Ahmed Khwaja, Gabriel Picone, Martin Salm, and Justin Trogdon, "A Comparison of Treatment Effects Estimators Using a Structural Model of AMI Treatment Choices and Severity of Illness Information from Hospital Charts", Journal of Applied Econometrics, Vol. 26, No. 5, 2011, pp. 825-853. The primary source of data is the Cooperative Cardiovascular Project (CCP). The CCP sample consists of randomly selected patient records for Medicare patients admitted to nonfederal acute-care hospitals in the US with a primary diagnosis of acute myocardial infarction (AMI), or heart attack, (ICD-9-CM 410, excluding a fifth digit of 2) over an eight-month period. All of the sampling occurred between February 1994 and July 1995. The sample includes all hospitals in the US that had not participated in a four-state pilot phase (Alabama, Connecticut, Iowa, and Wisconsin). Medical records for each sampled hospitalization were forwarded to clinical abstraction centers. Altogether, charts were abstracted for approximately 180,000 AMI patients. The CCP took steps to ensure consistent data collection, including comparison of randomly selected records across two independent data abstraction centers. The results indicate a high degree of consistency in data abstraction. Patient identifiers were randomized in the CCP. For use in this project, all hospital identifiers and zip codes for patients and hospitals have been removed from the analysis data. The Journal of Applied Econometrics requires authors to make their data available. The original data are stored at the Duke Clinical Research Institute (www.dcri.duke.edu) and cannot be accessed by public users. We will only be making our simulated data available. No patient, hospital, or zip code identifiers are included. We randomly draw 1000 people once from the CCP data and, using their age, race, and hospital choice sets, simulate all other variables in the empirical models using our structural data generating process. In this way, we simulate choices and transitions for this sample of 1000 individuals. We repeat this 100 times for each person, ending up with 100,000 simulated histories. Below is information on the simulated data: Filename: mcsim_late_5.dat Space delimited file in DOS format. Length = 100,000 observations Width = 14 variables The file mcsim_late_5.dat is zipped in kpst-data.zip. Unix/Linux users should use "unzip -a". Variables (in order): Constant (ones) Age: 1 (65-74), 2 (75-84), 3 (85+) Sex: 1 (Male), 2 (Female) Race: 1 (White), 2 (Non-white) MSA: 1 (No), 2 (Yes) Charlson index: 1 (1) , 2 (2-3), 3 (>3) Distance (decile): 1-10 [see Table 3 for represented distances] Hospital service: 1 (no service), 2 (cath-only), 3 (surgery) Hospital surgery volume: 1 (0-49), 2 (50-99), 3 (100-199), 4 (200+) Killip class: 1, 2, 3 Catheterization/Transfer: 1 (no tran/no cath), 2 (no tran/cath), 3 (tran/no cath), 4 (tran/cath) Blockage: 1 (normal), 2 (mild/moderate reduction), 3 (severe reduction LVEF) Surgery/Transfer: 1 (no tran/no surgery), 2 (no tran/surgery), 3 (tran/no surgery), 4 (tran/surgery) 1-year mortality: 1 (Alive), 2 (Dead) For questions about the data, please contact the authors: Ahmed Khwaja (ahmedk [AT] duke.edu) Gabriel Picone (GPICONE [AT] coba.usf.edu) Martin Salm (m.salm [AT] uvt.nl) Justin Trogdon (jtrogdon [AT] rti.org)