Impact of Clinical Pharmacist Practitioner-Driven High Opioid Dose Reevaluation in Veterans with Chronic Non-Cancer Pain

Abstract Veterans Health Administration implemented the Opioid Safety Initiative (OSI) in 2013 to promote safe/rational opioid prescribing. West Palm Beach VA Healthcare System (WPBVAHCS) has been an outlier for the percentage of Veterans with chronic non-cancer pain receiving ≥90 mg Morphine Equivalent Daily Dosing (MEDD) in Veterans Integrated Service Networks (VISN) 8 since the 2016 fiscal year. The purpose was to determine the utility of a Pain Clinical Pharmacist Practitioner (CPP) identifying, reviewing, and approaching Veterans utilizing high-dose opioids for dose reevaluation and the impact on OSI metric post-opioid reevaluation. Pain CPP opioid education resulted in 28% (11/39) of Veterans undergoing an average 17.7 mg MEDD opioid dose reduction. For Veterans evaluated by Pain CPP, 83% (15/18) reported no change or improvement in average pain and PEG score. Pain CPP’s implemented 48 interventions outside of opioid dose reduction, the most common related to naloxone. No documented opioid overdose events, hospitalizations for uncontrolled pain or mental health, suicide attempts or pain-related crisis interventions were reported. Pain CPP’s are equipped to provide opioid education, address risk mitigation strategies, reassess pain regimens, and refer for non-pharmacologic modalities. Utilization of Pain CPP resources helps improve OSI metrics while providing safe comprehensive medication management (CMM) for chronic pain.


Introduction
In 2013, Veterans Health Administration (VHA) implemented the Opioid Safety Initiative (OSI), an initiative to promote safe and effective opioid prescribing (1). Key clinical OSI metrics include the following: the percent of Veterans dispensed an opioid over time, percent of Veterans receiving combination opioid plus a benzodiazepine over time, percent of Veterans dispensed opioids long-term with a urine drug screen (UDS) completed over time, and stratification of the percent of Veterans dispensed oral Morphine Equivalent Daily Dosing (MEDD) with metrics updated on a quarterly basis. Veterans were stratified into the following groups: <100 mg MEDD, 100-199 mg MEDD, 200-299 mg MEDD, 300-399 mg MEDD, and ≥ 400 mg MEDD. This metric was simplified at the start of Fiscal year 2021 (FY21) (October 2020) to include all Veterans with MEDD ≥90mg. Another VA electronic database available to providers is called the Stratification Tool for Opioid Risk Mitigation (STORM) database, which is used to evaluate risk of suicide, overdose events, or death in the next 1 or 3 years (2). STORM is frequently updated using data pulled from the electronic medical record (EMR) (including diagnostic codes, active medications, etc.) and considers various disease states, medications, mental health comorbidities, and substance use disorders (including nicotine). Since 2013, there have been multiple guidelines published regarding the safe and effective use of opioids in patients with chronic non-cancer pain (3)(4)(5).
In 2016, the Centers for Disease Control and Prevention (CDC) published the CDC Guideline for Prescribing Opioids for Chronic Pain with MEDD target of ≤50 mg with a recommended ceiling dose of ≤90 mg to be prescribed in the primary care setting (3). In 2017, the VA and Department of Defense (DoD) published the VA/ DoD Clinical Practice Guideline for Opioid Therapy for Chronic Pain. While this guideline is similar to the CDC guidelines, there are a few notable differences. The VA/DoD recommendations focus on identifying the risk of developing substance use disorder (SUD). VA/DoD guidelines readily address the need to avoid long-term opioid therapy (LTOT) in patients under the age of 30 years, has a strong focus on suicide prevention, evaluates reassessment strategies for those with acute pain, and recommends a 90-day reassessment period for those prescribed chronic opioid therapy (COT) (4).
Most recently in 2019, the Department of Health and Human Services (HHS) published the HHS Guide for Clinicians on the Appropriate Dosage Reduction or Discontinuation of Long-Term Opioid Analgesics. This guide includes recommendations to engage in shared decision-making with patients and the need to individualize each opioid taper (5).
West Palm Beach VA Healthcare System (WPBVAHCS) has consistently been an outlier for the percentage of Veterans with chronic non-cancer pain receiving ≥90 mg MEDD in Veterans Integrated Service Networks (VISN) 8 since the 2016 fiscal year. In Quarter 4 of Fiscal Year 2020 (Q4FY20), the WPBVAHCS had 0.76% patients prescribed ≥90 mg MEDD compared to the VISN 8 average of 0.26% and VHA national average of 0.44%. As the WPBVAHCS's Pain Clinic has two Pain Clinical Pharmacists Practitioners (CPP) and one Post-Graduate Year 2 (PGY2) Pharmacy Resident detailed to the Pain Clinic longitudinally for six months of the residency year dedicated to the management of chronic pain patients, Pain CPP interventions may have the potential to impact OSI metrics. The purpose of this quality improvement project was to determine the utility of Pain CPP identifying, reviewing, and approaching Veterans utilizing high-dose opioids for opioid dose reevaluation, targeted opioid educational session(s), incorporation of non-opioid pharmacotherapy and non-pharmacologic approaches, and the impact on OSI metric post-opioid dose reevaluation.

Methods
This prospective quality improvement project occurred from October 1, 2020 through April 30, 2021. The project was deemed institutional review board exempt by WPBVAHCS Research and Development Committee on August 28, 2020. Information was obtained from the Computerized Patient Record System (CPRS) and the OSI-database generated spreadsheet, which excludes Veterans receiving buprenorphine products. Information was only accessible to resident and project advisors within a restricted electronic folder. The PGY2 Pain Pharmacy Resident acted as another Pain CPP for the project duration (under the supervision of Pain CPP's). Protected health information remained confidential per VA policy and HIPAA. Veterans were identified by code number in the database, and a separate electronic document was maintained to link this information. After data validation was completed, the code sheet was destroyed.
Veterans were originally identified by the OSI MEDD report generated on July 27, 2020, who had ≥90 mg MEDD, as calculated per the OSI dashboard. At the start of this project, ≥100mg MEDD was the original target based on OSI metrics, however in October 2020, the OSI MEDD metric was reduced to ≥90 mg MEDD, which aligns with the CDC recommendations, therefore the inclusion criteria was modified to align with this change (3). A manual MEDD calculation using the same OSI MEDD equation was completed on every Veteran originally identified, to ensure accuracy of the dashboard calculation ( Figure 1) (6). Veterans receiving primary care and other services primarily at WPBVAHCS with a chronic non-cancer pain diagnosis on an opioid prescribed by a WPBVAHCS provider at a MEDD of ≥90 mg for ≥3 months were eligible for inclusion. Veterans were excluded if they had active cancer, were receiving palliative or hospice care, actively tapering opioid or benzodiazepine at time of initial review, receiving opioid therapy from Community Care (care contracted out from VA to provider in non-VA sector) or private provider, and/or receiving opioid for acute pain management. All Veterans who met inclusion criteria were discussed with the primary opioid prescriber for approval and referral to Pain CPP. Once the Pain Clinic consult was received, Pain CPP contacted Veteran to discuss reason for referral, then clinic schedulers contacted Veteran to schedule appointment via face-to-face, telephone, or video modality based on Veteran's preference due to COVID-19 pandemic. During initial appointment with Pain CPP, the Veteran was educated on risks and benefits of long-term opioid use for chronic pain using documents available through the VA Academic Detailing service, reviewed and established functional goals, developed initial plan for opioid dose reduction, assessed pharmacologic pain regimen for potential pain adjuvants or non-pharmacologic interventions and addressed opioid risk mitigation strategies (i.e., naloxone education, adherence to sleep apnea equipment, urine toxicology). Veterans were provided with the following opioid educational materials available through VA Academic Detailing, regardless of clinic appointment modality: "Slowing Stopping Opioid Medications: Helpful Tips to Getting Off Your Opioids Successfully" IB 10-1016 P96884 (7), "Safe and Responsible Use of Opioids for Chronic Pain" IB 10-791 P96791 (8), and "Opioid Safety" IB 10-784 P96786 (9), to help encourage voluntary but individualized opioid taper and provide documentation on additional risk mitigation needs (7)(8)(9). For Veterans who declined appointment with Pain CPP, an electronic consult (E-consult) was completed by Pain CPP, which outlined opioid taper recommendation, pain adjuvant considerations, non-pharmacologic interventions, and opioid risk mitigation strategies for the opioid prescriber to consider implementing as clinically indicated.
Primary objectives of this quality improvement project were to evaluate the overall reduction in MEDD in Veterans with chronic non-cancer pain, optimization of non-opioid pharmacotherapy regimen, and change in numerical pain score, using numeric pain rating scale, and Pain, Enjoyment of Life and General Activity (PEG) scores (10). The numeric pain rating scale is a 0 to 10 scale, with zero being "no pain" and ten being "worst possible pain." Change in PEG scores only assessed in those seen by Pain CPP, whereas the other primary objective data was gathered for all Veterans referred to Pain CPP. Secondary objectives were to quantify and categorize Pain CPP interventions in improving quality metrics related to pain care. Safety assessments included quantifying opioid overdose events, hospitalizations for uncontrolled pain or mental health crises (at WPBVAHCS), suicide attempts, classification of STORM risk, and crisis line calls related to pain management.
Descriptive statistics were utilized to show trends in opioid dose reduction on average pain and PEG scores as well as non-opioid interventions performed by Pain CPP. The WPBVAHCS OSI dashboard metrics were reviewed on July 20, 2020 (pre-intervention) and May 6, 2021 (post-intervention) to compare the data along with data provided by the VISN 8 Pain/Mental Health Academic Detailer. EMR chart review was performed to quantify and categorize total number of Pain CPP interventions.

Results
One hundred nineteen Veterans were identified by OSI dashboard for initial review. Forty-six Veterans were excluded, leaving 73 Veterans who met eligibility criteria ( Figure 2). Thirty-four Veterans (46.6%) were not referred to Pain CPP because the primary opioid prescriber declined reevaluation. Thirty-nine Veterans (53.4%) were referred to Pain CPP for pain regimen reevaluation. Of those who were referred to Pain CPP, 46.2% (18/39) of Veterans were evaluated by Pain CPP at least one time. Conversely, 53.8% (21/39) of Veterans declined to be seen by Pain CPP, therefore an E-consult with recommendations by chart review were provided by Pain CPP to the referring opioid prescriber.
The baseline demographics show the most common chronic pain indication for LTOT was low back pain (Table 1). Approximately 49% of Veterans included had nicotine dependence. The top mental health comorbidities included depression and post-traumatic stress disorder (PTSD). Urine toxicology revealed 10% (4/39) of Veterans had an unexpected or aberrant toxicology result. Of those who had unexpected or aberrant toxicology results, two Veterans were positive for cannabinoids, one Veteran positive for methadone, and one Veteran positive for alcohol. The average time spent by Pain CPP to complete E-consult chart reviews was 89 minutes (range: 34-140 minutes; two E-consults lacked documentation of time spent and therefore were excluded from this calculation). For Veterans seen in clinic by Pain CPP, the average time spent with Pain CPP was 91 minutes (range: 35-270 minutes), which does not include clinic preparation or chart documentation time. Figure 3 shows the overall reduction in average MEDD for those who had Pain CPP involvement, Veterans with MEDD closer to target <90 mg MEDD were more willing to trial opioid dose reduction or rotation. Of those Veterans who underwent opioid dose reduction or rotation: 36% of Veterans (4/11) are below 90 mg MEDD, 36% of Veterans (4/11) are at 90 mg MEDD, and 27% of Veterans (3/11) remained above 90 mg MEDD at project completion. One Veteran had their opioid dose increased during the project by their opioid prescriber for worsening neck pain and depression despite the Pain CPP evaluation which assessed Veteran as not meeting functional goals to justify LTOT or opioid dose escalation. Most Veterans evaluated by Pain CPP (n = 18) reported the same or better average pain score and average PEG score (Figure 4). The STORM database was utilized to categorize Veteran's risk of overdose, suicide-related healthcare events, or death showing the trend in STORM 1-year and 3-year risk change over the project duration ( Figure 5) (2). One Veteran did not have a calculated 3-year initial STORM category. One Veteran was not included in the 1-year and 3-year final STORM categories, as he died during the project time frame. His death was unrelated to opioid dose reduction as opioid taper was not performed prior to his unrelated hospitalization. Pain CPP's implemented a total of 48 inventions related to non-pharmacologic and non-opioid pharmacotherapy. Nine non-pharmacologic modality referrals were made, which included four referrals to Physical Therapy for electrical stimulation device trial, one referral to pain psychology, and four referrals for acupuncture. The most common non-opioid pharmacotherapy intervention was renewing (n = 12) or prescribing  (n = 10) naloxone rescue kits. Other non-opioid pharmacotherapy interventions included initiation of duloxetine (n = 1), initiation of gabapentinoid (n = 2), initiation of topical NSAID (n = 2), initiation of topical anesthetic (n = 2), initiation of skeletal muscle relaxant (n = 1), initiation of acetaminophen (n = 1), adjustment of gabapentinoid (n = 3), adjustment of topical anesthetic (n = 2), discontinuation of gabapentinoid (n = 1), discontinuation of systemic NSAID (n = 1), and discontinuation of topical NSAID (n = 1). Prior to initiation of the project, the WPBVAHCS metric was 0.76% (255/33667) in Q4FY20. Over the project duration, the WPB VAHCS metric  initially increased to 0.84% (278/33261) in Q1FY21, then began to steadily decrease to 0.78% (259/33407) in Q2FY21 to 0.75% (251/33410) Q3FY21 to 0.67% (221/32898) in Q4FY21 ( Figure  6). OSI dashboard data indicated that the percent of patients prescribed ≥90 mg MEDD decreased by the completion of the project, with sustainability seen after project completion, as seen in Q4FY21 ( Figure 6). There were no documented opioid overdose events, hospitalizations for uncontrolled pain or mental health crises, suicide attempts, or crisis line calls related to pain management during the study period.

Discussion
This quality improvement project showed that Pain CPP opioid safety education targeted to Veterans on chronic opioid therapy resulted in safe, voluntary opioid dose reductions. Pain CPP's offered opioid rotation rather than just straight dose reduction in Veterans initially resistant to changes in pain regimen. Another alternative offered by the Pain CPP included possible rotation to buprenorphine products approved by the Food and Drug Administration (FDA) for chronic pain, specifically the buccal film, as Veterans targeted were not eligible for transdermal buprenorphine patch due to their already high MEDD. There were no Veterans at the completion of this project who were transitioned to a buprenorphine product. In Veterans who were willing to trial opioid dose reduction or rotation, the majority did not experience worsening pain or PEG score. As previously noted, there were two Veterans who failed an opioid dose reduction and/or rotation, which accounts for two out of three Veterans who reported worsening of pain and PEG scores after dose reduction. Due to a significant number of Veterans who declined evaluation by Pain CPP, recommendations provided to the opioid prescriber by Pain CPP were not counted as an intervention unless the opioid prescriber enacted the recommendations. Not all Pain CPP interventions are listed, as certain risk mitigation strategies such as ordering additional urine toxicology screens, completing STORM point of care notes, updating Informed Consent for Long-Term Opioid Use, and ensuring adequate bowel regimen are standard practices for Pain CPP at the WPB VAHCS.
One of the barriers identified was getting the opioid prescriber to refer to Pain CPP. Nearly half of Veterans who met inclusion criteria were not referred by the provider for Pain CPP review. After the conclusion of the project, the Interdisciplinary Pain Management Team (IPMT) committee leaders presented WPBVAHCS data compared to VISN 8 data at the facility medical staff meeting. After this medical staff meeting, physicians were then provided a list of all Veterans on their panel who are on LTOT and above the OSI metric. This led to an increase in opioid prescriber awareness to review their prescribing habits and Pain CPP's have seen another increase in referrals for opioid tapering as a result of this communication, which contributes to ongoing efforts to improve the target OSI metric. Facility administration was more encouraging of medical providers to reevaluate opioid prescribing practices after project conclusion.
Another barrier identified in this QI project was that Veterans who are on LTOT for chronic non-cancer pain demonstrated a lack of willingness or desire to change and lack of engagement in or desire for the biopsychosocial model for integrated pain care. While nine non-pharmacologic treatment modality referrals were placed, only one referral was to Pain Psychology services. Pain CPP's utilized the VA Pain Management Opioid Taper Decision Tool -A VA Clinician's Guide to assist with opioid dose reductions. In most cases, opioid dose was reduced by 5-20% every month with taper duration being extended pending Veteran's response to dose reduction (11). In some Veterans, opioid dose reduction intervals were extended for extenuating circumstances including upcoming interventional pain procedure or life events, such as a death in the family, to avoid destabilization. Veterans were given the option to initially reduce either the long-acting or short-acting opioid (when applicable), with mixed preferences. Pain CPP's were able to dedicate 60 minutes for each initial evaluation appointment, which is substantially more time that most opioid prescribers can spend with the Veteran, as most opioid prescribers' appointment last 15-30 minutes in duration and are not solely focused on pain medication management.
Both dashboard and manual calculations for MEDD were obtained. The dashboard data may be impacted based on when the provider enters the renewal for the opioid, thus resulting in most cases of an overestimation of the MEDD due to the computer detecting what it believes to be an overlap in the opioid prescription. Thus, manual calculation helped identify if a patient was indeed at the OSI MEDD cutoff or not, especially for those who are on the borderline of the 90 mg MEDD target.
Unexpectedly, STORM risk did worsen for some individuals. A change in any one of the STORM parameter categories results in a recalculated risk score which can change at any time (2). The estimated risk for overdose, suicide, or death is leveraged against risk mitigation strategies that have been employed in the last year to determine percent risk. This further confirms the need for routine risk mitigation strategies incorporated into chronic pain care provided for patients with concurrent high MEDDs to ensure safe opioid prescribing practices.
There are a few important limitations to note. First, safety data reported in the results section could only be obtained from the EMR locally. WPBVAHCS catchment area spans seven counties, and some Veterans may have presented to a private hospital for evaluation without sharing this information with WPB VAHCS. In an observational study published in 2021 by Agnoli et al., they found there was an association between opioid dose tapering with overdose or mental health crisis in those on LTOT (12). While this was not seen based on the information available, it cannot be excluded as an ongoing concern. Secondly, due to the rolling referral process from opioid prescribers, total number of encounters with Pain CPP is widely variable for those who elected ongoing management for opioid dose reevaluation, with some Veterans still actively engaged in treatment with the Pain CPP at the conclusion of data collection, so MEDDs may continue to reduce with time. As with nearly all opioid tapers, when not being performed due to safety concerns, it may take years to reduce MEDD to meet the OSI metric so the project timeline may not be sufficient to see the positive impact of final MEDD reduced dose. Third, this project was completed during the COVID-19 pandemic, which may have been an added stressor for patients; perhaps circumstances surrounding whether a patient was agreeable to reevaluate pain medication regimens may have been different pre-or post-pandemic. Fourth, statistical analysis was not performed due to small number of Veterans who were agreeable to undergo opioid dose reductions. Finally, while Pain CPP's offered multiple modalities for evaluation, referral for additional services were limited, either due to limited clinic appointments as a result of social distancing and enhanced cleaning protocols or were not being offered at all due to the COVID-19 pandemic.

Conclusion
Pain CPP's demonstrated an opioid safety educational intervention resulted in an average MEDD reduction of 17.7 mg allowing for lower chronic opioid usage without causing mental health crises or hospitalization for uncontrolled pain. Pain CPP's were able to safely reduce to MEDD ≤90 mg in 73% of Veterans (8/11) who voluntarily reduced opioids by project completion. Utilizing a Pain CPP to reevaluate pain medication regimens can improve OSI metrics while increasing utilization of non-opioid pain care modalities consistent with the biopsychosocial approach to pain care.