Abstract
Purpose: To characterize family physicians (FPs) who are stewards of care by consistently prescribing omeprazole over esomeprazole.
Methods: Cross-sectional analysis of physicians prescribing omeprazole or esomeprazole under Medicare Part D in 2014.
Results: There was a regional trend with 49% of Western FPs but only 6% of Southern FPs rarely prescribing esomeprazole. Physicians had increased odds of being a steward if they worked with a care coordinator (P < .001), at a patient-centered medical home (P < .001), or in a large practice (P < .001).
Conclusions: If these findings are replicated across multiple drugs, future outreach could be conducted based on provider prescribing patterns.
Responsible stewardship of health care expenditure is an important movement in an era of increasing costs. The Centers for Medicare and Medicaid Services publicly released comprehensive cost data for all medications in 2013 and 2014 under Medicare Part D.1,2 Brand-only esomeprazole was listed in the top 10 by cost despite the availability of generic therapeutic equivalent omeprazole that is well supported by medical literature.1⇓⇓⇓–5 Although esomeprazole contains only the active S-enantiomer whereas omeprazole contains an equal mix of the S-Enantiomer and R-enantiomer, many studies show no significant difference in efficacy.3⇓–5 Preferred and nonpreferred brand medications carry a higher total cost and copayment cost compared with generic medications under Medicare Part D prescription drug plans.6 Preferential prescribing of esomeprazole over omeprazole can result in a significant financial burden to Medicare Part D patients who paid an average of 10.5 times more for the branded medication.6 This data release presents an opportunity to evaluate prescribing habits of family physicians (FPs) to help target future evidence-based responsible stewardship interventions. The objective of this analysis was to identify and characterize stewards of cost-conscious prescribing.
Methods
We used Medicare Part D data from 2014 to examine prescribing patterns of physicians. This data lists which drugs each physician prescribed, how many days were prescribed, and the cost. We restricted this analysis to prescriptions of esomeprazole (introduced 2001) and generic prescriptions of omeprazole (introduced 1989). We defined stewards as physicians who prescribed esomeprazole less than 2% of the time rather than 0% as this accommodates rare events difficult to control by providers who may in normal circumstances never choose to prescribe esomeprazole. To exclude physicians who did not prescribe either drug frequently, we further restricted the analysis to physicians who prescribed 3650 days or more combined.
We used data from the American Board of Family Medicine to take a closer look at a sample of these physicians, board-certified FPs who recertified in 2014 or 2015. Information on practice size, working at a patient centered medical home (PCMH), and working alongside a care coordinator was self-reported as part of the recertification process. We excluded physicians initially certifying because we lacked data on their practice setting.
We obtained information about each physician’s sex, country of training, year of graduation from medical school, and type of degree obtained from the American Medical Association (AMA) Physician Masterfile. Roughly 94% of the physicians in our sample were successfully matched with AMA records. Each physician’s address was geocoded to determine region of the country and whether they practiced in a nonmetropolitan county as defined by a Rural-Urban Continuum Codes ≥ 4.
We calculated the percent of physicians identified as stewards across a number of physician demographics and practice characteristics. χ2 tests were used to identify statistically significant differences with a threshold of P < .05 used to determine significance. Multivariate logit regression was used to assess the relationship between the likelihood each physician was a steward and our identified physician and practice characteristics.
Results
There were 15,688 FPs who both appeared in Part D Data for 2014 and recertified in 2014 or 2015. There were 8,792 FPs who met our threshold for significant prescribing; this subset accounted for 94.2% of all omeprazole and esomeprazole recorded days supplied in the Part D data.
Overall, 15.0% of the prescriptions were for esomeprazole. The median physician prescribed esomeprazole 13.0% of the time. As defined as prescribing only esomeprazole less than 2% of the time, 2001 (22.8%) physicians met our criterion for being a steward (Figure 1). Approximately 27% of women were stewards, but only 20% of men were. (P < .001).
Distribution of physician prescribing behavior for choosing esomeprazole over omeprazole.
There were significant differences based on practice characteristics. Nearly half of physicians practicing in the West were stewards, but the same was true of only 1 in sixteen physicians from the South. (P < .001) Urban FPs were more likely to not prescribe esomeprazole than their rural counterparts. (P < .001) Physicians who practiced alongside 20 or more providers were 3 times as likely as solo practitioners. (P < .001) Physicians who worked in PCMHs were approximately 50% more likely to be stewards as those who did not (P < .001). The same was true for those working with care coordinators (P < .001).
Multivariate logit regression revealed that many of these differences remain significant when holding other factors constant (Figure 2). Region alone explains much of the variation. Compared with physicians in the Midwest, Southern physicians were one fifth as likely to be stewards (P < .001), and physicians in the Northeast were a little more than half as likely. In contrast, physicians practicing in the West were roughly 2.5 times as likely to be (P < .001). Practice size was also a key determinant. Physicians at medium and large practices were 63% (P < .001) and 145% (P < .001) more likely respectively to be stewards than solo practitioners. Independent of practice size, working with a care coordinator (P < .001) or in a PCMH (P < .001) each increased the likelihood of being a steward by approximately 50%. The effects of demographics were more modest with women being a third more likely than men to be stewards. Physicians who graduated medical school before 1980 were less likely to be stewards than those graduating since 2000.
Practice characteristics and other factors that increase likelihood of being a steward of health care expenditure as defined as prescribing generic omeprazole over esomeprazole (Nexium).
Discussion
In this analysis of Medicare Part D data, we found that 23% of our sample are prescribing stewards. Women, living in the West, and working in large practices that are PCMHs are associated with stewardship. Efforts to help physicians and patients engage in evidence-based conversations aimed at wise management and just distribution of finite resources were created with programs such as the Choosing Wisely campaign by the American Board of Internal Medicine Foundation.7,8 The campaign consists of recommendations based on topics of care but has not targeted specific geographic areas or populations based on physician or practice characteristics related to increased expenditure.
Although more research needs to be conducted, future outreach programs could be directed based on practice or provider specific data to improve prescribing habits. Academic detailing through evidence-based information by specially trained noncommercial outreach educators to physician offices has been shown to be effective in more than 60 randomized controlled trials and could be targeted to specific locations or practice characteristics from future studies such as this.9 Furthermore, with the emergence of health care systems, there are opportunities for information feedback where physicians could be rewarded for prescribing drugs optimally.9
Appendix
Demographics and Practice Characteristics
Notes
This article was externally peer reviewed.
Funding: none.
Conflict of interest: none declared.
To see this article online, please go to: http://jabfm.org/content/30/6/824.full.
- Received for publication February 5, 2017.
- Revision received May 12, 2017.
- Accepted for publication May 15, 2017.
