This section includes short summaries of papers that were useful for our study, written by COEC staff. It is intended to be a starting point for others seeking literature on the health effects of fish consumption. Click on the links to access the original abstracts and papers. This is not a comprehensive list of the available literature; please contact the COEC if you have suggestions for other papers to include.
- Health Effects Associated with Fish Consumption
- Fish Consumption and Advisory Awareness
- Assessment of Educational Strategies
Health Effects Associated with Fish Consumption
Ralston, N.V.C. 2008. Selenium health benefit values as seafood safety criteria. EcoHealth 5: 442 – 455.
A number of studies have been conducted to document the effects of maternal consumption of seafood containing methylmercury (MeHg) on children’s development. Some have shown serious detrimental effects, while others have demonstrated no effect, or even potential health benefits even with high MeHg concentrations. Conflicting results in such studies have led to a great deal of confusion about whether pregnant women should consume fish.
Studies of the mechanisms by which MeHg affects development have demonstrated that MeHg binds to Selenium (Se), making it unavailable to the brain. Se is a nutrient that is essential in many developmental functions, and without it children are unable to develop properly. Ralston hypothesizes that it’s not necessarily the amount of mercury in fish that influences a child’s development, but rather the ratio of MeHg to Se. If there is more mercury in food than Se available for uptake by the brain, then there will be detrimental health effects. Ralston briefly summarizes several famous cohort studies and demonstrates the support of this hypothesis in each example. Where real Se measurements were unavailable, estimates were made based on region and average concentrations in seafood.
Ralston concludes that women who are pregnant should stick to Se-rich seafoods; the research shows that, in many cases, the negative developmental effects resulting from not consuming seafood are significantly greater than the potential effects of MeHg. Ralston also emphasized that Se concentrations in seafood vary by region and waterbody. For example, most ocean fish is high in Se, while many freshwater lake species do not contain high amounts of the nutrient; these fish are likely to contain more Hg than Se.
Mozaffarian, D. and E.B. Rimm. 2006. Fish intake, contaminants, and human health: Evaluating the risks and the benefits. Journal of the American Medical Association 296(15): 1885 – 1899.
The authors conducted a review of scientific literature to clarify common misunderstandings regarding the risks and benefits of fish consumption. Fish, particularly oily species, contain high amounts of omega-3 fatty acids, which have been shown to decrease a person’s likelihood of death from coronary heart disease (CHD). Maternal consumption of fish has also demonstrated benefits through improved neurodevelopment in children. However, there are also health concerns with respect to methylmercury (MeHg), polychlorinated biphenyls (PCBs) and dioxins which have the potential to cause detrimental health effects. The authors assert that for the general population, the health benefits associated with moderate fish consumption far outweigh the risks. For children and women of childbearing age however, consumption should be decreased and limited to species known to contain fewer contaminants.
Oken, E., R.O. Wright, K.P. Kleinman, D. Bellinger, C.J. Amarasiriwardena, H. Hu, J.W. Rich-Edwards, and M.W. Gillman. 2005. Maternal fish consumption, hair mercury, and infant cognition in a U.S. cohort. Environmental Health Perspectives 113(10): 1376 – 1380.
Oken et al compared maternal fish consumption during pregnancy, maternal hair mercury at the infant’s birth, and infant cognition. While their study revealed a significant correlation between high fish consumption and high infant cognition, those infants whose mothers ate more than two meals of fish each week and had lower hair mercury levels significantly outperformed infants whose mothers ate more than two meals each week but had higher levels of mercury in their hair. The researchers conclude that mothers should indeed eat about two meals of fish each week during pregnancy, but that they should select fish lower in mercury.
Stewart, P., J. Reihman, E. Lonky, T. Darvill, and J. Pagano. 2000. Prenatal PCB exposure and neonatal behavioral assessment scale (NBAS) performance. Neorotoxicology and Teratology 22: 21 – 29.
The Neonatal Behavioral Assessment Scale (NBAS) examines three clusters of infant development that are associated with fish consumption: habituation, autonomic and reflex clusters. Stewart et al compared PCBs in cord blood of mothers who consumed Lake Ontario fish to infant NBAS scores and found a relationship between higher levels of heavily chlorinated PCB congeners (species) and impaired performance in the habituation and autonomic clusters. Although the level of PCBs in cord blood of mothers who consumed Lake Ontario fish did not significantly differ from levels in mothers who did not consume Lake Ontario fish, levels of highly-chlorinated PCB congeners were higher among fish consumers. In children born to mothers who had higher levels of heavily chlorinated PCBs, researchers saw an “increase in abnormal reflexes, increased startles, tremors, and other changes in autonomic functioning, and poorer habituation (response decrement) to repeated auditory, tactile, and visual stimulation.” Stewart et al argue that these findings suggest that “behavior changes associated with Lake Ontario fish consumption may be attributable to PCBs.” It is emphasized, however, that NBAS has not been analyzed as a predictor of future behaviors, making it difficult to predict lasting impacts of these deficits. The study also looked at DDE, HCB, mirex, lead and methylmercury, but found no association with changes in NBAS results.
J. F. Risher, H. E. Murray, G. R. Prince. 2002. Organic mercury compounds: human exposure and its relevance to public health. Toxicology and Industrial Health 18: 109 -160.
Although seafood is the typical vehicle of exposure to organic mercury in humans, there are many other mediums through which people may be exposed. The type of mercury that is found in fish and shellfish is methylmercury which bioaccumulates through food chains. Concentrations in tissues of top predators can be as much as 100,000 times higher than species at the bottom of the food chain. These authors suggest that hair collected from the scalp is the best biomarker for methylmercury. Other organic mercury sources may be topical disinfectants (thimerosal and merbromin), grain products (methyl and ethyl mercurials), older paints (phenyl mercury), and in some industrial processes or scientific laboratory equipment (dialkyl mercurials). Depending on the compound, these sources can enter the human body through inhalation, oral or dermal routes.
Bloom, M.S., J.E. Vena, M.K. Swanson, K.B. Moysich and J.R. Olson. 2005. Profiles of orthopolychlorinated biphenyl congeners, dichlorodiphenyldichloroethylene, hexachlorobenzene, and Mirex among male Lake Ontario sportfish consumers: the New York State Angler Cohort Study. Environmental Research 97: 178 – 194.
Bloom et al used frequency of consumption, whole-fish PCB estimates and meal size to develop an index for comparing consumer and non-consumer exposures to certain chemicals through Great Lakes sport fish consumption. The study revealed that long-term consumption of Great Lakes sport fish can increase organochlorine compounds (PCBs, Mirex, etc.) in the body, even at low levels of consumption. The authors note that this information can be used to compare other fish consumers, and may be useful in evaluating current consumption advisories for Great Lakes fish.
Fish Consumption and Advisory Awareness
R. Sechena, C. Nakano, S. Liao, N. Polissar, R. Lorenzana, S. Truong, and R. Fenske. 1999. Asian and Pacific Islander Seafood Consumption Study. (EPA 910/R-99-003).
As a preliminary study for an ultimate determination of risk assessment of fish consumption in the Pacific Northwest, members of the Environmental Protection Agency (EPA) created and conducted surveys investigating consumption rates, species, seafood parts consumed, and cooking methods of Asians and Pacific Islanders in King County, Washington. The study targeted 10 different ethnicities in this community to volunteer for a “Dietary Habits Study”. Additionally, participants were randomly chosen from lists offered by community groups and organizations. The results indicated that the API community members would receive the health messages better through select mediums being, in order of decreasing preference, books/pamphlets, verbal communication and video presentation. It was indicated that the preferred sources for these health messages were firstly community newspapers/newsletters, followed by television and word of mouth.
L. Knobeloch, H. A. Anderson, P. Imm, D. Peters, A. Smith. 2004. Fish consumptions, advisory awareness, and hair mercury levels among women of childbearing age. Environmental Research 97 (2005): 220 – 227.
Studies of the health effects of methylmercury (MeHg) suggest that it may have detrimental developmental effects. It is therefore recommended that women of childbearing age avoid consuming fish with high levels of mercury. This study was conducted to gauge the awareness of the fish consumption advisories amongst women who were of childbearing age, and to estimate the body burden of MeHg in women who consume fish. This study conducted surveys of 3,015 women of childbearing age in 12 U.S. states, and measured hair mercury in 414of the survey respondents. On average only 20% of participants were aware that their state had fish consumption advisories. Ninety-two percent of the participants had eaten fish within the past year, but only 29% reported eating sport caught fish. Thus the fish consumption advisories, formerly targeting sports fishermen need to be expanded to accommodate the more general population. Hair MeHg results did not allow the researchers to draw any significant conclusions regarding fish consumption and ethnicity because most participants were Caucasian, however the researchers did note a high frequency of consumption and high MeHg levels in Asian women. The majority of the participants reported that the most effective means of communicating health messages are TV news and newspapers.
K.K. Pflugh, L. Lurig, L.A. Von Hagen, S. Von Hagen, J. Burger. 1999. Urban anglers’ perception of risk from contaminated fish. The Science of the Total Environment 228: 203-218.
Pflugh et al interviewed anglers in the Newark Bay Complex, which consists of a number of marine waterbodies. Because of the “marine” characterization of many of the fishing sites, the anglers are not required to have a fishing license and thus do not necessarily receive advisory information that is contained within the Fish and Game Digest. Pflugh et al found that 60% of the anglers surveyed in the Newark Bay Complex reported having heard about advisories. However, many of the anglers that knew about the advisories did not believe them or were not concerned with the possible health detriments from eating certain fish species.
This study emphasizes that different income groups will respond to advisories in different ways. Some families must decide between obtaining food and avoiding potential detrimental health effects. This issue is important to consider when examining anglers’ belief in advisories. Anglers of the Newark Bay Complex considered to be “high-risk” gain information through different avenues. It was determined that a successful strategy for conveying these health messages is to use personal communication with anglers and have materials like press releases to newspapers that are both English and Spanish.
Q. Li, J.E. Vena, M.K. Swanson. 2005. Reliability of sport fish consumption in the New York State Angler Cohort Study. Environmental Research 97: 142 – 148.
Li et al assess the reliability of data collected in the 2001 New York State Angler Cohort Study (NYSACS). NYSACS is briefly described by the authors as a study in which anglers and their spouses/partners were randomly selected from the list of New York State licensed anglers. Participants were asked to complete self-surveys about species-specific fish consumption in the Great Lakes. About 100 randomly selected anglers were re-interviewed over the phone at a later date. This paper reports the reliability of using spousal and re-interview data to assess sport fish consumption.
The authors found that information provided by respondents and their spouses is comparable, and that spouses are a reasonable proxy data source when an individual is unavailable. The authors note that women are more likely to know their husband’s eating habits than vice versa. Likewise, re-interviewing participants is an effective way to determine the reliability of previous responses. The authors do note that “recall bias” is always a concern in such studies, and that the earliest information provided (i.e., the first interview/survey) is typically the most accurate.
Shayler, H.A. 2008. Mercury contamination in the northeastern United States: science-based decision making about fish consumption. – Master’s thesis, Cornell University
This Master’s thesis assessed the availability and clarity of fish consumption advisories and how this influences anglers’ decisions for consuming fish caught in the Adirondacks. The project also addressed data collection methods and risk assessment in Adirondack sportfish. In 2007, Shayler interviewed members of Adirondack fishing clubs and their families. These interviews contained questions about their fish consumption and were used to calculate approximate exposure to the risks of fish consumption (specifically mercury). Hair samples were also taken to measure actual mercury levels. Lastly, Shayler analyzed fish sample collection methods with respect to accurate mercury exposure/risk estimates. Shayler used existing fish mercury datasets with survey participant consumption estimates to approximate exposure of Hg to anglers who use private Adirondack waters (Adirondack Mountains, NY). Estimated exposures were compared to USEPA’s maximum recommended level (0.1 μg/kg) to determine whether estimated exposure exceeds this recommendation. Various datasets (Adirondack Fishery Research Program [AFRP], USFDA, etc) provided Hg estimates for fish caught from multiple sources or sold commercially. Known concentrations of Hg in sport caught fish were also used to calculate “safe” monthly consumption amounts for individuals.
Surveys also gathered information such as demographic characteristics, awareness of and familiarity with Hg and consumption advisories, and about factors for decision making about consumption. Survey results indicated that only half of the answers regarding consumption advisories were correct, and that only 3 out of 9 households responding to the survey answered all three consumption advisory questions correctly. Outreach (e.g., periodic newsletters) was actively conducted throughout the study to keep participants and community members up-to-date on the progress of the project, and to address questions raised by study participants.
Shayler collected hair samples from adult participants (volunteers) to measure Hg concentrations in the body. The mean Hg concentration exceeded the EPA threshold. However, it was found that most participants had Hg levels below the threshold. Participants exceeding this threshold reported consumption of more than twice as much fish than those who did not.
The amount of fish consumed is not the only factor influencing Hg concentrations in anglers. Shayler emphasizes the variability in Hg concentrations in fish (both between species, and in the same species from different areas). Understanding this variability she asserts that it is important to maintain an understanding of Hg and other contaminant concentrations by water bodies; current recommendations by species and region appear to be inadequate. Shayler argues that “more comprehensive datasets, together with detailed characteristics of fish consumption” are essential for accurately estimating Hg exposure. It is also noted that models that take participant body weight into account when estimating Hg exposure and hair Hg content are much more accurate than those that do not, indicating that body weight plays a significant role in concentration/exposure.
In summary, Shayler recommends a comprehensive approach to data collection – consumption,demographics, behaviors, knowledge of risk, available Hg datasets – that follows a pattern of:
- Targeted data collection
- Comprehensive risk assessment on an individual basis using Hg datasets, interviews and hair samples as a control for estimates
- Improved risk communication to encourage informed decision-making
Beehler, G.P., B.M. McGuinness, and J.E. Vena. 2001. Polluted fish, sources of knowledge, and the perception of risk: Contextualizing African Americans’ sport fishing practices. Human Organization 60(3): 288 – 297.
Beehler et al cite several studies detailing to the inaccessibility of consumption advisories, particularly for African Americans and other minorities. In 1999, the authors conducted four African American angler focus groups (FG) in Buffalo, NY to assess the perception of risk from an angler’s perspective. The authors report three central themes that emerged from FG discussions: 1) the context of sport fishing; 2) angler knowledge; and 3) pollution avoidance.
All FG participants mentioned “love of the sport” as a reason for fishing; no participants reported subsistence use of fisheries as a necessity. Participants referred to fishing as a means of relaxing and reducing stress, an important self-reported health benefit that is often overlooked. The heritage of fishing in African American culture was also emphasized during FG discussions. Participants note that most information is passed through generations, including where to fish, what to catch and keep, and how to prepare it. Likewise, African American anglers in Buffalo share information amongst themselves, with elders being respected as the most knowledgeable.
Risk communication from a scientific view is not part of the knowledge passed between African American anglers. Anglers are aware that some waters and fish are contaminated, but the “lay notion” of pollution discussed during FG is that pollution can be detected by the senses alone (so anglers avoid stagnant, littered or cloudy water; throw fish that appear sick along the banks to avoid contaminating other fish; and avoid eating fish that tastes bad). It is common for older anglers to advise on which waters to fish in for a ‘safe’ catch based on experience, though it is also reported that younger anglers tend to base safety on whether there are other anglers present. Beyond sensory cues and information from their peers, the participants of this study believed that the government (namely the EPA) regularly monitored fish and would warn of contamination./p>
Beehler et al conclude that scientific and lay definitions of what is safe differ significantly, and that new outreach and education methods should be used to better inform African American anglers in the Buffalo region. Although traditionally angler knowledge is based on past experience, the anglers practice a very adaptive knowledge system that can incorporate health information if presented appropriately. The authors note that elder anglers should be considered an education target for disseminating consumption advisories and information. Likewise, information should pair with existing knowledge (e.g., existing knowledge is that old fish are tough and have poor taste so they are thrown back – the science tells us that larger fish tend to have higher levels of contaminants).
Beehler, G.P., B.M. McGuinness, and J.E. Vena. 2001. Characterizing Latino Anglers' Environmental Risk Perceptions, Sport Fish Consumption, and Advisory Awareness. Medical Anthropology Quarterly 17(1): 99 – 116.
The same focus group (FG) study as described above was conducted with Latino anglers. In this case, six themes emerged from the discussions: 1) the benefits of fishing; 2) costs and risks associated with fishing; 3) Buffalo’s water quality; 4) perceptions of risk associated with consumption; 5) patterns of consumption and the distribution of fish; and 6) the efficacy of current risk-communication strategies. Post-group demographic surveys also revealed that most of the study participants had less than complete high school education, made less than $15,000 a year, consumed sport fish, were retired or not working, did not carry licenses and were (therefore) not aware of advisories.
Latino anglers in Buffalo primarily fish for leisure and as a social activity. Sport fish are considered a food source, but fishers discuss the financial limitations to fishing (e.g., cost of license [or fine with no license], bait, equipment) as reasons why it is not considered a “reliable subsistence strategy.” Likewise, licenses are seen as a financial or legal entity for the state, rather than an educational outlet, so accompanying advisories are typically ignored.
Elder anglers and one of two younger FGs relied on sensory cues to determine quality and therefore considered Great Lakes water to be clean. The fourth FG, consisting of young male anglers, strongly believed that lakes represent a closed system (likened to used bathwater) polluted by industry. Since all believed that fish are a “product of their environment,” three of four FG believed fish from the Great Lakes were uncontaminated, while the fourth FG did not and therefore refused to consume Great Lakes fish. Those who do consume the fish often do not prepare it as recommended – the skin is not removed, and it is battered and fried rather than baked, grilled or broiled. It is usually also not filleted.
As mentioned, most Latino anglers were unaware of the consumption advisories – 30% of FG participants had fishing licenses. Of these, few spoke fluent English. Participants noted that the guidebook handed out with licenses is an ineffective means of disseminating the information, and that strategies such as signs at fishing sites would be much more effective as is done in Canada.
Assessment of Education Strategies
G. Gliori, P. Imm, H.A. Anderson, L. Knobeloch. 2006. Fish Consumption and Advisory Awareness Among Expectant Women. Wisconsin Medical Journal 105: 41-44.
Brochures, posters and other outreach materials with health messages about the risks of developmental effects of fish consumption were distributed to healthcare clinics and facilities in the spring. In August of the same year, women who had given birth in June were surveyed. The survey asked whether the mother had seen such materials and asked questions regarding the content of outreach materials. It also characterized fish consumption of the mothers in the months leading up to their deliveries. 85% of the participants indicated that they had not seen the outreach materials and the average fish consumption was 3 meals per month. Many of the women that learned the messages modified their diets based on amount of fish consumed or types of fish consumed. Due to these findings it was determined that a more effective means of distributing the messages was necessary. The authors suggest a more vigorous method than simply mailing materials to a clinic, or some type of media campaign involving a public service announcement or advertisements to effectively communicate these health messages.
L. M. Westphal, M. Longoni, C. L. LeBlanc, A. Wali. 2008. Anglers’ Appraisals of the Risks of Eating Sport-Caught Fish from Industrial Areas: Lessons from Chicago’s Calumet Region. Human Ecology Review. 15.1: 46-62.
At 31 fishing sites in the Calumet region of northwest Indiana and southeast Chicago, researchers conducted an ethnographic study relying heavily on qualitative data contrary to most of these studies that use primarily quantitative data. This was based mostly on participant observation and unstructured interviews with fishermen. This purpose of the ethnographic study is to explain the impetus for social behaviors more than to reveal specific trends.
The authors found that the risk messages coming from various sources were sometimes contradictory and difficult to understand. They also found that there are differences in fish consumption patterns for different racial groups. They conclude by outlining the challenges of distributing and circulating the risk messages.
Several other studies in this bibliography also address the effectiveness of various outreach strategies, and include recommendations made by anglers through various interview-based studies.