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 David Tauben, MD Co-Investigator Chief of Pain Medicine University of Washington Mark Stephens Change Management Consulting Seattle, WA

Annual Opioid Prescribing Rates 2001-2015

November 26, 2018

• Vital Signs CDC MMWR July 7, 2017

Opioid Prescribing by Specialty, IMS Health, 2012 (Total Rx %)

Family Practice 22% Internal Medicine 22% General Practice 10% Non-physician 11% Surgery 2% Dentistry 1% Pain Medicine 1% Emergency Med 1% All Others 30%

Percent by Specialty

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 Learning from Effective Ambulatory Practices

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

commitments

• Implementation

blueprint (the Six Building Blocks)

• Assess for readiness and

identify barriers

• Conduct local consensus

discussions

• Practice facilitation
• Capture/share local

knowledge

• Learning collaborative
• Development of tools

for quality monitoring

• Identify/support clinical

champions

• Promote adaptability

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
• f 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
• n LtOT
• Proportion of

patients on higher dose LtOT

This Photo by Unknown Author is licensed under CC BY

11.8 11.2 11.4 11.6 11.4 10.1 10.3 9.9 10.3 10.0 10.0 9.4 9.0 8.9 8.8 14.0 14.7 15.3 15.7 15.5 14.9 14.6 14.4 14.0 13.8 13.6 13.5 13.1 12.8 12.8

8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Trend in Proportion >=100 MED

Diff of 2.2% Diff of 4.0%

Months %

Welcome Visit

Intervention Cohort Control Cohort

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 trend) 1 0.358 0.351 1.02 0.320 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 trend) 1

• 0.226

0.089

• 2.55

0.019 Change in level 1 1.592 0.478 3.33 0.003 Change in slope 1 0.242 0.094 2.58 0.018

2065 2004 1974 1974 1931 1942 1921 1881 1897 1847 1883 1835 1872 1824 1836 1776 1797 1801 1805 1777 1731 1758 1814 1817 1801 1789 1769 1765 1774 1768 1742 1712

1500 1600 1700 1800 1900 2000 2100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Number of LtOT Patients by Month

Control Cohort

Months

Intervention Cohort

Welcome Visit

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

November 26, 2018 70.0 75.0 80.0 85.0 90.0 95.0 100.0

• 8
• 7
• 6
• 5
• 4
• 3
• 2
• 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Percent of LtOT Patients

Study Month

% MED < 50 % MED 50-99 % MED ≥ 100

Welcome Visit

What others said about clinic life after implementing the Six Building Blocks:

“Everybody that works in this clinic says to me, ‘do you remember how much turmoil there was around it? Wow, we don’t have any of that anymore.” Medical Director “Hopefully there’s no going

• back. It works. I don’t think any
• ne of us wants to go back.”

Medical Assistant “The teamwork, there’s been a lot of teamwork regarding it. I wouldn’t say that was a surprise, but it’s been nice.” Nurse “I saw one of the high MED patients that I inherited… we got him down to 80... just for him to say, ‘You know, I’m more functional — my pain is not different, might be better.” Physician

www.improvingopioidcare.org

Thank YOU! www.improvingopioidcare.org

November 26, 2018 24

Michael Parchman, MD, MPH michael.x.parchman@kp.org