Sport-for-development programs claim to address key determinants of recreational physical activity participation and subsequent development outcomes in low-income settings. We conducted a natural experiment with pre-post measures taken from women in the 12 villages in Samoa, some of which voluntarily participated in the sport-for-development intervention. The intervention comprised a six-week netball league delivered by local volunteers who attended coaching workshops, received ongoing support from the national governing body and were provided with infrastructure and equipment to conduct local training sessions. Changes in netball participation, recreational physical activity, body composition, mental wellbeing and socio-ecological determinants of physical activity were compared between intervention and comparison villages using a univariate ANOVA. The intervention reached women who participated in little recreational physical activity and had poor physical and mental wellbeing. Program uptake was higher in villages with the strongest social support for netball participation. Local social support and capacity to independently organize netball activities increased. There were concurrent improvements in netball participation, physical activity levels, mental wellbeing and body weight in the intervention villages. Our findings support scaling-up of the intervention in similar settings but preceding this with formative evaluation to identify low active communities that are "primed" to participate in the proposed activity.
Keywords: Pacific Islands; developing countries; exercise; physical activity; sport; sport-for-development.