Use of Peripheral μ-Opioid Receptor Antagonists for Treating Opioid-Induced Constipation among US Medicare Beneficiaries from 2014 to 2018

Introduction

Opioid-induced constipation (OIC) is the most common adverse event associated with opioid use, and affects 40%–80% of patients receiving opioids.^1,2 It is associated with patient distress and decreased quality of life, and health care utilization.^3,4 Traditional bowel regimens of one or more laxatives, together with conservative use of opioids, and use of adjunctive pain management strategies, improved fluid intake, mobility, and other behavioral steps form first-line therapy for both prevention and treatment of OIC. ³ In some patients where these measures do not relieve OIC, rectal-based interventions or treatment with more novel and costly agents (such as peripheral μ-opioid receptor antagonists [PAMORAs]) may be employed. Three PAMORAs are U.S. Food and Drug Administration approved for managing OIC—methylnaltrexone (FDA approved in 2008), naloxegol (in 2014), and naldemedine (in 2017). The label advises use in patients with advanced illness receiving palliative care and/or those with chronic pain, and guidelines recommend their use only after failure of combination traditional laxative therapy.^5,6 Thus, PAMORAs may be expected to be used in select patients, and more commonly by pain management and palliative care clinicians. Spending associated with supportive care interventions has emerged as an important cause of financial toxicity.^7,8 We evaluated the overall and per-specialty/profession use and associated spending of PAMORAs.

Methods

We evaluated trends in use and expenditures for the three PAMORAs from 2014 to 2018, using the Medicare Part D Prescription Drug Event data. ⁹ This data set contains outpatient medication expenditures for ∼75% of all Medicare beneficiaries (there are ∼60 million Medicare beneficiaries, predominantly [85%] >65 years of age). These drugs do not have generic formulations. We did not include alvimopan (a PAMORA but not approved for OIC) and lubiprostone (approved for treating OIC, but not a PAMORA). For each PAMORA, we extracted annual spending, number of beneficiaries who received a PAMORA prescription, number of claims for the PAMORA, and spending per beneficiary and claim. Spending was adjusted for inflation and presented in 2018 U.S. dollars. ¹⁰ Overall, and for prescribers with >10 prescriptions of a PAMORA, we analyzed the distribution by prescriber specialty using the 2018 Part D Prescriber Public Use File. ¹¹ Because the study used publicly available deidentified patient data, it was granted an exempt status by the University of Texas Southwestern institutional review board. We used Microsoft Excel v16.0 (Redmond, WA) and GraphPad Prism v7.0 (San Diego, CA) for analyses.

Results

From 2014 to 2018, aggregate spending on PAMORAs increased, from $13.6 to $150.9 million, and use increased, from 4221 to 72,592 beneficiaries (Fig. 1A, B). After FDA approval in 2014, naloxegol overtook methylnaltrexone in the number of users in 2015 and spending in 2016 (Fig. 1A, B), and remained the most heavily used drug in 2018 (Fig. 1A–C). Among the three PAMORAs, the per unit spending was highest for methylnaltrexone: it peaked in 2016 ($178) and declined in 2017 and 2018 (Fig. 1D) after the introduction of naloxegol and the availability of a methylnaltrexone tablet (oral) formulation. Despite this, overall spending on methylnaltrexone increased from 2016 to 2018, associated with increases in number of beneficiaries and claims (Fig. 1A–C). Mirroring the spending/unit, spending/claim was also highest for methylnaltrexone (Fig. 1E).

OPEN IN VIEWER

FIG. 1.

U.S. Medicare Part D Trends in Utilization of the three individual PAMORA drugs from 2014 to 2018. Numbers in U.S. dollars are inflation adjusted for 2018. (A) Total aggregate spending. (B) Number of beneficiaries. (C) Number of claims. (D) Average spending per dosage unit. (E) Average spending per claim. PAMORA, peripheral μ-opioid receptor antagonist. Color image is available online.

In 2018, 6989 unique prescribers used any PAMORA. Among them, the most common specialties/professions were family practice (20.2%), internal medicine (18.0%), and nurse practitioner (15.4%). These three also represented the highest volume and spending on PAMORAs. Of all PAMORA prescribers, 597 (8.5%) prescribers prescribed PAMORAs to >10 beneficiaries. The most common specialties/professions among them were nurse practitioners (19.6%), anesthesiology (13.2%), and physician assistant (12.2%).

Discussion

Over a five-year period from 2014 to 2018, the number of beneficiaries served and overall spending on PAMORAs rose sharply among Medicare beneficiaries. Although market entry of newer PAMORA formulations was associated with decreased average spending per dosage unit of the existing PAMORA (methylnaltrexone), the overall spending on all PAMORAs steadily increased, driven by higher number of beneficiaries and claims, and PAMORA use extending across multiple medical specialties.

When placing rising PAMORA use along the extensive marketing of these agents in the wider context of chronic opioid use, the spending on PAMORAs reflects downstream costs of an opioid-based pain management strategy. An example of widespread marketing is an advertisement by a drug manufacturer during the 2016 Super Bowl. ¹² Our data do not account for the appropriateness of PAMORA therapy, which we could not evaluate using this database. The 2019 American Gastroenterological Association OIC guidelines recommend PAMORAs over no treatment when an adequate trial of traditional laxatives results in suboptimal symptom control (strong recommendation for naldemedine and naloxegol, conditional for methylnaltrexone). ⁵ They defined adequate trial as the combined use of two or more laxatives, on a scheduled (versus, as-needed basis). The guidelines do not support use of PAMORAs for primary prophylaxis of OIC and advise judicious use given their cost and lack of data beyond a few weeks. ⁵

In this study, specialties/professions with the highest PAMORA prescriptions mirrored those with the highest rates of opioid prescribing in an earlier analyses of IQVIA data from 2016 to 2017. ¹³ Notably, PAMORA use by hospice/palliative medicine clinicians and oncologists in this study, two groups traditionally thought to prescribe frequent opioids, was low. Anesthesiologists often serve to manage complex pain in the United States, and that may explain high rates of PAMORAs among them. Any efforts to promote thoughtful PAMORA use should involve and target multiple medical specialties.

It is also important for prescribers to note the limits of PAMORAs for treating OIC. Although OIC may be a prominent contributor to constipation among patients receiving opioids, patients can have multiple potential etiologies for constipation (dehydration, inactivity, concurrent use of anticholinergic medications, etc.), which are common in many hospice/palliative care patients. ¹⁴ Addressing these etiologies remains important even if patients are receiving PAMORAs, and may need additional nonpharmacologic and pharmacologic interventions.

This study has limitations. First, we did not account for manufacturer rebates/discounts. Second, methylnaltrexone is available in both oral and subcutaneous forms, and its spending was reported across both routes. The drop in per unit price of methylnaltrexone in 2017 and 2018 may be related to the emergence of competitor products, but also the availability and subsequent prescription of an oral methylnaltrexone formulation (which would be cheaper than a subcutaneous injection). Third, given the lack of clinical information in this database, we could not assess the appropriateness of therapy. Some of the spending on PAMORAs could represent appropriate care and a “good investment” if it prevented OIC-related distress to the patient and OIC-related health care use and expenditure to the health system. Lastly, prescribers could be misclassified (i.e., an oncology physician listed as an internal medicine physician) and we could not determine which specialties nurse practioners and physician assistants represented.

Conclusions

Our findings—significant and increasing expenditure on PAMORAs, and use across specialties—serve as a call for first defining and then implementing the appropriate use of PAMORAs, perhaps through incorporation into professional society and institution/health-system specific guidelines to improve prescribing practices. Certain institutions and payers already enforce checks (pharmacy approval, prior authorization, step therapy) to prevent over prescription. The role of marketing to patients and clinicians, and industry payments to clinicians as drivers of PAMORA use should be investigated.