I do not have access to the article so difficult to say but they do specify "community-dwelling" which I believe means they were in something like assisted livi

I do not have access to the article so difficult to say but they do specify "community-dwelling" which I believe means they were in something like assisted living facilities. This would make data collection on food intake much easier since the specific meals would be prepared by the facility and they must have records of which resident had which meal. While these are super interesting headlines, I think the chances of some hidden driver affecting animal protein intake is very likely. E.g., it is not uncommon for geriatric individuals to be unable to consume specific foods or have doctors specifically restrict their diets due to heart disease, blood pressure, cholesterol or any number of other factors. Without detailed analysis of confounding health factors, I would not read this as prescriptive. In fact, animal protein intake may be even be a trailing indicator.