Lower household income has been consistently associated with poorer diet quality. Household food purchases may be an important intervention target to improve diet quality among low income populations. Associations between household income and the diet quality of household food purchases were examined. Food purchase receipt data were collected for 14 days from 202 urban households participating in a study about food shopping. Purchase data were analyzed using NDS-R software and scored using the Healthy Eating Index 2010 (HEI 2010). HEI total and subscores, and proportion of grocery dollars spent on food categories (e.g. fruits, vegetables, sugar sweetened beverages) were examined by household income-to-poverty ratio. Compared to lower income households, after adjusting for education, marital status and race, higher income households had significantly higher HEI total scores (mean [sd] = 68.2 [13.3] versus 51.6 [13.9], respectively, adjusted p = 0.05), higher total vegetable scores (mean [sd] = 3.6 [1.4] versus 2.3 [1.6], respectively, adjusted p < .01), higher dairy scores (mean [sd] = 5.6 [3.0] versus 5.0 [3.3], p = .05) and lower proportion of grocery dollars spent on frozen desserts (1% [.02] versus 3% [.07], respectively, p = .02). Lower income households purchase less healthful foods compared with higher income households. Food purchasing patterns may mediate income differences in dietary intake quality. ClinicalTrials.gov identifier: NCT02073643.

Background: In many studies, it is of interest to identify population subgroups that are relatively homogeneous with respect to an outcome. The nature of these subgroups can provide insight into effect mechanisms and suggest targets for tailored interventions. However, identifying relevant subgroups can be challenging with standard statistical methods. Main text: We review the literature on decision trees, a family of techniques for partitioning the population, on the basis of covariates, into distinct subgroups who share similar values of an outcome variable. We compare two decision tree methods, the popular Classification and Regression tree (CART) technique and the newer Conditional Inference tree (CTree) technique, assessing their performance in a simulation study and using data from the Box Lunch Study, a randomized controlled trial of a portion size intervention. Both CART and CTree identify homogeneous population subgroups and offer improved prediction accuracy relative to regression-based approaches when subgroups are truly present in the data. An important distinction between CART and CTree is that the latter uses a formal statistical hypothesis testing framework in building decision trees, which simplifies the process of identifying and interpreting the final tree model. We also introduce a novel way to visualize the subgroups defined by decision trees. Our novel graphical visualization provides a more scientifically meaningful characterization of the subgroups identified by decision trees. Conclusions: Decision trees are a useful tool for identifying homogeneous subgroups defined by combinations of individual characteristics. While all decision tree techniques generate subgroups, we advocate the use of the newer CTree technique due to its simplicity and ease of interpretation.

Food purchasing is considered a key mediator between the food environment and eating behavior, and food purchasing patterns are increasingly measured in epidemiologic and intervention studies. However, the extent to which food purchases actually reflect individuals’ dietary intake has not been rigorously tested. This study examined cross-sectional agreement between estimates of diet quality and nutrient densities derived from objectively documented household food purchases and those derived from interviewer-administered 24-h diet recalls. A secondary aim was to identify moderator variables associated with attenuated agreement between purchases and dietary intake. Primary household food shoppers (N = 196) collected and annotated receipts for all household food and beverage purchases (16,356 total) over 14 days. Research staff visited participants’ homes four times to photograph the packaging and nutrition labels of each purchased item. Three or four multiple-pass 24-h diet recalls were performed within the same 14-d period. Nutrient densities and Healthy Eating Index-2010 (HEI-2010) scores were calculated from both food purchase and diet recall data. HEI-2010 scores derived from food purchases (median = 60.9, interquartile range 49.1–71.7) showed moderate agreement (ρc = .57, p < .0001) and minimal bias (-2.0) with HEI-2010 scores from 24-h recalls (median = 60.1, interquartile range 50.8–73.9). The degree of observed bias was unrelated to the number of food/beverage purchases reported or participant characteristics such as social desirability, household income, household size, and body mass. Concordance for individual nutrient densities from food purchases and 24-h diet recalls varied widely from ρc = .10 to .61, with the strongest associations observed for fiber (ρc = .61), whole fruit (ρc = .48), and vegetables (ρc = .39). Objectively documented household food purchases yield an unbiased and reasonably accurate estimate of overall diet quality as measured through 24-h diet recalls, but are generally less useful for characterizing dietary intake of specific nutrients. Thus, some degree of caution is warranted when interpreting food purchase data as a reflection of diet in epidemiological and clinical research. Future work should examine agreement between food purchases and nutritional biomarkers. ClinicalTrials.gov, NCT02073643 . Retrospectively registered.

Regular self-weighing has been a focus of attention recently in the obesity literature. It has received conflicting endorsement in that some researchers and practitioners recommend it as a key behavioral strategy for weight management, while others caution against its use due to its potential to cause negative psychological consequences associated with weight management failure. The evidence on frequent self-weighing, however, has not yet been synthesized. The purpose of this paper is to evaluate the evidence regarding the use of regular self-weighing for both weight loss and weight maintenance.

The present study examined income-related household food purchases among a sample of 90 households from the community.