Trial Outcomes & Findings for Physician Coaching to Reduce Opioid-related Harms (NCT NCT02433496)

NCT ID: NCT02433496

Last Updated: 2018-07-06

Results Overview

The proportion of patients with a chronic pain diagnosis receiving daily opioids.

• Recruitment status: COMPLETED
• Target enrollment: 53132 participants
• Primary outcome timeframe: Up to 3 years
• Results posted on: 2018-07-06

Participant Flow

Clinics were paired and randomly assigned to be grouped in the intervention or control group. The intervention group was invited. If clinics in this group refused to participate, the other clinic of the pair was invited to be a part of the intervention group. Once clinics agreed to take part in the study they were immediately assigned to a group.

Unit of analysis: clinics

Participant milestones

Measure	Physician Coaching This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group Clinics were invited to join the study, but refused to receive the intervention. Clinics reason for refusal was lack of staff or time. Data was not collected or analyzed for clinics that refused the intervention.
Overall Study STARTED	29990 4	13309 4	9833 3
Overall Study COMPLETED	29990 4	13309 4	9833 3
Overall Study NOT COMPLETED	0 0	0 0	0 0

Reasons for withdrawal

Withdrawal data not reported

Baseline Characteristics

This measure analyzes at the patient level. The population not analyzed for this measure are providers of the clinic. The Refused group was not analyzed due to desire not to participate in the study.

Baseline characteristics by cohort

Measure	Physician Coaching n=4 clinics This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=4 clinics This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group n=3 clinics This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.	Total n=11 clinics Total of all reporting groups
Age, Customized Age of participants · < 18 years old	0 Participants n=29990 Participants	0 Participants n=13309 Participants	0 Participants n=9833 Participants	0 Participants n=53132 Participants
Age, Customized Age of participants · >= 18 years old	29990 Participants n=29990 Participants	13309 Participants n=13309 Participants	9833 Participants n=9833 Participants	53132 Participants n=53132 Participants
Sex: Female, Male Female	14125 Participants n=29990 Participants	6615 Participants n=13309 Participants	4504 Participants n=9833 Participants	25244 Participants n=53132 Participants
Sex: Female, Male Male	15865 Participants n=29990 Participants	6694 Participants n=13309 Participants	5329 Participants n=9833 Participants	27888 Participants n=53132 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	660 Participants n=29990 Participants	333 Participants n=13309 Participants	246 Participants n=9833 Participants	1239 Participants n=53132 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	29330 Participants n=29990 Participants	12976 Participants n=13309 Participants	9587 Participants n=9833 Participants	51893 Participants n=53132 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	0 Participants n=29990 Participants	0 Participants n=13309 Participants	0 Participants n=9833 Participants	0 Participants n=53132 Participants
Race (NIH/OMB) American Indian or Alaska Native	150 Participants n=29990 Participants	67 Participants n=13309 Participants	49 Participants n=9833 Participants	266 Participants n=53132 Participants
Race (NIH/OMB) Asian	720 Participants n=29990 Participants	213 Participants n=13309 Participants	69 Participants n=9833 Participants	1002 Participants n=53132 Participants
Race (NIH/OMB) Native Hawaiian or Other Pacific Islander	0 Participants n=29990 Participants	0 Participants n=13309 Participants	0 Participants n=9833 Participants	0 Participants n=53132 Participants
Race (NIH/OMB) Black or African American	720 Participants n=29990 Participants	306 Participants n=13309 Participants	69 Participants n=9833 Participants	1095 Participants n=53132 Participants
Race (NIH/OMB) White	24831 Participants n=29990 Participants	11126 Participants n=13309 Participants	8525 Participants n=9833 Participants	44482 Participants n=53132 Participants
Race (NIH/OMB) More than one race	0 Participants n=29990 Participants	0 Participants n=13309 Participants	0 Participants n=9833 Participants	0 Participants n=53132 Participants
Race (NIH/OMB) Unknown or Not Reported	3569 Participants n=29990 Participants	1597 Participants n=13309 Participants	1121 Participants n=9833 Participants	6287 Participants n=53132 Participants
Proportion of patients with consistent opioid Rx	0.013 proportion of patients n=29956 Participants • This measure analyzes at the patient level. The population not analyzed for this measure are providers of the clinic. The Refused group was not analyzed due to desire not to participate in the study.	0.014 proportion of patients n=13296 Participants • This measure analyzes at the patient level. The population not analyzed for this measure are providers of the clinic. The Refused group was not analyzed due to desire not to participate in the study.	—	0.013 proportion of patients n=43252 Participants • This measure analyzes at the patient level. The population not analyzed for this measure are providers of the clinic. The Refused group was not analyzed due to desire not to participate in the study.
Proportion with mental health screen	0.226 proportion of patients n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	0.220 proportion of patients n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	0.224 proportion of patients n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.
Proportion with urine drug screen	0.399 proportion of patients n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	0.374 proportion of patients n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	0.391 proportion of patients n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.
Proportion with treatment agreement	0.428 proportion of patients n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	0.368 proportion of patients n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	0.409 proportion of patients n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.
Average morphine equivalent daily dose (MEDD)	86.3 milligram STANDARD_DEVIATION 18.7 • n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	58.8 milligram STANDARD_DEVIATION 10.1 • n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	72.5 milligram STANDARD_DEVIATION 16.9 • n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.
Proportion with MEDD >120 mg	0.215 proportion of patients n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	0.137 proportion of patients n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	0.190 proportion of patients n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.
Proportion with co-prescribed benzodiazepines	0.08 proportion of patients n=389 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	0.055 proportion of patients n=186 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.	—	0.072 proportion of patients n=575 Participants • This measure analyzes at the patient level and includes only the subset of patients with consistent opioid Rx.

PRIMARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients in each group.

The proportion of patients with a chronic pain diagnosis receiving daily opioids.

Outcome measures

Measure	Physician Coaching n=29956 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=13296 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Overall Rate of Opioid Prescribing	0.012 Proportion of patients	0.013 Proportion of patients	—

SECONDARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients who fall in the subset of consistent opioid use.

Proportion of patients with a chronic pain diagnosis receiving daily opioids and benzodiazepines concurrently.

Outcome measures

Measure	Physician Coaching n=359 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=160 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Rate of Opioid / Benzodiazepine Co-prescribing	0.074 Proportion of patients	0.061 Proportion of patients	—

SECONDARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients who fall in the subset of consistent opioid use.

Proportion of opioid patients completing urine drug screens prior to and during the study intervention

Outcome measures

Measure	Physician Coaching n=359 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=160 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Urine Drug Screening Rate	0.645 Proportion of patients	0.440 Proportion of patients	—

SECONDARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients who fall in the subset of consistent opioid use.

Proportion of opioid patients screened for mental health/substance use problems

Outcome measures

Measure	Physician Coaching n=359 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=160 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Mental Health Screening Rate	0.574 Proportion of patients	0.394 Proportion of patients	—

SECONDARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients who fall in the subset of consistent opioid use.

Proportion of opioid patients signing pain management agreements

Outcome measures

Measure	Physician Coaching n=359 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=160 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Use of Pain Management Agreements	0.782 Proportion of patients	0.542 Proportion of patients	—

SECONDARY outcome

Timeframe: Up to 3 years

Population: This is the number of patients who fall in the subset of consistent opioid use.

Proportion of opioid prescriptions above 120 mg daily morphine equivalent

Outcome measures

Measure	Physician Coaching n=359 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=160 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
High-dose Patients	0.191 Proportion of patients	0.125 Proportion of patients	—

SECONDARY outcome

Timeframe: 3 months

Population: This is at the prescriber level.

Number and percentage of providers who drop out of study at 3 months

Outcome measures

Measure	Physician Coaching n=34 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=13 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Provider Drop-out Rate	0 Participants	0 Participants	—

SECONDARY outcome

Timeframe: Up to 3 years

Characteristics of participating patients vs. general patient population (race, gender, ethnicity)

Outcome measures

Measure	Physician Coaching n=29956 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=13296 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Participating Patient Demographics % female	47.1 percentage of patients	49.7 percentage of patients	—
Participating Patient Demographics % hispanic	2.2 percentage of patients	2.5 percentage of patients	—
Participating Patient Demographics % asian	2.4 percentage of patients	1.6 percentage of patients	—
Participating Patient Demographics % black	2.4 percentage of patients	2.3 percentage of patients	—
Participating Patient Demographics % native	0.5 percentage of patients	0.5 percentage of patients	—
Participating Patient Demographics % other	11.9 percentage of patients	12.2 percentage of patients	—
Participating Patient Demographics % white	82.8 percentage of patients	83.6 percentage of patients	—

SECONDARY outcome

Timeframe: Up to 12 months

Characteristics of participating clinics vs. non-participating clinics (number of patients, number of providers, overall opioid prescribing rate)

Outcome measures

Measure	Physician Coaching n=29990 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=13309 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group n=9833 Participants This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Participating Clinic Characteristics number of providers	34 participants	13 participants	20 participants
Participating Clinic Characteristics number of patients	29956 participants	13296 participants	9813 participants
Participating Clinic Characteristics overall opioid prescribing rate	359 participants	160 participants	NA participants Refused clinic data could not be obtained and analyzed.

SECONDARY outcome

Timeframe: Up to 12 months

Population: These are clinic staff that made up the clinic change teams. Since change teams were specific to the physician coaching group, no change teams were constructed in the Control group, thus there is no data to be reported.

Characteristics of participating staff (profession)

Outcome measures

Measure	Physician Coaching n=27 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Participating Staff Characteristics Medical doctor	6 Participants	—	—
Participating Staff Characteristics Nurse practitioner	1 Participants	—	—
Participating Staff Characteristics Registered nurse	4 Participants	—	—
Participating Staff Characteristics Licensed practical nurse	3 Participants	—	—
Participating Staff Characteristics Clinic operations assistant	1 Participants	—	—
Participating Staff Characteristics Clinic operations manager	2 Participants	—	—
Participating Staff Characteristics Lab staff	2 Participants	—	—
Participating Staff Characteristics Reception	3 Participants	—	—
Participating Staff Characteristics Medical assistant	5 Participants	—	—

SECONDARY outcome

Timeframe: Up to 12 months

Population: Four clinics were analyzed for this measure. Since the intervention was specific to the physician coaching group, no time was spent in the Control group, thus there is no data to be reported.

Total hours of coaching delivered/received among all clinics that received physician coaching.

Outcome measures

Measure	Physician Coaching n=4 clinics This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Intervention Fidelity physician consultant hours	85.7 hours spent delivering implementation	—	—
Intervention Fidelity Facilitator hours	237.7 hours spent delivering implementation	—	—

SECONDARY outcome

Timeframe: Up to 12 months

Population: Four clinics were analyzed. No intervention was given to control group clinics, so there is no reported data.

Total cost of coaching intervention among all clinics that received physician coaching. At clinic level.

Outcome measures

Measure	Physician Coaching n=4 clinics This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Intervention Cost	29379 Dollars	—	—

SECONDARY outcome

Timeframe: Up to 12 months

Population: This is the number of patients who have a consistent opioid Rx.

The proportion of patients who have consistent opioid Rx above a morphine equivalent daily dose about 120 mg.

Outcome measures

Measure	Physician Coaching n=389 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=186 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Proportion With MEDD >120 mg	0.191 Proportion of participants	0.125 Proportion of participants	—

SECONDARY outcome

Timeframe: Up to 12 months

Population: This is the number of patients who have a consistent opioid Rx.

The average MEDD in milligrams for patients with consistent opioid Rx.

Outcome measures

Measure	Physician Coaching n=389 Participants This group included 4 intervention primary care clinics that received an organizational coaching intervention that included in-person site visits and phone/email communication. Each participating clinic was designated one primary care physician to act as a clinic lead in working with the coach to coordinate an initial site visit, a follow-up site visit, and communicating with the coach throughout the 6-month follow-up period. Physician coaching: The coach presented the latest research on the benefits and risks of long-term opioid use. The coach helped the clinic team flowchart clinical workflows and determined the best course for implementing aspects of a checklist-based implementation guide developed to support adoption of the guidelines for opioid prescribing. The coach helped the team implement ideas using Plan-Do-Study-Act change cycles and maintained contact with the clinic lead and clinic team after the initial site visit to monitor implementation progress and offer advice.	Control Group n=186 Participants This group included 4 control primary care clinics that did not receive any intervention. A de-identified dataset was created to examine differences in outcome variables between intervention and control clinics.	Refused Group This group included 3 primary care clinics that were approached to participate in the study, but refused to participate.
Average Morphine Equivalent Daily Dose (MEDD)	84.3 milligrams Standard Deviation 16.8	60.2 milligrams Standard Deviation 20.1	—

Adverse Events

Physician Coaching

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Control Group

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Andrew Quanbeck

University of Wisconsin - Madison

Phone: 6082627385

Email: arquanbe@wisc.edu

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place