Riverine fisheries in England are under pressure from a variety of activities, including increasing intensification of land-use, urbanisation, rising demands for water abstraction, pollution, proliferation of exotic species, climate change and recreational activities. As a result, the integrity of English rivers has changed. In this study, an attempt was made to measure the ecological health of 22 English lowland rivers from the Thames, Trent and Yorkshire Ouse catchments using a variety of tools. The objective was to modify the Index of Biotic Integrity (IBI) for use on English lowland rivers and compare it with existing indices. A number of diversity indices, Margalef (DMg), Simpson (Dsm) and ShannonWiener (H') were used to evaluate the status of fisheries in the study rivers. The Abundance / Biomass (ABC) method and computer-based multivariate analyses, UPGMA, TWINS PAN, DECORANA, were also used to evaluate the status of fish communities. In addition to these indices, the ABC method and multivariate analyses, the IBI, a multimetric index was also used to evaluate the ecological health of study rivers. The IBI is based on structural and functional attributes of fish communities and is capable of evaluating health and condition of an aquatic ecosystem. The IBI requires a reference condition with which to compare the output. In English rivers, no pristine (reference) sites were considered available, consequently best available data were used to develop a reference condition. In this study, the IBI was modified from Karr (1981), which was based on 12 metrics (community characteristics) of fish assemblages. For the study rivers, 15 metrics which described the status of the fish communities were selected to calculate the IBI. Each metric was scored on a simple scale from 0 (absence) to 5 (high quality). The sum of all the metrics (range 0 - 75) was used to assign sites to qualitative classes of biotic integrity. Six integrity classes on a continuous scale were chosen with the following class boundaries: Excellent (56 - 75), Good (42 - 55), Fair (28 - 41), Poor (16 - 27), Very Poor (1 - 15) and No Fish (0). In the study rivers, the DMg, Dsm and H' indices were unable to measure anthropogenic impacts on fish communities as all these indices were based on structural properties of fish communities. These indices also failed to take account of the presence of juveniles in the fish community in a river. Moreover, these indices were influenced by dominant species abundance and sampling strategies, giving an inaccurate assessment of the status of the fisheries. The ABC method was better at evaluating fish communities than diversity indices as the method considered fish abundance and biomass. However, the method did not include functional components of the fish community and was over influenced by juvenile fishes. Consequently, the ABC method was not considered a good indicator of ecosystem health based on fish assemblages. The UPGMA, TWINSPAN and DECORANA analyses, successfully grouped and separated river reaches with rich or poor fish stocks. These analyses however, did not take into account the functional attributes of the fish communities and were not sufficient to explain the status of a fishery without support from other indices. The IBI assessed the ecological health of the middle and lower reaches of the study rivers more accurately than the other diversity indices, ABC method and multivariate analyses. The selected IBI metrics were able to evaluate many perturbations and disturbances as the metrics represented both structural and functional attributes of fish communities. The DMg, Dsm, H', ABC, UPGMA, TWINSPAN and DECORANA were designed to highlight a specific attribute and lost information during calculation but the IBI included a greater variety of information and produced an appropriate index. Spearman's rank correlation indicated the IBI outputs were more similar to diversity indices than other measures, as significant relationships were found between the IBI and DMg, the IBI and Dsm, and the IBI and H' at a = 0.01 level. Significant relationships were probably due to the use of fish density and abundance in the models. However, this did not mean that all diversity indices and the IBI were similar in measuring ecological conditions of a river, rather it was probably numerical similarity. No significant relationship was found between the IBI and ABC, as the ABC index was a ratio of abundance and biomass while the IBI used absolute values of biomass and abundance separately. All the diversity indices, ABC method and multivariate analyses mentioned reinforced the view that the IBI developed in this study was an appropriate index at evaluating ecological health of the middle and lower reaches of the study rivers. The IBI, however, failed to predict the quality of the fisheries in headwater streams because of the exclusion of salmonid species, minor species and general low species diversity found in these zones. Consequently, it was identified that reference conditions and metrics chosen for the middle and lower reaches of the study rivers were not appropriate to assess the ecological health of headwaters. The existing monitoring programmes of the Environment Agency (EA) for fishery data collection, were considered appropriate for calculating IBIs. Sampling strategies of the EA, i.e. daytime, electric fishing both in summer and winter periods irrespective of lunar cycle and breeding season were also considered acceptable to calculate the IBI. Further research was recommended to test the IBI on a wide range of rivers to assess whether the IBI is appropriate for assessing ecological health of middle and lower reaches of rivers in all regions of the UK. Separate IBIs for headwaters, still waters and estuaries were proposed as these zones / waterbodies have different fish communities. Investigation should be directed at developing a simplified IBI using other cost-effective data sources if suitable resources are not available. It is also recommended that the possibility of including the IBI in wider aquatic resource monitoring programmes (e.g. WFD) be investigated. It is also recommended that the possibility of using the IBI to detect change in the pre and post implementation periods of any management action or anthropogenic activity be investigated. Research is also needed to integrate the IBI with other bioassessment methods (e.g. Habitat index, Diatom index, Microinvertebrate index, Chemical index and GQA index). For more effective application and understanding, the IBI may be built into a GIS (Geographical Information System) environment. It is suggested that a suitable computer package be developed to simplify calculations of the IBI. The interpretation should however, be carried out by the fishery manager or scientist.