From 2000 to 2009, 300 chemicals were analyzed in water, bed sediment and suspended particulate matter (SPM) at 17 sites throughout the Rhône river network by the regional water authority, providing a very rich and useful dataset. This work focuses on factors influencing the quality of this database. Few studies have been carried out on the analysis of hydrophobic organic contaminants and metals associated with SPM, we thus favoured critical review of those data. Data processing revealed several issues that can weaken the dataset quality. While the constancy of analytical methods is of utmost importance, data processing highlighted major temporal discontinuity for some chemical concentrations, either due to changes of the analytical methods, or due to inappropriate or shifting quantification limits. Furthermore, despite the fact that this dataset refers to a river network, no chemical concentrations were recorded during flood events, biasing any annual average concentration evaluation, especially for contaminants associated with SPM. Thereby, the sampling frequency or strategy should be carefully chosen in order to obtain information on potential variations of concentrations linked to the river hydrological cycle. Finally, sampling methods should be characterized and compared. In a recent study of the Rhone Sediment Observatory we focused on the comparison of different SPM sampling techniques. A difference in the particle size distribution of SPM sampled with the various techniques was noticed, which could have an effect on the concentrations of contaminants associated with SPM. To conclude, this study revealed that factors such as analytical methods, sampling frequency and sampling methods impact concentrations or annual average concentrations of chemicals in SPM, bed sediment and water, and thus impact the monitoring dataset quality. Consequently, the outcome of an environmental risk assessment based on such a monitoring dataset might be significantly affected. Interlaboratory comparison tests, quality controls, as well as a complete documentation on sampling and analytical methods are among possible solutions to build relevant datasets of chemical concentrations in rivers.