Association of new opioid continuation with surgical specialty and type in the United States.

BACKGROUND: The consequences of opioids-including post-surgical prescriptions-remain a critical public health issue. We sought to determine how procedure type and subspecialty group influence new opioid use after procedures. METHODS: We analyzed 2011-2015 IBM MarketScan Research Databases to identify opioid-naïve adults prescribed opioids for single surgical procedures. We defined new opioid continuation (primary outcome) a priori as receipt of prescription opioids between 90 and 180 days after the procedure. RESULTS: Among 912,882 individuals, new opioid continuation was higher for non-operating room compared to operating room procedures (13.1% versus 9.2%; aOR 1.61; 95% CI 1.59-1.64) and higher for subspecialties including colorectal surgery (aOR 1.35; 95% CI 1.26-1.43) and cardiovascular surgery (aOR 1.30; 95% CI 1.12-1.50) compared to urology as a referent. New opioid continuation was also associated with perioperative opioid prescription dosage, days' supply, preoperative receipt, and multiple prescriptions. CONCLUSIONS: Opioids prescriptions associated with non-operating room surgical exposures appear to confer higher risk regarding conversion to new long-term opioid use.

Use of non-insulin diabetes medicines after insulin initiation: A retrospective cohort study.

BACKGROUND: Clinical guidelines recommend that metformin be continued after insulin is initiated among patients with type 2 diabetes, yet little is known regarding how often metformin or other non-insulin diabetes medications are continued in this setting. METHODS: We conducted a retrospective cohort study to characterize rates and use patterns of six classes of non-insulin diabetes medications: biguanides (metformin), sulfonylureas, thiazolidinediones (TZDs), glucagon-like peptide 1 receptor agonists (GLP1 receptor agonists), dipeptidyl peptidase 4 inhibitors (DPP4 inhibitors), and sodium-glucose co-transporter inhibitors (SGLT2 inhibitors), among patients with type 2 diabetes initiating insulin. We used the 2010-2015 MarketScan Commercial Claims and Encounters data examining 72,971 patients with type 2 diabetes aged 18-65 years old who initiated insulin and had filled a prescription for a non-insulin diabetes medication in the 90 days prior to insulin initiation. Our primary outcome was the proportion of patients refilling the various non-insulin diabetes medications during the first 90 days after insulin initiation. We also used time-to-event analysis to characterize the time to discontinuation of specific medication classes. RESULTS: Metformin was the most common non-insulin medication used prior to insulin initiation (N = 53,017, 72.7%), followed by sulfonylureas (N = 25,439, 34.9%) and DPP4 inhibitors (N = 8,540, 11.7%). More than four out of five patients (N = 65,902, 84.7%) refilled prescriptions for any non-insulin diabetes medications within 90 days after insulin initiation. Within that period, metformin remained the most common medication with the highest continuation rate of 84.6%, followed by SGLT2 inhibitors (81.9%) and TZDs (79.3%). Sulfonylureas were the least likely medications to be continued (73.6% continuation) though they remained the second most common medication class used after insulin initiation. The median time to discontinuation varied by therapeutic class from the longest time to discontinuation of 26.4 months among metformin users to the shortest (3.0 months) among SGLT2 inhibitor users. CONCLUSION: While metformin was commonly continued among commercially insured adults starting insulin, rates of continuation of other non-insulin diabetes medications were also high. Further studies are needed to determine the comparative effectiveness and safety of continuing insulin secretagogues and newer diabetes medications after insulin initiation.

Using trajectory models to assess the effect of hydrocodone upscheduling among chronic hydrocodone users.

PURPOSE: In October 2014, the US Drug Enforcement Agency moved hydrocodone combination products (HCPs) from schedule III to II of the Controlled Substances Act, further restricting their access. The aim of the study is to quantify the effect of hydrocodone's "upscheduling" on the use of opioid and nonopioid analgesics among chronic users. METHODS: Using IQVIA LRx LifeLink anonymized pharmacy data 2013 to 2015, we performed interrupted time series analysis and group-based trajectory modeling to characterize the effect of rescheduling on 316 731 long-term hydrocodone users. Main measures were the number of prescriptions, patients, tablets, and morphine milligram equivalents of opioids and nonopioid analgesics pre and post the policy change. We used logistic regression to assess the relationship between sociodemographic characteristics and these measures. RESULTS: The schedule change was associated with significant declines in opioid prescriptions (20.9%, from 421 798 to 333 627) and the number of patients using opioids (11.4%, from 307 974 to 272 804). Majority of hydrocodone users filled prescriptions for nonopioid analgesics with some declines in the number of users after the schedule change (5.2%, from 181 085 to 171 758). Based on group-based trajectory models, majority of patients continued to fill HCP prescriptions consistently after the policy change, while 15.4% showed large declines in HCP use, accounting for two-thirds of the decrease in opioid volume. There was no evidence that the policy change was associated with significant increases in the use of alternative analgesics. CONCLUSIONS: The upscheduling of hydrocodone led to reductions in opioid use, which were concentrated among a small subset of chronic hydrocodone users, without evidence of commensurate increases in the use of alternative pharmacologic pain treatments.

Association Between Recombinant Human Bone Morphogenetic Proteins and Postoperative Opioid Use in Lumbar Fusion Procedure Patients: A Propensity Score-Matched Analysis.

BACKGROUND: Little is known about the effectiveness of recombinant human bone morphogenetic proteins (rhBMPs) in reducing the demand for opioids after surgery. We investigated the association between rhBMP use and the likelihood of achieving opioid independence and changes in opioid demand in the first postoperative year. METHODS: Using the Multi-Payer Claims Database from 2007 to 2010, patients aged >20 years who had undergone a degenerative disc disease-indicated lumbar fusion procedure and had had ≥1 opioid prescription in the 3 months before surgery were identified. Propensity score matching (1:1) of rhBMP-exposed and rhBMP-unexposed patients was used to mitigate confounding. The outcomes of interest were opioid independence and decreases in opioid doses in morphine equivalent units, assessed at 3-6 and 9-12 months after the procedure. Logistic regression and analysis of covariance models were used. RESULTS: The data from 318 patients were analyzed. Most patients were women (61%) and aged <65 years (68%). Few had achieved opioid independence at 3-6 (n = 71; 22.3%) or 9-12 (n = 115; 36.2%) months postoperatively. During the 3-6-month window, the rhBMP group reduced their opioid use rates (estimated mean difference. -28.4 vs. -19.5; P = 0.69) and achieved opioid independence (21.4% vs. 23.3%; odds ratio, 0.92; 95% confidence interval, 0.54-1.56; P = 0.74) at rates that were statistically comparable to their matched comparators. Similar patterns were observed during the 9-12-month window. CONCLUSION: We found no evidence to suggest that rhBMP use during spinal fusion procedures is associated with either the discontinuation or decrease of opioid analgesic therapy. The continued opioid use after surgery warrants further clinical and research attention.

Quality of Prescribing by Physicians, Nurse Practitioners, and Physician Assistants in the United States.

OBJECTIVE: Nurse practitioners (NPs) and physician assistants (PAs) have increasingly broad prescribing authority in the United States, yet little is known regarding how the quality of their prescribing practices compares with that of physicians. The objective of this study was to compare the quality of prescribing practices of physicians and nonphysician providers. METHODS: A serial cross-sectional analysis of the 2006-2012 National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey was performed. Ambulatory care services in physician offices, hospital emergency departments, and outpatient departments were evaluated using a nationally representative sample of patient visits to physicians, NPs, and PAs. Main outcome measures were 13 validated outpatient quality indicators focused on pharmacologic management of chronic diseases and appropriate medication use. RESULTS: A total of 701,499 sampled patient visits were included during the study period, representing ~8.33 billion visits nationwide. Physicians were the primary provider for 96.8% of all outpatient visits examined; NPs and PAs each accounted for 1.6% of these visits. The proportion of eligible visits where quality standards were met ranged from 34.1% (angiotensin-converting enzyme inhibitor use for congestive heart failure) to 89.5% (avoidance of inappropriate medications among elderly). The median overall performance across all indicators was 58.7%. On unadjusted analyses, differences in quality of care between nonphysicians and physicians for each indicator did not consistently favor one practitioner type over others. After adjustment for potentially confounding patient and provider characteristics, the quality of prescribing by NPs and PAs was similar to the care delivered by physicians for 10 of the 13 indicators evaluated, and no consistent directional association was found between provider type and indicator fulfillment for the remaining measures. CONCLUSIONS: Although significant shortfalls exist in the quality of ambulatory prescribing across all practitioner types, the quality of care delivered by nonphysicians and physicians was generally comparable.

Coverage of Nonpharmacologic Treatments for Low Back Pain Among US Public and Private Insurers.

Importance: Despite epidemic rates of addiction and death from prescription opioids in the United States, suggesting the importance of providing alternatives to opioids in the treatment of pain, little is known regarding how payers' coverage policies may facilitate or impede access to such treatments. Objective: To examine coverage policies for 5 nonpharmacologic approaches commonly used to treat acute or chronic low back pain among commercial and Medicare Advantage insurance plans, plus an additional 6 treatments among Medicaid plans. Design, Setting, and Participants: Cross-sectional study of 15 commercial, 15 Medicaid, and 15 Medicare Advantage health plans for the 2017 calendar year in 16 states representing more than half of the US population. Interviews were conducted with 43 senior medical and pharmacy health plan executives from representative plans. Main Outcomes and Measures: Medical necessity and coverage status for the treatments examined, as well as the use of utilization management tools and cost-sharing magnitude and structure. Results: Commercial and Medicare insurers consistently regarded physical and occupational therapy as medically necessary, but policies varied for other therapies examined. Payers most commonly covered physical therapy (98% [44 of 45 plans]), occupational therapy (96% [43 of 45 plans]), and chiropractic care (89% [40 of 45 plans]), while transcutaneous electrical nerve stimulation (67% [10 of 15 plans]) and steroid injections (60% [9 of 15 plans]) were the most commonly covered among the therapies examined for Medicaid plans only. Despite evidence in the literature to support use of acupuncture and psychological interventions, these therapies were either not covered by plans examined (67% of all plans [30 of 45] did not cover acupuncture) or lacked information about coverage (80% of Medicaid plans [12 of 15] lacked information about coverage of psychological interventions). Utilization management tools, such as prior authorization, were common, but criteria varied greatly with respect to which conditions and what quantity and duration of services were covered. Interviewees represented 6 Medicaid managed care organizations, 2 Medicare Advantage or Part D plans, 9 commercial plans, and 3 trade organizations (eg, Blue Cross Blue Shield Association). Interviews with plan executives indicated a low level of integration between the coverage decision-making processes for pharmacologic and nonpharmacologic therapies for chronic pain. Conclusions and Relevance: Wide variation in coverage of nonpharmacologic treatments for low back pain may be driven by the absence of best practices, the administrative complexities of developing and revising coverage policies, and payers' economic incentives. Such variation suggests an important opportunity to improve the accessibility of services, reduce opioid use, and ultimately improve the quality of care for individuals with chronic, noncancer pain while alleviating the burden of opioid addiction and overdose.

Relationship between high-risk patients receiving prescription opioids and high-volume opioid prescribers.

AIMS: (1) To characterize the amount of prescription opioids prescribed for high-risk patients by low-volume prescribers; (2) to quantify how high- and low-volume prescribers differ systematically in their prescribing patterns. DESIGN: Cross-sectional study using 2015 longitudinal, all-payer QuintilesIMS pharmacy claims. We conducted an aggregated analysis for the first aim and an individual-level analysis for the second aim. SETTING: California, Florida, Georgia, Maryland, and Washington, USA. PARTICIPANTS: Among 4 046 275 patients, we identified 375 848 concomitant users (filling more than 30-days of concomitant opioids and benzodiazepines), 150 814 chronic users (using 100+ morphine milligram equivalents (MMEs) per day for more than 90 days), and 3190 patients prescribed opioids by > 3 prescribers and filling opioids at > 3 pharmacies during any 90-day period. Among 192 126 prescribers, we identified 8023 high-volume prescribers, who comprised the highest fifth percentile of opioid volume during four calendar quarters. MEASUREMENTS: (1) MME dose per transaction, (2) days supplied per transaction, (3) total opioid volume per patient and (4) number of prescriptions per patient. We also examined differences in opioid dispensing between high- and low-volume prescribers among patients receiving opioids from both. FINDINGS: Low-volume prescribers accounted for 15-29% of opioid volume and 18-56% of opioid prescriptions for high-risk patients, compared with 28-37% and 53-58% for low-risk patients. After accounting for state of residence, comorbid burden, prescriber specialty and care sequence, patients were more likely to receive higher doses (60.9 versus 53.2 MMEs per day, P < 0.01), longer supplies (22.1 versus 15.6 days, P < 0.01), more prescriptions (4.0 versus 2.6 prescriptions, P < 0.01) and greater opioid volume (5.6 versus 1.9 g, P < 0.01) from high- than low-volume prescribers. CONCLUSIONS: In the United States, high-risk patients obtain a substantial proportion of prescription opioids from low-volume prescribers. The differences in prescribing patterns between high- and low-volume prescribers suggest the importance of interventions targeting prescriber behaviors.

Prescription Drug Coverage for Treatment of Low Back Pain Among US Medicaid, Medicare Advantage, and Commercial Insurers.

Importance: Despite unprecedented injuries and deaths from prescription opioids, little is known regarding medication coverage policies for the treatment of chronic noncancer pain among US insurers. Objective: To assess medication coverage policies for 62 products used to treat low back pain. Design, Setting, and Participants: A cross-sectional study of health plan documents from 15 Medicaid, 15 Medicare Advantage, and 20 commercial health plans in 2017 from 16 US states representing more than half the US population and 20 interviews with more than 43 senior medical and pharmacy health plan executives from representative plans. Data analysis was conducted from April 2017 to January 2018. Main Outcomes and Measures: Formulary coverage, utilization management, and patient out-of-pocket costs. Results: Of the 62 products examined, 30 were prescription opioids and 32 were nonopioid analgesics, including 10 nonsteroidal anti-inflammatory drugs, 10 antidepressants, 6 muscle relaxants, 4 anticonvulsants, and 2 topical analgesics. Medicaid plans covered a median of 19 opioids examined (interquartile range [IQR], 12-27; median, 63%; IQR, 40%-90%) and a median of 22 nonopioids examined (IQR, 21-27; median, 69%; IQR, 66%-83%). Medicare Advantage plans covered similar proportions (median [IQR], opioids: 17 [15-22]; 57% [50%-73%]; nonopioids: 22 [22-26]; 69% [69%-81%]), while commercial plans covered more opioids (median [IQR], 23 [21-25]; 77% [70%-84%]) and nonopioids (median [IQR], 26 [24-27]; 81% [74%-85%]). Utilization management strategies were common for opioids in Medicaid plans (median [IQR], 15 [11-20] opioids; 91% [74%-97%]), Medicare Advantage plans (median [IQR], 15 [9-18] opioids; 100% [100%-100%]), and commercial plans (median [IQR], 16 [11-20] opioids; 74% [53%-94%]), generally relying on 30-day quantity limits rather than prior authorization. Step therapy was especially uncommon. Many of the nonopioids examined also were subject to utilization management, especially quantity limits (24%-32% of products across payers) and prior authorization (median [IQR], commercial plans: 2 [0-3] nonopioids; 9% [0%-11%]; Medicare Advantage plans: 4 [3-5] nonopioids; 19% [10%-23%]; Medicaid plans: 6 [1-13] nonopioids; 38% [2%-52%]). Among commercial plans, the median plan placed 18 opioids (74%) and 20 nonopioids (81%) in tier 1, which was associated with a median out-of-pocket cost of $10 (IQR, $9-$10) per 30-day supply. Key informant interviews revealed an emphasis on increasing opioid utilization management and identifying high-risk prescribers and patients, rather than promoting comprehensive strategies to improve treatment of chronic pain or better integrating pharmacologic and nonpharmacologic alternatives to opioids. Conclusions and Relevance: Given the effect of coverage policies on drug utilization and health outcomes, these findings provide an important opportunity to evaluate how formulary placement, utilization management, copayments, and integration of nonpharmacologic treatments can be optimized to improve pain care while reducing opioid-related injuries and deaths.

Physician attitudes and experiences with Maryland's prescription drug monitoring program (PDMP).

AIMS: Physicians' use of prescription drug monitoring programs (PDMPs) varies by state. Among Maryland physicians, we sought to (1) estimate the PDMP impact on changes in opioid prescribing, (2) approximate the scope of PDMP utility and (3) determine the barriers to PDMP use after its 2013 implementation. DESIGN: Cross-sectional postal survey linking responses to state records of PDMP registration and use, randomly sampling physicians within specialty and registration strata. SETTING: Maryland, USA. PARTICIPANTS: A total of 1000 surveyed primary care, pain and emergency medicine physicians stratified into three subpopulations: PDMP non-registrants, PDMP registrants who were non-users and PDMP users; 405 respondents (44%) of 916 eligible physicians were analysed. MEASUREMENTS: Primary outcome measure was PDMP use. Key predictors were clinic characteristics, including type of practice and number of patients prescribed opioids. FINDINGS: No response-wave bias was identified. Seventy per cent of physicians believed PDMP access decreased their amount and increased their comfort level in prescribing opioids. Three-fourths (74%) of PDMP users reported the data very useful for informing opioid prescribing, although one-fifth (20%) reported difficulty accessing the data. Commonly reported barriers to PDMP use were lack of knowledge regarding its existence and registration process. In multivariable analysis after adjusting for key clinic characteristics, practicing at a managed care organization was associated with lower PDMP use [incidence rate ratio (IRR) = 0.19, 95% confidence interval (CI) = 0.05-0.73]. Conversely, physicians who prescribed opioids for more than 50 patients accessed the PDMP three times as often as those prescribing opioids for fewer than 10 patients monthly (IRR = 3.00, 95 % CI = 1.07-8.43). CONCLUSIONS: In this survey of Maryland, USA physicians, most participants reported that prescription drug monitoring programs (PDMPs) improved their opioid prescribing by decreasing prescription amounts and increasing comfort with prescribing opioids. Common barriers to PDMP use included not knowing about the program, registration difficulties and data access difficulties.