Return to Topics
The purpose of this project is to write the statistical analyses and power sections of a grant application. The PI of the grant application is Dana Dabelea, MD, PhD, who has kindly allowed us to use the grant for this class. A redacted version of the other sections of the grant are on the canvas website. Each of you will be responsible for only a subset of the hypotheses contained in the grant (see next specific aims page for assignments based on degree program). The timeline for this project is tighter than for the other projects. The amount of work is appropriate for the time given, but requires that you are proactive in getting information from the investigator the week of November 28th.
The investigator needs should restate the scientific hypotheses for which you are responsible, your attempt to map those to statistical hypotheses (recognizing that you will not have access to the investigator until then) the sample(s) you will have to address those hypotheses, the potential relevant outcome(s) and explanatory variables. I expect that you will have many questions for the investigator regarding exactly the questions of interest and the variables available.
By Wednesday, November 30th you should be able to fill in an outcome/exposure/covariate table for your hypotheses that resembles those we will discuss in class (a similar table already in the grant gives you a lot of information, but not all).
Written report: Use the general outline for the statistical section given in the notes November 30th. Be sure to cover all the topics discussed in that lecture and all the pieces of information important for the power analysis. Do not restate any of the information given in the other parts of the grant – your section should be compatible with the other sections, but do not repeat any of it (e.g. how the measure of visceral fat is obtained).
The prevalence of obesity has been increasing dramatically in the United States over the past decades (9-11). Moreover, overweight is now present at increasingly younger ages, indicating that risk factors for this condition start operating early in life (11). As in adults, obesity in childhood and adolescence is associated with adverse short- and especially long-term chronic outcomes, such as the insulin resistance syndrome, type 2 diabetes, cardiovascular disease, and increased cardiovascular mortality (12-14). Determining which and how early life factors operate among youth is the first step towards prevention of both childhood and adult obesity.
Fetal life is considered one of the critical (or sensitive) periods when an exposure may have lifelong effects on the structure or function of organs, tissues and body systems (through biological programming). These early effects may be modified by later life exposures (through biological interaction) to determine future chronic disease risk (15). Evidence exists that exposure to altered glucose-insulin metabolism (or frank diabetes) in utero leads to increased fetal growth, and may increase the risk for obesity later in life (2) . At the other end of the birth weight spectrum, lower birth weight (as marker of other less well characterized exposures) seems to be associated with measures of central adiposity and components of the insulinresistance syndrome later in life (16). Thus, it appears that increased adiposity in later life occurs among children born at both ends of the birth weight spectrum: generalized obesity with higher birth weight and increased visceral adiposity and its metabolic consequences at lower birth weights. Whether these phenotypes are detectable in exposed children and whether they have common biologic pathways is a focus of this proposal. Therefore, the specific aims of this proposal are:
Aim 1: To examine in parallel the long-term consequences of intrauterine exposure to excess nutrition (maternal diabetes) and under nutrition (growth restriction) on childhood body size, fat patterning and markers of insulin resistance, among children of different ethnic groups.
Aim 2: To explore the hypothesis of intrauterine hormonal “programming” through which fetal exposures could increase the later risk of obesity and insulin resistance in children.We hypothesize that:
Exposure to diabetes (DM) in utero programs the fetus for the development of generalized obesity. Consistent with the fetal origins hypothesis, the associations between exposure to DM in utero and childhood measures of obesity will be in addition to genetic susceptibility to obesity (marked by maternal pre-pregnancy body mass index). MPH
At the other extreme of birth weight distribution, intrauterine growth restraint (IUGR) is associated with the development of central obesity and insulin resistance (IR), when adjusted for attained body mass index (BMI). For a given BMI, children with IUGR will have higher IR-related measures and more visceral and muscle fat mass, when compared to children without IUGR. All
Cord blood leptin concentration is associated with fetal growth but is also a marker of fetal hormonal programming of future risk of obesity and IR in children.
Newborns exposed to DM in utero have higher leptin levels at birth, when compared with un-exposed offspring, even after adjustment for differences in birth weight. MPH
Newborns with IUGR have lower absolute leptin levels, when compared with newborns without IUGR; however, leptin levels are not reduced, and may even be increased, relative to their smaller body size. MS and PhD
Both exposure to DM and IUGR induce in utero leptin-resistance and promote defective leptin signaling. Leptin levels at birth mediate the association between each exposure and childhood outcomes. MS and PhD (NOTE: No power section expected for c)