A matched case-control study evaluates whether oral contraceptive use is associated with venous thrombosis. Each case is matched to one control, and the data are arranged as discordant pairs: 25 pairs where only the case used contraceptives and 10 pairs where only the control used them. The appropriate test for these paired binary observations is:

• A Standard chi-square test without correction
• B Fisher exact test on the full fourfold table
• C Paired t-test
• D McNemar test ✓

Explanation

McNemar test is designed for paired or matched dichotomous data and uses only the discordant pairs, computed as (b minus c) squared divided by (b plus c). Here it gives (25 minus 10) squared divided by 35, about 6.43, which exceeds the critical value of 3.84 at one degree of freedom. Applying an ordinary chi-square test to the full table ignores the matching and inflates the apparent sample size, violating independence assumptions. Paired t-test applies to continuous paired measurements.

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.