Skip to main content
. Author manuscript; available in PMC: 2017 Nov 1.
Published in final edited form as: Am Econ Rev. 2016 Dec;106(12):3730–3764. doi: 10.1257/aer.20140260

Table 2.

Summary statistics illustrating potential misweighting of risk factors

A. Fraction tested B. Test yield
Selected candidates for under-weighting
 Prostate cancer (CCW) 0.0370 0.1019
 No prostate cancer (CCW) 0.0380 0.0677
 Black 0.0313 0.0851
 Non-black 0.0385 0.0682
 History of PE 0.0726 0.1881
 No history of PE 0.0378 0.0686
 History of deep vein thrombosis 0.0507 0.1656
 No history of deep vein thrombosis 0.0378 0.0685
 Prior hospital visit within 30 days 0.0465 0.1976
 No prior hospital visit within 30 days 0.0377 0.0656
Selected candidates for over-weighting
 Chronic obstructive pulmonary disease (CCW) 0.0466 0.0524
 No chronic obstructive pulmonary disease (CCW) 0.0360 0.0742
 Ischemic heart disease (CCW) 0.0376 0.0566
 No ischemic heart disease (CCW) 0.0382 0.0786
 Atrial fibrillation (CCW) 0.0317 0.0520
 No atrial fibrillation (CCW) 0.0388 0.0713

Notes: Table reports summary statistics for selected comorbidities to motivate the examination of misweighting. Variables are selected on the Column A reports average rates of testing for patients with and without the listed conditions. Column B reports average rate of positive tests among tested patients with and without the listed conditions. CCW notes comorbidity is coded by the Chronic Condition Warehouse. Data is from the Medicare claims 2000–2009.