Depressive Disorder, Drug Abuse, Drug Addiction, Drug Dependence, Drug Habituation, Drug Use Disorders, Pain
Conditions
Brief summary
Background: Sixty million American adults suffer from moderate to severe chronic pain. Of these, 5 to 8 million currently use opioids long-term. With increased opioid prescribing for chronic pain, an epidemic of prescription opioid addiction and overdose has arisen. This necessitates action to stem opioid-related morbidity and mortality. Group Health (GH), a large nonprofit health plan, developed and implemented opioid risk reduction strategies for doctors and patients in some, but not all, of its clinics. The risk reduction initiative achieved large opioid dose reductions, near universal documentation of care plans, and marked increases in patient monitoring. Rigorous evaluation of patient outcomes resulting from the opioid risk reduction initiative, incorporating patient perspectives, is needed to guide health care improvement efforts to reduce opioid risks regionally and nationally. Research goal: The investigators will evaluate a major health plan initiative to reduce risks of long-term opioid use for chronic pain. Starting in 2008, some GH clinics reduced prescribing of high opioid doses. In 2010 the same clinics increased care planning and monitoring of chronic opioid therapy (COT) patients. Our research goal is to evaluate effects of this initiative on health and safety outcomes of COT patients. We will test whether the initiative influenced pain outcomes; patient-reported opioid benefits and problems; and opioid-related adverse events. Design and Outcomes: The investigators will assess effects of GH's opioid risk reduction initiative among COT patients using opioids long-term. The investigators will compare COT patients from clinics that implemented the initiative with COT patients from care settings that did not implement the initiative. The investigators will use survey data to assess patient-reported outcomes including pain severity, depressive symptoms, and patient perceptions of opioid benefits and problems, including validated measures of prescription opioid use disorder. They will interview and compare 800 COT patients using opioids long-term from clinics that implemented the risk reduction initiative and 800 COT patients from care settings that did not. Impact: This research will provide an urgently needed, rigorous evaluation of a major risk reduction initiative among COT patients. Evaluation results will guide efforts of health plans, clinicians and patients nationwide to ensure safe, effective and compassionate chronic pain care.
Detailed description
Detailed description not provided
Interventions
None listed
Sponsors
Kaiser Permanente
Study design
Observational model
COHORT
Time perspective
CROSS_SECTIONAL
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Enrolled for at least one year prior to sampling for survey * Received 70+ days supply of opioids in 2 of 4 quarters, including the most recent 90 days * Received 45+ days supply of opioids in the other 2 quarters
Exclusion criteria
* Received cancer diagnoses, other than non-melanoma skin cancer, at least twice in the prior year * Received any skilled nursing facility care in the prior year * Received any hospice care or * Received any opioids from an oncologist
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Proportion With Prescription Opioid Use Disorder (Defined by DSM5 Criteria). | 1 year prior to interview | Proportion with Prescription Opioid Use Disorder, which is defined by multiple indicators of opioid abuse and addiction from 9 criteria in the DSM5 manual of the American Psychiatric Association. In this research, it will be assessed using the Psychiatric Research Interview for Substance and Mental Disorders (PRISM5, Columbia University). Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Pain Severity (Intensity, Interference With Activities, Enjoyment): PEG Scale | 1 week prior to interview | PEG (Pain - Enjoyment - Interference with General Activities) pain scale consisting of the average of 3 0-10 ratings of pain intensity, interference with activities due to pain, and reduced enjoyment of life due to pain. Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The PEG score ranges from 0 to 30, with higher scores indicating greater pain severity. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation. |
| Depressive Symptoms | 2 weeks prior to interview | Patient Health Questionnaire (PHQ-8) depression scale. The PHQ-8 is estimated by summing the 8 scale items. The total score ranges from 0 to 24, with higher scores indicating greater depression severity. Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation. |
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Opioid Risk Reduction Initiative Opioid risk reduction initiatives (dose reduction and then risk stratification and monitoring) for chronic opioid therapy patients implemented in Group Health integrated group practice clinics. | 935 |
| Usual Care Usual care for management of chronic opioid therapy patients implemented in Group Health network care settings. | 653 |
| Total | 1,588 |
Baseline characteristics
| Characteristic | Opioid Risk Reduction Initiative | Total | Usual Care |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 428 Participants | 666 Participants | 238 Participants |
| Age, Categorical Between 18 and 65 years | 507 Participants | 922 Participants | 415 Participants |
| Age, Continuous | 62.6 years STANDARD_DEVIATION 12.1 | 62.0 years STANDARD_DEVIATION 12.1 | 61.0 years STANDARD_DEVIATION 12.1 |
| Alcohol use disorder, past 36 months No | 885 Participants | 1506 Participants | 621 Participants |
| Alcohol use disorder, past 36 months Yes | 50 Participants | 82 Participants | 32 Participants |
| Average daily opioid dose in morphine equivalents 120 milligrams or greater | 72 Participants | 183 Participants | 111 Participants |
| Average daily opioid dose in morphine equivalents 15 to less than 50 milligrams | 432 Participants | 728 Participants | 296 Participants |
| Average daily opioid dose in morphine equivalents 50 to less than 120 milligrams | 199 Participants | 349 Participants | 150 Participants |
| Average daily opioid dose in morphine equivalents Less than 15 milligrams | 232 Participants | 326 Participants | 94 Participants |
| Average daily opioid dose in morphine equivalents No opioid use | 0 Participants | 2 Participants | 2 Participants |
| Charlson Comorbidity Score (past 12 months) 0 | 426 Participants | 733 Participants | 307 Participants |
| Charlson Comorbidity Score (past 12 months) 1 | 75 Participants | 148 Participants | 73 Participants |
| Charlson Comorbidity Score (past 12 months) 2 | 144 Participants | 251 Participants | 107 Participants |
| Charlson Comorbidity Score (past 12 months) 3-5 | 195 Participants | 322 Participants | 127 Participants |
| Charlson Comorbidity Score (past 12 months) 6 or greater | 95 Participants | 134 Participants | 39 Participants |
| Continuously enrolled in health plan for past 24 months No | 48 Participants | 146 Participants | 98 Participants |
| Continuously enrolled in health plan for past 24 months Yes | 887 Participants | 1442 Participants | 555 Participants |
| Education College graduate or post-college graduate | 255 Participants | 407 Participants | 152 Participants |
| Education High school graduate or less | 243 Participants | 446 Participants | 203 Participants |
| Education Missing | 1 Participants | 3 Participants | 2 Participants |
| Education Some college or technical school | 436 Participants | 732 Participants | 296 Participants |
| Employment status Disabled | 182 Participants | 352 Participants | 170 Participants |
| Employment status Employed full time/part time | 314 Participants | 537 Participants | 223 Participants |
| Employment status Missing | 1 Participants | 1 Participants | 0 Participants |
| Employment status Other | 28 Participants | 43 Participants | 15 Participants |
| Employment status Retired | 410 Participants | 655 Participants | 245 Participants |
| Hepatitis C or cirrhosis diagnosis No | 851 Participants | 1458 Participants | 607 Participants |
| Hepatitis C or cirrhosis diagnosis Yes | 84 Participants | 130 Participants | 46 Participants |
| Location in Washington State Eastern Washington | 242 Participants | 537 Participants | 295 Participants |
| Location in Washington State Western Washington | 693 Participants | 1051 Participants | 358 Participants |
| Marital status Married | 615 Participants | 1053 Participants | 438 Participants |
| Marital status Missing | 2 Participants | 3 Participants | 1 Participants |
| Marital status Not married | 318 Participants | 532 Participants | 214 Participants |
| Mental disorder diagnosis past 36 months No | 312 Participants | 542 Participants | 230 Participants |
| Mental disorder diagnosis past 36 months Yes | 623 Participants | 1046 Participants | 423 Participants |
| Non-opioid drug use disorder diagnosis past 36 months No | 846 Participants | 1465 Participants | 619 Participants |
| Non-opioid drug use disorder diagnosis past 36 months Yes | 89 Participants | 123 Participants | 34 Participants |
| Opioid drug use disorder diagnosis, past 36 months No | 834 Participants | 1411 Participants | 577 Participants |
| Opioid drug use disorder diagnosis, past 36 months Yes | 101 Participants | 177 Participants | 76 Participants |
| Race/Ethnicity, Customized African American | 19 Participants | 22 Participants | 3 Participants |
| Race/Ethnicity, Customized Asian/Pacific Islander | 11 Participants | 15 Participants | 4 Participants |
| Race/Ethnicity, Customized Caucasian | 804 Participants | 1380 Participants | 576 Participants |
| Race/Ethnicity, Customized Missing | 11 Participants | 24 Participants | 13 Participants |
| Race/Ethnicity, Customized Multiple or other | 83 Participants | 137 Participants | 54 Participants |
| Race/Ethnicity, Customized Native American | 7 Participants | 10 Participants | 3 Participants |
| Received excess days supply of 20% or greater in any of past 4 quarters No | 734 Participants | 1227 Participants | 493 Participants |
| Received excess days supply of 20% or greater in any of past 4 quarters Yes | 201 Participants | 361 Participants | 160 Participants |
| Sex: Female, Male Female | 589 Participants | 1009 Participants | 420 Participants |
| Sex: Female, Male Male | 346 Participants | 579 Participants | 233 Participants |
| Tobacco use disorder diagnosis, past 36 months No | 711 Participants | 1213 Participants | 502 Participants |
| Tobacco use disorder diagnosis, past 36 months Yes | 224 Participants | 375 Participants | 151 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 935 | 0 / 653 |
| other Total, other adverse events | 0 / 935 | 0 / 653 |
| serious Total, serious adverse events | 0 / 935 | 0 / 653 |
Outcome results
Primary
Proportion With Prescription Opioid Use Disorder (Defined by DSM5 Criteria).
Proportion with Prescription Opioid Use Disorder, which is defined by multiple indicators of opioid abuse and addiction from 9 criteria in the DSM5 manual of the American Psychiatric Association. In this research, it will be assessed using the Psychiatric Research Interview for Substance and Mental Disorders (PRISM5, Columbia University). Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation.
Time frame: 1 year prior to interview
Population: Prevalent chronic opioid therapy (COT) patients defined as having received at least 70 days supply of opioids in the 90 days prior to sample selection, at least 70 days supply of opioids in at least one of the 3 other quarters in the preceding year, and at least 45 days supply of opioids in the other two quarters.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Opioid Risk Reduction Initiative | Proportion With Prescription Opioid Use Disorder (Defined by DSM5 Criteria). | 0.215 proportion of COT patients |
| Usual Care | Proportion With Prescription Opioid Use Disorder (Defined by DSM5 Criteria). | 0.239 proportion of COT patients |
Secondary
Depressive Symptoms
Patient Health Questionnaire (PHQ-8) depression scale. The PHQ-8 is estimated by summing the 8 scale items. The total score ranges from 0 to 24, with higher scores indicating greater depression severity. Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation.
Time frame: 2 weeks prior to interview
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Opioid Risk Reduction Initiative | Depressive Symptoms | 7.39 units on a scale |
| Usual Care | Depressive Symptoms | 7.91 units on a scale |
Secondary
Pain Severity (Intensity, Interference With Activities, Enjoyment): PEG Scale
PEG (Pain - Enjoyment - Interference with General Activities) pain scale consisting of the average of 3 0-10 ratings of pain intensity, interference with activities due to pain, and reduced enjoyment of life due to pain. Estimates were weighted to account for non-response, with weights based on baseline patient characteristics assessed with electronic health care data available for all chronic opioid therapy patients sampled for the survey. The PEG score ranges from 0 to 30, with higher scores indicating greater pain severity. The number analyzed differs from the Participant Flow module due to persons with missing item data who were dropped from this analysis. The numbers with missing items were deemed too few to justify item imputation.
Time frame: 1 week prior to interview
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Opioid Risk Reduction Initiative | Pain Severity (Intensity, Interference With Activities, Enjoyment): PEG Scale | 5.79 units on a scale |
| Usual Care | Pain Severity (Intensity, Interference With Activities, Enjoyment): PEG Scale | 5.82 units on a scale |