1-D Plot Submission
Partner: Sora Johnson
Title. Percentage of Calories from All Ultraprocessed Foods (UPF) and UPF Subtypes among Cancer Survivors in NHANES (1999-2018)
Legend. Data is from 24-hour dietary recalls for the 1999-2018 NHANES cycles combined, only considering individuals who indicated a prior cancer diagnosis (n=4527). Individual boxplots depict the percentage of calories from UPF for either all types of UPF combined or considering individual subtypes. Median percentages are depicted for each boxplot. Percentages for both all UPF and the subtypes are percentage of calories in the diet as a whole.
1-D Plot Revision
Partner: Sora Johnson
Title. Percentage of Calories from Ultraprocessed Foods Among Cancer Survivors in NHANES (1999-2016).
Legend. Self-reported 24-hour dietary recalls from cancer survivors in NHANES (1999-2016). Violin plots depict percentage of calories from all types of ultraprocessed foods combined and specific types. Food intake for specific types was not mutually exclusive; participant-reported foods could be grouped into all, some, or none of these types. Plots depict density of observations (participants) at different percentage levels. Median percentages depicted for each boxplot as well as lines for the 25th, 50th (median), and 75th percentiles. Outliers (points) are participants with consumption below 25th percentile minus 1.5 times the IQR, or above 75th percentile plus 1.5 times the IQR.
Abstract
Cancer survivors have an increased risk of developing chronic comorbidities due to the toxicity of the treatments they receive [1]. A survivor’s diet is a modifiable life factor known to prevent many of these comorbidities [2-4]. The consumption of ultraprocessed foods (UPF) is thus of concern, since these foods have little nutrients remaining after processing [5]. In the general population, the adverse effects of UPF are well established, while UPF consumption patterns in cancer survivors have not been evaluated [6-13]. We used dietary data from NHANES 1999-2018 to describe the total consumption of UPF and UPF subtypes, among participants who indicated a cancer diagnosis. We found that a median 54.34% of calories came from all UPF, while individual UPF subtype contribution to calories varied widely. There were many high outliers in each subtype despite low medians. Given this, further research is needed on subtype consumption among survivors.
1. American Cancer Society. Long-Term Side Effects of Cancer. Secondary Long-Term Side Effects of Cancer 2022. https://www.cancer.org/treatment/survivorship-during-and-after-treatment/long-term-health-concerns/long-term-side-effects-of-cancer.html.
2. Van Blarigan EL, Fuchs CS, Niedzwiecki D, et al. Association of Survival With Adherence to the American Cancer Society Nutrition and Physical Activity Guidelines for Cancer Survivors After Colon Cancer Diagnosis: The CALGB 89803/Alliance Trial. JAMA Oncol 2018;4(6):783-90 doi: 10.1001/jamaoncol.2018.0126.
3. Hurtado-Barroso S, Trius-Soler M, Lamuela-Raventós RM, Zamora-Ros R. Vegetable and Fruit Consumption and Prognosis Among Cancer Survivors: A Systematic Review and Meta-Analysis of Cohort Studies. Adv Nutr 2020;11(6):1569-82 doi: 10.1093/advances/nmaa082.
4. Schwedhelm C, Boeing H, Hoffmann G, Aleksandrova K, Schwingshackl L. Effect of diet on mortality and cancer recurrence among cancer survivors: a systematic review and meta-analysis of cohort studies. Nutr Rev 2016;74(12):737-48 doi: 10.1093/nutrit/nuw045.
5. Monteiro CA, Cannon G, Moubarac J-C, Levy RB, Louzada MLC, Jaime PC. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutrition 2018;21(1):5-17 doi: 10.1017/s1368980017000234.
6. Askari M, Heshmati J, Shahinfar H, Tripathi N, Daneshzad E. Ultra-processed food and the risk of overweight and obesity: a systematic review and meta-analysis of observational studies. International Journal of Obesity 2020;44(10):2080-91 doi: 10.1038/s41366-020-00650-z.
7. Schnabel L, Kesse-Guyot E, Allès B, et al. Association Between Ultraprocessed Food Consumption and Risk of Mortality Among Middle-aged Adults in France. JAMA Internal Medicine 2019;179(4):490-98 doi: 10.1001/jamainternmed.2018.7289.
8. Beslay M, Srour B, Méjean C, et al. Ultra-processed food intake in association with BMI change and risk of overweight and obesity: A prospective analysis of the French NutriNet-Santé cohort. PLOS Medicine 2020;17(8):e1003256 doi: 10.1371/journal.pmed.1003256.
9. Srour B, Fezeu LK, Kesse-Guyot E, et al. Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (NutriNet-Santé). BMJ 2019:l1451 doi: 10.1136/bmj.l1451.
Abstract
Considering the toxicity of radiation and chemotherapy treatments, and conversely, the protective and healing effects of a healthy diet, it is critical to better understand consumption of ultraprocessed foods among cancer survivors. [1,2]. These foods are high in unhealthy fats, sugars, and salts, and low in dietary fibers and micronutrients: in other words, poor dietary choices for this vulnerable population [3]. We used self-reported dietary data from NHANES (1999-2016) to describe total consumption of ultraprocessed foods as well as specific types. Among 4,527 cancer survivors, a median 54.34% of calories came from all ultraprocessed foods. When considering specific types, amounts varied widely with many outliers of high consumption in Non-Hispanic White participants. Food intake for specific types of ultraprocessed foods was not mutually exclusive, so future research should investigate intake within these types, including patterns of consumption and impacts of sociodemographic factors.
1. American Cancer Society. Long-Term Side Effects of Cancer. Secondary Long-Term Side Effects of Cancer 2022. https://www.cancer.org/treatment/survivorship-during-and-after-treatment/long-term-health-concerns/long-term-side-effects-of-cancer.html.
2. Van Blarigan EL, Fuchs CS, Niedzwiecki D, et al. Association of Survival With Adherence to the American Cancer Society Nutrition and Physical Activity Guidelines for Cancer Survivors After Colon Cancer Diagnosis: The CALGB 89803/Alliance Trial. JAMA Oncol 2018;4(6):783-90 doi: 10.1001/jamaoncol.2018.0126.
3. Monteiro CA, Cannon G, Moubarac J-C, Levy RB, Louzada MLC, Jaime PC. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutrition 2018;21(1):5-17 doi: 10.1017/s1368980017000234.
Keywords
cancer survivors; ultraprocessed foods; NHANES; nutrition; diet quality
Keywords
cancer survivors; ultraprocessed foods; NHANES; nutrition; diet quality
Highlights
- Diet quality in cancer survivors is lacking: survivors in NHANES (1999-2018) got over 50% of calories from ultraprocessed foods.
- Total ultraprocessed food consumption among cancer survivors in NHANES was high (over 50%) but varied when considering subtypes.
Highlights
- Diet quality in cancer survivors is lacking: survivors in NHANES (1999-2018) got over 50% of calories from ultraprocessed foods.
- Total ultraprocessed food consumption among cancer survivors in NHANES was high (over 50%) but varied when considering specific types.
Peer Feedback Review
When providing feedback on your partner’s initial submission, please comment on ways the visual can be improved with respect to the 4 E’s. Keep your responses to 1-2 sentences per principle. Use the template shown here to structure your feedback, which should be posted as a comment below.
Feedback Giver Name: Please enter your name here.
• Evidence: Please enter your comment here.
• Efficiency: Please enter your comment here.
• Emphasis: Please enter your comment here.
• Ethics Please enter your comment here.
Self Reflection
Please reflect on what revisions you have made to your visual and abstract. Respond to the following questions regarding questions you have made in 1-2 sentences each. Use the template shown here to structure your feedback, which should be posted as a comment below.
• What changes did you make when revising your visual?
I changed color to blue and red so that colorblind individuals could see better.
• What changes did you make when revising your general description? Why?
Please enter your comment here.
• What changes did you make when revising your scientific description? Why?
Please enter your comment here.
Feedback Giver Name: Emily Sanchez
• Evidence: Terrible do it again!
• Efficiency: Please enter your comment here.
• Emphasis: Please enter your comment here.
• Ethics Please enter your comment here.
Feedback Giver Name: Sora Johnson
• Evidence: Great visual! I’m impressed by all the components you included in this
• Efficiency: Please enter your comment here.
• Emphasis: Please enter your comment here.
• Ethics Please enter your comment here.
Great visual! I really appreciate the detail in your graphic. For a 1-D plot, there’s a lot of information!
Evidence: The sources are clearly indicated as coming from the NHANES 1999-2018 which is considered to be a reliable source. Something that is unclear about the data time/source is what “NHANES 1999-2018 cycles mean; from my understanding and a quick google search, the NHANES surveys are conducted annually. It would be helpful to clearly indicate if all data between 1999-2018 were included or if the data from the 1999 and the 2018 were included. If it’s just the two years, it may be beneficial to explain the reason for choosing those two years speifically. In terms of labelling, all axises were clearly labelled and eash to read.
Efficiency: The layout of this visualization is very intuitive and an efficient use of the space; for example, labelling just the y-axis of the left-hand-most graph reduces the clutter while not taking away the ability for the reader to read the other three plots in the same row. One thing I may suggest adding is a horizontal bar at the ends of each error bar (before the outliers are listed) so that it is clear to the reader where the error bar extends to and where the outliers begin. The black/gray/white background also adds to the clarity of this graph and maintains accessibility.
Emphasis: What I appreciated about this graph is that the title already explains the graph relatively well, followed by a concise but informative figure legend. The abstract also provided more context (specifically some background information about the NHANES survey and how UPF consumption data in individuals with cancer diagnosis is important), however, it may be beneficial to add a sentence or two to explain how UPF consumption and NCDs/comorbidities are related to clarifying the point of this visual. Something I would consider adding to the figure legend is an indication of the statistical analysis/significance level of the error bars. As noted in the abstract, coming back to the outliers component of the graph, it is definitely something that is a bit distracting and makes the emphasis of the graph a bit diluted; it is unclear if the message is to draw attention to the high variability (lots of outliers), or if it’s to the data distribution in the box. Again, I think this could be resolved by clearly stating the message in the figure legend or the abstract.
Ethics: Generally, based on the topics included (cancer, UPFs, comorbidities, NHANES), I would say the intended audience is probably someone in the realm of academia or someone with some background in NCDs and nutrition. That being said, I believe a layperson would also be able to understand the data in the graphic, but with less context to inform their understanding of the main message(s) of this visualization. I belive there will be an understanding that there is a great variability in the distribution of UPF consumption by foodgroup amongst individuals with a cancer diagnosis surveyed in the NHANES surveys in the included years. There is a possibility that due to the large number of outliers, there may be some critical thoughts about the relationships displayed here. The text definitely complements the visuals, however, it may be helpful to add a sentence or two to clearly state the take-away point of this visualization.
Overall, great visual! Most of my comments were focused on how the text relates to the visualization and about the data itself, but the visual aspects of the graphic were very clear and well-thought out.
What changes did you make when revising your visual?
Based on feedback I received in class, I chose to change the graph type to a violin plot rather than a boxplot, include a description of how to read the violin plot in the figure legend, and also superimpose the jittered outliers on top of the graph; I believe this allows for better visual interpretation of the true distribution and density of the data, including clarifying how many outliers there actually are, but also highlighting some selected characteristics of the outliers, since I chose to color these by self-reported race. This particular characterization of outliers will then flow well into my next 2-D visual and then into my planned multidimensional visual in terms of overall content/theme focus. In response to additional feedback that I received on separating the total UPF consumption plot from the individual types, I also enlarged the graph of total consumption compared to the types and further clarified the non-mutually exclusive nature of the types in my legend.
For changes with both my description above (the visual abstract) and my general and scientific policy briefs, I changed all my wording of “subtypes to “types,” to better clarify what a subtype of ultraprocessed foods was. I realized it made sense to me what I was trying to say, since I have been deeply entrenched in this topic, but it might not come across that way to a general reader. This was also reflected in other wording changes and legend descriptions that I used. Based on Corby’s feedback specifically, I also softened and expanded my description of ultraprocessed foods, and in general tried to draw the reader in more effectively with my first sentences of my 1-D policy briefs.