PURPOSE: To examine the effects of patient, physician, and health care system characteristics on primary care physicians' (PCPs') specialty referral decision making. METHODS: Physicians (n=142) and their practices (n=83) located in 30 states completed background questionnaires and collected survey data for all patient visits (n=34,069) made during 15 consecutive workdays. The authors modeled the occurrence of any specialty referral, which occurred during 5.2% of visits, as a function of patient, physician, and health care system structural characteristics. A subanalysis was done to examine determinants of referrals made for discretionary indications (17% of referrals), operationalized as problems commonly managed by PCPs, high level of diagnostic and therapeutic certainty, low urgency for specialist involvement, and cognitive assistance only requested from the specialist. RESULTS: Patient characteristics had the largest effects in the any-referral model. Other variables associated with an increased risk of referral included PCPs with less tolerance of uncertainty, larger practice size, health plans with gate-keeping arrangements, and practices with high levels of managed care. The risk of a referral being made for discretionary reasons was increased by capitated primary care payment, internal medicine specialty of the PCP, high concentration of specialists in the community, and higher levels of managed care in the practice. CONCLUSIONS: PCPs' referral decisions are influenced by a complex mix of patient, physician, and health care system structural characteristics. Factors associated with more discretionary referrals may lower PCPs' thresholds for referring problems that could have been managed in their entirety within primary care settings.
PURPOSE: To examine the effects of patient, physician, and health care system characteristics on primary care physicians' (PCPs') specialty referral decision making. METHODS: Physicians (n=142) and their practices (n=83) located in 30 states completed background questionnaires and collected survey data for all patient visits (n=34,069) made during 15 consecutive workdays. The authors modeled the occurrence of any specialty referral, which occurred during 5.2% of visits, as a function of patient, physician, and health care system structural characteristics. A subanalysis was done to examine determinants of referrals made for discretionary indications (17% of referrals), operationalized as problems commonly managed by PCPs, high level of diagnostic and therapeutic certainty, low urgency for specialist involvement, and cognitive assistance only requested from the specialist. RESULTS:Patient characteristics had the largest effects in the any-referral model. Other variables associated with an increased risk of referral included PCPs with less tolerance of uncertainty, larger practice size, health plans with gate-keeping arrangements, and practices with high levels of managed care. The risk of a referral being made for discretionary reasons was increased by capitated primary care payment, internal medicine specialty of the PCP, high concentration of specialists in the community, and higher levels of managed care in the practice. CONCLUSIONS: PCPs' referral decisions are influenced by a complex mix of patient, physician, and health care system structural characteristics. Factors associated with more discretionary referrals may lower PCPs' thresholds for referring problems that could have been managed in their entirety within primary care settings.
Authors: C Jessica Dine; Lisa M Bellini; Gretchen Diemer; Allison Ferris; Ashish Rana; Gina Simoncini; William Surkis; Charles Rothschild; David A Asch; Judy A Shea; Andrew J Epstein Journal: J Grad Med Educ Date: 2015-12
Authors: Megumi J Okumura; Michele Heisler; Matthew M Davis; Michael D Cabana; Sonya Demonner; Eve A Kerr Journal: J Gen Intern Med Date: 2008-07-26 Impact factor: 5.128
Authors: Eva H DuGoff; Justin E Bekelman; Elizabeth A Stuart; Katrina Armstrong; Craig Evan Pollack Journal: Health Serv Res Date: 2014-01-24 Impact factor: 3.402
Authors: Liesbeth Borgermans; Geert Goderis; Carine Van Den Broeke; Geert Verbeke; An Carbonez; Anna Ivanova; Chantal Mathieu; Bert Aertgeerts; Jan Heyrman; Richard Grol Journal: BMC Health Serv Res Date: 2009-10-07 Impact factor: 2.655