Fishing, fish consumption and advisory awareness among Louisiana’s recreational fishers

Fishing, fish consumption and advisory awareness among Louisiana’s recreational fishers

Environmental Research 111 (2011) 1037–1045 Contents lists available at ScienceDirect Environmental Research journal homepage:

366KB Sizes 1 Downloads 52 Views

Environmental Research 111 (2011) 1037–1045

Contents lists available at ScienceDirect

Environmental Research journal homepage:

Fishing, fish consumption and advisory awareness among Louisiana’s recreational fishers$ Adrienne Katner a,b,n, Ebenezer Ogunyinka c, Mei-Hung Sun b, Shannon Soileau b, David Lavergne c, Dianne Dugas b, Mel Suffet a a

Environmental Science and Engineering Program, University of California at Los Angeles, Los Angeles, CA 90095, United States Louisiana Department of Health and Hospitals, Office of Public Health, Section of Environmental Epidemiology and Toxicology, New Orleans, LA 70112, United States c Louisiana Department of Wildlife and Fisheries, Office of Management and Finance, Socioeconomic Research and Development Section, Baton Rouge, LA 70808, United States b

a r t i c l e i n f o


Article history: Received 8 February 2011 Received in revised form 22 July 2011 Accepted 1 August 2011

This paper presents results from the first known population-based survey of recreational fishers in Louisiana (n ¼ 1774). The ultimate goal of this study was to obtain data in support of the development of regional advisories for a high exposure population with unique seafood consumption patterns. Between July and August of 2008, a survey was mailed to a random sample of licensed recreational fishers to characterize local fishing habits, sportfish consumption, and advisory awareness. Eighty-eight percent of respondents reported eating sportfish. Respondents ate an estimated mean of four fish meals per month, of which, approximately half were sportfish. Over half of all sportfish meals (54%) were caught in the Gulf of Mexico or bordering brackish areas. Sportfish consumption varied by license and gender; and was highest among Sportsman’s Paradise license holders (2.8 7 0.2 meals per month), and males (2.2 7 0.1 meals per month). The most frequently consumed sportfish species were red drum, speckled trout, catfish, bass, crappie and bream. Advisory awareness rates varied by gender, ethnicity, geographic area, license type, age and education; and were lowest among women (53%), AfricanAmericans (43%), fishers from the southeast of Louisiana (50%), holders of Senior Hunting and Fishing licenses (51%), individuals between 15 and 19 years of age (41%), and individuals with less than a high school education (43%). Results were used to identify ways to optimize monitoring, advisory development and outreach activities. & 2011 Elsevier Inc. All rights reserved.

Keywords: Fish Consumption Advisory Mercury Louisiana

1. Introduction State fish advisory programs are responsible for monitoring regional fish, and developing fish consumption advisories, for the purpose of reducing the risk of contaminant exposure. State fish advisories are targeted to residents who frequently consume

Abbreviations: LA, Louisiana; LDEQ, LA Department of Environmental Quality; LDHH, LA Department of Health and Hospitals; LDWF, LA Department of Wildlife and Fisheries; OPH, Office of Public Health; U.S. EPA, U.S. Environmental Protection Agency; RfD, Reference Dose; g/d, grams per day. $ This study was funded by the LA Department of Health and Hospitals (LDHH), Office of Public Health (OPH) in cooperation with the LA Department of Environmental Quality (LDEQ). The Louisiana Department of Wildlife and Fisheries (LDWF) distributed the survey, and conducted a preliminary data analysis and report (LDWF, 2009). The views expressed in this paper do not imply the expressions of any opinion on the part of the LDEQ, LDWF, LDHH; Tulane University or the University of California at Los Angeles. n Corresponding author at: Environmental Science and Engineering Program, University of California at Los Angeles, Los Angeles, CA 90095, United States. Fax: þ 1 504 219 4582. E-mail address: [email protected] (A. Katner). 0013-9351/$ - see front matter & 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2011.08.001

recreationally caught fish (sportfish), as opposed to commercial fish. The targeted population includes recreational, commercial and subsistence fishers, as well as individuals sensitive to the effects of low level contaminant exposure, such as children and women of child-bearing age. As of 2010, state officials in Louisiana (LA), Unites States (U.S.), have issued 57 fish consumption advisories, most of which are for methylmercury (84%), a ubiquitous fish contaminant which can affect fetal and childhood neurodevelopment and suppress immune systems (Mergler et al., 2007). To protect women of child-bearing age and children against methylmercury, the U.S. Environmental Protection Agency (EPA) and U.S. Food and Drug Administration (FDA) issued a national advisory with meal limit recommendations for commercial fish. It recommends against eating shark, swordfish, tilefish, and king mackerel; and restricting albacore tuna to one meal per week (US EPA, 2004). The advisory also recommends that women of childbearing age and children limit consumption of local sportfish from waterbodies not under advisory to one meal per week. Limiting seafood consumption is not a simple proposition to LA residents, for whom seafood is an integral part of the culture and economy. LA’s total annual fisheries products are second only


A. Katner et al. / Environmental Research 111 (2011) 1037–1045

to Alaska. Louisiana residents consume more fish than the general U.S. population (Dellenbarger et al., 1993). This may be due to the quantity of fresh seafood available in local markets—an estimated 80% of the seafood in LA’s commercial markets and restaurants is from state waters (US Department of Energy, 1997). Evidence exists which suggests that LA residents who consume local fish may be at risk from dietary exposure to methylmercury (Belanger et al., 2000; Lincoln et al., 2010). In LA, high blood mercury levels were observed in frequent fish consumers and commercial fishers residing near advisory areas (n¼ 313) (Belanger et al., 2000). A recent study reported that 40% of a sample of coastal recreational fishers in LA (n¼402) had hair mercury levels that correspond to doses exceeding the U.S. EPA’s health-based methylmercury Reference Dose (RfD, 41 mg/g) (Lincoln et al., 2010). The authors of this study estimated that mercury intake by LA recreational fishers is dominated by sportfish, in contrast to the average U.S. fish consumer, whose exposure is dominated by commercial fish. These results suggest that either, state advisories on sportfish are not sufficiently protective, or that they are not known or not followed. Given the dominance of local seafood on the commercial market these results highlight the potential for a pervasive public health problem, which remains largely untracked. LA’s fish consumption advisories are based on meal limits derived from equations presented in EPA’s Guidance for Assessing Chemical Contaminant Data for Use in Fish Advisories (EPA, 2000). Fish tissue contaminant concentration data are used with healthbased standards and standard default exposure assumptions. When local data are unavailable, LA’s Fish Advisory Program (referred hereafter to as ‘‘Program’’) assumes that an individual eats no more than four 227-g meals per month, or an average of 30 grams per day (g/d), of species-specific sportfish from a particular waterbody for 30 years. For mercury, advisories for the sensitive population are based on the EPA’s RfD, which is based on a benchmark dose analysis of the developmental and neurological impairment of a developing fetus (Rice et al., 2000). Advisories for the general population are based on the Agency for Toxic Substances and Disease Registry’s (ATSDR) less conservative Minimal Risk Level (MRL), which falls within the range of uncertainty of the EPA’s RfD. Meal limits less than four meals per month trigger additional monitoring, analysis or advisories. Advisories recommend population-, species- and waterbody-specific meal limits. Thus, advisories placed on a specific waterbody may recommend different meal limits by species and target population. As speciesspecific sportfish consumption rates are not considered during the advisory process, the result is often advisories for species with high contaminant concentrations that are infrequently consumed; while frequently consumed species low in mercury may be overlooked. Recent estimates indicate that high consumption of fish with low-to-moderate mercury levels drive mercury intake in LA’s coastal recreational fishers (Lincoln et al., 2010). To address this issue, information on species-specific sportfish consumption rates for Gulf State fishers is needed. Consumption rate data are also of interest given a massive oil spill which occurred in April of 2010, in which an estimated 4 million barrels of oil was released into the Gulf of Mexico (or Gulf), from the BP-licensed Transocean drilling rig, Deepwater Horizon. Since this time, a significant amount of resources has been diverted to monitoring Gulf seafood for the purpose of estimating potential exposures and risks, and evaluating the need for advisories. In addition to influencing future monitoring and advisory decisions for the affected area, species-specific fish consumption data will have a significant bearing on estimates of economic impact for the region. Due to the lack of data in available literature, a survey was developed to solicit information from a cross-section of licensed LA recreational fishers. This is the first known population-based survey of licensed LA recreational fishers. The primary

objectives of this study were to characterize: (1) fishing habits, (2) fish consumption patterns, and (3) advisory awareness among LA recreational fishers.

2. Methods 2.1. Sample selection A random sample (n¼ 6064) was selected from 503,336 residents who purchased recreational fishing licenses from the LA Department of Wildlife and Fisheries (LDWF) in the 2008 license year (i.e., June 1, 2007–May 31, 2008). To minimize sampling error and ensure representative responses, resident fishers from the LDWF database were stratified by license type and resident location, and selected using population-based proportional sampling with the ‘‘surveyselect’’ procedure in SAS (v 9.1; SAS Institute, Cary, NC) (see Supplementary Material Table 1). License types include: (1) Hook and Line (allows fishing with cane pole in both freshwater and saltwater), (2) Basic (freshwater only), (3) Saltwater (both saltwater and freshwater), (4) Sportsman’s Paradise (allows hunting and basic and saltwater fishing with all recreational fishing gear except trawls over 16 feet long), (5) Senior Hunting and Fishing (4 60 years of age—for hunting and fishing), and (6) Lifetime (for fishing only, hunting only, or combination of fishing and hunting). License group samples were further stratified into three geographic areas (southeast, southwest and north), as determined by the administrative regions of the LA Department of Health and Hospitals’ Office of Public Health (Supplementary Material Fig. 1). Hook and line, sportsman’s paradise and lifetime license strata were oversampled by a factor of ten in all geographic areas due to the need to correct for small sample size (Supplementary Material Table 1).

2.2. Survey design and administration The Louisiana Recreational Fisherman and Health Advisory Survey (see Supplementary Material Appendix A) was developed by the Louisiana Department of Wildlife and Fisheries (LDWF), Socioeconomic Research and Development Section in consultation with the Louisiana Department of Environmental Quality (LDEQ) and Louisiana Department of Health and Hospitals (LDHH). Survey questions were based on guidance provided by EPA (2000). An informal pre-test with colleagues was used to ensure that the questionnaire was understood by respondents as the authors intended. The survey was designed to characterize demographics, fishing habits, fish consumption patterns and advisory awareness over a 12 month period prior to survey response. As such, responses are subject to recall bias. No information was obtained about portion sizes, body weight, children’s consumption frequency or preparation methods. Surveys were mailed out between July and August of 2008; and responses were received between July 2008 and January 2009. To ensure a robust response rate, a second mailing of 5475 additional questionnaires to the first set of non-respondents was conducted four weeks later. To evaluate exposure assumptions used by the Program to develop fish advisories, fishers were asked to indicate how many years they have fished (n ¼1546), which is assumed to approximate exposure duration; and to estimate their consumption frequency for all fish (n¼ 1713), and all sportfish (n¼1685). To evaluate exposure assumptions and monitoring plans, fishers were asked to indicate their three most frequently eaten sportfish (n¼ 1218); and three favorite freshwater and saltwater fishing spots (n¼ 901). The most popular fishing spots were those in which the overall annual sum of fishing trips for the population exceeded 100. Popular waterbodies and species were evaluated to identify potentially under-, and over-sampled waterbodies and fish species by comparing them against waterbodies and species sampled by the Program, as reported in the LA Department of Environmental Quality’s (LDEQ) Fish Tissue Mercury Database and the LA Department of Health and Hospital’s (LDHH) Fish Advisory Database. Summary statistics for total fish and total sportfish consumption frequencies were calculated only from individuals who reported eating fish (n¼ 1681) and sportfish (n¼1490), in line with the standard U.S. EPA approach (U.S. EPA, 1998). Total sportfish consumption frequency was defined as the product of ‘‘times per month you eat fish’’ and ‘‘percent of the fish you eat is caught by you’’. Species-specific sportfish consumption frequencies were provided by respondents for the three most frequently eaten self-caught sportfish in terms of ‘‘times eaten per month’’. These measures (total sportfish consumption and species-specific sportfish consumption) may be underestimates as they do not include consumption of sportfish caught by others. It is assumed that as the respondents are licensed recreational fishers, most of the sportfish they eat will be self-caught. In addition, only data from the top three most frequently eaten sportfish were obtained. About half of the fishers who responded to questions about species-specific sportfish consumption identified three species (53%). Over half (52%) of these individuals reported eating less than one meal per month of the third species. As such, while this measure may underestimate total species-specific sportfish consumption rates for the population, it may serve our purpose in identifying the most popular fish consumed. To evaluate advisory awareness and effectiveness, fishers were asked (1) if they have ever seen, heard or read about fish consumption warnings (n ¼1745);

A. Katner et al. / Environmental Research 111 (2011) 1037–1045 (2) if aware of advisories, to indicate all ways in which they were informed (n¼ 1028); (3) to indicate all ways in which they prefer to be informed about advisories (n¼ 1675); (4) if they were aware of advisories, if and how advisories affected their fishing (n¼1018) and fish consumption behavior (n¼ 1040); and (5) if aware of advisories, but behavior did not change (n¼ 733), to indicate the reasons why (n¼ 693). 2.3. Survey analysis Survey data were weighted before analysis to reflect the license-specific angler population selection probability using LDWF’s database of recreational fishers. Statistical analyses of prevalence estimates, chi-square, ANOVAs, T-tests, Wilcoxon rank sum, Kruskal–Wallis and Dunn’s tests were conducted using SAS, R (v 2.12.0, Free Software Foundation, Boston, MA) and Scout (v 1.0, National Exposure Research Laboratory, Las Vegas, NV). The probability of type I error was set at a ¼0.05. 2.3.1. Response rates Response rates are summarized in Supplementary Material Table 1. Samples were adjusted for non-deliverable surveys before response rates were derived. The overall response rate of 33% (1774 respondents) is comparable to past LDWF surveys. An analysis of fishers responding to the second mail request (n¼ 812) was conducted in an attempt to characterize ‘‘non-respondents’’. No significant differences were observed between the first (n¼962) and second (n¼812) wave of respondents with respect to license, location, education, ethnicity, age, income and gender (w2, p o 0.05).

3. Results 3.1. Sociodemographics Select sociodemographics of respondents are summarized in Table 1. Most respondents were Caucasian males between the ages of 40 and 59. Almost half of our population came from the southeast (47%). Forty-percent of our survey population is comprised of individuals with a saltwater fishing license (with both freshwater and saltwater fishing rights). Over one-third of our population had only a high school education (36%). Twelve percent of respondents were females of child-bearing age (between 18 and 45 years of age); and 20% had a child less than 8 years in the household within the 12 months preceding this survey. Twenty percent of all respondents were seniors ( 460 years); and 14% of all respondents reported being retired (n¼ 1726). The percent of respondents in each license group is presented by location in Fig. 1. Most Southern respondents hold a Saltwater license (in addition to a Basic license); and most Northern respondents hold a Basic license. 3.2. Fishing 3.2.1. Fishing purpose and location When asked to indicate all reasons they fish, 78% of respondents said to catch fresh fish for food and 10% said to reduce the amount my family spends on food (n¼ 1721). This may suggest a strong reliance on local seafood for sustenance—either necessary or by choice. In the year prior to the survey, 42% of respondents participated in freshwater activities only; 37% participated in both freshwater and saltwater activities; and 21% participated in saltwater activities only (n ¼1746). When broken down by geographical location, 97% of all fishing trips taken by respondents from the north occurred in freshwater areas (n ¼501). By comparison, 52% of all fishing trips taken by respondents from the south occurred in freshwater areas, while 48% occurred in saltwater areas (n ¼1245). By license types, respondents that took the most saltwater trips over the year prior to the survey were holders of a Saltwater license (10 saltwater fishing trips per person per year); Sportsman’s Paradise (8); and Lifetime licenses (7). Basic license holders took the most freshwater fishing trips over the year prior to the survey (18); followed by those with Sportsman’s Paradise licenses (16) (n¼ 1746). When asked to


list their three favorite freshwater and saltwater fishing spots (n ¼901), 30% of the most popular fishing spots in which the overall annual sum of fishing trips exceeded 100 (n ¼71), were under advisory. All of the most popular waterbodies were monitored either directly or upstream. Waterbodies with over 500 fishing trips for the group included, in descending order: Toledo Bend, Lake Pontchartrain, Red River, Atchafalaya Basin, Lake Claiborne, Henderson Lake and Bayou D’Arbonne Lake (Supplementary Material Table 2), of which two are under advisory (Toledo Bend and Henderson Lake). 3.2.2. Temporal fishing patterns Fishers were asked how many years they had fished in LA (Fig. 2). Overall, the average was 32 years (n¼1546). Senior respondents had the highest average with 41 years (n¼297). The average exposure period for women of child-bearing age was 18 years (n ¼175). While over half of the respondents (52%) had over 30 years of fishing experience, which is LA’s default exposure duration assumption, the assumption that exposure is constant over time may not be realistic for all species. Some species are not available year-round, or have seasonal capture restrictions. The most popular months for fishing were March through May. 3.3. Fish consumption 3.3.1. Total fish consumption Ninety-eight percent of respondents reported that they eat fish (commercial or sportfish) (n ¼1713). When asked how many times per month they ate fish, the mean monthly consumption frequency of all fish (both commercial and sportfish) reported by consumers was approximately four meals per month or 44 meals per year (3.770.10 meals per month). By comparison, the adult US fish consuming population reported eating 36 meals per year, based on a fish meal size of 227 g per meal and mean daily average consumption rate of 97.56 g of all fish (U.S. EPA, 2002). Twenty-four percent of respondents eat more than four meals per month of all fish (Fig. 3A). 3.3.2. Sportfish consumption Eighty-eight percent of respondents reported that they eat sportfish (n ¼1685). This is high compared to a survey of residents in the Great Lakes states, of which only 22% reported eating any sport-caught fish (Imm et al., 2005). Sportfish made up about half (48–51%) of total fish consumed in all geographic areas (Fig. 3A); which is lower than the proportion of sportfish eaten by coastal LA recreational fishers (64%) reported by Lincoln et al. (2010) (64%, n¼ 534). However, both estimates are higher than the proportion of sportfish eaten by the general public (Sunderland, 2007). The average sportfish consumption frequency for fishers in this study was approximately two meals per month (2.11 70.06 meals per month, Table 1), or 25 sportfish meals per year. This estimate is higher than annual average sportfish consumption reported for Great Lake sportfish consumers (13 sportfish meals per year; Imm et al., 2005). Sportfish consumption by LA fishers may be even higher considering the fact that this measure may have been under-estimated in our survey due to limitations referred to in the methods section. Nine percent of all sportfish consumers eat more than four total sportfish meals per month, which is LA’s default assumption for maximum consumption frequency (Fig. 3A). Fifty-four percent of all sportfish meals were identified as saltwater or brackish fish, obtained from the Gulf or bordering coastline. Ten percent of respondents were classified as ‘‘high consumers’’, who ate more than four sportfish meals per month. High consumers ate an average of seven sportfish meals per month, ranging from 5 to 30 sportfish meals per month (one


A. Katner et al. / Environmental Research 111 (2011) 1037–1045

Table 1 Socio-demographics, sportfish consumption and advisory awareness of survey participants. Sportfish consumption frequencya (meals/month)


Advisory awareness P-valueb

%c (N)


Sample size (N)



Mean 7SE


90th percentile


Total population




2.11 7 0.06




Gender Male Female Total

1309 465 1774

76 24 100

1147 343 1490

2.18 7 0.07 1.88 7 0.14

1.5 1.0

5.0 4.0

30.0 18.8

o 0.05f

59 (1300) 53 (445) 1745

o 0.05

Ethnicity/Race C AA Oth UR Total

1516 116 101 41 1774

85 7 6 2 100

1300 89 79 22 1490

2.07 7 0.07 2.63 7 0.36 2.04 7 0.23 2.74 7 0.68

1.5 1.5 1.5 1.4

4.8 7.8 4.0 7.0

30.0 18.8 9.5 12.0


60 (1506) 43 (112) 42 (99) 46 (28) 1745

o 0.05

State geographic area SE 826 SW 441 N 507 Total 1774

47 25 29 100

718 367 405 1490

2.13 7 0.09 2.08 7 0.12 2.11 7 0.13

1.5 1.3 1.5

4.8 5.0 5.0

19.8 16.0 30.0


50 (817) 67 (431) 62 (497) 1745

o 0.05

License type B H&L L SW Snr SP Total

393 114 173 709 247 138 1774

22 6 10 40 14 8 100

315 83 154 632 183 123 1490

1.87 7 0.11 1.54 7 0.24 2.16 7 0.18 2.13 7 0.10 2.24 7 0.24 2.78 7 0.24

1.5 0.9 1.6 1.5 1.0 2.0

4.0 3.8 5.0 4.8 5.4 6.0

10.8 16.0 12.0 19.8 30.0 15.0

o 0.05g

60 (384) 56 (109) 68 (171) 55 (702) 51 (242) 66 (137) 1745

o 0.05

Age (years) 15–19 20–39 40–59 4 ¼ 60 UR Total

44 448 891 343 48 1774

2 25 50 19 3 100

36 385 778 267 24 1490

1.88 7 0.29 1.94 7 0.12 2.08 7 0.08 2.43 7 0.20 2.67 7 0.65

1.5 1.2 1.5 1.5 1.4

4.0 4.0 5.0 6.0 7.0

7.5 18.0 19.8 30.0 13.0


41 (44) 58 (445) 59 (881) 56 (340) 43 (35) 1745

o 0.05

Education oHigh School High School Collegeh Post-graduate UR Total

94 620 892 131 37 1774

5 35 50 7 2 100

77 532 763 100 18 1490

1.83 7 0.24 2.08 7 0.10 2.16 7 0.10 2.07 7 0.27 2.33 7 0.64

1.0 1.5 1.5 1.0 1.3

5.4 4.8 5.0 6.0 7.0

9.0 18.8 30.0 13.0 10.8


43 (94) 53 (611) 62 (887) 68 (130) 26 (23) 1745

o 0.05

Total household income o$25,000 230 $25,000–$44,999 284 $45,000–$84,999 516 Z$85,000 546 UR 198 Total 1774

13 16 29 31 11 100

185 242 437 482 144 1490

2.09 7 0.21 2.00 70.14 2.09 7 0.12 2.28 7 0.12 1.84 7 0.17

1.0 1.5 1.5 1.6 1.0

5.5 4.0 5.0 5.4 4.0

18.8 16.0 19.8 30.0 12.0


50 (224) 57 (284) 61 (514) 59 (540) 55 (183) 1745


58 (1745)

Key: License: H&L ¼ Hook and Line; B ¼Basic (Freshwater only); SW ¼ Saltwater; SP ¼ Sportsman’s Paradise; Snr ¼Senior Hunting and Fishing; L ¼Lifetime licenses. Ethnicity/Race: C ¼ Caucasian; AA ¼ African American; Oth¼ Other (includes Asian, Hispanic, Native American and other); UR ¼ Unreported. Location: SE ¼Southeast LA; SW ¼Southwest LA; N ¼North LA. a

Fish consumption statistics were calculated only from those who reported sportfish consumption (n¼1490). P-values indicate the significance of differences in mean sportfish consumption or advisory awareness among subgroups. Bold P-values are statistically significant. c Percent prevalence advisory awareness. d Chi-square test. e One respondent reported eating 100 total fish meals per month and was removed as an outlier. f T-test. g ANOVAs test were conducted on natural logarithm transformation. h College includes some college, technical/vocational, and college degree. b

outlier reported eating 100 sportfish meals per month). Two percent of respondents were identified as high exposure individuals, who ate over four sportfish meals per month, and had fished for over 30 years. Almost half of the high exposure individuals had a Saltwater fishing license (44%). Seven percent of African-Americans were among the high exposure individuals

compared to 2% of Caucasians. Sixty-eight or 4% of respondents were defined as at-risk individuals who were both unaware of advisories and ate sportfish over four times per month. Almost half (41%) of the at-risk individuals were Saltwater license holders. Nine percent of all African Americans (n ¼10) were at risk, compared to 4% of Caucasians (n ¼54).


6 18







Sportsman's Paradise

8 9


Senior Hunt/Fish




Saltwater 8

51 11




3 8





Percent of All Respondents



60 All Fish Commercial Fish Sportfish

50 40 30 20 10

0 0

Hook & Line


3-4 5-6 7-8 Meals per Month



Basic (freshwater only)

Fig. 1. Percent of respondents in license groups by area. Key: n¼ North; SE¼ Southeast; SW¼ Southwest.

Percent of Sensitive Population

Percent of Respondents in Each License Group by Area

A. Katner et al. / Environmental Research 111 (2011) 1037–1045

70 All Fish


Commercial Fish



40 30 20 10 0 0







Meals / Month Fig. 3. Percent frequency distributions of monthly fish consumption frequency for (A) all respondents and (B) female respondents of child-bearing age. Note: The category of ‘‘all fish’’ includes both commercial and sportfish.

Fig. 2. Box plot of years of fishing in Louisiana by population type. Notes: Populations evaluated included seniors (460 years of age; n¼ 297); the sensitive population (women of child-bearing age, between 18 and 45 years; n¼ 175); and the general population (all other respondents; n¼ 1074). Box plot values represent (from top to bottom): maximum, 75th percentile, median (at the line), 25th percentile and minimum.

Statistically significant differences in sportfish consumption were observed by gender and license type (Table 1). With respect to the type of fishing license, Sportsman’s Paradise fishers had the highest proportion of individuals consuming over four sportfish meals per month (17%); and significantly higher mean sportfish consumption frequency than Basic and Hook and Line license holders (ANOVA, p o0.05) (see Table 1 and Fig. 4). Holders of Saltwater licenses also had significantly higher mean sportfish consumption frequency than Hook and Line license holders (ANOVA, po0.05). Female respondents had a significantly lower mean sportfish consumption frequency than males (T-test, p o0.05). When females were restricted to women of childbearing age, the average total fish and sportfish meals per month were three and one, respectively (n¼189) (Fig. 3B). Only 5% of all women of child-bearing age reported eating over four meals of sportfish per month, with a range of 5–14 meals per month. Species-specific sportfish consumption. Fishers were asked to indicate the three most frequently eaten sportfish, and to estimate associated monthly consumption frequency (n¼1218)

Fig. 4. Box plot of monthly sportfish consumption frequency by recreational fishing license. Key: B¼ Basic (n¼ 369); HL¼ Hook and Line (n¼ 105); L¼ Lifetime (n¼168); SP¼ Sportsman’s Paradise (n¼ 132); S¼ Senior Hunting and Fishing (n¼ 226); SW¼ Saltwater (n¼ 665). Note: Box plot values represent (from top to bottom): maximum, 75th percentile, median (at the line), 25th percentile and minimum.

(see Fig. 5 and Supplementary Material Table 3). Overall the most frequently consumed sportfish are: drum (mostly red drum), speckled trout, and catfish, which were also the most popular species among women of child-bearing age. Eighty-one percent of all fish-tissue samples analyzed were comprised of these species, based on samples listed in LDEQ’s Fish Tissue Mercury Database (Katner et al., 2010). These species were also the most frequently caught and eaten finfish reported in previous surveys of coastal


A. Katner et al. / Environmental Research 111 (2011) 1037–1045

Number of All Respondents 0




Drum Speckled trout Catfish Bass

expressed a wish to be contacted for more information. Most fishers (73%) indicated they were aware of the health benefits of fish (n¼1715). Fifty-three percent of women of child-bearing age were aware of advisories (n ¼198); while 75% were aware of the health benefits of eating fish (n ¼192). Awareness rates varied by gender, ethnicity, geographic area, license type, age and education (w2 test, po0.05; Table 1); and were lowest among women (53%), African-Americans (43%), fishers from the southeast (50%), holders of Senior Hunting and Fishing licenses (51%), individuals of age 15–19 (41%), and individuals with less than a high school education (43%). Due to the low sample size (n o15) for Asians and Hispanics, no conclusions could be reached about their advisory awareness.

Crappie <1 m/m Bream 1-2 m/m Flounder 3-4 m/m Snapper >4 m/m Tuna Croaker

Unknown / Forgot

Fig. 5. Frequency distribution of monthly sportfish consumption (meals per month or m/m) for the most frequently eaten fish by all respondents. Notes: Saltwater species were grouped as follows: Drum includes red and black drum. Snapper includes red and other snappers. Tuna includes blackfin, yellowfin, albacore and other tunas. Freshwater species were grouped as follows: Catfish includes flathead, channel and blue catfish. Bass includes largemouth, spotted, white and striped bass. Crappie includes white and black crappie. Bream includes bluegill and red ear sunfish.

LA recreational fishers (U.S. DOE, 1997; and Lincoln et al., 2010); and residents in Orleans Parish, LA (n ¼405; Anderson and Rice, 1993). Only 5% of all respondents ate more than four meals per month of a specific species or fish group. Almost 3% (2.5%) of all respondents ate more than four meals per month of speckled trout; 2% ate more than four meals per month of each of catfish, drum (red drum mostly), bass and crappie (mostly white crappie); and less than 1% ate more than four meals of bream per month (mostly bluegill and red ear sunfish). There were differences among the various locations and ethnicities in terms of the ranking of favorite species, but not in the specific species selected. In the south, the most frequently eaten sportfish were drum (primarily red drum) and speckled trout. In the north, the most frequently eaten species were catfish and crappie. The most frequently eaten by ethnicity were as follows: (1) for Caucasians—drum and speckled trout; (2) for Native Americans—catfish and crappie; and (3) for AfricanAmericans—catfish and drum. The most frequently eaten sportfish for Asians and Hispanics were catfish and drum; and drum and speckled trout, respectively; however, there was a low number of respondents (n o20) in these two ethnic groups.

3.4.2. Mode of awareness Respondents who were aware of advisories were asked to select all ways in which they became aware (n¼1028). Modes of awareness ranked as follows: (1) 53% of respondents became aware of advisories through newspapers or magazines; (2) television (48%), (3) family or friends (30%), (4) brochures or fishing regulation booklets (30%), (5) signs at bait shop, landing, boat launch and fishing sites (20%), and (6) radio (16%). Less than 10% of respondents have seen, read or heard about the warnings by means of any other methods, which include, but were not limited to, internet, library, doctor, clinic, school and billboard. When respondents were asked to select all the best ways to reach them (n¼1675), the most frequently selected methods by both all respondents and respondents unaware of advisories, were television, newspapers/magazines, mail outs and notification at the time of license purchase (Supplementary Material Fig. 2). 3.4.3. Effect of advisory on behavior When asked if they had changed their behavior because of advisories, only 30% of 1040 respondents reported that they did. These respondents were asked to select all ways in which their behavior changed (n¼1020). Thirty-three percent of the respondents who had changed their behavior reported that they stopped eating all fish from water bodies where health warnings have been issued. This is followed by 27% who said that they ate less fish from water bodies with health warnings; 25% who said that they ate more fish bought from a store or vendor; and 24% who said they stopped eating certain types of fish from waterbodies under advisory. About 14% of respondents switched to another location where there are no health warnings, while 11% practiced catch and release. These patterns were similar across license categories. Respondents who did not change their fish consumption habits (n ¼733) were asked to select all reasons why their fish consumption habits did not change (n ¼693) The reason provided by most respondents (44%) was there was no advisory where they fished; followed by 29% who do not usually eat fish from waterbodies with warnings; and 27% who do not eat enough fish for the warnings to apply to them. Fifteen percent of respondents who were aware of advisories and did not change their fish consumption behavior reported that ‘‘people have been eating here forever and they’re not sick’’; while over 10% reported they thought the health benefits exceeded the risks. Less than 5% did not think the advisories were accurate and less than 5% reported that they needed to fish to feed their families.

3.4. Advisory awareness and effect 4. Discussion 3.4.1. Awareness of advisories and health benefits Survey results indicated that over half (58%) of 1745 respondents had seen, read or heard about fish consumption warnings for LA (Table 1). Sixteen percent of 205 respondents who made reportable comments indicated a need for improvements in education on health warnings and recreational fishing; and 6%

Our results support previous studies in the conclusion that recreational fishers from LA represent an exposed subpopulation with a reliance on local seafood for sustenance, either as a necessity or choice (Lincoln et al., 2010; Belanger et al., 2000; Dellenbarger et al., 1993). A high potential for exposure, results

A. Katner et al. / Environmental Research 111 (2011) 1037–1045

from not only the importance of seafood to the LA culture and economy, but from the fact that over one quarter of the most popular waterbodies in LA are under advisory. These facts justify the need to ensure that monitoring activities, and advisories which are typically based on default exposure assumptions, are protective of this exposure group. To ensure protection of individuals without deterring healthy fishing and fish consumption activities, monitoring, advisory development and outreach are evaluated and discussed; and recommendations are made to optimize program activities. 4.1. Monitoring Monitoring resources were appropriately targeted by LA’s Fish Advisory Program to the most popular fishing locations, and most frequently eaten high mercury species (Katner et al., 2010). However, only 18% of LA’s fish samples were collected from saltwater or brackish areas, yet over half of all sportfish meals were for saltwater or brackish fish. Given the fact that approximately half of the potentially high-exposure population resides near, or fishes in, the Gulf; more fish samples in brackish and saltwater areas is encouraged. Louisiana recently developed a long-term monitoring plan (LA Dept. of Health and Hospitals, 2010) of brackish and saltwater areas affected by the BP oil spill. Species identified in the plan for monitoring were evaluated against the most frequently eaten species identified here. Species identified in the plan to be monitored include, but are not limited to: black and red drum, cobia, croaker, greater amberjack, grouper, Gulf menhaden, king mackerel, red snapper, sheepshead, southern flounder, speckled trout, striped mullet and tuna. Based on our estimates of species-specific consumption frequencies by recreational fishers, this monitoring plan covers all of the most frequently eaten species. Given the higher consumption frequencies of speckled trout, red drum, flounder, snapper, sheepshead, tuna and croaker, evaluation of these species is particularly critical. 4.2. Louisiana’s advisory development protocol LA currently assumes that individuals do not eat more than four area- and species-specific fish meals per month. This fish meal frequency may be an acceptable assumption for adult finfish consumption, in light of the fact that only 5% of all respondents ate more than four recreationally caught species-specific finfish meals per month (the most popular being speckled trout, catfish, red drum, bass, white crappie, and bluegill and red ear sunfish). However, given the considerable variation in estimates of sportfish consumption by location and species (Supplementary Material Table 3 and Fig. 3), one default consumption rate may not adequately reflect exposure. The use of species-specific fish consumption rates will increase the accuracy of exposure estimates upon which advisories are based; thereby increasing the relevance of fish consumption advisories. One case in point is bowfin- the majority of fish consumption advisories in LA restrict bowfin consumption (88%), a high trophic level species used to screen waterbodies (Katner et al., 2010). However, this species is infrequently consumed, with o1% of the respondents reporting it as one of their three most frequently consumed fish (Supplementary Material Table 3). Estimates of the contribution of bowfin consumption on mercury exposure were substantially lower than those for speckled trout, snapper and southern flounder (Lincoln et al., 2010)—species which are frequently eaten, but infrequently monitored and not under advisory (Katner et al., 2010). If bowfin consumption were considered during the advisory process, 15% of LA’s mercury-based fish advisories (i.e., those based on bowfin alone), may be found unnecessary (n ¼7), assuming that Program brochures with information about the generally high levels of


mercury in bowfin are provided to fishers when they apply for licenses (LDHH, 2009). It also follows that if species-specific consumption rates had a greater influence on monitoring and advisory development activities, many areas and species not frequently monitored or currently under advisory may require re-evaluation. This includes low mercury species that are frequently eaten, and low mercury areas that are popular fishing locations. Results also indicate that statewide species-specific consumption frequencies may not be an adequate representation of fish consumption for fishers from different areas of the state (Supplementary Material Table 3 and Fig. 3). The use of statewide rates for a coastal advisory may underestimate saltwater fish consumption for coastal populations. Due to this variability, it is strongly recommended that if advisories are to be location- and species-specific, then monitoring and advisory decisions should be based on regional species-specific consumption rates. Given the observation that the most popular species identified here were largely consistent across sub-populations and with other studies, it may be feasible to apply regional levels of fish consumption during the advisory process to redirect monitoring activities and re-evaluate advisory decisions. One suggestion is to adjust the recommended meal limit that triggers an advisory to one appropriate to each species (e.g., the 90th percentile speciesspecific consumption rate for the total population). LA uses a default of 4 meals per month meal limit to trigger an advisory. If the mercury levels in a species result in a recommended meal limit of 4 meals per month or more, no advisory is triggered. Meal limits of three or less trigger an advisory or further monitoring. This value may be raised in the case of frequently eaten species, such as shellfish and catfish, to an amount representing statewide or location-specific consumption. To illustrate this approach, if meal limits were based on the 90th percentile of area- and speciesspecific consumption rates (where sufficient data exist) the meal limit which would trigger an advisory could be raised to 6 meals per month for bass in the north (based on a 90th percentile consumption rate of 7 meals per month) (Supplementary Material Table 3). This approach would require more monitoring and advisories for frequently consumed low mercury high species, and may reduce the need for sometimes unnecessary advisories on infrequently consumed high mercury species. Species with high mercury that are infrequently eaten could be listed on advisory material with general meal limit recommendations. If this approach is taken more accurate area- and populationspecific fish consumption data are needed. Fish consumption surveys are generally not used for the advisory process. Since exposure is a key element in the estimation of risk, consumption surveys should be conducted. To enable characterization of a broader population, future surveys should be distributed to commercial and subsistence fishers in multiple languages; and include questions about meal size, children’s consumption patterns, individual body weight, and consumption of commercial seafood. Information about shellfish consumption is also needed. The authors of the Lincoln study estimated that 8% of mercury intake in LA’s coastal fishers came from eating recreationally caught shellfish (Lincoln et al., 2010). To keep costs down, a brief one-page survey could be distributed to fishers when they apply for a fishing license. Permit costs can be reduced as an incentive to provide survey information. Despite the uncertainties inherent in the exposure assessment process, all individuals should be provided with sufficient information to help them decide if and how they should limit their intake. To address this issue, a web-based tool that enables users to evaluate personal risks and receive individualized meal limit recommendations is proposed. Some species may not be under advisory because it is assumed that they are eaten less than five times per month—yet they could pose a risk to individuals


A. Katner et al. / Environmental Research 111 (2011) 1037–1045

consuming them in excess of this frequency. This tool would apply the equations used to develop advisories to available fish tissue contaminant data and user-provided fish consumption information. The user would be able to select the species and waterbody they are interested in from a drop down menu of areas and associated species sampled. This would provide the public with a way to quickly identify the waterbodies, species and contaminants which have been monitored. One assumption that fishers make is if it is not under advisory, it is safe. However not all waterbodies have been tested and even if they have been tested in most cases it has been for mercury alone. The user would then be asked to provide information such as their body weight; gender and age; average meal size; amount of time they have consumed that species from the waterbody in question; and their species-specific fish consumption frequency. They can also choose the summary contaminant level statistic most preferable to them; and a level of acceptable risk, where applicable. Information can be provided to explain the options, highlight the default values, and provide recommendations on healthy fish selection and preparation. A suggestion box can be provided for those who wish to recommend for sampling, the species, waterbodies and contaminants of concern to them to help guide monitoring activities to popular areas and associated species. This tool would provide users with a quick method for identifying a safe, and more realistic, consumption amount tailored to their needs; thereby reducing unnecessary fear and avoidance. It would also help individuals understand the assumptions made, and uncertainty involved, in determining meal limit recommendations; and would provide state officials with more realistic exposure data. The biggest limitation of this tool is the lack of Internet access for rural fishing populations who would need it the most. One possible solution is to provide residents with a tollfree number they can call to receive results by phone. 4.3. Outreach and risk communication Our results indicate that over half of the respondents had seen, read or heard about fish consumption warnings, a result similar to other studies (Burger and Waishwell, 2001; Imm et al., 2005). While mass media options (e.g., television, newspapers and magazines) and mail-outs were the most effective and preferred methods for receiving advisory information, limited resources may impede this option. The next most popular methods for receiving advisory information were to receive information at the time of license application, or via signs and fishing regulation booklets. While signs are generally posted at sites under advisory, only 20% of respondents became aware of advisories via signs at bait shops, landings, boat launches and fishing sites. Many respondents complained that they did not see signs along waterbodies under advisory. It is already known that there are problems with signs being shot down or removed, perhaps because of their potential effect on property values or tourism. Also, in some areas, it is not possible to put signs in every possible location where water can be accessed. With respect to fishing regulation booklets, only 30% of respondents became aware of advisories via the booklet which is supposed to be provided at the time of the license application. This finding suggests that there may be problems with the booklet’s availability, distribution or readability. However, receiving information at the time of license application was still preferred. One alternative is to include a number, email and website address for advisory information on the back of the fishing license receipts when issued. These receipts, must be on hand at all times, thus are less likely to be lost. Mass media should be used when resources are available. This is particularly important in LA, as an estimated 80% of LA’s commercial seafood comes from local sources (US Department of Energy, 1997). When resources

are limited, targeted outreach to the most exposed and most susceptible population is encouraged; particularly, prior to or during the months of March to May, as this is both the most popular time for fishing, as well as the state’s seafood festival season. A special effort should be made to increase awareness in coastal populations, given the low advisory awareness and high Gulf seafood consumption along the coast; as well as the fact that most fishing licenses are sold in coastal parishes (LDWF, 1990, 1991). Of particular concern, is the fact that about half of the women of child-bearing age in our survey (47%) were unaware of advisories. To address this issue, outreach and informational material should be provided to health clinics, pediatricians and gynecologists (Burger, 2005). Other groups to which outreach can be targeted include those that have high exposures or are unaware of advisories- this includes African-Americans and saltwater fishers. 4.4. General study limitations This study focused on the consumption of recreationally caught finfish, neglecting consumption of recreationally caught shellfish and commercial fin – and shellfish – the knowledge of which is essential for estimating total exposure and evaluating potential risk. In addition, the recommendations presented here are based on a single contaminant- results may differ if other contaminants are considered. As such, it is suggested that local seafood be evaluated for other contaminants—a combination of indicators will provide a better assessment of seafood health. Also, as children and commercial or subsistence fishers were not surveyed, conclusions presented here are only applicable to adult recreational fishers, who are expected to eat less fish than commercial and subsistence fishers. This survey may also underrepresent non-license holders, which can account for as much as 25% of fishers (U.S. EPA, 1998). This could include on-reservation Native American fishers; those who fish without licenses; migrant workers and homeless people. Some of these groups might contain a disproportionately high number of subsistence fishers and thus might be groups at higher overall risk (U.S. EPA, 1998). Given the inadequate representation of ethnic minorities we were not able to appropriately evaluate ethnic differences in species-specific sportfish consumption, but we suspect differences do exist based on previous findings in the literature (Burger et al., 1999). As this survey was only distributed in English, respondents with limited or no English were under-represented as well. An estimated 328,041 (8.21%) of Louisiana residents, five years and older, speak a language other than English at home (U.S. Census Bureau, 2010). Similarly, 9% of our respondents reported a primary language other than English. This may explain the low number of surveys returned by ethnic groups, such as the Vietnamese, for which there is a large proportion of non-English speaking subsistence and commercial fishers. According to the U.S. Census information, 2.2% of nonEnglish speakers in LA, 5 years and older, are Vietnamese with approximately 31.7% of the 2.2% speaking English ‘‘less than well.’’

5. Conclusions This study presents the results of the first known populationbased survey of LA recreational fishers. Based on the results reported here, LA recreational fishers represent an exposed subpopulation with a strong reliance on local seafood. Though monitoring was appropriately targeted to the most popular waterbodies and most frequently eaten high mercury species, reliance on default exposure scenarios could lead to flawed monitoring regimes and advisories. To ensure protection of atrisk individuals without deterring healthy fishing and fish consumption activities, more realistic advisories should be created by

A. Katner et al. / Environmental Research 111 (2011) 1037–1045

basing monitoring and advisory decisions on regional speciesspecific sportfish consumption levels; not just on contaminant levels alone. For example, meal limits could be based on the 90th percentile of area- and species-specific consumption rates, where sufficient data exist. If this approach is taken, frequently consumed species which are infrequently monitored due to low mercury levels would require more in-depth sampling (i.e., shellfish, and frequently eaten estuarine/marine species such as speckled trout and red drum). It will also be critical to obtain more realistic species-specific fish consumption data for different areas and populations. To enable characterization of a broader population, future surveys should be distributed to commercial and subsistence fishers in multiple languages; and include questions about meal size, children’s consumption patterns, individual body weight, and all finfish, shellfish and commercial seafood consumption. A brief one-page survey could be distributed to fishers when they apply for a fishing license, and discounts offered as an incentive to provide information. It is safe to assume that advisories alone will have little effect on reducing contaminant exposures, unless they are known and understood. While results indicate that over half of the respondents had seen, read or heard about fish consumption warnings, similar to other fishing groups, continued outreach is hindered by limited resources. One solution to this, and to limitations inherent in other traditional outreach methods, is to include a number, email and website address for advisory information on the back of the fishing license receipts, which must be on hand at all times. However, given the fact that the majority of commercial seafood in LA is locally derived, this approach may not be sufficient for reaching the wider population. Populations of potential concern who may benefit from targeted outreach include women of child-bearing age, coastal fishing communities and African-Americans.

Acknowledgments The authors wish to thank Chris Piehler from the LA Department of Environmental Quality for providing access to the state’s fish tissue mercury database; Dr. Wei Chen from Tulane University for generously agreeing to review this paper and providing guidance on statistical methods; and Drs. Richard Ambrose, Deborah Glik and Jennifer Jay from the University of California, Los Angeles; Dr. Raoult Ratard from the Louisiana Office of Public Health; and Dr. Rebecca Lincoln from Harvard School of Public Health, for their valuable suggestions.

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.envres.2011.08.001.

References Anderson, A.C., Rice, J.C., 1993. Survey of fish and shellfish consumption by residents of the Greater New Orleans area. Bull. Environ. Contamin. Toxicol. 51, 508–514.


Belanger, T.M., Caesar, E.M., Trachtman, L., 2000. Blood mercury levels and fish consumption in Louisiana. J. LA State Med. Soc. 152 (2), 64–73. Burger, J., Pflugh, K.K., Lurig, L., Von Hagen, L.A., Von Hagen, S., 1999. Fishing in urban New Jersey: ethnicity affects information sources, perception, and compliance. Risk Analysis 19, 217–229. Burger, J., Waishwell, L., 2001. Are we reaching the target audience: evaluation of a fish fact sheet. Sci. Tot. Environ. 277, 77–86. Burger, J., 2005. Fishing, fish consumption and knowledge about advisories in college students and others in central New Jersey. Environ. Res. 95, 268–275. Dellenbarger, L., Schupp, A., Kanjilal, B., 1993. Seafood consumption in coastal Louisiana. Louisiana Department of Environmental Quality, Baton Rouge, LA. Imm, P., Knobeloch, L., Anderson, H., 2005. Fish consumption and advisory awareness in the Great Lakes Basin. Environ. Health Perspect 113 (10), 1325–1329. Katner, A., Sun, M.H., Suffet, M., 2010. An evaluation of mercury levels in Louisiana fish: Trends and public health issues. Sci. Tot. Environ. 408, 5707–5714. LDHH (LA Department of Health and Hospitals), 2010. Louisiana seafood safety response and quality certification plan: Post-Mississippi Canyon 252 oil spill (Version 3). LA Department of Health and Hospitals, LA Dept. of Environmental Quality, LA Dept. of Wildlife and Fisheries, LA Dept. of Agriculture and Fisheries, LA Economic Development. Baton Rouge, LA. (Available: /http://, accessed December 7, 2010). LDHH, 2009. Mercury in fish: What everyone should know about eating locally caught fish. LDHH, Baton Rouge, LA Available: / viewer?url= MrcryNFsh_Rcrtnl_Eng.pdfS, (accessed February 7, 2011). LDWF (LA Department of Wildlife and Fisheries), 1990. Sport license sales 1989–1990. LDWF, Baton Rouge, LA. LDWF, 1991. Commercial license recap report for 1991: Parish breakdown. LDWF, Baton Rouge, LA. LDWF, 2009. The 2008 Louisiana recreational fisherman and health advisory survey report. LDWF, Baton Rouge, LA Available: /http://www.wlf.louisiana. gov/sites/default/files/pdf/publication/32724-2008-louisiana-recreationalfishermen-and-health-advisory-survey-report/survey_report.pdfS, (accessed June 12, 2009). Lincoln, R.A., Shine, J.S., Chesney, E.J., Vorhees, D.J., Grandjean, P., Senn, D.B., 2010. Fish consumption and mercury exposure among Louisiana recreational fishers. Environ. Health Perspect. Available: /, accessed 11/16/2010 (accessed December 2, 2010). Mergler, D., Anderson, H., Chan, L., Mahaffey, K., Murray, M., Sakamoto, M., Stern, A., 2007. Methylmercury exposure and health effects in humans: a worldwide concern. Ambio 36 (1), 3–11. Rice, G., Swartout, J., Mahaffey, K., et al., 2000. Derivation of US EPA’s oral Reference Dose (RfD) for methylmercury. Drug Chem Toxicol. 23 (1), 41–54. Sunderland, E.M., 2007. Mercury exposure from domestic and imported estuarine and marine fish in the U.S. seafood market. Environ. Health Perspect 115 (2), 235–242. U.S. Census Bureau, 2005–2009 American Community Survey Five Year Estimates. April, 2010. Available: / (accessed February 1, 2010). U.S. Department of Energy (DOE), 1997. Synthesis of seafood catch, distribution and consumption patterns in the Gulf of Mexico region. National Petroleum Office, US DOE, Tulsa, OK Available: / cover.jsp?purl=/9728-4GtY3X/webviewable/S(accessed December 9, 2010). U.S. Environmental Protection Agency (EPA), 1998. Guidance for conducting fish and wildlife consumption surveys. US EPA Office of Water, Washington DC EPA-823-B-007. U.S. EPA, 2000. Guidance for assessing chemical contaminant data for use in fish advisories, Vol. 1–3. US EPA, Office of Water, Washington DC. U.S. EPA, 2002. Estimated per capita fish consumption in the United States: 2002. US EPA, Washington DC Available: / files/consumption_report.pdfS, (accessed April 6, 2011). U.S. EPA. 2004 Fish advisories: What you need to know about mercury in fish and shellfish. Available: /, (accessed December 5, 2010).