Chronic opioid therapy (COT) is associated with various adverse outcomes, especially at higher doses, yet little is known about predictors of sustained higher-dose COT. This study aimed to ascertain, among higher-dose COT patients, the association of patient-perceived pros and cons of opioids with continued higher-dose use 1 year later. Patients (N=1229) in 2 large health plans prescribed ?50 mg morphine-equivalent dose (MED) per day for chronic noncancer pain completed a survey assessing opioid benefits and harms. The Prescribed Opioid Difficulties Scale questionnaire assessed psychosocial problems, concerns, benefits, and side effects related to opioid use. Logistic regression models estimated the associations of the reported benefits and problems with higher-dose continuation (?50 mg MED/d) versus dose reduction (<50 mg MED/d) 1 year later. Over 80% of participants continued higher-dose opioid use at 1 year, regardless of reported problems, concerns, side effects, pain reduction, or perceived helpfulness. Higher scores on the Prescribed Opioid Difficulties Scale Problems subscale (odds ratio=0.79, 95% confidence interval, 0.68-0.92) and Concerns subscale (odds ratio=0.76, 95% confidence interval, 0.65-0.90) were negatively associated with higher-dose use 1 year later. Other baseline measures (opioid helpfulness, reduction in pain, number of side effects, and side effect bothersomeness) were not significantly associated with continued higher-dose use. The large majority of patients continued using higher-dose opioids regardless of baseline characteristics. These findings suggest the difficulty of reducing opioid dose among chronic higher-dose opioid users.