Adults that were interested in the study were screened and enrolled in the study if all inclusion criteria were met. Baseline data collection was then scheduled for adults and children. After baseline data collection, study parents were informed which treatment group they were in. This intervention component sought to increase vegetable intake among children in the ECE programs. We collaborated with the Leadership Committee and Osage Nation ECE program site managers, teachers, cooks, and staff at the tribal farm to determine which produce was of interest and available for the study. Since the climate in Oklahoma is unpredictable during the intervention months, ranging from cold, icy winters to a warm, wet spring, we utilized the farm and supplemented any produce they were unable to grow from a local supermarket. More details regarding the tribal farm are described elsewhere. The farm-to-school nutrition and gardening curriculum was adapted for NA children from two curricula: Early Sprouts; and Watch Me Grow. More details regarding how we adapted the curriculum are described elsewhere. The FRESH farm to-school nutrition and gardening curriculum included knowledge, reading, gardening, and indoor and outdoor sensory activities, comprised of three themes that were taught for five weeks each: Harvest ; Explore ; and Sprout . The focus of the curriculum was on six target vegetables: tomatoes, bell peppers, spinach, squash, butter beans, and carrots. The weekly curriculum for each theme included a reading activity , indoor and outdoor sensory activity, and cooking activity, which included a take-home recipe kit. The intended duration for each activity varied: 5–30 min for reading activities,cannabis curing up to 60 min for indoor and outdoor sensory activities, and up to 75 min for cooking activities. All weekly lessons were assembled in a handbook and distributed to intervention teachers for implementation.

Garden beds for the outdoor sensory activities and cooking activities were built and maintained by the Harvest Land farm staff at Osage Nation. All intervention children also took home a family recipe kit, including ingredients and a recipe to repeat the cooking activity with their family to increase exposure to the vegetables. Although the FRESH study did not directly intervene upon dietary intake of parents, we did include a passive online and in-person hybrid parent curriculum, adapted from the Choose Health LA’s Healthy Parenting Workshops , with components from the First Nations Development Institute’s Food Sovereignty Assessment Tool and the Grassroots International’s Food for Thought and Action curriculum. The online curriculum comprised of 12 short video modules focused on providing parents with strategies to support their children in eating healthier foods and included healthy lifestyle education and healthy parenting practices. The in-person component included three in-person family night workshops that focused on food sovereignty in the community and community capacity building for health. The last component included menu modifications at the ECE programs. Further description of the menu modifications from the FRESH study can be found elsewhere. In short, fresh vegetables from Osage Nation’s Harvest Land farm were harvested and delivered to the ECEs to be incorporated into the ECE menus. The menus were modified to achieve best practices established by the Child and Adult Care Food Program , which included more vegetables and fruits as snacks, replacing whole grains for refined grains, reducing fried foods, and removing sugar-sweetened beverages. The menus included the six target vegetables from the farm-to-school curriculum provided from the Osage Nation farm two times weekly and provided to the children in meals or offered as snacks within each menu cycle.

Dietary intake for children was assessed by measuring the consumption of the six target vegetables in the FRESH farm-to-school curriculum using the weighed plate waste method to assess objective levels of vegetable consumption. During the plate waste administration, we also assessed preference, or willingness to try, target vegetables. Trained researchers rated each child’s interaction with each target vegetable using a five-point checklist to measure observed willingness to try. The rating options were: Did not remove vegetable from box, removed food, but did not bring to nose/mouth, removed food and brought to nose/mouth, but did not put food in mouth, put food in mouth, but did not swallow food , put food in mouth and swallowed. More information regarding the child food consumption methods is provided elsewhere . Dietary intake for adults was evaluated using the National Cancer Institute’s Automated Self-Administered 24 h Recall. Recalls were obtained either in-person or via phone by trained university staff. Recall data were used to estimate mean intake of total energy , total sugar , total fats , total fruits , and total vegetables between intervention and control groups. We also used the 7-item Fruit and Vegetables Behavior Checklist to assess combined fruit and vegetable intake in cups per day.The aims of the FRESH study were to improve dietary intake , BMI, systolic blood pressure , health status, and food insecurity among NA families. FRESH is one of the first comprehensive multi-component, multi-level CBPR studies to use a farm-to-school and parent curricula to build community capacity and reduce obesity risk among NA families attending ECE programs. Although the FRESH study did not improve BMI or other secondary outcomes among children, there were significant increases in vegetable intake. Previous studies looking at vegetable intake and BMI improvement among children showed varied results. Some randomized controlled trials found that nutrition interventions that resulted in significant increase of vegetable intake also found a decrease in BMI, whereas others found no change among BMI; the latter finding is consistent with our study.

One nutrition and gardening intervention that implemented a randomized controlled trial at school found that BMI significantly improved in the intervention group compared to the controls; however, this study involved older children and a longer intervention period. At follow-up, regarding the willingness to try scale, we found significant increases in scores for tomatoes for both treatment groups and increases in scores for beans in the intervention group. Our findings are similar to the Nutrition Matters! curriculum, which found a significant increase in willingness to try scores in three fruits and vegetables among the nutrition and gardening group,curing cannabis and is consistent with a previous study that found that repeated exposures to vegetables led to an increase in children’s willingness to try target vegetables. Among the adults, the FRESH study did not improve vegetable intake, BMI, blood pressure, or food security. However, in the intervention group, there was a trend toward increased fruit and vegetable intake from baseline to post-intervention. At follow-up, total sugar intake and total energy significantly improved among the intervention group compared to the controls. Our results differ from a previous online nutritional intervention among NA participants that found an increase in vegetable intake. However, as the FRESH intervention focused primarily on children with a secondary component including the parents, significant results were not expected. Our study had several strengths. This study used a randomized controlled design, able to compare intervention and control groups. Although the study was under powered, we still found a trend towards increased vegetable intake in intervention adults. In addition, we used objective measures of vegetable consumption in children rather than a dietary recall, providing a more comprehensive dietary intake. This was noted as a suggestion among authors in a systematic review on garden-based interventions among preschoolers. Another strength of our study is providing children with repeat taste exposure of vegetables, which past research has shown to be effective in increasing intake. Furthermore, our study focused on providing the ECE menus with local fresh vegetables, addressing the need for studies that intervene in the social determinants of health. The Osage Nation is a reservation that has limited access to healthy and fresh foods, and this study built upon and strengthened local resources by facilitating the process for supplying the ECEs with the local produce. The limitations of this study include the challenge of implementing some of the dietary measures among children.

For example, there were negative values on plate waste vegetables, which were determined to likely be the result of water condensation. In addition, we did not directly intervene with parents, which contributed to low participation rates in the study’s online component of the parent curriculum. Only 56% of parents attended the first week of the online curriculum and 12% attended the final week. However, in contrast, participation in the in-person component of the parent/family intervention was nearly twice as high as the online participation, although it also decreased as the intervention continued. Since the original implementation of the FRESH study, the Osage Nation Harvest Land farm built upon the lessons from the study process and findings to expand its produce to a greater number of tribal programs and services. The Harvest Land farm now features commercial-grade aquaponics systems and eight new state-of-the-art greenhouses, which are able to grow various vegetables year-round. The farm is now in the process of developing a tribally specific community-supported agriculture program, which aims to increase access and intake of fresh produce. The principal goal of environmental epidemiology is to characterize how environmental factors affect human health. Specifically, environmental epidemiologists seek to quantify a dose-response relationship between the level of a hazardous agent in the environment and the severity of its health impact on a population. To this end, epidemiologists have generally classified potential exposures to environmental agents on the basis of self-reported information. However, self-reported exposure surrogates are generally qualitative and they may not accurately reflect true environmental exposures. When the discrepancy between estimated and true exposure levels is substantial, the true relationship between exposure and disease will be obscured. As such, the development of quantitative and objective measures of exposure is a critical aspect of environmental epidemiology. Recently, investigators have considered estimating exposures to indoor contaminants using toxicant levels in residential-dust samples, because dust measurements are quantitative and objective. Although, in practice, few epidemiological studies have employed estimates of exposure using dust samples, this dissertation will show that concentrations of chemical contaminants in residential dust can be useful surrogates for indoor chemical exposures This dissertation will focus on three chemical classes which have been associated with either childhood leukemia in the Northern California Childhood Leukemia Study  or with developmental effects in other studies , namely, polybrominated diphenyl ethers , polychlorinated biphenyls , and polycyclic aromatic hydrocarbons , as well as nicotine . All of these chemicals are ubiquitous contaminants in residential dust due to their many indoor sources. Nicotine is a specific marker of cigarette smoke; PBDEs have been used as chemical flame retardants in consumer goods ; PCBs have been used in a host of consumer products, including fluorescent lights, televisions, and refrigerators ; and PAHs are produced by indoor combustion sources, including cigarette smoke, wood-burning fireplaces, and gas appliances . PBDE, PCB, PAH, and nicotine molecules on dust can enter the body via inhalation, via inadvertent ingestion after hand-to-mouth contact, or via direct absorption through the skin . For some individuals, notably children, dust likely contributes a substantial portion of the overall intake of PBDEs , PCBs , and PAHs and nicotine levels in dust offer a useful quantitative measure of cigarette smoking in the home .There have been several reviews of the use of dust as a medium for measuring chemical contamination in the home . Investigators have generally sampled residential dust by obtaining dust from subjects’ household vacuum cleaners or by collecting dust from floors, carpets or other surfaces using a standardized vacuum cleaner, such as a high volume surface sampler . Use of household vacuum cleaner bags eliminates the need for an in-home visit, which reduces study costs and minimizes the invasiveness of home sampling. In contrast, collecting dust with a standardized protocol allows investigators to know the location and time of dust collection. Some investigators have collected dust from household surfaces using a brush or broom . For example, Tan et al. collected dust from the upper surface of a fan blade to measure room-wide contamination. Researchers have made direct comparisons between chemicals measured in dust taken from household vacuum cleaners and dust collected using a standardized protocol in the same residence .