INFECTIOUS DISEASE TRANSMISSION

For a host to become infected with an infectious disease, a sequence of events – referred to as the Chain of Infection – needs to occur (United States Centers for Disease Control & Prevention, 2015). A pathogenic organism leaves its reservoir through a portal of exit, reaches a host via a mode of transmission, entering through a portal of entry to cause infection and disease in a susceptible host. Infection control strategies are focused on breaking one or more links in the chain.

Potential reservoirs for marine bathing-related infection include seawater – with or without anthropogenic input (Kay et al., Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994; Wade et al., Reference Wade, Pai, Eisenberg and Colford2003; Tolba et al., Reference Tolba, Loughrey, Goldsmith, Millar, Rooney and Moore2008; Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013), beach sand and sediment (Solo-Gabriele et al., Reference Solo-Gabriele, Wolfert, Desmarais and Palmer2000; Beversdorf et al., Reference Beversdorf, Bornstein-Forst and McLellan2007; Yamahara et al., Reference Yamahara, Layton, Santoro and Boehm2007), zooplankton (Huq et al., Reference Huq, Colwell, Rahman, Ali, Chowdhury, Parveen, Sack and Russek-Cohen1990), fellow bathers (Plano et al., Reference Plano, Garza, Shibata, Elmir, Kish, Sinigalliano, Gidley, Miller, Withum, Fleming and Solo-Gabriele2011) and animals (Wang et al., Reference Wang, Solo-Gabriele, Abdelzaher and Fleming2010). Gastrointestinal pathogens, for example norovirus and E. coli, leave human and animal hosts respectively through faeces (Ihekweazu et al., Reference Ihekweazu, Barlow, Roberts, Christensen, Guttridge, Lewis and Paynter2005), whereas skin shedding of methicillin-resistant Staphylococcus aureus may potentially increase the risk to fellow bathers (Tolba et al., Reference Tolba, Loughrey, Goldsmith, Millar, Rooney and Moore2008; Plano et al., Reference Plano, Garza, Shibata, Elmir, Kish, Sinigalliano, Gidley, Miller, Withum, Fleming and Solo-Gabriele2011, Reference Plano, Shibata, Garza, Kish, Fleisher, Sinigalliano, Gidley, Withum, Elmir and Hower2013). The mode of transmission for gastrointestinal pathogens is faeco-oral – once organisms leave an infected host subsequent ingestion can lead to infection. In addition to faeco-oral transmission, marine bathers may become ill through inhalation of pathogens in marine aerosols – leading to respiratory tract disease – and through direct contact with skin, ears and eyes.

Numerous host factors determine susceptibility to marine-derived pathogens and the severity of any subsequent illnesses. Breaches of innate defences (e.g. skin ulcers and penetrating trauma) increase the risk of bathing-related infection (Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013; Janda & Abbott, Reference Janda and Abbott2014); subsequent disease risk depends on age (Lepesteur et al., Reference Lepesteur, McComb and Moore2006; Wade et al., Reference Wade, Calderon, Brenner, Sams, Beach, Haugland, Wymer and Dufour2008); co-morbid conditions such as diabetes mellitus and impaired immunity (Frank et al., Reference Frank, Littman, Alpers and Hallauer2006; Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013); and the presence of immunity from previous infection (Prieto et al., Reference Prieto, Lopez, Juanes, Revilla, Llorca and Delgado-Rodriguez2001) or vaccination. The ability of an organism to cause disease, and the subsequent severity, is termed ‘virulence’; some pathogens, such as noroviruses, generally cause mild self-limiting illness, whereas ingestion of a limited number of Verotoxin-producing E. coli can lead to severe illness and even death.

The development of infectious disease in man is determined by complex interactions between the host, pathogens and the environment (Scholthof, Reference Scholthof2007). This interaction is apparent in marine bathing; host behaviour is associated with the illness, with activities that increase duration of exposure, and swallowing of water increasing risk (Wade et al., Reference Wade, Pai, Eisenberg and Colford2003; Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012; Harder-Lauridsen et al., Reference Harder-Lauridsen, Kuhn, Erichsen, Mølbak and Ethelberg2013); increased risk is suggested with surfing activities, diving and athletic events (Bradley & Hancock, Reference Bradley and Hancock2003; Wade et al., Reference Wade, Pai, Eisenberg and Colford2003; Schijven & de Roda Husman, Reference Schijven and de Roda Husman2006; Harder-Lauridsen et al., Reference Harder-Lauridsen, Kuhn, Erichsen, Mølbak and Ethelberg2013). Environmental conditions such as eutrophication, sunlight and rainfall will all determine the growth and fate of micro-organisms and so alter the probability of disease (Yau et al., Reference Yau, Schiff, Arnold, Griffith, Gruber, Wright, Wade, Burns, Hayes and McGee2014). In a single study, swallowing water was a risk factor for gastrointestinal illness only in the presence of high groundwater discharge and low solar radiation (Yau et al., Reference Yau, Schiff, Arnold, Griffith, Gruber, Wright, Wade, Burns, Hayes and McGee2014).

Bradley & Hancock (Reference Bradley and Hancock2003) elegantly demonstrated an example of the complex host-pathogen-environment relationship in marine bathing. During meteorological conditions with greater wave height; this is associated with higher numbers of bathers participating in high risk activity (surfing) and where levels of faecal indicator bacteria were significantly higher than during calm weather (Bradley & Hancock, Reference Bradley and Hancock2003).

SOURCES AND SURVIVAL OF PATHOGENS IN THE MARINE ENVIRONMENT

The predominant sources of human pathogens in the marine environment in the UK are considered to be from: human sewage, whether through poorly connected domestic sewers or combined sewage outflows or treated waste; agricultural run-off; urban area run-off; and animal and bird faeces (Environment Agency, UK, 2014).

Faecal indicator bacteria (FIB) are a limited number of organisms (typically Coliforms, E. coli, faecal Streptococci and Enterococci) used as markers for the presence of sewage containing human faeces. Not necessarily pathogenic per se, FIB are used as a proxy measure for the vast number of pathogens potentially present in human waste (Wade et al., Reference Wade, Calderon, Brenner, Sams, Beach, Haugland, Wymer and Dufour2008). Measurement of these organisms is undertaken widely and their association with disease described in areas with point sources of microbial pollution, but the discharge of pathogenic organisms into the sea and their persistence, survival and subsequent virulence are poorly understood. Microbial source tracking has demonstrated potential in identifying the sources of pathogen in the marine environment (Rusinol et al., Reference Rusinol, Fernandez-Cassi, Hundesa, Vieira, Kern, Eriksson, Ziros, Kay, Miagostovich and Vargha2014).

Current established methods for measuring FIB are based on culturing organisms; therefore requiring a minimum of 24–36 h to provide results. Novel molecular methods, specifically quantitative Polymerase Chain Reaction (qPCR), can provide measures of microbial contamination in bathing waters within a few hours and therefore have the potential to inform short-term public health interventions such as beach closures (Oliver et al., Reference Oliver, van, Kay, Heathwaite, Porter, Fleming, Kinzelman, Connolly, Cummins and McPhail2014). There is some epidemiological evidence linking qPCR measures of FIB and illness (Wade et al., Reference Wade, Sams, Brenner, Haugland, Chern, Beach, Wymer, Rankin, Love and Li2010), but further studies would be required if slower, culture-based measures are to be replaced (Oliver et al., Reference Oliver, van, Kay, Heathwaite, Porter, Fleming, Kinzelman, Connolly, Cummins and McPhail2014).

Many human pathogens have been identified in seawater and the marine environment, including: viruses – adenovirus (Pina et al., Reference Pina, Puig, Lucena, Jofre and Girones1998; Jiang & Chu, Reference Jiang and Chu2004; Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007), noroviruses (Victoria et al., Reference Victoria, Rigotto, Moresco, De Abreu Corrêa, Kolesnikovas, Leite, Miagostovich and Barardi2010), enteroviruses (Pina et al., Reference Pina, Puig, Lucena, Jofre and Girones1998; Jiang & Chu, Reference Jiang and Chu2004) and hepatitis A (Jiang & Chu, Reference Jiang and Chu2004); bacteria – S. aureus (including methicillin-resistant Staphlylococcus aureus (MRSA)) (Plano et al., Reference Plano, Shibata, Garza, Kish, Fleisher, Sinigalliano, Gidley, Withum, Elmir and Hower2013), Bacteroides spp. (Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007), Clostridia (Wade et al., Reference Wade, Sams, Brenner, Haugland, Chern, Beach, Wymer, Rankin, Love and Li2010), also demonstrated in sand (Heaney et al., Reference Heaney, Sams, Dufour, Brenner, Haugland, Chern, Wing, Marshall, Love, Serre, Noble and Wade2012; Plano et al., Reference Plano, Shibata, Garza, Kish, Fleisher, Sinigalliano, Gidley, Withum, Elmir and Hower2013), Pseudomonas spp. (Khan et al., Reference Khan, Ahsan, Taylor and Kogure2010); protozoa – Cryptosporidia in seawater (Fayer et al., Reference Fayer, Graczyk, Lewis, Trout and Farley1998) and giardia in the faeces of marine mammals (Measures & Olson, Reference Measures and Olson1999).

Salmonella bacteria have been found in small numbers of samples of bathing water classified using FIB measurements as excellent by European bathing water standards; Salmonella levels were positively correlated with FIB levels (Efstratiou & Tsirtsis, Reference Efstratiou and Tsirtsis2009).

In addition to pathogens from sewage, run-off and animal sources, there is potential for colonized individuals to infect others with potentially pathogenic organisms. Under experimental conditions mimicking marine bathing, individuals can release 10⁵–10⁶S. aureus per 15 min immersion; with 15–20% of isolates being resistant (MRSA), capable of causing serious soft-tissue infections (Plano et al., Reference Plano, Garza, Shibata, Elmir, Kish, Sinigalliano, Gidley, Miller, Withum, Fleming and Solo-Gabriele2011). Plano et al. (Reference Plano, Shibata, Garza, Kish, Fleisher, Sinigalliano, Gidley, Withum, Elmir and Hower2013) demonstrated a correlation between bather density and concentrations of S. aureus in sub-tropical recreational waters, with levels of 2–780 colony forming units (CFUs) per mL (Plano et al., Reference Plano, Shibata, Garza, Kish, Fleisher, Sinigalliano, Gidley, Withum, Elmir and Hower2013). Experimental evidence suggests that S. aureus may be able to persist in seawater for several days (Fujioka & Unutoa, Reference Fujioka and Unutoa2006). Further experimental work suggests that individuals shed in the region of 1 million enterococci per bathing episode (Elmir et al., Reference Elmir, Wright, Abdelzaher, Solo-Gabriele, Fleming, Miller, Rybolowik, Peter, Pillai, Cooper and Quaye2007, Reference Elmir, Shibata, Solo-Gabriele, Sinigalliano, Gidley, Miller, Plano, Kish, Withum and Fleming2009) and potentially up to the same number of Bacteroidales – a potential indicator of human contamination (Elmir et al., Reference Elmir, Shibata, Solo-Gabriele, Sinigalliano, Gidley, Miller, Plano, Kish, Withum and Fleming2009).

In addition to faecal pathogens from external sources, indigenous marine bacteria are capable of causing disease in humans; examples include Vibrio spp. (Huq et al., Reference Huq, Colwell, Rahman, Ali, Chowdhury, Parveen, Sack and Russek-Cohen1990; Andersson & Ekdahl, Reference Andersson and Ekdahl2006; Frank et al., Reference Frank, Littman, Alpers and Hallauer2006; Kushawaha et al., Reference Kushawaha, Mobarakai, Cooper, Rose and Awasum2010), Shewanella (Dominguez et al., Reference Dominguez, Vogel, Gram, Hoffmann and Schaebel1996; Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013; Srinivas et al., Reference Srinivas, Pillai, Vinod and Dinesh2015) and mycobacteria (Cheung et al., Reference Cheung, Fung, Wong and Ip2010). In a review of 171 patients from a coastal area of South Korea diagnosed with necrotizing fasciitis – a potentially fatal soft-tissue infection – marine bacteria were considered to be the causal pathogen in two-thirds of cases (Park et al., Reference Park, Jung, Jung, Shin and Hwang2009).

SURVEILLANCE OF INFECTIOUS DISEASES

Public health surveillance is defined as “the continuous, systematic collection, analysis and interpretation of health-related data needed for the planning, implementation, and evaluation of public health practice” (World Health Organization, 2015b); robust systems are fundamental to public health practice. Surveillance data are used to identify trends, both short- and long-term; identify disease outbreaks and emerging threats; provide hypotheses of causal associations; and provide policymakers with evidence used to prioritize resources. Surveillance can take many forms – both formal and informal – and recent advances in genomics have the potential to provide public health authorities with additional information with which to identify previously elusive shared exposures among cases of infectious disease (Dallman et al., Reference Dallman, Byrne, Ashton, Cowley, Perry, Adak, Petrovska, Ellis, Elson and Underwood2015).

Many European countries have well-developed infectious disease surveillance systems; as in many countries, in the UK registered medical practitioners are legally required to notify certain infectious diseases – whether suspected or confirmed. Notifiable diseases listed in the UK Public Health (Control of Disease) Act 1984 include several relevant to marine exposure, including Verotoxin-producing E. coli (Ihekweazu et al., Reference Ihekweazu, Barlow, Roberts, Christensen, Guttridge, Lewis and Paynter2005), Vibrio cholerae (Huq et al., Reference Huq, Colwell, Rahman, Ali, Chowdhury, Parveen, Sack and Russek-Cohen1990) and Campylobacter spp. (Harder-Lauridsen et al., Reference Harder-Lauridsen, Kuhn, Erichsen, Mølbak and Ethelberg2013). Despite robust systems being in place, cases reported represent only a small proportion of persons infected, particularly if their illness is mild and self-limiting; based on data from UK primary care, it is estimated that for every case of gastrointestinal infection reported to national surveillance, there are 10 primary care consultations and 147 cases in the community (Tam et al., Reference Tam, Rodrigues, Viviani, Dodds, Evans, Hunter, Gray, Letley, Rait and Tompkins2012). In addition to physician reporting, there is a reporting system which collects laboratory confirmed infections to public health; as described previously, this will only detect the tip of the ‘clinical iceberg’ in milder disease, but would be expected to detect a larger proportion of more severe infection such as Verotoxin-producing E. coli that can cause bloody diarrhoea and haemolytic-uraemic syndrome (Heymann, Reference Heymann2008).

Formal notification provides only part of the information used to monitor infectious diseases in the UK; many other sources of data are available. In England a syndromic surveillance team analyses health data from several sources in real-time, allowing rapid dissemination of trend data to identify emerging threats (Public Health England, 2014b); syndromes pertinent to marine exposure include gastroenteritis and acute respiratory infection. Syndromic surveillance has demonstrated research utility in the field of environmental and human health (Elliot et al., Reference Elliot, Hughes, Hughes, Locker, Brown, Sarran, Clewlow, Murray, Bone and Catchpole2013).

In addition to formal surveillance by public health authorities, additional sources of data can provide insights into the emergence of infectious disease outbreaks, particularly in areas where robust systems are absent. In 2003, the World Health Organization (WHO) was alerted to the emergence of the SARS epidemic in Guangdong Province, China by an early warning tool searching global media reports; this method detected the emergence of SARS significantly in advance of formal surveillance (Heymann & Rodier, Reference Heymann and Rodier2004; Keller et al., Reference Keller, Blench, Tolentino, Freifeld, Mandl, Mawudeku, Eysenbach and Brownstein2009). In additional to public health bodies, stakeholders (including charities), have set up surveillance systems. For example, Surfers Against Sewage (SAS), a UK environmental charity, collects reports of illness among bathers/surfers, and has conducted a self-reported beach user health survey in collaboration with the European Centre for Environment and Human Health (ECEHH; http://www.ecehh.org) in the UK; there is an antecedent to the use of online surveys to record water-associated illness (Turbow et al., Reference Turbow, Kent and Jiang2008).

Identifying infectious disease outbreaks related to recreational use of the marine environment poses challenges as exposed populations quickly disperse, particularly tourists. In response to notified cases of infectious diseases in the UK, questionnaires are administered in an attempt to identify shared exposures and therefore potential sources of disease; for example, persons diagnosed with Verotoxin-producing E. coli in England are specifically asked about any site and type of exposure to seawater. In 2006, an outbreak of seven cases of E. coli was identified associated with playing in the same beach stream (Ihekweazu et al., Reference Ihekweazu, Barlow, Roberts, Christensen, Guttridge, Lewis and Paynter2005). By their nature, diseases with longer and less precise incubation periods, such as some types of viral hepatitis, often identify a greater number of potential causal exposures; apparent clustering of cases related to a single exposure therefore becomes harder to identify.

The use of questionnaire data to identify epidemiological links requires considerable resource and cooperation from cases, often introducing additional delay in the process of identifying common exposures. Novel molecular techniques, such as whole-genomic sequencing, are beginning to provide opportunities to rapidly identify common exposures among populations that are temporally and geographically dispersed (Dallman et al., Reference Dallman, Byrne, Ashton, Cowley, Perry, Adak, Petrovska, Ellis, Elson and Underwood2015). Nevertheless this approach may be negated by outbreaks caused by several pathogens released following a particular environmental or anthropogenic event (Harder-Lauridsen et al., Reference Harder-Lauridsen, Kuhn, Erichsen, Mølbak and Ethelberg2013).

EVIDENCE FOR THE ASSOCIATION BETWEEN MARINE BATHING AND INFECTIOUS DISEASE

Numerous studies have demonstrated an increased risk of gastrointestinal illness (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993; Kay et al., Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994; Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007; Wade et al., Reference Wade, Calderon, Brenner, Sams, Beach, Haugland, Wymer and Dufour2008; Fleisher et al., Reference Fleisher, Fleming, Solo-Gabriele, Kish, Sinigalliano, Plano, Elmir, Wang, Withum, Shibata, Gidley, Abdelzaher, He, Ortega, Zhu, Wright, Hollenbeck and Backer2010; Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012; Yau et al., Reference Yau, Schiff, Arnold, Griffith, Gruber, Wright, Wade, Burns, Hayes and McGee2014), ear infections (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993; Fleisher et al., Reference Fleisher, Kay, Salmon, Jones, Wyer and Godfree1996; Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012; Wade et al., Reference Wade, Sams, Beach, Collier and Dufour2013), eye infections (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993; Fleisher et al., Reference Fleisher, Kay, Salmon, Jones, Wyer and Godfree1996; Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012), respiratory illness (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993; Prieto et al., Reference Prieto, Lopez, Juanes, Revilla, Llorca and Delgado-Rodriguez2001; Fleisher et al., Reference Fleisher, Fleming, Solo-Gabriele, Kish, Sinigalliano, Plano, Elmir, Wang, Withum, Shibata, Gidley, Abdelzaher, He, Ortega, Zhu, Wright, Hollenbeck and Backer2010; Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012) and skin complaints (Prieto et al., Reference Prieto, Lopez, Juanes, Revilla, Llorca and Delgado-Rodriguez2001; Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007; Fleisher et al., Reference Fleisher, Fleming, Solo-Gabriele, Kish, Sinigalliano, Plano, Elmir, Wang, Withum, Shibata, Gidley, Abdelzaher, He, Ortega, Zhu, Wright, Hollenbeck and Backer2010) associated with marine swimming – typically defined as head immersion – compared with non-swimmers. The overall incidence of any illness among swimmers in historical studies includes 2–3% (Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012), 4.5–6.1% (Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007), 7.5% (Prieto et al., Reference Prieto, Lopez, Juanes, Revilla, Llorca and Delgado-Rodriguez2001), 14.8% (Kay et al., Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994) and 24% (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993), with the relative risk of gastrointestinal illness compared with non-swimmers quantified at approximately 1.5 times higher (Balarajan et al., Reference Balarajan, Soni Raleigh, Yuen, Wheeler, Machin and Cartwright1991; Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993; Kay et al., Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994), double (Fleisher et al., Reference Fleisher, Fleming, Solo-Gabriele, Kish, Sinigalliano, Plano, Elmir, Wang, Withum, Shibata, Gidley, Abdelzaher, He, Ortega, Zhu, Wright, Hollenbeck and Backer2010; Arnold et al., Reference Arnold, Schiff, Griffith, Gruber, Yau, Wright, Wade, Burns, Hayes, McGee, Gold, Cao, Weisberg and Colford2013), and up to 3.6 times greater (Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012).

The association between levels of FIB and illness among swimmers has been studied. Kay et al. (Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994) demonstrated an association between any illness and faecal streptococci counts measured at chest level, suggesting a threshold level of 32 CFUs per 100 mL below which there is no increased risk (Kay et al., Reference Kay, Fleisher, Salmon, Jones, Wyer, Godfree, Zelenauch-Jacquotte and Shore1994). In a systematic review and meta-analysis, Wade et al. (Reference Wade, Pai, Eisenberg and Colford2003) estimated that for a log₁₀ unit increase in enterococci, the risk of gastrointestinal illness was increased by 1.34 times; additionally noting an association between enterovirus and illness in a small number of studies (Wade et al., Reference Wade, Pai, Eisenberg and Colford2003). The presence of any illness following marine bathing has been shown in a single study to be more closely associated with mean coliform counts than faecal streptococci levels (Corbett et al., Reference Corbett, Rubin, Curry and Kleinbaum1993). Furthermore, levels of enterococci in sand are positively associated with gastrointestinal illness among beachgoers digging or buried in sand (Heaney et al., Reference Heaney, Sams, Dufour, Brenner, Haugland, Chern, Wing, Marshall, Love, Serre, Noble and Wade2012).

Despite evidence of association between FIB levels and illness, this is not a consistent finding. Colford et al. (Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007) demonstrated higher rates of gastrointestinal and skin symptoms among a cohort of swimmers compared with non-swimmers, but no association between enterococci or faecal coliform levels and reported illness. Likewise, the BEACHES study found elevated incidence of gastrointestinal, acute febrile and skin illness in bathers compared with those non-bathing bathers under sub-tropical conditions, the only association with FIB levels was between enterococci levels and skin illness (Fleisher et al., Reference Fleisher, Fleming, Solo-Gabriele, Kish, Sinigalliano, Plano, Elmir, Wang, Withum, Shibata, Gidley, Abdelzaher, He, Ortega, Zhu, Wright, Hollenbeck and Backer2010). Both studies were carried out at beaches with a non-point source run-off, suggesting that where high exposure to an output of human waste is not the predominant source of pathogens traditional FIB may not be predictive of human health impacts (Colford et al., Reference Colford, Wade, Schiff, Wright, Griffith, Sandhu, Burns, Sobsey, Lovelace and Weisberg2007). Microbiological testing conducted in areas of California urban run-off have demonstrated a lack of association between adenovirus, a potential viral pathogen, and FIB levels (Jiang et al., Reference Jiang, Noble and Chu2001).

Under the European Union revised Bathing Water Directive (2006/7/EC), coastal water is classified as ‘excellent’ and awarded ‘blue flag status’ if E. coli are predominately below 250 per 100 mL and intestinal enterococci below 100 per 100 mL. Among bathers at sites with demonstrable excellent water quality by EU standards, there was still evidence of increased gastrointestinal, respiratory, eye and ear infections compared with non-bathers; with 3% of bathers having gastroenteritis compared with 0.7% of non-bathers in the 10 days following exposure (Papastergiou et al., Reference Papastergiou, Mouchtouri, Pinaka, Katsiaflaka, Rachiotis and Hadjichristodoulou2012). Furthermore, it is estimated that at the threshold for excellent bathing water quality for enterococci by European standards, it would be expected that 2.5–4.0% of bathers would develop gastrointestinal illness (Saliba & Helmer, Reference Saliba and Helmer1990). Estimates suggest that 71% of gastrointestinal episodes in Southern California occur when the water quality is considered safe for bathing (Brinks et al., Reference Brinks, Dwight, Osgood, Sharavanakumar, Turbow, El-Gohary, Caplan and Semenza2008), indicating that a ‘prevention paradox’ may exist – most cases of illness associated with marine bathing arise from the lowest risk exposure, but higher numbers of people exposed; this observation has implications for preventative strategies aimed at reducing the burden of disease.

The majority of epidemiological studies investigating the associations between marine bathing and illness have relied on self-reporting to define cases; however, there are clearly limitations to this approach. Bathers may be more likely to report symptoms if they believe they have had an exposure, such as head immersion, and in retrospective studies ‘recall bias’ may lead to increased reporting of events – such as swallowing water – in persons who have become unwell. A limited number of studies have attempted to obtain microbiological confirmation of the cause of illness; an association between exposure to seawater and confirmed staphylococcal skin infections has been shown (Charoenca & Fujioka, Reference Charoenca and Fujioka1995). Following heavy rainfall in Copenhagen, 42% of competitors in a sea-swimming event reported gastrointestinal illness; competitors who submitted stool samples had confirmed illness as a result of various pathogens including Campylobacter jejuni, Giardia lamblia and diarrheagenic E. coli. However, microbiological testing of the vast majority of stool samples submitted by competitors did not reveal any pathogens, suggesting a viral cause for illness in these competitors as very few specimens were tested for viruses (Harder-Lauridsen et al., Reference Harder-Lauridsen, Kuhn, Erichsen, Mølbak and Ethelberg2013).

Simulation models incorporating bathing water quality have estimated a considerable burden of disease related to bathing in coastal waters in Southern California, with 689,000 to over 4 million episodes of gastrointestinal disease and 693,000 episodes of respiratory illness among the 100 million visitors per year (Brinks et al., Reference Brinks, Dwight, Osgood, Sharavanakumar, Turbow, El-Gohary, Caplan and Semenza2008). Marine-bathing associated earache in the United States has the potential to result in 260,000 visits to the doctor, and direct health costs of nearly US$60 million (Wade et al., Reference Wade, Sams, Beach, Collier and Dufour2013).

EMERGING INFECTIOUS DISEASES AND THE MARINE INTERFACE

The term ‘emerging infectious diseases’ is used to describe: the emergence of novel or evolved pathogens, the spread of infections to new populations or geographic areas, or an increased threat resulting from antimicrobial resistance to previously treatable conditions (United States Centers for Disease Control & Prevention, 2014). Notable recent emerging human infectious diseases include avian influenza, Severe Acute Respiratory Syndrome (SARS) and Middle East respiratory syndrome coronavirus (MERS Co-V); all of which are considered to have emerged from hosts in the natural environment (Heymann, Reference Heymann2008). Examples of zoonotic (transferred between animals and humans) marine-related infections exist; a classical example being vibrio cholerae existing as a commensal organism in zooplankton, primarily copepods, conceivably leading to epidemics of cholera among humans (Huq et al., Reference Huq, Colwell, Rahman, Ali, Chowdhury, Parveen, Sack and Russek-Cohen1990).

Shewanella are bacteria that are found in marine environments worldwide (Holt et al., Reference Holt, Gahrn-Hansen and Bruun2005); over 50 species of this genus have been identified but recent developments in speciation have led to the identification of Shewanella putrefaciens and Shewanella algae as potential emerging human pathogens (Goyal et al., Reference Goyal, Kaur and Thakur2011; Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013; Janda & Abbott, Reference Janda and Abbott2014). Shewanella infection, of which 80% are caused by S. algae, can lead to skin and soft tissue infection, septicaemia, ear infections and death (Goyal et al., Reference Goyal, Kaur and Thakur2011; Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013; Janda & Abbott, Reference Janda and Abbott2014; Srinivas et al., Reference Srinivas, Pillai, Vinod and Dinesh2015). Case series of Shewanella infection, with case-fatality rates of 20–25%, have been reported in warmer climates, including Martinique and India (Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013; Srinivas et al., Reference Srinivas, Pillai, Vinod and Dinesh2015); although a systematic review by Vignier et al. (Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013) identified cases from northern Europe, including France, Denmark and the UK (Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013). The majority (Vignier et al., Reference Vignier, Barreau, Olive, Baubion, Theodose, Hochedez and Cabie2013) or all (Srinivas et al., Reference Srinivas, Pillai, Vinod and Dinesh2015) of Shewanella cases in reported series had skin breaks or penetrating injury providing a portal of entry; a risk factor often identified along with chronic health conditions with potential immunosuppression such as diabetes mellitus in people infected with marine indigenous bacteria (Gomez et al., Reference Gomez, Fajardo, Patino and Arias2003; Frank et al., Reference Frank, Littman, Alpers and Hallauer2006; Aigbivbalu & Maraqa, Reference Aigbivbalu and Maraqa2009; Cheung et al., Reference Cheung, Fung, Wong and Ip2010). Multidrug resistance has been demonstrated in some isolates of Shewanella (Holt et al., Reference Holt, Gahrn-Hansen and Bruun2005). During a notable warm summer in 1994 in Denmark, S. algae was isolated from ear swabs from 67 patients mostly aged between 3–15 years; most of the patients had symptoms of acute or chronic otitis (52%) or a non-specific aural discharge; S. algae was isolated from bathing sites in the same area where cases were identified (Holt et al., Reference Holt, Sogaard and Gahrn-Hansen1997).

Vibrios are bacteria found in estuarine and salt water environments; species that cause illness in man include V. cholerae – causing gastrointestinal illness (including cholera with toxin-producing forms); V. parahaemolyticus – gastroenteritis; V. vulnificus – skin infections, septicaemia and death; and V. alginolyticus – wound, ear and eye infections (Spicer, Reference Spicer2008). In 2006, a notably warm summer in Europe when sea temperatures across much of the Baltic Sea were >20°C, cases of V. vulnificus and non-toxin-producing V. cholerae were reported in Germany and Sweden, respectively; cases of vibrio infection were also reported in Denmark in the same year including a fatality associated with V. vulnificus – all cases identified in the three countries had bathed in the Baltic Sea (Frank et al., Reference Frank, Littman, Alpers and Hallauer2006). Vibrio spp. were identified widely in the Baltic Sea in July and August 2006 (Frank et al., Reference Frank, Littman, Alpers and Hallauer2006) and have since also been found to be common in Dutch waters (Schets et al., Reference Schets, van den Berg, Marchese, Garbom and de Roda Husman2011); cases of V. alginolyticus were detected in people bathing in Dutch Waters of the North Sea in 2006 (Schets et al., Reference Schets, Van den Berg, Demeulmeester, Van Dijk, Rutjes, van Hooijdonk and de Roda Husman2006). The pandemic strain of V. parahaemolyticus has recently been identified in UK waters for the first time (Powell et al., Reference Powell, Baker-Austin, Wagley, Bayley and Hartnell2013); it has been suggested that, as patterns of infection with Vibrio spp. are associated with climate trends, increasing sea temperatures have the potential to increase the range and burden of related disease. Furthermore, following Hurricane Katrina, an extreme weather event in 2005, there were five reported deaths from vibrio-related wound infections in persons from affected states (United States Centers for Disease Control & Prevention, 2005). Robust surveillance systems will be required to detect emerging trends (Baker-Austin et al., Reference Baker-Austin, Stockley, Rangdale and Martinez-Urtaza2010).

Antimicrobial resistance (AMR) is a major emerging global public health threat that challenges our ability to treat common infectious diseases; resistant strains can emerge and disseminate worldwide resulting in a highly significant disease burden (World Health Organization, 2015c). Escherichia coli resistant to antibiotics, notably third-generation cephalosporins, have been demonstrated in coastal waters across England and Wales, posing a potential risk to recreational users through ingestion; although the clinical significance of this exposure is uncertain (Leonard et al., Reference Leonard, Zhang, Balfour, Garside and Gaze2015). The ‘Beach Bums’ study, conducted by the ECEHH (http://www.ecehh.org), will use rectal swabs from regular surfers to investigate gut colonization by AMR bacteria in this group highly exposed to seawater.