Author Affiliations: Department of Medicine, The Pulmonary Center, Boston University School of Medicine, Boston, Massachusetts, and Center for Health Quality, Outcomes, and Economic Research, Edith Nourse Rogers Memorial VA Hospital, Bedford, Massachusetts (Dr Wiener); Department of Veterans Affairs, VA Outcomes Group, White River Junction, Vermont (Drs Schwartz and Woloshin); and Departments of Medicine and Community & Family Medicine, The Dartmouth Institute for Health Policy and Clinical Practice, and Dartmouth Medical School, Hanover, New Hampshire (Drs Wiener, Schwartz, and Woloshin).
We thank Ashrani and Heit for their interest in our study.1 Because direct evidence of overdiagnosis can only be obtained if individuals diagnosed as having a PE are observed without treating the PE until they die from other causes, using indirect evidence from population data is an established method of identifying overdiagnosis.2- 4 The pattern of rising incidence, stable mortality, and decreasing case fatality in the setting of the introduction of a highly sensitive test is strongly suggestive of overdiagnosis.5 While we agree with Ashrani and Heit about the general limitations inherent in administrative data (and noted many of these limitations in our article), we are not convinced by their specific comments about competing explanations for our findings.
Wiener RS, Schwartz LM, Woloshin S. Caution on Interpreting the Time Trends in Pulmonary Embolism as “Overdiagnosis”—Reply. Arch Intern Med. 2011;171(21):1963-1964. doi:10.1001/archinternmed.2011.550