Michelle S Keller1,2, Lyna Truong3, Allison M Mays4, Jack Needleman5, Mary Sue V Heilemann6, Teryl K Nuckols7. 1. Division of General Internal Medicine, Department of Medicine, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, CA, 90048, USA. Michelle.keller@cshs.org. 2. Department of Health Policy and Management, UCLA Fielding School of Public Health, Los Angeles, CA, USA. Michelle.keller@cshs.org. 3. Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, MI, USA. 4. Division of Geriatrics, Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, USA. 5. Department of Health Policy and Management, UCLA Fielding School of Public Health, Los Angeles, CA, USA. 6. UCLA School of Nursing, Los Angeles, CA, USA. 7. Division of General Internal Medicine, Department of Medicine, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, CA, 90048, USA.
Abstract
BACKGROUND: Given the risks of opioids, clinicians are under growing pressure to treat pain with non-opioid medications. Yet non-opioid analgesics such as non-steroidal anti-inflammatory drugs (NSAIDs) have their own risks: patients with kidney disease or gastrointestinal diseases can experience serious adverse events. We examined the likelihood that patients with back pain diagnoses and contraindications to NSAIDs and opioids received an opioid prescription in primary care. METHODS: We identified office visits for back pain from 2012 to 2017 and sampled the first office visit per patient per year (N = 24,543 visits). We created indicators reflecting contraindications for NSAIDs (kidney, liver, cardiovascular/cerebrovascular, and gastrointestinal diseases; concurrent or chronic use of anticoagulants/antiplatelets, chronic corticosteroid use) and opioids (depression, anxiety, substance use (SUD) and bipolar disorders, and concurrent benzodiazepines) and estimated four logistic regression models, with the one model including all patient visits and models 2-4 stratifying for previous opioid use. We estimated the population attributable risk for each contraindication. RESULTS: In our model with all patients-visits, patients received an opioid prescription at 4% of visits. The predicted probability (PP) of receiving an opioid was 4% without kidney disease vs. 7% with kidney disease; marginal effect (ME): 3%; 95%CI: 1-4%). For chronic or concurrent anticoagulant/antiplatelet prescriptions, the PPs were 4% vs. 6% (ME: 2%; 95%CI: 1-3%). For concurrent benzodiazepines, the PPs were 4% vs. 11% (ME: 7%, 95%CI: 5-9%) and for SUD, the PPs were 4% vs. 5% (ME: 1%, 95%CI: 0-3%). For the model including patients with previous long-term opioid use, the PPs for concurrent benzodiazepines were 25% vs. 24% (ME: -1%; 95%CI: - 18-16%). The population attributable risk (PAR) for NSAID and opioid contraindications was small. For kidney disease, the PAR was 0.16% (95%CI: 0.08-0.23%), 0.44% (95%CI: 0.30-0.58%) for anticoagulants and antiplatelets, 0.13% for substance use (95%CI: 0.03-0.22%) and 0.20% for concurrent benzodiazepine use (95%CI: 0.13-0.26%). CONCLUSIONS: Patients with diagnoses of kidney disease and concurrent use of anticoagulants/antiplatelet medications had a higher probability of receiving an opioid prescription at a primary care visit for low back pain, but these conditions do not explain a large proportion of the opioid prescriptions.
BACKGROUND: Given the risks of opioids, clinicians are under growing pressure to treat pain with non-opioid medications. Yet non-opioid analgesics such as non-steroidal anti-inflammatory drugs (NSAIDs) have their own risks: patients with kidney disease or gastrointestinal diseases can experience serious adverse events. We examined the likelihood that patients with back pain diagnoses and contraindications to NSAIDs and opioids received an opioid prescription in primary care. METHODS: We identified office visits for back pain from 2012 to 2017 and sampled the first office visit per patient per year (N = 24,543 visits). We created indicators reflecting contraindications for NSAIDs (kidney, liver, cardiovascular/cerebrovascular, and gastrointestinal diseases; concurrent or chronic use of anticoagulants/antiplatelets, chronic corticosteroid use) and opioids (depression, anxiety, substance use (SUD) and bipolar disorders, and concurrent benzodiazepines) and estimated four logistic regression models, with the one model including all patient visits and models 2-4 stratifying for previous opioid use. We estimated the population attributable risk for each contraindication. RESULTS: In our model with all patients-visits, patients received an opioid prescription at 4% of visits. The predicted probability (PP) of receiving an opioid was 4% without kidney disease vs. 7% with kidney disease; marginal effect (ME): 3%; 95%CI: 1-4%). For chronic or concurrent anticoagulant/antiplatelet prescriptions, the PPs were 4% vs. 6% (ME: 2%; 95%CI: 1-3%). For concurrent benzodiazepines, the PPs were 4% vs. 11% (ME: 7%, 95%CI: 5-9%) and for SUD, the PPs were 4% vs. 5% (ME: 1%, 95%CI: 0-3%). For the model including patients with previous long-term opioid use, the PPs for concurrent benzodiazepines were 25% vs. 24% (ME: -1%; 95%CI: - 18-16%). The population attributable risk (PAR) for NSAID and opioid contraindications was small. For kidney disease, the PAR was 0.16% (95%CI: 0.08-0.23%), 0.44% (95%CI: 0.30-0.58%) for anticoagulants and antiplatelets, 0.13% for substance use (95%CI: 0.03-0.22%) and 0.20% for concurrent benzodiazepine use (95%CI: 0.13-0.26%). CONCLUSIONS:Patients with diagnoses of kidney disease and concurrent use of anticoagulants/antiplatelet medications had a higher probability of receiving an opioid prescription at a primary care visit for low back pain, but these conditions do not explain a large proportion of the opioid prescriptions.
Authors: Joseph A Boscarino; Margaret Rukstalis; Stuart N Hoffman; John J Han; Porat M Erlich; Glenn S Gerhard; Walter F Stewart Journal: Addiction Date: 2010-08-16 Impact factor: 6.526
Authors: Paul M Ridker; Samuel Z Goldhaber; Ellie Danielson; Yves Rosenberg; Charles S Eby; Steven R Deitcher; Mary Cushman; Stephan Moll; Craig M Kessler; C Gregory Elliott; Rolf Paulson; Turnly Wong; Kenneth A Bauer; Bruce A Schwartz; Joseph P Miletich; Henri Bounameaux; Robert J Glynn Journal: N Engl J Med Date: 2003-02-24 Impact factor: 91.245
Authors: Katherine B Santosa; Yen-Ling Lai; Chad M Brummett; Jeremie D Oliver; Hsou-Mei Hu; Michael J Englesbe; Emilie M Blair; Jennifer F Waljee Journal: J Gen Intern Med Date: 2020-08-03 Impact factor: 5.128
Authors: Gary M Franklin; Bert D Stover; Judith A Turner; Deborah Fulton-Kehoe; Thomas M Wickizer Journal: Spine (Phila Pa 1976) Date: 2008-01-15 Impact factor: 3.468
Authors: Roger Chou; Richard Deyo; Janna Friedly; Andrea Skelly; Robin Hashimoto; Melissa Weimer; Rochelle Fu; Tracy Dana; Paul Kraegel; Jessica Griffin; Sara Grusing; Erika D Brodt Journal: Ann Intern Med Date: 2017-02-14 Impact factor: 25.391
Authors: Roger Chou; Richard Deyo; Janna Friedly; Andrea Skelly; Melissa Weimer; Rochelle Fu; Tracy Dana; Paul Kraegel; Jessica Griffin; Sara Grusing Journal: Ann Intern Med Date: 2017-02-14 Impact factor: 25.391