Tramadol use in the United States, 2004-2021

Background: Tramadol has been associated with increased risk of chronic opioid use, emergency room (ER) visits, and death. However, little is known about outpatient trends in tramadol use in the US, particularly over the past decade.

Objectives: To evaluate trends in tramadol use and patient characteristics.

Methods:

Methods: We conducted population-level trend analyses of tramadol use over time. We also conducted an individual-level retrospective cohort study of people initiating tramadol after a 180 day clean period. Data were sourced from Optum Clinformatics© commercial health insurance claims (2005-2021). We included adults aged 18 years and over with a tramadol dispensing in the analyses. National and state-specific trends in tramadol initiation and use were analyzed. Among people initiating tramadol, we compared patient sociodemographics, painful conditions, prescribers, concomitant use with other central nervous system (CNS) depressants. Amount of tramadol initiated was measured in morphine milligram equivalents (MMEs), days supplied, and persistent use. Analyses were stratified by age (18-64 vs ≥65 years) and time: 2005-2010, 2011-2015, and 2016-2021..

Results: Tramadol use increased from 2005-2014, decreasing thereafter, although not uniformly or in all states. Between 2005-2010 and 2016-2021, total MMEs of initial tramadol dispensations declined from median (IQR) of 300 (225-600) to 300 (150-300) in ages 18-64 years, and from 500 (300-700) to 300 (200-450) in ages≥65. Days supplied declined from median of 8 to 6 days in younger adults, and from 12 to 7 days in older adults. Tramadol initiation in primary care decreased from 45% to 31% in younger adults, and from 60% to 42% in older adults, increasing in ED settings from 6% to 11% among younger adults, and from 3% to 7% among older adults. Dental prescribing increased from 0.65% to 3.87% among ages 18-64 years, and 0.25% to 1.98% among ages 65+ years. Back and neck pain was the most common diagnosis, ranging 37-42% (younger-older) in 2016-2021. We observed increased abdominal pain, osteoarthritis, chronic kidney disease with dialysis, diabetic neuropathy, and bone fracture diagnoses. Opioid naive tramadol initiations rose from 64% (younger) and 66% (older), to 78% and 82%, respectively. Persistent tramadol use decreased from 6% to 4% (younger) and 11% to 7% (older). Co-dispensing of CNS drugs declined, except for gabapentinoids, rising from ~1.3% to ~2.5% in both.

Conclusions: Despite tightened federal regulations in 2014 and a subsequent decrease in use overall, tramadol use remains high, especially in older adults. The higher proportion of opioid naïve patients initiating tramadol after 2014 is concerning. Decreased tramadol prescribing in primary care and increased prescribing in ER and dental care settings may reflect sub-optimal care coordination between primary care providers and specialists. Declines in chronic tramadol use is promising yet requires further study. Rising gabapentinoid co-medication may be due to increased overall use of non-opioid pain medications. Shifting pain conditions and prescriber types among people initiating tramadol signal rapidly evolving use requiring safety surveillance. There are plenty of avenues to expand our understanding with targeted investments in research aimed at optimizing safety, effectiveness, and equity in opioid use.