Premium
Dietary compensation in randomized nutrition trials: Can we identify and adjust?
Author(s) -
Carpenter Catherine L,
Robbins Wendie,
Henning Susanne M
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.1153.11
Subject(s) - medicine , randomized controlled trial , psychological intervention , environmental health , calorie , clinical trial , intervention (counseling) , gerontology , psychiatry
Background Nutrition, an essential lifestyle factor that influences health and development of disease, exhibits several methodologic shortcomings in analyses of health‐related parameters. Human nutrition studies range from ecologic assessments to large‐scale multi‐center randomized trials, with randomized trials considered the most informative. Dietary compensation, a common methodologic factor that influences all nutrition trials, can be described as underlying changes in dietary consumption patterns resulting from administration of nutritional interventions. Randomized trials that compare nutritional supplementation in one group to usual diet in another are fairly common. Unless the study is conducted in a controlled in‐patient hospital setting, the addition of nutrients to an underlying diet may result in changes to the underlying diet. Effects resulting from nutritional supplementation may therefore be confounded by underlying dietary changes from consumption of other foods that occur in addition to changes from the nutritional intervention. Aims and Methods To assess dietary compensation, we utilized data from a previous three‐month phase II randomized controlled trial that was conducted on 117 young adult men attending a major university. The intervention group received 75 grams of whole‐shelled walnuts per day compared to controls who avoided tree nuts. Both groups consumed an ad libitum diet. We primarily determined whether walnut supplementation would affect male fertility parameters. To assess whether compensation was occurring, we conducted 24‐hour recalls every 2 weeks using the National Cancer Institute (NCI) ASA24 (Automated‐Self‐Administered 24‐hour Recall) assessment. Results Weight gain is one potential type of compensation that occurs in many trials with supplementation. Results indicate that men in the walnut group did not gain weight compared to the control group (p=0.90). The walnut supplementation group, by consuming 75 grams of walnuts, received almost 500 additional calories per day, suggesting that if weight remained constant, some compensation occurred in the walnut group. However when we assessed compensation patterns utilizing a modified repeated measures analysis of variance model, almost all nutrient patterns did not significantly differ between the intervention compared to the control group. Total fat compensation, marginally significantly different between intervention and control, was added as a covariate to the main effect model. Conclusion The method we developed for dietary compensation in the walnut supplementation and male fertility trial enabled us to identify dietary compensation and adjust for potential influences from dietary compensation. Support or Funding Information California Walnut Commission