@Article{info:doi/10.2196/57530, author="Dammas, Shaima and Weyde, Tillman and Tapper, Katy and Spanakis, Gerasimos and Roefs, Anne and Pothos, M. Emmanuel", title="Prediction of Snacking Behavior Involving Snacks Having High Levels of Saturated Fats, Salt, or Sugar Using Only Information on Previous Instances of Snacking: Survey- and App-Based Study", journal="JMIR Med Inform", year="2025", month="Apr", day="23", volume="13", pages="e57530", keywords="high fats, salt, or sugar snacks", keywords="machine learning algorithms", keywords="internet data collection", keywords="just-in-time interventions", abstract="Background: Consuming high amounts of foods or beverages with high levels of saturated fats, salt, or sugar (HFSS) can be harmful for health. Many snacks fall into this category (HFSS snacks). However, the palatability of these snacks means that people can sometimes struggle to reduce their intake. Machine learning algorithms could help in predicting the likely occurrence of HFSS snacking so that just-in-time adaptive interventions can be deployed. However, HFSS snacking data have certain characteristics, such as sparseness and incompleteness, which make snacking prediction a challenge for machine learning approaches. Previous attempts have employed several potential predictor variables and have achieved considerable success. Nevertheless, collecting information from several dimensions requires several potentially burdensome user questionnaires, and thus, this approach may be less acceptable for the general public. Objective: Our aim was to consider the capacity of standard (unmodified in any way; to tailor to the specific learning problem) machine learning algorithms to predict HFSS snacking based on the following minimal data that can be collected in a mostly automated way: day of the week, time of the day (divided into time bins), and location (divided into work, home, and other). Methods: A total of 111 participants in the United Kingdom were asked to record HFSS snacking occurrences and the location category over a period of 28 days, and this was considered the UK dataset. Data collection was facilitated by a purpose-specific app (Snack Tracker). Additionally, a similar dataset from the Netherlands was used (Dutch dataset). Both datasets were analyzed using machine learning methods, including random forest regressor, Extreme Gradient Boosting regressor, feed forward neural network, and long short-term memory. We additionally employed 2 baseline statistical models for prediction. In all cases, the prediction problem was the time to the next HFSS snack from the current one, and the evaluation metric was the mean absolute error. Results: The ability of machine learning methods to predict the time of the next HFSS snack was assessed. The quality of the prediction depended on the dataset, temporal resolution, and machine learning algorithm employed. In some cases, predictions were accurate to as low as 17 minutes on average. In general, machine learning methods outperformed the baseline models, but no machine learning method was clearly better than the others. Feed forward neural network showed a very marginal advantage. Conclusions: The prediction of HFSS snacking using sparse data is possible with reasonable accuracy. Our findings offer a foundation for further exploring how machine learning methods can be used in health psychology and provide directions for further research. ", doi="10.2196/57530", url="https://medinform.jmir.org/2025/1/e57530" } @Article{info:doi/10.2196/66970, author="Berm{\'u}dez-Mill{\'a}n, Angela and P{\'e}rez-Escamilla, Rafael and Segura-P{\'e}rez, Sofia and Grady, James and Feinn VI, S. Richard and Agresta, Hanako and Kim, Dean and Wagner, Ann Julie", title="The Monthly Cycling of Food Insecurity in Latinas at Risk for Diabetes: Methods, Retention, and Sample Characteristics for a Microlongitudinal Design", journal="JMIR Form Res", year="2025", month="Mar", day="28", volume="9", pages="e66970", keywords="food insecurity", keywords="monthly cycling", keywords="type 2 diabetes risk", keywords="quantitative methods", keywords="Latinas", keywords="endocrinology", keywords="nutrition", keywords="nutrition assistance", keywords="micro-longitudinal design", abstract="Background: Food insecurity (FI) is a risk factor for type 2 diabetes (T2D) that disproportionately affects Latinas. We conducted a microlongitudinal study to examine the relationship of monthly cycling of FI and diabetes risk factors. Objective: This study aimed to determine the quantitative methodology, recruitment and retention strategies, predictors of retention across time, and baseline sample demographics. Methods: Participants were adult Latinas living in Hartford, Connecticut who were recruited through a community agency, invited to participate if they were receiving Supplementary Nutrition Assistance Program (SNAP) benefits, screened positive for FI using the 2-item Hunger Vital Sign Screener, and had elevated risk factors for T2D using the American Diabetes Association risk factor test. Using a microlongitudinal design, we collected data twice per month for 3 months (week 2, which is a period of food budget adequacy; and week 4, which is a period of food budget inadequacy) to determine if the monthly cycling of FI was associated with near-term diabetes risk (fasting glucose, fructosamine, and glycosylated albumin) and long-term risk (BMI, waist circumference, and glycated hemoglobin) markers. We determined whether household food inventory, psychological distress, and binge eating mediated associations. We examined Health Action Process Approach model constructs. To assess the relationship between monthly cycling of FI with diabetes risk markers, we used repeated measures general linear mixed models. To assess the role of mediators, we performed a causal pathway analysis. Results: Participant enrollment was from April 1, 2021 to February 21, 2023. A total of 87 participants completed 420 assessments or a mean of 4.83 (SD 2.02) assessments. About half (47/87, 54\%) of the sample self-identified as Puerto Rican, mean age was 35.1 (SD 5.8) years, with 17.1 (SD 11.6) years in the mainland United States. Just under half (41/87, 47.1\%) spoke Spanish only, 69\% (60/87) had no formal schooling, and 31\% (27/87) had less than eighth grade education. Modal household size was 4 including 2 children; 44.8\% (39/87) were not living with a partner. About half (47/87, 54\%) were unemployed, 63.2\% (55/87) reported a monthly income