This study compared the characteristics of community nursing homes where veterans received their care with those of facilities that did not treat veterans from 1999 to 2002 using the Centers for Medicare and Medicaid Services (CMS) Online Survey Certification and Reporting system data merged with the CMS Minimum Data Set. A structure, process, and outcome model was used to examine whether the presence of per diem veterans had any impact on multidimensional quality measures. Facilities with any veterans were less likely to meet recommended nurse staffing standards; more likely to have patients with tube feeding, new catheterizations, and mobility restraints; and more likely to have actual harm citations and new pressure sores, plus quality-of-care, quality-of-life, and total deficiencies, than facilities without veterans. The implications of this study are that the U.S. Department of Veterans Affairs may need to examine its contracting policies with community facilities to understand both quality and selection effects that may be occurring. © 2007 Sage Publications.