This post is dedicated to Ted Kyle who is as obsessed with understanding and treating obesity as I am defending ultra-processed foods. He points out that many of the recommendations for overcoming obesity come from correlation studies. A principle of science is that correlation does not mean causation. And yet, on the internet correlation and cause are used interchangeably. To understand the distinction, we must understand differences in holism and reductionism.

This month I made distinctions between a holistic view of nutrition and food and a reductionistic one. Nowhere is there a starker distinction between the two as the contrast between whole and ultra-processed foods. Holism seeks a broader perspective of the nutrition landscape. It focuses on developing an understanding of the food system and how different aspects interact with each other. More information leads to a broader understanding. Patterns emerge that lead to possible solutions to complex problems. Reductionism seeks to narrow nutrition down to its essential elements. It focuses on nutrients and the role they play in maintaining health. More attention to detail leads to simple explanations. Reductionism aims to establish cause and effect.

Holism claims reductionism misses the big picture. Reductionism claims holism doesn’t solve real problems. Holism charges reductionism with simplistic solutions that cause more harm than good. Reductionism charges holism with idealistic thinking that fails to work in the real world. So where do we go from here? Maybe backing off strict holistic and reductionistic thinking and combining the two approaches would work. It might be worth a try.

Full disclosure. I have some personal experience with both holism and reductionism. My scientific education was strictly reductionistic. All my degrees are in Food Science. Food Science is not a scientific discipline. Rather it is informed by three overarching disciplines—chemistry, microbiology, and engineering. Food chemistry, food microbiology, and food engineering are all reductionistic. My research in graduate school was reductionistic. At the MS level I extracted plant pigments for use as colorants formulated beverages (Kool Aid). For my PhD I studied the interaction of lipid hydrolysis and oxidation in flounder muscle.

Upon graduation I embarked on a two-pronged mission. I continued to pursue cellular biology in understanding the role of lipid oxidation in the development of chilling injury in fresh vegetables, a reductionistic exercise. As part of an interdisciplinary research team, I focused on changes in quality of fresh fruits and vegetables from harvest to purchase. The team, led by an engineer, took a systems approach to understanding handling and distribution of fresh produce. This research became my introduction to holism.   

What holism taught me about distribution of fresh fruits and vegetables. Studying fresh produce from farm to market is challenging. The idea of treatment and control groups is illusory. In the tomato patch, there is no such thing as a standard tomato. One tomato on a plant within a plot of the same variety differed in ripeness and flavor from its neighbors on the same plant. We started out with high sample-to-sample variation. Conditions in the field from one day to the next, conditions in the truck, and the way they were handled varied beyond our control. Similar variations were observed between packinghouses, wholesale warehouses, retail outlets, and home storage practices. There are reasons scientists confine themselves in laboratories and avoid field work.

We were able to observe handling techniques that worked better than others and identify problem areas. We drew nice diagrams of a “handling system” from farm to consumer. We determined that the key sources of variation happened at harvest and at the point of sale. What we were not able to determine were causes and specific outcomes. Even what we called “handling systems” were actually supply chains. The variability from one run to the next was not clearly controlled. We had a much better big picture of tomato handling than reductionists had, but we could not identify causes of problems without reducing the problem into its essential elements. We were able to develop a window of maximum quality at retail based on color and firmness changes. [1]

What reductionism told me about the mechanism of chilling injury. In general, the closer we store fruits and vegetables to the freezing point without freezing them the longer their shelf life. Some fruits and vegetables are sensitive to cold temperatures and must be stored at higher temperatures to maintain quality. Bananas, peppers, and tomatoes are sensitive to very low storage temperatures. The prevailing theory as to why some produce is more susceptible to low temperatures is the fluidity of the lipids (fats) in their cellular membranes. If the membrane is not fluid enough at low temperatures, quality suffers.

Most of my research on chilling injury involved bell peppers and tomatoes. Our research established the importance of lipid oxidation in cellular membranes in the mechanism of chilling injury [2]. Mechanistic studies are very reductionistic and can point to cause and effect. Development of a mechanism, however, requires some holistic thought. Turns out that details of the mechanism of chilling injury were never established. The holistic solution of storing sensitive fruits or vegetables at or above the optimal storage temperature helped preserve quality.

Holism and nutrition. Holistic nutritionists have declared whole foods as healthy and ultra-processed foods as unhealthy. Such a binary choice seems rather unholisitic to me, but my biases trend reductionistic. Holistic nutrition tends to reject the consumption of foods from animals or at least in minimizing their consumption [3]. Holism rejects the importance of individual nutrients and stresses the interaction of nutrients both within the food matrix and in our bodies. The sociology of eating, meal preparation, and communion of eating in groups of people contribute to holistic health and wellbeing. Epidemiological studies demonstrate clear linkages between ultra-processed foods and cancer [4], diabetes [5], heart disease [6], and dementia [7]. A recent publication showed that red meat consumption is clearly associated with diabetes [8] in observational studies.

Reductionism and nutrition. Reductionistic studies identified diseases associated with deficiencies of individual vitamins and minerals. Further research uncovered the importance of diagnosing subclinical deficiencies and establishing balances between critical nutrients in health maintenance. Reductionism rejects epidemiology as conclusive when it suggests cause and effect. Epidemiology is very good at providing direct links between the offending food and the disease in a food poisoning outbreak. It is less useful in relating dietary behavior to specific chronic disease. Epidemiological studies provide hints as to where to look in mechanistic studies, but they do NOT demonstrate cause and effect.

In the research linking red meat and diabetes, data from mechanistic studies (reductionistic) do not support the observational data (holistic). Reductionists help overcome such data difficulties with covariate analysis. For example, people who eat large amounts of meat share many characteristics with each other. How do we know it is the meat that causes the diabetes or some other covariate(s) that are responsible for the disease. An understanding of the biochemistry (a very reductionistic discipline) of diabetes suggests that the answer lies in understanding carbohydrate metabolism. Carbohydrate concentrations are low in meats.   

An illustration from a completely different topic. My wife and I finally decided to quit pumping money into saving our 2000 Buick LeSabre that brought us safely out of Florida after Hurricane Ian and to South Carolina, the land of our youth. Turns out we had very different ideas of how to shop for a car. She wanted a wide-open search to narrow our choice to the best possible automobile. I wanted to limit the possibilities down to a very few characteristics and make the choice quick and surgical. We went to CarMax which offered a wide range of makes and models of used cars. She went shopping on Consumer Reports. I identified the three most important characteristics we both liked in a car: comfortable seats for long trips, smooth handling, and mechanical reliability to give us at least 15 years of driving pleasure. We had never owned anything other than a sedan.

We found a sedan that gave us the comfortable seats and wonderful handling, but Consumer Reports said no to reliability. We also tried a sedan that had stiffer seats and not so smooth a ride but was much more reliable. I had my sights set on a Buick LaCrosse that was supposed to be a successor to the LeSabre. We had to wait for it to be shipped from Atlanta to Upstate South Carolina. What a disappointment with its stiff seats and not-so-great handling! Then my wife started suggesting numerous alternatives—the more she suggested, the more fits I pitched. We test drove a SUV that had been recommended by family members and didn’t like it. We then drove a Honda CRV and both fell in love with it. No, its seats are not as cushy as the LeSabre, but we are both happy with it.

The point in all of this is that her approach was holistic and mine was extreme reductionism. I did not visualize our search in this manner until I started to write this post. Neither approach worked on its own, but we found a solution that combined the two philosophies.

Take home lesson. Holism doesn’t provide cause and effect. Reliance on correlation studies to condemn ultra-processed foods as the primary basis of unhealthiness of the Western diet obscures the issue. Reductionism can get lost in the weeds and miss the main point. The answer to a problem does not always have to be holistic or reductionistic. A combination of holism and reductionism can help move us forward in understanding diet, health, and chronic disease management. Dogmatic orthodoxy will slow us down whether we are trying to design healthy dietary patterns or buy a car.       

Coming soon: Who killed fake meat?

References:

[1] Shewfelt, R.L., S.E. Prussia, A.V.A. Resurreccion, W.C. Hurst, and D.T. Campbell, 1987. Quality changes of vine-ripened tomatoes within the postharvest handling system. Journal of Food Science 661-664, 672.

[2] Shewfelt, R.L. and A.C. Purvis, 1995. Toward a comprehensive model for lipid peroxidation in plant tissue disorders. HortScience 30(2): 213-218.

[3] Fardet, A. and E. Rock, 2022. Chronic diseases are first associated with the degradation and artificialization of food matrices rather than food composition: calorie quality matters more than calorie quantity. European Journal of Nutrition 61:2239–2253 https://doi.org/10.1007/s00394-021-02786-8

[4] Fiolet, T., B. Srour, L. Sellem, E. Kesse-Guyot, B. Allès, C. Méjean, M. Deschasaux, P. Fassier, P. Latino-Martel, M. Beslay, S. Hercberg, C. Lavalette, C.A. Monteiro, C. Julia, and M. Touvier, 2018. Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ 2018; 360 https://doi.org/10.1136/bmj.k322

[5] Nardocci, M., J.Y. Polsky, and J. C. Moubarac, JC. 2021. Consumption of ultra-processed foods is associated with obesity, diabetes and hypertension in Canadian adults. Can J Public Health 112, 421–429. https://doi.org/10.17269/s41997-020-00429-9

[6] Zhong, GC., H.T. Gu, Y. Peng, K. Wang, Y-Q-L. Wu, T-Y Hu, F-C. Jing, and F.B. Hao, 2021. Association of ultra-processed food consumption with cardiovascular mortality in the US population: long-term results from a large prospective multicenter study. Int J Behav Nutr Phys Act 18, 21 (2021). https://doi.org/10.1186/s12966-021-01081-3

[7] Li, H., S. Li, H. Yang, Y. Zhang, S. Zhang. Y. Ma, Y. Hou. X. Zhang, K. Niu, Y. Borné, and Y. Wang, 2022. Association of ultraprocessed food consumption with risk of dementia, a prospective cohort study. Neurology 99:E1056-E1066. DOI: https://doi.org/10.1212/WNL.0000000000200871

[8] Sanders, L.M., M.L. Wilcox, and K.C. Maki, 2023. Red meat consumption and risk factors for type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials. European Journal of Clinical Nutrition 77:156-165. https://doi.org/10.1038/s41430-022-01150-1