Fast Take: Ultraprocessed Foods and Mortality

Fast Take: Ultraprocessed Foods and Mortality


  • Foods described as ultraprocessed contain high levels of refined carbohydrates, added sugar, saturated fats and sodium, and are low in dietary fiber.
  • Ultraprocessed foods can also contain common food compounds such as titanium dioxide, emulsifiers, no- and low-calorie sweeteners, acrylamide and BPA.
  • Researchers analyzed self-reported food records of nearly 45,000 French adults (ages 45 and above) for an average of 7 years.
  • The results showed a positive association between ultraprocessed food consumption and all-cause mortality: for every 10 percent increase of ultraprocessed foods in the diet, the risk for mortality went up by 14 percent.
  • Certain processed foods should be consumed less often than others, because not all processed foods are created equal.

If you’ve seen recent food headlines, you may be wondering what ultraprocessed foods are and what exactly you should know about them. A recent study in JAMA Internal Medicine suggests they may increase our risk of death. It’s no secret that the foods we eat can have an impact on our health, so let’s take a closer look at what the study found.

What Is “Ultraprocessed”?

You’ve likely heard of processed food, but what makes something ultraprocessed? Typically, foods described as ultraprocessed contain additives, preservatives and may be fortified with vitamins and minerals. These foods can also contain high levels of refined carbohydrates, added sugar, saturated fats and sodium, and are low in dietary fiber. In other words, they are not considered nutrient-dense. Examples include foods like sugary drinks, many desserts, ready-to-eat or -heat meals, processed meats, some packaged snacks and even some types of bread. Diets high in these types of foods have been associated with higher incidence of non-communicable diseases such as diabetes, cardiovascular disease and cancer.

The authors rely on the NOVA classification system to determine which foods would be categorized as ultraprocessed. NOVA separates foods into four groups based on level and purpose of food processing, not nutritional value:

  1. Unprocessed or minimally processed
  2. Processed culinary ingredients
  3. Processed foods
  4. Ultraprocessed food and drink products

How Was the Study Done?

Researchers followed nearly 45,000 French adults (ages 45 and above) enrolled in the French NutriNet-Santé Study, an ongoing observational cohort study, for an average of 7 years. Study participants were 73 percent women and had an average age of 57 years at baseline. Every six months, study participants self-reported food intake by recording everything they ate and drank for 24 hours on three non-consecutive days. Each year participants also completed five questionnaires about themselves, including measurements like body weight and body mass index (BMI) and other key sociodemographic and health indicators.

What Did the Study Find?

The results showed a positive association between ultraprocessed food consumption and all-cause mortality: for every 10 percent increase of ultraprocessed foods in the diet, the risk for mortality went up by 14 percent. Those with higher intakes of ultraprocessed foods were younger, more likely to live alone, had a lower income, were less educated, were less physically active, had a higher BMI, and were less likely to adhere to French nutritional recommendations. Due to the nature of the study (e.g., an observational design with only French subjects), the authors caution against generalizing their results and note that more research is necessary in order to confirm causal effects.

Study Strengths and Limitations

Strengths of this study include its large sample size and the fact that participants filled out several dietary records over the course of several years, making it possible to measure dietary changes over time. Researchers also took into account many potential factors that could influence the results by adjusting for confounders like age, sex, monthly income, education level, physical activity, smoking, alcohol use and family health history.

However, there were a considerable number of limitations as well. In addition to the fact that ultraprocessed foods include a wide variety of foods with varying nutritional profiles (some with desirable benefits like breads fortified with folic acid, high-fiber breakfast cereals, and spreads made with healthful vegetable oils), this study lumps them all together in their analysis. Doing so makes it impossible to determine what specific foods or food components are more closely associated with increased mortality risk.

While the study does capture a large sample size, the group is not representative of the broader global population and may not even be comparable to the French people as a whole. Researchers found that their test group’s diet consisted of 14.4 percent ultraprocessed foods of the total weight of food and drink consumed, which was much lower than that of people in similar studies in countries such as Canada, the United States and the United Kingdom. It is likely that the participants are more health-conscious than the general French population because this study was done using a voluntary group of adults with long-term involvement in the French NutriNet-Santé study. These differences mean that the study results shouldn’t be generalized to everyone.

Each food product was categorized into the most likely NOVA group, but the authors disclose that they cannot rule out some misclassification of food items at the individual level. This means some foods that are not ultraprocessed could have been categorized as such, and vice versa.

Another pertinent limitation the authors disclose is the timetable in assessing the incidence of diet-related noncommunicable diseases. These types of diseases develop over decades, and the duration of this cohort study was only approximately seven years, with diet and lifestyle histories of the study participants being limited prior to baseline assessments. This limitation makes it difficult to determine whether health outcomes observed in this study are a result of food intake during the study timeframe or if they are the cumulative effect of a lifetime in which much less is known about participants’ diet quality.

Finally, observational study designs like this one, while useful in generating hypotheses for further study, do not demonstrate cause and effect. In fact, the study authors are careful to point out that “despite accounting for a wide range of confounding factors, we cannot exclude the hypothesis of residual confounding because of the observational design of this study. …Thus, no causality can be established for the observed associations.”

A Closer Look at Ultraprocessed Food Components

When reading a study that suggests eating simple grocery store items could cause serious health consequences, it benefits us to take a closer look at the accused food culprits.

Additives, including titanium dioxide, are flagged by the authors as being potentially linked to chronic intestinal inflammation and carcinogenesis.

Our take: This food ingredient has been deemed safe to use at current standard levels allowed in food production by the U.S. Food and Drug Administration (FDA) and other international public health authorities.

Emulsifiers are noted as having the potential to alter the composition of the gut microbiota, promote low-grade inflammation in the intestine and enhance cancer induction and metabolic syndrome.

Our take: The FDA has also confirmed the safe use of emulsifiers used in foods and beverages. All emulsifiers used in foods have passed a rigorous evaluation set by the FDA and are then certified “generally recognized as safe” (GRAS).

Low-calorie sweeteners consumption is mentioned by the authors as having the potential to alter the gut microbiota and play a role in the onset of type 2 diabetes and other metabolic diseases.

Our take: The evidence cited by the authors in support of these claims is not regarded as high-quality evidence by scientists around the world. All low- and no-calorie sweeteners used in food production and tabletop sweetener packets have been studied extensively by numerous global health authorities to determine safe levels of intake.

Acrylamide. The research group mentions that acrylamide may have carcinogenic properties, although they also underscore that the hypotheses indicating this notion are “speculative.”

Our take: Acrylamide exposure in foods, which occurs at very low levels, has not been associated with negative health effects in people. The FDA does not recommend eliminating any one particular food to avoid or reduce dietary exposure.

Endocrine disruptors (such as BPA). The paper calls out food packaging as being “suspected” to have endocrine-disrupting properties. Endocrine disruptors, which are chemicals that may interfere with the body’s endocrine system and produce adverse effects, can cause endocrine cancers and metabolic diseases like diabetes and obesity.

Our take: All food contact materials, including BPA, are regulated by the FDA for safety. Although some compound migration may occur over time, the detected levels are not high enough to be considered harmful.

Closing Thoughts

There is no denying that certain processed foods should be consumed less often than others. But not all processed foods are created equal: some contribute key vitamins, minerals and nutrients to our diet, and others provide very few if any. A healthy diet emphasizes whole foods like fruits, vegetables, nuts, seeds, whole grains, lean proteins and healthy oils. But that doesn’t mean healthy diets can’t include processed food. A healthy diet need not be a perfect diet. It should be enjoyable, sustainable, and flexible—and include a wide range of whole and processed foods. Healthy diets also need to be practical and take into account the limited amounts of time, culinary expertise and budget that many people face.