Abstract
Background and Aims: Problematic internet use, internet addiction, and internet gaming disorder all describe a global phenomenon where individuals have trouble limiting their use of internet to such an extent that their use has negative consequences. Past systematic reviews and meta-analyses have focused on estimating prevalence, but there has been no comprehensive research synthesis of the trajectory of the problem. The research objective was to create a pooled estimate of the persistence of problematic internet use. This review included studies using a longitudinal panel data design with a follow-up of at least a year. Studies had to be published before the end of the year 2017, in peer-reviewed academic journals, using English language. Samples from populations in any country were accepted, given they were of acceptable quality. Methods: A literature search was conducted in Web of Science, Pro Quest, and Scopus. Several definitions of problematic internet use were included. Inverse-variance, random-effect meta-analysis was used to estimate weighted summary means of persistence. Attrition and selection bias was investigated using pre-specified tools, and heterogeneity was assessed in subgroup analysis. Results: Nine studies fit the criteria, all using samples from Asian or Western countries. The aggregate estimate for 1-year persistence it was 50% (CI: 40-61%), but results were heterogeneous. Prevalence and persistence estimates were correlated and generally higher in Asian countries. Methodological differences only explain part of the heterogeneity. Conclusion: All included studies found individuals with persistent problems, but the between-studies variation is substantial.