The corporate food industry does a pretty admirable, if diabolical, job of keeping important scientific information from consumers regarding the types of foods one should eat for optimal health. This is because a lot of money rides on getting people to ignore evidence-based dietary health advice and eat more "value-added" foods.
"Value-addded" is a fancy moniker for refined and processed foods, which mean more money in the pockets of food company shareholders when consumers eat them. That being said, value-added foods are not, in and of themselves, always harmful to human health. In fact, many minimally processed foods can be harmful to human health, especially if they are derived from animals.
Take, for example, meat. Most meat is minimally processed in that the animals from which it is derived are fed optimal, yet artificial (as compared with their natural diets), foods designed to develop the animal's muscle tissue in such a way as to make it appealing to humans, so they will eat it. For example, beef cattle are fed corn* because it's cheap. This adds value to the corn by turning it into much more expensive cuts of meat. The muscle tissue is then rendered from the carcasses in slaughter houses, refrigerated, sometimes mechanically ground and/or treated with chemicals, and packaged, adding still more value to the meat. Such processing can be pretty basic and still increase the food's value immensely.
We know from the scientific research on diet that meat consumption is strongly correlated with poor health outcomes and premature death, even after controlling for confounding factors like caloric intake (meat is more calorie dense) and lifestyle factors (smoking, exercise, etc.). It's also true that the more processed meat is (cured meat, for example), the worse it is for health. Conversely, a plant-based diet that contains little to no animal-based foods (including dairy) has been clearly linked to longevity and a reduction in the incidence of chronic diseases like diabetes, cancer, and heart disease. This is true even though many completely plant-derived foods, such as sugar and flour (even tofu!), are highly processed and refined (to add value) by comparison to the processing and refinement that happens to meat-based foods.
The current scientific evidence strongly suggests that the degree of refinement and processing of a food is less of a predictor of health outcomes than whether a food is plant- or animal-derived. Another confounding factor is that many dietary food components are significantly correlated ("fries and soda with that cheeseburger?"). That is to say, people who eat a lot of meat generally do not eat meat exclusively, and meat-heavy meals are often supplemented with plant-based foods, often highly refined and processed. So, in any diet in which meat is consumed, epidemiological studies are likely to show adverse health outcomes linked to both the meat in the diet as well as the refined/processed plant-based foods in the diet. This is why studies of diet and health have to be really large to be able to tease out these slight statistical nuances, and it is also why the food industry can reasonably question and critique any scientific studies that purport to show adverse health consequences from their precious value-added foods.
As a result, some of the most compelling studies on the role of diet in health are the ones that look at overall dietary patterns in large groups of people (population studies), rather than individual foods, and then use factor analysis to determine statistically the effects on health of particular food groups (beef, poultry, fish, dairy, whole grains, nuts, oils, sugars, fruits, and vegetables, etc.) within those dietary patterns. It is studies like these that have given us the
Paleolithic, Mediterranean, Okinawan, Standard American, Vegan, Vegetarian, and other dietary patterns that we hear about in the media. Studies like these also have the advantage of being conducted with humans, rather than in mouse or rat models of diet (humans eat very differently than mice and rats, so it is literally like comparing apples to oranges...pun intended). These studies clearly link plant-based diets to better overall health outcomes in humans than meat/dairy-based diets. Factor analysis reveals that greater animal-derived foods in the diet are the main contrbutors to poorer health outcomes and death. Starchy, salty, oily junk food also correlates with poorer health outcomes, even when such foods are plant-based (potato chips and French fries, for example), but this effect is strongly linked to the greater consumption of meat in diets that have a high consumption of junk food. Junk food is so highly refined and processed that a consumer cannot get enough of the required nutrients (mainly protein) for survival without eating high-protein animal-based foods as well (in general). The mechanism of action of many of the effects of diet on health is inflammation in the body, which is beyond the scope of this post.
