You have accessJournal of UrologyProstate Cancer: Epidemiology & Natural History III1 Apr 2014MP78-02 CUMULATIVE INCIDENCE OF SKELETAL RELATED EVENTS IN MEN WITH METASTATIC PROSTATE CANCER BASED ON ALTERNATIVE ALGORITHMS Abdulla M. Abdulhalim, Eberechukwu Onukwugha, Corinne Woods, Yi Qian, Jorge Arellano, Arun Balakumaran, C. Daniel Mullins, and Arif Hussain Abdulla M. AbdulhalimAbdulla M. Abdulhalim More articles by this author , Eberechukwu OnukwughaEberechukwu Onukwugha More articles by this author , Corinne WoodsCorinne Woods More articles by this author , Yi QianYi Qian More articles by this author , Jorge ArellanoJorge Arellano More articles by this author , Arun BalakumaranArun Balakumaran More articles by this author , C. Daniel MullinsC. Daniel Mullins More articles by this author , and Arif HussainArif Hussain More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2014.02.2485AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Skeletal related events (SREs) are common in men with metastatic prostate cancer (mPC). Several algorithms have been proposed to identify SREs when using claims data. Our aim was to determine how key components of algorithms for identifying SREs from claims impact results on cumulative incidence of SREs in men with mPC. METHODS Algorithms for identifying SREs were developed and applied to data for men aged ≥66 years diagnosed with mPC between 2000-2009 in the SEER-Medicare datasets and followed through 12/31/2010 or until they were lost to follow up. Post-diagnosis SREs were identified using claims that indicated spinal cord compression (SCC), pathologic fracture (PF), surgery to bone (SB), or radiation (suggestive of bone palliative radiation, RAD). Two key components defined alternative algorithms: the window length to consolidate SRE claims into events (14-, 21-, and 28-day windows) and the diagnosis and procedure codes used to identify SREs (′conservative′ and ′extended,′ e.g., PF ′conservative′ definition used codes that specifically indicated pathologic fracture and PF ′extended′ definition used codes that indicated any fracture without accidents/falls within 14 days before fracture). The base case algorithm used a ′conservative′ definition with a 21-day window. RESULTS Among 8,997 mPC men, 3,618 (40.2%) experienced an SRE according to the base case algorithm. Among those with an SRE, 2076 (57%) experienced a subsequent SRE. A total of 8,550 SREs were observed during follow up, with RAD, PF, SCC, and SB constituting 57%, 25%, 10%, and 8%, respectively. (Table) Varying the code definition from ′conservative′ to ′extended′ increased the cumulative incidence of SRE by 65% (Range: 15% [RB] - 200% [PF]). Varying the window length ±7 days from the 21-day base case resulted in a 7% change in SRE cumulative incidence (Range: 0.2% [SB] - 10% [RAD]). CONCLUSIONS The estimated cumulative incidence of SREs using claims data is mostly driven by code definition, and to a lesser extent, by window length. Across the 4 SREs, RAD is most affected by window length and PF is most affected by code definition. Investigators using claims to identify SREs should explicitly define their algorithm for SREs, including the algorithm components based on SRE type. SRE Window Length 14 days 21 days 28 days Ext. Cons. Ext. Cons. Ext. Cons. All SREs 16,313 9,845 14,100 8,550 12,926 7,928 Radiation 6,855 (42) 5,910 (60) 5,638 (40) 4,907 (57) 5,073 (39) 4,439 (56) Pathologic fracture 7,342 (45) 2,318 (23) 6,505 (46) 2,125 (25) 5,990 (46) 2,029 (26) Spinal cord compression 1,473 (9) 974 (10) 1,319 (9) 880 (10) 1,226 (10) 823 (10) Surgery to bone∗ 643 (4) 643 (7) 638 (5) 638 (8) 637 (5) 637 (8) Ext.: Extended code definition; Cons.: Conservative code definition All numbers in parentheses are column %. ∗ Only one set of codes was used for surgery to bone © 2014FiguresReferencesRelatedDetails Volume 191Issue 4SApril 2014Page: e920-e921 Advertisement Copyright & Permissions© 2014MetricsAuthor Information Abdulla M. Abdulhalim More articles by this author Eberechukwu Onukwugha More articles by this author Corinne Woods More articles by this author Yi Qian More articles by this author Jorge Arellano More articles by this author Arun Balakumaran More articles by this author C. Daniel Mullins More articles by this author Arif Hussain More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...
Read full abstract