Search

Natural Medicine Journal

August 7, 2024

Ultraprocessed Foods Aren’t Good for You. We Knew That.

Surprising results from a population-based cohort study

Jacob Schor

ND, FABNO
Those who ate more than double the amount of ultraprocessed foods as others had only a “modest” 4% higher all-cause mortality

Reference

Fang Z, Rossato SL, Hang D, et al. Association of ultra-processed food consumption with all cause and cause specific mortality: population based cohort study. BMJ. 2024;385:e078476. 

Study Objective

To delineate the association between all-cause mortality and cause-specific mortality with the consumption of ultraprocessed food in the diet

Key Takeaway

Higher intake of ultraprocessed foods was associated with “modestly higher” all-cause mortality, driven by causes other than cancer and cardiovascular diseases (CVDs). No association was observed with cancer or CVD mortality.

Design

Population-based cohort study using data from the Nurses’ Health Study (1984–2018) and the Health Professionals Follow-Up Study (1986–2018)

Participants

The study included 74,563 women and 39,501 men with no history of cancer, CVDs, or diabetes at baseline. Investigators obtained medical and lifestyle information via mailed questionnaires completed every 2 years.

Intervention

Investigators evaluated participants’ diets using 2 separate scoring systems. They evaluated ultraprocessed food intake using the Nova classifications:

  • unprocessed or minimally processed foods
  • processed culinary ingredients
  • processed foods
  • ultraprocessed foods

Investigators also assessed the participants’ diets using the Alternative Healthy Eating Index–2010 (AHEI) score.

Study Parameters Assessed

In addition to categorizing participants’ diets using the 4 Nova classifications, investigators further categorized the ultraprocessed foods into 9 mutually exclusive subgroups:

  1. Ultraprocessed breads and breakfast foods 
  2. Fats, condiments, and sauces
  3. Packaged sweet snacks and desserts
  4. Sugar-sweetened and artificially sweetened beverages
  5. Ready-to-eat/heat mixed dishes
  6. Meat/poultry/seafood-based ready-to-eat products
  7. Packaged savory snacks
  8. Dairy-based desserts
  9. Other

Investigators did not consider alcohol an ultraprocessed food in the primary analysis. They also removed whole grain foods from the analysis even if they were ultraprocessed, as they are associated with lowering all-cause mortality.

These cohort studies have established policies that track deaths of members via contact with next of kin and vital records.

Primary Outcome

The primary outcome was all-cause mortality. Secondary outcomes included deaths from cancer, cardiovascular diseases, and other causes including respiratory diseases and neurodegenerative diseases.

Key Findings

Of the study participants, 30,188 women and 18,005 men died during a median of 34 and 31 years of follow-up, respectively. When compared with those in the lowest quarter of ultraprocessed food consumption, the participants in the highest quarter of ultraprocessed food consumption had a 4% higher all-cause mortality (hazard ratio [HR] 1.04, 95% confidence interval [CI] 1.01–1.07) than those in the lowest quarter of consumption.

Participants in the highest quartile of ultraprocessed food consumption had a 9% higher mortality from causes other than cancer or cardiovascular diseases (1.09, 1.05–1.13). The all-cause mortality rates among participants in the lowest vs the highest quartiles of consumption were 1,472 vs 1,536 deaths per 100,000 person years.

No association was seen with cancer or cardiovascular mortality and consumption of ultraprocessed foods.  Meat/poultry/seafood-based ready-to-eat products (for example, processed meat) consistently showed strong associations with mortality outcomes (HRs ranged from 1.06 to 1.43). Sugar-sweetened and artificially sweetened beverages (1.09, 1.07–1.12), dairy-based desserts (1.07, 1.04–1.10), and ultraprocessed breakfast food (1.04, 1.02–1.07) were also associated with higher all-cause mortality.

Practice Implications & Limitations

Let me admit at the onset of this discussion that I was surprised by these results; I had expected a more pronounced effect would be demonstrated. Instead, the effects were, to use the word the study’s authors chose, “modest.” 

Participants in the lowest-consumption quartile ate a median of 3.0 servings of ultraprocessed food a day. In contrast, those in the upper quartile ate a median of 7.4 servings per day. That is nearly double the amount of ultraprocessed food, yet their risk of dying increased only 4%. Modest is right.

The idea that consumption of ultraprocessed foods is detrimental to health first appeared in a paper written by Carlos Monteiro et al, in 2010.1 The researchers described a new food classification system, calling it the Nova classification, which assigned foods into separate groupings based on the degree of industrial processing applied to them. Foods were divided into 3 defined groups: unprocessed or minimally processed foods (group 1), processed culinary and food-industry ingredients (group 2), and ultraprocessed food products (group 3). They were initially looking for an explanation for the surge in childhood obesity. Monteiro’s group used their classification system to analyze dietary data gathered from nearly 50,000 Brazilian households and reported that as household income increased, so did consumption of ultraprocessed foods. 

In other words, the AHEI, which gauges a diet by the quality of the foods and the nutrients eaten, is a better measure of dietary impact on mortality and chronic disease than the Nova measure of food processing.

On average, a quarter of total energy came from consumption of group 3 foods, but in higher-income households, more than 1/3 of energy was derived from these foods. Monteiro pointed out that in comparison to group 1 and group 2 items, group 3 foods contained “more added sugar, more saturated fat, more sodium, less fibre and much higher energy density.” Given the generally harmful nutrition profile of group 3 foods, Monteiro strongly suggested “that governments and health authorities should use all possible methods, including legislation and statutory regulation, to halt and reverse the replacement of minimally processed foods…”2

Writing in 2013, Monteiro summed up the situation thusly: “Many forms of food processing are beneficial. But what is identified and defined here as ultra-processing, a type of process that has become increasingly dominant, at first in high-income countries, and now in middle-income countries, creates attractive, hyper-palatable, cheap, ready-to-consume food products that are characteristically energy-dense, fatty, sugary or salty and generally obesogenic.… It is proposed here that the main driving force now shaping the global food system is transnational food manufacturing, retailing and fast food service corporations whose businesses are based on very profitable, heavily promoted ultra-processed products, many in snack form.”3

In 2013, Monteiro, now the principal investigator, reported that the Canadian diet was dominated by ultraprocessed foods, supplying 61.7% of dietary energy. Such dominance left Canadians exceeding the World Health Organization (WHO) limits for fat, saturated fat, sugar, and sodium while leaving them deficient in fiber.4

Scholarly and authoritative groups might suggest what a healthier diet should look like, but food processing and incorporation of fat, salt, and sugar with refined carbohydrates can create food products that are nearly addictive. Ultraprocessed foods have long shelf lives and are inexpensive to manufacture. They are easy to sell, profitable, and, in the opinion of many, addictive. But the trend toward increasing consumption of group 3 foods has continued unabated.

The trend toward evaluating diets using this Nova system has grown with similar speed, and many studies have been published evaluating associated health risks. PubMed currently lists about 500 citations in response to a search for “Nova Classification ultra-processed foods.”

Newer studies published using this Nova classification no longer use numbered groupings to specify extent of processing and instead employ abbreviations of their descriptions (ultraprocessed foods=UPFs and unprocessed or minimally processed foods=MPFs).

Those who have followed the research on ultraprocessed foods were expecting, based on earlier studies, more dramatic results from this Accordino study. 

In 2021, Suksatan and a team of Iranian researchers reported on a meta-analysis they had done that included 207,291 participants. Their results showed ultraprocessed food consumption was associated with a 21% increase in all-cause mortality (HR 1.21; 95% CI, 1.13, 1.30; P<0.001) and a 50% increase in death from cardiovascular disease (HR 1.50; 95% CI, 1.37, 1.63; I2=0.0%; P<0.001) but not cancer caused mortality.5

In an umbrella review of epidemiological meta-analyses published in February 2024, Lane et al reported far more striking effects. They identified 45 unique pooled analyses, including dose-response associations and 32 nondose-response associations (N=9,888,373). They report that associations were found between consumption of ultraprocessed foods and 32 health parameters, including mortality, cancer, and cardiovascular disease. They reported finding convincing evidence (class I) supporting a direct association between greater ultraprocessed-food exposure and a 50% higher risk of CVD mortality (risk ratio [RR] 1.50, 95% CI, 1.37–1.63;). Highly suggestive (class II) evidence suggested that greater consumption was associated with a 21% increase in risk of all-cause mortality (RR 1.21, 1.15–1.27; low) and heart disease mortality (HR 1.66, CI 1.51 to–1.84; low).6

How do we go from conclusions like those to this current report of no association with CVD? Or, for that matter, the “modest” (or should we say meager?) 4% increase in mortality in comparison to Lane’s reported 21% increase?

The Lane et al findings were not outliers. A separate review was also published in February 2024, by a German team led by Janett Barbaresko. Again, this was an umbrella review gathering data from prior meta-analyses that had combined findings from separate studies. 

Barbaresko et al identified 16 publications. Moderate certainty of evidence was found for all-cause mortality (summary RR per 50 g: 1.02; 95% CI, 1.01, 1.03), cardiovascular disease incidence and mortality (per 50 g/d: 1.04; 95% CI, 1.02, 1.06, and 1.05; 95% CI, 1.01, 1.08), type 2 diabetes incidence (per 10%: 1.12; 95% CI: 1.10, 1.13), and colorectal cancer (per 10%: 1.04; 95% CI, 1.01, 1.07).7

Similar associations were seen for a number of other conditions including “inflammatory bowel diseases, obesity, metabolic syndrome, nonalcoholic fatty liver disease, mental health…. but certainty of evidence was limited.” 

A January 2024 systematic review on food addiction in children, “characterized by behavioral changes related to the consumption of palatable foods, marked by dependence, impulsivity, and compulsion,” concluded that the more ultraprocessed foods consumed, the more vulnerable children are to food addiction.8

A December 2023 meta-analysis put some rather specific numbers to the association between UPF and chronic disease. Bestari and colleagues summarized 18 previously published research articles. Their results showed that UPF consumption significantly increased daily intake of carbohydrates, sugar, saturated fat, sodium, energy, cholesterol, and total fat while lowering fiber. A 10% increase in UPF consumption was associated with an 11.5% increased risk of diabetes (HR ± 95% CI: 1.115 ± 0.044); a 1% increase in CVD (1.096 ± 0.053); a similar 1% increase in obesity (1.068 ± 0.050); and a 2% increase in cancer risk (1.020 ± 0.020). While these increases in risk seem small, bear in mind these increases in risk were per small incremental increase in UPF consumption.9 The average American eats a lot of UPF. In 2016 Martinez Steele et al reported that between 60% and 90% of the standard American diet consisted of foods and beverages that are highly processed.10,11

It would be so tempting to simply ignore Zhe Fang et al’s modest results or assume some confounder had snuck into their calculations. One has to wonder whether Fang’s team did anything that might have eliminated confounders that earlier researchers didn’t think of.

Fang offers several possible explanations for their weaker results. They did not include some foods in their analysis that usually are included. They did not include ultraprocessed whole grain foods because it has already been shown that these provide health benefits despite the processing. They also did not include alcohol even though it is generally included as it is highly processed and is an already proven risk. 

Admittedly, the modest findings that Fang reported are still significant, even if not as strong as what has been published elsewhere. A 9% increase in the risk of neurodegenerative disease is still quite undesirable. We find ourselves wondering which of these reports is closer to the actual truth. Keep in mind that the Nova system is a very rough classification system.

Two other parts to Fang’s study need to be mentioned. The first is that the ultraprocessed food group was further categorized into 9 separate subgroups, as mentioned earlier:

  1. Ultraprocessed breads and breakfast foods
  2. Fats, condiments, and sauces
  3. Packaged sweet snacks and desserts
  4. Sugar-sweetened and artificially sweetened beverages
  5. Ready-to eat/heat mixed dishes
  6. Meat/poultry/seafood ready-to eat products (eg, processed meats)
  7. Packaged savory snacks
  8. Dairy-based desserts
  9. Others

The positive associations with mortality were mainly driven by the meat/poultry/seafood-based ready-to-eat products, sugar-sweetened and artificially sweetened beverages, dairy-based desserts, and ultraprocessed breakfast foods.

The second aspect to Fang’s study that needs mention is that along with scoring the study participants’ diets according to this expanded Nova classification, the investigators also scored their diets using the Alternative Healthy Eating Index–2010 score. This will take a little explaining. 

Over the years, researchers have devised numerous systems to evaluate whether diets are healthy or not. Back in 1995 the United States Department of Agriculture (USDA) developed a method called the Healthy Eating Index (HEI). It was a way to assess how closely a person’s diet aligned with the recommendations in the Dietary Guidelines for Americans that the USDA promulgates. “The HEI score captures key nutrients and food groups that reflects current evidence on the dietary components that are healthful.”12 Some individuals might critique these dietary guidelines as having been a compromise between scientists and food-industry lobbyists whose goals were not necessarily congruent, with the former interested in preserving health while the latter group’s assignment was to preserve the profit margins of their clients.13

In response to these 2 competing agendas, the Harvard T.H. Chan School of Public Health created the AHEI purposefully as an alternative to the USDA’s HEI.14 The AHEI uses evidence-based recommendations that incorporate additional components focusing on foods and nutrients that predict the risk of chronic disease. This alternative scoring system works better at predicting risk of chronic disease than the HEI. The AHEI has already been shown to be highly predictive of mortality and chronic disease risk.

For example, in 2011, Akbaraly et al reported that in a group of 7,319 participants, those who scored in the top third of the AHEI scores compared to those in the bottom third had 25% lower all-cause mortality (HR, 0.76; 95% CI, 0.61, 0.95) and a greater than 40% lower mortality from cardiovascular disease (HR, 0.58; 95% CI, 0.37, 0.91).15

Another example of the AHEI’s utility is Chiuve et al’s 2012 study, which scored the diets of 71,495 women from the Nurses’ Health Study and 41,029 men from the Health Professionals Follow-Up Study, using both systems, the HEI and AHEI. Both were predictive of chronic disease, but the AHEI was more accurate.16,17 Think of the AHEI as a measure of food quality in contrast to the Nova system, which is simply a classification of the degree of processing the food has undergone. 

As Fang is a doctoral student at T.H. Chan and his professors were involved in developing the AHEI, it is not surprising that his study’s participants were also evaluated using this dietary scoring for comparison.

In Fang’s analysis of data, it appears that the AHEI score was more predictive of mortality than the Nova scores.  For those participants whose score for consuming ultraprocessed foods was in the upper quartile, they were (modestly) more likely to die than those in the lowest quarter of scores. Risk of dying, however, was not clearly associated with scores between quarters or within quarters. “No consistent associations between ultra-processed foods and mortality were observed within each quarter of dietary quality assessed by the Alternative Healthy Eating Index-2010 score, whereas better dietary quality showed an inverse association with mortality within each quarter of ultra-processed foods.”

The AHEI scores were significant predictors of greater mortality risk. In other words, the AHEI, which gauges a diet by the quality of the foods and the nutrients eaten, is a better measure of dietary impact on mortality and chronic disease than the Nova measure of food processing. Thus, logically or scientifically we might conclude the AHEI is better, but still, there is something about this Nova system and the story of modern food processing that is compelling. 

The concept that ultraprocessed foods lack some element required for good health resonates with the traditions of naturopathic medicine that have long valued natural foods and the assumption that any degree of processing lessens the innate, health-bestowing qualities of the original food. Living foods—plants in a state of growth or retaining their potential for growth, such as vegetables, fruits, seeds, and nuts—are considered of great value. Processing reduces something in foods that is vital for good health. These assumptions are so well-established that we take them as facts. It is easy for us to accept that there is truth to the Nova classification system. The Nova system is also a concept that most patients who seek out naturopathic advice will accept and willingly believe.

It is also a simple story, which will stick in people’s minds, easily turning into a dietary approach that patients will believe and adapt to. It’s easy to believe that the more processed a food is, the less healthy it is. It’s an easy “rule” to incorporate into one’s food choices. It may not be as true as we thought, but any belief that gets people to turn away from the addictive, highly processed foods that have become common fare is probably a good thing.

Categorized Under

Jacob Schor

ND, FABNO

Jacob Schor, ND, FABNO, is a graduate of National College of Naturopathic Medicine (since renamed National University of Naturopathic Medicine), Portland, Oregon, and recently retired from his practice in Denver, Colorado. He served as president to the Colorado Association of Naturopathic Physicians and is a past member of the board of directors of the Oncology Association of Naturopathic Physicians and American Association of Naturopathic Physicians. He is recognized as a fellow by the American Board of Naturopathic Oncology. He serves on the editorial board for the International Journal of Naturopathic Medicine, Naturopathic Doctor News and Review (NDNR), and Integrative Medicine: A Clinician's Journal. In 2008, Dr Schor was awarded the Vis Award by the American Association of Naturopathic Physicians. His writing appears regularly in NDNR, the Townsend Letter, and Natural Medicine Journal, where he is the past Abstracts & Commentary editor.

References

  1. Monteiro CA, Levy RB, Claro RM, Castro IR, Cannon G. A new classification of foods based on the extent and purpose of their processing. Cad Saude Publica. 2010;26(11):2039-2049. 

  2. Monteiro CA, Levy RB, Claro RM, de Castro IR, Cannon G. Increasing consumption of ultra-processed foods and likely impact on human health: evidence from Brazil. Public Health Nutr. 2011;14(1):5-13. 

  3. Monteiro CA, Moubarac JC, Cannon G, Ng SW, Popkin B. Ultra-processed products are becoming dominant in the global food system. Obes Rev. 2013;14 Suppl 2:21-28. 

  4. Moubarac JC, Martins AP, Claro RM, Levy RB, Cannon G, Monteiro CA. Consumption of ultra-processed foods and likely impact on human health. Evidence from Canada. Public Health Nutr. 2013;16(12):2240-2248. 

  5. Suksatan W, Moradi S, Naeini F, et al. Ultra-processed food consumption and adult mortality risk: a systematic review and dose-response meta-analysis of 207,291 participants. Nutrients. 2021;14(1):174. 

  6. Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;384:e077310. 

  7. Barbaresko J, Bröder J, Conrad J, Szczerba E, Lang A, Schlesinger S. Ultra-processed food consumption and human health: an umbrella review of systematic reviews with meta-analyses. Crit Rev Food Sci Nutr. 2024:1-9. 

  8. Jurema Santos GC, de Sousa Fernandes MS, Carniel PG, da Silva Garcêz A, Góis Leandro C, Canuto R. Dietary intake in children and adolescents with food addiction: a systematic review. Addict Behav Rep. 2024;19:100531.

     Bestari FF, Andarwulan N, Palupi E. Synthesis of effect sizes on dose response from ultra-processed food consumption against various noncommunicable diseases. Foods. 2023;12(24):4457.

  9. Martínez Steele E, Baraldi LG, Louzada ML, Moubarac JC, Mozaffarian D, Monteiro CA. Ultra-processed foods and added sugars in the US diet: evidence from a nationally representative cross-sectional study. BMJ Open. 2016;6(3):e009892

  10. Henry Ford Health Staff. Ultra-processed foods are sabotaging your diet. Here’s how to cut them out. Henry Ford Health website. https://www.henryford.com/blog/2023/09/ultra-processed-foods#. Accessed May 22, 2024.

  11. Al-Ibrahim AA, Jackson RT. Healthy eating index versus alternate healthy index in relation to diabetes status and health markers in U.S. adults: NHANES 2007-2010. Nutr J. 2019;18(1):26. 

  12. McGinley-Gieser D. Lobbyists move to weaken the dietary guidelines; help us protect them. American Institute of Cancer Research website. https://www.aicr.org/resources/blog/lobbyists-move-to-weaken-the-dietary-guidelines-help-us-protect-them/. Accessed May 25, 2024.

  13. McCullough ML, Willett WC. Evaluating adherence to recommended diets in adults: the Alternate Healthy Eating Index. Public Health Nutr. 2006;9(1A):152-157. 

  14. Akbaraly TN, Ferrie JE, Berr C, et al. Alternative Healthy Eating Index and mortality over 18 y of follow-up: results from the Whitehall II cohort. Am J Clin Nutr. 2011;94(1):247-253.

  15. Chiuve SE, Fung TT, Rimm EB, et al. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012;142(6):1009-1018. 

Diversified Communications
© 2024 Diversified Communications. All rights reserved.