Base on the following Table, taken from a linear regression from the article, Jo

Question

Base on the following Table, taken from a linear regression from the article, Joossens JV et al: Dietary salt, nitrate and stomach cancer mortality in 24 countries. Int J Epid. 1996 25(3): 494-504. The outcome is age-adjusted stomach mortality rate. Remember that this is a linear regression. (You do NOT need the article).

Which of the following can be deduced from Model 1 alone?

A. Nitrates are not a confounder of the sodium-stomach cancer association.

B. Nitrates are not an effect modifier of the sodium-stomach cancer association.

C. After adjusting for Nitrates, sodium remains a highly significant predictor of stomach cancer rates.

D. A and C

Table 3: Linear Multiple Regression of Stomach Cancer Mortality, Age-Adjusted 45-74 years, Rate/100,000 year, Against Average Daily Sodium (Na) and Nitrate Excretion (mmol/24-hrs) 20-49 Years, N-24 Countries Men Coeff (SE) Adjusted R2 Model 1 Stomach cancer mortality related to median 24-hour urinary sodiurn and nitrate Na Nitrate 0.679 (0.176) 21.97 (7.17)* 0.61 Model 2 Stomach cancer mortality related to median 24-hour urinary sodium and nitrate and their interaction Na Nitrate Na Nitrate 0.219 -130 0.77 0.763 (0.199)* For Clarity, the following is copied from the Table 3 above so you can read it: Model 1 interaction term Coefficient 0.679 Na Nitrate 21.97 Model 2 Parameter Na Nitrate Na X Nitrate nteraction term Coefficient 0.219 130 0.763**

Explanation / Answer

First let us understand the terms "Cofounding" and "Effect Modification".

Confounding: A situation in which the effect or association between an exposure and outcome is distorted by the presence of another variable. Positive confounding (when the observed association is biased away from the null) and negative confounding (when the observed association is biased toward the null) both occur. If an observed association is not correct because a different (lurking) variable is associated with both the potential risk factor and the outcome, but it is not a causal factor itself, Its Cofounding.

Effect modification : a variable that differentially (positively and negatively) modifies the observed effect of a risk factor on disease status. Different groups have different risk estimates when effect modification is present. If an effect is real but the magnitude of the effect is different for different groups of individuals (e.g., males vs females or blacks vs whites). It is effect modification.

-----------------------------

Now note that both the variables in model 1 are significant (both have at least two star marks), so we can't say that either of them is lurking variable. So, there is no cofounder.

As the variable "Nitrate" is significant, so its presence has effect on measure of association of sodium-stomach cancer association. So, it is an effect modifier.

But as we can see that the predictor sodium is highly significant after adjusting for nitrates.

So, option (C) is correct.

Related Questions

Navigate

• Browse (All)
• Browse B
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.