River flow regimes, controlled by climatic and catchment factors, vary over a wide range of temporal and spatial scales. This hydrological dynamism is important in determining the structure and functioning of riverine ecosystems; however, such hydroecological associations remain poorly quantified. This paper explores and models relationships between a suite of flow regime predictors and macroinvertebrate community metrics from 83 rivers in England and Wales. A two-stage analytical approach was employed: (1) classification of 83 river basins based upon the magnitude and shape (form) of their long-term (1980–1999) average annual regime to group basins with similar flow responses; and (2) examination of relationships between a total of 201 flow regime descriptors identified by previous researchers and macroinvertebrate community metrics for the whole data set and long-term flow regime classes over an 11-year period (1990–2000). The classification method highlighted large-scale patterns in river flow regimes, identifying five magnitude classes and three shape classes. A west-east trend of flow regime magnitude (high-low) and timing (early-late peak) was displayed across the study area, reflecting climatic gradients and basin controls (e.g. lithology). From the suite of hydrological variables, those associated with the magnitude of the flow regime consistently produced the strongest relationships with macroinvertebrate community metrics for all sites and for the long-term regime composite classes. The results indicate that the classification (subdivision) of rivers into flow regime regions potentially offers a means of increasing predictive capacity and, in turn, better management of fluvial hydrosystems. © Crown copyright 2006. Reproduced with the permission of Her Majesty's Stationery Office. Published by John Wiley & Sons, Ltd.