This paper presents the results obtained from the application of a data integration approach used to study water quality variations in the distribution system. The data integration approach allows the simultaneous analysis of water quality, system operation and maintenance data, supervisory control and data acquisition (SCADA) data, hydraulic model and geographical information systems for a complete analysis and understanding of the parameters that may cause water quality degradation. This research, part of an on-going AwwaRF project #2764 - Integrated Water Quality Management, was conducted in selected areas of 2 large US water utilities, Greater Cincinnati Water Works (Cincinnati, Ohio) and the Denver Water Department (Denver, Colo.). From the available data set, customer complaints were selected for study as these can provide an indication of DS water quality variation, most often of an aesthetic nature. Using the data integration approach with historical databases, about 25% of the complaints could be explained. In Cincinnati, highly probable causes of complaints were identified as repair of main breaks and/or valve operations related to the shut down of water mains. In Denver, these causes included fire hydrant replacements and routine maintenance, spot flushing, main repairs and valve operation for hydraulic testing. Includes 6 references, tables, figures.