Background: With increasing recognition of the wider social and environmental determinants of public health, decision-makers and researchers need to access evidence on the effectiveness of ‘upstream’ interventions. One way of accessing at least some of this is by searching for existing reviews. Objectives: This paper aims to inform search terms for systematic reviews of ‘upstream’ interventions and, secondly, to assess the relative value of searching different sources when looking for review level evidence in this area. Methods: We analysed the index terms used to describe ‘upstream interventions’ in 11 databases (across health, social science, physical activity and transport) for over 50 reviews included in a systematic ‘map’ describing research evidence on the effectiveness of ‘upstream’ interventions for childhood obesity and overweight. We analysed sources of included studies to identify those with the richest yield of relevant studies. Searches were sensitive and wide-ranging. We included reviews written in English which investigated the effectiveness of ‘upstream’ interventions on obesity and overweight and ‘obesity-relevant’ behaviours (physical activity, sedentary behaviour and eating). Results: Searches generated 5492 unique ‘hits’. We excluded 5164 studies on abstract, retrieved 328 full texts for a second screening and included over 50 reviews. A search of just three databases would have found over 80% of included studies. The characteristics of the reviews found in these databases were similar to the 20% found from other sources. Index terms for and free-text descriptions of ‘upstream’ interventions were varied. We will present the distribution of terms used and discuss (a) the merits of building searches around specific interventions versus a general ‘upstream intervention’ concept and (b) the implication of using specific terms on the type of evidence found. Conclusions: There may be a few high-yield sources for review level evidence about ‘upstream’ interventions which could be targeted in resource-limited searching. Researchers working in this field could profitably create a pool of search terminology from work on substantive reviews in order to inform and improve future searches for ‘upstream’ interventions.