Readers know that I am highly-skeptical of all reports of studies.
Everybody remembers being told, by the scientists, not to eat eggs and red meat. Of course, now that is all known to be bogus. Unless you do manual labor or are significantly underweight, there is no reason to eat grains or cereals for breakfast except for fun. Oats and cholesterol? All now known to be bogus too. And real butter is good for you.
Just one reason I distrust reports has to do with data-mining. I was once involved with a medical study which applied data-mining to a large amount of data. Inevitably, we found some correlations which passed the t-test. That's not science, but academics get papers published that way and papers can lead to grants and tenure.
Is social isolation a good or bad thing? It seems to me to depend on the reason for it. Many people do not enjoy a social life while many suffer from loneliness. It's probably a bell curve, like most things.
This report, I feel, confuses causation with correlation. That's a rookie error: How Your Relationships Affect Your Health