Correlation still does not mean causation: Mental health and nutritional quality

[sue_frantz]

In an example of archival research, researchers analyzed data from the U.S. National Health and Nutrition Examination Survey for the years 2007 to 2012 (Hecht et al., 2022). They found that after controlling for “age, gender, race/ethnicity, BMI, poverty level, smoking status and physical activity,” (p. 3) survey participants “with higher intakes of UPF [ultra-processed foods] report significantly more mild depression, as well as more mentally unhealthy and anxious days per month, and less zero mentally unhealthy or anxious days per month” (p. 7). So far, so good.

The researchers go on to say, “it can be hypothesised that a diet high in UPF provides an unfavourable combination of biologically active food additives with low essential nutrient content which together have an adverse effect on mental health symptoms” (p. 7). I don’t disagree with that. It is one hypothesis. By controlling for their identified covariates, they address some possible third variables, such as poverty. However, at no place in their article do they acknowledge that the direction can be reversed. For example, it can also be hypothesized that people who are experiencing the adverse effects of mental health symptoms have a more difficult time consuming foods high in nutritional quality. Anyone who battles the symptoms of mental illness or who is close to someone who does knows that sometimes the best you can do for dinner is a hotdog or a frozen pizza—or if you can bring yourself to pick up your phone—pizza delivery.  

They do, however, include reference to an experiment: “[I]n one randomized trial, which provides the most reliable evidence for small to moderate effects, those assigned to a 3-month healthy dietary intervention reported significant decreases in moderate-to-severe depression.” The evidence from that experiment looks pretty good (Jacka et al., 2017), although their groups were not equivalent on diet at baseline: the group that got the dietary counseling scored much lower on their dietary measure than did the group that got social support. Also, those who received social support during the study did, in the end, have better mental health scores and better diet scores than they did at baseline, although all we have are the means. I don’t know if the differences are statistically significant. All of that is to say is that the possibility remains that reducing the symptoms of mental illness may also increase nutritional quality.

Both the Jacka et al. experiment and the Hecht et al. correlational study are freely available.  

You may also want to read the Science Daily summary of the Hecht et al. study where the author (or editor?) writes, “Do you love those sugary-sweet beverages, reconstituted meat products and packaged snacks? You may want to reconsider based on a new study that explored whether individuals who consume higher amounts of ultra-processed food have more adverse mental health symptoms.”

If you’d like to use this in your Intro Psych class, after covering correlations and experiments, ask your students to read the Science Daily summary. Ask your students two questions. 1) Is this a correlational study or an experiment? 2) From this study, can we conclude that ultra-processed foods negatively affect mental health? These questions lend themselves well for use with in-class student response systems (e.g., Clickers, Plickers). Lastly, you may want to share with your students more information about both the Hecht et al. study and the Jacka et al. experiment. If time allows, give your students an opportunity to design an experiment that would test this hypothesis: Improved mental health symptoms causes better nutritional consumption.

References

Hecht, E. M., Rabil, A., Martinez Steele, E., Abrams, G. A., Ware, D., Landy, D. C., & Hennekens, C. H. (2022). Cross-sectional examination of ultra-processed food consumption and adverse mental health symptoms. Public Health Nutrition, 1–10. https://doi.org/10.1017/S1368980022001586

Jacka, F. N., O’Neil, A., Opie, R., Itsiopoulos, C., Cotton, S., Mohebbi, M., Castle, D., Dash, S., Mihalopoulos, C., Chatterton, M. L., Brazionis, L., Dean, O. M., Hodge, A. M., & Berk, M. (2017). A randomised controlled trial of dietary improvement for adults with major depression (the ‘SMILES’ trial). BMC Medicine, 15(1), 23. https://doi.org/10.1186/s12916-017-0791-y