Biological invasions are regarded as one of the largest threats to native biodiversity. The eradication of non-native parasites by culling of hosts are a controversial conservation strategy, particularly when using indiscriminate methods involving whole ecosystem collateral damage. While short-term effects are abundantly documented, long-term surveys are needed to detect potential wider ecosystem effects. Here, we report a six-year study on effects of the piscicide rotenone on invertebrate communities from a Norwegian water course using a Before-After-Control-Impact design. Kick-net samples of benthic invertebrates were collected from three lentic sampling stations and two lotic stations two to four times per year in both a control and a treated watershed. In general, only relatively minor short-term effects immediately after the treatment on species turnover, measured as temporal beta-diversity, of benthic invertebrates were observed both in lentic and lotic locations. However, the lotic fauna was temporarily severely negatively affected following a period of rotenone exposure from an upstream lake. Species turnover co-varied markedly between control and treatment locations, indicating that natural environmental variation override effects of rotenone treatment. Likewise, the abundance of invertebrate taxa varied considerably both over time and between control and treatment locations. Our study indicates minor short-term (i.e. < one month) or long-term (i.e. four years) effects of rotenone treatment on benthic invertebrates, but severe effects on the lotic fauna eight months after treatment. However, long-term effects are likely to be taxa-specific and vary depending on habitat connectivity and thus potential for re-colonization and will differ among locations and among taxa.