Good prospective cohort studies and meta-analyses of them are some of the best evidence we can possibly get for diet, because doing lifespan-scale human RCTs is completely impractical, especially for diet. The meta analyses you shared still showed an association between reduced meat consumption and reduction of CVD risk, just with low effect size and low statistical confidence, so this is really not strongly contrary to the conclusion that red meat is better to eliminate or reduce.

I hear the “correlation is not causation” argument loud and clear, but meat is expensive, meat consumption correlates with wealth, and wealth correlates strongly with health and longevity. Diets like the Atkins diet and the Paleo diet have also been relatively popular among health-conscious people since the 70’s. The animal agriculture industry also has enormous incentive to fund studies which show animal products in a positive light. So I really don’t see any strong reason to think that confounding factors/biases would systematically favor low-meat diets rather than high-meat diets. If meat were truly neutral in terms of CVD risk, we would expect to see a near equal number of studies showing that it is protective for CVD as we see studies showing it increases CVD risk, but this is not what we see in practice.

RCTs measuring CVD biomarkers also pretty consistently show that reducing meat consumption, especially red meat, decreases well-established CVD biomarkers, so, if these results combined don’t convince you, what results would convince you?

And yes, the evidence for cancer with unprocessed red meat is weaker, but the evidence for CVD risk and CVD biomarkers is pretty consistent across both RCTs and observational studies. The small effect size you mention is consistent with the expected heterogeneous results of some studies, and the fact that the effect size is weaker than something like smoking does not mean the effect does not exist.

The CVD hazard ratio may be close to 1.11 for red meat, which means just 100g a day increases your risk for CVD by 11% (Red meat consumption, cardiovascular diseases, and diabetes: a systematic review and meta-analysis). In an animal based low carb/carnivore diet you would need to be eating 10x that much (whether or not youre incorporating other animal foods also high in saturated fat), and there’s no compelling reason to expect the results to be better by consuming more of something with demonstrable long term cardiovascular system-damaging effects, whether in the context of an extremely low carbohydrate diet or not. Studies like Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis also show us that a moderate level of carbohydrate consumption give better results than either extreme high carbohydrate or extreme low carbohydrate diets. “There was a U-shaped relationship between carbohydrate intake and mortality in the Atherosclerosis Risk in Communities cohort, a finding that was consistent in the meta-analysis combining these data with those from the other cohorts. When assessing total carbohydrate without regard to specific food source, diets with high (>70%) or low (<40%) percentage of energy from carbohydrates were associated with increased mortality, with minimal risk observed between 50–55%. Low carbohydrate dietary patterns that replaced carbohydrate with animal-derived protein or fat were associated with greater mortality risk, whereas this association was inverse when energy from carbohydrate was replaced with plant-derived protein or fat.“

What you define as “dangerous,” is somewhat subjective, but if you care about things like minimizing all-cause mortality and heart disease risk (which you probably should, given that CVD is the number 1 killer), then it’s pretty clear that eating ASF in large amounts is suboptimal.