BackgroundMany children in low- and middle-income countries lack the stimulation needed to support healthy growth and development. Sensitive interactions between caregivers and infants may promote healthy movement behaviours in infants, which could improve childhood growth and development. However, reliable measures for such interactions require testing in the South African context. The aim of this study was to test the acceptability and feasibility of the headcam caregiver-infant interaction assessment tool in mothers from Soweto, South Africa.MethodsNineteen mother and infant (6–24 months) pairs were asked to wear headcams (first-person observation) while participating in group and individual activities. Detailed instructions on headcam use were provided before and during these activities. Mothers were then asked to use the headcams, as well as photoframe cameras (which provided context of the interactions), in at least three, 5-min mother-infant engagement sessions at home over a 1-week period. Thereafter, focus group discussions (FGDs) were conducted to explore mothers’ experiences of using the tool in the home setting. The feasibility of the headcam mother-infant interaction tool was assessed according to a priori criteria which scored (i) technical reliability of the devices and (ii) usability of the recorded footage. Acceptability was assessed according to emerging themes which were coded from the FGDs using a constant comparison method by two researchers.ResultsThe headcam mother-infant assessment tool was found to be feasible in Soweto, and sufficient data was available to code. Three main themes emerged from the FGD analysis: use of the headcam, using the headcams in the home environment and using the photoframe vs. the headcam. Mothers remarked on the ease of using the tool across daily activities, the normality of their infant’s behaviour during recording and the acceptability by other members of the household. Large amounts of wasted unusable recordings were produced, and challenges related to switching the cameras on and off and to headcam placement were discussed.ConclusionsOur study shows that headcams are both an acceptable and feasible method for assessing mother-infant interactions in Soweto. However, improvements to the usability of the tool and the quality of the data collected should be made prior to future work.