Association of Ultra-processed Food and Unprocessed or Minimally Processed Food Consumption With Bowel Habits Among US Adults

Background & Aims: Ultra-processed foods (UPFs) may have a negative impact on bowel habits. We aimed to assess the association between UPF and unprocessed or minimally processed food (MPF) intake and bowel habits among adults in the United States (U.S.). Methods: We performed a cross-sectional study using data from the National Health and Nutrition Examination Survey (2005-2010). We used two 24-hour dietary recalls and, based on the Nova classification, calculated intakes of UPFs and MPFs. Constipation and diarrhea were defined using the Bristol Stool Form Scale and stool frequency. We performed survey-weighted logistic regression and substitution analysis to estimate the odds ratios (ORs) and 95% confidence intervals (CIs). Results: Among 12,716 U.S. adults, there were 1290 cases of constipation and 1067 cases of diarrhea. Median UPF and MPF intakes were 26.5% and 66.2% of total grams per day, respectively. Greater UPF consumption (in % gram/d) was associated with higher odds of constipation (adjusted OR [aOR(Q4 vs Q1)], 2.20; 95% CI, 1.76-2.74) (P-trend < .001) but not diarrhea (aOR(Q4 vs Q1), 0.82; 95% CI, 0.62-1.09) (P-trend = .12). Increased MPF consumption was associated with lower odds of constipation (aOR(Q4 vs Q1), 0.46; 95% CI, 0.370-0.57) (P-trend < .001). Associations with constipation were attenuated after adjusting for diet quality (aOR(Q4 vs Q1), UPF, 1.53; MPF, 0.69). Substituting 10% of UPF intake with an equivalent proportion of MPFs was associated with lower odds of constipation (aOR, 0.90; 95% CI, 0.87-0.93). Conclusions: UPF intake was associated with higher odds of constipation, whereas the odds were lower with greater MPF consumption. The effect of food processing on bowel habits was independent of diet quality.