Opioids are a crucial part of acute pain management and palliative care, and an accepted element of chronic pain management for some patients. However, patients with substance use disorders (SUDs) or a history of SUDs are more likely to experience chronic pain compared to non-substance using patients, and if they receive opioids, they are more likely to misuse their medication and even develop or relapse to opioid dependence or addiction. This places patients with SUDs at increased risk of experiencing a variety of adverse health events associated with opioid use.
The 2010 VA/DoD Clinical Practice Guideline for Opioid Therapy for Chronic Pain (the CPG for OT) recommends prescribing practices intended to maximize the safe and effective use of opioids to treat chronic pain and minimize their likelihood of misuse. In a prior QUERI funded project (Trafton, RRP10-106) we used guidance from expert clinicians to operationalize key recommended prescribing practices in the 2010 CPG for OT as a series of metrics that could be applied to Veteran medical record data. Use of recommended care practices is considered essential for minimizing negative consequences of opioid prescribing without reversing gains made in improving pain management. The importance of adhering to recommended practices for patients with SUDs who receive opioids was also highlighted in a recent report from the Office of Inspector General (VAOIG-14-00895-163). There is currently limited data on the patient, medication, and facility characteristics associated with the use of recommended practices to reduce the risk of triggering medication misuse or an addictive disorder and improve patient outcomes among the SUD population.
Combining data from the metrics with other available data on Veterans with SUDs who received one or more opioid prescriptions, medication-level data on opioids prescribed in VHA, and surveys of the policies, staffing and services, and organizational structure of both specialty pain management and SUD treatment in VHA allows completion of the following aims:
Aim 1: Identify patient, facility, and medication characteristics associated with (a) fewer serious adverse health events, (b) lower mean pain numeric rating scale (NRS) scores post opioid prescription, and (c) increased likelihood of receiving clinical practice guideline-recommended opioid therapy practices.
Aim 2: Identify clinical practice guideline-recommended opioid therapy practices associated with (a) fewer serious adverse health events, and (b) lower mean pain NRS scores post opioid prescription.
The project addressed SUD QUERI strategic goal 2 of improving the quality and efficiency of SUD treatment within multiple VA medical settings, including primary care (where many opioids are prescribed).
First, we combined the information in the metrics with additional information on patient, medication, and facility characteristics. Second, we will analyze the data to identify (1) patient, medication, and facility characteristics associated with improved patient outcomes and receipt of recommended practices, and (2) associations between delivery of recommended practices and patient outcomes. All multi-level analyses will be performed using mixed-effects regression models, with a random effect for facility included to account for the clustering of patients within VHA facilities.
We have preliminary data from the completed dataset. A total of 176,105 patients in fiscal year 2010 received both an opioid prescription and a diagnosis of a substance use disorder. The demographics of the cohort were 95% male, 74% between the ages of 45-64, and 41% identified as racial and ethnic minorities. A majority - 65% - had a diagnosis consistent with neck or spine pain, 46% had a diagnosis consistent with neuropathic pain, and 10% had a diagnosis consistent with headaches. Virtually every patient (99.4%) had at least one pain NRS recorded; the average pain score value was 4 (SD = 3) and the average number of pain scores in FY10 was 13 (SD = 18). Although generally contra-indicated in the CPG for OT recommendations, 41% of patients had received greater than 90 days supply of short acting opioids and 9% had received a long acting opioid.
These patients received guideline-recommended care at widely varying rates. For example, 82% participated in at least one rehabilitation treatment and 63% had received adjunctive non-opoiod pharmacotherapy. However, only 36% had a documented medication management encounter in FY10 and 46% had a urine drug screen for the presence of the opioid. Patients experienced serious adverse events (SAE) including falls and overdoses at a rate of 7.0%. Patients at facilities offering cognitive behavioral pain mangement as likely to be over-prescribed acetaminophen, and only slightly less likely to be co-prescribed a sedative medication.
Opioid prescribing has been the subject of multiple and overlapping new initiatives since this project began. We anticipate that the results of the final project will have an impact on facilities by allowing us to identify and disseminate key organizational practices and policies associated with guideline adherent opioid prescribing and reduced adverse events in patients with SUDs who receive opioids. While improved management of this population of patients is a recommendation in the OIG report, they do not provide specific guidance on how to do this. In addition to responding to the OIG report described above, the metrics from the CPG for OT are available to the field through VSSC , the Opioid Safety Initiative has been released, and signature informed consent for opioid therapy has been adopted.
- Lewis ET. What we know about opioid adherence might surprise you: findings from a qualitative study of opioid use behaviors. HSR&D Spotlight on Pain Management. [Cyberseminar]. 2014 Nov 4.