 |
The 130th Annual Meeting of APHA |
|
The 130th Annual Meeting of APHA |
Christopher H. Young, PhD, IMR, AdvancePCS, Inc., 9501 E. Shea Blvd., Scottsdale, AZ 85260, 480-314-8224, Christopher.Young@pcshs.com
Following the lead of the literature on network analysis (White, Boorman, and Breiger, 1976), a network of physician ties is constructed by treating every patient who sees more than one physician as a “tie” between two physicians. A hypothesis based on information exchange would suggest that when a patient is referred from one physician to another that this would be an opportunity for an exchange of information as well. If this is the case, different patterns of prescribing should exist between physicians with and without network ties. The data for this investigation was obtained from the AdvancePCS warehouse of prescription claims. AdvancePCS provides healthcare coverage for approximately 75 million Americans, and processes over 1 million claims a day. From this I extracted a dataset of six months of prescription claims for Proton Pump Inhibitors (PPIs) as well as the H2 Antagonists. The period of time under investigation was from April 1, 2001 to September 30, 2001, the initial launch of Nexium. In order to avoid tautological conclusions, the data was divided into two datasets of three months each – the first used to construct the network data, and the second, counts of prescriptions. Physicians were categorized into those with no referrals to or from other physicians, those with one to three referrals, and those with four or more. Of the 221,983 physicians in the study, 96,628 (44%) had one or more ties to other physicians. Further, gastroenterologists were distinguished from primary care physicians, and from all other specialties. Physicians who had prescribed one of the newer PPIs (Protonix, Aciphex, and Nexium) were distinguished those who had not. An analysis of network ties of physicians revealed that 82% of the physicians with four or more network ties had prescribed the newer PPIs in contrast to only 41% of physicians with no network ties and 49% of those with up to three network ties. Similarly, 86% of gastroenterologists had prescribed one or more of the newer PPIs, while 60% of primary care physicians, and 32% of others had prescribed these drugs. Of Gastroenterologists with four or more ties to other physicians, 98% had prescribed the newer PPIs. Eighty-five percent of primary care physicians with four or more ties had prescribed the newer PPIs. These results suggest that physicians who have more ties to other physicians may be more likely to use the newer drugs than physicians who are relatively isolated from their colleagues.
Learning Objectives:
Keywords: Behavioral Research, Outcomes Research
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.