Assignment
PUBH5010 Epidemiology Methods and Uses
Due Date and Time: Sunday 5th April 2026, 23:59 AEST
Assignment Category: Submitted Work
Assignment Sub-category: Assignment
Compulsory Assessment: No
Weighting: 0 or 25%
Plagiarism Policy
You must complete your assignment alone. Submitting assignments that have been jointly done is not
acceptable. Copying someone else’s work or quoting from text without adequate attribution of the source is
plagiarism and is not acceptable. All assignments will be verified by plagiarism detection software. Penalties
will be applied for plagiarism. The University’s policy on academic honesty can be found at the following site:
http://sydney.edu.au/policies/showdoc.aspx?recnum=PDOC2012/254&RendNum=0

Use of Artificial Intelligence
Permitted uses of artificial intelligence (AI) include:
• Searching for and summarising relevant literature.
• Translation
• Checking grammar
• Editing, polishing, and proofreading initial drafts.
You must follow the University’s requirements for Generative Artificial Intelligence.

Late Penalties & Special Consideration & Assessment Policy
Unless you have an approved simple extension, special consideration, or academic plan 5% (5 marks out of
100) will be deducted per calendar day late, until Wednesday 15th April, 23:59 AEST or until the mark reaches
50%. Assignments submitted after this date without approved Special Consideration may not be accepted and
may be given a zero (0) mark. For students seeking ‘Special Consideration’ please use the following site:
Semester 1, 2026
Page 2 of 8

http://sydney.edu.au/current_students/special_consideration/apply.shtml
Requests for simple extensions of up to five working days can be submitted via the following site:
https://www.sydney.edu.au/students/simple-extensions.html
For more information regarding various assignment policies and procedures please see the following site:
https://www.sydney.edu.au/students/browse.html?category=your-studies

Instructions to Students
• You are not expected to do a literature search for this assignment.
• Assignments must be submitted via the Canvas assignment dropboxes (Assessments overview >
Assessment 1 - Information > and use the respective dropboxes to submit Section A and Section B of
the assignment).
• This assignment has two sections: Section A and Section B.
• Section A and Section B must be completed in separate documents.
• Section A should be uploaded to the Assignment Dropbox–Section A and Section B to the Assignment
Dropbox–Section B.
• Your submission will automatically submit through Turnitin. You should name your file to include your
student ID, unit code and the assignment section, e.g., 305111222_PUBH5010_SectionA. Please put
your student number first and don’t forget to specify the assignment section (Section A or B) in the
name of the submitted file.
• Assignments are marked anonymously. Please do not include your name on the title page or in the
header/footer of the documents. Only use your student ID number.



If you have any administrative questions, please post them on the Assignment discussion board on
Canvas. Go to Discussions > Assignment discussion board. Alternatively contact the Unit Co-
ordinators on (E: epi.teaching@sydney.edu.au.)
When you have successfully submitted a document into Turnitin via the dropboxes you will only
receive an on-screen receipt; this will not be emailed to you.
Things to consider before beginning
The purpose of this section is to guide you before you complete your assignment. The following feedback is
based on common strengths and limitations of previous student assignment answers.
The primary aim of the assignment is to allow you to demonstrate your early epidemiological knowledge and to
learn what is expected of you in PUBH5010.
Semester 1, 2026
Page 3 of 8

The assignment is not compulsory, and the mark does not count for your final mark if your exam mark is higher
than your assignment mark. However, we highly recommend that you submit an answer for both Parts A and B
to receive feedback and assess your progress.
Commentary
• Marking in PUBH5010 is criterion-based. We do not mark to a specific distribution.
• Critical Appraisal is a difficult but very important skill and we want students to demonstrate their
understanding of epidemiological principles. So, the key aspect we look for is that students provide a
justification for each point that they make. Students find this a challenge and it is this area that is
typically where students lose most marks in the assignment and the exam. Critical appraisal marks are
generally broken into two sections: content-based marks (i.e., the relevant point and justification for
the point) and a quality mark (a judgement based on how many appropriate points are discussed and
how well they are justified). To get full marks for content-based marks, you would need to identify the
issue and justify why it is an issue. For the quality section of a Critical Appraisal question you must
identify the majority (but not necessarily all) of the major points for a given section and adequately
justify each point.
• We DON’T expect a completely structured response (which reflects how our marking is conducted, i.e.,
we do not expect a formal essay). That said, having some structure or logical flow to your answer will
help ensure you are commenting on all relevant aspects and adequately demonstrating your
understanding.
• There is no negative marking. If you write something incorrect you won’t have marks deducted, but
you would expect to be awarded fewer marks in the quality section.
• There are no page limits to assignments, but we expect students to write succinctly. Our page limits
are only provided as a guide.
• Epidemiology is not always clear cut. Therefore, our marking criteria allow for other points to be
raised, as long as they are justified appropriately. We expect students to show insight, rather than just
regurgitating material.
• We do NOT expect references to be required to answer any of the questions and so do not assign any
marks to references, citations or quotes. However, if for some reason you do choose to quote from
somewhere you must of course properly cite the reference.
Creating Flow Charts
We highly recommend that you use word processing software to create the requested flow diagram. If you
need guidance on how to do so there are various tutorials available on the internet such as
http://www.makeuseof.com/tag/create-stunning-flowcharts-microsoft-word/or
https://www.youtube.com/watch?v=iiS7aAFI2Cs. If you cannot create a flow diagram in Microsoft Word, we
recommend you draw one by hand on blank white paper then scan and insert the image into your Word
document. You must submit a single document for each section.

Semester 1, 2026
Page 4 of 8

Background information for the questions
Two studies (Study A and Study B) were conducted to examine the relationship between exposure to talc and
the risk of cancer. One study (Study A) examined the risk of ovarian cancer from using talcum powder for
personal hygiene. The other study (Study B) was of talc miners and millers and examined the risk of lung
cancer. For the purposes of Study A, assume that increasing age, a history of being diagnosed with
endometriosis, exposure to asbestos and a family history of ovarian cancer are known to increase the risk of
ovarian cancer (and of course there might be other, unknown, risk factors). Also, assume that using talcum
powder is more common in women with higher socio-economic status. For the purposes of Study B, assume
that increasing age, smoking and various occupational exposures including asbestos are known to increase the
risk of lung cancer (and of course there might be other, unknown, risk factors). For both studies, it may be
relevant that talc is commonly contaminated with asbestos because the two minerals often occur naturally in
the same geologic deposits.

SECTION A (55 Marks)
Study A
Study A was a case-control study that examined the risk of ovarian cancer arising from the use of talcum
powder for personal hygiene. The study was conducted in Sharmaville, the capital city of Catland. Sharmaville
had three large hospitals (Hospital Blue, Hospital Red and Hospital Green). Nearly all people who required
admission to hospital would attend one of these three hospitals and nearly all people diagnosed with ovarian
cancer were admitted to a hospital at the time of diagnosis or very soon afterwards. All included subjects were
female.

Eligible cases were all women aged 20 years and over who were living in Sharmaville and were newly
diagnosed with ovarian cancer between 1 January 2020 and 31 December 2025 at either Hospital Blue or
Hospital Red. Cases were identified from the medical records of Hospital Blue and Hospital Red. These records
were known to be very accurate in terms of diagnoses of cancer. All people who were cases identified from
the medical records were invited to participate. Of the 300 invited cases, 205 agreed to participate. Those
who didn’t participate either died before interview (20), were too ill to participate (25) or refused to participate
(50). The cases who didn’t participate were more likely to come from a lower socioeconomic group than cases
who did participate.

Semester 1, 2026
Page 5 of 8

Eligible controls were all females aged 20 years and over who lived in Sharmaville at some time between 1
January 2020 and 31 December 2025 and who had not been diagnosed with ovarian cancer. Controls were
randomly selected from the electoral role for Sharmaville. This electoral roll was known to include 98% of
eligible people living in Sharmaville and essentially no-one who was not eligible. Each time a case was
identified in the medical records, a control was randomly selected. If the selected control declined to
participate, another control was randomly selected. This continued until the selected control agreed to
participate. Of 650 controls who were invited to participate, 201 actually took part in the study. Those who
didn’t participate were deceased at the time of contact (30) or refused to participate. The controls who didn’t
participate were more likely to come from a higher socioeconomic group than controls who did participate.

People were classed as exposed if they had used talcum powder for personal hygiene reasons for at least one
year and that this use had occurred more than 10 years before they were diagnosed with ovarian cancer (for
cases) or more than 10 years before they were selected for the study (the controls). One hundred and thirty-
four of the 205 people with ovarian cancer reported using talcum powder for personal hygiene reasons for at
least one year more than 10 years before they were diagnosed with ovarian cancer. Seventy-three of the 201
controls reported using talcum powder for personal hygiene reasons for at least one year more than 10 years
before they were selected for the study.

1) Study A. (35 marks)
(i) Do you think the study base in Study A is well defined or not well defined? Justify your answer.
(15 marks)

(ii) Do you think selection issues in Study A are likely to have biased the estimate of exposure (using
talcum powder for personal hygiene reasons, based on their self-report) in persons with ovarian
cancer? If so, do you think this would have led to an underestimate or an overestimate in the
estimate of exposure or are you unable to tell? Justify your answer. (10 marks)

(iii) Do you think selection issues in Study A are likely to have biased the estimate of exposure (using
talcum powder for personal hygiene reasons, based on their self-report) in persons without
ovarian cancer? If so, do you think this would have led to an underestimate or an overestimate in
the estimate of exposure or are you unable to tell? Justify your answer. (10 marks)


Semester 1, 2026
Page 6 of 8

2) Study A. (20 marks)
Regardless of your answer to Question 1, now assume there was NO important bias arising in the
study (i.e. no important selection bias, measurement bias or confounding). Calculate the appropriate
measure of effect that shows the relationship between using talcum powder for personal hygiene
reasons and the risk of developing ovarian cancer. To do this you will need to draw a 2*2 table.
(i) Show this table. (5 marks)
(ii) Calculate the appropriate measure of effect - show your working. (5 marks).
(iii) Interpret the result you obtained (in Question 2 (ii)) in words that someone who has not studied
epidemiology would understand. (5 marks)
(iv) Assuming the relationship found between using talcum powder for personal hygiene reasons and
the risk of developing ovarian cancer was causal, what proportion of the cases of ovarian cancer in
the community would be caused by the use of talcum powder for personal hygiene reasons?
Show your working (5 marks)


Semester 1, 2026
Page 7 of 8

SECTION B (45 Marks)
Study B
Study B was a cohort study of talc miners and millers (‘talc workers’) and a comparison group of workers at a
nearby cardboard box manufacturing factory. The study commenced in 1960. There were 1,500 workers at
the talc facility - 1,200 were invited to join the study and 1,100 agreed to join. There were 1,300 workers at the
cardboard box factory – 1,000 were invited and 800 agreed to join. Anyone with a history of cancer was
considered ineligible and was excluded from the study. This left 1,020 talc workers and 720 factory workers
who participated in the study. Compared to the factory workers, the talc workers were of a similar age,
smoked more and were more likely to have had an occupational exposure that was known to increase the risk
of lung cancer. The study ran for 50 years, during which participants were contacted every five years to
determine if they had been diagnosed with lung cancer and to obtain information on their smoking status and
occupational history. Participants were followed up until the earliest of being diagnosed with lung cancer,
death, or the end of the study. Of the talc workers who started the study and were not known to have been
diagnosed with lung cancer during the study, 120 died, 45 withdrew and 30 were lost to follow-up. Of the
factory workers who started the study and were not known to have been diagnosed with lung cancer during
the study, 60 died, 120 withdrew and 99 were lost to follow-up. This left 825 talc workers and 441 factory
workers who completed the study. The talc workers who completed the study were of a similar age, smoked
more and were more likely to have had an occupational exposure that was known to increase the risk of lung
cancer compared to the factory workers who completed the study.

By the end of the study, 95 of the talc workers had been diagnosed with lung cancer and 36 of the factory
workers had been diagnosed with lung cancer.


3) Draw a flow diagram that summarises what happened in Study B. (15 marks)


Semester 1, 2026
Page 8 of 8

4) Study B. (20 marks)
Assume there was NO important bias arising in the study (i.e. no important selection bias,
measurement bias or confounding). Calculate the appropriate measure of effect that compares the
risk of developing lung cancer between talc workers and cardboard box factory workers, based on the
data provided for Study B (not taking into account any issues with selection bias). To do this you will
need to draw a 2*2 table.
(i) Show this table. (5 marks)
(ii) Calculate the appropriate measure of effect - show your working. (5 marks).
(iii) Interpret the result you obtained (in Question 5(ii)) in words that someone who has not studied
epidemiology would understand. (5 marks)
(iv) Assuming the relationship found between working as a talc worker and the risk of lung cancer
(compared to cardboard box factory workers) was causal, what proportion of the cases of lung
cancer in talc workers was due to them being talc workers (rather than cardboard box factory
workers)? Show your working (5 marks)


5) Study B. (10 marks)
(i) Do you think the cardboard factory workers were an appropriate comparison group for the talc
workers? Justify your answer. (10 marks)



学霸联盟