Mouse over boxplot for 25th, median and 75th percentile.
Min. 1st Qu. Median Mean 3rd Qu. Max.
50.00 65.00 73.00 70.62 77.00 84.00
In this section, we tried to fit a logistic regression over the data with the probalibity of having intraoperative Bradycardia/Hypotension/Hypertension as the response(y-axis) and the balloon inflation time as the predictor.
Call:
glm(formula = intra_op_bradycardia ~ balloon_inflation_time,
family = binomial, data = record)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.517 -1.176 0.851 1.071 1.290
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.5182 0.8874 -0.584 0.559
balloon_inflation_time 0.2577 0.2983 0.864 0.388
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 33.104 on 23 degrees of freedom
Residual deviance: 32.327 on 22 degrees of freedom
(14 observations deleted due to missingness)
AIC: 36.327
Number of Fisher Scoring iterations: 4
Call:
glm(formula = intra_op_hypotension ~ balloon_inflation_time,
family = binomial, data = record)
Deviance Residuals:
Min 1Q Median 3Q Max
-1.4323 -1.2178 0.9423 1.1375 1.2944
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) 0.8255 0.8988 0.918 0.358
balloon_inflation_time -0.2436 0.2939 -0.829 0.407
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 33.104 on 23 degrees of freedom
Residual deviance: 32.395 on 22 degrees of freedom
(14 observations deleted due to missingness)
AIC: 36.395
Number of Fisher Scoring iterations: 4
Call:
glm(formula = intra_op_hypertension ~ balloon_inflation_time,
family = binomial, data = record)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.2597 -1.0310 0.6407 0.8702 1.3313
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.9201 0.9548 -0.964 0.335
balloon_inflation_time 0.5654 0.3542 1.596 0.110
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 31.755 on 23 degrees of freedom
Residual deviance: 28.742 on 22 degrees of freedom
(14 observations deleted due to missingness)
AIC: 32.742
Number of Fisher Scoring iterations: 3
We can see from the data plot, the resulting logistic regression line does not fit well with the data. The p-values of the fitted coefficients are larger than 5% which implies its non-significance. Therefore, we can conclude that balloon inflation time does not correlate to intraoperative Bradycardia/Hypotension/Hypertension
60% of the patients remain hypertensive after taking hydralazine.
41.3793103 of patients remain bradycardiac after taking atropine or glycol