Introduction:
This study examined the blood pressure during the day in women who had normal blood pressure and then those that had preeclampsia. All of the participants were similar in age, weight, and mean duration of gestation (35 weeks). The researchers collected both systolic and diastolic BP readings during the day and at night. We will focus on systolic blood pressures during the day and the night.
Methods:
Researchers analyzed similar women. The Diastolic and Systolic BP's were recorded for a normotensive group and a preeclampsia group.
Analysis: The data we were given is interval/ratio data, which is appropriate for Pearson correlation. The 1st thing we must do in order to determine if there is a relationship between the systolic blood pressure during the day and at night is to produce a scatter plot.
One way anova was done to see if there was a relationship between the groups. The p value of
Analysis of Variance results:
Data stored in separate columns.
Column statistics
Column |
n |
Mean |
Std. Dev. |
Std. Error |
Diastolic (day) |
48 |
80.104167 |
10.929639 |
1.5775575 |
Diastolic (night) |
48 |
67.041667 |
16.150829 |
2.3311714 |
ANOVA table
Source |
DF |
SS |
MS |
F-Stat |
P-value |
Columns |
1 |
4095.0938 |
4095.0938 |
21.535766 |
<0.0001 |
Error |
94 |
17874.396 |
190.15315 |
||
Total |
95 |
21969.49 |
In order to estimate slope and intercept we conduct the linear regression.
Simple linear regression results for Group=Normotensive:
Dependent Variable: Diastolic (night)
Independent Variable: Diastolic (day)
Diastolic (night) = 33.74505 + 0.28960396 Diastolic (day)
Sample size: 24
R (correlation coefficient) = 0.34737345
R-sq = 0.12066832
Estimate of error standard deviation: 4.3250183
Parameter estimates:
Parameter |
Estimate |
Std. Err. |
DF |
95% L. Limit |
95% U. Limit |
Intercept |
33.74505 |
11.977689 |
22 |
8.9048439 |
58.585255 |
Slope |
0.28960396 |
0.16667593 |
22 |
-0.056060763 |
0.63526868 |
Analysis of variance table for regression model:
Source |
DF |
SS |
MS |
F-stat |
P-value |
Model |
1 |
56.472772 |
56.472772 |
3.0190007 |
0.0963 |
Error |
22 |
411.52723 |
18.705783 |
||
Total |
23 |
468 |
Simple linear regression results for Group=Preeclampsia:
Dependent Variable: Diastolic (night)
Independent Variable: Diastolic (day)
Diastolic (night) = -45.181135 + 1.4091046 Diastolic (day)
Sample size: 24
R (correlation coefficient) = 0.84460369
R-sq = 0.71335539
Estimate of error standard deviation: 7.4342533
Parameter estimates:
Parameter |
Estimate |
Std. Err. |
DF |
95% L. Limit |
95% U. Limit |
Intercept |
-45.181135 |
16.929738 |
22 |
-80.291262 |
-10.071008 |
Slope |
1.4091046 |
0.19043677 |
22 |
1.0141629 |
1.8040463 |
Analysis of variance table for regression model:
Source |
DF |
SS |
MS |
F-stat |
P-value |
Model |
1 |
3025.9347 |
3025.9347 |
54.750091 |
<0.0001 |
Error |
22 |
1215.8987 |
55.268122 |
||
Total |
23 |
4241.8333 |
Future Studies:
I think that a larger sample size should be used. I also think that we should examine if they have had treatment or any medications that they are on. We should also look at their exercise levels and which activities they are participating in and how they are sleeping. Their stress levels should also be evaluated. There is a lot that could interfere and affect the BP here.
Already a member? Sign in.
Nov 20, 2017
Hi Candace,
This is a good start. A few comments:
1. You state that you are going to examine the relationship between systolic blood pressure - day vs. night, but you use the diastolic in your analysis.
2. You should omit the One-Way ANOVA results as it is not appropriate here. To test for a correlation, we need to use Simple Linear Regression. (We use One-Way ANOVA to test for mean differences between 2 or more independent groups.)
3. You should check the assumptions for regression analysis. We need to check normality and constant variability. These can be checked with a "QQ Plot of Residuals" and "Residuals vs. X-values", respectively.
4. You were expected to interpret the estimates for slope. . For the preeclampsia group (comparing systolic day vs systolic night), a correct interpretation of the slope: With 95% confidence, we estimate the slope of our true regression line to be between 0.81 and 1.52. Putting this in context of the problem, we can say that a one BPM increase in daytime Systolic BP will correspond to a 0.81 to 1.52 BPM increase in the AVERAGE night-time Systolic BP for preeclampsia women. (We would interpret the slope for the normotensive group similarly.)
5. You were asked to make some predictions of your choosing which is done by selecting "Prediction of Y" and entering a value for X when conducting the Linear Regression procedure. For example, if we wish to make an individual prediction for X=120, a correct interpretation would be: For a preeclampsia individual with 120bpm daytime SBP, the estimated nighttime SBP is between 88.5 and 126.3bpm for the nighttime measurement. Remember that a PI is interpreted for an INDIVDUAL – not average.
Please review the solutions and let me know if you have any questions.