THURS-096 - Adapting Infant Feeding Observations for a Virtual World: Insights and Innovations

Brief Abstract Summary: Learn about the feasibility and validity of remote assessments of infant feeding observations. The COVID-19 pandemic necessitated adaptation of in-person observational methods for remote and virtual settings but also provided a natural experiment for evaluating the effectiveness of remote assessments. This study was a secondary analysis of studies of infant feeding and growth wherein both in-person and remote observations of mother-infant feeding interactions occurred. Dyads were video recorded while feeding in their homes before the pandemic or virtually on Zoom during the pandemic. We also developed methods for assessing infant intake by analyzing photos of bottles taken before and after the feeding. No significant differences were seen between in-person and virtual observations, and analysis of bottle photos provided reliable assessments of bottle weight. Remote infant feeding assessments may increase the accessibility of infant feeding studies without compromising data quality.

Detailed abstract description: Observing parent-infant feeding interactions allows for objective assessment of parent feeding practices and infant eating behaviors but also offers insights into the quality of early relationships and broader cognitive and socioemotional development. Despite their value, traditional in-person observations can be burdensome and limit participant diversity. Laboratory observations, while controlled, may lack ecological validity, and home-based observations, though convenient, can suffer from lower quality due to environmental factors. Additionally, in-person methods are less accessible to families with socioeconomic or logistical challenges. Advances in video conferencing provide a feasible alternative, allowing remote observations that reduce burden and enhance accessibility without sacrificing data quality. The COVID-19 pandemic highlighted the importance of remote assessments and accelerated the adoption of these technologies, making video conferencing more familiar and widely available. Remote assessments keep immunocompromised individuals (e.g., infants) safe from infectious diseases but may also provide more ecologically valid data by allowing for the evaluation of infant behavior within familiar home environments. To this end, we describe our experience adapting feeding observation protocols to a virtual format and evaluate whether these protocols yield data comparable to that collected in person. Data come from secondary analyses of infant feeding and growth studies. Mothers (n=76) and their young infants (age 12.7 ± 6.1 wk) were video recorded while feeding in their homes before the pandemic (n=36) or virtually on Zoom during the pandemic (n=40). Mother and infant contributions to the feeding interactions were assessed using the Nursing Child Assessment Parent-Child Interaction Feeding Scale. No significant differences were seen between in-person and virtual assessments for mothers’ sensitivity to infant cues, responsiveness to distress, socioemotional growth fostering, and cognitive growth fostering or infants’ clarity of cues and responsiveness to the mother. Within a subsequent study, we developed methods for assessing infant intake by analyzing photos of bottles taken before and after the feeding. Research assistants took digital photographs of infant feeding bottles with varying amounts of milk (n=100) and assessed milk weight using an adapted Remote Food Photography Method (RFPM). Specifically, image processing software was used to measure milk area in pixels. Actual milk weight was then regressed on estimated milk area and bottle parameters to create an equation to estimate milk weight from digitally derived data. The mean difference between RFPM estimated milk weight and actual milk weight was equivalent within 5% equivalence bounds via two one-sided tests of equivalence. These findings suggest the quality of data from remote observations of mother-infant feeding interactions is comparable to the observations done in person.