SKIP for PA Study: Team and Leadership Level Implementation Support for Collaborative Care (SKIPforPA)

Care Team and Practice Level Implementation Strategies to Optimize Pediatric Collaborative Care: A Cluster-Randomized Trial

National Institute of Mental Health (NIMH), American Academy of Pediatrics, Drexel University, NYU Langone Health, University of Michigan, Pennsylvania Chapter American Academy of Pediatrics

In a prior application (MH064372), the investigators' treatment research program (Services for Kids In Primary-care, SKIP) developed and tested a chronic care model-based intervention, called Doctor Office Collaborative Care (DOCC), that was found to be effective in the management of childhood behavior problems and comorbid ADHD. In the "SKIP for PA Study", the investigators propose to conduct a randomized clinical trial to evaluate the effects of team- and practice leadership-level implementation strategies designed to enhance the use and uptake of DOCC in diverse pediatric primary care offices.

This study is a randomized, hybrid type 3 effectiveness-implementation trial to support the adoption of a chronic care model (CCM)-based intervention in pediatric primary care settings by testing the impact of implementation strategies directed towards the provider care team (TEAM) or practice leadership (LEAD) level. The treatment investigators seek to deliver here is called Doctor Office Collaborative Care (DOCC), an evidence-based intervention for the management of child behavior problems and comorbid ADHD. The implementation strategies being tested to enhance DOCC uptake include TEAM coaching/consultation strategies, which will be delivered to care team providers and target provider competency to deliver DOCC, and LEAD facilitation strategies, which will be delivered to practice leaders and target organizational support of DOCC delivery. These multi-level implementation strategies have not been formally evaluated to learn about their separate and combined effects in any randomized clinical trial conducted in pediatric primary care. Such information is needed to optimize our approaches to promoting the implementation of a CCM-based intervention in pediatric practice. The statewide sample includes 24 primary care practices from the Medical Home Program of the Pennsylvania Chapter of the American Academy of Pediatrics. After standard training in the DOCC EBP, all practices will be randomized to one of four implementation conditions: 1) No TEAM or LEAD (ongoing technical support only); 2) TEAM implementation; 3) LEAD implementation, or 4) TEAM+LEAD implementation. TEAM and LEAD implementation will be delivered via videoconference on a graded schedule. Care teams will deliver DOCC to up to 25 children who meet a clinical cutoff for modest behavior problems and their caregivers. Investigators will collect practice/provider measures from enrolled practice staff (0, 6, 12, 18, 24 months) and caregivers over several timepoints (0, 3, 6, 12 months) to support all analyses evaluating implementation and treatment outcomes, mediation, and moderation. By proposing one of the first large pragmatic pediatric trials of a CCM-based evidence-based intervention to address these aims in response to RFA-MH-18-701 and the NIMH's Strategic Plan (4.2), this research will advance the implementation science knowledge needed to optimize promising strategies for promoting the delivery and scale-up of DOCC in a pediatric medical home.

The investigators propose a randomized trial using a factorial design to test effects of implementation support strategies on uptake of an evidence-based intervention (DOCC) in 24 primary care practices from the Medical Home Program of the Pennsylvania Chapter of the American Academy of Pediatrics. Practices will receive DOCC training in accord with the Replicating Effective Programs model. Practices will be randomized to 1 of 4 implementation support conditions: 1) No TEAM/LEAD (training and technical support only); 2) provider care team consultation (TEAM); 3) practice leadership facilitation (LEAD); and 4) combination of provider care team and leadership implementation (TEAM+LEAD). Care teams in each practice will identify families and deliver DOCC to 20 caregivers and their children with externalizing behavior. Investigators will collect measures from 120 practice staff over 5 assessments (24 months) and 480 caregivers over 4 assessments (12 months) to support analyses.

Participants are not directly informed of the implementation condition to which the practice site has been randomized. All formal standardized assessments will be conducted via secure electronic web-based data collection systems, limiting the potential for assessment bias. In the event that a caregiver requests that a staff member administer an assessment via phone, the assessor would be unaware of the implementation condition assigned to the child's practice where the caregiver receives care.

Practices in this arm will receive DOCC materials/training and technical support, but will not receive care team coaching/consultation (TEAM) or practice leadership facilitation (LEAD) after the training phase.

Practices in this arm will receive DOCC training and materials and one type of implementation support after the training: coaching/consultation for the provider care team (TEAM).

Practices in this arm will receive DOCC training and materials and only one type of implementation support after the training: facilitation for practice leadership (LEAD).

Practices in this arm will receive DOCC training and materials and both types of implementation support after the training: coaching/consultation for the provider care team (TEAM) and facilitation for practice leadership (LEAD).

Practices will learn and then deliver DOCC in treatment sessions with caregivers and/or children. The content covers key topics related to the treatment of behavior problems (e.g., self-management, positive parenting) and ADHD (e.g., psychoeducation, medication).

TEAM: Implementation support strategies at the care team level following standard implementation of DOCC

Practices will learn and then deliver DOCC in treatment sessions with caregivers and/or children. Coaching and consultation will be provided to the provider care team to support the use of collaborative care for behavior problems and ADHD. The TEAM intervention includes regular virtual meetings or calls with the providers (about once/month, on average) that cover core chronic care model functions, including registry use, case-finding, collaborative care team roles, and workflows outlining how DOCC is delivered in the practice.

Practices will learn and then deliver DOCC in treatment sessions with caregivers and/or children. Practice facilitation will be provided to practice leaders to help them support the care team's use of collaborative care for behavior problems and ADHD. The LEAD intervention includes regular virtual meetings or calls with practice leaders (about once/month, on average) that cover the assessment of practice capacity/barriers, ways to overcome organizational barriers and support staff use of DOCC, promoting innovation, and leveraging practice resources to support DOCC delivery and maintenance in the practice.

As part of a short checklist-style progress note completed after each service contact and housed in the study's web-based patient registry dashboard, providers will document each DOCC encounter delivered to each family. These encounters include referral, assessment, treatment or medication delivery, medication or treatment review/monitoring, case management, review of emergent symptoms, and care management and consultation meetings involving a family member. Investigators will report the total number of DOCC encounters delivered to each family during the intervention phase.

The Mental Health Practice Readiness Inventory will be completed to document the degree to which a practice has the organizational and individual competencies needed to support integrated BH care. All 32 items reflect the diverse activities (e.g., workflows, financing, service delivery, care coordination) suggested as functions for PCPs (scale: 0 = no function exists; 1 = some function; 2 = function is complete). Thus, total scale scores can range from 0 to 64. Investigators will use scores for each of the 32 competency items (range 0-2) and the total score (range 0-64) to describe the overall level of collaborative care competencies achieved per practice.

The Mental Health Practice Readiness Inventory will be completed to document the degree to which a practice has the organizational and individual competencies needed to support integrated BH care. All 32 items reflect the diverse activities (e.g., workflows, financing, service delivery, care coordination) suggested as functions for PCPs (scale: 0 = no function exists; 1 = some function; 2 = function is complete). Thus, total scale scores can range from 0 to 64. Investigators will use scores for each of the 32 competency items (range 0-2) and the total score (range 0-64) to describe the overall level of collaborative care competencies achieved per practice.

The Mental Health Practice Readiness Inventory will be completed to document the degree to which a practice has the organizational and individual competencies needed to support integrated BH care. All 32 items reflect the diverse activities (e.g., workflows, financing, service delivery, care coordination) suggested as functions for PCPs (scale: 0 = no function exists; 1 = some function; 2 = function is complete). Thus, total scale scores can range from 0 to 64. Investigators will use scores for each of the 32 competency items (range 0-2) and the total score (range 0-64) to describe the overall level of collaborative care competencies achieved per practice.

The Mental Health Practice Readiness Inventory will be completed to document the degree to which a practice has the organizational and individual competencies needed to support integrated BH care. All 32 items reflect the diverse activities (e.g., workflows, financing, service delivery, care coordination) suggested as functions for PCPs (scale: 0 = no function exists; 1 = some function; 2 = function is complete). Thus, total scale scores can range from 0 to 64. Investigators will use scores for each of the 32 competency items (range 0-2) and the total score (range 0-64) to describe the overall level of collaborative care competencies achieved per practice.

The Mental Health Practice Readiness Inventory will be completed to document the degree to which a practice has the organizational and individual competencies needed to support integrated BH care. All 32 items reflect the diverse activities (e.g., workflows, financing, service delivery, care coordination) suggested as functions for PCPs (scale: 0 = no function exists; 1 = some function; 2 = function is complete). Thus, total scale scores can range from 0 to 64. Investigators will use scores for each of the 32 competency items (range 0-2) and the total score (range 0-64) to describe the overall level of collaborative care competencies achieved per practice.

To assess the severity of the child's behavioral and emotional problems, the Vanderbilt ADHD Diagnostic Rating Scale will be completed by caregivers. The VADPRS includes 5 symptom severity subscales, each with a varying number of items: (hyperactivity/impulsivity (n=9), inattention (n=9), oppositional behavior (n=7), conduct problems (n=15), and anxiety/depression (n=7). The scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance (impairment) subscale (n=7) used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional behavior (21), conduct problems (45), anxiety/depression (21) ,and performance (24).

Change from baseline in severity of ADHD, ODD, CD, and ANX/DEP symptoms at home and in community at 3 months

To assess the severity of the child's behavioral and emotional problems, the Vanderbilt ADHD Diagnostic Rating Scale will be completed by caregivers. The VADPRS includes 5 symptom severity subscales, each with a varying number of items: (hyperactivity/impulsivity (n=9), inattention (n=9), oppositional behavior (n=7), conduct problems (n=15), and anxiety/depression (n=7). The scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance (impairment) subscale (n=7) used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional behavior (21), conduct problems (45), anxiety/depression (21) ,and performance (24).

Change from baseline in severity of ADHD, ODD, CD, and ANX/DEP symptoms at home and in community at 6 months

To assess the severity of the child's behavioral and emotional problems, the Vanderbilt ADHD Diagnostic Rating Scale will be completed by caregivers. The VADPRS includes 5 symptom severity subscales, each with a varying number of items: (hyperactivity/impulsivity (n=9), inattention (n=9), oppositional behavior (n=7), conduct problems (n=15), and anxiety/depression (n=7). The scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance (impairment) subscale (n=7) used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional behavior (21), conduct problems (45), anxiety/depression (21) ,and performance (24).

Change from baseline in severity of ADHD, ODD, CD, and ANX/DEP symptoms at home and in community at 12 months

To assess the severity of the child's behavioral and emotional problems, the Vanderbilt ADHD Diagnostic Rating Scale will be completed by caregivers. The VADPRS includes 5 symptom severity subscales, each with a varying number of items: (hyperactivity/impulsivity (n=9), inattention (n=9), oppositional behavior (n=7), conduct problems (n=15), and anxiety/depression (n=7). The scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance (impairment) subscale (n=7) used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional behavior (21), conduct problems (45), anxiety/depression (21) ,and performance (24).

Paralleling the VADPRS, the VADTRS will be completed by teachers. This version includes 3 of the same subscales in the parent version -- hyperactivity/impulsivity (n=9), inattention (n=9), and anxiety/depression (n=7) -- but it also includes an aggregated oppositional/conduct scale (n=10 items). Scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance subscale (n=8 items) that is used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional/conduct problems (30), anxiety/depression (21) ,and performance (24). Practices will continue to separately request their own forms during services (e.g., for medication monitoring).

Paralleling the VADPRS, the VADTRS will be completed by teachers. This version includes 3 of the same subscales in the parent version -- hyperactivity/impulsivity (n=9), inattention (n=9), and anxiety/depression (n=7) -- but it also includes an aggregated oppositional/conduct scale (n=10 items). Scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance subscale (n=8 items) that is used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional/conduct problems (30), anxiety/depression (21) ,and performance (24). Practices will continue to separately request their own forms during services (e.g., for medication monitoring).

Paralleling the VADPRS, the VADTRS will be completed by teachers. This version includes 3 of the same subscales in the parent version -- hyperactivity/impulsivity (n=9), inattention (n=9), and anxiety/depression (n=7) -- but it also includes an aggregated oppositional/conduct scale (n=10 items). Scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance subscale (n=8 items) that is used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional/conduct problems (30), anxiety/depression (21) ,and performance (24). Practices will continue to separately request their own forms during services (e.g., for medication monitoring).

Paralleling the VADPRS, the VADTRS will be completed by teachers. This version includes 3 of the same subscales in the parent version -- hyperactivity/impulsivity (n=9), inattention (n=9), and anxiety/depression (n=7) -- but it also includes an aggregated oppositional/conduct scale (n=10 items). Scores for the hyperactivity/impulsivity and inattention subscales are often aggregated to create an overall ADHD scale factor total score. Individual items are rated on a 4-point Likert scale (0 = never; 1 = occasionally; 2 = often; 3 = very often). There is also a performance subscale (n=8 items) that is used to determine whether a child would meet clinical criteria for a given disorder. The range of scores per factor are as follows: hyperactivity/impulsivity (27), inattention (27), oppositional/conduct problems (30), anxiety/depression (21) ,and performance (24). Practices will continue to separately request their own forms during services (e.g., for medication monitoring).

Caregivers will complete the Pediatric Quality of Life (PEDS-QL) to measure health-related quality of life which includes 4 subscales: health/physical status (n=8), emotional health (n=5), school health (n=5), and social health (n=5). All items are rated on a 5-point Likert scale (0 = never a problem; 1 = almost never a problem; 2 = sometimes a problem; 3 = often a problem; 4 = almost always a problem). These subscales have very good reliability and treatment validity and are sensitive to the DOCC EBP.

Caregivers will complete the Pediatric Quality of Life (PEDS-QL) to measure health-related quality of life which includes 4 subscales: health/physical status (n=8), emotional health (n=5), school health (n=5), and social health (n=5). All items are rated on a 5-point Likert scale (0 = never a problem; 1 = almost never a problem; 2 = sometimes a problem; 3 = often a problem; 4 = almost always a problem). These subscales have very good reliability and treatment validity and are sensitive to the DOCC EBP.

Caregivers will complete the Pediatric Quality of Life (PEDS-QL) to measure health-related quality of life which includes 4 subscales: health/physical status (n=8), emotional health (n=5), school health (n=5), and social health (n=5). All items are rated on a 5-point Likert scale (0 = never a problem; 1 = almost never a problem; 2 = sometimes a problem; 3 = often a problem; 4 = almost always a problem). These subscales have very good reliability and treatment validity and are sensitive to the DOCC EBP.

Caregivers will complete the Pediatric Quality of Life (PEDS-QL) to measure health-related quality of life which includes 4 subscales: health/physical status (n=8), emotional health (n=5), school health (n=5), and social health (n=5). All items are rated on a 5-point Likert scale (0 = never a problem; 1 = almost never a problem; 2 = sometimes a problem; 3 = often a problem; 4 = almost always a problem). These subscales have very good reliability and treatment validity and are sensitive to the DOCC EBP.

The investigators propose to recruit 4 types of practice provider participants (PCP = primary care provider, CM = care manager, SL = Senior Leader, PM = practice manager) as well as caregiver participants. Inclusion Criteria: PCP Participants: Employed at one of the twenty-four (24) pediatric primary care practices identified by the PA Medical Home Program at the PA AAP. Identified by the practice as a Primary Care Provider CM Participants: Employed at one of the twenty-four (24) pediatric primary care practices identified by the PA Medical Home Program at the PA AAP. Identified by the practice as a Behavioral Health Resource who delivers and coordinates behavioral health care in the practice, who will function in the study as a care manager. SL Participants: Employed at one of the twenty-four (24) pediatric primary care practices identified by the PA Medical Home Program at the PA AAP. Identified by the practice as the Senior Leader. Have a practice-level leadership role such as Medical Director or a clinical/practice leader Have administrative responsibilities related to patient care and/or the operations/management of the practice PM Participants: Employed at one of the twenty-four (24) pediatric primary care practices identified by the PA Medical Home Program at the PA AAP. Identified by the practice as the Practice Manager or equivalent position Are responsible for day-to-day practice operations, such as personnel management, billing, and compliance with regulations, in the pediatric practice. Caregiver Participants: Have a child age 5-12 years old who exhibits at least a modest level of behavior problems (Caregiver Participants) Are at least 18 years of age (Caregiver participants) Have parental rights for this child (Caregiver participants) Exclusion Criteria: Caregivers Already enrolled in the study as the caregiver to a different child (e.g., sibling) (Caregiver participants)

All requests for study data will follow NIMH's data sharing and data use policies. The final completely de-identified dataset(s) will include demographic and clinical data at baseline, and primary and secondary outcomes for all studies. These analytic datasets may also include derived variables with documentation. Our form datasets will include original case data, a detailed codebook of variable names, value labels, and programming formats and all study documentation including the protocol and manual of procedures. Study investigators/study staff will upload descriptive/raw data to NIMH's National Database for Clinical Trials Related to Mental Health Illness (NDCT) on a semi-annual basis. Data will be released to NDCT after manuscripts reporting primary findings are accepted for publication.

These data will be released to the NDCT soon after each project's "main outcomes" manuscript is accepted for publication. The investigators plan to make these data available indefinitely, with no anticipated time limit.

In addition to public access to the NDCT, data can also be accessed by contacting the principal investigator.

Kolko DJ, Campo JV, Kelleher K, Cheng Y. Improving access to care and clinical outcome for pediatric behavioral problems: a randomized trial of a nurse-administered intervention in primary care. J Dev Behav Pediatr. 2010 Jun;31(5):393-404. doi: 10.1097/DBP.0b013e3181dff307.

Kolko DJ, Campo JV, Kilbourne AM, Kelleher K. Doctor-office collaborative care for pediatric behavioral problems: a preliminary clinical trial. Arch Pediatr Adolesc Med. 2012 Mar;166(3):224-31. doi: 10.1001/archpediatrics.2011.201. Epub 2011 Nov 7.

Kolko DJ, Campo J, Kilbourne AM, Hart J, Sakolsky D, Wisniewski S. Collaborative care outcomes for pediatric behavioral health problems: a cluster randomized trial. Pediatrics. 2014 Apr;133(4):e981-92. doi: 10.1542/peds.2013-2516. Epub 2014 Mar 24.

Kolko DJ, Hart JA, Campo J, Sakolsky D, Rounds J, Wolraich ML, Wisniewski SR. Effects of Collaborative Care for Comorbid Attention Deficit Hyperactivity Disorder Among Children With Behavior Problems in Pediatric Primary Care. Clin Pediatr (Phila). 2020 Jul;59(8):787-800. doi: 10.1177/0009922820920013.

McGuier EA, Kolko DJ, Ramsook KA, Huh AS, Berkout OV, Campo JV. Effects of Primary Care Provider Characteristics on Changes in Behavioral Health Delivery During a Collaborative Care Trial. Acad Pediatr. 2020 Apr;20(3):399-404. doi: 10.1016/j.acap.2019.11.008. Epub 2019 Nov 21.

Yu H, Kolko DJ, Torres E. Collaborative mental health care for pediatric behavior disorders in primary care: Does it reduce mental health care costs? Fam Syst Health. 2017 Mar;35(1):46-57. doi: 10.1037/fsh0000251.

McGuier EA, Kolko DJ, Klem ML, Feldman J, Kinkler G, Diabes MA, Weingart LR, Wolk CB. Team functioning and implementation of innovations in healthcare and human service settings: a systematic review protocol. Syst Rev. 2021 Jun 26;10(1):189. doi: 10.1186/s13643-021-01747-w.

Kolko DJ, McGuier EA, Turchi R, Thompson E, Iyengar S, Smith SN, Hoagwood K, Liebrecht C, Bennett IM, Powell BJ, Kelleher K, Silva M, Kilbourne AM. Care team and practice-level implementation strategies to optimize pediatric collaborative care: study protocol for a cluster-randomized hybrid type III trial. Implement Sci. 2022 Feb 22;17(1):20. doi: 10.1186/s13012-022-01195-7.