Race/Ethnicity, Nativity, andu.S. Adult Mortality* Robert A. Hummer, [Jniuersity of Texas at Austin Richard G. Rogers, IJniuersity of Colorado at Bowlder Charles B. Nam, Florida State [Jniuersity Felicia B. LeClere, Notre Dame lJniuersity .Obiectiue. This paper investigates how racelethnicity is associated with the mor- talitv risks of u's. adults, while considering th. infl,r.r,.e .i;;;;;;. Methods. Data come from the National Health Interview Surveyalational Death Index linked data set. we use discrete-time hazards models ,o .rri*",. the association be_ tween racelethnicity, nativity, and mortality. Results. Native-born blacks exhibit the highest odds of death among all age groups of adults, *hil; i.r.ig"-Lorn bl".k, and Asian Americans display especially row odds of death, p".ri-*r"ay among older adults. Mexican Americans and other Hispanics display intermediate risks of mortality similar to non-Hispanic whites. RaceTethnic diiferences in mortality are influenced bv nativitS with the groups having the highest p.;;;;r;il;f foreign_ born individuals experiencing lower mortaht| risks it.r, ,.'iglr, otf;erwise be the case. Foreign-born persons arso exhibit lower mortality tharinative-born persons across.nearly all age groups, with the magnitude of the advantage varying by racelethnicity. Conclusions. Race/ethnic aduit mortality differentials"are wide and are influenced by nativity, social factors, and health Lehaviors. B;.;;r" foreign- born individuals, compared to their native-born counterpartq ofrcn e*hibit differential health behavior patterns and health outcomes, researchers and policy_ makers must be aware of immigrant composition and in how the compositional mix affects mortality for racial/ethnic groups and for the whole popot*io., over many decades, researchers have documented mortality differentiars across U.S. racial/ethnic subpopulations. Most often investigated are dif_ ferences between non-Hispanic blacks and non-Hirp;i. ;hites, but _ "Direct.all correspondence to Robert A. Hummer, popuration Research center, university of Texas at Austin, 1800 Main Building, Austin, TX zgliz qe-maiL.tru--.rop...,rtexas.edu.) The data used here are for public us!-and a.. au"il"tt. f.;;'rl* ffio.,,u;'dJrrr.iro. H.rltr, statistics (NCHS). The authors gratefully acknowledge John Horm ;;Nci{il;;iris work on construding the linked data set used herl. rwe also tt"r"l st".ri"j ruilr- f;;;;;ting assis_ tance; Parker Frisbie, Bob rreymeyer, three anonymous reviewers, and the sso ediror for commenrs on an earlier version of the. p_aper; Nancy Mann for ediroriai "*i.r*..i"a supporr provided by grants from the National-Scienc. Fou.rd"tior, (SBR -9GL7760 "nJ in'[-letzlosy. Editor's nore: Reviewers were Daniel Lichter, Douglas S. Massey, ""a vi[i"- ii-*. l9!l+! SCIENCE QUARTERL! Volume 80, Number 1.,March 1999 @1'999 by the university of Texas press, p.o. box 7g19, Austin. TX 7g713-7g1,9