Physical inactivity (PI), a leading modifiable cause of disease and injury, is endemic in industrialised nations. Although considerable research has been undertaken in this field, we lack a system to synthesise the research literature to inform policy and identify research needs. The aims of this study were to (1) develop a system to classify physical inactivity intervention studies, (2) examine the distribution of PI interventions published in the peer-reviewed health literature using the system, and (3) consider implications for future research. We developed the Physical Inactivity Matrix (PIM), with 12 intervention points, created by the intersection of two dimensions: the intervention target (individual, physical environment and social/cultural environment) and the activity focus (transport, work/school, leisure and consumer). A formal search of the health research literature identified 529 eligible studies and each was classified into one of the 12 cells of the PIM. Most studies were categorised as: individual-leisure (68%), individual-work/school (12%) or social/cultural environment-leisure (13%). Only 4% targeted the physical environment. The findings of this initial application of the PIM support the call for greater investment in policies, interventions and research that focus on the relationship between the environment and PI, and transportation in particular. There would be merit in establishing the inter-rater reliability of the PIM and applying it to a wider variety of studies, including those published in the transportation and urban planning literatures. The PIM could be a useful tool for monitoring trends in research directions and funding levels over time and across countries.