School of Postgraduate Studies
Master of Public Health and Master of
Science in Epidemiology and
Biostatistics
MEB 632–Statistical Methods in
Epidemiology
[PART-TIME]
Group Assignment
Instructions:
1. Group yourselves in groups of between 5 and 8
students.
2. Use the assignment cover given on the next page.
3. The assignment must be typed using Times New
Roman font type, Font size 12 and 1.5 line spacing. Do
NOT convert your word document to pdf.
Page 1 of 5
4. Submit your assignment by 23:59 hrs on 17th March
2023 using the Moodle Learning Management System
ONLY. The submission should be made by the Group
Leader.
5. Late submission of the assignment and copying from
each other will attract a mark of zero.
February 2023
UNIVERSITY OF LUSAKA
School of Postgraduate Studies
Master of Public Health
and
Master of Science in Epidemiology and
Biostatistics
MEB632- Statistical Methods in
Epidemiology
[PART-TIME AND DISTANCE]
Group Assignment
Names and Student Numbers:
Page 2 of 5
…………………………………………………………
…………………………………………………………
…………………………………………………………
…………………………………………………………
…………………………………………………………
Lecturer: Prof. Eustarckio Kazonga
Due Date: 17th March 2023
QUESTION ONE (1)
(a) The command describe in Stata gives information about the
variables and how they are defined in the dataset. The
results of the command show that there are 11 variables
defined as:
» id : identification code
» low: low birth weight, birthweight < 2500g (1=yes,
0=no)- This is a dependent variable being predicted
by the following independent variable
» age: age of mother (years)
» lwt: weight at last mentstrual period
» race: race of the mother, 1=white, 2=black,
3=other
» smoke: smoked during pregnancy, yes=1, no=0
Page 3 of 5
» ptl: premature labour history (counts)
» ht: has history of hypeternsion, 1=yes, 0=no.
» ui: presence of uterine irritability (1=yes, 0=no)
» ftv: number of visits to physician during first trimester
(count)
» bwt: birth weight (grams as continuous)
The best predictor model is obtained as shown in Table 1.3.
Table 1.3: Predictors of Low birth weight
Adjusted
Predictor OR p-value 95% CI
Weight at last 0.98 0.014 (0.97, 1.00)
Menstrual period
Race
White 1 (ref) n/a n/a
Black 3.76 0.011 (1.35, 10.4)
Other 2.53 0.031 (1.09, 5.87)
Smoke
No 1 (ref) n/a n/a
Yes 2.82 0.008 (1.31, 6.08)
History of hypertension
No 1 (ref) n/a n/a
Yes 6.49 0.007 (1.68, 25.1)
Presence uterine
irritability 1 (ref) n/a n/a
No 2.47 0.043 (1.03, 5.94)
Yes
REQUIRED
Interpret the information in Table 1.3.
(15 Marks)
(b) Smith, Delgado and Rutledge (1976) report data on ovarian
carcinoma. Individuals had different numbers of courses of
chemotherapy. The 5- year survival data for those with 1-4 and 10
or more courses of chemotherapy are:
Page 4 of 5
Five-Year Status
Courses Dead Alive
1-4 21 2
≥10 2 8
Source: Fisher LD and Van Belle G. Biostatistics: A Methodology
for the Health Sciences. New York: Wiley, 1993.
Chapter 6 problem 5, page 232.
REQUIRED
Using Fisher’s Exact test, is there a statistically significant
association (p < 0.05) in this table? In 1-2 sentences, write a clear
interpretation of your hypothesis test.
(10 marks)
[TOTAL: 25 MARKS]
QUESTION TWO (2)
(a) One way of identifying confounding is to examine the primary
association of interest at different levels of a potential
confounding factor. The side by side tables below examine
the relationship between obesity and incident CVD in persons
less than 50 years of age and in persons 50 years of age and
older, separately.
Table 2: Obesity and Incident Cardiovascular Disease by Age
Group
Age ‹ 50 Age ≥ 50
Tota Tota
CVD No CVD CVD No CVD
l l
Obese 90 10 100 Obese 164 36 200
Not Not
465 35 500 175 25 200
Obese Obese
Total 555 45 600 Total 339 61 400
REQUIRED
Page 5 of 5
Calculate Mantel-Haenszel estimate for the odds ratio and
interpret your answer.
(10 marks)
(b) The following Caesarian Births data is defined as follows:
Csec= number of C-sections performed
Hosp= type of hospital public or private, coded as (0-public
or 1-private)
Birth= number of births at the hospital
Case No. Csec Hosp Birth
1 8 0 236
2 16 1 739
3 15 1 970
4 23 1 2371
5 5 1 309
6 13 1 679
7 4 0 26
8 19 1 1272
9 33 1 3246
10 19 1 1904
11 10 1 357
12 16 1 1080
13 22 1 1027
14 2 0 28
15 22 1 2507
16 2 0 138
17 18 1 502
18 21 1 1501
19 24 1 2750
20 9 1 192
REQUIRED
Formulate an appropriate Poisson regression model for the
number of C-section births and interpret your model.
(15 marks)
[TOTAL: 25 MARKS]
Page 6 of 5
END OF ASSIGNMENT
Page 7 of 5
