Diet quality is positively associated with intake of traditional foods and does not differ by season in remote Yup’ik communities

ABSTRACT

This study evaluated whether traditional food intake and diet quality differed by season in Yup’ik communities and examined the relationship between intake of traditional food groups and diet quality. Data were collected from 38 participants, ages 14–79 years, from two Yup’ik communities in Southwest Alaska from 2008 to 2010. Self-reported intake (24-h recalls) and dietary biomarker (nitrogen stable isotope ratio) data were collected twice in distinct seasons. Diet quality was assessed using the Healthy Eating Index. A paired sample t-test was used to test for seasonal differences in traditional food intake and diet quality, and linear regression was used to evaluate associations between traditional food intake and diet quality. Total traditional food intake and overall diet quality did not significantly differ by season, but there were differences in traditional food group intake and diet quality component scores. Diet quality was strongly associated with intake of traditional food groups including fish, tundra greens, and berries. Given the strong relationship between traditional food intake and diet quality, policies should aim to ensure continued access to traditional foods in Yup’ik communities amid environmental changes in the circumpolar North.

KEYWORDS:

• Healthy eating index
• Alaska native
• Yup’ik
• seasonal variation
• indigenous communities
• dietary biomarker

Introduction

Traditional foods are fundamental to the health of Yup’ik communities in the Yukon-Kuskokwim Delta (YKD) region of Southwest Alaska. A high intake of traditional foods is associated with lower levels of metabolic syndrome [1] and cardiometabolic risk factors [2]. Ties to traditional food systems provide additional health benefits through adherence to cultural and spiritual practices and increased physical activity [3,4]. In Yup’ik communities, traditional foods are also the primary source of key nutrients including vitamin D, omega-3 fatty acids and other vitamins [5,6] and are closely related to food security and diet quality [7–9]. Traditional Yup’ik foods are harvested from land and water and include food groups such as tundra greens, berries, birds, land mammals, marine mammals and fish. Fish, such as salmon, are the cornerstone of a traditional Yup’ik diet, contributing an estimated 50% of the energy from traditional foods [5] and holding significance for cultural beliefs and practices [3].

Intake of traditional foods in Yup’ik communities has dropped in past decades amid a nutritional transition from traditional subsistence foods to western market foods resulting from colonisation and other socioeconomic and climatic pressures [10–12]. Western foods in Yup’ik communities are sourced from small stores that commonly provide shelf-stable, energy-dense processed foods with fewer options for nutrient-dense vegetables and fruits due to the geographic remoteness of the YKD region. However, the extent of fresh and canned vegetable and fruit options in the stores vary from community to community. Nutrition transitions are associated with drops in diet quality and are implicated in a number of health disparities faced by Indigenous People, such as a high burden of cancer, obesity, and diabetes [13–15]. Since traditional food intake is strongly linked to diet quality and overall health and is protective of chronic disease development, some efforts to address health disparities in Yup’ik communities have focused on strengthening adolescents’ ties to traditional food systems [16]. One study found that leveraging the cultural and physical resources of the traditional food system in a Fish-to-School program, Neqa Elicarvigmun, significantly improved students’ diet quality [16].

Culturally appropriate dietary interventions rely on an understanding of the local food environment [17], including the seasonal availability of traditional foods. In addition, dietary interventions that aim to observe changes in diet may be biased if seasonal variation is not taken into account [18,19]. Subsistence activities (i.e. fishing, harvesting, and hunting) are highly seasonal [20,21]. In the YKD, tundra plants and berries are harvested in the summer or fall during a short growing season. Salmon and halibut fishing occurs in late spring or summer, while other fish are harvested at different times of the year, including through the ice in the winter. Intake of some traditional foods can extend to all seasons by means of food preservation: berries can be frozen and fish can be dried, frozen or salt-cured. However, total traditional food intake in Yup’ik communities can vary based on climate change, snow and ice coverage, strength of fish runs, federal or state fishing restrictions that limit hunting or fishing openings, what foods can be preserved, and storage capacity limitations. Diet quality in Yup’ik communities may also differ by season given that traditional food intake is strongly predictive of diet quality [7–9].

The primary aims of this study were to determine whether total traditional food intake and overall diet quality differed by season in two Yup’ik communities and to examine the association between intake of traditional food groups and overall diet quality. Findings from this study can be used to inform ongoing dietary interventions aimed at improving the health of Yup’ik communities by understanding how seasonality influences traditional food intake and diet quality.

Materials and methods

Recruitment and study participants

This study is a secondary analysis of existing data from the Negem Nallunailkutaa (“The Foods’ Marker”) study [22–25]. The study measured dietary intake in two seasons (winter [October to December] and spring [March to June]) and offers the chance to study the seasonality of traditional food intake and diet quality in Yup’ik communities. Yup’ik participants aged 14 years or older were recruited via posters and word of mouth from two coastal communities in Southwest Alaska. Seventy participants were enrolled in the primary study and 38 were included at both time points. Data were collected in each community at two time points. In the first community, recruitment and baseline data collection took place from October to December 2008 and follow-up data collection took place from May to June 2010. In the second community, recruitment and baseline data collection took place from March to April 2009 and follow-up data collection took place from November to December 2009. Written consent was obtained from adults; minors signed an assent form and their parent or guardian signed a consent form. Participants received $75 at each time point upon completion of all study procedures ($150 total). This study was approved by the University of Alaska Fairbanks Institutional Review Board and the Yukon-Kuskokwim Health Corporation Human Studies Committee and Executive Board of Directors.

Data collection

Sociodemographic survey

At study recruitment, participants completed a short demographic survey that asked about age, sex, education, income range, experience of food shortage, food assistance (use of any program including Women, Infants, and Children [WIC] or Supplemental Nutrition Assistance Program [SNAP]), and adherence to the traditional Yup’ik way of life. Yup’ik way of life was measured with a single item on the demographic survey: “People talk to us about traditional ways of living. How much do you follow the traditional Yup’ik way of life?”

24-hour dietary recall

To measure self-reported intake, participants completed eight 24-h recalls over the entire study period (four recalls at each time point). The 24-h recalls were administered by certified interviewers using computer-assisted software [Nutrition Data System for Research (NDSR) software 2008, University of Minnesota]. Participants were asked to recall all food and beverages consumed the day prior to the interview using a multiple-pass approach. Participants were given portion estimation tools (measuring cups, rulers, and food models or portion estimation guides; Fred Hutchinson Cancer Research Center, Seattle, WA). A native Yup’ik speaker conducted interviews for participants who did not speak English. Dietary interviews were conducted on average every 9 days (standard deviation: 5 days), with a minimum of 2 days between recalls.

Dietary biomarker

During the first visit, participants provided hair and blood samples to measure a validated dietary biomarker of fish and traditional Yup’ik foods, the stable isotope ratio of nitrogen (NIR) [25]. During the second visit, participants provided only hair samples because hair is less invasive to collect and findings indicated that the NIR of hair is highly correlated with the NIR of blood [22,26]. All hair and blood samples were prepared and analysed at the Alaska Stable Isotope Facility as previously described [23]. Using data from participants who had both blood and hair samples available, a linear regression model was created to convert the NIR of hair to the NIR of blood [26,27]. The observed regression model was used to convert all NIR samples to the NIR of blood to allow for comparability of data between study visits.

Variables

Traditional food intake

Traditional food intake was assessed using both self-reported intake data and dietary biomarker data. For the self-reported intake variable, all ingredients reported in the 24-h recalls were coded as either traditional foods or non-traditional foods. Each traditional food was then assigned to one of the following traditional food groups: tundra greens, berries, birds, land mammals, marine mammals, or fish. Alaska Native mixed dishes that contain both traditional and non-traditional foods (e.g. moose stew or akutaq, which is a Yup’ik food made with different recipes that may include berries, sugar, fats [seal oil or shortening], and sometimes fish or meat) were disaggregated into their respective ingredients. Finally, traditional food intake was operationalised as percent of energy from traditional foods: (kilocalories from traditional foods/kilocalories from all foods) · 100%.

The traditional food intake biomarker variable based on the NIR was presented as standard delta values as δX = (R_sample − R_standard)/(R_standard) · 1000‰, where R is the ratio of heavy to light isotope [15]N/ [14]N) and the standard is atmospheric nitrogen. In a validation study, the hair traditional food biomarker varied from 6.9 to 15.2‰, corresponding to diets in which the percent of energy from fish and marine mammals ranged from 0% to 57% [25].

Diet quality

The Healthy Eating Index (HEI) was used to assess diet quality based on data from the 24-h recalls. The HEI provides an overall diet quality measure based on consistency with the recommendations set by the Dietary Guidelines for Americans [28] and 13 diet quality components that measure adequacy and moderation in the diet. Overall diet quality and each diet quality component are scored from 0 to 100 where a higher score represents a diet that is more closely aligned with dietary recommendations [28]. The average HEI for all adults in the US is 59 [29].

Statistical analysis

Descriptive statistics were generated for study participants who completed data collection at each time point. Differences in diet quality and traditional food intake by sociodemographic characteristics were tested using Student’s t-test for binary variables (e.g. sex) and ANOVA for variables with more than two categories (e.g. education and income). A complete-case analysis was used to account for participants with missing sociodemographic data. The paired sample t-test was used to test whether total traditional food intake and overall diet quality differed by season. Associations between diet quality and traditional food intake were evaluated using linear regression. Based on work that has suggested Yup’ik individuals who eat fish also eat other traditional food groups [19] and that fish intake is the main source of energy from traditional foods [5], we also evaluated whether associations between other traditional food groups and diet quality were confounded by fish intake using multivariable linear regression. A significance level of α = 0.05 was used throughout the analyses. Statistical analysis was performed using SPSS Statistics 25 (IBM Corp.; 2017 Armonk, NY).

Results

Traditional food intake and diet quality by demographic characteristics

Overall, traditional food intake accounted for 20.9% of energy and overall diet quality measured by the HEI was 47.8 (Table 1). Traditional food intake was strongly positively associated with age group (p < .001). In comparison, diet quality increased with age group, but did not attain statistical significance (p = .06). Traditional food intake significantly differed by self-reported adherence to the traditional Yup’ik way of life (p = .03).

Table 1. Diet quality and traditional food intake reported by sociodemographic characteristics for Yup’ik participants (n = 38).

	Yup’ik Study Participants n (%)	Traditional Food Intake % of Energy Mean (SD)	p^b	Diet Quality Mean (SD)^c	p
Overall		20.9 (17.4)		47.8 (8.2)
Age group			<.001		.06
14–19 years	6 (16%)	6.5 (7.1)		42.8 (9.0)
20–39 years	12 (32%)	10.4 (9.2)		45.1 (7.7)
40–59 years	15 (39%)	25.8 (13.4)		49.3 (8.0)
60–79 years	5 (13%)	48.6 (13.6)		54.8 (4.0)
Sex			.57		.49
Male	18 (47%)	22.6 (19.0)		48.8 (1.9)
Female	20 (53%)	19.4 (16.2)		46.9 (1.9)
Education, years			.96		.61
<12 years	14 (37%)	21.0 (23.1)		46.0 (9.5)
12 years	14 (37%)	20.0 (14.2)		48.4 (2.2)
>12 years	10 (26%)	22.1 (13.4)		49.3 (7.9)
Income			.74		.67
<$10,000	7 (18%)	23.4 (13.5)		51.1 (8.2)
$10,000–24,999	5 (13%)	20.7 (20.2)		47.0 (8.9)
$25,000–49,999	6 (16%)	14.4 (15.3)		46.8 (6.9)
$50,000–99,999	5 (13%)	15.4 (14.4)		46.3 (6.5)
Unknown/No Response	15 (39%)	23.8 (20.2)		47.3 (9.4)
Food Money Shortage			.55		.03
Never	6 (16%)	22.0 (9.9)		52.1 (6.5)
Seldom	4 (10%)	13.3 (12.9)		42.4 (2.6)
Sometimes	11 (29%)	24.3 (17.6)		50.8 (6.5)
Most Times	9 (24%)	17.0 (13.1)		43.4 (8.6)
Always	6 (16%)	29.9 (29.0)		53.0 (8.2)
Unknown	2 (5%)	4.5 (1.9)		36.3 (0.7)
Receives Any Food Assistance			.59		.18
Yes	16 (42%)	19.1 (16.1)		45.6 (1.7)
No	22 (58%)	22.2 (18.5)		49.6 (2.1)
Adherence to the Traditional Yup’ik Way of Life			.03		.07
A lot	12 (32%)	31.5 (18.5)		52.2 (6.2)
Some	24 (63%)	16.5 (15.0)		45.6 (8.0)
Not at all	2 (5%)	9.9 (14.1)		47.1 (16.0)

Note: SD, standard deviation.

^bP-values are testing for differences in diet quality and traditional food intake by sociodemographic characteristics and were generated with Student’s t-test for binary variables and ANOVA for variables with more than two categories. A complete-case analysis was used to account for participants with missing data for a sociodemographic characteristic.

^cDiet quality is measured using the healthy eating index which is scored from 0-100 where a higher value represents a diet that is more closely aligned with dietary recommendations.

Traditional food intake by season

Total traditional food intake did not differ between spring and winter when estimated via either self-reported intake (20.3% of total energy in spring vs. 21.5% of total energy in winter, p = .55.) or via the biomarker of traditional food intake (9.2‰ in spring vs. 9.4‰ in winter, p = .30) (Table 2). The only traditional food group that differed between spring and winter was land mammals, which accounted for 1.4% of total energy in spring and 2.8% in winter (p = .03). Intake of the traditional food groups tundra greens, berries, and birds each contributed less than 2% of total energy in each season. Marine mammal intake and fish intake contributed the greatest percent of total energy (6–10%) in spring and winter.

Table 2. Total traditional food intake and intake of traditional food groups compared between spring and winter for Yup’ik study participants.

	Spring Mean (SD)	Winter Mean (SD)	p^b
Total Traditional Food Intake, % of Energy (n = 38)	2.3 (17.8)	21. 5 (19.2)	.55
Traditional Food Groups, % of Energy (n = 38)
Tundra Greens	<.1 (.1)	0.1 (0.2)	.23
Berries	.7 (1.2)	0.4 (0.8)	.05
Birds	2.0 (3.5)	0.8 (2.3)	.10
Land Mammals	1.4 (2.0)	2.8 (3.9)	.03
Marine Mammals	6.4 (8.3)	7.7 (9.2)	.37
Fish	9.8 (1.3)	9.9 (9.7)	.93
Traditional Food Biomarker, ‰ (n = 33)^a	9.2 (1.8)	9.4 (1.8)	.30

Note: SD, standard deviation.

^aThe traditional food biomarker is the nitrogen stable isotope ratio of blood.

^bp-values are reported for paired sample t-test between spring and winter.

Diet quality by season

Overall diet quality did not differ by season and was low (47.6 in spring vs. 48.0 in winter, p = .85). The diet quality component score for total vegetables was significantly lower in the spring than in the winter (20.5 vs 29.9, p = .002), whereas the whole grain and sodium diet quality component scores were significantly higher in the spring than in winter (45.5 vs. 23.7, p = .04 and 66.6 vs. 51.5 p = .02, respectively) (Figure 1). There were no significant differences in any other diet quality component scores. Averaged across seasons, diet quality component scores were low for greens and beans (10.5), total fruits (17.9), whole fruits (21.2), and total vegetables (25.2). In comparison, the average diet quality component scores were much higher for refined grains (75.3), seafood and plant proteins (85.9), and total protein (97.1).

Figure 1. Comparison of diet quality component scores between spring and winter for Yup’ik study participants (N = 38).

Note: *P < .05, **P < .01.

^aEach diet quality component score is measured using the healthy eating index, which is scored from 0 to 100 where a higher value represents a diet that is more closely aligned with dietary recommendations.

^bSignificant differences in each component score between spring and winter were tested with the paired sample t-test.

Relationship between diet quality and traditional food intake

Since total traditional food intake and overall diet quality did not differ by season, spring and winter data were combined to evaluate the association between diet quality and traditional food intake. Diet quality was associated with total traditional food intake measured via both self-reported intake (beta = 0.3, p < .001) and the traditional food biomarker (beta = 2.6, p = .001) (Table 3). Diet quality was also strongly associated with an intake of tundra greens (beta = 20.2, p = .04) and berries (beta = 2.7, p = .07). However, after adjusting for fish intake, the association between berries and diet quality was no longer statistically significant.

Table 3. Associations between overall diet quality and traditional food intake among Yup’ik study participants (N = 38).

	Overall Diet Quality		Overall Diet Quality
	Beta (95% CI)	p	Adjusted Beta^a(95% CI)	p
Total Traditional Food Intake	0.3 (0.16, 0.42)	<.001	-	-
Traditional Food Groups
Tundra Greens	23.7 (2.00, 45.45)	.03	20.2 (0.44, 39.97)	.04
Berries	4.0 (1.30 6.78)	.005	2.7 (−0.20, 5.64)	.07
Birds	1.5 (0.25, 2.76)	.02	0.8 (−0.50, 2.17)	.21
Land Mammals	0.5 (−0.72, 1.63)	.43	−0.1 (−1.26, 1.01)	.83
Marine Mammals	0.7 (0.43, 1.00)	<.001	0.7 (0.30, 1.09)	.001
Fish	0.4 (0.15, 0.72)	.004	-	-
Traditional Food Biomarker	2.6 (1.15, 4.06)	.001	-	-

Note: ^aEstimates were adjusted for fish intake given that fish intake was a hypothesised confounder in the association.

^bThe traditional food biomarker is the nitrogen stable isotope ratio of blood.

Discussion

This study highlights the critical importance of access and intake of traditional foods to the diet quality of Indigenous communities [7,9,30,31]. We found that fish intake was strongly related to diet quality in coastal Yup’ik communities of Southwest Alaska and fish intake did not differ between spring and winter. This finding is particularly important in the light of the environmental changes the YKD has experienced in the past decades that raise concerns about traditional food security. For example, in 2021, a salmon fishery collapse on the Yukon River resulted in a salmon run that was only about a tenth of normal years, leading to inadequate numbers of fish to meet subsistence needs [32]. In addition, salmon body sizes have begun to decline in the YKD region [33], and some communities have had to relocate due to rising sea levels causing erosion [34]. The decline in availability of fish is likely to have a serious and negative impact on diet quality, further exacerbating health disparities. The instrumental role of traditional foods such as fish in the diet quality and overall health of Indigenous communities underscores the urgent need to implement policies that ensure continued access to traditional foods, including joint management of fishery resources and support for Indigenous food sovereignty, as led by the Inuit Circumpolar Council [35].

We did not detect any seasonal differences in total traditional food intake nor in overall diet quality in this study; however, there were seasonal differences in some traditional food subgroups and diet quality component scores. To our knowledge, this is the first study to examine whether there is seasonal variation in diet quality in Indigenous communities. However, the consistency in diet quality that we observed across seasons is not surprising given that we also observed seasonal consistency in total traditional food intake, which has been shown to strongly predict diet quality [7,9,36]. In addition, we found that traditional food intake was higher among older age groups, a well-documented relationship in the literature [12], which could potentially explain the generally higher diet quality among older age groups in this study. Previous work has explored seasonal variations in diet for Alaska Native communities [20,21,37–39]. One study found that protein sources differ by season in Alaska Native communities [20], which is consistent with our finding that suggested intake of traditional protein sources such as land mammals significantly differed by season. One explanation for the lack of seasonal variation in total traditional food intake could be the consistent intake of fish between seasons, which is made possible due to salmon preservation practices (e.g. drying and freezing). This is also supported by the fact that the study communities were coastal and likely had low intake of land mammals generally compared to fish intake.

The overall diet quality score for Yup’ik study participants was lower compared to the overall diet quality score for a nationally representative sample of US adults. Specifically, the average overall diet quality score was about 48 for the Yup’ik study participants and is about 59 for all American adults [29]. Disparities in the overall diet quality scores could be explained by the relatively low component scores for vegetable and fruit components for the Yup’ik study participants. Averaged across seasons, diet quality component scores for greens and beans, total fruits, whole fruits, and total vegetables ranged from 10 to 25. Corresponding metrics for US adults range from about 50 to 70 [29]. Similar disparities in diet quality indices have been reported for other Indigenous communities compared to non-Indigenous communities. For instance, results from the 2004 Canadian Health Survey and the 2015 Canadian Health Survey show persistent disparities in diet quality for Indigenous populations relative to non-Indigenous populations [40]. Income is another important predictor for diet quality [41], and the relatively low-income for the study region could also contribute to the disparity in diet quality.

The low vegetable and fruit diet quality component scores among study participants are consistent with previous work [42,43]. Vegetable diet quality component scores were particularly low in the spring. It is not clear why there were seasonal differences in vegetable intake, but it could be because of seasonal deficiencies in the supply and/or the purchasing of market vegetables; only the vegetable diet quality component score differed by season, while the corresponding vegetable traditional food group (tundra greens) did not. Fresh produce options in small stores in these communities are limited and expensive due to constraints to the food distribution system in remote regions that depend on bush planes to transport food. There may also be fluctuations in income around the year that influence the purchasing of produce from small stores. For example, distributions of the Alaska permanent fund dividend (an annual permanent fund from oil revenue distributed to all Alaska residents), other Native corporation dividends, and timing of food assistance programs could influence purchasing behaviours seasonally [19]. Additional research aimed at understanding the reasons behind seasonal differences in vegetable intake could lend insight into how to increase vegetable intake and subsequently improve diet quality in Yup’ik communities.

This is the first study to show that the overall diet quality in Yup’ik communities is not only associated with total traditional food intake in Yup’ik communities but also with individual traditional food groups including tundra greens and berries. These findings suggest that interventions aimed at improving diet quality in Yup’ik communities could be successful by promoting intake of any traditional food group, especially the traditional food groups that are currently being consumed at low levels and that fall below the corresponding diet quality recommendations (i.e. vegetables, fruit, tundra greens and berries). Since food preservation may be important for year-round intake of traditional foods, dietary interventions aimed at increasing traditional food intake in these communities could be most successful when subsistence practices are supported at every level, from resources and equipment for harvesting to preservation. Such programs can also include families and communities in multi-level interventions that leverage the knowledge and perspective of Yup’ik elders [44–46].

Strengths and limitations

This study had some key strengths. First, the high-quality dietary data included four unannounced 24-h dietary recalls, which are considered the gold standard of dietary assessment. Second, the dietary biomarker data also strengthened our assessment of dietary intake as it is not subject to self-report bias. The data source also presents another strength as it included data collected over multiple seasons, which is not always represented in dietary studies among Indigenous communities.

There are also limitations to the study. First, these data were collected a decade ago and may not reflect current dietary intake. Nevertheless, these data represent the most recent and only data available and provide a unique and important opportunity to examine seasonality in the Yup’ik diet. In addition, we do not expect that the associations we tested in this study (association of traditional food intake with diet quality) to change over time because any relationship observed is driven by the nutrient profile of the food, which is a static characteristic. Second, the study only included 38 participants, aged 14–70 years, in coastal communities, which limits the generalisability of our findings. Future research should explore seasonal differences in a larger sample and include Yup’ik children as well as inland communities. Third, the data presented here does not consider intake in late summer and early fall months and data were only collected in one winter and one spring. Seasonal changes in diet during late summer and fall could differ from spring and winter and the years selected for this study could not have been representative of other years. Finally, a complex system influences Indigenous food access including factors like food security, biodiversity migration and hunting patterns, which were not examined in this study. Future research should consider these factors in evaluations of traditional food intake.

Conclusion

In this study of diet quality in two coastal Yup’ik communities, we found that fish intake is an important driver of diet quality and is consistent across seasons. The findings also suggest that overall diet quality could be improved by increasing intake of both traditional and market vegetables and fruits and moderating intake of saturated fats, refined grains, and processed foods rich in sodium and added sugar. To successfully decrease the diet-related health disparities experienced by Yup’ik communities, significant public health and research investments will need to be made across the individual, community, and systemic factors that impact access and intake of traditional foods and vegetables and fruit. In addition, to overcome the threat to traditional food security posed by climate change that has triggered loss of biodiversity, changes to animal migration patterns, and loss of hunting opportunities [47,48] collaborative solutions aimed at ensuring access to traditional foods are urgently needed. This will require partnerships between public health professionals, wildlife and environmental agencies, policymakers, and most importantly, communities, in order to uplift Indigenous food sovereignty.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This publication was supported by an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences from the National Institutes of Health under grant number 2P20GM103395.