here the documents for the assignment:
1-The assignment what to do
2-the guideline for the teacher
3-the articles I want to talk about: epidemiologic research
4-a little text about where I’m working
5-down here a copy of my book of the course
Assignment 2: Application of Epidemiology in Community Health Nursing | Value 20%
The purpose of this assignment is to provide you with an opportunity to demonstrate utilization of epidemiology research related to the health of the community.
Determine a specific population that you would like to have more information about, this should be the same population chosen for assignment 1 and the one you anticipate to use for assignment 3 and 4).
Use the Athabasca University Library Databases to search the epidemiology literature. Students will:
Find an applicable research article on your chosen topic.
Summarize and critique your chosen research article.
Consider how this research could be integrated into nursing practice.
Your work will achieve the maximum value/grade if it is succinct and insightful and clearly shows how you have applied the theory learned to a practical situation.
Your critique should include, but is not limited to the following:
What type of study design is it? Provide rationale.
Are the findings reported consistent with information/knowledge that you have? Do the reported relationships make sense?
If reported, how strong is the observed association?
Would you incorporate the findings of this study into your community health nursing practice (i.e., the health promotion program that you are planning) or recommend this study to others? Provide rationale.
What further research question(s) would you develop in relation to this study and/or your observations?
When submitting in the assignment dropbox include a copy of the article in PDF along with your completed assignment.
· Each element of the assignment guidelines is addressed.
· Ability to analyze, evaluate, create, and engage in critical inquiry is evident throughout.
· Adheres to APA 7th edition scholarly format – limit of 5 pages (excluding title, reference, and appendix pages).
Tips from the teacher:
I am sharing some Assignment 2 tips to help summarize the strategies for this assignment.
· Choosing an epidemiologic study: Epidemiology research studies characteristics include:
– what happens to the people:
-the incidence, causes and effects of diseases in populations, and trends and patterns
-investigates factors that determine the presence or absence of diseases and disorders
-reveals risk factors for a particular disease.
-helps us to understand how many people have a disease, if those numbers are changing, and how the disorder affects society and even the economy. As for knowing the correct type of study to choose, the major types of epidemiology study designs are randomized controlled trials, and nonexperimental study types including cohort, case control, cross-sectional and ecological studies.
· Present this assignment as if you think the audience has no background knowledge on the topic- so present as you would in a conference
· Introduction-state the purpose and strategy for the organization and set up the logical flow of the assignment so the reader is engaged and knows what you are doing. Include the purpose as related to the assignment guidelines- the purpose of this assignment is to demonstrate utilization of epidemiology research related to the health of your selected community.
· Provide the rationale for the choice of the epidemiologic research in relation to your community/population that you chose for all the assignments.
· Identify the specific population and selection of the study design with rationale.
· Critique the study with the focus of addressing the points in the marking criteria. i.e. -Critique – the strong points, the weak points, the gaps, omissions and assumptions.
· Are the findings reported consistent with information/knowledge that is available? Do the reported relationships make sense? If reported, how strong is the observed association? (Strength of associations should include statistics depending on the type of measurements in the study.)
· Specify-Would you incorporate the findings of this study into your community health nursing practice (i.e., the health promotion class that you may be planning) or recommend this study to others? Provide rationale. Why or why not? If so, how?
· Use critical inquiry-demonstrate the ability to read and comprehend what the researchers were doing and relate it back to the reader in your own words; paraphrasing is important, and you can think about challenges that incorporating the research might hold to your nursing practice. i.e. How does this study influence your practice as a nurse- i.e. shift how you care for the population or consider providing health promotion programs that are not in existence? Say how and why.
Assignment 2: Application of Epidemiology in Community Health Nursing
Student: Date Received:
Marked by: Marlyss Valiant
· The research article been summarized clearly and concisely
· The research article has been summarized to a(n) insufficent/beginning/satisfactory/good/excellent level
· The type of study design was articulated
· Rationale provided for study design
· The type of study design was articulated to a(n) insufficient/beginning /satisfactory /good/excellent level
· Rationale was provided to a(n) insufficient/ beginning /satisfactory/ good/excellent level
· Are the findings reported consistent with information/knowledge that is available?
· Do the reported relationships make sense?
· If reported, how strong is the observed association?
· The findings reported are explained in relation to previous information/knowledge to a(n) insufficient/beginning /satisfactory/ good/excellent level
· The reported relationships are articulated to a(n) insufficient/beginning /satisfactory/ good/excellent level
· The observed association are/are not discussed
Integration into Practice
· Explanation of integration into practice was evident
· Recommendations of this study to others was included
· Rationale provided (ie: strengths and limitations of research)
· Integration into practice was evident and explained to a(n) insufficient/beginning /satisfactory/good/excellent level
· Recommendations of this study to others was included/not included
· Rationale was provided and strengths and limitations of the research were included to a(n) insufficient/beginning /satisfactory /good/excellent level
· Were further research questions developed in relation to this study?
· Further research questions developed in relation to study to a(n) insufficient/ beginning/ satisfactory/ good/excellent level
· Ability to analyze, evaluate, create, and engage in critical inquiry is evident throughout
· All elements of the assignment guidelines are addressed
· The ability to analyze, evaluate, create, and engage in critical inquiry is performed to a(n) insufficient/beginning /satisfactory /good/excellent level
· All elements of the assignment guidelines are addressed to a(n) insufficient/beginning /satisfactory /good/excellent level
· Paper is within page limit (5 pages excluding title and reference page).
· Paper uses correct spelling and grammar
· The paper is informative and within page limit
· The paper is organized in a(n) insufficient/ beginning/satisfactory/good/excellent manner
· The paper has many/some/few/no errors in spelling, punctuation, grammar, sentence structure
Adheres to Current APA
· All elements of current APA edition have been met
· The paper was written in a scholarly format including: title page, introduction, phrasing, headings, and conclusion to a(n) insufficient/beginning/ satisfactory/good/ excellent level
· References are published within last 10 years and appropriate for the assignment—completed to a(n) insufficient/beginning/ satisfactory/good/excellent level
· APA current edition format for in text citation and reference page was used to a(n)
insufficient/ beginning /satisfactory /good/excellent level
If you have any questions or need further clarification, please contact me directly via course mail or telephone.
1Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
contaminant mixtures among
adults with type 2 diabetes in the
cree first nation communities of
Eeyou Istchee, canada
Aleksandra M. Zuk1*, Leonard J. S. tsuji1, evert Nieboer2, ian D. Martin1 & eric n. Liberda3
Type 2 diabetes mellitus (T2DM) disproportionately affects Indigenous populations. It is possible that
exposure to complex mixtures of environmental contaminants contribute to T2DM development. This
study examined the association between complex environmental contaminant mixtures and T2DM
among canadian indigenous communities from the Eeyou Istchee territory, Quebec, Canada. Using
data from the cross-sectional Multi-Community Environment-and-Health Study (2005–2009) Principal
Component Analysis (PCA) was used to reduce the dimensionality of the following contaminants:
9-polychlorinated biphenyl congeners; 7-organic pesticides; and 4-metal/metalloids. Following this
data reduction technique, we estimated T2DM prevalence ratios (PR) and 95% confidence intervals
using modified Poisson regression with robust error variance across derived principal components,
adjusting for a priori covariates. For both First Nation adult males (n = 303) and females (n = 419), factor
loadings showed dichlorodiphenyltrichloroethane (DDT) and lead (Pb) highly loaded on the second
principal component (PC) axis: DDT negatively loaded, and Pb positively loaded. T2DM was significantly
associated with PC-2 across all adjusted models. Because PCA produces orthogonal axes, increasing
PC-2 scores in the fully adjusted model for females and males showed (PR = 0.84; 95% CI 0.72, 0.98)
and (PR = 0.78; 95% CI 0.62, 0.98), respectively. This cross-sectional study suggests that our observed
association with T2DM is the result of DDT, and less likely the result of Pb exposure. Further, detectable
levels of DDT among individuals may possibly contribute to disease etiology.
Globally, diabetes continues to be a growing concern1. In Canada, Indigenous peoples are disproportionately
affected by diabetes mellitus (T2DM)2. The lifetime risk of diabetes is estimated to be 8 in 10 among First Nations
persons, and 5 in 10 among non-First Nations persons over 18 years of age2. However, the etiology and patho-
genesis of diabetes mellitus is yet to be fully understood. Exposure to environmental contaminants and the risk of
diabetes has received much research attention as persistent organochlorine pollutants (POPs) have been shown
to be associated with type 2 diabetes mellitus3–7.
In an extensive review of predominantly cross-sectional studies, Taylor et al.7 reports a positive association
between T2DM and some organochlorine pollutants (e.g., trans-nonachlor, dichlorodiphenyldichloroethylene
(DDE), polychlorinated biphenyls (PCBs)). Similarly, in an updated, and globally-relevant review, Kuo et al.8
confirmed the positive association between organochlorine compounds (OCs) and T2DM. More specifically, Pal
et al.9 observed higher plasma concentrations of OCs among persons with diabetes from First Nation commu-
nities in northern Canada. Similar to organic contaminants, long-term environmental exposure to toxic metals
and/or deficiency of essential metals may possibly also contribute to the development of diabetes10. The role of
various inorganic metals and metalloids on type 2 diabetes is complex. For example, Khan and Awan10 note that
poor glycemic control and diabetes may alter the level of various essential trace elements due to polyuria. Chen
1Health Studies, and the Department of Physical and Environmental Sciences, University of Toronto Scarborough,
Toronto, Ontario, Canada. 2Department of Biochemistry and Biomedical Sciences, McMaster University, Hamilton,
Ontario, Canada. 3School of Occupational and Public Health, Ryerson University, Toronto, Ontario, Canada. *email:
2Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
et al.11 suggested that some heavy metals may play a role in T2DM etiology by adversely affecting islet function.
Cross-sectional data show that metals such as cadmium may contribute to hypertriglyceridemia12. Commonly,
type 2 diabetes is complicated with dyslipidemia and other factors associated with metabolic conditions which
increase the risk of developing cardiovascular diseases and T2DM among Indigenous populations13. Further,
toxic metals may act as endocrine disrupters that contribute to adiposity14, which is a risk factor that exacerbates
metabolic and physiologic abnormalities associated with T2DM.
In Canada, Indigenous populations have a higher risk of developing T2DM and health-related complications
compared to general Canadian population2,15. Therefore, examining the associations between environmental con-
taminants and T2DM is a priority, especially among Indigenous communities, where higher body burdens of
complex mixtures exist. In this study, we examined the association between complex environmental contaminant
mixtures and prevalent type 2 diabetes status among Canadian Cree communities residing in the Eeyou Istchee
territory, in northern Quebec, Canada.
Materials and Methods
Data sources. The Eeyou Istchee territory, located in the James Bay Region of northern Quebec, Canada
consists of nine Cree communities (Fig. 1). The Nituuchischaayihtitaau Aschii – Multi-Community Environment-
and-Health Study aim was to provide assessment and surveillance among people of Eeyou Istchee. Eligibility for
enrollment in the study included any person living on reserve. The Environment-and-Health Study stratified
participants by age: children (0–7 years, and 8–14 years), adults (15–39 years, and 40 years and older). The main
objectives examined the health effects of lifestyle factors, (including diet), environmental contaminants exposure,
and environmental change on wildlife and aquatic ecosystems resulting from mining, forestry, and hydro-elec-
tric developments. In total, nine Cree communities were sampled. However, two of the nine communities were
for an initial pilot study on preliminary health assessments conducted between 2002–2005, and not part of the
analysis undertaken in this study. The remaining seven communities were studied between 2005–2009, which
focused on participant health measures including exposure to environmental contaminates. Due to the time
required to travel to between remote communities and collect all the necessary data, field data collection took
place a two-to-four-week period over spring/summer. Specifically, one community was sampled in 2005, two
in 2007, two in 2008, and two in 2009 (community names withheld at the request of the Cree Board of Health
and Social Services of James Bay. Full details about the Multi-Community Environment-and-Health Study are
provided16–19. As part of the Nituuchischaayihtitaau Aschii – Multi-Community Environment-and-Health Study,
trained research nurses were integral to the study data collection. Participants underwent a physical examina-
tion, completed health and dietary surveys, and provided tissue and blood samples for laboratory analysis. An
additional medical chart review was performed by a research nurse who had been involved in the clinical field
work to verify individual health-related information ascertained from health-questionnaires for all consenting
adults. Informed consent was obtained from all participants or their guardians in Cree, English, or French. The
Nituuchischaayihtitaau Aschii – Multi-Community Environment-and-Health Study was conducted in accordance
with relevant guidelines, regulations, and research agreements. All work conducted was approved by the research
Figure 1. Eeyou Istchee Territory, Quebec, Canada.
3Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
ethics boards of McGill University and Laval University, in partnership with the Cree Board of Health and Social
Services of James Bay and McMaster University.
Study population. In the 2005–2009 Environment-and-Health Study, 1750 participants were recruited.
Our analysis included the following adults over the age of 20 years of age who had: (1) medical-chart verified
T2DM diagnoses; (2) complete environmental contaminant exposure profiles; (3) undergone physical examina-
tion, completed interviewed health questionnaires, and underwent a phlebotomy blood draw were retained for
analyses. Medical chart reviews were conducted in only seven of the nine communities for self-reported medical
conditions. Therefore, adults who had not undergone medical chart review were excluded from analysis. Adults
with type 1 diabetes were also excluded. This resulted in a total of 722 cases, representing seven of the nine com-
munities from the Eeyou Istchee territory. A flow chart of the sample is presented in the Supplemental Fig. S1.
environmental contaminant analyses. Details concerning the analytical methods and related QA/QC
are provided in Liberda et al.20. Briefly, OCs were recovered from blood plasma using solid-phase extraction
and cleaned on a florisil columns prior to high resolution gas chromatography-mass spectrometry (HRGC-MS)
analysis. Limits of detection (LODs) were based on a signal-to-noise ratio of 3:1 also as previously reported.
Polychlorinated biphenyl (PCBs) congeners (CBs 99, 187, 183, 180, 170, 153, 128, 118, and 105), organic pes-
ticides (cis-Nonachlor, Dichlorodiphenyltrichloroethane [p,p’-DDT], Dichlorodiphenyldichloroethylene
[p,p’-DDE], Hexachlorobenzene [HCB], Mirex, oxy-chlordane, trans-Nonachlor) were all assessed for their
Whole blood samples were drawn from participants to measure concentrations of a selection of elements
(Lead [Pb], total mercury [Hg], cadmium [Cd], and selenium [Se]) and were kept frozen until analyzed at the
Institut National de Santé Publique du Québec (INSPQ) Human Toxicology Laboratory using inductively cou-
pled plasma mass spectrometry (ICP–MS) as detailed in Nieboer et al.21. All limits of detection (LODs) and the
analytical methods are also described therein.
Contaminants that were detected in less than 10% of the total participants were excluded from analyses.
Several methods exist for imputing missing values, however, the most common methods used for imputing sam-
ples with limits below the detection limit are using half the detection limit or one over the square root of two
multiplied by the detection limit22. Non-detections for all individuals’ contaminant body burdens were imputed
as half the detection limit as is recommended by the United States Environmental Protection Agency23. Due to
year-to-year analytical detection limit variation (i.e., lower detection limits due to better technology and stand-
ards), we utilized the highest detection limit through all years to prevent false differences owing the improvement
of the limit of detection overtime, and hence community.
outcome assessment. Health information was initially obtained through interviewer-administered ques-
tionnaires. Medical chart reviews were conducted for each consenting participant to confirm and gain additional
health information (i.e., medication use and medical history). All participants who were diagnosed with type 2
diabetes were confirmed through medical chart review.
Risk factor covariates. Covariate measures were ascertained through either self-report, by interviewer
administered health questionnaires, or via direct physical examination, which included a blood draw. Detailed
aspects of each are provided in Nieboer et al.19. Age was categorized into the following groups: 20–39, 40–59,
and ≥60 years. Educational attainment was self-reported and defined according to the following groups: com-
pleted less than high school, high school, and some or more college. The survey questionnaire collected infor-
mation on smoking habits, which classified participants as “current, former and never smoker.” Due to the low
prevalence of ‘never-smokers” in our analysis smoking status is a composite measure of “current and occasional
smokers” compared to “former or non-smokers. Standing height and body weight was measured at the time of
the physical examination. Body mass index (BMI) was calculated according to weight in kilograms (kg) divided
by height measures (meters squared, m2). Total lipids concentrations were determined using methods described
by Rylander et al.24.
Statistical methods. Statistical analysis. Descriptive statistics were calculated for all contaminant con-
centrations and covariates, stratified by sex and diagnosis of T2DM. Continuous variables were reported as
means ± standard deviations (SD) or geometric means, where appropriate. Categorical data are reported as fre-
quencies and percentages. Using SAS PROC GENMOD procedures, we separately estimated adjusted prevalence
ratios (PR) using modified Poisson regression with robust error variance25,26. Multivariable models examined the
association between T2DM (a non-rare binary outcome) and derived principal components (PCs) adjusted for
the following a priori covariates: age, plasma lipid concentrations, BMI, smoking status, and education. Overall,
the following covariates were missing among females and males, respectively; education: 1.7% (n = 7) and 2.97%
(n = 9); BMI: 1.9% (n = 8) and 5.3% (n = 16) and; smoking status: 1.4% (n = 6) and 2.97% (n = 9). Consequently,
numbers of individuals in subsequent regression analyses were reduced slightly depending upon the number of
valid observed covariates. Statistical analyses were carried out using SAS v9.4 (SAS Institute, Inc., Cary, NC) and
all figures were generated using R (version 3.5.2; Vienna, Austria).
Principal component analysis. Principal component analysis (PCA) was used to transform an initial set of 21
plasma or whole blood contaminant variables (i.e., PCB congeners, organic pesticides, and metals/metalloid) into
a reduced number of uncorrelated (i.e., orthogonal) predictor variables by maximizing the variance of the original
variables into derived fewer dimensions or principal components (PC)27,28. We used the correlation matrix of con-
taminant variables as the input matrix for PCA, and put all original variables on a common scale. Components
with eigenvalues exceeding 1.0 were retained and used to define independent summary axes. Therefore, the first
4Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
principal component axes (i.e., PC-1) will account for the largest variance in the data, and any subsequent PCs
(i.e., orthogonal to the first) will account for a portion of the variance not accounted for in the preceding compo-
nent. The new derived PCs (i.e., scores) are linear combinations of all original variables. Values, or scores, for indi-
viduals on these new PC variables are measures of shared exposure to the original contaminant concentrations.
Prior to the PCA, contaminant concentrations were log10-transformed (variate + 1), improving normality of the
distribution29,30. Separate PCAs were performed for female and male cohorts, owing to differential prevalence
of T2DM between sexes and differing levels of exposure for females and males31. Absolute component loadings
of 0.50 or greater were identified as important for a given principal component. Thus, signs of loadings are arbi-
trary, only the relative magnitude and patterns are meaningful32. Separate-sex principal component (PC) scores
summarized new, synthetic measures of contaminant burdens for both males and females. These uncorrelated PC
score variables were then used as independent predictors in the regression analysis of T2DM.
Sensitivity analysis. Based on findings from the regression models, we examined the frequency of detection
of two variables (i.e., DDT and Pb) by diagnosed T2DM status in contingency analysis. Adjusted Standardized
Residuals (ARS) of contaminant levels measured above or below the limit of detection were calculated for T2DM.
An association between the frequency of detectability of contaminants, above or below the limit of detection with
T2DM status was explored by examining overall chi-square significance and ASRs greater than (|1.96|) in a 2 × 2
contingency table. Complete-case analysis was also performed as a sensitivity check, which found no appreciable
difference in results using the same modified Poisson regression models33.
ethics approval and consent to participate. All work conducted was approved by the research eth-
ics boards of McGill University and Laval University, in partnership with the Cree Board of Health and Social
Services of James Bay and McMaster University. Informed consent was obtained from all participants or their
guardians in Cree, English, or French.
Descriptive results. Summary statistics of Cree population data for demographic, risk factors variables and
contaminants are presented in Table 1. In total, there were 722 participants, 419 females, and 303 males. The prev-
alence of T2DM among females and males was 23% (n = 95) and 16.5% (n = 50), respectively.
The mean age among participants with T2DM was 47.9 years and 56.2 years for females and males with
diagnosed type 2 diabetes, respectively. Among female respondents, 24.2% self-reported attaining some or
more college education whereas among males, 17% had attained some form of college education. Among adults
with diabetes, body mass index (BMI) at the time of examination was higher for females (39 kg/m2) than males
(34.5 kg/m2). As well, there was a two-fold higher prevalence of self-reported smoking status (i.e., current and
occasional compared to former or never) among females with T2DM. The total mean lipid concentrations also
differed among females and males among adults with diagnosed T2DM, 6.4 g/L and 5.9 g/L, respectively.
Contaminant principal component analysis (PCA) loadings. Sex-stratified contaminant PC loadings
are shown in Fig. 2. Among females, eigenvalues greater than 1 were found for the first two components. PC-1
explained 73% of the total variance in the original log transformed concentrations, which for increasing PCA
scores, resulted in high positive loadings for PCBs, organochlorines, and Hg. On the second axis, PC-2 accounted
for 5% of the variation, showing that DDT had a negative loading relative to the positive loading of Pb on PC-2
among females. However, for decreasing PCA scores, DDT loadings are interpreted as positive relative to Pb,
which is interpreted as having negative loadings.
Among males, contaminant PCA revealed three orthogonal axes, which explained 72%, 6%, and 5%, of the
variation for PC-1, PC-2, and PC-3, respectively. Similar to females, PC-1 was highly positively loaded by PCBs,
most organochlorines, and Hg. DDT had a strong negative loading, but high positive loading for lead Pb, and a
moderate loading for Hg for the second PC axis. Lastly, cadmium and selenium loaded positively on the third PC
axis for males.
PCA biplot of orthogonal PC axes overlaid with T2DM status among Indigenous Cree adults in the Eeyou
Istchee territory (Fig. 3).
Main effects of the association between PCA scores and type 2 diabetes. Multivariable modified
Poisson regression analyses are presented in Tables 2 and 3, which investigates the relationship between prevalent
T2DM and the extracted orthogonal principal components of contaminant exposure, for both adult females and
males, respectively. Among adult females, the prevalence ratio for PC-2 (but not for PC-1), was significantly associ-
ated with T2DM across all adjusted models. After fully adjusting for covariates, the final model for females shows,
PC-2 (PR = 0.84; 95% CI: 0.72, 0.97) was significantly associated with T2DM. For increasing PC-2 scores, DDT neg-
atively loaded, and Pb was positively loaded, on the second axes. Therefore, as PC-2 axis scores increase (i.e., DDT
loadings decreases and Pb loadings increases), resulting in a PR significantly less than 1. Conversely, as PC-2 axis
score decreases (i.e., DDT loadings increase and Pb loadings decrease), PC-2 is significantly associated with preva-
lent T2DM (PR = 1.19; 1.03, 1.38) among females. Similarly, among adult males, PC-2 (explaining 5% of variation)
also was significantly associated with prevalent T2DM across all adjusted models. PC-2 was shown to have a strong
positive and negative loading for Pb and DDT, respectively. In the fully adjusted model, PC-2 was significantly asso-
ciated with T2DM (RR = 0.78, 95%; CI: 0.62–0.98). As above, since a decrease in PC-2 score indicates an increase in
DDT loadings, resulting in a similar significant association between (PR = 1.27; 95% CI: 1.02–1.60).
Sensitivity analysis results. Based on the significant associations observed in PC-2, we further explored
the role of DDT and Pb on T2DM. Rather than examining only concentration values, we inspected the fre-
quency of detection for DDT and Pb on T2DM status by contingency analysis. An examination of the adjusted
5Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
Sex stratified Total population (n = 722) % ≤ LODa
Type 2 Diabetes Status
N (%); or Mean ± SD N (%); or Mean ± SD
Sex (n, %)
Female 419 (58%) 95 (23%) 324 (77%)
Male 303 (42%) 50 (16.5%) 253 (83.5%)
Female 47.9 ± 14.7 38.5 (13.9)
Male 56.2 ± 14.9 40.6 ± 14.6
Less than High school Female 412 34 (37.4%) 55 (17.1%)
Some or completed High school 35 (38.5%) 181 (56.4%)
Some or completed College or higher (R) 22 (24.2%) 85 (26.5%)
Less than High school Male 294 19 (40.4%) 47 (19.0%)
High school 20 (42.6%) 156 (63.2%)
Some College or higher 8 (17.0%) 44 (17.8%)
Female 411 39 ± 8.6 34.5 ± 6.5
Male 287 34.5 ± 5.8 31.7 ± 5.6
Smoking status, current/occasional
smoker compared to former/never (R)
Female 413 28 (30.8%) 176 (54.7%)
Male 294 6 (12.8%) 127 (51.4)
Total lipids (g/L)b
Female 419 6.4 ± 1.6 5.8 ± 1.1
Male 302 5.9 ± 1.1 6.5 ± 1.9
Female 40.6% 0.116 ± 4.061 0.044 ± 3.470
Male 32.3% 0.108 ± 3.333 0.054 ± 3.438
Female 49.2% 0.058 ± 3.331 0.025 ± 2.563
Male 47.2% 0.046 ± 2.832 3.438 ± 2.345
Female 15.7% 0.262 ± 4.583 0.074 ± 4.508
Male 10.9% 0.236 ± 3.930 0.086 ± 3.991
Female 83.3% 0.019 ± 1.639 0.015 ± 1.331
Male 77.6% 0.017 ± 1.490 0.016 ± 1.403
Female 5.0% 0.565 ± 4.641 0.175 ± 5.077
Male 1.6% 0.678 ± 3.854 0.275 ± 4.404
Female 1.4% 1.204 ± 4.877 0.376 ± 5.586
Male 0.3% 1.695 ± 4.185 0.666 ± 4.686
Female 16.9% 0.259 ± 4.592 0.093 ± 4.751
Male 6.6% 0.407 ± 4.264 0.168 ± 4.485
Female 2.63% 0.887 ± 5.069 0.290 ± 5.677
Male 1.3% 1.483 ± 4.583 0.572 ± 4.998
Female 30.1% 0.112 ± 3.919 0.044 ± 3.442
Male 17.8% 0.131 ± 3.570 0.063 ± 3.520
Female 13.4% 0.348 ± 4.863 0.117 ± 5.077
Male 6.3% 0.528 ± 4.293 0.212 ± 4.695
Female 50.6% 0.052 ± 2.945 0.024 ± 2.405
Male 38.0% 0.064 ± 3.074 0.031 ± 2.628
Female 0.7% 2.958 ± 3.305 1.043 ± 3.746
Male 0.3% 2.993 ± 2.911 1.386 ± 2.866
Female 88.3% 0.035 ± 1.811 0.027 ± 1.326
Male 90.0% 0.031 ± 1.626 0.027 ± 1.372
Female 31.5% 0.120 ± 3.054 0.055 ± 2.878
Male 21.8% 0.120 ± 2.476 0.071 ± 2.666
Female 25.1% 0.161 ± 4.620 0.062 ± 4.432
Male 14.2% 0.266 ± 4.681 0.113 ± 4.934
Female 30.8% 0.085 ± 3.249 0.035 ± 2.862
Male 18.5% 0.104 ± 3.103 0.048 ± 2.851
6Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
standardized residuals showed that adults with T2DM were significantly more often than expected to have detect-
able levels of DDT (females: 55%, ASR 5.8, p < 0.001; males: 30%, ASR 2.1, chi-square p = 0.036) compared to
only 18.4% (ASR −5.8) and 15% (ASR −2.1) with levels below detection for DDT among females and males,
respectively. Lead (Pb) detection frequencies between T2DM states were not found to be statistically significant.
Single-pollutant models for DDT and Pb levels are also provided in the Supplemental Material to aid interpreting
of the PCA results (Supplementary Material, Table S4).
In this cross-sectional Multi-Community Environment-and-Health Study among adult Indigenous peoples, we
show that DDT and Pb load oppositely to each other on the second component axis indicating differential expo-
sures. PCA has distributed the variation of the contaminants on what appears to be similar groups of lipophilic
or hydrophilic compounds and/or their sources. As DDT and Pb are considered together on the second axis,
Sex stratified Total population (n = 722) % ≤ LODa
Type 2 Diabetes Status
N (%); or Mean ± SD N (%); or Mean ± SD
Female 20.8% 0.150 ± 3.527 0.052 ± 3.510
Male 9.2% 0.212 ± 3.456 0.084 ± 3.422
Cadmium, Cd (nmol /L)
Female 40.1% 5.563 ± 2.729 8.849 ± 2.782
Male 51.2% 3.926 ± 2.338 8.192 ± 3.104
Total mercury, Hg (nmol /L)
Female 15.7% 28.275 ± 3.798 14.862 ± 3.809
Male 13.2% 44.208 ± 3.383 20.433 ± 3.990
Lead, Pb (µmol /L)
Female 29.1% 0.131 ± 2.848 0.119 ± 3.003
Male 9.2% 0.164 ± 2.680 0.196 ± 2.528
Selenium, Se (µmol /L)
Female 0.2% 2.209 ± 1.218 2.118 ± 1.157
Male 0.3% 2.244 ± 1.151 2.235 ± 1.140
Table 1. Participant characteristics according to sex and type 2 diabetes status: results among
Nituuchischaayihtitaau Aschii – Multi-Community Environment-and-Health Study (2005–2009). Missing
values among adult females; Education (n = 7, 1.7%); BMI (n = 8, 1.9%); Smoking status (n = 6, 1.4%).
Missing values among adult males; Education (n = 9, 2.97%); BMI (n = 16, 5.3%); Smoking status (n = 9,
2.97%). Abbreviations: N, frequency value; %, percentage; BMI, Body mass index; R, reference category;
PCB, Polychlorinated biphenyl congeners; p,p’-DDT, Dichlorodiphenyltrichloroethane; p,p’-DDT,
Dichlorodiphenyldichloroethylene; aPercentage of contaminants below the level of detection (LOD). bTotal lipid
concentrations were determined using methods described by Rylander et al. 2012. cPresented are geometric
mean ± standard deviation (SD).
Figure 2. Principal Component (PC) Loadings for Males and Females
7Scientific RepoRtS | (2019) 9:15909 | https://doi.org/10.1038/s41598-019-52200-x
decreasing PC-2 axes scores manifest as increasing DDT and decreasing Pb loadings. Positive DDT loadings were
associated with type 2 diabetes with decreasing axes scores for both females and males.
Experimentally, it has been shown that perinatal DDT exposure in mice may contribute to insulin resist-
ance and metabolic syndrome in adult female offspring34. A separate experiment in male mice exposed to DDT
reported significant reductions in glucose tolerance and pancreatic activity35. More recently, a systematic review and
meta-analysis of observational human studies reported a significant overall increased risk between DDT and type
2 diabetes (odds ratio 1.79 [95% CI 1.31, 2.4]36. Additionally, exposure to other organochlorine pesticides were also
shown to be associated with T2DM36. However, specific sex-related differences were not examined in the systematic
review, and the responsible mechanisms of action for sex-dependent findings remains yet to be elucidated. One
possible mechanism may be due to DDT’s role as an estrogen agonist and an androgen antagonist34,37.
This is the first study to examine complex body burdens using a reduction method technique on prevalent
type 2 diabetes among Indigenous peoples in the Eeyou Istchee territory of northern, Quebec. The combined
biological adverse effects of body burden are a concern to human health, particularly among areas of north-
ern Canada, where environmental contaminants are reported to be present, often at greater concentrations38–40.
Figure 3. PCA Biplot of Principal Component (PC) Axes and Type 2 Diabetes Status among …
More about Chisasibi
Chisasibi is the biggest Cree Community in James Bay. There is around 7000 people.I am a nurse in this beautiful community for 2 years. I’m working at the emergency of Chisasibi hospital.
Chisasi is only 40 years. Before they used to live in an island at 15 km from Chisasibi but the Gouvernment of Quebec moved them to built hydroelectric central. They were living there for more many many years. It was the land of ancestors. In northern Quebec there is a lot of mines, dozen of hydroelectric dam so the Gouvernment kind of destroy the forest, there land and bring up lot of minerals of toxic gas and minerals. All these things had bad influence on the water they drink, animals they hunt and eat and change their life
There is 9 Cree communities in James Bay (Eeyou Istchee) and I’m working to Chisasibi; the one where hydro-quebec and The government of Quebec had the most impact (bad impact)
I see a lot of patient with diabetes and uncontrol diabetes.
They often don’t take their medication or take it when they feel like it.
This research is related directly to the community where I’m working so I can talk about this research and introduce it when I’m doing teaching about diabetes. I think they will be also interesting to know that all the contaminant can be a cause of diabetes and not also the bad habits.