3a: Risk/benefit communication related to fish and seafood consumption

FDA and EPA have issued updated final advice on fish consumption. The advice updates the 2014 draft consumption advice and recommends that women who are pregnant (or might become pregnant) or nursing and young children eat certain amounts and types of fish in order to improve health and developmental outcomes while minimizing risk from methylmercury in fish. Since fish species vary in the extent to which eating the fish poses a risk of methylmercury exposure to the consumer, FDA and EPA developed recommended consumption frequencies to minimize the risk from methylmercury exposure while encouraging fish consumption. The advice graphically shows fish separated into three categories: Best Choices to be eaten 2-3 times a week, Good Choices to be eaten once a week only, and Fish to Avoid which the agencies are recommending the target audience of women of childbearing age and young children not consume. Accompanying Qs As are grouped in sections on the use of the chart, serving sizes, consumption by children, nutrients in fish, contaminants in fish, specific information on tuna, fish caught by family and friends, and other tips on eating fish. The advice and other related documents are available on the agencies websites at www.fda.gov/fishadvice and www.epa.gov/fishadvice.

The 2015 Dietary Guidelines Advisory Committee report recommends the consumption of seafood by pregnant women, following a similar recommendation by the 2010 DGAC. That advice is based on the exceptionally well grounded evidence for the need for certain nutrients, principally long chain omega-3 fatty acids, as the structural material required to develop a proper brain. Unlike most other foods, the DGAC focused on nutrient composition as a key issue, emphasizing the practical difference between nutrients in farmed versus wild fish induced by intensive feeding practices. It also emphasized the importance of minimizing environmental pollutants. The DGAC report is a technical document written to translate science to implications for practice. The Dietary Guidelines for Americans report is written based on the DGAC report and input from the public including industry. The DGA is then translated by practitioners such as registered dieticians to advice to the level of the consumer. The average consumer uses the words beef, pork, chicken, and pork to refer to single species, but uses the word fish (aka seafood) to refer to all sea animals, perhaps 100 species. Translation of nutrition advice to women of childbearing age must use consumer words that can be reliably implemented by, for instance, pregnant teenagers. Recommendations that distinguish between types of a particular fish (e.g. albacore vs light tuna) is lost on most consumers and leads to avoidance of all tuna. Advice which has been proposed to distinguish between subspecies of similarly named fish (e.g. bigeye vs skipjack tuna) that are seldom even identified on a product label is impossible even for the most careful consumers to reliably implement. The food selections that consumers make in the marketplace are a decision between one dish and another, though most scientific studies treat fish like an optional vitamin pill that may, or may not be consumed, rather than a food that will substitute for another food. Advice from the DGAC is simple: eat a variety of fish and seafood, 8-12 oz per week. Integrating this advice with the evidence that Americans consume excess red meat leads to the simple message to substitute 2-3 dishes per week with fish.

Initially fish consumption advisories focused upon a toxic constituent in fish to ensure that consumers were not exposed to greater than some daily dose limit of that constituent (e.g., the USEPA reference dose or RfD). This focus on risk avoidance created a potential loss of benefits since the predominant source of certain nutrients such as omega-3 fatty acids (O-3 FAs) is fish. We have been developing a quantitative risk/benefit model that provides an estimate of the net effect of positive and negative constituents in fish. Given that the focus of many fish advisories is on the potential for methyl mercury (meHg) to adversely affect neurodevelopment from prenatal exposure, our modeling has evaluated the counterbalancing effects of meHg and O-3 FAs on brain development. Using dose-response slopes for the benefit of O-3 FAs and the risk of meHg we have been able to simulate the net effect of fish ingestion on neurodevelopmental outcomes, with specific focus on visual recognition memory (VRM), an early life indicator of IQ. This quantitative modeling approach has enabled the evaluation of both locally caught and market purchased fish species based upon datasets describing the O-3 FA and Hg content of these species. The implications of such an approach for fish consumption advisories are that: 1) it provides a tool to evaluate the health effects of fish consumption in addition to the meHg RfD (risk only approach); 2) it provides an evidence-based rationale for advising greater fish consumption in those cases where in spite of the presence of meHg, the modeling predicts a clear benefit; 3)it allows for ready comparison of risks and benefits across seafood choices; and 4) it provides a framework for evaluating fish that can be improved as new toxicology/epidemiology data are gathered for these and other fish constituents. This presentation will compare this modeling approach with those developed by FAO/WHO, the USFDA and others, it will provide a framework for incorporating quantitative risk/benefit modeling information into the setting of consumption advisories, it will discuss the implications of the modeling for specific market purchased fish species, and it will highlight the uncertainties and data gaps which still limit its utility. Key among these are aspects of fish that may provide benefits and risks that are not captured by the focus on only two constituents (O-3 FAs and meHg). Recent efforts to expand the model to capture the neurodevelopmental risks associated with PCBs and the benefit from other constituents in fish will be described.

High levels of Hg in meat and organs from pilot whales, which are an important traditional food source for the Faroese were first reported in 1977. This finding led to the first consumption advisory for the general Faroese population from the Chief Medical Officer to limit the consumption of pilot whale to one meal per week and to completely avoid pilot whale liver and kidney. Since 1980, pregnant women were specifically advised to limit their consumption of pilot whale meat and blubber. In 1989, additional information on high levels of organochlorine contaminants in the blubber of pilot whales led to the consumption advisory that not more than 200 g of whale meat and blubber (each) should be consumed per month, and that liver and kidney should be avoided completely. In 1998, another advisory followed due to demonstrated effects of Hg and PCB exposure on the health of the fetus and newborns. This advisory focused on adults and most specifically on young and pregnant women. In 2008, the Faroese health authority concluded that pilot whales currently exceed limits for acceptable concentrations of toxic ontaminants and can no longer be recommended for human consumption. In the case of the Faroe Islands, the risk communication efforts appears to have been successful in convincing pregnant women to consume less pilot whale than before. While Hg levels in pilot whales have not decreased over the last three decades, concentrations in the blood of pregnant women have decreased significantly. While a dietary shift can be caused by several factors, it is likely that the risk communication undertaken in the Faroe Islands was the driving force for the decreases in human tissue levels of Hg and PCB for several reasons. For example, associated with the extensive cohort studies that have been ongoing since 1985, risk communication was continuous throughout the years and reached all areas of the islands. Further, risk communication messages were always restricted to pilot whale consumption, and several fish species with low contaminant concentrations were available and recommended as alternative dietary choices. However, health effects are still measurable even at these falling levels of exposure. Additionally, the success of the risk communication efforts and lower levels of contaminants in the Faroese population comes at a cost of loss of cultural identity for the Faroese people, who have relied on pilot whales as a staple part of their diet for hundreds of years.

Many women of childbearing age (WCBA) avoid eating fish due to concerns about mercury exposure, and therefore miss out on the benefits of fish consumption for themselves and their potential offspring. To increase consumption of fish low in mercury, we tested the impact of a short brochure using a personal narrative (messages communicated as part of a story about a hypothetical woman of childbearing age) to supplement traditional risk/benefit information about fish consumption among licensed WCBA anglers in the Great Lakes coastal region of the US. We conducted a randomized, two-wave longitudinal experiment, involving 1,135 WCBA drawn from a sample of licensed anglers, between May 18th, 2014 and September 5th, 2015. Participants reported their fish consumption in summer 2014 by completing an online diary for relevant meals. We then randomly assigned women to three groups for spring 2015: no-exposure control; sent brochure with a short personal narrative; sent brochure with no narrative. All WCBA participants completed a fish consumption diary again in summer 2015. Women who ate the least fish (< 0.7 meals/week at baseline) stood to benefit the most from increasing their fish consumption. Among these women who confirmed that they saw the brochure, those who were exposed to the brochure that included the narrative ate more fish the next year. These women increased their fish consumption largely by eating more low-mercury, purchased fish. These women did not increase their consumption of more contaminated fish. We also found that women who ate too much fish (>2.8 meals/week at baseline) were also influenced by the narrative form of the brochure. They ate fewer meals after receiving the brochure, but did not decrease their consumption sufficiently to be within advisory recommendations. It is not known how these changes impacted mercury ingestion. In the next stage of our analysis we will estimate actual mercury ingestion by linking data from womens diaries with available data on the mercury levels in different types of fish. We will use ordinary least squares regression to test the effect of the brochures on changes in ingested mercury. The results will be useful for government agencies responsible for communicating the benefits and risks of fish consumption.

BACKGROUND: Asians have higher methyl mercury (MeHg) levels compared to other U.S. racial/ethnic groups due to frequent seafood (fish and shellfish) consumption. There is little information about how seafood consumption attitudes and knowledge among Asians contribute to their consumption behavior. The aim of this analysis is to assess how beliefs about seafood consumption, as well as other social-demographic characteristics, impact seafood consumption behavior among three Asian ethnic groups in Chicago, using preliminary data from the Fish Intervention for Health (FISH) study.

METHOD: Fish consumption and social-demographic data were collected through in-person interviews from Chinese, Korean and Vietnamese adults who consumed seafood at least monthly. Social-demographic characteristics include age, sex, BMI, income, ethnicity, marital status, English proficiency and years in the US. We elicited awareness about benefits and risks of seafood consumption using open ended questions, and responses were classified into seventeen components. Logistic regression was used to examine the associations of knowledge about fish consumption risks and benefits with seafood consumption frequency, adjusting for covariates.

RESULTS: A total of 275 participants were included in this analysis, of which 229 (83.3%) were women. The mean age of the participants was 44.2 years old (range from 18 to 80). Eighty-nine (32.4%), 61 (22.1%) and 125 (45.5%) participants were Korean, Vietnamese and Chinese respectively. Of the 275 participants, 123 (44.7%) reported eating seafood at least 3 times/week, 128 (46.7%) 1-2 times/week and 24 (8.6%) 1-3 times/month. The most commonly reported known benefit about seafood consumption was that seafood is low in fat (36.4%) and the least reported was that seafood is good for breast feeding (1.5%). The most commonly reported risk about seafood consumption was elevated Hg in fish(30.2%) and the least reported risk was to avoid certain fish parts (1.5%). Among social-demographic characteristics, Korean ethnicity (OR: 3.3; 95%CI: 1.6-7.1; P=0.02) and receiving food stamps (OR: 0.5; 95%CI: 0.3-0.9; P=0.04) were associated significantly with seafood consumption frequency. Of the risks and benefits factors, only fish is good for the heart was associated significantly with increased seafood consumption (OR: 1.8; 95%CI: 0.9-3..7; P=0.09) at P level less than 0.1.

CONCLUSION: Knowledge about the benefits and risks of seafood consumption among study participants varied and did not influence seafood consumption behavior. Interventions to increase the awareness of the seafood advisories to lower Hg and PCB exposure while maintaining or improving consumption of omega-3 FAs from seafood are warranted in this study population.

Multiple studies have shown that physicians continue to be the most trusted source of health information, even with easy access to information on the internet. Physicians in the US receive little instruction on environmental health and specifically on the health benefits and risks of fish consumption. A 2011 survey of 2,625 obstetricians found that less than 25% take an environmental health history from their patients. The fish consumption message is complex since advice includes information on frequency of consumption as well as specific species to avoid and others to recommend. Methods to increase the capacity of physicians to incorporate environmental health into their clinical practices include offering courses for Continuing Medical Education credits, providing them with environmental screening forms, adding screening questions to the electronic health record, and including environmental health questions on the specialty board certification exams. In this presentation I will cover my recent efforts to increase physicians capacity to advise patients about healthy fish consumption. With funding from the Great Lakes National Program Office at USEPA, my team developed the Healthy Fish Choices on-line curriculum for healthcare providers that uses adult learning modalities to promote change in clinical practice to include advise on fish consumption. Of 105 providers from the Great Lakes states who evaluated the curriculum immediately after completing the modules, 89.2% stated that they would implement modifications in their clinical practice. Six months later, 79.1% had adopted new practices to enhance their care regarding fish consumption. With short patient visits and increased demands to address disease prevention as well as treatment, succinct screening questions can help physicians determine risk quickly and therefore indicate which patients need further testing or messaging. I will present results from a project in which we are testing a fish consumption screening question to predict elevated blood mercury in prenatal patients. These findings will inform interventions aimed at busy clinicians who dont have time for extensive risk assessments in everyday practice. Finally, I will describe the efforts of the EPA/ATSDR-funded Pediatric Environmental Health Specialty Units (PEHSUs) to reach out to prenatal providers and to function as expert consultants in reproductive environmental health in their federal regions. In the last three years each PEHSU has added a reproductive healthcare provider to its unit. These midwives and obstetricians are working to increase their own knowledge and expertise in reproductive environmental health while providing a much-needed link to their colleagues and professional organizations.

Humans exposed to methylmercury (MeHg) can suffer from adverse neurological impacts, and because eating fish is the primary mechanism of MeHg exposure, federal and state agencies issue fish consumption advisories (FCAs) to inform the public about the risks of eating contaminated fish. However, fish is also a good source of omega-3s which promotes infants neurological development. Health care providers (HCPs) provide similar information to their pregnant patients. The information from FCAs and HCPs enable pregnant women to make better food choices; however, the difficulty in communicating both the risks and benefits of eating fish can lead to over-restricted fish consumption.

To measure the effectiveness of Maines FCAs, we conducted a survey to assess the awareness of the advisory and measure advisory-induced changes in (1) knowledge of the risks and benefits of eating fish, (2) knowledge of the benefits of omega-3 fish oil consumption and (3) species-specific fish consumption behavior. We focus on examining differences between readers and non-readers of the advisory. We find the FCA successfully increased womens knowledge of both the benefits and risks of consuming fish. It also increased their: ability to differentiate fish by their MeHg content, knowledge of both low and high-MeHg fish and knowledge of detailed attributes of seemingly substitutable goods, such as white and light tuna. Non-readers could not identify fish that provide: health benefits like Omega-3 fatty acids or health risks like MeHg. Readers increased ability to make substitutions to minimize risk while maintaining the benefits of eating fish suggests the advisory may reduce MeHg-related health risks while avoiding the drop in fish consumption show in other studies.

Since HCPs are an important source of food information for pregnant women, we were interested in learning about their knowledge and perception of important food safety and nutrition topics. We used data from a national survey of OBGYNs, nurse practitioners, nurse midwifes, physician assistants, and WIC educators to examine the perceptions held by HCPs regarding their role to provide fish consumption advice during pregnancy. We found HCPs are not systematically providing all recommended advice about each of these topics to their pregnant patients. Rather, within each category HCPs are more likely to supply some advice over others. These findings suggest that, especially in multi-provider practices, clarifying the role of providing food safety and nutrition information to patients is important.