Research on physical exercise addiction is becoming more frequent due to the importance of excessive physical activity on health in general. Different studies have investigated the prevalence of risk of exercise addiction (REA) and its consequences. Furthermore, there exist a series of contradictions regarding the relationship between REA and other variables associated with physical training. One goal of this systematic review and meta-analysis consists of analysing possible differences in prevalence, age, general health (mental and physical quality of life, eating disorders) and physical training (hours/week) between groups with REA and non-addicted groups. The Exercise Addiction Inventory (EAI) was used to undertake this research. Research was carried out in electronic databases such as Pubmed, SPORTDiscus or Scopus. Inclusion criteria: Studies were eligible as long as participants were measured with EAI, results showed prevalence of REA and/or EAI score, and the study was observational. Twenty studies met the established eligibility criteria for inclusion in the systematic review, whereas seventeen studies were included in the meta-analysis. Regarding mental quality of life, results showed lower values for the exercise addiction risk group, compared with the non-addicted group. The exercise addiction group was younger than the non-addicted group and dedicated more weekly hours to physical training. Subjects with REA have a lower health profile than those with non-REA. However, more research is required, given the lack of consensus on how to measure exercise addiction and the scarce number of studies to date.