Opioid Abuse, Opioid Use, Opioid-use Disorder
Conditions
Brief summary
The objective of this research is to assess the effects of electronic health record (EHR)-based decision support tools on primary care provider (PCP) decision-making around pain treatment and opioid prescribing. The decision support tools are informed by principles of behavioral economics, whereby clinicians are nudged, though never forced, towards guideline-concordant care.
Detailed description
To test the effects of these decision support tools for improving the quality of care for pain treatment, the investigators will implement a pragmatic clinic-randomized trial across the primary care clinics of Fairview Medical Group and University of Minnesota Physicians. The study has two parallel components. The decision support tools to be tested will differ somewhat depending on whether a given patient is opioid-naïve, or whether a given patient is a current opioid-user. Four sets of analyses will be conducted separately: one for the opioid-naïve group using EHR data, one for the current opioid-user group using EHR data, one at the PCP-level using web survey data, and one at the PCP-level using MN Prescription Drug Monitoring Program (PDMP) data.
Interventions
BEHAVIORALChoice Architecture Nudge
During the choice architecture nudge intervention, Primary Care Providers (PCPs) will be sent alerts in the Electronic Health Record (EHR) system when they initiate an opioid order for a patient who has not had an opioid prescription within the past six months. The alert provides guidance language about opioid prescribing and prompts the PCP to open the SmartSet to order non-opioid treatment alternatives. PCPs can choose to ignore this, but opening the SmartSet is the default option. When the SmartSet is opened, PCPs can choose to click on a variety of treatment order options, including both non-opioid pharmacological options and non-pharmacological options (e.g., referral to physical therapy or pain clinic).
BEHAVIORALPMP Integration & Nudge
During the Prescription Drug Monitoring Program (PMP) integration & nudge intervention, Primary Care Providers (PCPs) will have integrated access to the PMP embedded within the EHR. All clinicians can already access the PMP to look up a patient's prior opioid prescriptions and prescription fills. However, this process involves signing in to the separate PMP website and can be complicated and time-consuming within typical clinical workflow. The integrated PMP tool makes it much easier and faster for a PCP to access the PMP information for a given patient.
Sponsors
National Institute on Drug Abuse (NIDA)
University of Minnesota
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
DOUBLE (Investigator, Outcomes Assessor)
Intervention model description
43 Primary Care Clinics will be randomized to be in one of 4 arms: 1) Care as usual, 2) Choice architecture nudge, 3) Prescription Drug Monitoring Program (PMP) Integration & nudge, 4) Choice architecture nudge + PMP Integration & nudge
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
\- All primary care providers from all of the Fairview and University of Minnesota Physicians study clinics
Exclusion criteria
\- Primary care providers who work less than 20% full time equivalent (FTE)
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Opioid Prescription Rate | 12 months | Outcome reported as the percent of Primary Care Appointments (PCAs) at each clinic during which an opioid is prescribed, without currently receiving a non-opioid alternative pain treatment (including a new order for a non-opioid pain treatment). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Rate of Non-Opioid Treatment Prescription | 12 months | Outcome reported as the percent of Primary Care Appointments (PCAs) at each clinic during which a CDC-recommended non-opioid treatment in ordered |
| Opioid Prescription Length | 12 months | Outcome reported as the mean length of opioid prescription (in days) given during Primary Care Appointments |
| Opioid Prescription MME | 12 months | Outcome reported as the mean Milligram Morphine Equivalents (MME) of opioid prescriptions given during Primary Care Appointments |
Countries
United States
Participant flow
Pre-assignment details
The original study protocol included a potential second randomization period after six months into the trial. This secondary randomization did not occur, per the recommendation of the Data Monitoring Committee
Participants by arm
| Arm | Count |
|---|---|
| Care as Usual Clinics assigned to this arm will continue to care for the patients as usual in regards to opioid prescribing. | 126 |
| Choice Architecture Nudge Clinics in this arm will receive the choice architecture nudge intervention.
Choice Architecture Nudge: During the choice architecture nudge intervention, Primary Care Providers (PCPs) will be sent alerts in the Electronic Health Record (EHR) system when they initiate an opioid order for a patient who has not had an opioid prescription within the past six months. The alert provides guidance language about opioid prescribing and prompts the PCP to open the SmartSet to order non-opioid treatment alternatives. PCPs can choose to ignore this, but opening the SmartSet is the default option. When the SmartSet is opened, PCPs can choose to click on a variety of treatment order options, including both non-opioid pharmacological options and non-pharmacological options (e.g., referral to physical therapy or pain clinic). | 112 |
| PMP Integration & Nudge Clinics in this arm will receive the Prescription Drug Monitoring (PMP) Integration & Nudge intervention.
PMP Integration & Nudge: During the Prescription Drug Monitoring Program (PMP) integration & nudge intervention, Primary Care Providers (PCPs) will have integrated access to the PMP embedded within the EHR. All clinicians can already access the PMP to look up a patient's prior opioid prescriptions and prescription fills. However, this process involves signing in to the separate PMP website and can be complicated and time-consuming within typical clinical workflow. The integrated PMP tool makes it much easier and faster for a PCP to access the PMP information for a given patient. | 104 |
| Choice Architecture Nudge + PMP Integration & Nudge Clinics in this arm will receive both the choice architecture nudge and prescription drug monitoring (PMP) integration & nudge interventions.
Choice Architecture Nudge: During the choice architecture nudge intervention, Primary Care Providers (PCPs) will be sent alerts in the Electronic Health Record (EHR) system when they initiate an opioid order for a patient who has not had an opioid prescription within the past six months. The alert provides guidance language about opioid prescribing and prompts the PCP to open the SmartSet to order non-opioid treatment alternatives. PCPs can choose to ignore this, but opening the SmartSet is the default option. When the SmartSet is opened, PCPs can choose to click on a variety of treatment order options, including both non-opioid pharmacological options and non-pharmacological options (e.g., referral to physical therapy or pain clinic).
PMP Integration & Nudge: During the Prescription Drug Monitoring Program (PMP) integration & nudge intervention, Primary Care Providers (PCPs) will have integrated access to the PMP embedded within the EHR. All clinicians can already access the PMP to look up a patient's prior opioid prescriptions and prescription fills. However, this process involves signing in to the separate PMP website and can be complicated and time-consuming within typical clinical workflow. The integrated PMP tool makes it much easier and faster for a PCP to access the PMP information for a given patient. | 131 |
| Total | 473 |
Baseline characteristics
| Characteristic | Care as Usual | Choice Architecture Nudge | PMP Integration & Nudge | Choice Architecture Nudge + PMP Integration & Nudge | Total |
|---|---|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical Between 18 and 65 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race and Ethnicity Not Collected | — | — | — | — | 0 Participants |
| Sex/Gender, Customized Female | 70 Participants | 64 Participants | 66 Participants | 86 Participants | 286 Participants |
| Sex/Gender, Customized Male | 49 Participants | 44 Participants | 38 Participants | 42 Participants | 173 Participants |
| Sex/Gender, Customized Unkown | 7 Participants | 4 Participants | 0 Participants | 3 Participants | 14 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk | EG003 affected / at risk |
|---|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 0 | 0 / 0 | 0 / 0 | 0 / 0 |
| other Total, other adverse events | 0 / 0 | 0 / 0 | 0 / 0 | 0 / 0 |
| serious Total, serious adverse events | 0 / 0 | 0 / 0 | 0 / 0 | 0 / 0 |
Outcome results
Primary
Opioid Prescription Rate
Outcome reported as the percent of Primary Care Appointments (PCAs) at each clinic during which an opioid is prescribed, without currently receiving a non-opioid alternative pain treatment (including a new order for a non-opioid pain treatment).
Time frame: 12 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Care as Usual | Opioid Prescription Rate | 0.0159 percent | Standard Deviation 0.125 |
| Choice Architecture Nudge | Opioid Prescription Rate | 0.0141 percent | Standard Deviation 0.118 |
| PMP Integration & Nudge | Opioid Prescription Rate | 0.0161 percent | Standard Deviation 0.129 |
| Choice Architecture Nudge + PMP Integration & Nudge | Opioid Prescription Rate | 0.0120 percent | Standard Deviation 0.109 |
Secondary
Opioid Prescription Length
Outcome reported as the mean length of opioid prescription (in days) given during Primary Care Appointments
Time frame: 12 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Care as Usual | Opioid Prescription Length | 9.63 Days | Standard Deviation 18.66 |
| Choice Architecture Nudge | Opioid Prescription Length | 11.85 Days | Standard Deviation 21.4 |
| PMP Integration & Nudge | Opioid Prescription Length | 13.43 Days | Standard Deviation 35.86 |
| Choice Architecture Nudge + PMP Integration & Nudge | Opioid Prescription Length | 11.75 Days | Standard Deviation 25.39 |
Secondary
Opioid Prescription MME
Outcome reported as the mean Milligram Morphine Equivalents (MME) of opioid prescriptions given during Primary Care Appointments
Time frame: 12 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Care as Usual | Opioid Prescription MME | 23.41 Milligram Morphine Equivalents | Standard Deviation 14.9 |
| Choice Architecture Nudge | Opioid Prescription MME | 23.89 Milligram Morphine Equivalents | Standard Deviation 18.3 |
| PMP Integration & Nudge | Opioid Prescription MME | 21.6 Milligram Morphine Equivalents | Standard Deviation 12.1 |
| Choice Architecture Nudge + PMP Integration & Nudge | Opioid Prescription MME | 23.89 Milligram Morphine Equivalents | Standard Deviation 15.9 |
Secondary
Rate of Non-Opioid Treatment Prescription
Outcome reported as the percent of Primary Care Appointments (PCAs) at each clinic during which a CDC-recommended non-opioid treatment in ordered
Time frame: 12 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Care as Usual | Rate of Non-Opioid Treatment Prescription | 0.239 percent | Standard Deviation 0.426 |
| Choice Architecture Nudge | Rate of Non-Opioid Treatment Prescription | 0.227 percent | Standard Deviation 0.419 |
| PMP Integration & Nudge | Rate of Non-Opioid Treatment Prescription | 0.23 percent | Standard Deviation 0.421 |
| Choice Architecture Nudge + PMP Integration & Nudge | Rate of Non-Opioid Treatment Prescription | 0.228 percent | Standard Deviation 0.419 |