The Republic of the Marshall Islands (RMI) has been affected by marine pollution from militarization and urbanization. To address concerns raised by the Marshall Islands Marine Resources Authority, this study examined concentrations of dissolved contaminants in reef and pelagic fishes in the RMI and assessed potential associated risks. Metals, organochlorine pesticides, polychlorinated biphenyls (PCBs), and polycyclic aromatic hydrocarbons (PAHs) were examined in reef and pelagic fishes from six atolls: Kwajalein, Majuro, Jaluit, Utirik, Rongelap, and Wotje. Clear trophic patterns emerged for metals. Total arsenic was highest in higher trophic level reef fishes, particularly in the camouflage grouper (Epinephelus polyphekadion) (>100 μg g-1 total As), but inorganic arsenic was negligible in higher trophic levels and showed an inverse trend with the highest percentages present in parrotfishes and herbivores. Copper and mercury were elevated in higher trophic level reef and pelagic fishes, respectively, and the maximum mercury concentrations (6.45 μg g-1 in Gymnosarda unicolor) were among the highest reported in the Pacific. Conversely, cadmium and lead were highest in lower trophic levels, like surgeonfishes and parrotfishes. PCBs were more clearly linked to locations and were highest at two atolls with military history (Kwajalein and Jaluit) (>U.S. EPA Screening Value of 2.5 ppb). PAHs were ubiquitous across taxa (detected in 97% of samples), but the highest concentrations were in lower trophic levels. Organochlorine pesticides were detected at very low concentrations that do not likely pose a risk. We compare concentrations to established thresholds for human health and find that - for specific locations and species - contaminant concentrations may pose a risk to fish and other marine taxa, as well as human consumers. This study provides baseline information that aids the development of marine conservation and public health recommendations and addresses a data gap that persists for marine pollution throughout the Pacific Islands.

Abstract Background Genetics plays an important role in the development of type-2 diabetes (T2D). Polygenic risk scores (PRS) are increasingly used to quantify genetic risk of T2D in epidemiological studies. These scores, when integrated into analyses of modifiable lifestyle factors, may improve understanding of T2D etiology, as the strength of association with T2D and some lifestyle or demographic factors may vary according to genetic predisposition. Methods We examined PRS-lifestyle factor interactions on T2D with data from the United States Health and Retirement Study (HRS), a prospective longitudinal cohort of older adults (≥50 at baseline). HRS contains nationally representative samples of Black and White Americans with pre-calculated PRS for T2D (N = 14,001). Covariates included sex, education, BMI, smoking, alcohol, and physical activity. Predicted prevalence and incidence of T2D were calculated with logistic regression models. Nonparametric bootstrap method was performed to calculate differences in T2D prevalence and incidence by PRS percentiles and interaction variables. Results Significant interaction (p_interaction=0.0096) was detected between PRS and physical activity among Whites only. In those with the lowest decile of PRS, T2D prevalence was similar (∼10%) for those reporting no physical activity compared to low or moderate activity. In those with the top decile of PRS, lower T2D prevalence (17%, 95%CI:14.8,19.6) was observed among those with moderate compared to no activity (24%, 95%CI:20.4,27.5). Incident T2D in Whites followed a similar pattern (p_interaction=0.0194). Among Black participants, no significant interaction with any lifestyle variables was detected. Conclusions Interaction of different genetic risk profiles with lifestyle factors may inform understanding of why certain inventions are more or less effective in different groups of people, potentially improving clinical and prevention interventions. Key messages Protection conferred by physical activity on T2D varied by underlying genetic risk. Gene-environment interaction studies provide insights on why lifestyle factors vary in their associations with T2D.

Women experience worse physical function and greater physical decline than men at similar ages. These sex differences are heterogeneous across settings and plausibly linked to gender inequality, with evidence of increasing disadvantage for women in increasingly iniquitous societies. As described in “Age at natural menopause and physical function in older women from Albania, Brazil, Colombia and Canada: A life-course perspective” [Velez et al., 2019] we assessed the association between age at natural menopause (ANM) and objectives markers of physical function (i.e., gait speed and grip strength) in older women from the International Mobility in Aging Study (IMIAS). For all sites combined, women with ANM ≥55 had higher gait speed than those with ANM 50–54. Women with ANM <40 had significantly lower grip strength compared with all other groups. In this article, we describe the region-specific associations between ANM, gait speed, and grip strength in 775 women aged 65–74, from the Southeastern European site (Tirana, Albania), Latin American sites (Manizales, Colombia and Natal, Brazil), and Canadian sites (Kingston, Ontario and Saint-Hyacinthe, Quebec). In region-specific analyses, ANM was associated with grip strength in Albania and Latin America and with gait speed in Albania only. No associations were observed in Canada.

Investigate the association of age at first childbirth (AFB) with Framingham Laboratory Risk Scores (FRS) for CVD. Multivariate linear regression analyses were conducted with baseline data from a multisite study of community-dwelling older female adults (N=868). Women with adolescent childbirth (AFB <19) were compared to those with AFB of 19–24, 25–29, 30+ and nulliparous women. Lifetime parity was considered. Adolescent childbirth was positively associated with FRS. Women with adolescent childbirths had 6.8 points higher mean FRS (95%CI 3.8–9.9) than women in the lowest AFB risk-group (25-29yr). Women with adolescent childbirths also had significantly higher scores compared to women with AFB 19–24, AFB 30+ and nulliparous women (β4.4, β4.3, β7.1, respectively). No association with parity was observed. To our knowledge, this is the first study to document a positive association between young AFB and greater FRS in later life. Young AFB in women may contribute to later life CVD risk.

Examine risks clusters for metabolic syndrome (MetS) and its components (hypertension, abdominal obesity, dyslipidemia, insulin resistance) in highly distinct societies (St. Hyacinthe and Kingston, Canada; Tirana, Albania; Manizales, Colombia; Natal, Brazil). With baseline data from a multisite study (N=1587), we employed model-based recursive partitioning to cluster subjects into age-adjusted groups according to: sex, living arrangements, smoking, childhood adversity, education, income and work status. Women were separately analyzed to investigate if age at first childbirth and parity contributed to clustering. MetS was observed in 43% of participants. The highest predicted prevalence was among women from middle-income sites (58–69%). For all components, except high density lipoprotein, study site was the primary partitioning variable, followed by sex. MetS was particularly prevalent among middle-income women, but female reproductive characteristics were not explanatory variables. Individual risk characteristics appeared to so strongly conditioned by study site that they could not be captured in the models.

We used longitudinal data from the International Mobility in Aging Study (n=1,355) to explore pathways between FOF measured by the Falls Efficacy Scale-International and incidence of functional limitation in old age. Outcomes were incidence cases of: 1) self-reported difficulty climbing a flight of stairs or walking 400 metres (mobility limitation), and 2) scoring <9 in the Short Physical Performance Battery [SPPB] (physical performance limitation). The potential pathways (gait speed, physical activity, balance, depression, and grip strength) were selected based on available theories and examined using mediation analysis adjusting for age, sex, site, cognition, and comorbidities. Total and direct effects of FOF on both outcomes were significant in all models. After adjustment, the relationship between FOF and mobility limitation was mediated only through gait speed. In women, no mediator linked FOF to SPPB; however, in men depression was a significant mediator. FOF has a strong direct effect on disability in older adults. Depression and gait speed partially account for incidence of disability.

Review of the literature on age at first childbirth (AFB) and CVD, including hypothesized causal pathways. PubMed and Web of Science databases were searched for observational studies published between 1980–2015. Data were extracted using a pre-defined list. 19 publications, reporting on 30 associations, were included in the review, all from high-income settings. Eight of 13 studies that defined young AFB as ≤20 found a positive association between early AFB and CVD, while two reported a positive association between later first childbirth and CVD. Substantial methodological limitations were observed related to: operationalization of risk categories, choice of reference category, sample size, follow-up time and statistical over-adjustment. Only two studies explicitly reported employing a theoretical framework when investigating the relationship. Early AFB is possibly related to CVD. Better conceptualization of causal pathways linking reproductive timing to CVD risk is needed.

Fear of falling (FOF) is highly prevalent among older adults. We examine the relationship between fear of falling and average minutes walking per day among community dwelling older adults. Statistically significant differences in walking were observed between site locations, sex, age, educational attainment, functional fitness levels, fall history, and FOF. Participants with low functional fitness walk less on average compared to high functional fitness(β: -9.49, 95%CI: -14.74, -4.23). Participants with high FOF walk less on average compared to low FOF(β: -10.11, 95%CI: -15.13, -5.10). Fall history is not associated with walking in multivariate analyses. When stratified by functional fitness, among those with high functional fitness, those with high FOF walk less than those with low FOF(β: -10.90, 95%CI: -16.67, -5.14). The relationship is not observed among those with low functional fitness. FOF may be a barrier to walking for those with high functional fitness.

A substantial body of research investigates potentially-modifiable individual risk factors for cardiovascular (CVD) diseases including tobacco use, physical inactivity, and clustered metabolic risk factors. In older adults, however, many of these “modifiable” factors are conditioned by exposure to a lifetime of socially- and culturally-established determinants. Some of the most intractable are tied to gender norms, which strongly influence reproductive timing and frequency and are among the most life-altering events in the human experience, both physiologically (women) and socially. To date, numerous studies investigate female lifetime parity and CVD, as well as adolescent childbirth and all-cause mortality. Many observe robust associations and attribute the findings to the physiological consequences of pregnancy. Almost all of this research takes place in high-income settings and few interrogate the complex constellation of influences on reproduction and its long-term health consequences. Very few look at male populations. In this symposium, we present findings from middle and high-income populations investigating age at first childbirth (AFB) and number of children on cardiometabolic diseases. We begin with results of a systematic review investigating associations between AFB and CVD. It highlights conceptual and methodological gaps in previous approaches to the topic. The next two studies present results from an international cohort of community-dwelling older adults with sites in Canada and middle-income countries, each characterized by divergent gender norms and human develop indices. Finally, we present results on lifetime number of children in men and women in the United States and associations with heart disease and stroke.

Examine the association of number of lifetime children with heart disease and stroke in United States older adults. We analyzed the 2012 wave of the Health and Retirement Study, a representative study with six cohorts born between 1890–1959. Participants self-reported number of children and doctor-diagnosed heart disease/stroke. In two sex-stratified logistic regression models (n≈19000), we included: cohort, census region, race/ethnicity, and then education. Compared to nulliparous women, those with 3–4, 5–6, and 7+ children had 1.21, 1.26, & 1.46 the odds of heart disease (p<0.05). With education, parity was no longer associated to heart disease. Similar results were observed for stroke. In men, compared to no children, having 3–4 and 5–6 children was significantly positively associated with heart disease (OR:1.20 & 1.56). For stroke, an association was only observed for 5–6 children. Adding education did not change the results. Findings suggest social pathways linking parenthood to cardiovascular health.