Assisted Identification and Navigation of Early Mental Health Symptoms in Children
Assisted Identification and Navigation of Early Mental Health Symptoms in Children
The randomized, two-arm pragmatic trial will test the effectiveness of offering 6-months of telephonic support from a mental health (MH) navigator to promote early access, engagement, coordination, and personalization of mental health treatment and services for children naïve to such treatments and services, and who are identified as being at risk for behavioral health concerns.
The model includes: (a) automated identification of early symptoms for children meeting criteria for behavioral health problems using a previously developed Natural Language Processing (NLP) program and predictive algorithm; (b) standardized instruments for assessment and diagnosis of mental health disorders (c) 30 minute assessment appointments with a study psychologist (d) creation of an Epic "reporting workbench" and Epic "smart form" to facilitate the outreach, monitoring and follow-up of families/children by the MH navigator; (e) use of MH Navigators (e.g., clinical social workers) to conduct family outreach, and coordination with and between clinicians; and (f) the offer of one to four clinic-to-home videoconferencing brief therapy sessions to bridge families/children unwilling or unable to access in-person MH services.
The specific aim of the pragmatic trial is to:
Test a MH navigator model to promote early access, engagement, coordination, and personalization of mental health treatment and services as soon as early symptoms of mental health problems are detected in children.
The navigator model and implementation to be tested include:
This study will employ an encouragement trial design, an alternative to a traditional RCT design. Guardians of randomized youth are offered the opportunity to receive additional services (navigation, FAST brief therapy sessions), but allowed to choose whether to receive the services. That is, acceptance of or adherence to the offered study services is not a condition of participation. Whether or not guardians choose to accept intervention services from the beginning is a measure of the effectiveness of the program. Guardians who decline the additional services (navigation, FAST sessions) must remain in the intervention arm in order to measure the effectiveness of the intervention without bias. This allows an intent-to-treat comparison between arms. That is, patients remain enrolled in the study regardless of the navigators' ability to contact the family by phone or the family's perceived need for or willingness to accept or participate in study-based services.
Investigators will conduct a two-group, randomized and pragmatic encouragement trial to compare outcomes between study arms (intervention; usual care control) in 2 health care systems. The follow up period for the trial will be 6-months. A parallel arm design is used to eliminate response bias stemming from the nature of the study.
The MH navigator will review the medical record for each patient automatically identified for the study. This review will occur prior to any arm assignment. Those found to have exclusions will be marked ineligible. Patients deemed eligible will then be randomly assigned to one of two study arms.
Eligibility review will be completed no earlier than 14 calendar days after the index event. This is to allow for any follow-up after the clinic visit (e.g., patient initiates MH specialty treatment). The review will occur no later than 21 days after the index event.
Study staff will use a predetermined randomization schedule. The lead biostatistician at KP Washington will generate the master assignment sequence before the first patient is randomized. Randomization allocation tables will be created by the lead biostatistician, who will oversee randomization at each site. Randomization will be stratified by age (4-6, 7-9, 10-11/12 (depending on site, KPNC 10-11. KPWA 10-12)) and sex.
Only subjects assigned to the intervention will be contacted by the navigator and offered the intervention. Outcomes for all subjects entering the intervention arm, regardless of whether the family accepted the MH navigation program, will be attributed to the intervention. That is, outcomes for subjects will be assigned on an intent-to-treat basis regardless of the extent to which they engage with the study clinician or the study navigator. Cross over prevention is therefore functionally not-applicable due to the study design and the analytic plan.
Each intervention arm subject's primary care provider will be sent an introductory staff message by the study navigator prior to contacting the patient. The message will invite the provider to respond within 3 days if they do not wish their patient to be outreached by the study for any reason.
Families contacted by the navigator may decline mental health navigation at the onset or at any time during the follow up period. Acceptance will be tracked.
Families who agree to navigation will first be offered a formal assessment in order to ascertain the primary mental health diagnosis and any co-occurring mental health disorders (e.g., substance use). After completing the online parent/guardian self-assessment, parents/guardians will meet for 30 minutes with a mental health clinician to review their answers, discuss diagnoses, and refer the family back to the navigator for connection to care or study-provided tele-video visits.
Families offered tele-video therapy visits may decline it at any time. Acceptance and adherence will be tracked as outcomes for those to whom it is offered.
The study clinician may withdraw an offer of tele-video visits if concerns regarding privacy or safety of the child/family arise with regard to the treatment modality. If this occurs, the family/child will be navigated to sessions with an in-person therapist.
Control Arm:
Data will be collected at baseline and 6-month follow-up from the standardized instruments and automated sources for all eligible subjects randomized to the control arm. Safety monitoring data will be collected from automated data 3 times per year for NIMH DSMB reports.
Families who agree to participate will first be offered a formal assessment in order to ascertain the primary mental health diagnosis and any co-occurring mental health disorders (e.g., substance use). After completing the online parent/guardian self-assessment, parents/guardians will meet for 30 minutes with a mental health clinician to review their answers, discuss diagnoses, and refer the family back to their primary care provider.
Study clinicians will document clinically relevant information from their assessment in a telephone encounter and route these to the child's KP primary care provider in Epic as well as the site navigator.
Study telephone encounters are not billable visits. Only clinically relevant information is incorporated into telephone visit notes.
Control arm participants do not receive navigation.
Intervention Arm:
Six months of follow up data will be collected from the standardized instruments and automated sources for all eligible subjects randomized to the intervention arm. Safety monitoring data will be collected from automated data 3 times per year for NIMH DSMB reports.
An Epic reporting workbench and smart form will support the MH navigator's in managing their active panels. Navigators will be prompted to follow up with subjects at the appropriate time points, log outreach attempts and outcomes, note subjects' acceptance or refusal of navigation, and (if offered) tele-video therapy visits, and track subjects ongoing acceptance of the intervention.
The MH navigator will contact the patient's primary care/pediatrics provider via Epic staff message within 2 business days of a new intervention subject appearing on their reporting workbench.
Navigators will use a standardized message (Epic Smartphrase). The message will (a) introduce the MH navigator to the provider, (b) briefly describe the extra support the study can offer, and (c) inform the provider that s/he may reply to the message to decline navigation for their patient for any reason.
Navigators will enter in their Epic smart form:
Providers are allowed 3 business days to actively decline the offer of study support before the navigator outreaches the family.
If a provider declines navigation on their patient's behalf the patient will not be contacted. The subject remains in the study; 6-months of automated data will be collected for the intent to treat analysis. The navigator will enter onto the smart form:
o Provider Agreed MHN = Declined
If a provider accepts or does not respond in the 3-day timeframe, subject outreach will begin. The navigator will use the Patient Care Notes field as appropriate and set the smart form to:
MH navigators at each site will outreach intervention arm families by phone to offer up the navigation program.
Up to 3 cycles of outreach, 2 weeks apart, will be made. For each of the 3 cycles of outreach, the Navigator will make up to 3 attempts to reach the family with in a 1-week period, with up to two messages left.
If not reached after 3 cycles of outreach are completed, the smart form will be coded:
o Family Agreed MHN = No Response
If reached and family agrees, the smart form will be coded:
o Family Agreed MHN = Agreed
If reached and family declines, the smart form will be coded:
o Family Agreed MHN = Declined. The navigator will notify the provider to inform him/her that the program was declined; the study will not outreach the family further.
Families who agree may choose to discontinue navigation at any time during the 6-month follow up period. If a family agrees and then opts to end early, the smart form will be coded:
o MHN Ended Early = Date ended will be entered. The reason ended early will also be captured.
Families/patients may report that the index MH symptoms have resolved. These patients/families will be offered a 'watchful waiting' approach. Subjects accepting 'watchful waiting' will be outreached by phone at least quarterly but not past 6-months post-randomization.
Families who agree to participate will first be offered a formal assessment in order to ascertain the primary mental health diagnosis and any co-occurring mental health disorders (e.g., substance use). After completing the online parent/guardian self-assessment, parents/guardians will meet for 30 minutes with a mental health clinician to review their answers, discuss diagnoses, and refer the family back to the navigator for connection to care or study-provided tele-video visits.
Study clinicians will document clinically relevant information from their assessment in a telephone encounter and route these to the child's KP primary care provider in Epic as well as the site navigator. Study telephone encounters are not billable visits. Only clinically relevant information is incorporated into telephone visit notes.
Navigators will document clinically relevant discussion (post assessment) in a telephone encounter and route these to the child's KP primary care provider in Epic. Study telephone encounters are not billable visits. Study-related (administrative) notes remain in the smart form; only clinically relevant information is incorporated into telephone visit notes.
Navigators will outreach families in the program by phone at least monthly during the 6-month study period. The navigator will inquire about the child's status and re-evaluate barriers to access on follow up navigation calls. They will use motivational interviewing techniques (e.g., reflective listening, asking open ended questions) with the child and/or parents (as age appropriate) to encourage them to engage in psychosocial treatments, attempt to motivate them to start treatment (as needed), and identify and help problem solve ways to minimize barriers to starting therapy. When barriers exist, the navigator will offer assistance in facilitating specialty treatment initiation and assess appropriateness of tele-video therapy sessions until ongoing MH services can be arranged.
Families/patients reporting that index symptoms have resolved will be offered a 'watchful waiting' approach. Subjects accepting 'watchful waiting' will be outreached by phone at least quarterly, but not beyond the end of the 6-month follow-up period.
Inclusions
Patient is ≥ 4 and less than 7 months from the mental health emancipation age in the respective State on the date of randomization. That age is 13 years of age in WA and 12 years of age in CA. Therefore, we will include these subjects:
Treated in the internal delivery system and enrolled in a Kaiser Permanente insurance plan for at least 3 months prior to the PC note or MH diagnosis (index event);
"Mental Health treatment naïve" per KP electronic records, defined as:
AND at least one of the following at the index encounter per KP records:
Exclusions
Past ICD-10 diagnosis for a mental health condition associated with any encounter type or prescription order, internal or external to the system);
Past mental health prescription medication ordered (including patient reported medication use, inpatient administration, past order for proposed off-label purpose, etc.);
Documentation of mental health issues in the encounter note describes someone other than the child (e.g., a sibling or parent's issue)
Note is from specialty, emergency or urgent care (i.e., not primary care or pediatrics).
Subject is flagged as "interpreter needed" in Epic.
o Subject having a sibling from the same index encounter for mental health complaint
Describe how individuals will be screened for eligibility.
We will identify potential participants automatically using diagnostic information and Natural Language Processing (NLP) of signed clinical notes from electronic health records (EHRs).
All potentially eligible subjects will be automatically added to an Epic reporting workbench (patient list) once per week. Study staff will confirm the eligibility of each listed patient via chart review. All participants confirmed eligible will be randomly assigned to a study arm (intervention, usual care control). Outcomes data will be collected using the standardized instruments and from the electronic records of all randomized participants for study analysis.
The MH navigator will review the medical record for each patient automatically identified for the study. This review will occur prior to any arm assignment. Those found to have exclusions will be marked ineligible. Participants deemed eligible will then be randomly assigned to one of two study arms.
Eligibility review will occur no later than 10 days following the index event.