3. Consider the following results from Rich-Edwards et al. (2005, Table 1) [the paper is available on Canvas, but you do not need to read the paper to answer these questions]. Specifically, consider the data highlighted in the red box. These data are hazard ratios, which you can, for our purposes, interpret as relative risks (the difference between hazard ratios and relative risks depends on whether the data are point-in-time or over time). As you hopefully recall from our study of the root causes of mortality (Mokdad et al., 2004), the relative risk is the ratio of the probability of an event occurring in an exposed group to the probability of the event occurring in a comparison (non-exposed) group. In this case, a comparison is being made on the relative incidence of coronary heart disease across nurses by birth weight, relative to those of average birth weight. Birth weight (g) 2495-3175 >3175-3856 >3856-4536 >4536 Per kg. across all Extremely Low Very Low Birth Low Birth Average Birth categories Coronary heart disease Birth Weight Weight Weight Weight Events 21 59 329 392 91 26 918 Person years 8493 25 174 158 727 237 378 58 718 12 518 501 008 Adjusted hazard ratio: For age 1.31 (0.84 to 2.04) 1.48 (1.12 to 1.95) 1.26 (1.09 to 1.46) 1.00 0.86 (0.68 to 1.08) 0.98 (0.66 to 1.46) 0.77 (0.69 to 0.87) For age and BMI 1.31 (0.84 to 2.03) 1.50 (1.14 to 1.98) 1.30 (1.12 to 1.51) 1.00 0.84 (0.67 to 1.06) 0.93 (0.63 to 1.39) 0.75 (0.67 to 0.84) (a) Is it a coincidence that all nurses born at average birth weight suffer from coronary heart disease, stroke, and all cardiovascular disease at the exact same rate? (b) Summarize the highlighted findings in one or two concise sentences. (c) Discuss the finding in light of what you know about the thrifty phenotype hypothesis. (d) BONUS (up to 3 points): A confidence interval is provided for the data. Of the three categories of low birth weight, which, if any, are not statistically significant from 1.00? Explain