Reduced Opioid Prescribing by Oncologists: Progress Made, or Ground Lost?

The United States is in the midst of an opioid epidemic that is killing 130 people each day (1). The origins of this epidemic are complex, arising in part from aggressive pharmaceutical marketing of opioids in the late 1990s, campaigns to screen for pain as the “fifth vital sign,” and misleading narratives implying that opioids were safe and effective for noncancer pain (2–4). These powerful yet largely unrecognized forces led to a quadrupling of opioid prescribing, which peaked in 2010-2012 (5). Simultaneously, there was increasing recognition that cancer pain was undertreated (6, 7), and alleviating it became a priority for oncologists and the emerging field of palliative care (8). Influential organizations, including the World Health Organization (9), the Agency for Healthcare Research and Quality (10), and the Institute of Medicine (11), took notice, issuing guidelines emphasizing the importance of preserving opioid access for cancer patients, particularly at the end of life. As progress was made against cancer pain (12), an emerging epidemic went largely unnoticed until the mid-to-late 2000s, when deaths from prescription opioids surpassed those from heroin and cocaine combined (13–15).

As the scope of the opioid addiction crisis was recognized, multiple legislative and policy efforts were implemented to clamp down on inappropriate prescribing (16–18). These regulations have reduced opioid prescribing in recent years, although rates remain 3-fold higher than they were in the late 1990s (5, 19). While regulations increased and prescribing decreased, reports emerged that cancer patients were having progressive difficulty accessing opioid analgesics (20). Advocacy organizations raised concerns that cancer patients were inadequately protected from heightened restrictions, and hard-fought progress against cancer pain was being lost (20–22). Despite these concerns, little has been known about national trends in opioid access for cancer patients.

In a timely examination of Medicare Part D provider data published in this issue of the Journal, Jairam and colleagues (23) report that national rates of opioid prescribing by oncologists fell by 20.7% from 2013 to 2017, corroborating a recent report from Agrawal and colleagues (24). Trends were similar among non oncologists, whereas opioid prescribing by palliative care physicians increased by 15%. Reductions were primarily driven by a 30% decline in hydrocodone-acetaminophen prescribing (corresponding to its reclassification as a schedule III medication in 2014) and by similar declines in long-acting opioid prescriptions—most notably, oxycontin. Jairam et al. (23) also note wide geographic variability in trends, with 5 states experiencing greater reductions in opioid prescribing among oncologists than nononcologists.

Although it is not possible from these data to determine the cause of these trends, a reasonable assumption is that it is related to the seismic shifts in prescribing regulations and attitudes toward opioids. The period studied overlaps with the rapid expansion of state opioid legislation establishing prescription drug–monitoring programs, mandating provider education, and requiring patient identification and pharmacist verification before opioid dispensing (16, 17). Prescription drug plans also began imposing limits on the quantity, dose, or duration of opioid prescriptions (25), further reducing prescribing (26). Although a cancer diagnosis permits exceptions from some of these regulations, onerous amounts of additional work are often required of providers. Writing an opioid prescription has become a complex process involving signing controlled substance agreements, checking prescription drug–monitoring programs, responding to phone queries from conscientious pharmacists, filing prior authorization paperwork, and even rewriting prescriptions to comply with seemingly arbitrary and sometimes conflicting rules set by states, insurers, and pharmacies. These burdens likely disincentivize oncologists from prescribing, potentially shifting this responsibility to palliative care—as evidenced by the 15% increase in opioid prescribing observed among palliative care physicians. Palliative care is still a relatively small and primarily hospital-based specialty; whether it is capable of taking over pain management for all of these patients remains to be seen. It is important to note that opioid regulations have increased substantially since the tail end of the study period when the Centers for Disease Control and Prevention published influential guidelines on opioid therapy for chronic noncancer pain (27). Following these guidelines, more than one-half of states have implemented 7- to 14-day limits on initial opioid prescriptions (16), and several commercial pharmacies followed suit with their own restrictions (28). It will therefore be critical to continue monitoring trends in oncologists’ opioid prescribing as more recent data become available. To protect cancer patients’ access, it is critical for policy solutions to lessen rather than add to oncology providers’ workload.

An important unanswered question is whether the reductions in opioid prescribing observed were appropriate or inappropriate. To answer this, it is essential to determine whether the reductions occurred in patients on active treatment, cancer survivors, or those with advanced disease—for whom preserving opioid access is of utmost importance. Using Medicare beneficiary data from 2007 to 2016, we recently reported a 15% decline in the proportion of patients with poor prognosis cancers filling 1 or more opioid prescriptions near the end of life (29) and a 30% decline in the proportion filling a long-acting opioid. Taken together, these findings suggest that at least some of the prescribing reductions reported here have translated into reduced access for terminally ill cancer patients. Although experts generally agree that opioids are appropriate treatments for advanced cancer pain, the issue is more complicated for patients undergoing curative treatment or survivors (30), who are not immune to the risks of opioid tolerance and addiction (31, 32). Future research is needed to better understand how cancer patients at various phases of disease are affected by evolving access, whether there are disparities in access among vulnerable patient groups (eg, racial or ethnic minorities, or those residing in rural areas), and ultimately whether reductions in oncologists’ opioid prescribing represent progress made or gains lost in the fight against cancer pain.

Notes

Disclosures: The authors have no conflicts of interest to disclose.