Assignment:
QUESTIONS:
The following exercise was adapted from:
Bayona, M, and Olsen C. "Measures in Epidemiology" - The Young Epidemiology Scholars Program. Supported by the Robert Wood Johnson Foundation and administered by the College Board.
1) A study of the health effect of magnetic fields (MFs) of extremely low frequency was conducted in the San Francisco area from October 1996 through October 1998. The investigators in this study were interested in the potential effects of magnetic fields on human pregnancies. Specifically they were interested in miscarriage. The investigators interviewed pregnant women and asked them to wear a device that would measure their exposure to magnetic field strength for 24 hours and also to keep a diary of their activities. The investigators did not observe an association between the risk of miscarriage and the average level of magnetic fields but did find an association between high magnetic field levels and the incidence of miscarriage.
After an initial look at their data, they focused on the relationship between whether or not the women were exposed to magnetic fields above the level of 16 milligauss and the subsequent incidence of miscarriage. The following table presents the data on the maximum magnetic field versus whether or not the women miscarried
A) Give the formula and calculate the incidence rate in the exposed (Ie) per 100 pregnant women
B) Give the formula and calculate the incidence rate in the unexposed (Iu) per 100 pregnant women
C) Give the formula and calculate the risk difference per 100 pregnant women
D) Interpret your answer for risk difference per 100 pregnant women
E) Give the formula and calculate the attributable risk % (AR%)
F) Interpret your answer for attributable risk %
G) Give the formula and calculate the population attributable risk (PAR) per 100 pregnant women (hint: incidence rate for the population (Ip) = (a+c)/(a+b+c+d)
H) Interpret your answer for population attributable risk per 100 pregnant women
I) Give the formula and calculate the relative risk (RR) for miscarriage comparing women exposed to magnetic fields above the level of 16 mG to those with less than 16 mG
J) Interpret your answer for the relative risk
K) If the confidence interval for the relative risk you calculated in I was = (0.88, 2.56) would you conclude that it was significant? Why?
L) If the confidence interval for the relative risk you calculated in I was = (1.45, 1.99) would you conclude that it was significant? Why?
M) If you wanted to use data from this study to do a statistical test to see if there was actually a statistically significant difference between the incidence rate in the exposed compared to the unexposed. What would your NULL HYPOTHESIS BE? Be specific.