In a case-control study of oral cancer, investigators individually match each case with a control on the basis of tobacco chewing status, intending to strengthen the study. A major consequence of this choice is that the study:

• A Will automatically show a stronger odds ratio for tobacco chewing
• B Cannot estimate the independent effect of tobacco chewing on oral cancer risk ✓
• C Becomes immune to recall bias regarding past tobacco use
• D Must be analysed using unconditional logistic regression

Explanation

Once cases and controls are matched on a variable, its distribution is forced to be identical in both groups, so no information about its association with disease remains in the data. The matched factor can be examined only if the matching is broken or the design allows discordant pairs to be modelled separately. Overmatching on a factor closely linked to the exposure can also bias the odds ratio toward the null.

Reference: Park's Textbook of Preventive and Social Medicine, 27th ed.

High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP

Written and medically reviewed by the StethoPrep medical team.