Abstract
Recreational water quality guidelines protect the public from health risks associated with water recreation by helping to prevent unacceptable concentrations of pathogenic organisms in ambient water. However, illness risk is associated with both the concentration of pathogens in the water and the degree of contact with those pathogens. Different recreational activities can result in different levels of contact with ambient water containing water-borne pathogens.
We conducted a systematic literature review and meta-analysis to evaluate risks of illness associated with different recreational activities and different levels of contact to ambient surface waters. We screened 8,618 potentially relevant studies for quantitative measures of risk using inclusion/exclusion criteria established in advance. We categorized recreational activities as swimming, sports-related contact, minimal contact, and sand contact. We combined relative risks using a random effects meta-analysis for adverse health outcome categories representing gastrointestinal illness, respiratory illness, skin, eye, ear, nose, throat, and cold/flu illness.
We identified 92 studies meeting our inclusion criteria. Pooled risk estimates indicate significant elevation of gastrointestinal illness with the recreational activity categories swimming (2.19, 95% CI: 1.82, 2.63) and sports-related contact (2.69, 95% CI: 1.04, 6.92), and nonsignificant elevation of gastrointestinal illness with minimal contact (1.27, 95% CI: 0.74, 2.16). We also found a significant elevation of respiratory illness with swimming (1.78, 95% CI: 1.38, 2.29) and sports-related contact (1.49, 95% CI: 1.00, 2.24), and no elevation of respiratory illness with minimal contact (0.90, 95% CI: 0.71, 1.14).
This study suggests that exposures associated with different types of recreational activities are important characteristics of the exposure pathway when assessing illness risk associated with recreation in ambient surface waters.
Keywords: primary, secondary, contact, illness, guidelines, standards
Graphical Abstract
1. Introduction
Fecal contamination of ambient surface waters is a public health concern (USEPA 2012, WHO 2003). Epidemiological studies have reported an increased risk of illness associated with exposure to fecally-contaminated recreational water (Arnold et al. 2013, Arnold et al. 2016, Colford et al. 2007, Fleisher et al. 2010, Lamparelli et al. 2015, Wade et al. 2008, Wade et al. 2006, Wade et al. 2003, Wade et al. 2010, Yau et al. 2014). Swimming, paddling, boating and fishing account for more than 90 million cases of gastrointestinal, respiratory, ear, eye and skin-related illnesses per year in the United States, with an estimated annual cost of $2.9 billion (DeFlorio-Barker et al. 2018). To protect the public from illnesses associated with recreational water exposures, national water quality guidelines have been developed by assessing relationships between the concentration of indicators of fecal contamination (serving as an indirect measurement of fecal pathogens) and rates of illness. A level of indicator corresponding to an illness rate is then chosen (Australian Government National Health and Medical Research Council 2008, Brazil National Standards 2005, Health Canada 2012, New Zealand Ministry for the Environment 2002, Republic of South Africa Department of Environmental Affairs 2012, USEPA 2012, WHO 2003).
Recreational water quality guidelines provide the allowable concentration of specific pathogen indicators in surface waters used for different recreational activities by the public. However, the degree of illness risk is related to both the concentration of pathogens in the water and the degree of contact with those pathogens (see USEPA (2014) for a review of dose-response relationships for pathogens). Thus, individuals engaging in activities with a greater degree of contact to ambient recreational water potentially have a higher risk of illness compared to individuals engaging in activities with a lower degree of contact (Wade et al. 2008, Wade et al. 2006, Wade et al. 2003, Wade et al. 2010).
Most water quality guidelines were established to protect the public during recreational activities described as swimming or bathing with a high degree of water contact. Some water quality guidelines have also been established in the United States (Table 1) and internationally (Brazil National Standards 2005, Health Canada 2012) to protect activities presumed to be associated with a lower degree of water contact. However, there is a lack of consistency in the way recreational activities and presumed levels of water contact are defined or grouped (Table 2). For example, some guidelines refer to recreational activities with a high degree of water contact as primary contact, whole-body contact, total body contact, immersion contact, or simply contact recreation. Other guidelines refer to recreational activities with a low degree of water contact as secondary contact, limited water contact, limited body contact, partial body contact, incidental contact, or limited contact recreation. Some guidelines consider a third category of recreational activities with minimal potential contact with water they define as no contact, not contact, or non-contact recreation.
Table 1.
States in the United States with specialized water quality guidelines for recreational activities presumed to be associated with a lesser degree water contact than swimming or bathing.
Table 2.
Definitions of recreational activities and presumed levels of water contact.
Different recreational activity categories may be associated with different water quality values. Some water quality guidelines assign less stringent water quality values for activity categories with lower potential for water contact on the basis of the presumption that less contact with pathogens in water compensates for a higher concentration of those pathogens in water resulting in the same overall risk of illness. Recreational activities with less water contact may still carry risk due to inadvertent contact with water, poor hygiene following water contact, and hand-to-mouth or hand-to-eye contact (Dorevitch et al. 2015). However, there is limited evidence about the magnitude of health risks associated with recreational activities commonly identified as resulting in less water contact compared to recreational activities commonly identified as resulting in greater water contact (Dorevitch et al. 2012b, Fewtrell et al. 1992, Lee et al. 1997).
Previous systematic reviews and meta-analyses focused on the risk of illness associated with swimming (Leonard et al. 2018, Mannocci et al. 2016, Wade et al. 2003, Yau et al. 2009). However, these studies did not address differences in the risk of illness between different types of recreational activities. The purpose of this systematic review and meta-analysis is to evaluate the current scientific evidence for differences in risk of illness between recreational activities typically associated with different levels of contact with water. The objectives of this study are to: (1) assess and summarize the scientific literature on the risk of illness associated with different types of recreational activities and different levels of water contact during recreation; (2) quantitatively estimate the pooled risk of illness associated with different categories of recreational activities and levels of water contact; and (3) evaluate risk of illness across activity and water contact categories to better understand illness risk associated with different types of recreation in ambient surface waters.
2. Material and methods
2.1. Protocol
The protocol for this systematic review and meta-analysis was developed with input from a panel of external peer-reviewers (the protocol is available upon request from the corresponding author) and conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Liberati et al. 2009). The protocol specified the literature search strategy for identification of relevant studies and management of retrieved studies, reference databases searched, search terms, procedures for hand searches, procedures for assessing grey literature, information requests, inclusion criteria, methodology for risk of bias assessment, and general methods of synthesis, analysis, and interpretation of results. The hazard of concern in our study was exposure to ambient surface waters during recreation. Although sand activities typically do not involve direct contact with water, we included sand contact as a potential exposure route because sand contact often occurs in conjunction with water recreation, especially among children (Arnold et al. 2016, DeFlorio-Barker et al. 2018, Heaney et al. 2012, Heaney et al. 2009, Solo-Gabriele et al. 2016).
2.2. Literature search
We searched PubMed, Web of Science, and TOXLINE for peer-reviewed articles in any language. We limited articles to those published during or after 1950 to match Wade et al. (2003). Studies of interest were those related to exposure to ambient surface water during recreational activities and any health outcomes of interest (see Supplementary Material Table S1 for all search terms and the Boolean operations we applied to them). After the first phase of study selection identifying potentially relevant studies (described below), we reviewed the content and reference lists of potentially relevant studies for additional relevant literature. We also manually searched for articles by examining the reference lists of review articles addressing health effects associated with recreational activities; however, we did not use results presented in any review articles unless they were derived from data that were not previously published. We stored all citations (with available abstracts) in an electronic database.
2.3. Study selection
Our study selection procedure had two phases. During the first phase, we identified potentially relevant studies returned from our literature searches for studies related to the occurrence of any illness following exposure to ambient surface waters during any type of recreational activity. We selected potentially relevant studies based on their titles and abstracts, and then confirmed those selections by obtaining and reviewing their full text. We included grey literature only if publicly available and if they contained enough supporting methodological information that could allow a peer review as if they were manuscripts being considered for publication by a scientific journal. During the second phase of study selection, we selected those studies identified in the first phase of study selection that met all the following criteria:
2.3.1. Study participants
Study participants were human subjects engaging in activities involving exposure to ambient surface waters.
2.3.2. Health outcomes
At least one health outcome was evaluated. Health outcomes included, but were not limited to, vomiting, diarrhea, stomachache, nausea, fever, loose stool, sore throat, cough, runny nose, rash, itchy skin, eye ailments, earache (i.e., ear pain, ear infection, or runny ears), symptoms of infection, and non-symptomatic health outcomes such as antibody responses to infection with waterborne pathogens.
2.3.3. Water exposures
Related to ambient surface water exposure. We included studies involving exposure to marine (i.e., ocean) or fresh waters (i.e., lakes, rivers, ponds), and excluded studies involving exposure to chlorinated waters (e.g., swimming pools, hot tubs, and splash parks).
2.3.4. Study designs
Study designs were observational epidemiologic studies (e.g., cohort, case-control, cross-sectional, ecological, survey, and time-series designs) or randomized trials.
2.3.5. Data availability
Studies reported crude or adjusted effect estimates derived from data not previously published or included enough unpublished data to allow calculation of crude relative risk. If a study reported both crude and adjusted effect estimates, we used only the adjusted estimates.
2.4. Data extraction
We extracted study design, methodology, and results from studies meeting our inclusion criteria. The data we extracted included water type (marine or fresh), water quality (analytical methods, quantitative values, and qualitative descriptions), time over which water quality measurements represented exposure (i.e., same day, at the time of swimming or over the entire study period), study population, study size, study design, geographic location, health outcome(s) measured, covariate(s) measured, water exposure/activity description(s), reported measures of risk, and data we could use to calculate crude relative risk. Data was extracted by two team members, one to identify and enter the data into a database, and the other to verify accuracy. A third team member supervised data extraction and resolved disagreements.
2.5. Study quality
We evaluated study quality using a checklist of 13 questions that assess risk of bias in 7 domains (Supplementary Material Table S2). We formulated the list of questions from the National Toxicology Program Office of Health Assessment and Translation Risk of Bias Tool (OHAT 2015), Agency for Healthcare Research and Quality (Viswanathan et al. 2012), Cochrane Handbook (Higgins et al. 2011) and the CLARITY Group at McMaster University (2013). Two team members independently rated each study on each of the 13 questions using a four-point scale (definitely low risk, probably low risk, probably high risk, and definitely high risk). We compiled and compared all ratings from the two team members, and any disagreements between the two team members were resolved by a third senior team member. We developed a classification system to determine whether each study has either a low risk of bias or high risk of bias by focusing on 6 of the 13 risk of bias questions that reflect exposure or outcome misclassification, biased methods, and lack of adequate adjustment for potential confounders.
They are:
Is the comparison (i.e., control or reference) group appropriate?
Does the study design or analysis account for important confounding and modifying variables?
Do researchers adjust or control for other exposures that are anticipated to bias results?
Are we confident in the exposure characterization?
Are we confident in the outcome assessment?
Are there any other potential threats to internal validity (e.g., inappropriate statistical methods)?
We excluded from the meta-analysis any study rated as definitely high risk of bias for any of the above 6 questions. We included studies in the meta-analysis but classified them as having a high risk of bias if they fulfilled any of the following criteria: a) A rating of probably high risk of bias for all questions 1, 2, 3, 4, and 5; b) A rating of probably high risk of bias for any 2 of questions 1, 2, 3 with a rating of probably high risk of bias for question 6; c) A rating of probably high risk of bias for 3, 5, and 6; d) A rating of probably high risk of bias for question 3 with a rating of probably high risk of bias for at least one of questions 1, 2, 3, and 6; e) A rating of probably high risk of bias for question 4 with a rating of probably high risk of bias for any 2 of questions 1, 2, 3, 5, or 6. We classified studies as having a low risk of bias if they did not fulfill any of the above high risk of bias criteria.
2.6. Meta-analysis
Descriptions of recreational activities, study participants, health outcomes, water quality, age groups, and types of water contact substantially varied among the studies meeting our inclusion criteria. To perform meta-analyses, we categorized each extracted effect estimate based on recreational activity, degree of water contact, health outcome, water quality, study subject age, and water type. Table 3 shows the study-specific descriptions and how we categorized them.
Table 3.
Categories of recreational activity, water exposure, health outcome, water quality, study subject age, and water type used for meta-analysis.
| Category | Study Description |
|---|---|
|
| |
| Recreational Activities | |
| Swimming | Playing in the water; Playing in shallow water or swimming; Swimming; Swimming in a triathlon; Swimming or wading; Swimming, snorkelinga, or diving; Swimming, wading, or boogie-boardingb; Swimming, wading, or playing in the water; Playing/bathing; Swimming, wading, or bathing; Swim in or go into any natural water; Wading; Water play or swimming; Water contact; Swimming or bathing. |
| Sports-related | Snorkeling; Surfing; Surfing or diving; Surfing or windsurfing; Rafting; Washing; Whitewater canoeing or whitewater kayaking; Whitewater canoeing or whitewater rafting; Windsurfing; Body surfingc. |
| Minimal | Canoeing; Canoeing or kayaking; Canoeing or rowing; Canoeing, kayaking, boating, fishing, or rowing; Canoeing, kayaking, or standup paddle-boarding; Fetching water; Fishing; Kayaking or rafting; Rowing; Playing, bathing, washing blankets, laundry, fishing, collecting water, or crossing. |
| Sand | Body buried in the sand; Digging in sand; Sitting or lying down in sand; Sand exposure; Playing on wet sand. |
| Mixed | Canoeing, sailing, windsurfing, scuba diving, fishing, caving, or laying in the water; Swimming, paddling, or playing by the water; Swimming, canoeing, kayaking, or standup paddle-boarding; Swimming, canoeing, sailboarding, or water skiing; Contact with floodwater; Swimming and/or fishing; Going to rivers/lakes. |
| Exposures | |
| Head immersion | Head below the water line; Completely submerged their head and body beneath the surface of the water; Head immersion; Swimming; Submerging in flood water while river rafting; Immersion of the face and head in water; Individuals who completely immersed their body and head in the water; Submerged head; All upper body orifices exposed to the water; Natation; At least 10 minutes in the water and to immerse their heads three times; Spend 15 min in knee-deep water and to immerse their head three times completely under water; Immersed their heads in the ocean water; Entered a defined and roped-off 20 m area of water for at least 10 min, completely immersing their heads at least three times: Accidental swimming in slalom course; Immersing the head; Participants in a swim-bike-run triathlon; Swam for at least 10 minutes with head immersion; Snorkel swimming; Swimmers (head in); Usually submerges head during recreational swimming; Swimmers in Olympic distance triathlons; 10 minutes bathing with at least three head immersions; Single head immersion; Full body contact with water including head immersion; exposure to upper body orifices were significantly in contact with water and head immersion; Significant exposure of upper body orifices (i.e., the head) to beach water, or young children who spent time in the water or at the water’s edge and whose faces were splashed by wave action or other means. |
| Body immersion | Water contact above the waist; Waders; Washing body; Bathingd: Body immersion for at least 10 min; Swimmers (no head immersion); Baignade; Waded in a river or stream; Swimmers (head out); Contact with floodwater; Swimming (body immersion); Waders (those who swam but did not immerse their head); Playing in shallow water or swimming. |
| Face splashing | Water on face. |
| Sand contact | Sitting, playing, or lying in the sand (exposed to dry sand only or exposed to wet sand only); Sand exposure (collecting shells, making sand castles, burying body in sand, and getting sand into mouth); Digging in the sand; Body buried in the sand; Sits or lies down directly on the sand; Playing on wet sand. |
| Unspecified direct water exposure | Swimming in rivers; Swimmers, waders, surfers, or divers; Swimming, snorkeling, or diving; People entering the water without significant exposure to either wet or dry sand; Wading; Carried by parents; Boogie boarding; History of seawater exposure; Any water contact; Any recreational water activity; Play in natural source of water, either fresh or saltwater; Windsurfing; Surfing; A person whose head had been immersed in the water, or who had swallowed seawater, or whose face had been splashed by waves; White-water canoeing; Surfers; Swims in rivers; Recreational water sports; Swimming; Ocean swim; Whitewater canoeists and rafters; Wade, swim or play in the water; Seabathing; Swam, waded, or bathed in a river or stream; Swimming in water from natural sources; All swimmers; Recreational, swam in pond; Recreational freshwater swimming; Recreational fresh water contact; Swimming in pond; Swimming in recreational fresh water lakes; Individuals who entered the water to waist height or beyond; Any flood exposure; Floodwater touching any part of body; Any contact; Those who reported immersing their body to their waist or higher; Immersion to the waist or higher; Swimmers with body immersion (waist or higher); Swimmers with head immersion; Bathers; Surfing or swimming; Playing/bathing. |
| Unspecified indirect water exposure | Canoeing, kayaking, boating, fishing, or rowing; Canoeing; Kayaking or rafting; Rowing; Fishing; Canoeing or rowing; Canoeing, kayaking, or standup paddleboarding; Canoeing or kayaking; Paddling, rowing, fishing, or boating. |
| Health outcomes | |
| HCGI | HCGI; HCGI-2; Objective acute gastrointestinal diseases with vomiting and/or diarrhea. |
| Diarrhea | Diarrhea; Diarrhea within 3 days; Diarrhea within ten days; Diarrhea and at least two of: vomiting, cramps, fever, nausea, dizziness, or headache; Dysentery; Diarrhea more than 2 times; Diarrhea/vomiting. |
| General RI | Respiratory (general).; Respiratory infection; Respiratory tract infection (sore throat, heavy cough, and running nose); Respiratory tract infection (having a cold, coughing or dyspnoea); Overall respiratory symptoms. |
| SRD | Acute febrile respiratory illness; Highly credible respiratory illness; Significant respiratory disease. |
| Cough | Cough. |
| Skin | Buruli ulcer; Skin rash; Rash; Skin infection; Skin/mucosa; Staphylococcus; Infected cuts; Cutaneous infection; Buruli ulcer disease; Infection of open wound; Itchy skin, skin rash; Dermatological complaints. |
| Cold/flu | Cold; Common cold; Rheum (Runny nose); Febrile illness; Fever; Fever/headache; Flu; Headache; Lack of energy; Loss of appetite; Avian influenza; Influenza-like illness. |
| ENT | Ear (general); Ear discharge; Ear infection; Earache; Ears, nose, and throat; Throat; Ear/eye symptoms; Otitis; Dysphagia (odynophagia); Otitis externa; Sinus pain or infection; Earache or infection; Affections ORL; Sore throat/coughs; Ear infection/discharges. |
| Eye | Eye (general); Eye infection; Eye irritation; Eye pain or burn; Tear secretion (eyes discharge); Eye redness. |
| Any | Any; Any illness; At least one symptom, all symptoms; Any infectious symptom; Total (any health complaint); Illness; Total illness. |
| Other | Allergy: Leptospirosis; Medical consultation; Medication receiving; Other; Urinary tract infection; Urinary Schistosomiasis. |
| Water quality | |
| Good | La Nina winter/low runoff; Enterococcus < 33 CFU/100 mL; Less than median; Berm closed/good water quality; Better water quality; Clean beach; First quartile; First quintile; First sextile; First tertile; Good; Least polluted beaches; Low pathogen concentration; Lower pollution; Clostridium spp. < median; Fecal index tertile 1; fecal strep 0–13CFU/100 mL; Santa Ana Beach; Dry weather. |
| Poor | Days with high groundwater flow; Days with high groundwater flow (≥ 163 m3); El Nino winter/high runoff; > 35 CFU/100 ml Enterococcus; Enterococcus ≥ 33 CFU/100 ml; >=Median; Berm open/poor water quality; Fifth quintile; Fourth quartile; High pathogen concentration; Higher pollution; Moderately polluted beach; Most polluted beaches; Poor; Sixth sextile; Third tertile; Wet weather; Moderate/High/Very high pollution exposure; Red tide. |
| Intermediate | Fecal streptococci: 40–79/ CFU100 mL; E. coli: 35–320 CFU/100 mL; second tertile (Enterococcus); Clostridium spp. tertile 2; Second and third quartiles (Enterococcus); Fecal coliforms 178–5623 CFU/100 mL. |
| Any | All groundwater flow days; Any; Berm open and closed; Combined; Not specified. |
| Age | |
| Children | 4 months - 16 years; < 10 years; < 19 years; 0–10 years; 0–12 years; 0–15 years; 0–4 years; 0–5 years; 11–15 years; 16–20 years; 1st and 2nd graders; 2–9.5 years; 5–12 years; 6–10 years; 6–11 years; Children; 5–17 years; Children less than 10; School children within the age of 7 to 14 years; Primary school students. |
| Adults | >55 yrs; ≥ 10 yrs; 16–55 years; 18+ yrs; 19–60 yrs; 19–65 yrs; 20–64 yrs; 21–40 yrs; 21–49 yrs; 30+ yrs; 35–69 yrs; 41–70 yrs; 55+ yrs; Adults. |
| Any | 0–67 years; 0–70 years; 0–74 years; 0–87 years; 11+ years; ≥ 10 years; 11–40 years; 11–54 years; 12–30 years; 12–65 years; 13–45 years; 15–50+ years; 15–54 years; 1–88 years; 2–53 years; 4–89 years; 5+ years; 5–54+ years; 5–64 years; 7–51 years; 8–68 years; All ages; Children and adults; Families with at least one child less than 10 years; Not specified. |
Abbreviations: ENT, ears, nose, and throat; CFU, colony forming units; GI, gastrointestinal illness; HCGI, highly credible gastrointestinal illness; ORL, oral afflictions; RI, respiratory illness; SRD, significant respiratory disease.
Categorized as Swimming because information about the study locations suggest a slim minority of study subjects were likely to only engage in snorkeling.
Categorized as Swimming because only 2 out of the 53 subjects in the study group used a boogie board.
Categorized as sports-related contact because of its similarity to surfing but with an even higher likelihood of water ingestion.
Ritual or routine washing of the body.
2.6.1. Recreational activities
We designed a system for categorizing recreational activities by first evaluating how currently available recreational water quality guidelines defined them. We compiled terms and phrases used to define recreational activities by international health organizations, national governments, and by individual states and territories in the United States. Most guidelines that distinguish between different categories of recreational activities define two categories on the basis of the probability of contact with ambient water. Thus, we specified in our protocol a strategy to categorize recreational activities into two categories and perform a meta-analysis to compare the risk of illness associated with those two categories.
When considering the descriptions of recreational activities extracted from the studies meeting our inclusion criteria, however, we found a wide range of recreational activities that we believe were not adequately represented by only two categories. Instead, we grouped recreational activities into 5 specific categories (Table 3). We categorized activities as “swimming” if they involved the immersion of any part of the body in water that did not involve the use of specialized sport equipment. We categorized activities as “sports-related contact” if they involved immersion or a high probability of immersion or splashing of water on any part of the body while using specialized sport equipment (e.g., surfing, water skiing, or canoeing, kayaking, or rafting in rough water). We categorized activities as “minimal contact” if they involved floating on or being near ambient water (except for sand contact) with a low probability of immersion or splashing of water on any part of the body (e.g., fishing, or canoeing, kayaking, or rafting in tranquil waters). We categorized activities as “sand contact” if the activity only involved contact with sand. Although we considered combining sand contact and minimal contact activities together, we decided to keep them as separate categories because the activities in each of those categories generally involve different types of beachgoers (children versus adults) and different media (sand versus water). Finally, we categorized activities as “mixed contact” if a study combined participants engaging in activities from more than one of the above activity categories. Because the primary goal of this systematic review and meta-analysis is to evaluate the risk of illness associated with different recreational activity categories representing different levels of water contact, we did not analyze effect estimates in the category mixed contact.
2.6.2. Levels of water contact
A subset of studies meeting our inclusion criteria provided estimates of illness risk associated with one or more levels of water contact independent of the recreational activity underlying the contact. To analyze water contact-related estimates of risk, we independently designated a water contact category to every effect estimate, including effect estimates associated with a specific recreational activity (Table 3). We categorized water contact as “body immersion,” “head immersion,” “face splashing,” or “swallowing water,” if a study specifically described those categories of water contact. In addition, we categorized water contact as “unspecified direct contact” if a study did not specify a category of water contact but did specify a recreational activity we categorized as either swimming or sports-related contact or described water contact consisting of more than one of the specific water contact categories body immersion, head immersion, face splashing, or swallowing water. We categorized water contact as “unspecified indirect contact” if a study did not specify a water contact category but did specify a recreational activity we categorized as minimal contact. For analytical purposes, we combined all data from specific water contact categories (body immersion, head immersion, face splashing, or swallowing water) together with data from the category unspecified direct contact into the general water contact category “all direct contact.” We also defined the general water contact category “all indirect contact” to compare with the general water contact category all direct contact even though the category all indirect contact represents the same set of data as the category unspecified indirect contact because no studies defined a specific water contact level we considered indirect water contact. Finally, we categorized exposure as “sand contact” if a study reported the effect estimate as related to contact with sand.
2.6.3. Health outcomes
We selected gastrointestinal illness (GI) and respiratory illness (RI) as the health outcomes of primary interest before performing our literature search. The studies meeting our inclusion criteria contained a total of 143 health outcome descriptors. We categorized the GI and RI health outcome descriptors as either “general GI”, highly credible gastrointestinal illness (“HCGI”), “diarrhea,” “general RI,” significant respiratory disease (“SRD”), or “cough.” Health outcomes of secondary interest were skin symptoms (“skin”), ear symptoms (“ear”), combined ear, nose, and throat symptoms (“ENT”), eye symptoms (“eye”), cold or flu symptoms (“cold/flu”), other symptoms (“other”), and unspecified symptoms (“any”). We used the category HCGI because it is a specific category of GI used by several studies to represent a combination of unmistakably recognized individual symptoms of gastrointestinal illness (Cabelli 1983). HCGI includes the presence of a specific set of symptoms such as vomiting, diarrhea with a fever or disabling condition, or stomachache or nausea accompanied by a fever. The category diarrhea refers to GI specifically described as diarrhea or a set of symptoms where diarrhea is a distinguishing characteristic. The category general GI refers to any one or collection of GI symptoms studies did not specifically categorize as HCGI or diarrhea. We used the category SRD because it is a category of RI used by several studies to clearly identify the occurrences of consequential RI events. The category SRD includes a specific set of clearly identifiable RI symptoms, although there were small variations in the precise definition of SRD among studies. The category cough refers to a health outcome that studies simply described as a cough. The category general RI includes any one or collection of RI symptoms studies did not specifically categorize as SRD or cough.
2.6.4. Water quality
Because previous studies have shown an association between water quality and rate of illness, water quality could be an effect modifier when evaluating risk of illness associated with different recreational activities. To evaluate the effects of water quality, we extracted water quality data and information from studies meeting our inclusion criteria. We categorized water quality as “good,” “intermediate,” or “poor” based on quantitative measures of fecal indicator bacteria reported in the studies. If a study reported multiple risk estimates each linked to a different concentration of fecal indicator bacteria, we categorized risk estimates linked to the lowest concentration of indicator bacteria as good water quality, the highest concentration of indicator as poor water quality, and densities between good and poor as intermediate water quality.
Many studies did not report quantitative measures of water quality but did provide narrative descriptions we used to categorize water quality as good or poor. For example, we categorized water quality as good if a study described water quality as low pathogen concentration, least polluted, or lower pollution, and we categorized water quality as poor if a study described water quality as high pathogen concentration, most polluted, or higher pollution (see Table 3). We also categorized water quality as good or poor based on physical or climatological conditions studies associated with good or poor water quality. For example, we categorized water quality as good if the study claimed good water quality due to the presence of a sand berm preventing contaminated water from reaching the study area, and we categorized water quality as poor if the study claimed poor water quality due to the absence of a sand berm preventing contaminated water from reaching the study area. We categorized water quality as “any” if the study did not provide water quality information or did not relate effect estimates to water quality.
2.6.5. Age
Because children may be at a higher risk of illness when recreating in contaminated waters compared to adults (Arnold et al. 2016, Heaney et al. 2012, Heaney et al. 2009, Wade et al. 2013), we evaluated age as an effect modifier on the risk of illness associated with different recreational activities. We generally attempted to categorize subjects below 18 to 20 years of age as “children” and subjects above 18 to 20 years of age as “adults” when studies provided numerical age information. We accepted study-specific categorizations of children or adults if the study did not provide numerical age information but nonetheless described subjects as children or adults. We categorized the age of the study subjects as “any” if the study did not report any age information or reported age information but did not relate effect estimates with specific age categories. We assumed all subjects from studies of athletes from competitive sporting events (e.g., triathlons) were adults.
2.7. Statistical analysis
We quantitatively evaluated the risk of illness associated with different categories of recreational activities and water contact by combining the results of multiple studies using meta-analysis (Egger et al. 2008, Littell et al. 2008). We considered odds ratios to be reasonable estimates of relative risk because the probability of infection due to water-based recreational activities is generally low. We calculated a crude relative risk and 95% confidence interval (CI) when a study did not report effect estimates but reported sufficient data to calculate one.
When analyses contained 5 or more studies, we used robust variance estimation meta-regression (Tanner-Smith and Tipton 2014, Tipton 2015) to allow the utilization of all effect estimates in a single study that fulfill the criteria of the category being evaluated. We assumed a within-study correlation of 0.8 (Tanner-Smith and Tipton 2014) because our sensitivity analysis using within-study correlations of 0.1, 0.3, 0.5, 0.7, 0.8, and 0.9 resulted in point estimates and confidence interval limits that varied by less than 0.001. When analyses contained fewer than 5 studies, we used the DerSimonian and Laird random effects approach (DerSimonian and Laird 1986) to first estimate study-specific pooled effects and then the overall pooled effect. We used Excel and Stata for data management and the metan, metareg and robumeta packages in Stata version 14.0 (StataCorp 2015) for all meta-analyses. We considered pooled risk estimates significant if their 95% confidence intervals did not overlap 1, and we considered pooled risk estimates significantly different from each other if their 95% confidence intervals did not overlap each other.
We evaluated heterogeneity by using forest plots and the Cochran Q test for heterogeneity in conjunction with the I2 index to quantify the degree of variation. When heterogeneity was high (I2 index > 80%), we conducted sensitivity analyses to assess the potential sources of heterogeneity, including study quality, study design, activity category, water contact category, and water type. We evaluated impacts of water quality on pooled risk estimates by performing meta-regression analyses that include water quality as a covariate and by calculating water quality-specific pooled estimates.
We analyzed publication bias by creating funnel plots of pooled risk estimates from each individual study versus the log standard error of those estimates and inspecting for asymmetry. Because robust variance estimation meta-regression does not report pooled effect estimates for each individual study, we used standard meta-regression to calculate the pooled risk for each study in the funnel plot. We assessed asymmetry in the funnel plot using Eggers’ test (Egger et al. 2008). We considered the results to indicate publication bias if p < 0.10 for the bias coefficient in Egger’s test. We conducted sensitivity trim and fill analyses (Duval and Tweedie 2000a, b) to estimate the number of missing studies in a meta-analysis affected by publication bias and the effect of those studies on the effect estimate. Before performing our analysis, we declared any meta-analysis result derived from fewer than three studies as inconclusive.
3. Results
3.1. Literature search summary
Figure 1 illustrates the study selection process. Our systematic literature search returned 8,618 articles. During the first phase of study selection we identified 968 studies as potentially relevant based on their titles and abstracts. We identified 629 of those studies as potentially relevant after evaluating their full text. During the second phase of study selection, we identified 112 of 629 studies as meeting our inclusion criteria. Of those 112 studies, we excluded three studies (Ferrer et al. 2012, Kundu et al. 2013, Rijal et al. 2011) because they were quantitative microbial risk assessments and did not provide evidence of quantitative associations between recreational activities and health outcomes. We excluded an additional 15 studies (Amin et al. 2017, Appleton and Bailey 1990, Cabelli et al. 1982, DeFlorio-Barker et al. 2018, Eze et al. 2014, Fewtrell and Delahunty 1985, Fleming et al. 2004, Kueh et al. 1995, Lepesteur et al. 2006, Papastergiou et al. 2011, Philipp et al. 1989, Seyfried et al. 1985b, Stevenson 1953, Turbow et al. 2008, Veleizan et al. 2016) because they did not report quantitative results or sufficient data to calculate crude relative risk. We also excluded one study (von Schirnding et al. 1992) because it contained pilot data published in a follow up paper (von Schirnding et al. 1993). We considered the results of two publications (Cabelli 1983, Cabelli et al. 1983) as one study because they each reported a subset of results from a single study. The remaining 92 studies are the subject of this systematic review and meta-analysis.
Figure 1.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart of study selection process.
3.2. Study characteristics
Figure 2 summarizes the key the characteristics of the 92 studies meeting our inclusion criteria (a complete list of the studies and their characteristics is provided in Table 4). Most (81/92) studies evaluated at least one of several symptoms related to GI that met our category definition of general GI. Because 11 studies independently evaluated the rates of illness associated with a more narrowly defined set of GI symptoms termed highly credible gastrointestinal illness (HCGI or HCGI-2), and 18 studies independently evaluated risk of diarrhea, we analyzed those health outcomes as both separate categories (HCGI and diarrhea), and together combined with general GI into the single category “all GI outcomes combined.” A minority (18/92) of studies evaluated at least one of several symptoms related to RI that met our category definition of general RI. Because 9 studies independently evaluated the rates of illness associated with the more narrowly defined set of RI symptoms SRD, and 4 studies independently evaluated risk of cough, we analyzed those health outcomes as both separate categories (SRD and cough), and together combined with general RI into the single category “all RI outcomes combined.”
Figure 2.
Summary of key characteristics of the 92 studies meeting our inclusion criteria. A. Number of studies reporting illness risk estimates for each health outcome category. B. Number of studies with experimental designs meeting our inclusion criteria. C. Number of studies reporting illness risk estimates for each recreational activity category. D. Number of studies reporting illness risk estimates for each category of water contact. E. Number of studies reporting illness risk estimates for each water quality category. F. Number of studies reporting effect estimates from each type of surface water. Studies that provide illness risk estimates from combined fresh and marine waters calculate illness risk estimates from subjects exposed to either fresh or marine waters. Estuarine waters are included in the category marine. Totals are greater than 92 because some studies reported illness risk estimates for more than one category. Abbreviations: GI, gastrointestinal illness; HCGI, highly credible gastrointestinal illness; RI, respiratory illness; SRD, significant respiratory disease; ENT, ears nose and throat; UTI, urinary tract infection.
Table 4.
Key Characteristics of Studies Meeting Inclusion Criteria
| Activities |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Study | Country | Study Design | Ages | Health outcomes | Water type | Water quality | Study descriptions | Analytical categories | Exposure categories | Risk of Bias |
|
| ||||||||||
| (Aiga et al. 2004) | Ghana | case control | any | skin | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Alexander et al. 1992) | United Kingdom | prospective cohort | children | GI; skin; cold/flu | marine; | any | swimming, paddling, or playing by the water | Mixed | Any contact; | low |
| (Arnold et al. 2013) | United States | prospective cohort | any | GI; diarrhea | marine; | any | swimming | Swimming | Body immersion; Head immersion; Swallowing water; | low |
| (Arnold et al. 2017) | United States | prospective cohort | adults | GI; diarrhea; ENT; skin; cold/flu; RI; any | marine; | good; poor | surfing | Sports-related | Unspecified direct; | low |
| (Atalabi et al. 2016) | Nigeria | cross-sectional | any | UTI | fresh; | any | water play or swimming | Swimming | Body immersion; | low |
| (Balarajan et al. 1991) | United Kingdom | prospective cohort | any | GI; diarrhea; RI; eye; ENT | marine; | any | swimming, wading, surfing, or diving; wading; swimming; surfing or diving | Mixed; Sports-related; Swimming | Body immersion; Unspecified direct; | high |
| (Bonilla et al. 2007) | United States | prospective cohort | any | GI | marine; | any | swimming, snorkeling, or diving; sand contact | Swimming; Sand | Unspecified direct; Sand; | high |
| (Brown et al. 2002) | China | case control | adults | GI | fresh; | any | washing | Sports-related; Swimming | Body immersion; | low |
| (Cabelli 1983, Cabelli et al. 1983) | United States and Egypt | prospective cohort | any | GI; HCGI; skin; eye; RI; diarrhea; cold/flu | marine; | any | swimming | Swimming | Head immersion; | low |
| (Calderon et al. 1991) | United States | prospective cohort | any | GI | fresh; | any | swimming | Swimming | Head immersion; | low |
| (Charoenca and Fujioka 1995) | United States | case control | children | skin | marine; | any; poor | swimming, wading, or boogie boarding | Swimming | Unspecified direct; | low |
| (Cheung et al. 1990) | Hong Kong | case control | adults; any; children | cold/flu; RI | marine; | any | swimming | Swimming | Head immersion; | low |
| (Chusri et al. 2012) | Thailand | prospective cohort | any | GI | fresh; | any | rafting | Sports-related | Head immersion; | low |
| (Colford et al. 2007) | United States | prospective cohort | children; adults; any | cold/flu; cough; diarrhea; ear; eye; HCGI; GI; SRD; skin; | marine; | any | swimming | Swimming | Unspecified direct; Swallowing water; Face splashing; | low |
| (Colford et al. 2012) | United States | prospective cohort | any | cold/flu; cough; diarrhea; ENT; eye; GI; HCGI; SRD; skin; UTI | marine; | any; good; poor | swimming | Swimming | Body immersion; Head immersion; Swallowing water; | low |
| (Corbett et al. 1993) | Australia | prospective cohort | any | any; cold/flu; cough, ear, eye; GI | marine; | good; poor | swimming | Swimming | Head immersion; | low |
| (Cordero et al. 2012) | United States | prospective cohort | any | GI | marine; | any | swimming; sand contact | Sand; Swimming | Sand; Head immersion; | low |
| (D’Alessio et al. 1981) | United States | case control | children | cold/flu | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Denno et al. 2009) | United States | case control | children | GI | fresh and marine; | any | playing in water | Swimming | Unspecified direct; | low |
| (Dewailly et al. 1986) | Canada | cross-sectional | adults | any; diarrhea, GI; skin | marine; | any | windsurfing | Sports-related | Unspecified direct; | low |
| (Dorevitch et al. 2012a) | United States | prospective cohort | any | GI | fresh; | any | canoeing, kayaking, boating, fishing, or rowing | Minimal | Unspecified indirect; Head immersion; Swallowing water; | low |
| (Dorevitch et al. 2012b) | United States | prospective cohort | any | GI; ear; SRD; skin; | fresh; | any | canoeing, kayaking, boating, fishing, or rowing; canoeing; kayaking or rafting; rowing; fishing | Minimal | Unspecified indirect; | low |
| (Dorevitch et al. 2015) | United States | prospective cohort | any | GI | fresh; | any | fishing | Minimal | Unspecified indirect; | low |
| (Dufour 1984) | United States | prospective cohort | any | GI; HCGI; other; RI | fresh; | any | swimming | Swimming | Head immersion; | low |
| (Dwight et al. 2004) | United States | cross-sectional | adults | any; diarrhea; eye; HCGI; SRD; skin | marine; | good; poor | surfing | Sports-related | Unspecified direct; | low |
| (Fattal et al. 1986) | Israel | prospective cohort | any; children | any; GI; RI; skin; | marine; | good; poor | swimming | Swimming | Unspecified direct; | low |
| (Feldman et al. 2002) | United States | retrospective cohort | children | GI | fresh; | any | swimming | Swimming | Swallowing water; | high |
| (Feleke et al. 2017) | Ethiopia | cross-sectional | children | GI | fresh; | any | swimming, fishing | Minimal; Swimming | Unspecified direct; Unspecified indirect; | low |
| (Ferley et al. 1989) | France | retrospective cohort | any | any; GI: HCGI; skin | fresh; | good, poor | swimming | Swimming | Unspecified direct; | low |
| (Fewtrell et al. 1992) | United Kingdom | prospective cohort | any | any; cold/flu; ENT; GI; RI; skin | fresh; | good; poor | whitewater canoeing or whitewater kayaking | Sports-related | Unspecified direct; | low |
| (Fewtrell et al. 1994) | United Kingdom | prospective cohort | adults | any; GI | fresh and marine; | any | canoeing or rowing | Minimal | Unspecified indirect; Swallowing water; | high |
| (Fleisher et al. 1993) | United Kingdom | controlled trial | adults | GI | marine; | good; intermediate; poor | swimming | Swimming | Head immersion; | low |
| (Fleisher et al. 1996) | United Kingdom | controlled trial | adults | eye; ear; SRD | marine; | any | swimming | Swimming | Head immersion; | low |
| (Fleisher et al. 2010) | United States | controlled trial | adults | GI; SRD; skin | marine; | any | swimming | Swimming | Head immersion; | low |
| (Galappaththi-Arachchige et al. 2016) | South Africa | cross-sectional | any | UTI | fresh; | any | playing, bathing, washing blankets, laundry, fishing, collecting water, crossing | Minimal | Unspecified indirect; | low |
| (Gammie and Wyn-Jones 1997) | United Kingdom | case control | any | other | marine; | any | surfing or windsurfing | Sports-related | Unspecified direct; | low |
| (Goodman et al. 1996) | United Kingdom | cross-sectional | children | GI | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Gray et al. 1994) | United Kingdom | case control | any | GI | fresh and marine; | any | swimming; canoeing, sailing, windsurfing, scuba diving, fishing, caving, or playing in the water | Mixed; Swimming | Unspecified direct; Any contact; | low |
| (Haile et al. 1999) | United States | prospective cohort | any | diarrhea; ear; HCGI; SRD; skin | marine; | good, intermediate, poor | swimming | Swimming | Head immersion; | low |
| (Hall et al. 2017) | United Kingdom | retrospective cohort | any | GI | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Hamner et al. 2006) | India | cross-sectional | any; children | any; diarrhea; GI; | fresh; | any | washing | Swimming | Body immersion; | low |
| (Harder-Lauridsen et al. 2013) | Denmark | retrospective cohort | any | diarrhea; GI | marine; | any | swimming | Swimming | Swallowing water; | low |
| (Harding et al. 2015) | United States | cross-sectional | any | diarrhea; ENT; GI; cold/flu | marine; | any | surfing, body surfing | Sports-related | Unspecified direct; | low |
| (Harrington et al. 1993) | Australia | prospective cohort | any | any; cold/flu; GI; RI; | marine; | any | surfing | Sports-related | Unspecified direct; | high |
| (Heaney et al. 2009) | United States | prospective cohort | any; children | ear; diarrhea; GI; SRD; skin | fresh and marine; | any | sand contact | Sand | Sand; | low |
| (Heaney et al. 2012) | United States | prospective cohort | any | diarrhea; GI; | marine; | good; intermediate; poor | sand contact | Sand | Sand; | low |
| (Jessop et al. 1985) | United Kingdom | case control | any | GI | fresh; | any | swimming, canoeing, sailboarding, or water skiing | Mixed | Any contact; | low |
| (Joosten et al. 2017) | Netherlands | prospective cohort | adults; any | GI; any; RI | fresh; | any | swimming | swimming | Head immersion; Swallowing water; | low |
| (Kay et al. 1994) | United Kingdom | controlled trial | adults | GI | marine; | good; intermediate; poor | swimming | Swimming | Head immersion; | low |
| (Lamparelli et al. 2015) | Brazil | prospective cohort | children; any | GI | marine; | any; poor | swimming; sand contact | Sand; Swimming | Sand; Swallowing water; | low |
| (Lee et al. 1997) | United Kingdom | prospective cohort | adults | GI | fresh; | any; intermediate; poor | whitewater canoeing or whitewater rafting | Sports-related | Unspecified direct; Head immersion; Swallowing water; | low |
| (Lepesteur et al. 2006) | Australia | prospective cohort | any | any | marine; | any | swimming, paddling, or playing by the water | Mixed | Any contact; | high |
| (Ly et al. 2016) | Cambodia | case control | children | cold/flu | fresh; | any | swimming or bathing | Swimming | Unspecified direct; | low |
| (H et al. 2016) | Netherlands | cross-sectional | adults | GI; cold/flu; skin | fresh; | any | contact with floodwater | Mixed | Any contact; | low |
| (Marion et al. 2010) | United States | prospective cohort | any | GI; HCGI | fresh; | any; intermediate; poor | swimming, wading, or playing in the water | Swimming | Unspecified direct; | low |
| (McBride et al. 1998) | New Zealand | prospective cohort | any | GI; HCGI; RI; | marine; | any; good, intermediate, poor | swimming; canoeing, kayaking, or standup paddle boarding | Mixed | Any contact; | low |
| (Medema et al. 1995) | Netherlands | prospective cohort | adults | any; cold/flu; GI; RI; skin | fresh; marine; | any | swimming | Swimming | Head immersion; | low |
| (Nardone et al. 2001) | France | case control | any | GI | fresh and marine; | any | canoeing or kayaking; fishing; swimming | Minimal; Swimming | Head immersion; Body immersion; Unspecified indirect; | low |
| (O’Halloran et al. 2017) | United States | prospective cohort | adults | RI | marine; | poor | surfing | Sports-related | Unspecified direct; | low |
| (Papastergiou et al. 2012) | Greece | prospective cohort | any | cough; diarrhea; ear; ENT; eye; GI; medical consultation; RI | marine; | any | swimming | Swimming | Body immersion; | low |
| (Parkkali et al. 2017) | Netherlands | retrospective cohort | any | GI | fresh; | any | swimming | Swimming | Swallowing water; Unspecified direct; | low |
| (Pham-Duc et al. 2014) | Vietnam | prospective cohort | any | diarrhea | fresh; | any | Washing clothes; fishing | Mixed | Any contact; | low |
| (Philipp et al. 1985) | United Kingdom | prospective cohort | any | GI | marine; | poor | snorkeling | Sports-related | Head immersion; | high |
| (Polkowska et al. 2018) | Finland | retrospective cohort | children; adults | GI | fresh; | any | swimming, head immersion, sand contact | Swimming; Sand | Sand; Head immersion; Unspecified direct; | low |
| (Prieto et al. 2001) | Spain | prospective cohort | any | any; GI; skin; | marine; | any; good; intermediate; poor | swimming | Swimming | Unspecified direct; Body immersion; Head immersion; | low |
| (Raghunathan et al. 2005) | Ghana | case control | any | skin | fresh; | any | fishing; swimming, wading, or bathing; wading | Minimal; Swimming | Unspecified direct; Unspecified indirect; Body immersion; | low |
| (Ravel et al. 2016) | Canada | case control | any | GI | fresh; | any | swimming | Swimming | Body immersion; | low |
| (Ross et al. 2017) | Philippines | cross-sectional | any | GI | fresh; | any | water contact | Swimming | Any contact; | low |
| (Sanchez-Nazario et al. 2014) | United States | prospective cohort | any | GI; RI; skin; cold/flu; ear | marine; | any | swimming | Swimming | Head immersion; | low |
| (Schonberg-Norio et al. 2004) | Finland | case control | any | GI | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Schonberg-Norio et al. 2006) | Finland | cross-sectional | any | GI | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Seyfried et al. 1985a) | Canada | prospective cohort | adults; any; children | allergy; any; ear; GI; RI; skin; | fresh; | any | swimming | Swimming | Unspecified direct; Body immersion; Head immersion; | low |
| (Slutsker et al. 1998) | United States | case control | any | diarrhea | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Soraas et al. 2013) | Norway | case control | adults | UTI | fresh; marine; | any | swimming | Swimming | Unspecified direct; Head immersion; | low |
| (Stuart et al. 2003) | United Kingdom | case control | any | GI | fresh; marine; | any | swimming | Swimming | Unspecified direct; Head immersion; Swallowing water; | low |
| (Taylor et al. 1995) | South Africa | cross-sectional | any | GI; other | fresh; | poor | canoeing | Minimal | Unspecified indirect; | low |
| (Umar et al. 2017) | Nigeria | cross-sectional | children | UTI | fresh; | any | fetching water, playing, bathing, swimming | Minimal; Swimming | Any contact; | low |
| (Unicomb et al. 2008) | United States | case control | any | GI | fresh; | any | swimming or wading | Swimming | Unspecified direct; | low |
| (van Asperen et al. 1995) | Netherlands | case control | any | ear | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (van Asperen et al. 1998) | Netherlands | prospective cohort | adults | GI | fresh; | any | swimming | Swimming | Head immersion; | low |
| (von Schirnding et al. 1993) | South Africa | prospective cohort | any | GI; RI; skin | marine; | good; poor | swimming | Swimming | Unspecified direct; | low |
| (Wade et al. 2004) | United States | prospective cohort | any; children | HCGI | fresh; | any | wading | Swimming | Unspecified direct; Unspecified indirect; Body immersion; | low |
| (Wade et al. 2006) | United States | prospective cohort | any | GI | fresh; | any | swimming | Swimming | Unspecified direct; Body immersion; Head immersion; | low |
| (Wade et al. 2008) | United States | prospective cohort | adults; any; children | ear; eye; GI; SRI; skin | fresh; | any; good; poor | swimming | Swimming | Unspecified direct; | low |
| (Wade et al. 2010) | United States | prospective cohort | any | diarrhea; ear; eye; SRD; skin | marine; | poor | swimming | Swimming | Unspecified direct; | low |
| (Wade et al. 2013) | United States | prospective cohort | any | ear | fresh; marine; fresh and marine; | any | swimming; wading | Swimming | Body immersion; Head immersion; | low |
| (Wade et al. 2018) | United States | prospective cohort | any | GI | marine; | any | swimming | swimming | Head immersion; | low |
| (Wiedenmann et al. 2006) | Germany | controlled trial | any | cold/flu; ear; eye; GI; RI; skin; UTI | fresh; | any | swimming | Swimming | Head immersion; | low |
| (Yapi et al. 2014) | Ivory Coast | cross-sectional | children | GI | fresh; | any | swimming, fishing | Mixed; Swimming | Any contact; | low |
| (Yau et al. 2014) | United States | prospective cohort | any | GI | marine; | any; poor | swimming | Swimming | Unspecified direct; Swallowing water; | low |
| (Zlot et al. 2015) | United States | retrospective cohort | any | GI | fresh; | any | swimming | Swimming | Unspecified direct; | low |
| (Zmirou et al. 1990) | France | retrospective cohort | any | any; ENT; eye; GI; HCGI; skin | fresh; | any | swimming | Swimming | Unspecified direct; | low |
Abbreviations: GI, gastrointestinal illness; HCGI, highly credible gastrointestinal illness; RI, respiratory illness; SRD, significant respiratory disease; ENT, ears, nose, and throat; UTI, urinary tract infection;
The majority (66/92) of studies meeting our inclusion criteria reported at least one risk estimate in the recreational activity category swimming. Most (55/66) of those studies reported risk estimates only for swimming. Substantially fewer studies reported risk estimates for sports-related contact (13/92), minimal contact (10/92), and sand contact (6/92).
The specific water contact category that studies meeting our inclusion criteria most often explicitly defined was head immersion (31/92). Fewer studies reported risk estimates for body immersion (15/92), swallowing water (13/92), and face splashing (1/92). Many (44/92) studies meeting our inclusion criteria reported at least one risk estimate in the water contact category unspecified direct contact, and fewer (10/92) studies reported at least one risk estimate in the water contact category unspecified indirect contact. About one quarter (21/92) of studies meeting our inclusion criteria provided illness risk estimates for more than one water contact category.
Of the 92 studies included in our meta-analysis, we assessed study quality as low risk of bias for 85 studies and high risk of bias for 7 studies. Risk of bias assessments for each study are provided in Supplementary Material Table S3.
3.3. Recreational activities
3.3.1. Gastrointestinal illness (GI)
Figure 3 shows the pooled risk of GI for each activity category using no contact as the reference group. We observe a consistent elevation in risks of GI for all activity categories.
Figure 3.
Pooled risk estimates of gastrointestinal illness (GI) for recreational activity categories swimming, sports-related contact, minimal contact, and sand contact using no contact as the reference group. Estimates are displayed on a logarithmic scale for each of the three GI outcomes categories (top) and for all data from the three GI outcome categories combined (bottom). Note: we consider results in gray inconclusive because they were derived from fewer than three studies. Arrowheads indicates ninety-five percent confidence intervals that are beyond the limits of the graph. Abbreviations: CI, confidence interval; N, total number of contributing studies; n, total number of contributing effect estimates from all contributing studies; I2, statistic indicating percentage of variance attributable to study heterogeneity; HCGI, highly credible gastrointestinal illness.
Our literature search retrieved the greatest amount of data for GI health outcomes we categorized as general GI. We found a significant elevation in risk of general GI associated with sports-related contact (2.70, 95% CI: 1.08, 6.78) and swimming (2.20, 95% CI: 1.82, 2.65), and a nonsignificant elevation in risk of general GI associated with sand contact (1.41, 95% CI: 0.94, 2.11) and minimal contact (1.19, 95% CI: 0.62, 2.31).
Although our literature search retrieved substantially fewer studies evaluating health outcomes we categorized as HCGI or diarrhea, our meta-analysis of the available data suggests a pattern of elevation in risk for these health outcome categories consistent with the pattern we found for general GI. We found a significant elevation in risk of HCGI associated with swimming (1.58, 95% CI: 1.08, 2.31) and a nonsignificant elevation in risk associated with minimal contact (1.38, 95% CI: 0.87, 2.21). We also found a significant elevation in risk of diarrhea associated with sports-related contact (6.70, 95% CI: 1.09, 41.25), swimming (2.03, 95% CI: 1.40, 2.95), and sand contact (1.39, 95% CI: 1.03, 1.86). No studies provided risk estimates for HCGI associated with the activity categories sports-related contact or sand contact, or for diarrhea associated with the activity category minimal contact. As specified in our protocol, we deemed the pooled risk estimates for HCGI associated with the category minimal contact and the pooled risk estimates for diarrhea associated with the activity categories sports-related contact and sand contact as inconclusive because they were derived from fewer than 3 studies.
Because of the relatively few estimates of illness risk for HCGI and diarrhea, we combined data from all 3 GI-related health outcome categories for each activity category separately and calculated the pooled risk for the health outcome category all GI outcomes combined. As expected, the pattern of pooled risk estimates for the category all GI outcomes combined generally follows the pattern of pooled risk estimates for the health outcome category general GI that contributed most of the data. We found a significant elevation in risk for all GI outcomes combined associated with sports-related contact (2.69, 95% CI: 1.04, 6.92) and swimming (2.19, 95% CI: 1.82, 2.63), and a nonsignificant elevation in risk associated with sand contact (1.42, 95% CI: 0.97, 2.07) and minimal contact (1.27, 95% CI: 0.74, 2.16).
We observed high heterogeneity in pooled risk estimates for general GI and all GI outcomes combined associated with the activity category swimming (I2 = 74% and 76%, respectively). To evaluate potential sources of heterogeneity, we performed a sensitivity analysis to evaluate the impact of study design, study quality, and water type (we evaluated the effect of water quality separately because many studies did not report water quality). Figure 4 shows the pooled risk estimate for all GI outcomes combined for the activity category swimming using all data (top row), followed by pooled risk estimates of subgroups selected by study design, study quality, and water type. Comparison of pooled risk estimates by study design and water type suggests these variables likely contribute to the heterogeneity enumerated by the I2 values. Evaluation of heterogeneity due to study quality is inconclusive because we assessed all the studies contributing data except for one that contributed to the analysis of all GI outcomes combined in the activity category swimming as low risk of bias.
Figure 4.
Sensitivity analyses of gastrointestinal illness (GI) associated with the swimming category for study design, study quality, risk of bias, and water type. Results in gray inconclusive because they were derived from fewer than three studies. Note that the sum of the number of studies for different water types is greater than 43 because 2 studies provided effect estimates for more than one water type. Abbreviations are the same as in Figure 3.
We evaluated publication bias by calculating pooled risk estimates for all GI outcomes combined in each activity category and the standard error of those estimates for each individual study (we used meta-regression to derive a study-specific effect estimate and standard error if a study provided multiple effect estimates for an activity category). Figure 5 shows the risk of GI obtained from each study plotted against the standard error of those estimates. Although scatter in the funnel plot for the category swimming appears asymmetrical, Egger’s test fails to reject the null hypothesis of a symmetrical distribution (p = 0.84). Funnel plots for the categories minimal contact and sand contact do not appear asymmetric, and Egger’s test fails to reject the null hypothesis of symmetrical distributions (p = 0.25 and 0.59, respectively). Scatter in the funnel plot for sports-related contact appears asymmetrical with fewer than expected smaller studies showing smaller risks. Egger’s test rejects the null hypothesis of symmetrical distributions (p = 0.03) suggesting publication bias. Trim and fill analysis suggests adding three studies may prevent small study publication bias resulting in a pooled risk estimate for sports-related contact of 1.52 (95% CI: 0.84, 2.83).
Figure 5.
Funnel plots of study-specific estimates of relative risk of gastrointestinal illness (horizontal axis) against the standard error of those estimates (vertical axis). Solid line indicates the pooled relative risk estimate for all studies combined. Dashed lines indicate pseudo 95% confidence intervals.
3.3.2. Respiratory illness (RI)
Figure 6 shows the pooled risk estimates of RI for different categories of recreational activities using no contact as the reference. We found fewer studies reporting risk estimates for RI (16 studies) compared to GI (43 studies). We found a significant elevation in risk of general RI associated with swimming (1.99, 95% CI: 1.53, 2.58), and sports-related contact (1.49, 95% CI: 1.00, 2.24). We found a nonsignificant increase in risk of general RI associated with minimal contact (1.33, 95% CI: 0.59, 3.03), although we deemed this result inconclusive because it was derived from fewer than 3 studies. We found a nonsignificant elevation in risk of SRD (1.40, 95% CI: 0.82, 2.39) and cough (1.15, 95% CI: 0.78, 1.71) for the activity category swimming. We found no elevation in risk of SRD for the activity category minimal contact (0.87, 95% CI: 0.69, 1.09) and a small nonsignificant elevation in risk of SRD for the activity category sand contact (1.01, 95% CI: 0.93, 1.09); however, we deemed these results inconclusive because they were derived from fewer than 3 studies. Like the outcome category general RI, we observed a significant elevation in risk of All RI outcomes combined associated with swimming (1.78, 95% CI: 1.38, 2.29) and sports-related contact (1.49, 95% CI: 1.00, 2.24). We found no elevation in risk associated with minimal contact (0.90, 95% CI: 0.71, 1.14), and a small nonsignificant elevation in risk associated with sand contact (1.01, 95% CI: 0.93, 1.09).
Figure 6.
Pooled risk estimates of respiratory illness (RI) for recreational activity categories swimming, sports-related contact, minimal contact, and sand contact using no contact as the reference group. Results in gray are inconclusive because they were derived from fewer than three studies. SRD, significant respiratory disease. All other abbreviations are the same as in Figure 3.
We found high heterogeneity among risk estimates for general RI and all RI outcomes combined in the activity category sports-related contact (I2 = 80% for both health outcome categories because they comprise the same set of data) and for cough in the activity category swimming (I2 = 84%). However, we could not perform a sensitivity analysis because of the small number of illness risk estimates in subgroups corresponding to study design, study quality, and water type.
Analysis of publication bias suggests a tendency toward publication of positive results for RI. Figure 7 shows the funnel plots for the overall risk of RI reported in each study for swimming and sports-related contact (we do not show funnel plots for minimal contact and sand contact because there were fewer than 3 studies providing illness risk estimates for those activity categories). The funnel plot for sports-related contact does not appear asymmetric, and Egger’s test fails to reject the null hypothesis of symmetrical distributions (p = 0.399). Scatter in the funnel plot for swimming appears asymmetrical with fewer than expected smaller studies showing smaller risks. Egger’s test suggests publication bias (p = 0.002), and a trim and fill analysis suggests adding three additional studies can prevent small study publication bias resulting in a pooled risk estimate for swimming of 1.51 (95% CI: 1.24, 1.83).
Figure 7.
Funnel plot of pooled effects for respiratory illness (horizontal axis) associated with swimming against the standard error of those estimate (vertical axis). Funnel plots for sports-related contact, minimal contact, and sand contact are not presented because the systematic literature search retrieved two or fewer studies providing risk estimates for those activity categories. All conventions are the same as in Figure 5.
3.4. Effect of water quality
We evaluated the effect of water quality on risk of GI and RI associated with different recreational activities by estimating pooled risk estimates from subsets of illness risk estimates we categorized as related to either good or poor water quality. There were relatively few studies contributing to this analysis because only 15 of 92 of the studies meeting our inclusion criteria provided information about water quality. As a result, we focused our analysis on two health outcome categories, all GI outcomes combined and all RI outcomes combined, which represent the health outcome data sets with the greatest amount of data.
Figure 8 shows the pooled risk estimates for each activity category stratified by good and poor water quality. We found a significant elevation in risk of GI associated with the activity category swimming for the subgroup poor water quality (2.00, 95% CI: 1.59, 2.51), and a smaller nonsignificant elevation in risk of GI associated with swimming for the subgroup good water quality (1.49, 95% CI: 0.91, 2.46). We also found a significant elevation in risk of GI associated with the activity category sports-related contact (3.72, 95% CI: 1.31, 10.60), and a smaller but significant elevation in risk of GI associated with sports-related contact in the subgroup good water quality (1.29, 95% CI: 1.04, 1.59), although we deemed the pooled risk estimate of GI associated with sports-related contact in the subgroup good water quality as inconclusive because it was derived from fewer than 3 studies.
Figure 8.
Pooled risk estimates for gastrointestinal illness (GI) and respiratory illness (RI) associated with “good” and “poor” water quality. We consider results in gray inconclusive because they were derived from fewer than three studies. Arrowheads indicates ninety-five percent confidence intervals that are beyond the limits of the graph. All abbreviations are the same as in Figure 3.
We found a nonsignificant elevation in risk of RI associated with the activity category swimming for the subgroup poor water quality (2.03, 95% CI: 0.97, 4.45) and a smaller nonsignificant elevation in risk associated with the activity category swimming for the subgroup good water quality (1.41, 95% CI: 0.92, 2.16). We found a significant elevation in risk of RI associated with the activity category sports-related contact for the subgroup poor water quality (1.65, 95% CI: 1.10, 2.47) and a smaller nonsignificant elevation in risk of RI associated with the activity category sports-related contact for the subgroup good water quality (1.24, 95% CI: 0.82, 1.89), although we deemed the smaller nonsignificant elevation in risk of RI associated with the activity category sports-related contact for the subgroup good water quality as inconclusive because it was derived from fewer than 3 studies.
3.5. Level of water contact
We analyzed the relationship between illness risk and level of water contact by calculating the pooled risk of GI and RI for different degrees of water contact regardless of the type of recreational activity underlying the water contact. Because a limited number of studies reported illness risk estimates related to water contact level, we focused on pooled risk estimates for GI by combining all GI health outcomes, and pooled risk estimates for RI by combining all RI health outcomes.
Figure 9 shows water contact-related pooled risk estimates using no contact as the reference group for all GI outcomes combined (top) and all RI outcomes combined (bottom). The figure presents pooled risk estimates for categories of water contact that studies explicitly described (swallowing water, head immersion, body immersion, face splashing, sand contact), and for categories of water contact that studies did not explicitly define but we could infer from information provided in each study (unspecified direct contact and unspecified indirect contact) (see 2. Materials and methods).
Figure 9.
Pooled risk estimates for gastrointestinal illness (GI) and respiratory illness (RI) with different levels of water contact. Results in gray are inconclusive because they were derived from fewer than three studies. Note: For each health outcome, the categories “unspecified indirect contact” and “all indirect contact” represent the same set of data (see text for explanation). All abbreviations are the same as in Figure 3.
We found elevation in risk of GI for all water contact categories, with the magnitude of GI risk generally following the extent of water contact. The top half of Figure 9 shows a significant elevation in risk of GI associated with the water contact categories body immersion (1.89, 95% CI: 1.27, 2.83), head immersion (1.98, 95% CI: 1.58, 2.46), and swallowing water (1.91, 95% CI: 1.27, 2.85), and a nonsignificant elevation in risk for the water contact categories sand contact (1.20, 95% CI: 0.80, 1.78) and face splashing (1.15, 95% CI: 0.99, 1.33). We also found a significant elevation in risk of GI associated with the water contact category unspecified direct contact (2.28, 95% CI: 1.74, 2.99) and a smaller nonsignificant elevation in risk of GI associated with the water contact category unspecified indirect contact (1.27, 95% CI: 0.74, 2.99).
To provide a general assessment of the association between level of water contact and GI risk, we combined all data we considered direct water contact into the general water contact category all direct contact and all data we considered indirect water contact into the general water contact category all indirect contact, and calculated separate pooled risk estimates for all GI outcomes combined and all RI outcomes combined for each general water contact category. We found a significant elevation in risk of GI associated with the water contact category all direct contact (2.23, 95% CI: 1.88, 2.65) and a smaller nonsignificant elevation in risk of GI associated with the water contact category all indirect contact (1.27, 95% CI: 0.74, 2.16).
The bottom half of Figure 9 shows the pooled risk estimates of RI with different levels of water contact generally followed the same pattern as GI except for the water contact category swallowing water. We observed a significant elevation in risk of RI associated with the water contact category head immersion (2.05, 95% CI: 1.48, 2.83), and a smaller nonsignificant elevation in risk of RI associated with the water contact category body immersion (1.66, 95% CI: 0.85, 3.24). We also found a small nonsignificant elevation in risk of RI associated with the water contact category sand contact (1.01, 95% CI: 0.93, 1.09), although we deemed this result inconclusive because it was derived from fewer than 3 studies. We found no elevation in the pooled risk estimates of RI for the water contact categories face splashing (0.89, 95% CI: (0.69, 1.15)) and swallowing water (0.99, 95% CI: 0.84, 1.16), although we deemed the result for face splashing as inconclusive because it was derived from fewer than 3 studies. We found a significant elevation in risk of RI associated with the water contact category unspecified direct contact (1.50, 95% CI: 1.13, 1.86) and no elevation in risk of RI associated with the water contact category unspecified indirect contact (0.90, 95% CI: 0.72, 1.12), although we deemed the result for the water contact category unspecified indirect contact as inconclusive because it was derived from fewer than 3 studies.
We found a significant elevation in risk of RI associated with the general water contact category all direct contact (1.70, 95% CI: 1.39, 2.07) and a smaller nonsignificant elevation in risk of RI associated with the general water contact category all indirect contact (0.90, 95% CI: 0.72, 1.12), although we deemed the result for the water contact category all indirect contact as inconclusive because it was derived from fewer than 3 studies.
3.6. Additional analyses
3.6.1. Children versus adults
We evaluated differences in risk of illness for children and adults by calculating pooled risk estimates for children and adults separately. Thirty-six studies meeting our inclusion criteria reported risk estimates for specific participant ages, age ranges, or age categories. Figure 10 shows the pooled risk estimates using no-contact as the reference for two health outcome categories, all GI outcomes combined and all RI outcomes combined, and stratified into the two subgroups children and adults. Because a limited number of studies reported illness risk estimates related to age, we focused on pooled risk estimates for GI by combining all GI health outcomes, and pooled risk estimates for RI by combining all RI health outcomes for the activity category swimming which is the activity category with the greatest amount of data. We found a significant elevation in risk of GI for the age subgroup children (2.53, 95% CI: 1.41, 4.55) and a smaller significant elevation in risk for the age subgroup adults (1.81, 95% CI: 1.35, 2.43). Although both pooled risk estimates are significantly greater than 1, they are not significantly different from each other. We found no increase in risk of RI for the age subgroup children (0.97, 95% CI: 0.69, 1.37) and a nonsignificant elevation in risk of RI for the age subgroup adults (1.58, 95% CI: 0.69, 3.60).
Figure 10.
Pooled risk estimates for all gastrointestinal illness (GI) outcomes (all GI outcomes combined) and all respiratory illness (RI) health outcomes (all RI outcomes combined) for the recreational activity category swimming stratified by age (adults or children). Recreational activity categories sports-related contact, minimal contact, and sand contact are not presented because the systematic literature search retrieved two or fewer studies providing risk estimates for those activity categories that also described study participants as either only adults or only children. All other conventions and abbreviations are the same as in Figure 3.
3.6.2. Other health outcomes
About one half (45/92) of the studies meeting our inclusion criteria reported at least one risk estimate for health outcomes other than GI or RI. We categorized these health outcomes of secondary interest as morbidities related to skin, ENT, eyes, and cold/flu (see Table 3) and performed meta-analysis on all available data. We performed meta-analysis on the activity category swimming and water contact categories body immersion, head immersion, and swallowing water because they were the only categories with data related to health outcomes of secondary interest.
In most cases we found a moderate elevation in risk for health outcome categories skin, ENT, eye, and cold/flu. Figure 11 shows the pooled risk estimates for these 4 health outcome categories. We observed a significant elevation in risk for the health outcome category skin associated with the activity category swimming (2.37, 95% CI: 1.77, 3.17) and water contact categories body immersion (1.68, 95% CI: (1.03, 2.75), head immersion (2.79, 95% CI: 1.48, 5.27), and swallowing water (1.51, 95% CI: 1.13, 2.00). We found a significant elevation in risk for the health outcome category ENT associated with the activity category swimming (1.83, 95% CI: 1.31, 2.56) and water contact category head immersion (1.80, 95% CI: 1.33, 2.42), and a nonsignificant elevation in risk associated with the water contact category body immersion (1.30, 95% CI: 0.96, 1.77). We found no elevation in risk for the health outcome category ENT associated with the water contact category swallowing water (0.97, 95% CI: 0.78, 1.20), although we deemed this result inconclusive because it was derived from fewer than 3 studies. We found a significant elevation in risk for the health outcome category eye associated with the activity category swimming (1.65, 95% CI: 1.17, 2.34), and water contact categories head immersion (2.79, 95% CI: 1.48, 5.27), and body immersion (2.24, 95% CI: 1.32, 3.81). We found a significant elevation in risk for the health outcome category eye associated with the water contact category swallowing water (1.77, 95% CI: 1.14, 2.24), but deemed this result inconclusive because it was derived from fewer than 3 studies. Finally, we found a significant elevation in risk for the health outcome cold/flu associated with the activity category swimming (1.60, 95% CI: 1.03, 2.47), and nonsignificant elevations in risk for the water contact categories head immersion (1.59, 95% CI: 0.74, 3.38) and swallowing water (1.08, 95% CI: 0.85, 1.36). We found no elevation in risk for the health outcome cold/flu associated with the water contact category body immersion (0.86, 95% CI: 0.61, 1.22), although we deemed this result inconclusive because it was derived from fewer than 3 studies.
Figure 11.
Pooled risk estimates of secondary health outcomes for different categories of recreational activities and water contact levels using no contact as the reference group. Results in gray are inconclusive because they were derived from fewer than three studies. All abbreviations are the same as in Figure 3.
4. Discussion
This is the first systematic review and meta-analysis comparing the risk of illness associated with different categories of recreational activities and different levels of contact with ambient surface waters. Our meta-analysis confirms a strong association between swimming and both GI and RI. Swimming approximately doubles the risk of GI and increases the risk of RI by approximately 75% compared to no-contact controls. This result is consistent with previous meta-analyses exclusively for swimming showing a strong relationship between swimming and GI (Leonard et al. 2018, Wade et al. 2003) and between swimming and RI (Mannocci et al. 2016). Our meta-analysis is also consistent with previous meta-analyses exclusively for swimming showing an association between swimming and skin-related health outcomes (Yau et al. 2009). This consistency is likely due to the relatively large body of scientific literature on the risk of illness associated with swimming.
In contrast to previous meta-analyses exclusively for swimming, our meta-analysis additionally suggests differences in risk of illness associated with recreational activities that correspond to the likelihood of water ingestion during recreation. We categorized recreational activities such as canoeing, kayaking, rowing, boating, fishing, standup paddle-boarding, or rafting when those activities occur in relatively tranquil waters as minimal contact, and activities such as surfing, diving, snorkeling, windsurfing, or canoeing, kayaking, or rafting in relatively rough waters as sports-related contact. We found an elevation in the risk of GI associated with minimal contact activities compared to a no contact reference that was less than the risk of GI associated with swimming. These results are consistent with studies suggesting that people who canoe, kayak, fish, motorboat, or row are less likely to ingest ambient water than swimmers (Dorevitch et al. 2011). We found an elevation in risk of GI associated with sports-related contact that was greater than the risk of GI associated with swimming (although publication bias may be contributing to our higher risk estimate). These results suggest that activity intensity may be an important determinant of exposure to water-borne pathogens and risk of illness.
Ingestion of water and subsequent risk of GI are presumed to be lower during minimal contact recreation. Still, it is expected that there is some degree of risk of acquiring illness through these activities. Inadvertent immersion can result in whole body contact, and splashing can lead to a variety of water contact scenarios. Illnesses affecting the skin and perhaps the mucous membranes of the eyes and ears may be of relatively greater importance for minimal contact recreation (USEPA 2003). Although the results of our meta-analysis of the secondary health outcomes skin, ENT, eye, and cold/flu suggests an elevation in risk with swimming and high levels of water contact, we could not calculate risk estimates for minimal contact or low levels of water contact for these health outcomes because none of the studies meeting our inclusion criteria contained relevant data. The precise risk of illness with activities we categorize as minimal contact is likely to be dependent on the potential for water contact during the specific activity.
We found no conclusive evidence of a risk of RI associated with minimal contact activities compared to no-contact controls. Although the pooled risk estimates of RI for minimal contact were derived from only two studies our results suggest the risk of RI is low during minimal contact activities and is lower than the risk associated with swimming. On the one hand, the results for RI are counterintuitive because RI presumably involves inhalation exposure of sprays or aerosols that could also occur during minimal contact activities. On the other hand, we define minimal contact activities as occurring on relatively tranquil waters where spray or aerosol generation may be less likely compared to waters where participants engaged in swimming or sports-related contact activities and where rough water or breaking waves may result in more spray or aerosols.
We separately estimated the risk of illnesses associated with recreational activities that are generally not characterized as swimming but nonetheless have a high degree of water contact. We categorized activities such as surfing, windsurfing, whitewater canoeing, whitewater kayaking, whitewater rafting, diving, and snorkeling as sports-related contact. We evaluated the risk of illness associated with this category of recreational activities separately because they are not commonly recognized as swimming but have the potential for similar or even greater exposure than the activities we categorized as swimming. Our results suggest the risk of GI associated with activities we categorize as sports-related contact is the same or greater than the risk of illness associated with swimming. Several outbreaks of GI and other illnesses have been associated with swimming during sports events such as triathlons which may occur in unmonitored or unregulated waterbodies such as rivers or canals with the potential for poor water quality (Davis and CDC 1998, Hall et al. 2017, Harder-Lauridsen et al. 2013, Parkkali et al. 2017). Our meta-analysis of GI risk stratified by water quality is consistent with greater GI risk in poor water quality conditions.
Our estimates of GI risk associated with different levels of water contact are consistent with the pattern of GI risk we found associated with different categories of recreational activities. Among explicitly defined levels of water contact we found greater GI risk associated with levels of contact with a high probability of water ingestion (body immersion, head immersion, and swallowing water) compared to contact with lower probability of water ingestion (sand contact and face splashing). We also found greater risk of GI associated with direct contact (all direct contact) compared to indirect contact (all indirect contact). These results are consistent with the occurrence of GI as associated with the amount of self-reported water exposure (Dorevitch et al. 2012b). Furthermore, the level of GI risk associated with body immersion, head immersion, and swallowing water are quantitatively similar, suggesting that body immersion alone may be enough to put recreators at significant risk of GI. This conclusion is consistent with a sensitivity analysis in Leonard et al. (2018) that reported the pooled risk of illness associated with any definition of swimming and the pooled risk of illness associated with swimming defined as including head immersion are not significantly different from each other.
The pooled risk estimates for RI generally followed the same pattern as GI. We found the highest risk of RI associated with swimming and sports-related contact whereas we found little or no elevation in risk for minimal contact. However, we considered the pooled risk estimates for RI associated with minimal contact as inconclusive because it was derived from fewer than 3 studies.
Although our meta-analysis includes 92 individual studies, it is nonetheless limited by the relatively small number of available studies with estimates of illness risk associated with activities we categorized as minimal contact and sports-related contact. Most studies meeting our inclusion criteria focused on swimming because regulatory entities are primarily concerned with protecting beachgoers from human fecal pollution in designated recreational waters where swimming is the most likely activity. Assessments of illness risk associated with swimming will likely continue to receive most research funding because it represents the largest at-risk population. However, increasing recognition of illness risk associated with non-swimming recreational activities and interest in developing water quality guidelines specifically designed to protect recreators engaged in those activities may prompt additional studies of non-swimming activities.
Our study was also limited by a lack of available data to adequately account for effect modifiers such as water quality, duration and frequency of water contact, age, type of surface water (i.e., marine or fresh), source of contamination (e.g., point versus non-point), naturally occurring organisms that can cause human illness, and heterogeneity in the way those effect modifiers are defined across studies. Most studies meeting our inclusion criteria either did not specify those parameters, did not stratify participants on the basis of those parameters, or did so in a manner that was difficult to categorize for meta-analysis. For example, defining children across a broad age range could potentially obscure elevated risk in young children. Greater attention to water quality and recreator age are relatively simple parameters that could provide important information on the risks associated with water-based recreation.
5. Conclusions
Recreational activities associated with less ambient water contact (e.g. fishing, boating, or kayaking in tranquil waters) may result in a lower risk of illness compared to recreational activities associated with greater ambient water contact (e.g. swimming, surfing, or whitewater kayaking). This study suggests that the level of exposure with different types of recreational activities are important characteristics when assessing the risk of illness associated with recreation in ambient surface waters.
Supplementary Material
6. Acknowledgements
This research was exclusively funded by the United States Environmental Protection Agency Office of Science and Technology in the Office of Water. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. This research has been subject to formal Agency review. The views expressed in this article are those of the authors and do not necessarily represent the views or policies of the United States Environmental Protection Agency. We are grateful to Brian Schnitker and Samar Khoury for reviewing the manuscript.
8. References
- Aiga H, Amano T, Cairncross S, Adomako J, Nanas OK and Coleman S (2004) Assessing water-related risk factors for Buruli ulcer: a case-control study in Ghana. Am J Trop Med Hyg 71(4), 387–392. [PubMed] [Google Scholar]
- Alaska Department of Environmental Conservation (2008) Water Quality Standards, Author, Juneau, Alaska. [Google Scholar]
- Alexander LM, Heaven A, Tennant A and Morris R (1992) Symptomatology of children in contact with sea water contaminated with sewage. J Epidemiol Community Health 46(4), 340–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Amin RM, Goweida MB, El Goweini HF, Bedda AM, Lotfy WM, Gaballah AH, Nadar AA and Radwan AE (2017) Trematodal granulomatous uveitis in paediatric Egyptian patients: a case series. Br J Ophthalmol 101(8), 999–1002. [DOI] [PubMed] [Google Scholar]
- Appleton CC and Bailey IW (1990) Canoeists and waterborne diseases in South Africa. S Afr Med J 78(6), 323–326. [PubMed] [Google Scholar]
- Arkansas Pollution Control and Ecology Commission (2018) Regulation Establishing Water Quality Standards for Surface Waters of the State of Arkansas.
- Arnold BF, Schiff KC, Ercumen A, Benjamin-Chung J, Steele JA, Griffith JF, Steinberg SJ, Smith P, McGee CD, Wilson R, Nelsen C, Weisberg SB and Colford JM Jr. (2017) Acute Illness Among Surfers After Exposure to Seawater in Dry- and Wet-Weather Conditions. Am J Epidemiol 186(7), 866–875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Arnold BF, Schiff KC, Griffith JF, Gruber JS, Yau V, Wright CC, Wade TJ, Burns S, Hayes JM, McGee C, Gold M, Cao Y, Weisberg SB and Colford JM Jr. (2013) Swimmer illness associated with marine water exposure and water quality indicators: impact of widely used assumptions. Epidemiology 24(6), 845–853. [DOI] [PubMed] [Google Scholar]
- Arnold BF, Wade TJ, Benjamin-Chung J, Schiff KC, Griffith JF, Dufour AP, Weisberg SB and Colford JM Jr. (2016) Acute Gastroenteritis and Recreational Water: Highest Burden Among Young US Children. Am J Public Health 106(9), 1690–1697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Atalabi TE, Lawal U and Akinluyi FO (2016) Urogenital schistosomiasis and associated determinant factors among senior high school students in the Dutsin-Ma and Safana Local Government Areas of Katsina State, Nigeria. Infect Dis Poverty 5(1), 69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Australian Government National Health and Medical Research Council (2008) Guidelines for managing risks in recreational water., Author, Canberra, Australia. [Google Scholar]
- Balarajan R, Soni Raleigh V, Yuen P, Wheeler D, Machin D and Cartwright R (1991) Health risks associated with bathing in sea water. BMJ 303(6815), 1444–1445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bonilla TD, Nowosielski K, Cuvelier M, Hartz A, Green M, Esiobu N, McCorquodale DS, Fleisher JM and Rogerson A (2007) Prevalence and distribution of fecal indicator organisms in South Florida beach sand and preliminary assessment of health effects associated with beach sand exposure. Mar Pollut Bull 54(9), 1472–1482. [DOI] [PubMed] [Google Scholar]
- Brazil National Standards (2005) Brazil National Standards CONAMA Resolution. Standards BN. (ed). [Google Scholar]
- Brown LM, Thomas TL, Ma JL, Chang YS, You WC, Liu WD, Zhang L, Pee D and Gail MH (2002) Helicobacter pylori infection in rural China: demographic, lifestyle and environmental factors. Int J Epidemiol 31(3), 638–645. [DOI] [PubMed] [Google Scholar]
- Cabelli VJ (1983) Health Effects Criteria for Marine Recreational Waters, U.S. Environmental Protection Agency, Cincinnati, OH. [Google Scholar]
- Cabelli VJ, Dufour AP, McCabe LJ and Levin MA (1982) Swimming-associated gastroenteritis and water quality. Am J Epidemiol 115(4), 606–616. [DOI] [PubMed] [Google Scholar]
- Cabelli VJ, Dufour AP, Mccabe LJ and Levin MA (1983) A Marine Recreational Water-Quality Criterion Consistent with Indicator Concepts and Risk Analysis. Journal Water Pollution Control Federation 55(10), 1306–1314. [Google Scholar]
- Calderon RL, Mood EW and Dufour AP (1991) Health effects of swimmers and nonpoint sources of contaminated water. Int J Environ Health Res 1(1), 21–31. [DOI] [PubMed] [Google Scholar]
- California Water Quality Control Board Central Coast Region (2017) Water Quality Control Plan for the Central Coastal Basin, California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- California Water Quality Control Board Colorado River Basin (2019) Water Quality Control Plan For The Colorado River Basin, California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- California Water Quality Control Board Los Angeles Region (2013) Water Quality Control Plan Los Angeles Region, California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- California Water Quality Control Board San Diego Region (2016) Water Quality Control Plan For The San Diego Basin, California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- California Water Quality Control Board San Francisco Region (2017) San Francisco Bay Basin (Region 2) Water Quality Control Plan (Basin Plan), California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- California Water Quality Control Board Santa Ana Region (2016) Water Quality Control Plan For The Santa Ana River Basin, California Environmental Protection Agency, Sacramento, California. [Google Scholar]
- Charoenca N and Fujioka RS (1995) Association of Staphylococcal Skin Infections and Swimming. Water Science and Technology 31(5–6), 11–17. [Google Scholar]
- Cheung WH, Chang KC, Hung RP and Kleevens JW (1990) Health effects of beach water pollution in Hong Kong. Epidemiol Infect 105(1), 139–162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chusri S, Sritrairatchai S, Hortiwahul T, Charoenmak B and Silpapojakul K (2012) Leptospirosis among river water rafters in Satoon, southern Thailand. J Med Assoc Thai 95(7), 874–877. [PubMed] [Google Scholar]
- CLARITY Group at McMaster University (2013) Tools to assess risk of bias in cohort and case control studies; randomized controlled trials; and longitudinal symptom research studies aimed at the general population. [Google Scholar]
- Colford JM Jr., Schiff KC, Griffith JF, Yau V, Arnold BF, Wright CC, Gruber JS, Wade TJ, Burns S, Hayes J, McGee C, Gold M, Cao Y, Noble RT, Haugland R and Weisberg SB (2012) Using rapid indicators for Enterococcus to assess the risk of illness after exposure to urban runoff contaminated marine water. Water Res 46(7), 2176–2186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Colford JM Jr., Wade TJ, Schiff KC, Wright CC, Griffith JF, Sandhu SK, Burns S, Sobsey M, Lovelace G and Weisberg SB (2007) Water quality indicators and the risk of illness at beaches with nonpoint sources of fecal contamination. Epidemiology 18(1), 27–35. [DOI] [PubMed] [Google Scholar]
- Corbett SJ, Rubin GL, Curry GK and Kleinbaum DG (1993) The health effects of swimming at Sydney beaches. The Sydney Beach Users Study Advisory Group. Am J Public Health 83(12), 1701–1706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cordero L, Norat J, Mattei H and Nazario C (2012) Seasonal variations in the risk of gastrointestinal illness on a tropical recreational beach. J Water Health 10(4), 579–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- D’Alessio D, Minor TE, Allen CI, Tsiatis AA and Nelson DB (1981) A study of the proportions of swimmers among well controls and children with enterovirus-like illness shedding or not shedding an enterovirus. Am J Epidemiol 113(5), 533–541. [DOI] [PubMed] [Google Scholar]
- Davis B and CDC (1998) Outbreak of acute febrile illness among athletes participating in triathlons - Wisconsin and Illinois, 1998 (Reprinted from MMWR vol 47, pg 585–588, 1998). J Am Med Assoc 280(17), 1473–1474. [Google Scholar]
- DeFlorio-Barker S, Wing C, Jones RM and Dorevitch S (2018) Estimate of incidence and cost of recreational waterborne illness on United States surface waters. Environ Health 17(1), 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Delaware Department of Resources and Environmental Control (2017) Surface Water Quality Standards.
- Denno DM, Keene WE, Hutter CM, Koepsell JK, Patnode M, Flodin-Hursh D, Stewart LK, Duchin JS, Rasmussen L, Jones R and Tarr PI (2009) Tri-county comprehensive assessment of risk factors for sporadic reportable bacterial enteric infection in children. J Infect Dis 199(4), 467–476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- DerSimonian R and Laird N (1986) Meta-analysis in clinical trials. Control Clin Trials 7(3), 177–188. [DOI] [PubMed] [Google Scholar]
- Dewailly E, Poirier C and Meyer FM (1986) Health hazards associated with windsurfing on polluted water. Am J Public Health 76(6), 690–691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- District Department of the Environment (2013) Water Quality Standards.
- Dorevitch S, DeFlorio-Barker S, Jones RM and Liu L (2015) Water quality as a predictor of gastrointestinal illness following incidental contact water recreation. Water Res 83, 94–103. [DOI] [PubMed] [Google Scholar]
- Dorevitch S, Dworkin MS, Deflorio SA, Janda WM, Wuellner J and Hershow RC (2012a) Enteric pathogens in stool samples of Chicago-area water recreators with new-onset gastrointestinal symptoms. Water Res 46(16), 4961–4972. [DOI] [PubMed] [Google Scholar]
- Dorevitch S, Panthi S, Huang Y, Li H, Michalek AM, Pratap P, Wroblewski M, Liu L, Scheff PA and Li A (2011) Water ingestion during water recreation. Water Res 45(5), 2020–2028. [DOI] [PubMed] [Google Scholar]
- Dorevitch S, Pratap P, Wroblewski M, Hryhorczuk DO, Li H, Liu LC and Scheff PA (2012b) Health risks of limited-contact water recreation. Environ Health Perspect 120(2), 192–197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dufour AP (1984) Health Effects Criteria for Fresh Recreational Waters, U.S. Environmental Protection Agency, Cincinatti, OH. [Google Scholar]
- Duval S and Tweedie R (2000a) A nonparametric “trim and fill” method of accounting for publication bias in meta-analysis. Journal of the American Statistical Association 95(449), 89–98. [Google Scholar]
- Duval S and Tweedie R (2000b) Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56(2), 455–463. [DOI] [PubMed] [Google Scholar]
- Dwight RH, Baker DB, Semenza JC and Olson BH (2004) Health effects associated with recreational coastal water use: urban versus rural California. Am J Public Health 94(4), 565–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Egger M, Smith G, Altman D and (eds) (2008) Systematic reviews in health care: meta-analysis in context, BMJ Publishing Group, London, UK. [Google Scholar]
- Eze JI, Scott EM, Pollock KG, Stidson R, Miller CA and Lee D (2014) The association of weather and bathing water quality on the incidence of gastrointestinal illness in the west of Scotland. Epidemiol Infect 142(6), 1289–1299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fattal B, Pelegolevsky E, Yoshpepurer Y and Shuval HI (1986) The Association between Morbidity among Bathers and Microbial Quality of Seawater. Water Science and Technology 18(11), 59–69. [Google Scholar]
- Feldman KA, Mohle-Boetani JC, Ward J, Furst K, Abbott SL, Ferrero DV, Olsen A and Werner SB (2002) A cluster of Escherichia coli O157: nonmotile infections associated with recreational exposure to lake water. Public Health Rep 117(4), 380–385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Feleke DG, Arega S, Tekleweini M, Kindie K and Gedefie A (2017) Schistosoma mansoni and other helminthes infections at Haike primary school children, North-East, Ethiopia: a cross-sectional study. BMC Res Notes 10(1), 609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ferley JP, Zmirou D, Balducci F, Baleux B, Fera P, Larbaigt G, Jacq E, Moissonnier B, Blineau A and Boudot J (1989) Epidemiological significance of microbiological pollution criteria for river recreational waters. Int J Epidemiol 18(1), 198–205. [DOI] [PubMed] [Google Scholar]
- Ferrer A, Nguyen-Viet H and Zinsstag J (2012) Quantification of diarrhea risk related to wastewater contact in Thailand. Ecohealth 9(1), 49–59. [DOI] [PubMed] [Google Scholar]
- Fewtrell L and Delahunty A (1985) The incidence of Cryptosporidiosis in comparison with other gastro-intestinal illnesses in Blackpool, Wyre, and Fylde. JCIWEM 9, 598–601. [Google Scholar]
- Fewtrell L, Godfree AF, Jones F, Kay D, Salmon RL and Wyer MD (1992) Health effects of white-water canoeing. Lancet 339(8809), 1587–1589. [DOI] [PubMed] [Google Scholar]
- Fewtrell L, Kay D, Salmon R, Wyer M, Newman G and Bowering G (1994) The health effects of low-contact water activities in fresh and estuarine waters. Water Environ J 8(1), 97–101. [Google Scholar]
- Fleisher JM, Fleming LE, Solo-Gabriele HM, Kish JK, Sinigalliano CD, Plano L, Elmir SM, Wang JD, Withum K, Shibata T, Gidley ML, Abdelzaher A, He G, Ortega C, Zhu X, Wright M, Hollenbeck J and Backer LC (2010) The BEACHES Study: health effects and exposures from non-point source microbial contaminants in subtropical recreational marine waters. Int J Epidemiol 39(5), 1291–1298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fleisher JM, Jones F, Kay D, Stanwell-Smith R, Wyer M and Morano R (1993) Water and non-water-related risk factors for gastroenteritis among bathers exposed to sewage-contaminated marine waters. Int J Epidemiol 22(4), 698–708. [DOI] [PubMed] [Google Scholar]
- Fleisher JM, Kay D, Salmon RL, Jones F, Wyer MD and Godfree AF (1996) Marine waters contaminated with domestic sewage: nonenteric illnesses associated with bather exposure in the United Kingdom. Am J Public Health 86(9), 1228–1234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fleming LE, Solo GH, Elmir S, Shibata T, Squicciarini D Jr., Quirino W, Arguello M and Van de Bogart G (2004) A Pilot Study of Microbial Contamination of Subtropical Recreational Waters. Fla J Environ Health 184, 29. [PMC free article] [PubMed] [Google Scholar]
- Galappaththi-Arachchige HN, Amlie Hegertun IE, Holmen S, Qvigstad E, Kleppa E, Sebitloane M, Ndhlovu PD, Vennervald BJ, Gundersen SG, Taylor M and Kjetland EF (2016) Association of Urogenital Symptoms with History of Water Contact in Young Women in Areas Endemic for S. haematobium. A Cross-Sectional Study in Rural South Africa. Int J Environ Res Public Health 13(11). [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gammie A and Wyn-Jones A (1997) Does hepatitis A pose a significant health risk to recreational water users? Water Science and Technology 35(11–12), 171–177. [Google Scholar]
- Goodman KJ, Correa P, Tengana Aux HJ, Ramirez H, DeLany JP, Guerrero Pepinosa O, Lopez Quinones M and Collazos Parra T (1996) Helicobacter pylori infection in the Colombian Andes: a population-based study of transmission pathways. Am J Epidemiol 144(3), 290–299. [DOI] [PubMed] [Google Scholar]
- Gray SF, Gunnell DJ and Peters TJ (1994) Risk factors for giardiasis: a case-control study in Avon and Somerset. Epidemiol Infect 113(1), 95–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Guam Environmental Protection Agency (2017) Water Quality Standards.
- H DEM., Gras Mughini, Schimmer L, Friesema B, AM IH, D.E.R.H. and V.A.N.P W. (2016) Gastrointestinal, influenza-like illness and dermatological complaints following exposure to floodwater: a cross-sectional survey in The Netherlands. Epidemiol Infect 144(7), 1445–1454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haile RW, Witte JS, Gold M, Cressey R, McGee C, Millikan RC, Glasser A, Harawa N, Ervin C, Harmon P, Harper J, Dermand J, Alamillo J, Barrett K, Nides M and Wang G (1999) The health effects of swimming in ocean water contaminated by storm drain runoff. Epidemiology 10(4), 355–363. [PubMed] [Google Scholar]
- Hall V, Taye A, Walsh B, Maguire H, Dave J, Wright A, Anderson C and Crook P (2017) A large outbreak of gastrointestinal illness at an open-water swimming event in the River Thames, London. Epidemiol Infect 145(6), 1246–1255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hamner S, Tripathi A, Mishra RK, Bouskill N, Broadaway SC, Pyle BH and Ford TE (2006) The role of water use patterns and sewage pollution in incidence of water-borne/enteric diseases along the Ganges river in Varanasi, India. Int J Environ Health Res 16(2), 113–132. [DOI] [PubMed] [Google Scholar]
- Harder-Lauridsen NM, Kuhn KG, Erichsen AC, Molbak K and Ethelberg S (2013) Gastrointestinal illness among triathletes swimming in non-polluted versus polluted seawater affected by heavy rainfall, Denmark, 2010–2011. PLoS One 8(11), e78371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harding AK, Stone DL, Cardenas A and Lesser V (2015) Risk behaviors and self-reported illnesses among Pacific Northwest surfers. J Water Health 13(1), 230–242. [DOI] [PubMed] [Google Scholar]
- Harrington J, Wilcox D, Giles P, Ashbolt N, Evans J and Kirton H (1993) The health of Sydney surfers: an epidemiological study. Water Science and Technology 27(3–4), 175–181. [Google Scholar]
- Health Canada (2012) Guidelines for Canadian Recreational Water Quality, 3rd Edition, Minister of Health, Ottawa, Ontario, Canada. [Google Scholar]
- Heaney CD, Sams E, Dufour AP, Brenner KP, Haugland RA, Chern E, Wing S, Marshall S, Love DC, Serre M, Noble R and Wade TJ (2012) Fecal indicators in sand, sand contact, and risk of enteric illness among beachgoers. Epidemiology 23(1), 95–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heaney CD, Sams E, Wing S, Marshall S, Brenner K, Dufour AP and Wade TJ (2009) Contact with beach sand among beachgoers and risk of illness. Am J Epidemiol 170(2), 164–172. [DOI] [PubMed] [Google Scholar]
- Higgins JPT, Green S and The Cochrane Collaboration (2011) Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. [Google Scholar]
- Idaho Department of Environmental Quality (2000) Water Quality Standards.
- Illinois Pollution Control Board (2011) Water Use Designations and Site-specific Water Quality Standards. Board IPC. (ed).
- Iowa Department of Natural Resources (2018) Water Quality Standards. [Google Scholar]
- Jessop EG, Horsley SD and Wood L (1985) Recreational use of inland water and health: are Windermere and Coniston water a health hazard? Public Health 99(6), 338–342. [DOI] [PubMed] [Google Scholar]
- Joosten R, Sonder G, Parkkali S, Brandwagt D, Fanoy E, Mughini-Gras L, Lodder W, Ruland E, Siedenburg E, Kliffen S and van Pelt W (2017) Risk factors for gastroenteritis associated with canal swimming in two cities in the Netherlands during the summer of 2015: A prospective study. PLoS One 12(4), e0174732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kansas Department of Health and Environment (2018) Surface Water Quality Criteria, Topeka, Kansas. [Google Scholar]
- Kay D, Fleisher JM, Salmon RL, Jones F, Wyer MD, Godfree AF, Zelenauch-Jacquotte Z and Shore R (1994) Predicting likelihood of gastroenteritis from sea bathing: results from randomised exposure. Lancet 344(8927), 905–909. [DOI] [PubMed] [Google Scholar]
- Energy Kentucky and Cabinet Environment (2019) Chapter 10 Water Quality Standards. [Google Scholar]
- Kueh CSW, Tam TY, Lee T, Wong SL, Llyod OL, Yu ITS, Wong TW, Tam JS and Bassett DCJ (1995) Epidemiological study of swimming-associated illnesses relating to bathing-beach water quality. Wat. Sci. Tech. 31(5–6), 1–4. [Google Scholar]
- Kundu A, McBride G and Wuertz S (2013) Adenovirus-associated health risks for recreational activities in a multi-use coastal watershed based on site-specific quantitative microbial risk assessment. Water Res 47(16), 6309–6325. [DOI] [PubMed] [Google Scholar]
- Lamparelli CC, Pogreba-Brown K, Verhougstraete M, Sato MI, Bruni Ade C, Wade TJ and Eisenberg JN (2015) Are fecal indicator bacteria appropriate measures of recreational water risks in the tropics: A cohort study of beach goers in Brazil? Water Res 87, 59–68. [DOI] [PubMed] [Google Scholar]
- Lee JV, Dawson SR, Ward S, Surman SB and Neal KR (1997) Bacteriophages are a better indicator of illness rates than bacteria amongst users of a white water course fed by a lowland river. Water Science and Technology 35(11–12), 165–170. [Google Scholar]
- Leonard AFC, Singer A, Ukoumunne OC, Gaze WH and Garside R (2018) Is it safe to go back into the water? A systematic review and meta-analysis of the risk of acquiring infections from recreational exposure to seawater. Int J Epidemiol 47(2), 572–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lepesteur M, McComb AJ and Moore SA (2006) Do we all face the same risk when bathing in the estuary? Water Res 40(14), 2787–2795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J and Moher D (2009) The PRISMA Statement for Reporting Systematic Reviews and Meta-Analyses of Studies That Evaluate Health Care Interventions: Explanation and Elaboration. PLoS Medicine 6(7), e1000100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Littell JH, Corcoran J and Pillai V (2008) Systematic reviews and meta-analysis, Oxford University Press, USA. [Google Scholar]
- Louisiana Department of Environmental Quality (2018) Water Quality.
- Ly S, Vong S, Cavailler P, Mumford E, Mey C, Rith S, Van Kerkhove MD, Sorn S, Sok T, Tarantola A and Buchy P (2016) Environmental contamination and risk factors for transmission of highly pathogenic avian influenza A(H5N1) to humans, Cambodia, 2006–2010. BMC Infect Dis 16(1), 631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mannocci A, La Torre G, Spagnoli A, Solimini AG, Palazzo C and De Giusti M (2016) Is swimming in recreational water associated with the occurrence of respiratory illness? A systematic review and meta-analysis. J Water Health 14(4), 590–599. [DOI] [PubMed] [Google Scholar]
- Marion JW, Lee J, Lemeshow S and Buckley TJ (2010) Association of gastrointestinal illness and recreational water exposure at an inland U.S. beach. Water Res 44(16), 4796–4804. [DOI] [PubMed] [Google Scholar]
- Massachusetts Department of Environmental Protection (2013) Massachusetts Surface Water Quality Standards, Massachusetts Department of Environmental Protection, Boston, Massachusetts. [Google Scholar]
- McBride G, Salmond C, Bandaranayake D, Turner S, Lewis G and Till D (1998) Health effects of marine bathing in New Zealand. International Journal of Environmental Health Research 8(3), 173–189. [Google Scholar]
- Medema G, Van Asperen I, Klokman-Houweling J, Nooitgedagt A, Van de Laar M and Havelaar A (1995) The relationship between health effects in triathletes and microbiological quality of freshwater. Water Science and Technology 31(5–6), 19–26. [Google Scholar]
- Michigan Department of Environment Great Lakes and Energy (2006) Part 4. Water Quality Standards. [Google Scholar]
- Minnesota Pollution Control Agency (2018) Water Quality Standards for Protection of Waters of the State. [Google Scholar]
- Missouri Department of Natural Resources (2011) Missouri Water Quality Standards. [Google Scholar]
- Montana Department of Environmental Quality (2014) Water Quality. [Google Scholar]
- Nardone A, Campese C and Capek I (2001) Les facteurs de risques de leptospirose en France métropolitaine. Sanitaire I.N.d.M.A.a.I.d.V. (ed), Saint-Maurice, France. [Google Scholar]
- Nevada Division of Environmental Protection (2018) Standards for Water Quality. [Google Scholar]
- New Jersey Department of Environmental Protection (2016) Surface Water Quality Standards. [Google Scholar]
- New Mexico Water Quality Control Division (2017) Standards for Interstate and Intrastate Surface Waters. [Google Scholar]
- New Zealand Ministry for the Environment (2002) Microbiological water quality guidelines for marine and freshwater recreational areas, Author, Wellington, New Zealand. [Google Scholar]
- O’Halloran C, Silver MW, Lahiff M and Colford J Jr. (2017) Respiratory Problems Associated with Surfing in Coastal Waters. Ecohealth; 14(1), 40–47. [DOI] [PubMed] [Google Scholar]
- OHAT (2015) OHAT Risk of Bias Rating Tool for Human and Animal Studies, National Toxicology Program, National Institute of Environmental Health Sciences, Rockville, Maryland. [Google Scholar]
- Ohio Environmental Protection Agency (2018) Water Quality Standards. [Google Scholar]
- Papastergiou P, Mouchtouri V, Pinaka O, Katsiaflaka A, Rachiotis G and Hadjichristodoulou C (2012) Elevated bathing-associated disease risks despite certified water quality: a cohort study. Int J Environ Res Public Health 9(5), 1548–1565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Papastergiou P, Mouchtouri VA, Rachiotis G, Pinaka O, Katsiaflaka A and Hadjichristodoulou C (2011) Bather density as a predominant factor for health effects related to recreational bathing: results from the Greek bathers cohort study. Mar Pollut Bull 62(3), 590–595. [DOI] [PubMed] [Google Scholar]
- Parkkali S, Joosten R, Fanoy E, Pijnacker R, V.A.N.B J., Brandwagt D. and V.A.N.P W. (2017) Outbreak of diarrhoea among participants of a triathlon and a duathlon on 12 July 2015 in Utrecht, The Netherlands. Epidemiol Infect 145(10), 2176–2184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pham-Duc P, Nguyen-Viet H, Hattendorf J, Cam PD, Zurbrugg C, Zinsstag J and Odermatt P (2014) Diarrhoeal diseases among adult population in an agricultural community Hanam province, Vietnam, with high wastewater and excreta re-use. Bmc Public Health 14, 978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Philipp R, Evans EJ, Hughes AO, Grisdale SK, Enticott RG and Jephcott AE (1985) Health risks of snorkel swimming in untreated water. Int J Epidemiol 14(4), 624–627. [DOI] [PubMed] [Google Scholar]
- Philipp R, Waitkins S, Caul O, Roome A, McMahon S and Enticott R (1989) Leptospiral and hepatitis A antibodies amongst windsurfers and waterskiers in Bristol City Docks. Public Health 103(2), 123–129. [DOI] [PubMed] [Google Scholar]
- Polkowska A, Rasanen S, Al-Hello H, Bojang M, Lyytikainen O, Nuorti JP and Jalava K (2018) An outbreak of Norovirus infections associated with recreational lake water in Western Finland, 2014. Epidemiol Infect 146(5), 544–550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Prieto MD, Lopez B, Juanes JA, Revilla JA, Llorca J and Delgado-Rodriguez M (2001) Recreation in coastal waters: health risks associated with bathing in sea water. J Epidemiol Community Health 55(6), 442–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Raghunathan PL, Whitney EA, Asamoa K, Stienstra Y, Taylor TH Jr., Amofah GK, Ofori-Adjei D, Dobos K, Guarner J, Martin S, Pathak S, Klutse E, Etuaful S, van der Graaf WT, van der Werf TS, King CH, Tappero JW and Ashford DA (2005) Risk factors for Buruli ulcer disease (Mycobacterium ulcerans Infection): results from a case-control study in Ghana. Clin Infect Dis 40(10), 1445–1453. [DOI] [PubMed] [Google Scholar]
- Ravel A, Pintar K, Nesbitt A and Pollari F (2016) Non food-related risk factors of campylobacteriosis in Canada: a matched case-control study. Bmc Public Health 16(1), 1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Republic of South Africa Department of Environmental Affairs (2012) South African water quality guidelines for coastal marine waters. Volume 2: Guidelines for recreational use, Author, Cape Town, South Africa. [Google Scholar]
- Rijal G, Tolson JK, Petropoulou C, Granato TC, Glymph A, Gerba C, Deflaun MF, O’Connor C, Kollias L and Lanyon R (2011) Microbial risk assessment for recreational use of the Chicago Area Waterway System. J Water Health 9(1), 169–186. [DOI] [PubMed] [Google Scholar]
- Ross AG, Olveda RM, McManus DP, Harn DA, Chy D, Li Y, Tallo V and Ng SK (2017) Risk factors for human helminthiases in rural Philippines. Int J Infect Dis 54, 150–155. [DOI] [PubMed] [Google Scholar]
- Sanchez-Nazario EE, Santiago-Rodriguez TM and Toranzos GA (2014) Prospective epidemiological pilot study on the morbidity of bathers exposed to tropical recreational waters and sand. J Water Health 12(2), 220–229. [DOI] [PubMed] [Google Scholar]
- Schonberg-Norio D, Sarna S, Hanninen ML, Katila ML, Kaukoranta SS and Rautelin H (2006) Strain and host characteristics of Campylobacter jejuni infections in Finland. Clin Microbiol Infect 12(8), 754–760. [DOI] [PubMed] [Google Scholar]
- Schonberg-Norio D, Takkinen J, Hanninen ML, Katila ML, Kaukoranta SS, Mattila L and Rautelin H (2004) Swimming and Campylobacter infections. Emerg Infect Dis 10(8), 1474–1477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seyfried PL, Tobin RS, Brown NE and Ness PF (1985a) A prospective study of swimming-related illness. I. Swimming-associated health risk. Am J Public Health 75(9), 1068–1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seyfried PL, Tobin RS, Brown NE and Ness PF (1985b) A prospective study of swimming-related illness. II. Morbidity and the microbiological quality of water. Am J Public Health 75(9), 1071–1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Slutsker L, Ries AA, Maloney K, Wells JG, Greene KD and Griffin PM (1998) A nationwide case-control study of Escherichia coli O157:H7 infection in the United States. J Infect Dis 177(4), 962–966. [DOI] [PubMed] [Google Scholar]
- Solo-Gabriele HM, Harwood VJ, Kay D, Fujioka RS, Sadowsky MJ, Whitman RL, Wither A, Canica M, Da Fonseca RC, Duarte A, Edge TA, Gargate MJ, Gunde-Cimerman N, Hagen F, Mclellan SL, Da Silva AN, Babic MN, Prada S, Rodrigues R, Romao D, Sabino R, Samson RA, Segal E, Staley C, Taylor HD, Verissimo C, Viegas C, Barroso H and Brandao JC (2016) Beach sand and the potential for infectious disease transmission: observations and recommendations. Journal of the Marine Biological Association of the United Kingdom 96(1), 101–120. [Google Scholar]
- Soraas A, Sundsfjord A, Sandven I, Brunborg C and Jenum PA (2013) Risk factors for community-acquired urinary tract infections caused by ESBL-producing enterobacteriaceae--a case-control study in a low prevalence country. PLoS One 8(7), e69581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- South Dakota Department of Environment and Natural Resources (2016) Surface Water Quality Standards.
- StataCorp (2015) Stata Statistical Software: Release 14, StataCorp LP, College Station, TX.
- Stevenson AH (1953) Studies of bathing water quality and health. Am J Public Health Nations Health 43(5 Pt 1), 529–538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stuart JM, Orr HJ, Warburton FG, Jeyakanth S, Pugh C, Morris I, Sarangi J and Nichols G (2003) Risk factors for sporadic giardiasis: a case-control study in southwestern England. Emerg Infect Dis 9(2), 229–233. [DOI] [PubMed] [Google Scholar]
- Tanner-Smith EE and Tipton E (2014) Robust variance estimation with dependent effect sizes: practical considerations including a software tutorial in Stata and SPSS. Res Synth Methods 5(1), 13–30. [DOI] [PubMed] [Google Scholar]
- Taylor MB, Becker PJ, Van Rensburg EJ, Harris BN, Bailey IW and Grabow WO (1995) A serosurvey of water-borne pathogens amongst canoeists in South Africa. Epidemiol Infect 115(2), 299–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Texas Commission on Environmental Quality (2009) Surface Water Quality Standards.
- Tipton E (2015) Small sample adjustments for robust variance estimation with meta-regression. Psychol Methods 20(3), 375–393. [DOI] [PubMed] [Google Scholar]
- Turbow DJ, Kent EE and Jiang SC (2008) Web-based investigation of water associated illness in marine bathers. Environ Res 106(1), 101–109. [DOI] [PubMed] [Google Scholar]
- Umar S, Shinkafi SH, Hudu SA, Neela V, Suresh K, Nordin SA and Malina O (2017) Prevalence and molecular characterisation of Schistosoma haematobium among primary school children in Kebbi State, Nigeria. Ann Parasitol 63(2), 133–139. [DOI] [PubMed] [Google Scholar]
- Unicomb LE, Dalton CB, Gilbert GL, Becker NG and Patel MS (2008) Age-specific risk factors for sporadic Campylobacter infection in regional Australia. Foodborne Pathog Dis 5(1), 79–85. [DOI] [PubMed] [Google Scholar]
- USEPA (2003) Bacterial Water Quality Standards for Recreational Waters Freshwater and Marine Waters Status Report, Author, Washington DC. [Google Scholar]
- USEPA (2012) Recreational Water Quality Criteria, Author, Washington DC. [Google Scholar]
- USEPA (2014) Microbiological Risk Assessment (MRA) tools, methods, and approaches for water media, Author, Washington, DC. [Google Scholar]
- Utah Department of Environmental Quality (2018) Standards of Quality for Waters of the State.
- van Asperen IA, de Rover CM, Schijven JF, Oetomo SB, Schellekens JF, van Leeuwen NJ, Colle C, Havelaar AH, Kromhout D and Sprenger MW (1995) Risk of otitis externa after swimming in recreational fresh water lakes containing Pseudomonas aeruginosa. BMJ 311(7017), 1407–1410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- van Asperen IA, Medema G, Borgdorff MW, Sprenger MJ and Havelaar AH (1998) Risk of gastroenteritis among triathletes in relation to faecal pollution of fresh waters. Int J Epidemiol 27(2), 309–315. [DOI] [PubMed] [Google Scholar]
- Veleizan AA, Flores VR and Viozzi GP (2016) [Cercarial dermatitis of bathers in northern Argentine Patagonia]. Medicina (B Aires) 76(1), 25–29. [PubMed] [Google Scholar]
- Virginia Department of Environmental Quality (2017) Water Quality Standards. [Google Scholar]
- Viswanathan M, Ansari M, Berkman N, Chang S, Hartling L, McPheeters M, Santaguida P, Shamliyan T, Singh K, Tsertsvadze A and Treadwell J (2012) Assessing the Risk of Bias of Individual Studies in Systematic Reviews of Health Care Interventions. [PubMed] [Google Scholar]
- von Schirnding YE, Kfir R, Cabelli V, Franklin L and Joubert G (1992) Morbidity among bathers exposed to polluted seawater. A prospective epidemiological study. S Afr Med J 81(11), 543–546. [PubMed] [Google Scholar]
- von Schirnding YER, Strauss N, Robertson P, Kfir R, Fattal B, Mathee A, Franck M and Cabelli VJ (1993) Bather morbidity from recreational exposure to sea water. Wat. Sci. Tech. 27(3–4), 183–186. [Google Scholar]
- Wade TJ, Augustine SAJ, Griffin SM, Sams EA, Oshima KH, Egorov AI, Simmons KJ, Eason TN and Dufour AP (2018) Asymptomatic norovirus infection associated with swimming at a tropical beach: A prospective cohort study. PLoS One 13(3), e0195056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wade TJ, Calderon RL, Brenner KP, Sams E, Beach M, Haugland R, Wymer L and Dufour AP (2008) High sensitivity of children to swimming-associated gastrointestinal illness: results using a rapid assay of recreational water quality. Epidemiology 19(3), 375–383. [DOI] [PubMed] [Google Scholar]
- Wade TJ, Calderon RL, Sams E, Beach M, Brenner KP, Williams AH and Dufour AP (2006) Rapidly measured indicators of recreational water quality are predictive of swimming-associated gastrointestinal illness. Environ Health Perspect 114(1), 24–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wade TJ, Pai N, Eisenberg JN and Colford JM Jr. (2003) Do U.S. Environmental Protection Agency water quality guidelines for recreational waters prevent gastrointestinal illness? A systematic review and meta-analysis. Environ Health Perspect 111(8), 1102–1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wade TJ, Sams E, Brenner KP, Haugland R, Chern E, Beach M, Wymer L, Rankin CC, Love D, Li Q, Noble R and Dufour AP (2010) Rapidly measured indicators of recreational water quality and swimming-associated illness at marine beaches: a prospective cohort study. Environ Health 9, 66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wade TJ, Sams EA, Beach MJ, Collier SA and Dufour AP (2013) The incidence and health burden of earaches attributable to recreational swimming in natural waters: a prospective cohort study. Environ Health 12, 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wade TJ, Sandhu SK, Levy D, Lee S, LeChevallier MW, Katz L and Colford JM Jr. (2004) Did a severe flood in the Midwest cause an increase in the incidence of gastrointestinal symptoms? Am J Epidemiol 159(4), 398–405. [DOI] [PubMed] [Google Scholar]
- WHO (2003) Guidelines for safe recreational water environments. Volume 1, Coastal and fresh waters., Malta. [PubMed] [Google Scholar]
- Wiedenmann A, Kruger P, Dietz K, Lopez-Pila JM, Szewzyk R and Botzenhart K (2006) A randomized controlled trial assessing infectious disease risks from bathing in fresh recreational waters in relation to the concentration of Escherichia coli, intestinal enterococci, Clostridium perfringens, and somatic coliphages. Environ Health Perspect 114(2), 228–236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wyoming Department of Environmental Quality (2018) Wyoming Surface Water Quality Standards. [Google Scholar]
- Yapi RB, Hurlimann E, Houngbedji CA, Ndri PB, Silue KD, Soro G, Kouame FN, Vounatsou P, Furst T, N’Goran EK, Utzinger J and Raso G (2014) Infection and co-infection with helminths and Plasmodium among school children in Cote d’Ivoire: results from a National Cross-Sectional Survey. PLoS Negl Trop Dis 8(6), e2913. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yau V, Wade TJ, de Wilde CK and Colford JM Jr. (2009) Skin-related symptoms following exposure to recreational water: a systematic review and meta-analysis. Water Qual Expo Health 1, 79–103. [Google Scholar]
- Yau VM, Schiff KC, Arnold BF, Griffith JF, Gruber JS, Wright CC, Wade TJ, Burns S, Hayes JM, McGee C, Gold M, Cao Y, Boehm AB, Weisberg SB and Colford JM Jr. (2014) Effect of submarine groundwater discharge on bacterial indicators and swimmer health at Avalon Beach, CA, USA. Water Res 59, 23–36. [DOI] [PubMed] [Google Scholar]
- Zlot A, Simckes M, Vines J, Reynolds L, Sullivan A, Scott AK, McLuckie JM, Kromer D, Hill VR, Yoder JS, Hlavsa MC, Centers for Disease C and Prevention (2015) Norovirus outbreak associated with a natural lake used for recreation - Oregon, 2014. MMWR Morb Mortal Wkly Rep 64(18), 485–490. [PMC free article] [PubMed] [Google Scholar]
- Zmirou D, Ferley JP, Balducci F, Baleux B, Fera P, Larbaigt G, Jacq E, Moissonnier B, Blineau A and Boudot J (1990) Evaluation of the microbial indicators of swimming-related health risk in recreational rivers. Rev. Epidem. et Sante Publ. 38, 101–110. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Studies reported crude or adjusted effect estimates derived from data not previously published or included enough unpublished data to allow calculation of crude relative risk. If a study reported both crude and adjusted effect estimates, we used only the adjusted estimates.