The CDC's Opioid Guideline Cut Prescribing Sharply. Among Older Cancer Patients, It Barely Moved.

When the CDC released its opioid prescribing guideline in 2016, the loudest warning was not about the patients it targeted but the ones it did not. The guidance explicitly excluded cancer pain, palliative care and end-of-life treatment, and clinicians, ethicists and patient groups spent the decade since arguing about whether that line would hold in ordinary practice.

A study published October 9 in Supportive Care in Cancer suggests it did. Using linked Medicare and cancer registry data, researchers found opioid prescribing fell steeply among older adults without cancer after the guideline took effect, while prescribing for older adults with breast, colon or lung cancer barely changed.

A Natural Experiment Buried in Medicare Claims

The team, led by Olajumoke A. Olateju of the University of Houston's PREMIER Center with collaborators at Penn State and Sungkyunkwan University, drew on the SEER-Medicare database, which joins population-level cancer registry records to federal insurance claims. They identified 42,931 beneficiaries with breast, colon or lung cancer and 108,218 patients without cancer, all continuously enrolled in Medicare Parts A, B and D and none in Medicare Advantage, whose prescription records would have been obscured. The pre-guideline window ran from January 2014 through November 2015; the post-guideline window from April 2016 through September 2017, with a three-month buffer so the transition months contaminated neither side.

Four outcomes were tracked within three months of a diagnosis or index claim: the monthly prescribing rate, days of opioid supply, average daily dose in morphine milligram equivalents, and high-risk co-prescribing with benzodiazepines, antidepressants or other sedatives. A difference-in-differences model adjusted for age, sex, race, comorbidities and prior medication use.

The divergence was sharp. Prescribing declined in both groups, but cancer patients' monthly prescribing rate fell 0.14 percentage points less, and their monthly days of supply dropped 3.07 days less. Average daily dose did not differ significantly between the groups, with a point estimate of 16.73 MME and a confidence interval running from minus 43.36 to 76.82. The guideline changed whether and how long opioids were dispensed far more than how strong the daily dose was for the patients who received them.

The Safety Message Landed, Just Not Here

Co-prescribing tells the same story with a harder edge. Combining opioids with benzodiazepines or other sedatives carries the FDA's strongest boxed warning, because both classes suppress breathing. Those pairings declined across the study population after the guideline, but significantly less among cancer patients, with p-values below 0.0001 for each comparison.

Whether that is reassuring depends on which explanation you accept. The 2016 guideline said plainly that it was aimed at chronic pain outside cancer care and end-of-life settings, and oncology bodies maintain separate pain frameworks that treat cancer pain as its own clinical problem. Clinicians appear to have honored the carve-out.

The alternative reading is less comfortable. Older adults are unusually vulnerable to sedative burden, with elevated risks of falls, fractures, cognitive impairment and respiratory depression, and benzodiazepines and antidepressants are frequently prescribed in that population for legitimate reasons. The study's claims data cannot distinguish appropriate prioritization of symptom control from a missed medication-safety review in a patient whose prescriptions several specialists manage at once.

The CDC revised the guideline in 2022, softening several of the 2016 recommendations. For anyone whose opioid use has moved past medical supervision into prescription drug misuse, the difference between a demographic trend and a personal trajectory is the whole story. The options are the same ones that apply to opioid addiction treatment more broadly: an assessment, behavioral therapy and, for many people, medication. A guideline can be written with an exception and still behave as intended.