27 Oct Study of Neural Tube Defects and Vitamin Use

Table 2: 2 by 2 Table of Study of Neural Tube Defects and Vitamin Use

NTD Other Births Total

Supplementary Vitamin Use 105 463 568

No Supplementary Vitamin Use 84 137 221

Total 189 600 789

 

 

 

 

Problem 3:  Cigarette Smoking and Incidence of Miscarriages (10 points)

 

A study was conducted many years ago on a sample of young women in Omaha, Nebraska, to determine if there was an association between cigarette smoking during pregnancy and the incidence of miscarriages.  Four hundred prospective mothers between 18 and 24 years of age were followed for the entire course of their pregnancies.  Fifteen of the prospective mothers were lost to follow-up or died due to causes other than complications of pregnancy during the follow-up period.  Therefore, data were available for only 385 mothers in the sample.  Of the 385 mothers, 50 reported smoking cigarettes during pregnancy and 10 of these had miscarriages.  A total of 25 miscarriages occurred altogether.  The results are shown in Table 3.

 

 

 

Table 3: Miscarriages and Smoking during Pregnancy

 

 

 

 

 

Had a Miscarriage

 

Did not have a Miscarriage

 

Total

 

Smoked 

 

10

 

40

 

50

 

Did not Smoke

 

15

 

320

 

335

 

Total

 

25

 

360

 

385

 

 

 

What was the exposure variable in this study? (1 point)

What was the outcome variable? (1 point)

What was the study design? (2 points)

Calculate the most appropriate measure of association for this study design comparing the smoking and nonsmoking groups of mothers on their incidence rates of miscarriages.  Name the type of ratio you are calculating.  Show all steps in the calculation, including numerator and denominator, as well as the final answer, rounded to two places past the decimal. (3 points)

Using 2.00 and 0.50 as cut-off points for determining whether an association is strong or weak, from these data, does it appear that smoking is (1 point):

A strong risk factor for miscarriages.

A weak risk factor for miscarriages.

Not associated with miscarriages.

A weak protective factor against miscarriages.

A strong protective factor against miscarriages.

Given what is written in the description, what particular type of bias might be possible in this study? (2 points)

 

 

Problem 4: Auto Accidents and Smoking (13 points)

 

An epidemiologist wants to test the hypothesis that cigarette smoking increases the risk of fatal automobile injuries.  To test this hypothesis, he selects 100 people who have died as a result of automobile injuries from police and other official records in his county, and 200 other people matched to those who had died by age and sex, using random sampling procedures.  Both groups consisted of licensed drivers at the time of the fatal accident or when recruited into the study (if still a living driver).  Interviews with spouses, friends, and co-workers of all the participants indicated that 68 of those who died in an automobile crash and 104 of those who didn’t were cigarette smokers. 

 

What is the exposure of interest in this study? (1 point)

What is the disease or outcome of interest? (1 point)

Did the investigators start by identifying participants (1 point):

 with and without the exposure

 with and without the outcome

or was both the exposure and outcome unknown at the time the sample was assembled?

What study design is this? (2 points)

Draw and label a  2 by 2 table to display these data.  Include the total column and total row. (2 points)

From the 2 by 2 table that you have constructed, calculate the most appropriate estimate of the association between the exposure and outcome.  In other words, if this is a cohort study, calculate a relative risk.  If this is a case-control study, calculate an odds ratio. If this is a cross-sectional study, calculate a prevalence ratio. State which ratio is most appropriate here. Show all steps when calculating it, including the numerator and denominator. Round the final answer to two places after the decimal. (3 points)

In one sentence, and following the examples from the lecture notes, precisely interpret the relative risk, odds ratio or prevalence ratio you calculated.   (2 points)

Using 2.00 and 0.50 as cut-off points for determining whether an association is strong or weak, from these data, does it appear that smoking is (1 point):

A strong risk factor for automobile fatalities.

A weak risk factor for automobile fatalities.

Not associated with automobile fatalities.

A weak protective factor against automobile fatalities.

A strong protective factor against automobile fatalities.

 

 

 

 

Part 3:  Interpretation of Ratios when Used as Measures of Association (10 points)

 

 

 

Table 4: Direction and Strength of Measures of Association

 

 

 

Prevalence Ratio, Odds Ratio or Relative Risk Positive (Risk Factor) or Negative (Protective Factor) Association

Strong or Weak Association

 

(Use 2.00 and 0.50 as the borders of strong versus weak.)

 

6.81  

0.34  

0.94  

1.15  

0.14  

3.68  

1.26  

0.88  

10.99  

0.27  

 

 

 

 

Please fill in the middle and right columns of Table 4 with the correct answers.  Simply type “positive” or “negative” in the middle column and “strong” or “weak” in the column on the right.