Researchers found that those who drank the wine, most notably red wine, had a reduction in cardiometabolic risk factors, or those for heart disease, diabetes or stroke. This was especially true in patients who had certain genotypes. Further, no one had any significant adverse effects from being randomized to drink the alcohol.