The Health Risks of Bathing in Recreational Waters

Abstract

The European Bathing Directive (2006/7/EC) stipulates water quality standards for recreational bathing waters based on specified limits of faecal indicator organisms (FIOs). Presence of FIOs above the limits is considered to be indicative of poor water quality and to present a risk to bathers’ health. The European Bathing Directive (2006) is to be reviewed in 2020.

We conducted a rapid evidence assessment on recreational bathing waters and gastrointestinal illness (GI) to identify the extent of the literature published since the previous review period in 2003 and to determine whether there is any new evidence which may indicate that a revision to the Directive would be justified.

Overall, 21 papers (from 16 studies), including two RCTs, met the inclusion criteria; 12 were conducted in marine waters and four were conducted in freshwater. Considerable heterogeneity existed between study protocols and the majority had significant methodological limitations, including self-selection and misclassification biases. Moreover, there was limited variation in water quality among studies, providing a limited evidence base on which to assess the classification standards.

Overall, there appears to be a consistent significant relationship between faecal indicator organisms and GI in freshwater, but not marine water studies. Given the apparent lack of relationship between GI and water quality, it is unclear whether the boundaries of the Bathing Waters Directive are supported by studies published in the post-2003 period. We suggest that more epidemiological evidence is needed to disprove or confirm the original work that was used to derive these boundaries for marine waters.

The quality standards in the current EU Bathing Waters Directive (2006/6/EC) are based, in part, on epidemiological research reviewed in 2003 by the World Health Organization, but they are due to be reexamined in 2020. This rapid evidence assessment sought to evaluate the current epidemiological literature that examines the relationship between recreational water use (i.e., exposure to marine water and freshwater recreational waters) and gastrointestinal illness (GI), and to highlight any significant new research and/or evidence gaps, which may help inform future bathing water quality guidelines. Specifically, it focused on literature which presented water quality information based on the concentration of faecal indicator organisms (E.coli and enterococci) and gastrointestinal illness (GI) in order to answer the following research questions:

1. What is the post-2003 evidence for the health risks of recreational bathing in general—and to specific groups of bathers in particular?

2. What is the evidence to support the different classification standards outlined in the European Bathing Directive?

The methodology of the review followed a systematic review process, limited only by searching for studies published from 2003 onwards (hence it is termed a rapid evidence assessment rather than a full systematic review). At least two reviewers were involved in each stage of the review process, and a third reviewer checked any decisions, thus limiting the potential for reviewer error and bias.

Overall, 21 papers (from 16 studies), including two randomised controlled trials and 14 observational studies, met the inclusion criteria of our review. Twelve of these studies were conducted in marine waters (11 were conducted in Mediterranean type or subtropical climates and one in a coastal lagoon), and four were conducted in fresh waters (all in temperate climates). Thus, while it is likely that some of the results from the freshwater studies may be directly applicable to the UK, very few of the reported results for marine studies may be applicable to the temperate British climate.

1. What is the post 2003 evidence for the health risks of recreational bathing in general—and also to specific groups of bathers?

Based on studies included in our review, there is continuing evidence that bathing in recreational water poses some increased risk of GI to bathers compared with non-bathers. Most studies evaluated the risk of bathing in beachgoers of all age groups. Only two studies reported results separately by age group of bathers, and only one recent study investigated the risk of GI among other water users (e.g., in people canoeing, fishing, kayaking, motor boating, or rowing), so the data on these specific population groups remain limited. Interestingly, our review of studies published since 2003 found that:

• There appears to be little or no significant difference between GI in bathers compared with non-bathers at marine beaches.

• In contrast, there appears to be a consistent and significantly higher risk of GI in bathers compared with non-bathers in freshwater sites in temperate climates (up to 3.2 times higher).

• There is some evidence to suggest that increased bather exposure (i.e., head immersion or swallowing water) results in a higher risk of GI, particularly for freshwater bathers.

• There is evidence to suggest that an increase in time spent in water is associated with an increase in GI.

• There is very little evidence on how the risk of GI varies with age.

• There is a lack of recent studies that have evaluated the risk of GI in recreational water users other than bathers (e.g., in people canoeing, fishing, kayaking, motor boating, or rowing).

2. What is the evidence to support the different classification standards outlined in the European Bathing Directive?

It was possible to approximate the water quality in eight of the studies (six marine and two freshwater) against the European Bathing Directive classifications. For the six marine studies, the water quality in one study could be classified as “poor” and in two as “excellent” or “good.” For the remaining three studies, the water quality varied. For the two freshwater studies, both could be classified as “excellent.”

To evaluate current bathing indicator standards, this review considered studies that examined a dose response—a relationship between increasing numbers/density of faecal organisms in the water, either as a continuous measure or as a cut-off value, and increased risk of GI. We also considered studies that reported the risk of GI in waters with differing pollution levels. This evidence was required to investigate the relationship between the concentration of faecal indicator organisms (FIOs) in water and GI, and to infer whether or not the literature supports the European Bathing Directive (2006) boundaries. Our review of studies published from 2003 onwards found that:

• There is little evidence for a significant dose response between faecal indicator organisms and GI in marine water.

• There appears to be a significant dose response between faecal indicator organisms and GI in freshwater.

• Very high levels of pollution due to heavy rainfall and urban run-off or sewage contamination are associated with increased GI.

Overall, it is difficult to draw any firm conclusions from this evidence because of the heterogeneity of study protocols and methodological limitations, including self-selection and misclassification biases. Thus, the various results presented by the study authors could be an artefact of the range of methods used. Moreover, there was limited variation in water quality among studies. In particular, few studies were conducted in “poor” quality water, and none were conducted in “sufficient” quality water, thus providing a limited evidence base on which to assess the classification standards.

However, two methodologically robust studies (randomised controlled trials [RCTs]) identified in our rapid evidence review were well conducted, and their results are likely to be reliable and worthy of mention. One study, conducted in “poor” quality marine water in Florida, United States (semi-tropical climate), found that bathers were almost two times more likely to report an episode of GI following water exposure than non-bathers, although the results were not statistically significant. We note, however, that this study also evaluated other illnesses and that the authors concluded that bathers may be at a significantly increased risk of skin illnesses relative to non-bathers. The other randomised controlled trial was conducted in “excellent” quality freshwater sites in Germany (temperate climate). This study found that the crude relative risk of GI was, significantly, more than two times greater in bathers compared with non-bathers. This increased to more than three and half times when bathers who were exposed to a (defined) higher level of enterococci concentration were compared with non-bathers.

With the methodological limitations of all of the included studies in mind, the following general conclusions may be made:

• Based on 16 studies published since 2003, there appears to be a consistent significant relationship between faecal indicator organisms (used to measure water quality) and GI in freshwater studies, but not in marine water studies.

• Given the apparent lack of relationship between GI and water quality levels meeting different boundaries, it is unclear whether the boundaries of the 2006/7/EC Bathing Waters Directive are supported by studies published in the post-2003 period.

We suggest that more UK epidemiological evidence is needed to disprove or confirm the findings of the original studies that were used to derive these boundaries for marine waters.