Human–Nature Interaction Types in Randomized Trials of Health: A Scoping Review and Evidence Map

Abstract

Time in nature is linked to better health, but most evidence is observational and rarely shows what people actually do in nature. We mapped randomized controlled trials (RCTs) of direct nature contact by type of interaction and health domain to guide nature-based health programmes. Following JBI guidance and PRISMA-ScR (protocol on OSF Registries), we searched four databases (October 2026) for English-language RCTs published 2010–2026, re-checked records first excluded on title and abstract against the full text, and screened the reference lists of five earlier reviews. Interventions were coded to seven interaction types and outcomes to six health domains. We included 325 reports of 295 RCTs (about 27,600 participants), 70% published since 2020. Most trials involved active (63%), restorative/therapeutic (32%) or passive (29%) interactions and measured mental (86%) or positive psychological (73%) outcomes; social outcomes appeared in 7%. Wildlife activities were tested in 11 trials and conservation activities in 3. Only 17 trials (6%) tested a mediating mechanism; 46% tested a single session and 87% had no follow-up. Randomized evidence is concentrated in a few interaction types and short-term psychological outcomes. Trials of wildlife and conservation activities, social and longer-term outcomes, under-represented populations and pre-specified mediators are needed.