Trends and Trajectories of Gabapentinoid Use and Associations with Subsequent Health Outcomes and Healthcare Expenditures in the United States

Abstract

Background: Increasing use of gabapentinoids (i.e., gabapentin and pregabalin), especially in off-label use or concurrent use with opioids, has raised concerns of misuse/abuse of gabapentinoids in the United States (US). Little is known about the patient and prescriber characteristics most associated with gabapentinoid use, as well as utilization patterns of gabapentinoids and their associations with subsequent adverse health outcomes and healthcare expenditures in the US. Objectives: This dissertation aimed to examine (1) trends, patient and prescriber characteristics, and potential off-label use of gabapentinoids among the US ambulatory care visits, (2) dual-trajectories of opioid and gabapentinoid use and risk of adverse health outcomes among US Medicare beneficiaries, and (3) association between dual-trajectories of opioid and gabapentinoid use and healthcare expenditures in US Medicare. Methods: This dissertation included the data source from (1) National Ambulatory Medical Care Survey (NAMCS) data from 2003-2015 and (2) 5% national representative sample of Medicare data from 2011-2016. A multivariable logistic regression, group-based multi-trajectory models along with inverse probability of treatment weighted multivariable Cox proportional hazard model and inverse probability of treatment weighted multivariable generalized linear regression, with log link and gamma distribution, were used to examine the three objectives of this dissertation, respectively. Results: This dissertation yielded three important insights into the patterns of gabapentinoid use. First, we found that the ambulatory visits involving gabapentinoids quadrupled from 2003-2015 in US ambulatory settings. Half of the gabapentinoid visits had concurrent opioids and/or benzodiazepine use. Gabapentinoids were mainly prescribed by primary care physicians, and potential off-label use of gabapentinoids was overwhelmingly high. Second, we identified ten distinct dual-trajectories of opioid and gabapentinoid use among fee-for-service Medicare beneficiaries with fibromyalgia, low back pain, neuropathy, or osteoarthritis. Consistent high-dose opioid-only users and all consistent opioid and gabapentinoid users (regardless of doses) were associated with more than doubled risk of drug overdose, compared to opioid-only early discontinuers. Third, they also had higher concurrent healthcare expenditures among the identified distinct opioid and gabapentinoid trajectories among Medicare beneficiaries. Conclusions: The increasing trend and extensive off-label use of gabapentinoids, especially concurrent use with opioids identified from NAMCS data, highlight the greater need for an understanding of long-term safety of gabapentinoid use. The distinct opioid and gabapentinoid dose and duration patterns among fee-for-service Medicare beneficiaries were associated with different risk of adverse health outcomes and healthcare expenditures. High-dose opioid-only users and all consistent opioid and gabapentinoid users (regardless of doses) were associated with a higher risk of adverse health outcomes and healthcare expenditures, compared to opioid-only early discontinuers. Healthcare providers should consider carefully when prescribing the concurrent opioid and gabapentinoid use in clinical practice. When the co-administration is necessary, patients should be monitored closely and assessed the benefit-risk profiles on a regular basis.