More Advanced Methods; Survival Analysis

Question 1

• _______ Time is the time to an event/time starting from a defined point to the occurence of a given event.
• In this what can an “event” be?

Answer 1

• Survival Time
• Death, Disease Occurence, Disease Recurrence, Recovery, Other experience of study interest such as adverse drug reaction or development of a new disease.

Question 2

What are some special features in Survival Time Data?

Answer 2

• rarely normally distributed
• often skewed
• typically with many early events and relatively few late ones

Question 3

What are Censored Observations of Survival Time?

Answer 3

• Those who have not yet reached the terminal event by the end of the study.
• When the information about their survival time is incomplete for reasons such as; pt not yet experienced event by close of study, pt lost to follow-up during study, pt experiences different event

Question 4

What is the issue with Censored Observations in Survival Time Analysis?

Answer 4

They will underestimate the true time to the event because it will occur beyond the end of the study.

Question 5

What is a Survival Analysis?

Answer 5

Collective term for the methods of analysis for survival data invented specifically to deal with time to an event problems.

Question 6

What are some examples of Survival Analysis methods used?

Answer 6

• Kaplan-Meier Curve
• Median Survival Time
• Mean Survival Time
• Hazard Ratio
• Log-Rank Test
• Cox Model

Question 7

Kaplan-Meier Curve:

• Widely used in clinical research to visualize the estimate of the ________ over _____.
• Shows what the probability of an event is at a certain time interval.
• The __-axis displays “time”; the __-axis displays proportion surviving
• Is a ____ function, as the cumulative survival remains the same until the day another person experiences the event.

Answer 7

• survival over time
• x-axis=”time”, y-axis=”proportion surving”
• step

Question 8

Kaplan-Meier Curve with Censored Data:

• The __-axis displays “time”; the __-axis displays proportion surviving.
• -Is a ____ function, as the cumulative survival remains the same until the day another person experiences the event.
• _______ observations are indicated on the K-M curve as tick marks.
• Censored observations ____ terminate the interval.

Answer 8

• x-axis=”time”, y-axis=”proportion surving”
• step
• censored
• do not

Question 9

Kaplan-Meier Curve for Two Groups:

• Visualizes the difference between two _______ curves.
• What is the purpose?

Answer 9

• survival curves
• Purpose is to compare the survival curves to a study questions like “Do patients get better faster under treatment A or B?”

Question 10

Median Survival Time:

• Is estimated as the _________ survival time for which the survival function is less than or equal to 0.5
• Can be estimated using the K-M curve by doing what?

Answer 10

• smallest
• Finding the 50% mark on the proportion axis, drawing a horizontal line at the 50% to find the crossing point with the K-M curve, drawing a vertical line at the crossing point down to the time axis to read time.

Question 11

Mean Survival Time:

• Is estimated as the area ______ the survival curve.
• May not be the best estimate for the samples or survival times that are frequently highly skewed.
• The _______ survival time is generally a better measure of central location than the mean.

Answer 11

• under

- median

Question 12

Is Median or Mean Survival Time Analysis better generally?

Answer 12

-Median

Question 13

Hazard Rate:

• Is the rate of a subject to experience a particular event at the time it has reached.
• Is a measure of how ________ a subject gets to experience a particular event.
• Can be estimated as being a slope of a line fitted through the ___ curve.

Answer 13

• rapidly

- K-M

Question 14

Hazard Ratio:

• Is a measure of how often a particular event happens in one group compared to another.
• Is a ratio of the ________ in one group over the __________ in another group.
• In clinical trials, it measures the survival at any point in time in a group of patients who have been given a specific treatment compared to a control group.
• HR of 1 = ?
• HR > 1 = ?
• HR < 1 = ?

Answer 14

• hazard rate, hazard rate
• HR of 1 = event rates are the same in both groups
• HR > 1 = event rate in treatment group is faster than control group
• HR < 1 = event rate in treatment group is slower than in the control group

Question 15

• HR of 1 = ?
• HR > 1 = ?
• HR < 1 = ?

Answer 15

• HR of 1 = event rates are the same in both groups
• HR > 1 = event rate in treatment group is faster than control group
• HR < 1 = event rate in treatment group is slower than in the control group

Question 16

The farther the HR is from 1, the _____ the difference between the 2 groups.

Answer 16

larger

Question 17

Log-Rank Test:

• One of the most commonly used tests that compare 2 or more samples with survival data in the presence of _______ observations.
• Log-Rank Test can fail if the two curves _______ (fail = no statistical power)
• Lower of Log-Rank Test is at its best for proportional hazards.

Answer 17

• censored observations

- cross (=no statistical power)

Question 18

What are the assumptions of Log-Rank Tests?

Answer 18

-Hazard rates of the groups to compare must be proportional. (no crossing)

Question 19

What are the limitations of Log-Rank Tests?

Answer 19

-Can only test one variable at a time.

Question 20

Cox Proportional Hazard Model:

• The most commonly used method that compares two or more samples with survival data in the presence of ________ observations.
• Cox model can accommodate any number of _________ variables.
• Cox model provides the estimate ___ with its associated 95% CI.

Answer 20

• censored
• cofounding
• HR

Question 21

What are the assumptions of Cox Proportional Hazard Models?

Answer 21

-The ratio of the hazard functions for any two observations does not vary with time.

Question 22

Quiz 1:

• What statistical hypothesis tests would you consider to use when your study aim is:
  1. ) To compare the mean blood pressure between men and women.
  2. ) To compare the mean blood pressure between members of three different groups of low, medium, and high BMI.
  3. ) To compare the mean age of onset of colorectal cancer between the group who use aspirin and the group who use ibuprofen.

Answer 22

1. ) independent t-test OR Mann-Whitney U Test when nonparametric
2. ) analysis of variance OR Kruskal-Wallis H Test when nonparametric
3. ) Log-Rank Test

Question 23

Quiz 2:

• In a study, two treatments were compared in their survival time and the data collected for all 15 patients were displayed in table 1. In the data, the column ‘Treatment’ recorded the treatment type each patient had received, the column ‘Survival time’ the number of days they had survived, the column ‘Outcome’ the status of each patient at the end of study.
  1. ) How many are the censored observations?
  2. ) What graph would be the best to visualize the survival time for each treatment group?
  3. ) Which statistical hypothesis tests would be appropriate to compare the two treatment groups with their survival time?

Answer 23

1. ) 5 censored observations
2. ) Kaplan-Meier Curve
3. ) Log-Rank Test OR Cox Proportional Hazard Model when there is a need to control for confounding variable

Question 24

Quiz 3:

• In the same study as Quiz 2, Kaplan-Meier Curves were drawn and displayed.
  1. ) What is the median survival time for treatment 1?
  2. ) What is the median survival time for treatment 2?
  3. ) Are the hazard rates of the two treatments proportional?
  4. ) Which treatment shows the slower hazard rate?
  5. ) What statistical method would be appropriate to test the different survival time between the two treatments?
  6. ) What statistical method would be appropriate to test the different survival time between the two treatments controlling for a confounding variable ‘Age’?

Answer 24

1. ) 37 days
2. ) 5 days
3. ) Yes, they do not cross
4. ) Treatment 1
5. ) Log-Rank Test
6. ) Cox Proportional Hazard Model

Question 25

Quiz 4:

• In a study for the association between body mass index (BMI) and time to incident cardiovascular disease (CVD), participants’ BMI was measured, at baseline, in four categories of underweight, normal, overweight, and obesity along with a known clinical risk factor for CVD, blood pressure. Participants were followed for up to 10 years for the development of CVD. Using ( ), they found that the time to a CVD is significantly different between the different BMI groups after controlling for blood pressure. Which statistical technique would be a best choice?
  1. Student t-test
  2. ) Log-Rank Test
  3. ) Cox Proportional Hazard Model
  4. ) ANOVA

Answer 25

3.) Cox Proportional Hazard Model

Question 26

Summary For More Advanced Methods; Survival Analysis:

• _________ ________ is a collective term for the methods of analysis for survival data as being used to analyze the time to an event data in the presence of censored observations.
• __________ ________ are those who have not yet reached the terminal event by the end of the study so whose information about their survival time is incomplete.
• Standard comparison tests such as analysis of variance or regression methods cannot be used for survival analysis because survival times are typically not ________ ________ and come with the presence of _______ observations.
• ______-________ survival curve is widely used in clinical research to visualize the estimate of the survival over time.
• Log-rank test is appropriately used to compare two or more independent groups with the time to an event data.
• ____ ________ ________ _____ is appropriately used to compare two or more independent groups for the time to an event data controlling for a confounding variable and it also estimates a HR with its 95% confidence interval.

Answer 26

• Survival Analysis
• Censored Observations
• normally distributed, censored
• Kaplan-Meier
• Cox Proportional Hazard Model