Microbial spoilage of salmon occurs during extended refrigerated storage and is often accompanied by unpleasant aromas. When spoilage is detected, it is assumed that consumers will reject the product for consumption. Because sensory panels of trained individuals or consumers are expensive and labor intensive, identification of microbiological or chemical indicators to characterize the extent to which fish has spoiled is needed when experimental process and storage treatments are being evaluated. A consumer panel of 53 senior citizens (60 to 85 years of age) evaluated in duplicate raw salmon subjected to 10 storage conditions, and the fish quality was targeted to range from fresh to very spoiled. This population group was chosen because they would be expected to have a greater prevalence of olfactory impairments and higher odor thresholds than the general population; in turn, a shorter safety margin or time period between product rejection due to spoilage and the generation of Clostridium botulinum toxins would be likely. Low hedonic scores for aroma and overall acceptance (2 or 3 of 9), corresponding to "dislike very much" to "dislike moderately," did not equate with unwillingness to prepare the sample for consumption by up to seven panelists (13%) when the product was presumed to have already been purchased. Despite these outliers, significant models (P = 0.0000) were developed for the willingness of consumers to prepare the sample for consumption and the sample's aerobic and anaerobic microbiological populations and two volatile peaks with Kovats indices of 640 and 753. However, these models revealed that the levels of microbiological and chemical markers must be very high before some consumers would reject the sample; hence, spoilage detection by smell would likely not be an adequate safeguard against consuming salmon in which C. botulinum toxin had been generated.