Probability Exercises Biostatistics - UdL - Academic year 2023-24

Exercise 1. Hypertension (H) is associated with myocardial infarction (I). Which of the
following answers corresponds to the event ‘Individual with hypertension and no myocardial
infarction’ ?

(a) H ∩ I
(b) H ∩ I
(c) H ∩ I
(d) H ∩ I

Exercise 2. If F is the event family history of myocardial infarction (M I), how would
you write the event individual with hypertension (H) that suffered a M I and does not have
family history of M I? Represent the event using Venn diagrams.

(a) (H ∪ M I) ∩ F
(b) H ∩ M I ∩ F
(c) H ∪ M I ∪ F
(d) H ∩ M I

Exercise 3. There is evidence that hypertension (H), hyperuricemia (U ) and hypercholes-

terolemia (C) are risk factors for a particular disease. Look at the following events and
represent them using set theory (union (∪), intersection (∩), ...). It is recommendable that
you represent the event graphically using Venn diagrams.

(a) Individuals who do not present any of the risk factors considered.
(b) Individuals who are hypertensive or hypercholesterolemic. Comment on the difference
between this event and the previous event.
(c) Individuals who only have one of the risk factors considered.
(d) Individuals who are hypertensive and hypercholesterolemic.
(e) Individuals who have more than one of the risk factors considered.

Exercise 4. The frequency of three diseases, A, B and C, in a studied population is indicated

in the figure below. Estimate the frequency that corresponds to the following events:

1
 A

8%

18% 16%

10%

10% 14%
20%

B C

(a) Having A and C.

(b) Having A and B but not C.
(c) Having only disease A.
(d) Having A or B.
(e) Having any of the three diseases.
(f) Having A or B but not C.
(g) Having the three of them or none.
(h) Having disease C.
(i) Having only one of the three diseases.
(j) Not having any of the three diseases.

Exercise 5. In a specific population, the percentages of adult individuals who suffered from
any of three childhood diseases D1, D2 and D3 are:

D1

8%

12% 14%

7%

10% 8%
15%

D2 D3

Calculate the probability of the following events:

(a) Not having had D3, if one had neither D1 nor D2.
(b) Having had D1 or D2, if one did not have D3.
(c) Having had both D1 and D2.

2
 (d) Having had disease D1, if one had disease D2.
(e) Not having had D1 or D2, if one had D2.
(f) Not having had D1, if one did not have D2.
(g) Not having had D1, if one had D2.

Exercise 6. In a study about a disease (D) and a risk factor (R) the following probabilities
were obtained:

D D Total
R 0.08 0.11 0.19
R 0.33 0.48 0.81
Total 0.41 0.59 1.00

(a) What is the probability that a person with the risk factor presents the disease?
(b) What is the probability that a person without the risk factor presents the disease?
(c) Calculate the relative risk (RR).
(d) Do you think that R is a risk factor for the disease?

Exercise 7. Data from a hospital show that 10% of patients who have a particular disease
die within an interval of two years. Suppose that 5 patients have been admitted to the
hospital recently.

(a) Describe the sample space.

(b) What is the probability that ‘No patient dies within this period of two years’.
(c) What is the probability that ‘Exactly one patient dies in this period’.

Exercise 8. The table below shows the results of evaluating a diagnostic test for a particular
disease. Two samples were chosen for this evaluation, a sample of 522 patients with the
disease and an independent sample of 522 healthy individuals.

D D
T+ 446 255
T− 76 267
Total 522 522

(a) Determine the sensitivity of the test. 446/522= 0,8544

(b) Determine the specificity of the test. 267/522 = 0,5114
0,85*0,08 (c) If the disease prevalence, P (D), is 0.08, which is the positive predictive value (PPV) of
the test? 0,1319
0,85*0,08+(1-0,511)*(1-0,08)
(d) If the disease prevalence, P (D), is 0.08, which is the negative predictive value (NPV)
of the test?
0,5114*(1-0,08)
0,9757
(1-0,8544)*0,08+0,5114*(1-0,08)

3
Exercise 9. The probability that a person aged 74 years or more has vitamin B12 deficiency
is 0.22. Among people aged 74 years or more with vitamin B12 deficiency, the likelihood of
having folate deficiency is 0.48.
P(A)=0,22 P (B|A) = 0,48
(a) What is the probability that a person aged 74 years or more has simultaneously defi-
ciency of vitamin B12 and folate? P ( B|A) = P(B A) entre P(A) = 0,1
(b) If A represents the event ‘to have vitamin B12 deficiency’ and B represents the event
‘to have folate deficiency’, to obtain the previous result, which formula did you use? (1)
P (A) = P (A | B)P (B); (2) P (A∩B) = P (B | A)P (A); (3) P (B) = P (B∪A)+P (B∪A);
or (4) P (A ∪ B) = P (A) + P (B) − P (A ∩ B).

Exercise 10. In a given population, the probability of having the risk factor A is 0.11, the
probability of having the risk factor B is 0.19 and the probability of having both is 0.05.
P(A B)
(a) If a person has the risk factor B, what is the probability of having also the risk factor
A?
P( A/B) = P(A B) / P(B) =
0,05/0,19=0,26
Exercise 11. The positive predictive value (PPV) of a diagnostic test is:

(a) The probability that a person has the disease if the test comes out positive.
(b) The probability that the test is negative in people without the disease.
(c) The probability that the test is positive in people with the disease.
(d) The probability that a person does not have the disease if the test comes out negative.

Exercise 12. The positive predictive value (PPV):

(a) It is not related to the prevalence of the disease.

(b) Depends on the prevalence of the disease.
(c) It is only related to the sensitivity.
(d) It is only related to the specifity.

Exercise 13. The negative predictive value of a diagnostic test is:

(a) The probability that a person does not have the disease if the test comes out negative
(b) The probability that the test is positive in people with the disease
(c) The probability that a person has the disease if the test comes out positive
(d) The probability that the test is negative in people without the disease

Exercise 14. The sensitivity of a diagnostic test is:

(a) The probability that the test is positive in those with the disease.

4
 (b) The probability that the test is negative in those without the disease.
(c) The probability that a person has the disease if the test comes out positive.
(d) The probability that a person does not have the disease if the test comes out negative.

Exercise 15. The probability that a person with the disease has a negative diagnosis is:

(a) 1-sensitivity Falso negativo

(b) sensitivity
(c) 1-positive predictive value
(d) 1-specificity

0,116
Exercise 16. A diagnostic test for a disease with a prevalence of 11.6% has a probability of
giving a false positive result equal to 0.18 and a probability of giving a false negative result
equal to 0.07.T - | D 1-sens T+|D 1-esp
Sens= 0,93
(a) What is the probability that a person who has obtained a positive result, has the
disease?
0,93*0,116 0,404
VPP =
0,93*0,116+(0,18)*(1-0,116)
Exercise 17. A diagnostic test has a sensitivity equal to 0.7 and a specificity equal to 0.96.
0,7/
(a) What is the likelihood ratio for a positive result (LR+ )?
1-0,96 17,5
(b) What is the likelihood ratio for a negative result (LR− )? 1-0,7 0,3125
0,96
(c) If the prevalence of the disease is 0.1 and the test gives a positive result, what is the
odds post-test of having the disease? Prev LR 0,1 0,111*17,5
1-prev 1-0,1 1,9425
(d) If the prevalence of the disease is 0.1, what is the post-test probability of having the
disease if the test gave a negative result?

0,111*0,3125= 0,03468

Exercise 18. Below are two tables adapted from the Catalan Life Tables produced by the
Catalan Health Department and the Institute of Statistics (IDESCAT). For the questions
below (except the first one), we suggest that if you are male you use Table 1, if you are female
you use Table 2. If you are working in a group and/or have the time, do the questions for
both. Use what you have learnt about probability and apply it to these questions:
Adapted from the web page of the University of South Western, Sidney (Australia), QMP tutorials. https://moodle.
telt.unsw.edu.au/enrol/index.php?id=7699

(a) What do they tell you? What differences can you see between the two? mujeres viven más
(b) What is the probability that an individual chosen at random from this group will survive
to age 20? Si es hombre 994/1000 Si es mujer 995/1000
(c) What is the probability that this individual dies before age 20? Which property of
probability does this calculation depend on? P(TM < 20) = 1-0,994) P(TM< 20) = 1- 0,995
(d) Let’s say you are 20 years old. What is the probability that you survive to age 40?
(e) Let’s say you are 20 years old. What is the probability that you survive to age 60?

Hombres
P (T>40|T>20) =983/994 5
P (T>40| T>20) =
 P (T>60|T>20)= 991/994 Hombre

Age in Number Number of Conditional Age in Number Number of Conditional

years surviving deaths prob. of death years surviving deaths prob. of death
x lx dx qx x lx dx qx
0 1000 4 0.004 0 1000 3 0.003
10 996 2 0.002 10 997 2 0.002
20 994 4 0.004 20 995 2 0.002
30 990 7 0.0071 30 993 3 0.003
40 983 19 0.0193 40 990 11 0.0111
50 964 51 0.0529 50 979 23 0.0235
60 913 108 0.1183 60 956 46 0.0481
70 805 220 0.2733 70 910 129 0.1418
80 585 375 0.641 80 781 383 0.4904
90 + 210 210 1 90 + 398 398 1

Table 1: Number of men out of 1000 re- Table 2: Number of women out of 1000
maining alive at 10 year intervals remaining alive at 10 year intervals

(f) What is the probability that you and a friend (also aged 20 and of the same sex) both
survive to age 60?
P(TW>60|TW>20)
(g) Out of 100 medical students in your year, how many would we expect to attain the age
of 80? To simplify things let’s assume they are all aged 20 and of the same gender (the
one you are working on). We assume here that a male or female medical student has
an average life expectancy / rate of death - but this is probably not so. What are the
major factors that influence death rates?
(h) What is the probability that you will die in the 3rd decade?
(i) You have a great idea and think that it would be good to know your probability of
dying in each decade (a wee bit morbid, but necessary for this question!). Calculate the
probability of dying in each decade for an individual chosen at random from this group.
Tabulate it. Take a look at what you have done. What is this distribution called and
why?