Historically, fecal indicator bacteria (including total and fecal coliforms and enterococci) have been used as indicators for the presence of bacterial, viral and protozoan pathogens (Savichtchevaa et al., 2005
). These microorganisms are of fecal origin from mammals and birds, and their presence in water may indicate fecal pollution and possible association with enteric pathogens. However, there are major problems with these bacterial indicators, including: short survival in water bodies (McFeters et al., 1974
; McFeters, 1990
); non-fecal source (Scott et al., 2002
; Simpson et al., 2002
); ability to multiply after release into the water column (Desmarais et al., 2002
; Solo-Gabriele et al., 2000
); susceptibility to disinfection processes (Hurst et al., 2002
); an inability to be used to identify the source of fecal contamination (Field et al., 2003
); low levels of correlation with the presence of pathogens (Deetz et al. 1984
; Gerba and Rose 1990
; Jiang et al., 2001
; Griffin et al., 2003
; Jiang and Chu 2004
; Noble and Fuhrman 2005
); and a low sensitivity of detection methods (Horman et al., 2004
; Winfield and Groisman, 2003
). As a result, none of the bacterial indicators currently used meet all the criteria for an ideal indicator. Furthermore, the only detection methodology currently accepted for regulatory purposes (i.e. enterococci) depends upon culture-based growth and enrichment of the target organisms for an incubation period of at least 18 to 24 hours, so current regulatory methods lack the ability to assess the same-day water quality status of tested water bodies.
In the epidemiologic portion of this study (Fleisher et al., in review
), we have shown that bathers had a significantly greater risk of reporting gastrointestinal, respiratory and skin illnesses when exposed to non-point source subtropical recreational marine waters than non-bathers. In the environmental monitoring portion of this study, despite performing a range of FIB and environmental parameter monitoring, we only found a dose response relationship between: a) enterococci measured by membrane filtration vs. report of skin illness; b) 24 hour antecedent rain vs. report of skin illness; and c) water temperature vs. report of acute respiratory febrile illness among the bathers (inverse relationship). Nevertheless, these data suggest that even in the absence of a known point source of sewage contamination, exposure to recreational waters may have public health impacts at subtropical marine beaches.
Enterococci is traditionally used to track GI illness in recreational marine waters (Wade et al., 2003
; Pruss, 1998
; Shuval, 2003
); however, for the current study, enterococci tracked only with skin illnesses. In addition, there was a positive association between skin illness and 24 hour antecedent rain in this study. Enterococci are not typically associated with skin disease; therefore, the correlation with skin illnesses may suggest that environmental conditions that impact enterococci at this recreational beach may also affect an unknown skin pathogen in a similar fashion. Measurements at this study site showed a greater number of enterococci in the water column immediately after storm events (within 6 hours); the hydrology of the site should be evaluated further to understand the sources that could contribute possible pathogens, in particular skin pathogens, a period of 24 hours after a storm event. Also of interest is that the skin pathogen measured as part of the study, S. aureus
, did not show a statistically significant relationship with skin or any other illnesses. Part of the lack of relationship between skin illness and S. aureus
may be due to the lack of virulence of the environmentally isolated strains, patchiness of the detection (only 37% of the water samples were positive), and relatively small sample sizes to evaluate dose response relationships.
The inverse relationship between water temperature and acute respiratory febrile illness is of particular interest. One hypothesis that could potentially explain this inverse relationship is that viruses or bacteria which cause acute respiratory febrile illness in the participants may be sensitive to water temperature. In particular, viruses can be vulnerable to higher water temperatures (Wetz et al., 2004
), therefore perhaps the increase in water temperature in this study resulted in the decreased virulence of viruses which cause respiratory-related illnesses.
The unique feature of the environmental monitoring program implemented during this study was the availability of a wide range of bacteria and environmental parameters data on a per bather basis. This differs from all prior epidemiologic studies which required clustering illness reports to a set of environmental measurements. In this study, the basic unit of measurement was the individual bather rather than the average rates of Illness among many bathers on different study days. This use of an aggregated measure of individual exposures (i.e. rates or rate differences) as the basic unit of measurement in all previous prospective cohort designs would lead to significant misclassification of exposure bias (Fleisher et al., 1991
). Thus, the design of the study reported herein would minimize this source of misclassification of exposure bias. Moreover, the variability of fecal indicator measurements has been well documented (EPA, 2007
; Boehm et al., 2009
), and the sampling strategy implemented in the current study minimizes the impacts from temporal and spatial variations in fecal indicator organisms through collection of a water sample at the time and location of exposure.
Although the health data were based on self-reported symptoms, it should be noted that the bathers and the researchers were inherently blinded to the levels of bacteria in the water and to the other environmental measures, since they were not aware of the results from the monitoring data at the time of exposure. The randomization of individual study participants who reported bathing regularly in recreational marine waters into exposed and unexposed groups solely for the purposes of the study avoided a bias possibly inherent in the prospective cohort designs used in previous recreational bather studies: the hypothesized phenomena that persons who report regular non-bathing might constitute a less healthy group relative to regular bathers, leading to a self-selection bias (Kay et al., 2004
; Fleisher et al., 2006
). While this particular study had sufficient power to resolve the main question of relative health risk between bathers and non-bathers, the relative rareness of the individual diseases required a much larger sample size than that used here to elucidate clear health associations between the alternative markers or methods and reported illnesses. Investigations based on other study designs using the same methods and markers utilized in this study have suggested relationships that were not seen here. Finally, the bather cohort used in this study consisted of regular adult bathers; thus care must be used when interpreting these findings, especially with respect to susceptible subpopulations (such as small children or anyone with a compromised immune system) (EPA, 2007
The disadvantage of the environmental monitoring program used in this study was the fact that the water samples collected included not only the ambient water quality, but also potentially the individual bather’s impacts from their individual shedding. It is not clear whether shedding by an individual can cause illness in that same individual. For example in the case of S. aureus, a person colonized with this bacteria could potentially become self-infected in another area of their body, perhaps in an open wound, through the release of their own bacteria into water in which they swim. So, it is not clear whether exposure to one’s own bacterial flora in recreational marine waters can impact health outcomes.
In addition, practical considerations of the study design (i.e. cost) limited the total number of participants that could be included, and thus limited the power of the study to elucidate finer-scale associations, such as associations between specific alternative assays and specific health outcomes. Such associations have been seen previously for some other epidemiologic studies of different design (Wade et al.,2006
). However, it should also be noted that for these other studies the health associations with alternative assays were observed for beaches with exposure to known point-sources of human fecal contamination. To date there has been relatively little similar investigation of non-point-source beaches that utilize these same alternative assays, but at least one other such study at a non.point-source beach has observed no associations between any of the traditional or alternative FIB assay methods and reported gastrointestinal illness (Colford et al., 2007
). Those observations are supportive of the similar findings the authors report here in this study. Overall, the authors of this study believe that the advantages gained from the epidemiologic study design used in this study outweigh the disadvantages. At a minimum, this study can be used as a comparison against the more traditional study designs used to evaluate recreational waters in the U.S. Similarities in the results between the current epidemiologic design and the traditional study design provide support to the conclusions reached from both.
In conclusion, it is apparent that exposure to even non-point source beaches in subtropical marine environments may have associated human health effects. Furthermore, in these non-point source subtropical marine waters, currently recommended and possibly future recommended fecal indicator bacteria may not be sufficient for the prevention of human illness. Further epidemiologic studies are needed to confirm these results, as well as explore in greater depth possible dose-response relationships between human illness with monitoring organisms and environmental parameters.