De-Implementing Long-term Opioids for Chronic Pain Michael - PowerPoint PPT Presentation

De-Implementing Long-term Opioids for Chronic Pain Michael Parchman, MD, MPH Laura-Mae Baldwin MD, MPH Robert Penfold, PhD Brooke Ike, MPH David Tauben, MD Kari Stephens, PhD Mark Stephens, MA Funded by the Agency for Healthcare Research & Quality (R18HS023750) Kaiser Permanente Washington Health Research Institute

Disclosures • No conflicts to disclose November 26, 2018

The Kaiser Permanente WA Health Research Institute- University of Washington Research Team Michael Parchman, MD, MPH, Principal Investigator Director, MacColl Center for Innovation Kaiser Permanente WA Health Research Institute Laura-Mae Baldwin, MD, MPH, Co-Investigator Professor, Department of Family Medicine, University of Washington Director, WWAMI region Practice and Research Network Brooke Ike, MPH, Project Manager and Practice Facilitator WWAMI region Practice and Research Network Coordinating Center University of Washington Mark Stephens David Tauben, MD Change Management Co-Investigator Consulting Chief of Pain Medicine Seattle, WA University of Washington

Annual Opioid Prescribing Rates 2001-2015 -Vital Signs CDC MMWR July 7, 2017 November 26, 2018

Opioid Prescribing by Specialty, IMS Health, 2012 (Total Rx %) Percent by Specialty All Others Family Practice 30% 22% Emergency Med 1% Internal Pain Medicine Medicine 1% 22% Dentistry Surgery 1% 2% Non-physician General Practice 11% 10% Levy B. Am J Prev Med 2015

Six ix Buil ildin ing Blo locks for Team-Based Opio ioid Management A quality improvement roadmap to help primary care teams implement effective, guideline- driven care for their chronic pain and long-term opioid therapy patients. November 26, 2018 6

LEAP: 30 Innovative Primary Care Practice Models for Improving Team-based Care L earning from E ffective A mbulatory P ractices

The Six Building Blocks Program Parchman ML. J Am Board Fam Med 2017;30:44 – 51.

The Six Building Blocks Program

Stu tudy Setti ting: Six ix Rural-Serving Healt lth Care Organizati tions with ith 20 clin linic ic sit ites

In Intervention: Expert Recommendatio ions for Im Imple lementing Change (E (ERIC) Strategies • Obtain formal • Capture/share local commitments knowledge • Implementation • Learning collaborative blueprint (the Six • Development of tools Building Blocks) for quality monitoring • Assess for readiness and • Identify/support clinical identify barriers champions • Conduct local consensus • Promote adaptability discussions • Practice facilitation 15 Months of Active Support

Stu tudy Subjects • LtOT patients defined as: • Any patient who has two opioid prescriptions at least 28 days apart in the past 90 days • Any patient who has one non-schedule 2 opioid prescription with at least one refill in the last 90 days • Patients with dx of cancer excluded

Control Subjects • Identified from a large regional health plan’s pharmacy claims data over same time period as the study intervention. • Patients enrolled in this health plan from same Primary Care Service Area as enrolled clinics. • Same inclusion/exclusion criteria applied to identify controls

Data Collection • Data extract from electronic health record of each organization • “Prescribed” not “Filled” medication data • Covered a period of time 12 months before the start of the intervention until 15 months after the start of the intervention in the last enrolled organization.

Analyses • Interrupted time series with segmented regression • Three analyses: • Within study sample changes in the percentage of LtOT patients with MED ≥ 100; • A difference-in-differences (DiD) analysis of the change in percentage of LtOT patients with MED ≥ 100 compared to the control group; • The total number of study patients on LtOT.

Result lts • Significant Declines in both • Number of patients on LtOT • Proportion of patients on higher dose LtOT This Photo by Unknown Author is licensed under CC BY

Trend in Proportion >=100 MED 17.0 16.0 15.7 15.5 15.3 15.0 14.9 14.7 Control Cohort 14.6 14.4 14.0 14.0 14.0 13.8 13.6 13.5 13.1 13.0 Diff of 2.2% % 12.8 12.8 12.0 11.8 11.6 11.4 11.4 11.2 11.0 Diff of 4.0% 10.3 10.3 10.1 10.0 10.0 10.0 9.9 Intervention Cohort 9.4 9.0 9.0 8.9 8.8 8.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Welcome Visit Months

Proportio ion on Hig igh Dose (> (>=100 MED) Within Study Sites Variable DF Estimate S.E. t-value p-value Intercept 1 11.929 0.329 36.3 <.0001 Time (secular 1 0.358 0.351 1.02 0.320 trend) Change in level 1 -0.054 0.065 -0.82 0.420 Change in slope 1 -0.138 0.069 -2 0.060 Study Sites versus Controls Variable DF Estimate S.E. t-value p-value Intercept 1 3.943 0.448 8.8 <.0001 Time (secular 1 -0.226 0.089 -2.55 0.019 trend) Change in level 1 1.592 0.478 3.33 0.003 Change in slope 1 0.242 0.094 2.58 0.018

Number of LtOT Patients by Month 2100 2065 2004 2000 1974 1974 Intervention Cohort 1942 1931 1921 1897 1900 1883 1881 1872 1847 1836 1835 1824 1817 1814 1805 1801 1801 1797 1789 1800 1777 1776 1774 1769 1768 1765 1758 1742 1731 Control Cohort 1712 1700 1600 1500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Welcome Visit Months

Within Study Site Number of Patients on LtOT Interrupted Time Series within Study Sites Variable DF Estimate S.E. t-value p-value Intercept 1 1680 38.1 44.1 <.0001 Time (secular trend) 1 -133.0 40.7 -3.27 0.004 Change in level 1 60.7 7.5 8.04 <.0001 Change in slope 1 -75.8 8.0 -9.48 <.0001 November 26, 2018

MED Trend by Categories 100.0 95.0 90.0 Percent of LtOT Patients 85.0 Welcome Visit 80.0 75.0 70.0 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Study Month % MED < 50 % MED 50-99 % MED ≥ 100 November 26, 2018

What others said about clinic life after implementing the Six Building Blocks: “Everybody that works in this “Hopefully there’s no going clinic says to me, ‘do you back. It works. I don’t think any remember how much turmoil one of us wants to go back.” there was around it? Wow, we Medical Assistant don’t have any of that anymore.” Medical Director “I saw one of the high MED patients that I inherited… we got “The teamwork, there’s him down to 80... just for him to say, ‘You know, I’m more been a lot of teamwork functional — my pain is not regarding it. I wouldn’t say different, might be better.” that was a surprise, but it’s been nice.” Physician Nurse

www.improvingopioidcare.org

Thank YOU! www.improvingopioidcare.org Michael Parchman, MD, MPH michael.x.parchman@kp.org November 26, 2018 24