1. Lecture 3 Survival analysis

2. Problem Do patients survive longer after treatment A than after treatment B? Possible solutions: –ANOVA on mean survival time? –ANOVA on median survival time?

3. Progressively censored observations Current life table –Completed dataset Cohort life table –Analysis “on the fly”

4. First example of the day

5. Person-year of observation In total: 15.122 days ~ 41.4y 11 patients died: 11/41.4y = 0.266 y -1 26.6 death/100y 1000 patients in 1 y or 100 patients in 10y

6. Mortality rates 11 of 25 patients died 11/25 = 44% When is the analysis done?

7. 1-year survival rate 6 patients dies the first year 25 patients started 24%

8. 1-year survival rate 3 patients less than 1 year 6/(25-3) = 27% Patient 7 24% -27%

9. Actuarial / life table anelysis Treatment for lung cancer

10. Actuarial / life table anelysis A sub-set of 13 patients undergoing the same treatment

11. Actuarial / life table anelysis Time interval chosen to be 3 months n i number of patients starting a given period

12. Actuarial / life table anelysis d i number of terminal events, in this example; progression/response w i number of patients that have not yet been in the study long enough to finish this period

13. Actuarial / life table anelysis Number exposed to risk: n i – w i /2 Assuming that patients withdraw in the middle of the period on average.

14. Actuarial / life table anelysis q i = d i /(n i – w i /2) Proportion of patients terminating in the period

15. Actuarial / life table anelysis p i = 1 - q i Proportion of patients surviving

16. Actuarial / life table anelysis S i = p i p i-1...p i-N Cumulative proportion of surviving Conditional probability

17. Survival curves How long will a lung canser patient survive on this particular treatment?

18. Kaplan-Meier Simple example with only 2 ”terminal-events”.

19. Confidence interval of the Kaplan- Meier method Fx after 32 months

20. Confidence interval of the Kaplan- Meier method Survival plot for all data on treatment 1 Are there differences between the treatments?

21. Comparing Two Survival Curves One could use the confidence intervals… But what if the confidence intervals are not overlapping only at some points? Logrank-stats –Hazard ratio Mantel-Haenszel methods

22. Comparing Two Survival Curves The logrank statistics Aka Mantel-logrank statistics Aka Cox-Mantel-logrank statistics

23. Comparing Two Survival Curves Five steps to the logrank statistics table 1.Divide the data into intervals (eg. 10 months) 2.Count the number of patients at risk in the groups and in total 3.Count the number of terminal events in the groups and in total 4.Calculate the expected numbers of terminal events e.g. (31-40) 44 in grp1 and 46 in grp2, 4 terminal events. expected terminal events 4x(44/90) and 4x(46/90) 5.Calculate the total

24. Comparing Two Survival Curves Smells like Chi-Square statistics

25. Comparing Two Survival Curves Hazard ratio

26. Comparing Two Survival Curves Mantel Haenszel test Is the OR significant different from 1? Look at cell (1,1) Estimated value, E(a i ) Variance, V(a i ) row total * column total grand total

27. Comparing Two Survival Curves Mantel Haenszel test df = 1; p>0.05

28. Hazard function d is the number of terminal events  f is the sum of failure times  c is the sum of censured times