Center for Promotion of Child Development through Primary Care

The Child Health and Development Interactive System (CHADIS)

Created by The Center for Promotion of Child Development through Primary Care

www.childhealthcare.org

Baltimore, Maryland

Raymond Sturner, MD, Barbara Howard, MD Tanya Morrel PhD, Kathline Albus PhD, Christopher Schultz, BS, Luke Mejewski, BS, Stephanie Porcaro, BA

Thanks to support from NICHD SBIR Phase II grant 5R44HD038599-03

Overview

The Child Health and Development Interactive System (CHADIS) is a web based clinical assessment and decision support tool to facilitate current guidelines for comprehensive child health care, including mental health, development and health risks. It was developed and tested partly under SBIR funding.

How It Works

CHADIS comprises:

  • Pre-visit web based computerized parent interview following algorithms for:
    • New guidelines for child mental health diagnoses (DSM-4 and DSM-PC for primary care)
    • Required health risk information e.g. Early & Periodic Screening Diagnosis & Treatment data
  • Data is computer analyzed and instantly presented on an electronic worksheet for the clinician in the form of:
    • Presumptive diagnoses
    • Risk factors
  • Presumptive diagnoses and risk factors are linked to:
    • Decision support/education at the point of care: definitions, key points, differential diagnosis and comorbidity, normal development and anticipatory guidance, clinical guide algorithms, management suggestions, other tools, citations
    • Resources specific to the problem or to promote strengths in child/family- parent text for handouts books, videotapes, games support groups providers e.g mental health counselors searchable by insurance accepted, zipcode, problems seen, languages spoken agencies e.g. substance abuse, legal aid mentors, coaches, tutors.

Preliminary Data

Preliminary data from sub-samples of > 800 patients in several pediatric practices support:

Validity for making diagnoses

Presumptive diagnoses of a child psychiatric disorder by CHADIS correctly classified children (N = 71 of a group of referred and non-referred children from the same pediatric practice) compared to a gold standard criterion for child mental health disorder (Diagnostic Interview of Children and Adolescents or DICA) in 91% of cases and was more predictive of this outcome than an extensively used and studied comparison measure (Child Behavior Checklist or CBCL) that has other limitations including not identifying the new DSM-PC diagnoses.
  • Valid in predicting the correct level of mental health diagnosis
  • Parents give similar responses over 2-week interval (reliability)
  • "Problem" level diagnoses lead to disorders one year later without intervention
Sensitivity Specificity % Agreement
CBCL-D .77 .88 83
CHADIS-D .84 .97 91
CHADIS-P&D .94 .74 83

Validity and prevalence of the new mental health categories

Validity of the new DSM-PC through CHADIS by comparison to independent concurrent measures (CBCL, DICA disorder) of behavior problem and impact on families, (impact items of Strengths and Difficulties Questionnaire). CHADIS-DSM-PC identified a disorder in 27% of the children; problem in 28%; variation in 21% and no diagnosis in 23% . The rates are similar to other surveys of disorder (e. g., Shaffer, et. al., 1996) and measures of parental concern (e. g., Sturner, et. al., 1980).

Predictive study showing that the new at-risk category of disorder sub-threshold diagnoses of "Problem" was a precursor to a disorder diagnosis in 39% of children (compared to zero disorders in the no diagnosis category) and 71% of children in the problem category had a disorder or problem one year later in cases without CHADIS interventions.

T–1 T-2 Disorder T-2 Disorder or Problem
No Dx (N=17) 0 12%
Variation (N=15) 20% 40%
Problem (N=38) 39% 71%
Disorder (N=35) 68% 84%

Acceptability to families

  • 9/10 parents in focus group would prefer to see a doctor using CHADIS previsit
  • 89% of parents who used CHADIS saw value & would like to use it again at a subsequent visit even without doctor feedback (N=102)

Desirability for physicians

  • 86% MDs viewing demo wanted to use it clinically (N=86)
  • Hopkins Community Physicians have been using it at 2 sites and now want to use it at all sites

Time efficiency characteristics

  • Top parent concern takes average of 7 minutes
  • Range to complete top 5 areas of concern 10-40 minutes
  • Parents did not judge this as too long

Current Usage

  • 6 sites
  • > 800 patients assessed

Innovation

  • New rules-based engine that chooses questions to ask caregiver based on attributes already collected and rules for diagnoses
    • Trained clinical users rather than programmers will easily be able to enter and modify questionnairesInterfaces with clinician worksheet
  • First application allowing widespread use of DSM-PC categories
  • First data on the new DSM-PC problem and variation diagnoses
  • First data on course of Problem diagnosis over time
  • Unique integration of parent questionnaires results with linked decision support and resource access
  • New comprehensive database for common childhood problems constitutes primary care research network

Features Under Development

  • Questionnaires for parents of 0-3 year olds implementing Diagnostic Classification 0-3R in collaboration with Zero to Three, Inc.
  • Questionnaires for teens based on PHQA and CHAMPS in collaboration with Columbia University Child Psychiatry and Adolescent Medicine Departments
  • CHADIS Brief version comprising Pediatric Symptom Checklist plus EPSDT health risks
  • Family background questionnaires including Maternal Depression
  • Developmental screen for 0-5 either Ages & Stages Questionnaire or Infant Behavior Inventory from Child Development Inventory plus M-CHAT for autism
  • School/Parent/Child web site
  • Health Passport and Memory Book print outs for families
  • Patient Portal for each family to view information and use resource search engine

Availability

CHADIS is now available for distribution free standing or linked into an inexpensive electronic medical record system.

Feedback

American Academy of Pediatrics is reviewing it at its Board meeting 8/11/05 for endorsement or partnering.

Comments from leaders in the field:

  • "CHADIS… is the best example of the what future primary care pediatrics should look like" Neal Halfon, MD, MPH, Professor of Pediatrics, UCLA in address entitled: "New and Emerging Models for Pediatric Primary Care" in Symposium titled "The Future of Primary Care Pediatrics" at the Pediatrics Academic Societies Meeting May, 17, 2005
  • "(CHADIS)..is exactly what pediatricians should be doing" Donald Cook, MD Past President of the Academy of Pediatrics at American Academy of Pediatrics Annual Meeting
  • "Every pediatrician in the country should be using CHADIS" Johns Forbes, COO, American Academy of Pediatrics, 7/13/2005
  • "This (CHADIS)..could take pediatrics into the 21st Century" Kelly Kelleher, MD, MPH, member of American Academy of Pediatrics Mental Health Task Force meeting May, 2005

Sturner, R., Howard, B. J., & Morrel, T. (2003). Preliminary Validation of the DSM-PC. Presented at the annual meeting of the Society of Developmental-Behavioral Pediatrics. JDBP, 393-394.

Sturner, R., Howard, B. J., Morrel, T., & Rogers-Senuta, K. (2003). Validation of a Computerized Parent Questionnaire for Identifying Child Mental Health Disorders and Implementing DSM-PC. PAS Meeting.

Sturner, R., Morrel, T., & Howard, B. J. (2004). Mental Health Diagnoses among Children being seen for Child Health Supervision Visits: Typical Practice and DSM-PC Diagnoses. PAS Meeting.

Sturner, R., Morrel, T., & Howard, B. J. (2005). DSM-PC Diagnoses Are Not "Outgrown" in One Year: Typical Practice and DSM-PC Diagnoses. PAS Meeting.

Sturner, R.A., Ferholt, J., Klatskin, E., & Granger, R. (1980). Mental Health Aspects of Routine Pediatric Examinations, Clinical Pediatrics, 251-260.


Sturner, R., R. Howard, B. J., & Morrel, Albus, K, Schultz, C, Majewski, L, and Porcaro, S. The Child Health and Development Interactive System (CHADIS) presented at the 7th Annual NIG SBIR/STTR Conference, 2005.