1.) See if the methodology was appropriate. Was reporting done by food diary or control food sampling? How big was the population pool? How were test subjects selected? How did they guard against bias? Placebo effect?
2.) Look at the statistics. You would be surprise at how like a 9% increase from control is construed to a 100% outcome in the media. For instance, only 33% of the US population have salt-sensitivity hypertension. For 66% of the population, the amount of salt you consume doesn't impact your blood pressure. And yet since there is such a sizable population that is, they put a blanket statement stating that one should reduce their sodium intake to reduce their blood pressure.
3.) What's the dosage? Strawberries can cure cancer! If you eat 13 pints a day, every day, for the rest of your life...
4.) What is a proposed biological mechanism? Correlation is not equal to causation.
Look, as a researcher I have read a LOT of bad research. It's up to you to read the original article and decided if there is enough evidence to persuade you to agree with them.
Until then, it's all about moderation. So for me it's enjoying the occasional soy chai latte or cream in my coffee.