1 Harvard University School of Public Health, Boston, Massachusetts 02115
The generalizations which Dr. Nichols has made (or inferred) by applying his (incorrect and incomplete) definitions to data from Galveston and Philadelphia are affected by an enormous sampling bias. How many other places in the United States which are not medical centers report the outcomes of pregnancy in the same way? Perhaps there is even more "under-reporting" (as Dr. Nichols defines it) in the U.S.A. as a whole than in the Netherlands as a whole in spite of certain differences in "official" reporting requirements (even though the difference Dr. Nichols specifies no longer holds)? Furthermore, the Galveston-Philadelphia data Dr. Nichols quotes are not those reported to the Vital Statistics Divisions of the cities of Philadelphia and Galveston.