The purpose of this study was to evaluate whether home-delivered meals, and the frequency of delivery, reduces self-reported falls among homebound older adults. Data come from a randomized parallel three-arm study of 371 older adults on seven Meals on Wheels programs’ waiting lists. Participants were randomly assigned to receive (a) daily meal delivery (n = 139); (b) once weekly, frozen meal delivery (n = 106); or (c) control, remain on the waiting list for meals (n = 126). Participants were surveyed at baseline and 15 weeks post randomization. At follow-up, 36 (28.6%) in the control group, 29 (27.4%) receiving once weekly delivered meals, and 33 (23.7%) receiving daily delivered meals reported a fall (compared with control, daily meal risk ratio [RR] = 0.83, 95% confidence limits [CL] = [0.55, 1.25]; frozen meal RR = 0.96, 95%CL = [0.63, 1.45]). Our study suggests that daily delivered meals may reduce the risk of falls. Additional work is needed to understand the effect of meals on falls, particularly among previous fallers, a high-risk subgroup.