Social Science & Medicine 56 (2003) 1727–1736 The influence of cost-effectiveness information on physicians’ cancer screening recommendations PeterA.Ubel a,b,c, *,ChristopherJepson d ,JonathanBaron e,f,g ,JohnC.Hershey e,f,g , David A. Asch d,e,g,h a VA Health Services Research & Development Center of Excellence, VA Ann Arbor Healthcare System, Ann Arbor, MI, USA b Division of General Internal Medicine, University of Michigan, USA c Program for Improving Heath Care Decisions, University of Michigan, 300 North Ingalls, Room 7B20 Ann Arbor, MI 48109-0429, USA d Division of General Internal Medicine, University of Pennsylvania School of Medicine, USA e Leonard Davis Institute of Health Economics, University of Pennsylvania, USA f Department of Psychology, University of Pennsylvania, USA g The Wharton School, University of Pennsylvania, USA h Veterans Affairs Medical Center, Philadelphia, Pennsylvania, USA Abstract Physicians are increasingly faced with choices in which one screening strategy is both more effective and more expensivethananother.Onewaytomakesuchchoicesistoexaminethecost-effectivenessofthemorecostlystrategy over the less costly one. However, little is known about how cost-effectiveness information influences physicians’ screening decisions. We surveyed 900 primary care US physicians, and presented each with a hypothetical cancer- screeningscenario.Wecreatedthreefamiliarscreeningscenarios,involvingcervical,colon,andbreastcancer.Wealso created three unfamiliar screening scenarios. Physicians were randomized to receive one of nine questionnaires, each containing one screening scenario. Three questionnaires posed one of the familiar screening scenarios without cost- effectivenessinformation,threeposedoneofthefamiliarscenarioswithcost-effectivenessinformation,andthreeposed one of the unfamiliar scenarios with cost-effectiveness information. The cost-effectiveness information for familiar scenarios was drawn from the medical literature. The cost-effectiveness information for unfamiliar scenarios was fabricated to match that of a corresponding familiar scenario. In all questionnaires, physicians were asked what screening alternative they would recommend. A total of 560 physicians responded (65%). For familiar scenarios, providing cost-effectiveness information had at most a small influence on physicians’ screening recommendations; it reduced the proportion of physicians recommending annual Pap smears ðp ¼ 0:003Þ; butdidnotsignificantlyalterthe aggressivenessofcoloncancerandbreastcancerscreening(both p 0 so0.1).Forallthreeunfamiliarscenarios,physicians were significantly less likely to recommend expensive screening strategies than in corresponding familiar scenarios (all p 0 so0.001). Physicians’ written explanations revealed a number of factors that moderated the influence of cost- effectivenessinformationontheirscreeningrecommendations.Providingphysicianswithcost-effectivenessinformation hadonlyamoderateinfluenceontheirscreeningrecommendationsforcervical,colon,andbreastcancer.Significantly, fewerphysiciansrecommendedaggressivescreeningforunfamiliarcancersthanforfamiliarones,despitesimilarcost- effectiveness. Physicians are relatively reluctant to abandon common screening strategies, even when they learn that theyareexpensive,andarehesitanttoadoptunfamiliarscreeningstrategies,evenwhentheylearnthattheyareinexpensive. Published by Elsevier Science Ltd. Keywords: Decision making; Physician survey; Cost-effectiveness; USA *Correspondingauthor.ProgramforImprovingHeathCareDecisions,UniversityofMichigan,300NorthIngalls,Room7B20Ann Arbor, MI 48109-0429, USA. Fax: +1-734-936-8944. E-mail address: paubel@umich.edu (P.A. Ubel). 0277-9536/03/$-see front matter Published by Elsevier Science Ltd. PII:S0277-9536(02)00167-3