Scribd
Upload
29 views3 pages

Assignment

The document discusses the Kaplan-Meier estimator for estimating survival functions from lifetime data, providing calculations for two drugs, A and B. Drug A has a median survival time of 32 days and a cumulative hazard of 0.1335 at 20 days, while Drug B has a median survival time of 15 days and a cumulative hazard of 0.2877. Overall, Drug A is determined to be more effective than Drug B based on these metrics.

Uploaded by

ellyotieno856
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
29 views3 pages

Assignment

The document discusses the Kaplan-Meier estimator for estimating survival functions from lifetime data, providing calculations for two drugs, A and B. Drug A has a median survival time of 32 days and a cumulative hazard of 0.1335 at 20 days, while Drug B has a median survival time of 15 days and a cumulative hazard of 0.2877. Overall, Drug A is determined to be more effective than Drug B based on these metrics.

Uploaded by

ellyotieno856
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

i) Kaplan-Meier Product-Limit Estimate and Survival Curves

The Kaplan-Meier estimator is used to estimate the survival function from lifetime data. The formula for
the Kaplan-Meier estimator is:

S^(t)=∏i=1k(1−dini)\hat{S}(t) = \prod_{i=1}^{k} \left(1 - \frac{d_i}{n_i}\right)

where:

 S^(t)\hat{S}(t) is the estimated survival function at time tt.

 did_i is the number of events (deaths) at time tit_i.

 nin_i is the number of individuals at risk just before time tit_i.

Let's compute the Kaplan-Meier estimates for both Drug A and Drug B.

Drug A:

 Survival times: 17, 21+, 23, 24+, 25, 32, 37+, 39

 Censored times: 21+, 24+, 37+

Time Number at Risk Events


Survival Probability S^(t)\hat{S}(t)
(days) nin_i did_i

17 8 1 1×(1−18)=0.8751 \times \left(1 - \frac{1}{8}\right) = 0.875

21+ 7 0 0.8750.875

0.875×(1−17)=0.750.875 \times \left(1 - \frac{1}{7}\right) =


23 7 1
0.75

24+ 6 0 0.750.75

0.75×(1−16)=0.6250.75 \times \left(1 - \frac{1}{6}\right) =


25 6 1
0.625

0.625×(1−15)=0.50.625 \times \left(1 - \frac{1}{5}\right) =


32 5 1
0.5

37+ 4 0 0.50.5

0.5×(1−14)=0.3750.5 \times \left(1 - \frac{1}{4}\right) =


39 4 1
0.375
Drug B:

 Survival times: 9, 10, 12+, 13, 15, 19+, 21, 28

 Censored times: 12+, 19+

Time Number at Risk Events


Survival Probability S^(t)\hat{S}(t)
(days) nin_i did_i

9 8 1 1×(1−18)=0.8751 \times \left(1 - \frac{1}{8}\right) = 0.875

0.875×(1−17)=0.750.875 \times \left(1 - \frac{1}{7}\right) =


10 7 1
0.75

12+ 6 0 0.750.75

0.75×(1−16)=0.6250.75 \times \left(1 - \frac{1}{6}\right) =


13 6 1
0.625

0.625×(1−15)=0.50.625 \times \left(1 - \frac{1}{5}\right) =


15 5 1
0.5

19+ 4 0 0.50.5

0.5×(1−14)=0.3750.5 \times \left(1 - \frac{1}{4}\right) =


21 4 1
0.375

0.375×(1−13)=0.250.375 \times \left(1 - \frac{1}{3}\right) =


28 3 1
0.25

Survival Curves

To plot the survival curves, we can use the survival probabilities calculated above.

ii) Median Survival Time

The median survival time is the time at which the survival probability is 0.5.

 For Drug A, the median survival time is 32 days.

 For Drug B, the median survival time is 15 days.

iii) Cumulative Hazard at 20 Days

The cumulative hazard function H(t)H(t) is related to the survival function S(t)S(t) by the formula:

H(t)=−ln⁡(S(t))H(t) = -\ln(S(t))

For Drug A at 20 days:

 S(20)≈0.875S(20) \approx 0.875

 H(20)=−ln⁡(0.875)≈0.1335H(20) = -\ln(0.875) \approx 0.1335


For Drug B at 20 days:

 S(20)≈0.75S(20) \approx 0.75

 H(20)=−ln⁡(0.75)≈0.2877H(20) = -\ln(0.75) \approx 0.2877

Which Drug is More Effective?

Drug A appears to be more effective as it has a higher median survival time (32 days vs. 15 days) and a
lower cumulative hazard at 20 days (0.1335 vs. 0.2877). This suggests that mice treated with Drug A have
a better survival rate compared to those treated with Drug B.

You might also like