Industry Payments to Physicians and Prescribing Branded Memantine and Donepezil Combination

Abstract

Background

Once-daily extended-released memantine with donepezil (hereafter memantine/donepezil) may improve medication adherence but has a 60-fold higher cost compared with combined generic components. Little is known about factors associated with prescribing memantine/donepezil. We examined the association between pharmaceutical industry payments to physicians and prescribing memantine/donepezil in Medicare.

Methods

A cross-sectional study was conducted. Using 2015–2016 Centers for Medicare and Medicaid Services Open Payments and Part D prescription databases, we identified unique physicians who prescribed ≥11 memantine/donepezil prescriptions from 2015 to 2016. Outcome variable was the number of memantine/donepezil prescriptions written per physician per year. The key independent variable was physician receipt of industry payments defined in 2 models: (1) number of payments and (2) amount of payment ($100 units) for memantine/donepezil received per physician per year. Multivariable Poisson regression was used, adjusting for potential confounders.

Results

Among 4,895 unique eligible physicians in 2015–2016, the median number of memantine/donepezil prescriptions per physician per year was 19.5 (25th percentile 13, 75th percentile 32). Physicians received between 0 and 75 payments per year (median 1, 25th percentile 0, 75th percentile 2.5) for memantine/donepezil, totaling an average of $92 per year (median $10.5, 25th percentile $0, 75th percentile $33.20). Every 1 additional payment received was associated with a 2% increase in new memantine/donepezil prescriptions prescribed per physician per year (rate ratio [RR] 1.02, 95% confidence interval [CI] 1.02–1.02). Every $100 increase in payment for memantine/donepezil was associated with a 0.3% increase in new memantine/donepezil prescriptions prescribed per physician per pear (RR 1.003, 95% CI 1.002–1.004).

Conclusions

Receipt of industry payments for memantine/donepezil was independently associated with increased likelihood of physician prescribing memantine/donepezil in Medicare.

The US Food and Drug Administration (FDA) approved once-daily extended-released memantine/donepezil combination (Namzaric; Dublin, Ireland; hereafter memantine/donepezil) for moderate to severe Alzheimer dementia in patients stabilized on donepezil and/or memantine therapy in December 2014.¹ This brand product has a 60-fold higher monthly cost (per the 2018 National Average Drug Acquisition Cost) compared with its combined generic components (cost for 30-day supply: memantine/donepezil combination, $405; memantine 10 mg twice daily, $5.40, donepezil 10 mg once daily, $1.35).² Convenient administration and improved adherence (by combining 2 medications, one of which is dosed twice daily, into a single once-daily product) are the only potential benefits of such a combination, but at a considerably higher medication cost.¹ Previous literature reviews have highlighted that medication adherence is a common challenge in patients with Alzheimer disease, although it is a relatively understudied area.^3,4 Because pharmaceutical industry payments to physicians have been shown to be associated with increased prescribing of high-cost medications, we sought to determine whether industry influence was a driver of prescriptions for memantine/donepezil.⁵ In this study, we examined the association between pharmaceutical industry payments to physicians for memantine/donepezil and prescribing memantine/donepezil in Medicare.

Methods

This cross-sectional study linked the ProPublica's Dollars for Docs news application data set (hereafter Dollars for Docs) to Medicare Part D Prescriber Public Use Files (MPDPUF) for the years 2015 and 2016 using National Provider Identifiers (NPIs).^6,7 Dollars for Docs is based on CMS's Open Payments data and includes drug and manufacturer names and NPIs, allowing identification of physician-level pharmaceutical industry payments for memantine/donepezil.⁶ MPDPUF includes physician-level data on number of drugs prescribed per year to Medicare Part D beneficiaries and identifies providers by their NPI.⁷

We first identified unique physicians who prescribed ≥11 memantine/donepezil prescription in 2015 and/or 2016. Physicians with fewer than 11 prescriptions are not included in MPDPUF due to the CMS' data suppression policy. The outcome variable was number of prescriptions for memantine/donepezil written per physician per year. The independent variable was physician receipt of industry payments for memantine/donepezil measured separately in 2 models: (1) number of payments for memantine/donepezil received per physician per year and (2) amount of payment (every $100 unit) for memantine/donepezil received per physician per year.

Covariates included physician specialty (primary care, geriatrics, internal medicine specialties, neurology, and others), physician sex, annual physician prescribing volume for Alzheimer disease prescriptions (excluding memantine/donepezil but including the individual drugs, memantine and donepezil), physician census region (Northeast, Midwest, South, Pacific West, and Mountain West), rural/urban county based on the 2013 Rural-Urban Continuum Codes (ers.usda.gov/data-products/rural-urban-continuum-codes.aspx), and type of payment (food and beverage, promotional speaking, and travel and lodging).

We compared physician characteristics by those receiving vs not receiving payments for memantine/donepezil using the Student t test for continuous variables and the χ² test for categorical variables. Under the assumption that physicians included in the analysis were actively practicing each year, we used multivariable Poisson regression with robust standard errors to examine the association between pharmaceutical industry payments to physicians and prescribing memantine/donepezil, controlling for all covariates. We also conducted stratified analyses examining the association between pharmaceutical industry payments to physicians and prescribing memantine/donepezil by physician specialty. We restricted these analyses to specialties with sufficient numbers of physicians and those most likely to prescribe memantine/donepezil for clinical relevance. For comparison, we also conducted multivariable Poisson regression to examine the association between pharmaceutical industry payments to physicians and prescribing memantine and donepezil single-ingredient products, controlling for all covariates. However, there were too few payments made for donepezil, so only results for memantine are reported. Adjusted rate ratios (RRs) with 95% confidence intervals (CIs) were reported. All statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Cary, NC).

Standard Protocol Approvals, Registrations, and Patient Consents

The University of Washington Institutional Review Board determined this study to be exempt.

Data Availability

Data not provided in the article per ProPublica Data Terms of Use.

Results

Among the 4,895 unique eligible physicians who prescribed memantine/donepezil in 2015 and/or 2016, 52.9% (n = 2,589) received at least 1 payment for memantine/donepezil (table 1). Compared with those not receiving any payment for memantine/donepezil, physicians receiving at least 1 payment were more likely to be male, a neurologist, and to have a higher annual prescribing volume for Alzheimer disease prescriptions.

Table 1.

Physician Characteristics Prescribing Memantine/Donepezil by Receipt of Industry Payment

Histograms are reported of the total number of memantine/donepezil prescriptions per physician per year, total number of payments for memantine/donepezil per physician per year, and total amount of payments for memantine/donepezil per physician per year in figures 1, 2, and 3, respectively. Among the eligible physicians, the median number of memantine/donepezil prescriptions per physician per year was 19.5 prescriptions (25th percentile 13, 75th percentile 32). Physicians received between 0 and 75 payments per year (median 1, 25th percentile 0, 75th percentile 2.5) for memantine/donepezil, totaling an average of $92 per year (median $10.5, 25th percentile $0, 75th percentile $33.20) (total number of payments and total number of physicians who received a payment for memantine/donepezil combination and individual drug components by year in table e-1, links.lww.com/CPJ/A178). All physicians (100%) receiving payment for memantine/donepezil received food and beverage, followed by promotional speaking (2.7%) and travel and lodging (1.9%) (table 1, footnote).

Figure 1. Number of Memantine/Donepezil Prescriptions Per Physician Per Year in Medicare (2015–2016).

Figure 2. Number of Payments for Memantine/Donepezil Per Physician Per Year (2015–2016).

Figure 3. Amount of Payments for Memantine/Donepezil Per Physician Per Year (2015–2016).

Every 1 additional payment received was associated with a 2% increase in new memantine/donepezil prescriptions prescribed per physician per year (RR 1.02, 95% CI 1.02–1.02) (table 2). Every $100 increase in payment for memantine/donepezil was associated with a 0.3% increase in new memantine/donepezil prescriptions prescribed per physician per pear (RR 1.003, 95% CI 1.002–1.004). Results were consistent when excluding outlying values (>99th percentile) of industry payments for memantine/donepezil (data not shown).

Table 2.

Industry Payments to Physicians and Prescribing Memantine/Donepezil^a

Stratification Analyses by Prescriber Specialty

As shown in table e-2 (links.lww.com/CPJ/A178), the primary results were largely consistent among primary care physicians, internal medicine physicians, and neurologists whereby industry payments for memantine/donepezil were independently associated with increased likelihood of physician prescribing memantine/donepezil.

Industry Payments and Prescribing Memantine

Every 1 additional payment received for memantine was associated with a 5% increase in new memantine prescriptions prescribed per physician per year (RR 1.05, 95% CI 1.05–1.05) (table e-3, links.lww.com/CPJ/A178). Every $100 increase in payment for memantine was associated with a 6% increase in new memantine prescriptions prescribed per physician per pear (RR 1.06, 95% CI 1.06–1.06).

Discussion

This study examined the association between pharmaceutical industry payments to physicians for memantine/donepezil—a branded drug with a 60-fold higher cost and unsubstantiated therapeutic advantages compared with its combined generic components—and prescribing memantine/donepezil in Medicare Part D. We found that among physicians who prescribed memantine/donepezil prescriptions from 2015 to 2016, receipt of industry payments for memantine/donepezil, measured with either number or amount of payments, was independently associated with increased likelihood of physician prescribing memantine/donepezil. In addition, these associations were largely consistent among primary care physicians, internal medicine physicians, and neurologists. These associations were also consistent when examining the association between industry payments for memantine and prescribing memantine. Although the estimates of association observed in this analysis are small in magnitude, extrapolating these findings to the population level could be meaningful.

Pharmaceutical industry payments to physicians have been shown to increase prescribing of high-cost medications for conditions such as hypertension and diabetes,⁵ but little is known about the use of memantine/donepezil. It is important to understand factors driving the prescribing of high-cost drugs, especially in care of Alzheimer disease where health care costs are rising.⁸ Memantine/donepezil was marketed as an option to reduce pill burden and in turn to improve medication adherence.¹ Although evidence from other therapeutic areas suggests that fixed-dose combination products may be associated with improved medication adherence,⁹ we are not aware of any empirical evidence showing that memantine/donepezil improves adherence. Moreover, given the drastically higher costs compared with its combined generic components, the benefit of memantine/donepezil in the treatment of Alzheimer disease is unclear.

Previous research has shown that industry-sponsored meals are associated with increased prescribing of the promoted brand.¹⁰ Of note, Ahlawat et al. conducted a scoping descriptive analysis of pharmaceutical industry payments to neurologists in 2015 using CMS's Open Payments data and reported that 51% of neurologists received payments, with food and beverage as the most common type of payment.¹¹ Our analyses showed that every physician who received at least 1 payment for memantine/donepezil received food and beverage, with few being paid for promotional speaking or travel and lodging.

There are important limitations worth mentioning. We were unable to identify the temporal relationship between physician prescribing memantine/donepezil and receiving pharmaceutical industry payments from these cross-sectional analyses. Reverse causality (e.g., higher prescribers being more likely to be targeted for payment from industry) cannot be ruled out. In addition, we cannot rule out unmeasured confounders (e.g., practice setting, key opinion leaders, and patient preference) in these data. However, we share this limitation with previous studies using the same data.¹⁰ We are also not able to examine the association by type of neurologist (e.g., cognitive neurologist vs other subspecialty neurologist) due to lack of this information. Finally, we only observed 2 years of payment and prescribing data; it is unknown whether industry payments for or prescribing of memantine/donepezil have changed since 2016. We intentionally selected 2015 and 2016 as the years immediately following FDA approval of memantine/donepezil (in December 2014) to represent the period of time during which pharmaceutical industry payments would likely be most frequent and thus have the greatest effect, if any.

In conclusion, we confirm previous results showing an association between pharmaceutical industry payments to physicians and prescribing of high-cost medications using memantine/donepezil as an example. Further studies should examine whether this fixed dose combination product improves adherence among patients with Alzheimer disease.

Appendix. Authors

Study Funding

Z. A. Marcum was supported by the Agency for Healthcare Research & Quality under Grant K12HS022982. The funding source had no involvement in the study design, collection, analysis, or interpretation of data, writing of the report, or the decision to submit the article for publication.

Disclosure

The authors report no relevant disclosures. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.

TAKE-HOME POINTS

• → Physician receipt of pharmaceutical industry payment was independently associated with increased likelihood of physician prescribing branded memantine and donepezil combination (memantine/donepezil) in Medicare.

• → Every additional payment a physician received for memantine/donepezil was associated with a 2% increase in memantine/donepezil prescriptions prescribed per physician per year.

• → Every $100 increase in payment a physician received for memantine/donepezil was associated with a 0.3% increase in memantine/donepezil prescriptions prescribed per physician per year.