2x2 with structural zeroes?

Hi All,

Sorry to bother you, but could anyone help with a question, please?

Suppose I have N patients who tested positive for a condition and I want to verify using a second type of test all N patients who tested positive by the first test. That would yield the following cells:

Test1 Test2
+ +
+ -

Is it possible to analyze this table using McNemar's test or other test?

Sorry if this is a really stupid question. I am not a statistician...

Thanks much,
If you have a table with two rows and two columns, McNemar's test is probably not the route to go. It would be easier to de Fisher's exact test or the Pearson chi-square