In the three communities combined, 101 participants provided hair samples and diet data (42 from Aklavik, NT; 32 from Old Crow, YT; and 24 from Fort McPherson, NT). The mean age was 52 (SD: 15.7; Range: 10-86) years. Participants were nearly all Indigenous, predominantly identifying as either Gwich’in (60%; 60/101) or Inuvialuit (30%; 30/101). A small proportion of participants were of European descent (6%; 6/101) but had been residing in the community for at least 5 years. The study population was disproportionately female (63%; 64/101), none pregnant or breastfeeding at the time of data collection. Assuming an average growth rate of 1 cm/month, the exposure periods represented in the collected hair samples ranged from approximately 3 weeks to almost 9 years (median: 1.1 year; IQR: 2.1 years).
Almost all participants (96%; 97/101) reported eating fish or marine mammals in the past 12 months. The data obtained from the fish-focussed FFQ was consistent with input from community planning committees, which identified the summer as the main season during which community members consume fish/whale. However, there was considerable variation by species and community (Figure 1). Table 1 shows the 17 aquatic species participants reported consuming in the previous 12 months. The fish species consumed by the largest proportion of participants was Broad Whitefish (C.nasus) (83%), followed by Inconnu (S.nelma) (42%) and Dolly Varden (S.malma) (33%). A large proportion of participants also ate Beluga Whale (D.leucas) (42%), with 71% of those who reported eating Beluga Whale in the past 12 months were from Aklavik, NT (30/42), the community with the largest proportion of Inuit residents.
Table 1: Fish and marine mammal species consumed at least once in the past 12 months by community, 101 western Canadian Arctic residents, 2016
Species
|
|
Proportion that Consumed Each Species
in the Past 12 Months
|
Scientific Name
|
Common Name
|
|
Aklavik (n=45)
|
|
Old Crow (n=32)
|
|
Fort McPherson (n=24)
|
n
|
%
|
|
n
|
%
|
|
n
|
%
|
Salmonidae Family
|
Salvelinus aplinus
|
Arctic Char
|
11
|
24
|
|
3
|
9
|
|
1
|
4
|
Salvelinus malma
|
Dolly Varden
|
30
|
67
|
|
0
|
0
|
|
3
|
13
|
Salvelinus namaycush
|
Lake Trout
|
1
|
2
|
|
0
|
0
|
|
5
|
21
|
Coregonus nasus
|
Broad Whitefish
|
36
|
80
|
|
26
|
81
|
|
22
|
92
|
Coregonus clupeaformis
|
Lake Whitefish
|
2
|
4
|
|
5
|
16
|
|
0
|
0
|
Coregonus autumnalis
|
Arctic Cisco
|
18
|
40
|
|
0
|
0
|
|
1
|
4
|
Oncorhynchus tshawytscha
|
Chinook Salmon
|
6
|
13
|
|
25
|
78
|
|
1
|
4
|
Oncorhynchus keta
|
Chum Salmon
|
1
|
2
|
|
7
|
22
|
|
3
|
13
|
Oncorhynchus kisutch
|
Coho Salmon
|
3
|
7
|
|
5
|
16
|
|
0
|
0
|
Oncorhynchus nerka
|
Sockeye Salmon
|
0
|
0
|
|
4
|
13
|
|
1
|
4
|
Oncorhynchus gorbuscha
|
Pink Salmon
|
2
|
4
|
|
0
|
0
|
|
2
|
8
|
Thymallus arcticas
|
Arctic Grayling
|
0
|
0
|
|
9
|
28
|
|
0
|
0
|
Stenodus nelma
|
Inconnu
|
24
|
53
|
|
1
|
3
|
|
17
|
71
|
Lotidae Family
|
Lota Lota
|
Burbot
|
12
|
27
|
|
7
|
22
|
|
10
|
42
|
Osmeridae Family
|
Thaleichthys pacificus
|
Eulachon
|
0
|
0
|
|
1
|
3
|
|
0
|
0
|
Percidae Family
|
Sander vitreus
|
Walleye
|
1
|
2
|
|
0
|
0
|
|
0
|
0
|
Monodontidae Family
|
Delphinapterus leucas
|
Beluga Whale
|
30
|
67
|
|
8
|
25
|
|
4
|
17
|
Table 2 shows the five most frequent species consumed ≥ 1 time/week by community and season. The mean number of different species eaten by participants was 3.5 (SD: 1.9; Range: 0-9). The main waterways and sites from which participants reported harvesting fish and whale are shown in Figure 2. On average, participants reported harvesting most of the species they consume from local rivers, followed by the ocean and nearby lakes. The mean proportions of harvesting sites for reported species consumed were: rivers 66.7% (SD: 32.9%; Range: 0-100); the ocean 21.7% (SD: 27.4; Range: 0-100); and lakes 1.8% (SD: 8.2; Range: 0-50). The mean proportion of consumed species purchased from the store was 2.0% (SD: 7.6; Range: 0-33).
Table 2: Five most frequent aquatic species consumed at least once per week by season and community, 101 western Canadian Arctic residents, 2016
Season
|
|
Aklavik, NT
(n=45)
|
|
Old Crow, YT
(n=32)
|
|
Fort McPherson, NT
(n=24)
|
Species
|
n
|
%
|
|
Species
|
n
|
%
|
|
Species
|
n
|
%
|
Winter
|
|
S. nelma
C. nasus
D. leucas
S. malma
C. autumnalis
|
9
8
7
5
4
|
20
18
16
11
9
|
|
C. nasus
O. tshawytscha
C. clupeaformis
O. kisutch
T. arcticas
|
8
6
3
2
2
|
25
19
9
6
6
|
|
C. nasus
S. nelma
|
9
5
|
38
21
|
Spring
|
|
S. nelma
D. leucas
C. nasus
S. malma
C. autumnalis
|
8
8
7
5
4
|
18
18
16
11
9
|
|
C. nasus
O. tshawytscha
C. clupeaformis
O. kisutch
T. arcticas
|
8
7
4
2
2
|
25
22
13
6
6
|
|
C. nasus
S. nelma
|
11
6
|
46
25
|
Summer
|
|
D. leucas
S. malma
C. nasus
C. autumnalis
S. nelma
|
17
14
14
13
10
|
38
31
31
29
22
|
|
O. tshawytscha
C. nasus
O. keta
S. aplinus
O. kisutch
T. arcticas
|
20
9
3
2
2
2
|
63
28
9
6
6
6
|
|
C. nasus
S. nelma
S. aplinus
D. leucas
|
14
10
1
1
|
58
42
4
4
|
Fall
|
|
C. nasus
S. nelma
D. leucas
S. malma
C. autumnalis
|
9
8
6
6
4
|
20
18
13
13
9
|
|
C. nasus
O. tshawytscha
O. keta
L. Lota
C. clupeaformis
T. arcticas
|
9
7
5
4
3
3
|
28
22
16
13
9
9
|
|
C. nasus
S. nelma
L. Lota
S. aplinus
S. malma
|
15
8
5
1
1
|
63
33
21
4
4
|
Among participants from all communities combined, the mean concentration of MeHg in hair samples was 0.60 μg/g (SD: 0.47; Range: 0.059-2.07). This varied slightly across communities, with mean values from Aklavik, NT, Old Crow, YT and Fort McPherson, NT of 0.51 μg/g (SD: 0.44; Range: 0.06-2.07), 0.54 μg/g (SD: 0.35; Range: 0.11-1.51) and 0.84 μg/g (SD: 0.58; Range: 0.06-1.90), respectively. The distributions of MeHg in hair samples across the entire study population and stratified by community are shown in Figure 3. Mean hair mercury levels (μg/g) ± SD stratified by population characteristics are shown in Tables 3 and 4. No participants had hair mercury levels that exceeded the exposure maximum defined by Health Canada.
Table 3: Distribution of demographic characteristics and stratum-specific mean MeHg concentrations (µg/g) by community, 101 western Canadian Arctic residents, 2016
Demographic Characteristics
|
Total
(n=101)
|
|
Aklavik, NT
(n=45)
|
|
Old Crow, YT
(n=32)
|
|
Fort McPherson, NT (n=24)
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
Age
≤ 30 years
|
9 (9)
|
0.258 ± 0.205
|
|
5 (11)
|
0.167 ± 0.080
|
|
4 (12.5)
|
0.372 ± 0.269
|
|
0 (0)
|
-
|
31-40 years
|
15 (15)
|
0.449 ± 0.600
|
|
8 (18)
|
0.504 ± 0.695
|
|
4 (12.5)
|
0.526 ± 0.661
|
|
3 (12.5)
|
0.196 ± 0.233
|
41-50 years
|
14 (14)
|
0.486 ± 0.341
|
|
8 (18)
|
0.414 ± 0.348
|
|
4 (12.5)
|
0.579 ± 0.354
|
|
2 (8)
|
0.586 ± 0.441
|
52-60 years
|
35 (35)
|
0.791 ± 0.515
|
|
14 (31)
|
0.603 ± 0.345
|
|
7 (22)
|
0.571 ± 0.441
|
|
14 (58)
|
1.089 ± 0.570
|
61-70 years
|
17 (17)
|
0.574 ± 0.307
|
|
6 (13)
|
0.781 ± 0.427
|
|
9 (28)
|
0.469 ± 0.156
|
|
2 (8)
|
0.424 ± 0.004
|
≥ 71 years
|
11 (11)
|
0.641 ± 0.391
|
|
4 (9)
|
0.421 ± 0.357
|
|
4 (12.5)
|
0.777 ± 0.265
|
|
3 (12.5)
|
0.752 ± 0.561
|
Sex
Male
|
37 (37)
|
0.742 ± 0.508
|
|
13 (29)
|
0.529 ± 0.462
|
|
17 (53)
|
0.681 ± 0.403
|
|
7 (29)
|
1.282 ± 0.490
|
Female
|
64 (63)
|
0.514 ± 0.424
|
|
32 (71)
|
0.504 ± 0.432
|
|
15 (47)
|
0.377 ± 0.196
|
|
17 (71)
|
0.655 ± 0.523
|
Table 4: Distribution of permanent hair treatment use and stratum-specific MeHg concentrations (µg/g) by community, 101 western Canadian Arctic residents, 2016
Hair Treatments
|
Total
(n=101)
|
|
Aklavik, NT
(n=45)
|
|
Old Crow, YT
(n=32)
|
|
Fort McPherson, NT (n=24)
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
Dyed
No
|
74 (73)
|
0.630 ± 0.479
|
|
33 (73)
|
0.483 ± 0.372
|
|
20 (62.5)
|
0.624 ± 0.400
|
|
21 (87.5)
|
0.869 ± 0.607
|
Yes
|
27 (27)
|
0.507 ± 0.429
|
|
12 (27)
|
0.589 ± 0.590
|
|
12 (37.5)
|
0.397 ± 0.205
|
|
3 (12.5)
|
0.621 ± 0.341
|
Perm
No
|
95 (94)
|
0.605 ± 0.476
|
|
45 (100)
|
0.511 ± 0.436
|
|
31 (97)
|
0.547 ± 0.356
|
|
19 (79)
|
0.923 ± 0.611
|
Yes
|
6 (6)
|
0.478 ± 0.298
|
|
0 (0)
|
-
|
|
1 (3)
|
0.275
|
|
5 (21)
|
0.518 ± 0.314
|
Dye or Perm
No
|
70 (69)
|
0.636 ± 0.486
|
|
33 (73)
|
0.483 ± 0.372
|
|
20 (62.5)
|
0.624 ± 0.399
|
|
17 (71)
|
0.947 ± 0.635
|
Yes
|
31 (31)
|
0.512 ± 0.417
|
|
12 (27)
|
0.589 ± 0.590
|
|
12 (37.5)
|
0.397 ± 0.205
|
|
7 (29)
|
0.575 ± 0.323
|
Because, data were insufficient for estimation of species-specific effects on hair mercury levels, this analysis was limited to effects of total fish/whale consumption. Table 5 shows MeHg levels stratified by total fish/whale consumption frequency across seasons. The strong correlations between season-specific fish/whale consumption variables prohibited including them in the same model, so season-specific effects were estimated in separate models. Model building procedures yielded the same set of adjustment variables for fish/whale consumption in each season: sex, hair length, use of hair dye or other permanent hair treatments, and the proportion of fish/whale meals usually prepared by cooking. There was no evidence of statistical interaction between fish/whale consumption and any of thed model covariates. Visual inspection of the lowess plots representing the locally weighted regression of MeHg concentration on exposure variables for each season showed that the relationships were not sufficiently linear to justify modeling exposures as continuous variables.
Table 5: Hair MeHg concentrations (µg/g) within categories of fish/whale consumption by season and community, 101 western Canadian Arctic residents, 2016
Total Fish/Whale Consumption
|
|
Total
(N=101)
|
|
Aklavik, NWT
(N=45)
|
|
Old Crow, YT
(N=32)
|
|
Fort McPherson, NWT (N=24)
|
Season
|
Intake Category
|
|
n
|
Mean ± SD
|
|
n
|
Mean ± SD
|
|
n
|
Mean ± SD
|
|
n
|
Mean ± SD
|
|
Winter
|
< 1 meal/week
|
|
55
|
0.512 ± 0.463
|
|
23
|
0.408 ± 0.387
|
|
17
|
0.398 ± 0.238
|
|
15
|
0.801 ± 0.634
|
|
1-2 meals/week
|
|
27
|
0.613 ± 0.380
|
|
13
|
0.571 ± 0.350
|
|
9
|
0.669 ± 0.424
|
|
5
|
0.624 ± 0.448
|
|
≥ 3 meals/week
|
|
19
|
0.822 ± 0.535
|
|
9
|
0.689 ± 0.613
|
|
6
|
0.741 ± 0.398
|
|
4
|
1.244 ± 0.382
|
|
Spring
|
< 1 meal/week
|
|
51
|
0.409 ± 0.343
|
|
23
|
0.343 ± 0.301
|
|
16
|
0.373 ± 0.240
|
|
12
|
0.585 ± 0.481
|
|
1-2 meals/week
|
|
28
|
0.708 ± 0.455
|
|
12
|
0.568 ± 0.365
|
|
10
|
0.701 ± 0.412
|
|
6
|
0.999 ± 0.609
|
|
≥ 3 meals/week
|
|
22
|
0.893 ± 0.547
|
|
10
|
0.829 ± 0.595
|
|
6
|
0.708 ± 0.349
|
|
6
|
1.184 ± 0.588
|
|
Summer
|
< 1 meal/week
|
|
28
|
0.443 ± 0.381
|
|
11
|
0.373 ± 0.326
|
|
8
|
0.413 ± 0.279
|
|
9
|
0.554 ± 0.517
|
|
1-2 meals/week
|
|
24
|
0.524 ± 0.493
|
|
11
|
0.375 ± 0.375
|
|
9
|
0.341 ± 0.215
|
|
4
|
1.343 ± 0.444
|
|
3-4 meals/week
|
|
18
|
0.616 ± 0.451
|
|
9
|
0.600 ± 0.437
|
|
4
|
0.423 ± 0.156
|
|
5
|
0.799 ± 0.620
|
|
≥ 5 meals/week
|
|
31
|
0.784 ± 0.482
|
|
14
|
0.669 ± 0.521
|
|
11
|
0.834 ± 0.374
|
|
6
|
0.961 ± 0.577
|
|
Fall
|
< 1 meal/week
|
|
45
|
0.419 ± 0.358
|
|
23
|
0.369 ± 0.331
|
|
14
|
0.412 ± 0.261
|
|
8
|
0.574 ± 0.549
|
|
1-2 meals/week
|
|
17
|
0.589 ±0.491
|
|
8
|
0.499 ± 0.347
|
|
6
|
0.411 ± 0.338
|
|
3
|
1.182 ± 0.753
|
|
3-4 meals/week
|
|
23
|
0.792 ±0.467
|
|
8
|
0.633 ± 0.460
|
|
8
|
0.868 ± 0.408
|
|
7
|
0.888 ± 0.554
|
|
≥ 5 meals/week
|
|
16
|
0.830 ±0.539
|
|
6
|
0.910 ± 0.646
|
|
4
|
0.514 ± 0.143
|
|
6
|
0.961 ± 0.577
|
|
Each season-specific model showed that hair MeHg concentration increased with increasing fish/whale consumption frequency (p-values for trend ranging from <0.0001 to 0.005). Tables 6-9 show multivariable random effects regression results. Hair mercury concentrations among participants who had the highest level of fish/whale consumption in each season ranged from 0.30 to 0.50 μg/g higher than those who consumed <1 meal/week. In the model for each season, the magnitude of the effect of each consumption category decreased slightly following adjustment for covariates. The magnitude of the change in hair MeHg concentration corresponding to the contrast of ≥ 3 vs. <1 meal/week was highest for intake during the spring (β: 0.40; 95%CI: 0.20, 0.60). Conversely, the magnitude of this effect was lowest for intake during the winter (β: 0.28; 95%CI: 0.07, 0.50). Tables 6-9 also show the intercepts from these models, representing the expected mean concentration of mercury (μg/g) if all covariates are at their reference level, and corresponding SDs, representing the variation in these values associated with clustering in communities, which reflect residual clustering in each model, suggesting that variation in baseline hair-mercury concentrations across communities is not fully explained by the variables in the model.
Table 6: Random effects regression of hair MeHg concentrations (µg/g) on fish/whale consumption frequency during the spring season, 101 western Canadian Arctic residents, 2016
|
Unadjusted ∞
|
|
Adjusted Φ
|
p-value for Trend
|
|
β
|
95% CI
|
|
β
|
95% CI
|
Sex
|
|
|
|
|
|
|
Male
|
Reference
|
|
|
Reference
|
|
|
Female
|
-0.250
|
-0.430, -0.070
|
|
-0.054
|
-0.248, 0.141
|
|
Hair Length (cm)
|
-0.006
|
-0.0103, -0.0017
|
|
-0.004
|
-0.009, 0.0003
|
|
Dye or Perm
|
|
|
|
|
|
|
No
|
Reference
|
|
|
Reference
|
|
|
Yes
|
-0.124
|
-0.314, 0.0664
|
|
-0.169
|
-0.334, -0.004
|
|
Proportion Cooked
|
0.004
|
0.00013, 0.0076
|
|
0.003
|
0.00004, 0.007
|
|
Spring
|
|
|
|
|
|
|
< 1 meal/week
|
Reference
|
|
|
Reference
|
|
|
1-2 meals/week
|
0.302
|
0.115, 0.489
|
|
0.258
|
0.076, 0.439
|
|
≥ 3 meals/week
|
0.476
|
0.274, 0.679
|
|
0.406
|
0.204, 0.609
|
0.000
|
Random intercept for community from adjusted model: SD 0.131 (95%CI: 0.045, 0.379)
|
∞ Each model included a random intercept for the effect of clustering in communities
Φ Adjusted for sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared by cooking, and fish/whale consumption frequency in the spring
|
Table 7: Random effects regression of hair MeHg concentrations (µg/g) on fish/whale consumption during the summer season, 101 western Canadian Arctic residents, 2016
|
Unadjusted ∞
|
|
Adjusted Φ
|
p-value for Trend
|
|
β
|
95% CI
|
|
β
|
95% CI
|
Sex
|
|
|
|
|
|
|
Male
|
Reference
|
|
|
Reference
|
|
|
Female
|
-0.250
|
-0.430, -0.070
|
|
-0.144
|
-0.341, 0.053
|
|
Hair Length (cm)
|
-0.006
|
-0.0103, -0.0017
|
|
-0.004
|
-0.009, 0.0002
|
|
Dye or Perm
|
|
|
|
|
|
|
No
|
Reference
|
|
|
Reference
|
|
|
Yes
|
-0.124
|
-0.314, 0.0664
|
|
-0.157
|
-0.333, 0.019
|
|
Proportion Cooked
|
0.004
|
0.00013, 0.0076
|
|
0.004
|
0.0009, 0.008
|
|
Summer
|
|
|
|
|
|
|
< 1 meal/week
|
Reference
|
|
|
Reference
|
|
|
1-2 meals/week
|
0.120
|
-0.113, 0.354
|
|
0.044
|
-0.183, 0.270
|
|
3-4 meals/week
|
0.187
|
-0.0669, 0.593
|
|
0.183
|
-0.065, 0.432
|
|
≥ 5 meals/week
|
0.374
|
0.156, 0.593
|
|
0.320
|
0.114, 0.526
|
0.001
|
Random intercept for community from adjusted model: SD 0.147 (95%CI: 0.053, 0.409)
|
∞ Each model included a random intercept for the effect of clustering in communities
Φ Model covariates: sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared by cooking, and fish/whale consumption frequency in the summer
|
Table 8: Random effects regression of hair MeHg concentrations (µg/g) on fish/whale consumption during the fall season, 101 western Canadian Arctic residents, 2016
|
Unadjusted ∞
|
|
Adjusted Φ
|
p-value for Trend
|
|
β
|
95% CI
|
|
β
|
95% CI
|
Sex
|
|
|
|
|
|
|
Male
|
Reference
|
|
|
Reference
|
|
|
Female
|
-0.250
|
-0.430, -0.070
|
|
-0.093
|
-0.291, 0.105
|
|
Hair Length (cm)
|
-0.006
|
-0.0103, -0.0017
|
|
-0.004
|
-0.009, 0.0005
|
|
Dye or Perm
|
|
|
|
|
|
|
No
|
Reference
|
|
|
Reference
|
|
|
Yes
|
-0.124
|
-0.314, 0.0664
|
|
-0.220
|
-0.394, -0.046
|
|
Proportion Cooked
|
0.004
|
0.00013, 0.0076
|
|
0.004
|
0.0002, 0.007
|
|
Fall
|
|
|
|
|
|
|
< 1 meal/week
|
Reference
|
|
|
Reference
|
|
|
1-2 meals/week
|
0.170
|
-0.0663, 0.406
|
|
0.150
|
-0.069, 0.369
|
|
3-4 meals/week
|
0.358
|
0.144, 0.572
|
|
0.323
|
0.113, 0.534
|
|
≥ 5 meals/week
|
0.387
|
0.144, 0.631
|
|
0.376
|
0.143, 0.610
|
0.000
|
Random intercept for community from adjusted model: SD 0.097 (95%CI: 0.025, 0.369)
|
∞ Each model included a random intercept for the effect of clustering in communities
Φ Model covariates: sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared by cooking, and fish/whale consumption frequency in the fall
|
Table 9: Random effects regression of hair MeHg concentrations (µg/g) on fish/whale consumption frequency during the winter season, 101 western Canadian Arctic residents, 2016
|
Unadjusted ∞
|
|
Adjusted Φ
|
p-value for Trend
|
|
β
|
95% CI
|
|
β
|
95% CI
|
Sex
|
|
|
|
|
|
|
Male
|
Reference
|
|
|
Reference
|
|
|
Female
|
-0.250
|
-0.430, -0.070
|
|
-0.131
|
-0.332, 0.070
|
|
Hair Length (cm)
|
-0.006
|
-0.0103, -0.0017
|
|
-0.004
|
-0.009, 0.0004
|
|
Dye or Perm
|
|
|
|
|
|
|
No
|
Reference
|
|
|
Reference
|
|
|
Yes
|
-0.124
|
-0.314, 0.0664
|
|
-0.165
|
-0.339, 0.009
|
|
Proportion Cooked
|
0.004
|
0.00013, 0.0076
|
|
0.004
|
0.0009, 0.008
|
|
Winter
|
|
|
|
|
|
|
< 1 meal/week
|
Reference
|
|
|
Reference
|
|
|
1-2 meals/week
|
0.123
|
-0.0777, 0.323
|
|
0.112
|
-0.077, 0.301
|
|
≥ 3 meals/week
|
0.326
|
0.0992, 0.552
|
|
0.284
|
0.071, 0.497
|
0.005
|
Random intercept for community from adjusted model: SD 0.143 (95%CI: 0.050, 0.405)
|
∞ Each model included a random intercept for the effect of clustering in communities
Φ Model covariates: sex, hair length, use of hair dyes or permanent treatments, the proportion of fish meals usually prepared by cooking, and fish/whale consumption frequency in the winter
|
Table 10 shows the distribution of ascertained diet components overall and by community. Few participants reported consuming more than 2 daily servings of fruit and vegetables. Conversely, a large proportion of participants (45%; 45/101) reported consuming dairy products more than once per day. Of 101 participants, 40 reported regular use of dietary supplements (vitamins calcium, fish oil, omega-3 and fibre). Table 10 also shows MeHg concentrations within intake categories of dietary components. Inspection of these distributions does not reveal clear patterns of association between intake frequencies and hair MeHg levels, consistent with results of the model building procedures, which did not identify these dietary factors as important confounders or effect-measure modifiers of the relationship under investigation in the study population. It should be noted, however, that the observed
Table 10: Hair MeHg concentrations (µg/g) within categories of year-round average dietary component intake by community, 101 western Canadian Arctic residents, 2016
|
Total
(N=101)
|
|
Aklavik, NWT
(N=45)
|
|
Old Crow, YT
(N=32)
|
|
Fort McPherson, NWT (N=24)
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
|
n (%)
|
Mean ± SD
|
Fruit & Vegetables
1-3 times/week
|
20 (20)
|
0.689 ± 0.517
|
|
11 (24)
|
0.530 ± 0.417
|
|
5 (16)
|
0.476 ± 0.183
|
|
4 (17)
|
1.391 ± 0.493
|
4-7 times/week
|
27 (27)
|
0.611 ± 0.509
|
|
12 (27)
|
0.490 ± 0.393
|
|
5 (16)
|
0.512 ± 0.441
|
|
10 (42)
|
0.807 ± 0.637
|
> 1 time/day;
< 2 times/day
|
30 (30)
|
0.531 ± 0.399
|
|
13 (29)
|
0.358 ± 0.284
|
|
11 (34)
|
0.667 ± 0.443
|
|
6 (25)
|
0.654 ± 0.448
|
≥ 2 times/day
|
24 (24)
|
0.589 ± 0.470
|
|
9 (20)
|
0.736 ± 0.630
|
|
11 (34)
|
0.450 ± 0.266
|
|
4 (17)
|
0.640 ± 0.512
|
Dairy
≤ 1 time/week
|
29 (29)
|
0.610 ± 0.504
|
|
13 (29)
|
0.531 ± 0.398
|
|
10 (31)
|
0.502 ± 0.430
|
|
6 (25)
|
0.964 ± 0.716
|
2-4 times/week
|
14 (14)
|
0.672 ± 0.418
|
|
9 (20)
|
0.590 ± 0.399
|
|
1 (3)
|
0.593
|
|
4 (17)
|
0.875 ± 0.503
|
5-7 times/week
|
13 (13)
|
0.252 ± 0.156
|
|
9 (20)
|
0.231 ± 0.132
|
|
2 (6)
|
0.369 ± 0.253
|
|
2 (8)
|
0.229 ± 0.240
|
> 1 time/day
|
45 (45)
|
0.666 ± 0.485
|
|
14 (31)
|
0.622 ± 0.561
|
|
19 (59)
|
0.573 ± 0.338
|
|
12 (50)
|
0.864 ± 0.567
|
Uses Dietary Supplements
No
|
62 (61)
|
0.577 ± 0.466
|
|
33 (73)
|
0.458 ± 0.361
|
|
19 (59)
|
0.592 ± 0.436
|
|
10 (42)
|
0.940 ± 0.655
|
Yes
|
39 (39)
|
0.630 ± 0.473
|
|
12 (27)
|
0.657 ± 0.590
|
|
13 (41)
|
0.460 ± 0.160
|
|
14 (58)
|
0.765 ± 0.536
|
Hair length was inversely correlated with MeHg concentration, with 1 cm increases in length corresponding to slight reductions in μg/g of mercury after adjusting for sex, permanent hair treatment use, fish/whale consumption and the proportion of fish/whale meals prepared by cooking (tables 6-9). Multivariable random effects regression yielded evidence of reduced hair mercury concentration among those who reported recent use of permanent hair treatments relative to those with untreated hair (tables 6-9). Figure 4 shows the adjusted effects of consuming different quantities of fish/whale in each season on hair mercury concentration, stratified by use of permanent hair treatments. These graphs show that within categories of fish/whale intake, participants who used permanent hair treatments had lower hair mercury concentrations relative to those who did not use such treatments.
Tables 6-9 show SDs and 95% CIs representing the random effect for clustering in communities. Visual comparison of community-specific patterns of fish/whale intake (table 2) revealed the following species as having the most divergent consumption patterns across communities: Beluga Whale (D.leucas), Arctic Grayling (T.arcticas), Chinook Salmon (O.tshawytscha) and Burbot (L.lota). Assessment of the linearity of the relationships between intake frequency and hair MeHg concentration indicated that each of these variables could be modeled as continuous. Table 11 shows the SDs and 95% CIs representing residual variation across communities for listed sets of covariates. The largest reduction in the SD was observed in the fall model, with the addition of Chinook Salmon (O.tshawytscha) and Arctic Grayling (T.arcticas) intake modeled as continuous variables. Variation across communities increased with the inclusion of beluga whale consumption in all models, likely because whale consumption was almost exclusively reported by residents of Aklavik, NT. Inclusion of this variable highlighted a strong association between Beluga Whale (D.leucas) consumption and hair MeHg concentration. The beta coefficients (95% CI) showing the increase in hair MeHg concentration for each one-meal-per-week increase in beluga whale intake in the spring, summer, fall and winter seasons were 0.2 (95%CI: 0.01, 0.4), 0.04 (95%CI: -0.05, 0.1), 0.3 (95%CI: 0.05, 0.5) and 0.3 (95%CI: -0.04, 0.6), respectively.
Of the 22 samples selected at random in the lab for duplicate MeHg concentration measurement, 4 came from individuals also among the 10 selected at random as blind duplicates, yielding 28 participants with duplicate measurements; the median percent change in MeHg concentration between measurements was 14.67% (IQR: 10.75) with 36% (10/28) having second values that were higher than the initial value. The maximum percent change of 159% corresponded to a woman with hair that had been dyed one month before sample collection. Excluding this extreme outlier, the mean percent change in MeHg concentration between measurements was 15.84% (SD: 9.95; Range: 3.32 - 43.77). We performed a bias analysis using MeHg values adjusted by the mean percent change observed in the validation subsample after excluding the outlier; we increased by 15.84% the values of 36% of the study population selected at random and decreased by the same amount the values of the remaining 64%. The mean of the adjusted MeHg values was 0.58 μg/g (SD: 0.47; Range: 0.05-2.05).
Table 11: Sensitivity analysis: Relationship of selected variables to residual between-community heterogeneity, 101 western Canadian Arctic residents, 2016
Model
|
|
SD
|
|
95%CI
|
Spring
|
Model 1: Sex, Hair Length, Proportion Cooked, Permanent Hair
Treatments, Total Fish/Whale Consumption in Spring
|
|
0.131
|
|
0.045, 0.379
|
Model 1 + Beluga Whale Consumption
|
|
0.144
|
|
0.053, 0.393
|
Model 1 + Arctic Grayling Consumption
|
|
0.124
|
|
0.042, 0.370
|
Model 1 + Chinook Salmon Consumption
|
|
0.116
|
|
0.037, 0.363
|
Model 1 + Chinook Salmon & Arctic Grayling Consumption
|
|
0.115
|
|
0.037, 0.361
|
|
|
|
|
|
Summer
|
Model 2: Sex, Hair Length, Proportion Cooked, Permanent Hair
Treatments, Total Fish/Whale Consumption in Summer
|
|
0.156
|
|
0.057, 0.423
|
Model 2 + Beluga Whale Consumption
|
|
0.159
|
|
0.059, 0.426
|
Model 2 + Chinook Salmon Consumption
|
|
0.155
|
|
0.054, 0.445
|
Model 2 + Arctic Grayling Consumption
|
|
0.150
|
|
0.054, 0.413
|
Model 2 + Chinook Salmon & Arctic Grayling Consumption
|
|
0.154
|
|
0.054, 0.442
|
|
|
|
|
|
Fall
|
Model 3: Sex, Hair Length, Proportion Cooked, Permanent Hair
Treatments, Total Fish/Whale Consumption in Fall
|
|
0.105
|
|
0.030, 0.371
|
Model 3 + Burbot Consumption
|
|
0.109
|
|
0.032, 0.371
|
Model 3 + Beluga Whale Consumption
|
|
0.119
|
|
0.038, 0.368
|
Model 3 + Arctic Grayling Consumption
|
|
0.092
|
|
0.022, 0.382
|
Model 3 + Chinook Salmon Consumption
|
|
0.070
|
|
0.010, 0.473
|
Model 3 + Chinook Salmon & Arctic Grayling Consumption
|
|
0.064
|
|
0.008, 0.534
|
|
|
|
|
|
Winter
|
Model 4: Sex, Hair Length, Proportion Cooked, Permanent Hair
Treatments, Total Fish/Whale Consumption in Winter
|
|
0.151
|
|
0.054, 0.417
|
Model 4 + Burbot Consumption
|
|
0.146
|
|
0.052, 0.411
|
Model 4 + Arctic Grayling Consumption
|
|
0.146
|
|
0.052, 0.410
|
Model 4 + Chinook Salmon Consumption
|
|
0.140
|
|
0.048, 0.404
|
Model 4 + Chinook Salmon & Arctic Grayling Consumption
|
|
0.139
|
|
0.048, 0.403
|
Model 4 + Chinook Salmon, Arctic Grayling & Burbot Consumption
|
|
0.133
|
|
0.045, 0.394
|
We inspected the mean percent change in repeat measurements by participant characteristics within the validation subsample and identified use of hair dyes or permanent hair treatments as possibly related to the magnitude of the difference between repeated measurements. The mean percent change among those who reported recent use of hair dye or other permanent hair treatments was 38.5 (SD: 59.5), compared to 16.2 (SD: 10.4) among those who did not. Among those who used permanent hair treatments, 50% had second values that were higher than initial values, while among those with untreated hair, 32% had second values that were higher than initial values. We performed a second bias analysis using MeHg values adjusted by the observed mean percent change by hair treatment status, increasing by 38.5% the values of 50% of those with treated hair selected at ransom and decreasing by the same amount the values of the other 50%, and increasing by 16.2% the values of 32% of those with untreated hair selected at random and decreasing by the same amount the remaining participants with untreated hair. This adjustment resulted in a mean MeHg concentration of 0.57 μg/g (SD: 0.50; Range: 0.05-2.87). Under each of these scenarios, all participants remained at levels below those thought to pose serious health risks. Table 12 compares results of regression models using the originally measured MeHg concentrations to results of models using values adjusted for measurement error. These comparisons indicate that error in laboratory measurement of MeHg is not likely to have impacted inferences drawn from this analysis.
Table 12: Sensitivity Analysis: Multivariable random effects regression of hair MeHg concentrations (µg/g) on fish/whale intake, comparing models using measured MeHg values to models using values adjusted for the degree of variability in MeHg measurements observed in subsets of the study population, 101 western Canadian Arctic residents, 2016
Fish/Whale
Intake (meals/week)
|
Original MeHg Measurement
|
|
Adjusted MeHg Measurement 1 ⌘
|
|
Adjusted MeHg Measurement 2§
|
β
|
95% CI
|
Trend
p-value
|
|
β
|
95% CI
|
Trend
p-value
|
|
β
|
95% CI
|
Trend
p-value
|
Spring
|
|
|
|
|
|
|
|
|
|
|
|
< 1
|
Ref
|
|
0.000
|
|
Ref
|
|
0.000
|
|
Ref
|
|
0.000
|
1-2
|
0.258
|
0.076, 0.439
|
|
0.231
|
0.048, 0.413
|
|
0.288
|
0.085, 0.4900
|
≥ 3
|
0.406
|
0.204, 0.609
|
|
0.377
|
0.174, 0.579
|
|
0.444
|
0.220, 0.668
|
Summer
|
|
|
|
|
|
|
|
|
|
|
|
< 1
|
Ref
|
|
0.001
|
|
Ref
|
|
0.001
|
|
Ref
|
|
0.001
|
1-2
|
0.044
|
-0.183, 0.270
|
|
0.017
|
-0.207, 0.241
|
|
0.082
|
-0.167, 0.331
|
3-4
|
0.183
|
-0.065, 0.432
|
|
0.104
|
-0.133, 0.342
|
|
0.059
|
-0.204, 0.323
|
≥ 5
|
0.320
|
0.114, 0.526
|
|
0.293
|
0.089, 0.498
|
|
0.365
|
0.138, 0.592
|
Fall
|
|
|
|
|
|
|
|
|
|
|
|
< 1
|
Ref
|
|
0.000
|
|
Ref
|
|
0.000
|
|
Ref
|
|
0.000
|
1-2
|
0.150
|
-0.069, 0.369
|
|
0.135
|
-0.089, 0.358
|
|
0.185
|
-0.064, 0.434
|
3-4
|
0.323
|
0.113, 0.534
|
|
0.216
|
0.006, 0.427
|
|
0.180
|
-0.053, 0.414
|
≥ 5
|
0.376
|
0.143, 0.610
|
|
0.274
|
0.041, 0.507
|
|
0.400
|
0.142, 0.658
|
Winter
|
|
|
|
|
|
|
|
|
|
|
|
< 1
|
Ref
|
|
0.005
|
|
Ref
|
|
0.004
|
|
Ref
|
|
0.008
|
1-2
|
0.112
|
-0.077, 0.301
|
|
0.087
|
-0.099, 0.274
|
|
0.107
|
-0.103, 0.317
|
≥ 3
|
0.284
|
0.071, 0.497
|
|
0.291
|
0.081, 0.501
|
|
0.302
|
0.066, 0.538
|
⌘ MeHg concentrations adjusted by the mean percent change of 15.8% estimated among 27 individuals with repeat measurements (excluding 1 outlier). MeHg concentration was increased by 15.8% for a random selection of 36% of participants and decreased by 15.8% for the remaining 64%, based on the distribution of increased or decreased values in the validation subset.
- MeHg concentrations adjusted according to reported use of permanent hair treatments: Among participants who used hair treatments, 50% were increased by the 38.5% mean percent change observed among 6 participants in the validation subset with reported use of permanent hair treatments; among those who did not use permanent hair treatments, 32% were increased by the 16.2% mean percent change observed among 22 participants in the validation subset who did not use permanent hair treatments.