SURVIVAL ANALYSIS

Question 19:The table below was obtained from a study of 345 cancer patients who were started on a new treatment and followed for survival. The study population was followed for 6 years. Using the Actuarial Method, complete the table and answer the questions pertaining to it. Note:Carry your calculations in the table to 4 decimal places (i.e., 0.1234).

Survival of Patients with Cancer Following Diagnosis - Actuarial Method
Interval (yrs)	# Alive at beginning of interval	# Dead during interval (d)	# censored (c)	# at risk (n)	Survival probability for interval	Survival probability cumulative
0	345	-	-	-	-	1.00
1	345	61	8
2	276	54	18
3	204	26	22
4	156	31	12
5	113	21	23
6	69	10	17
Survival = (n - d) / n ; where n = alive -1/2(c)

A. The probability of surviving the first year is:

B. Among those surviving year 2, the probability of surviving the third year is:

C. The probability of surviving from study onset through end of the fourth year is:

D. What is the probability that a person enrolled in the study will survive to the end of the sixth year?

E. What are the assumptions of this type of analysis?

Question 20: The tables below were obtained from a study of AIDS patients who were started on two different treatment regimens (Treatments A & B) and followed for survival for 49 months. Using the Kaplan-Meier method, complete the following tables. Carry your calculations in the table to 4 decimal places (i.e., 0.1234).

Analysis of AIDS Survival - Kaplan Meier Method
Time to event (months)	# Alive before death occurred	# of deaths (d)	# censored before death(c)	# at risk (n)	Survival probability for interval	Survival probability cumulative
Treatment A
0	33	-	-	-	-	1.00
1	33	5	3
2	25	4	0
3	21	2	1
7	18	3	3
13	12	2	1
25	9	1	2
37	6	1	5
49	0
Survival = (n - d) / n ; where n = (# alive before event - c)

Analysis of AIDS Survival - Kaplan Meier Method

Time to event (months)	# Alive before death occurred	# of deaths (d)	# censored before death (c)	# at risk (n)	Survival probability for interval	Survival probability cumulative
Treatment B
0	32	-	-	-	-	1.00
1	32	4	0
2	28	6	0
3	22	7	1
7	14	2	0
13	12	1	6
25	5	2	1
37	2	0	1
49	1
Survival = (n - d) / n ; where n = (# alive before event - c)

a. Which treatment worked better?

Bonus Question (1 point):

Graph the cumulative survival probability for both treatment A and B on the graph below

Cumulative Survival Probability for Treatment A and B Kaplan Meier Method 0.9 Treatment A 0.8 Treatment B 0.7 CUMULATIVE SURVIVAL (%) 0.1 0 10 20 30 40 50 TIME (months)