Addiction
Conditions
Keywords
Organizational change, process improvement, addiction treatment, cost effectiveness study
Brief summary
Addiction treatment is often characterized by long delays between first contact and treatment as well as high no-show and drop out rates leading to unused capacity in apparently full agencies. Patients do not get needed care and agency financial stability is threatened. The Network for Improvement of Addiction Treatment (NIATx) began as a high-intensity improvement collaborative of 39 addiction treatment agencies distributed across 25 states. NIATx substantially improved time to treatment and continuation in treatment by making improvements to organizational processes (such as first contact, intake and assessment, engagement, level of care transitions, paperwork, social support, outreach, and scheduling) in preliminary studies. While the results are very encouraging, they have, by intent, been obtained from a select group of agencies using a high-cost combination of services. A more practical diffusion model is needed to spread process improvements across the spectrum of treatment agencies. This study is a cluster-randomized trial to test the effectiveness and cost of less expensive combinations of the services that make up the NIATx collaborative (interest circles, coach calls, coach visits and learning sessions).
Detailed description
This cluster-RCT randomly assign 201 treatment agencies in 5 states to four experimental arms. The agencies were randomized to an intervention for 18 months with a 9 month sustainability period. The study aimed to: 1) Determine whether a state-based strategy can (with NIATx support) can lead mainstream treatment agencies to implement and sustain process changes that improve the study's primary outcomes: time to treatment, annual clinic admissions, and continuation in treatment; and 2) Evaluate the effectiveness and cost of the services making up NIATx. This study aims to create a practical model for improving efficiency and effectiveness of addiction treatment.
Interventions
OTHERLearning Session
Learning Sessions occur bi-annually as change teams convene to learn and gather support from each other and outside experts who offer advice on how best to adopt the innovations and learn about new directions for the collaborative (e.g., the need to create business cases for improvements). Learning Sessions and Interest Circles (see below) have similar objectives-to help agencies learn and gather support from each other and from outside experts.
OTHERInterest Circle Calls
Interest Circles are monthly teleconferences where agency change leaders discuss change-related issues and progress. Circles address how to improve timeliness, continuation, admissions, dropouts and transitions. They also address specialty topics (e.g., programs for women, adolescents). Participants discuss successes, failures, and challenges, and get advice and assignments for their improvement plans. Meeting summaries appear on the Web site. Interest Circles are inexpensive, but are they are sufficient? Should Interest Circles prove effective, they would provide a low-cost, convenient diffusion approach
OTHERCoaching
Coaching assigns an expert in process improvement to work with an agency to make, sustain, and spread process improvement efforts. Consultations focus on executive directors, change leaders and improvement teams. Coaches help agencies address key issues, but also broker relationships with other agencies, offer process improvement training, and promote the innovations to make and how to make them. Coaching takes place during site visits, monthly phone conferences, and via email.
OTHERWebsite
The NIATx Web site features resources central to improvement. The site includes: 1) a catalog of change ideas and case studies; 2) a toolbox providing just-in-time training on topics such as conducting a walk-through and key innovations; 3) on-line tools to assess organizational (or project) readiness for and ability to sustain change; 4) electronic communication services to ask questions of experts, and participate in peer discussion groups; 5) links to relevant process improvement Web sites; and 6) a secure portion for treatment agencies to report and track progress. Hence, our control group will have access to the entire website.
Sponsors
Oregon Health and Science University
National Institute on Drug Abuse (NIDA)
University of Wisconsin, Madison
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Healthy volunteers
Yes
Inclusion criteria
* at least 60 admissions/year * provide outpatient and intensive outpatient levels of care (as defined by ASAM) * provide or use detox services provided by others * have tax-exempt or government status or rely on public funding (e.g., block grants, Medicare, Medicaid, local government, private philanthropy) for at least 50% of their budget * have adopted no more than two of the planned interventions
Exclusion criteria
* are current NIATx members
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in Average Waiting Time From First Contact to Treatment | Baseline and 15 months | The average length of time in days it takes from when a patient first calls for help to the time a patient was able to meet a clinician. In this quality improvement study, changes in this measure over time are reported. Estimates of improvement show the average days of improvement per month based on a best linear unbiased predictor estimate for each site. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis. |
| Change in Annual Number of Patient Admissions | 48 months (2 year baseline period and 2 year post-intervention period) | We aimed to increase clinics' treatment capacity in this quality improvement study. Capacity was measured by counting clinics' annual number of patient admissions. We monitored changes in admission counts, per clinic, in a pre-post analysis. Changes in the natural logarithm of annual admissions are presented, which approximates the average percentage change (year-to-year) in the number of new patient admissions per clinic. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis. |
| Change in Average Continuation Rate Through the Fourth Treatment Session | Baseline and 21 months | This outcome represents change in the rate at which a clinic's patients continue in treatment. Continuation rate is defined as the percentage of patients that make at least 4 visits to the clinic, on different days, before being discharged. Estimates of improvement show the average percentage points of improvement per month based on a best linear unbiased predictor estimate for each site. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cost of Group | Baseline and 18 months | The goal of the economic analysis was to estimate costs of each group for governmental authorities who might organize improvement collaboratives. We collected the cost of personnel (state employees, NIATx employees, coaches and consultants), data management, buildings and facilities, lodging, travel, telephone calls and miscellaneous costs. Costs were categorized as group specific (such as hotel costs for the learning sessions group) or non-group-specific, which included state-incurred costs for outreach, data management and infrastructure, encouraging participation and administration. Cost data were collected three times during the study period and aggregated to create a total cost estimate. Figures reported below represent costs at the arm/group level (costs were not assessed at the organizational level). Measure type is Number. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Interest Circle Call + Website Interest Circles are monthly teleconferences where agency change leaders discuss change-related issues and progress. Circles address how to improve timeliness, continuation, admissions, dropouts and transitions. They also address specialty topics (e.g., programs for women, adolescents). Participants discuss successes, failures, and challenges, and get advice and assignments for their improvement plans. Meeting summaries appear on the Web site. Interest Circles are inexpensive, but are they are sufficient? Should Interest Circles prove effective, they would provide a low-cost, convenient diffusion approach. | 49 |
| Coaching + Website Coaching assigns an expert in process improvement to work with an agency to make, sustain, and spread process improvement efforts. Consultations focus on executive directors, change leaders and improvement teams. Coaches help agencies address key issues, but also broker relationships with other agencies, offer process improvement training, and promote the innovations to make and how to make them. Coaching takes place during site visits, monthly phone conferences, and via email. | 50 |
| Full: LS, Coaching, ICC, Website Learning Session, Coaching, Interest Circle Calls, Website, see descriptions above | 48 |
| Learning Session + Website Learning Sessions occur bi-annually as change teams convene to learn and gather support from each other and outside experts who offer advice on how best to adopt the innovations and learn about new directions for the collaborative (e.g., the need to create business cases for improvements). Learning Sessions and Interest Circles (see below) have similar objectives-to help agencies learn and gather support from each other and from outside experts. | 54 |
| Total | 201 |
Baseline characteristics
| Characteristic | Interest Circle Call + Website | Coaching + Website | Full: LS, Coaching, ICC, Website | Learning Session + Website | Total |
|---|---|---|---|---|---|
| Age Continuous | 33.36 years STANDARD_DEVIATION 11.81 | 31.89 years STANDARD_DEVIATION 12.02 | 32.71 years STANDARD_DEVIATION 11.81 | 32.08 years STANDARD_DEVIATION 12.19 | 32.53 years STANDARD_DEVIATION 11.96 |
| Clinics' annual number of patient admissions per year | 500 Patients per year STANDARD_DEVIATION 485 | 578 Patients per year STANDARD_DEVIATION 684 | 488 Patients per year STANDARD_DEVIATION 437 | 671 Patients per year STANDARD_DEVIATION 1031 | 559.25 Patients per year STANDARD_DEVIATION 659.25 |
| Organizational structure Private for-profit | 0 participants | 0 participants | 0 participants | 3 participants | 3 participants |
| Organizational structure Private not-for-profit | 37 participants | 41 participants | 41 participants | 43 participants | 162 participants |
| Organizational structure Unit of other government | 7 participants | 6 participants | 1 participants | 3 participants | 17 participants |
| Organizational structure Unit of state government | 4 participants | 1 participants | 4 participants | 3 participants | 12 participants |
| Organizational structure Unit of tribal government | 1 participants | 2 participants | 2 participants | 2 participants | 7 participants |
| Primary setting Community mental health clinic | 7 participants | 13 participants | 5 participants | 10 participants | 35 participants |
| Primary setting Corrections | 3 participants | 0 participants | 1 participants | 1 participants | 5 participants |
| Primary setting Family or children's service agency | 3 participants | 2 participants | 1 participants | 2 participants | 8 participants |
| Primary setting Free-standing alcohol or drug treatment clinic | 26 participants | 25 participants | 26 participants | 32 participants | 109 participants |
| Primary setting Hospital/Health Center (including primary setting) | 4 participants | 6 participants | 5 participants | 3 participants | 18 participants |
| Primary setting Other or unreported | 5 participants | 2 participants | 7 participants | 4 participants | 18 participants |
| Primary setting Social services agency | 1 participants | 2 participants | 3 participants | 2 participants | 8 participants |
| Race/Ethnicity, Customized American Indian or Alaska Native | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Race/Ethnicity, Customized Asian | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Race/Ethnicity, Customized Black or African American | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Race/Ethnicity, Customized More than one race | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Race/Ethnicity, Customized Native Hawaiian or Other Pacific Islander | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Race/Ethnicity, Customized Unknown or Not Reported | 49 Non-white or Hispanic participants | 50 Non-white or Hispanic participants | 48 Non-white or Hispanic participants | 54 Non-white or Hispanic participants | 201 Non-white or Hispanic participants |
| Race/Ethnicity, Customized White | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants | 0 Non-white or Hispanic participants |
| Rating of management quality | 3.02 Units on a scale between 1 and 5 STANDARD_DEVIATION 0.54 | 2.97 Units on a scale between 1 and 5 STANDARD_DEVIATION 0.64 | 2.99 Units on a scale between 1 and 5 STANDARD_DEVIATION 0.57 | 2.95 Units on a scale between 1 and 5 STANDARD_DEVIATION 0.73 | 2.98 Units on a scale between 1 and 5 STANDARD_DEVIATION 0.62 |
| Region of Enrollment United States | 49 participants | 50 participants | 48 participants | 54 participants | 201 participants |
| Sex: Female, Male Female | 16 Participants | 17 Participants | 16 Participants | 17 Participants | 66 Participants |
| Sex: Female, Male Male | 33 Participants | 33 Participants | 32 Participants | 37 Participants | 135 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 | — / — | — / — | — / — | — / — |
| other Total, other adverse events | 0 / 0 | 0 / 0 | 0 / 0 | 0 / 0 |
| serious Total, serious adverse events | 0 / 49 | 0 / 50 | 0 / 48 | 0 / 54 |
Outcome results
Primary
Change in Annual Number of Patient Admissions
We aimed to increase clinics' treatment capacity in this quality improvement study. Capacity was measured by counting clinics' annual number of patient admissions. We monitored changes in admission counts, per clinic, in a pre-post analysis. Changes in the natural logarithm of annual admissions are presented, which approximates the average percentage change (year-to-year) in the number of new patient admissions per clinic. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis.
Time frame: 48 months (2 year baseline period and 2 year post-intervention period)
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Interest Circle Call + Website | Change in Annual Number of Patient Admissions | -3.6 Percent change (approx.) | 95% Confidence Interval 0.04 |
| Coaching + Website | Change in Annual Number of Patient Admissions | 19.5 Percent change (approx.) | 95% Confidence Interval 0.089 |
| Full: LS, Coaching, ICC, Website | Change in Annual Number of Patient Admissions | 0.0 Percent change (approx.) | 95% Confidence Interval 0.065 |
| Learning Session + Website | Change in Annual Number of Patient Admissions | 8.9 Percent change (approx.) | 95% Confidence Interval 0.041 |
Primary
Change in Average Continuation Rate Through the Fourth Treatment Session
This outcome represents change in the rate at which a clinic's patients continue in treatment. Continuation rate is defined as the percentage of patients that make at least 4 visits to the clinic, on different days, before being discharged. Estimates of improvement show the average percentage points of improvement per month based on a best linear unbiased predictor estimate for each site. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis.
Time frame: Baseline and 21 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Interest Circle Call + Website | Change in Average Continuation Rate Through the Fourth Treatment Session | -.019 Change in continuation rate | Standard Error 0.021 |
| Coaching + Website | Change in Average Continuation Rate Through the Fourth Treatment Session | -.02 Change in continuation rate | Standard Error 0.016 |
| Full: LS, Coaching, ICC, Website | Change in Average Continuation Rate Through the Fourth Treatment Session | -.004 Change in continuation rate | Standard Error 0.017 |
| Learning Session + Website | Change in Average Continuation Rate Through the Fourth Treatment Session | .013 Change in continuation rate | Standard Error 0.018 |
Primary
Change in Average Waiting Time From First Contact to Treatment
The average length of time in days it takes from when a patient first calls for help to the time a patient was able to meet a clinician. In this quality improvement study, changes in this measure over time are reported. Estimates of improvement show the average days of improvement per month based on a best linear unbiased predictor estimate for each site. Note: this study has three primary outcomes. The number of participants analyzed varies for each outcome. The (higher) number of clinics shown in the flow diagram results because clinics may have been analyzed on a subset of the three primary outcomes (e.g., analyzed for waiting time and continuation, but not for annual number of new patients). To be considered analyzed in the flow diagram, a clinic must have been included in at least one primary outcomes analysis.
Time frame: Baseline and 15 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Interest Circle Call + Website | Change in Average Waiting Time From First Contact to Treatment | .253 Change in days | Standard Error 1.514 |
| Coaching + Website | Change in Average Waiting Time From First Contact to Treatment | 4.610 Change in days | Standard Error 1.414 |
| Full: LS, Coaching, ICC, Website | Change in Average Waiting Time From First Contact to Treatment | 4.723 Change in days | Standard Error 1.421 |
| Learning Session + Website | Change in Average Waiting Time From First Contact to Treatment | 3.520 Change in days | Standard Error 1.407 |
Secondary
Cost of Group
The goal of the economic analysis was to estimate costs of each group for governmental authorities who might organize improvement collaboratives. We collected the cost of personnel (state employees, NIATx employees, coaches and consultants), data management, buildings and facilities, lodging, travel, telephone calls and miscellaneous costs. Costs were categorized as group specific (such as hotel costs for the learning sessions group) or non-group-specific, which included state-incurred costs for outreach, data management and infrastructure, encouraging participation and administration. Cost data were collected three times during the study period and aggregated to create a total cost estimate. Figures reported below represent costs at the arm/group level (costs were not assessed at the organizational level). Measure type is Number.
Time frame: Baseline and 18 months
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Interest Circle Call + Website | Cost of Group | 59790 USD ($) |
| Coaching + Website | Cost of Group | 135244 USD ($) |
| Full: LS, Coaching, ICC, Website | Cost of Group | 380652 USD ($) |
| Learning Session + Website | Cost of Group | 229229 USD ($) |