In: Math
The article entitled “Effect of metformin therapy on 2-h
post-glucose insulin levels in patients of polycystic ovarian
syndrome” by Saxena et al. discussed the influence of metformin on
insulin levels in women with polycystic ovarian syndrome (POCS),
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3017330/?tool=pubmed
(Links to an external site.).
Summarize the article in terms of the statistics that were used and
calculate the following statistics for i) the body mass index, ii)
the FSH levels, and iii) the mean hirsutism scores (15 points):
a. The 95% and 99% confidence intervals.
b. Test the null and alternative hypothesis that the after treatment results are significantly different from the before treatment results.
Sample size=40
Pre-Treatment |
Post- Treatment |
|||||
Average |
SD (±) |
Average |
SD (±) |
|||
BMI |
25.5 |
4.4 |
23.6 |
3.7 |
||
FSH |
5.7 |
2.6 |
4.5 |
2.7 |
||
Mean Hirsutism |
13.8 |
4.2 |
9.2 |
2.2 |
(a)
BMH:
Pre Treatment:
One-Sample T
N Mean StDev SE
Mean 95% CI
40 25.500 4.400 0.696 (24.093, 26.907)
One-Sample T
N Mean StDev SE
Mean 99% CI
40 25.500 4.400 0.696 (23.616, 27.384)
Post Treatment:
One-Sample T
N Mean StDev SE
Mean 95% CI
40 23.600 3.700 0.585 (22.417, 24.783)
One-Sample T
N Mean StDev SE
Mean 99% CI
40 23.600 3.700 0.585 (22.016, 25.184)
FSH:
Pre Treatment:
One-Sample T
N Mean StDev SE Mean
95% CI
40 5.700 2.600 0.411 (4.868, 6.532)
One-Sample T
N Mean StDev SE Mean
99% CI
40 5.700 2.600 0.411 (4.587, 6.813)
Post Treatment:
One-Sample T
N Mean StDev SE Mean
95% CI
40 4.500 2.700 0.427 (3.636, 5.364)
One-Sample T
N Mean StDev SE Mean
99% CI
40 4.500 2.700 0.427 (3.344, 5.656)
Mean Hirsutism:
Pre Treatment
One-Sample T
N Mean StDev SE
Mean 95% CI
40 13.800 4.200 0.664 (12.457, 15.143)
One-Sample T
N Mean StDev SE
Mean 99% CI
40 13.800 4.200 0.664 (12.002, 15.598)
Post Treatment
One-Sample T
N Mean StDev SE Mean
95% CI
40 9.200 2.200 0.348 (8.496, 9.904)
One-Sample T
N Mean StDev SE
Mean 99% CI
40 9.200 2.200 0.348 (8.258, 10.142)
b. From above 95% and 99% CIs for Pre treatment and Post treatment of BMI we see that there are not disjoint hence at 1% level of significance we conclude that there is insufficient evidence that after treatment results are different from the before treatment results.
From above 95% and 99% CIs for Pre treatment and Post treatment of FSH we see that there are not disjoint hence at 1% level of significance we conclude that there is insufficient evidence that after treatment results are different from the before treatment results.
But from above 95% and 99% CIs for Pre treatment and Post treatment of Mean Hirsuitism we see that there are disjoint hence at 1% level of significance we conclude that there is sufficient evidence that after treatment results are significantly different from the before treatment results.