Opioid Prescribing and Opioid-Related Health Outcomes Among Cancer Survivors

The US population of older long-term cancer survivors—Americans who are free of cancer 5-years post-cancer diagnosis and not receiving cancer treatment—is growing. The prevalence of pain among cancer survivors after curative treatment is approximately 40% and opioids are frequently prescribed to manage the pain. The purpose of this dissertation is to explore long-term opioid therapy and opioid-related harms in cancer survivors using Surveillance Epidemiology and End Results – Medicare linked datasets. First, we explored the temporal and geographical variation in long-term opioid therapy among cancer survivors in the United States. We found that long-term opioid therapy rates were highest in the south and lowest in the northeast and that long-term opioid therapy rates peaked in 2012 but declined until 2016. Second, we assessed if patient level pain conditions and provider specialties seen at outpatient visits by cancer were associated with long-term opioid therapy. We found that cancer survivors who had been diagnosed with chronic pain or noncancer pain conditions and who were treated by noncancer specialists were more likely to receive long-term opioid therapy. Third, we assessed if cancer survivors were more likely than noncancer controls—matched on age, gender, race, pain conditions, previous opioid use—to experience an opioid-related emergency department visit or hospitalization. We found that the incidence of opioid-related adverse events were five times higher among cancer survivors who used opioids previously than opioid naïve cancer survivors. We found cancer survivors were as likely as persons without cancer to experience an opioid-related emergency department visit or hospitalization. In conclusion, we found high prevalence rates of long-term opioid therapy that differed by time and US geographical region and the risk of an opioid-related emergency department visit and hospitalization is comparable between cancer survivors and persons without a history of cancer. Our findings support the idea that policies and guidelines should continue to promote and incentivize the use of nonpharmacological and nonopioid interventions for managing pain among older adults.