Kindly find the attached documents.It has all the instructions in it. Simple sentences Please.
homework_6.doc
lecture_16.pptx
Unformatted Attachment Preview
Name: _______________________________________________________________________________
Homework #6: Study Design II: Observational Studies
Due Tuesday, November 26th by the End of the Day (11:59 p.m.)
43 Total Points (7% of Grade)
This is an individual homework assignment. While you are allowed to use any notes, textbooks or
readings, and to consult with the faculty instructor, if having trouble, you are expected to think
through this assignment and do it on your own without the collaboration of classmates. You
should wait until after you have listened to Lectures #13, 14, 15 and 16, and read pages 153-158 of
the textbook before starting this homework assignment. Remember to have your notes in front of
you while doing this homework.
Absolutely no late homework assignments will be accepted. Please type your answers,
numbering all of the questions.
Part 1: Neural Tube Defects and Multivitamins (11 points in total)
In 1988 Joseph Mulinare and his fellow investigators reported their findings concerning the
association of neural tube defects (type of birth defects) and periconceptional use of
multivitamins (taking multivitamins around the time they became pregnant) (JAMA 1988; 260:
3141-3145). They selected several groups of infants. One group consisted of “all live-born or
stillborn infants with a diagnosis of anecephaly or spina bifida (the two major kinds of neural
tube defects) born during the years 1968 through 1980 who were registered in the Metropolitan
Atlanta Congenital Defects Program.” The second group consisted of “live-born babies without
birth defects who were randomly chosen from all live births that occurred in the metropolitan
Atlanta surveillance area.” Data on multivitamin use was then obtained. Multivitamin use was
defined as “multivitamin or prenatal vitamin consumption during every month of the entire six
month period …from three months before conception through the third month of pregnancy by
the mother.” The following table presents partial results from that report. (A number of changes
and adjustments have been made to simplify the table for purposes of this example.)
2 by 2 Table for Maternal Multivitamin Use and Neural Tube Defects
Mother took
multivitamins
Mother did not take
multivitamins
Total
Infant born with
Neural Tube Defect
24
Infant not born with
birth defect
411
Total
159
1092
1251
183
1503
1686
435
1) What is the exposure in this study? (1 point)
2) What is the disease outcome in this study? (1 point)
1
Name: _______________________________________________________________________________
3) According to the description above, what were the two comparison groups? (2 points)
4) Were the two comparison groups those with and without the outcome or those with and
without the exposure? (1 point)
5) What type of study design is this? (Choose either case-control or cohort.) (1 point)
6) If this is a cohort study, calculate a relative risk. If it is a case-control study, calculate an
odds ratio. Calculate only the one that goes with this study design, and state which one it
is. Show all steps, including the formula, numerator and denominator. Round off the
final answer to two places after the decimal. (3 points)
7) Is taking multivitamins around the time of conception a possible protective or risk factor
for having a child with neural tube defects? (1 point)
8) Is taking multivitamins strongly associated with neural tube defects (i.e. a strong risk
factor or a strong protective factor) or is it just weakly associated? (Hint: consider a
strong risk factor as being an odds ratio or relative risk > 2.00, and a strong protective
factor as an odds ratio or relative risk < 0.5.) (1 point)
Part 2: Maternal Drug Consumption and Birth Defects (12 points in total)
As an epidemiologist you are going to investigate the effect of a drug suspected of causing
malformations in newborn infants when the drug in question is taken by pregnant women
during the course of their pregnancies. As your sample, you will use the next 250 single
births occurring in a given hospital. You will follow these mothers during the entire
course of their pregnancies, maintaining a complete and accurate record of drug use during
pregnancy. This will be achieved not only through interviewing the women, but in addition,
you will review medical records to verify use of the drug. At the end of the study, the
resultant data turn out to be: Fifty mothers have taken the suspected drug during their
pregnancies. Of these mothers, 42 have delivered malformed infants. In addition, there are 15
other infants born with malfunctions to women who have not taken the drug.
1)
2)
3)
4)
5)
What is the exposure in this study? (1 point)
What is the disease outcome in this study? (1 point)
What type of study design is this? (Choose either case-control or cohort.) (1 point)
Explain how you know that it is this study design. (2 points)
Construct a two by two table with all labeled columns and rows, including totals for
this study. (2 points)
6) If this is a cohort study, calculate a relative risk. If it is a case-control study, calculate
an odds ratio. Calculate only the one that goes with this study design, and state which
one it is. Show all steps, including the formula, numerator and denominator. Round
off the final answer to two places after the decimal. (3 points)
7) Is the exposure possible protective or risk factor for the outcome? (1 point)
8) Is the exposure strongly associated with the outcome (i.e. a strong risk factor or a
strong protective factor) or is it just weakly associated? (Hint: consider a strong risk
2
Name: _______________________________________________________________________________
factor as being an odds ratio or relative risk > 2.00, and a strong protective factor as
an odds ratio or relative risk < 0.5.) (1 point)
Part 3: Study Design Identification (20 points in total, 2 points each)
For each of the descriptions below, identify the study design. Choose only one study design per
description: ecologic, cross-sectional, case-control, or cohort study. Write the type
of study design that applies below each description.
Example:
A) The level of unemployment was used as a measure of economic distress in Germany.
Researchers examined the association between economic distress and general anxiety
syndrome across states in Germany.
Study Design: Ecologic
1. Young women who had suffered a stroke were compared with those who had not had a
stroke on their use of oral contraceptives.
Study Design:
2. A study was designed to find out if high rates of tuberculosis were associated with high rates
of AIDS in a population. The rates of tuberculosis in 50 countries was correlated with the
rates of AIDS in those same countries.
Study Design:
3. Parents of 863 one-month old Australian children were interviewed about the infant’s
sleeping environment. When these families were interviewed again 7 years later, it was
found that those infants who had used synthetic bedding items at one month of age were
more likely to develop asthma by the time they were seven years old than those whose
bedding items were completely made of natural materials.
Study Design:
4. A survey was conducted to discover the prevalence of overweight among Texas
schoolchildren.
Study Design:
3
Name: _______________________________________________________________________________
5. The entire population of a given community is examined and all who are judged free of
bowel cancer are questioned extensively about their diet. These people are then followed for
several years to see whether their eating habits will predict their risk of developing bowel
cancer.
Study Design:
6. An investigator collects information on the size of factories and their rates of accidents. She
finds that the five largest factories have accident rates that are 50% higher than the five
smallest factories.
Study Design:
7. A random sample of 300 people over the age of 65 with Parkinson’s Disease was obtained
through the National Parkinson’s Foundation. These were compared with 300 people over
age 65 who were randomly sampled from Medicare files, and had no diagnosis of
Parkinson’s Disease, on a variety of lifestyle factors.
Study Design:
8. The prevalence of genital infection with chlamydia trachomatis among women of
childbearing age was ascertained in a survey conducted in Houston.
Study Design:
9. Two hundred babies were followed from birth to age five to determine if those whose
mothers smoked during pregnancy were more likely to have respiratory infections in the first
five years of life than those whose mothers did not smoke during pregnancy. The findings
supported the investigators’ assumption that smoking during pregnancy increases the
frequency of respiratory infections in young children.
Study Design:
10. Fifty patients recently diagnosed with hepatitis A were selected along with another 50
patients without hepatitis A from among those attending a local clinic. The participants in
the two groups were then interviewed to determine if they had any body piercing within the
last two months.
Study Design:
4
COHORT STUDIES
Lecture 16
Cohors = Latin: Warriors,
1/10 of a Roman Legion
◦ The term cohort is said to
originate from the Latin
cohors, which referred to
one of ten divisions of an
ancient Roman legion, part
of the Roman army.
Epidemiologist’s
Definition:
◦ A designated group of
persons who are
followed or traced over
a period of time
Cohort Studies:
Longitudinal Studies with Data on Individuals
Epidemiology
Observational
Individuals
Groups
Ecologic
Experimental
CrossSectional
CaseControl
Cohort
Groups
Individuals
Community
Interventions
Clinical
Trials
◦ An analytic study attempts to identify
causes or risk factors that explain healthrelated states or events.
Cohort
Studies are
Analytic
◦ It tests specific a priori (previously formed)
hypotheses often developed in descriptive
studies.
◦ Has a Comparison group.
◦ Examines associations between exposure(s)
and outcome(s)
◦ It begins with people with the
exposure and compares them to
people without the exposure
(comparison group) over time in
respect to incidence (or mortality)
of the disease.
Chief
Characteristic
of a Cohort
Study
Design of a Cohort Study
Disease
Exposed
Source
Population
People without
the disease:
Sample of
the Population
at Risk
Not Exposed
No
Disease
Disease
No
Disease
Start with the Population at Risk
Population of interest should be
reviewed to ascertain those people or
groups that are susceptible to
becoming cases.
Individuals who already have a disease outcome
of interest (prevalent cases) or who are not at risk
(e.g., they have had an organ removed such
that they cannot become a case) should be
excluded from the study.
Cohort Studies: Characteristics
◦ Start with a group of participants who lack a positive
history of the outcome of interest and are at risk for the
outcome.
◦ Include at least two observation points:
1. one to determine eligibility and exposure status
2. a second (or more) to determine the number of incident
cases.
◦ The same individuals are followed over time.
Example 1 of Cohort Study
The Alameda County Study
• Involved residents of Alameda County, CA,
ages 16-94 years.
• Data collected through mailed
questionnaires; telephone interviews or home
interviews of non-respondents.
• Began in 1965.
• Follow-up in
◦
◦
◦
◦
◦
1973,
1985,
1988,
1994, and
1999.
Source of Map:
https://upload.wikimedia.org/wikipedia/commo
ns/4/4c/California_map_showing_Alameda_Co
unty.png
“ The Alameda 7”
Studied lifestyle factors associated with health and mortality.
1. Having Never Smoked
2. Drinking Less Than 5 Drinks at One Sitting
3. Sleeping 7- 8 Hours a Night
4. Exercising
5. Maintaining Desirable Weight for Height (Normal BMI)
6. Avoiding Snacks
7. Eating Breakfast Regularly
Example 2 of a Cohort Study: Framingham Heart Study
◦ Conducted in Framingham, Massachusetts.
◦ Used a random sample of 5,209 from targeted age range of 30 to 62
years in 1948.
◦ Longest Ongoing study of Coronary Heart Disease in the world
◦ In the 3rd Generation of Participants.
◦ http://www.framinghamheartstudy.org/
Source of Map:
https://upload.wikimedia.org/wikipedia/comm
ons/3/3f/Framingham_ma_highlight.png
Source of Logo:
https://upload.wikimedia.org/wikipedia/commons/thumb/0
/03/US-NIH-NHLBI-OldLogo.svg/602px-US-NIH-NHLBIOldLogo.svg.png
Framingham Study
(Film)
◦ Study of heart disease and stroke risk
factors.
◦ Ongoing study of Coronary Heart Disease
initiated in 1948.
◦ Later recruited children and grandchildren
of original participants
◦ Examine participants every two years
Some Research Milestones from
the Framingham Heart Study
◦ 1960: Cigarette smoking found to increase risk of heart disease.
◦ 1961: Cholesterol level, blood pressure, and electrocardiogram abnormalities found to increase risk
of heart disease.
◦ 1967: Physical activity found to reduce risk of heart disease and obesity to increase risk of heart
disease.
◦ 1970: High blood pressure found to increase risk of stroke.
◦ 1976: Menopause found to increase risk of heart disease.
◦ 1978: Psychosocial factors found to affect heart disease.
◦ 1978: High levels of HDL cholesterol found to reduce risk of death.
◦ 1999: Lifetime risk at age 40 of developing coronary heart disease is 50% for men and 33.3% for
women
◦ 2002: Lifetime risk of developing high blood pressure in middle-aged adults is 90%
https://www.framinghamheartstudy.org/fhs-about/research-milestones/
Begun in 1995, the first questionnaire
was mailed to subscribers to
Essence magazine and members of
several professional organizations.
59,000 women who completed and
returned that questionnaire are
being followed.
To study many illnesses, such as
hypertension, breast cancer, stroke,
and lupus in black women.
https://www.bu.edu/bwhs/
Example 3:
The Black
Women’s
Health Study
(BWHS)
Example #4: The Nurses’ Health Study I, II and III
◦
◦
◦
◦
◦
◦
◦
◦
◦
◦
Nurse’s Health Study I:
Began in 1976 with 121,700 nurses
Originally studied oral contraceptive use; expanded to women’s health.
Married female R.N.s ages 30-55 years.
Data collected through mailed questionnaires.
Follow-up every 2 years; toenail sample at year 6 and blood sample at
year 13.
Nurses’ Health Study II:
started in 1989 with 116,430 female nurses between 25 and 42 years old
Nurses’ Health Study III: both male and female nurses, web-based, started
in 2010, still recruiting
http://www.channing.harvard.edu/nhs/
• A study in which a
group of people is
followed over time.
• The group is made
up of people who have
the exposure of interest
and people who do
not have the exposure
of interest.
Cohort Study
• Exposed and
unexposed people are
followed over time to
determine whether
they experience the
outcome.
Chief Characteristic of a Cohort Study
◦It begins with people with the exposure and
compares them to people without the
exposure (comparison group) over time in
respect to incidence of the disease.
HOW IS THE DESIGN OF A
COHORT STUDY DIFFERENT
THAN THE DESIGN OF A
CASE-CONTROL STUDY?
Design of Case Control Study: Retrospective
Time
Direction of inquiry: backwards
Exposed
Not exposed
Exposed
Not exposed
Cases
(diseased)
Population
Controls
(disease free)
Design of a Cohort Study
Direction of Inquiry: Forward in Time
Disease
Exposed
No
Disease
Population
People without
the disease
Disease
Not Exposed
Time
Direction of inquiry: forward
No
Disease
Comparison of Case-Control and Cohort Studies
Case-Control Study
◦ Comparison Groups: Those with
and without the outcome
◦ Data collected from the cases
(with the outcome) and the
controls (without the outcome) at
one point in time.
◦ They may be asked about the
past, but are not followed over
time.
◦ The disease (or other outcome)
has already occurred when the
study begins.
Cohort Study
◦ Comparison Groups: Those with
and without the exposure
◦ Data collected from the exposed
and not exposed (entire cohort)
at several points in time.
◦ They are followed over time to see
what happens to them.
◦ The disease (or other outcome)
has not occurred when the study
begins.
Cohort Studies permit calculation of incidence (and mortality) rates.
◦ Can be thought of as going from cause to effect.
Earlier Exposure
Later Disease or Death
Incidence
◦ Rate of development of disease
during a given period of time
◦ Three key elements:
◦ Only new cases included in
numerator
◦ Total population at risk in the
denominator
◦ Time element – period over which
new cases developed
Cohort
Studies:
Measures of
Association
Relative risk provides a direct
measure of association
between exposure and
outcome.
Relative risk is the ratio of the
incidence of disease in the
exposed group to the
incidence in the non-exposed
group.
Risk
◦ The probability that an
event will occur, for
example, that an
individual will become
ill or die, within a
specified period of time
Relative Risk
◦ A way of
quantifying the
relationship
between two risks
◦ Tells us the number
of times one risk is
larger or smaller
than another
Ratio
◦ A fraction with no specified relationship between
numerator and denominator
◦ Range: 0 to
◦ A/B
◦ Examples
◦ Prevalence Ratio (ratio of prevalence proportions)
◦ Odds Ratio (ratio of two sets of odds)
◦ Relative Risk (ratio of two incidences)
Relative Risk
Relative risk =
Incidence rate in the exposed
Incidence rate in the non-exposed
◦ Research Question:
Is there an association between
child abuse and suicide attempts
among chemically dependent
adolescents?
◦ Source: Deykin EY, Buka SL. Am J
Public Health. 1994;84:634-639.
Cohort
Studies:
Relative Risks Sample
Calculation
Numerator and Denominator
Numerator
Denominator
Incidence of the Outcome
Among Exposed
Incidence of the Outcome
Among unexposed
Incidence (Ie) = a / (a+b)
Incidence (Io) = c / (c+d)
With
Without
Outcome Outcome
Total
Exposed
A
B
A+B
Not
Exposed
C
D
C+D
Total
A+C
B+D
N
(A+B+C+D)
Relative Risk is
calculated
like the
Prevalence
Ratio, but has
a different
meaning.
◦ Relative Risk
= [A/(A+B)] /
[C/(C+D)]
Suicide Attempt and
History of Sexual
Abuse among Male
Chemically
Dependent
Adolescents
History of Suicide
Attempt
Sexual
Abuse
No Suicide Totals
Attempt
Yes
A = 14
B=9
A + B = 23
No
C = 49
D = 149
C + D = 198
Total
A+C = 63
B+D = 158
N = 221
Relative Risk =
(14/23) /
(49/198) =
0.6087 / 0.2475
History of Suicide
Attempt
Sexual
Abuse
No Suicide Totals
Attempt
Yes
A = 14
B=9
A + B = 23
No
C = 49
D = 149
C + D = 198
Total
A+C = 63
B+D = 158
N = 221
= 2.46
Is the Exposure Protective or a Risk Factor?
Positive
Association
Negative or Inverse
Association
Protective
0
No
Association
1
Risk Factor
How to Interpret Relative Risks
Positive
Association
Negative or Inverse
Association
Possibly
Protective
Exposure
Possible Risk
Factor
No Association
1
0
Stronger
Weaker
2.46
Stronger
How to tell from the Relative Risk if the exposure may be protective
or a risk factor
◦ RR=1 implies no association.
◦ Assuming statistical significance
◦ (which you only know from a p-value):
◦ RR >1 suggests the exposure is a risk factor for the outcome
◦ RR<1 suggests a protective factor
◦ The further away from one (smaller or larger) that the RR is,
the stronger the association is.
Calculated like a
Prevalence Ratio, but…
◦ It is two incidence rates (or risks) that
you are comparing – not two
prevalence proportions
◦ Since incidence is ONLY influenced by
the exposure – not by duration of
disease
◦ A relative risk is a stronger measure of a
possible causal relationship.
Advantages of Cohort Studies
◦ Permit direct determination of incidence (risk)
◦ Time sequencing of expos ...
Purchase answer to see full
attachment