Abstract
Shopping cart handles have recently been targeted by the popular press as a highly contaminated public surface. Both retailers and consumers are concerned due to the heavy utilization of the grocery industry within the United States. Recent epidemiological investigations have implicated riding in shopping carts as risk factors for food-borne pathogen infection in infants. Studies have also revealed the existence of food-borne pathogens on the surface of food and packaging as well as cross-contamination from surfaces and hands that have not been effectively cleaned. Additionally, transfer of fecal-borne bacteria from fomite surfaces to hands has been well studied. Taken together, data from all of these studies suggest that a shopper could acquire pathogens from contact with grocery shopping cart handles that have been contaminated. Although inferences have been drawn, no studies to date have been published that have specifically investigated the type and levels of fecal-borne and food-borne bacteria present on grocery shopping cart handles. The purpose of this study was to fill this informational gap by evaluating the prevalence of fecal-bacterial and food-borne pathogen contamination on local shopping cart handles utilizing a combination of molecular methods and traditional cultivation techniques. Additionally, we evaluated if there was a difference in the prevalence of bacteria between traditional grocery establishments and discount/outlet type grocery establishments. Specifically, this study investigated the total aerobic bacterial populations present, as well as identified the presence of Escherichia coli spp., Salmonella spp., and Shiga toxin producing Escherichia coli (STEC) on shopping cart handles from retail grocery stores in the Sacramento Region. A multiplex PCR method was successfully developed to detect the presence of STEC and Salmonella from a mixed culture. This method was used in tandem with PCR to identify generic Escherichia coli as an indicator of fecal contamination. These methods were applied in a survey of 12 stores in three regional locations east of Sacramento. Between March and October 2009, each store was visited five times and ten carts were selected at random and sampled at each visit. In total, 600 bacterial samples were successfully collected and analyzed utilizing PCR, multiplex PCR, and traditional cultivation for this study. Out of 600 samples, one sample (0.17%) tested positive for STEC’s stx1 and one sample tested positive for Salmonella spp. (0.17%). For the fecal contamination indicator test, 582 were found to be positive for E.coli spp. (97%). The total number of aerobic bacteria found on the cart handles varied from 0 to over 53,000 colony forming units. Finally, comparisons of the means of total aerobic counts were made between regional locations and grocery establishment type using a 2-way ANOVA test. Statistical analysis of the mean total aerobic counts revealed no significant difference between regional locations. There was also no statistically significant difference between the two grocery establishment-types tested (P > 0.05).