Early Identification of Developmental and Social Problems: Screening in Pediatric Primary Care Settings

Disease prevention and health promotion rely on early identification of risk factors or early markers of disease process.  In the domain of early social and behavioral development, pediatric primary care settings provide a unique opportunity to find those individuals who might benefit from interventions that would promote long-term health and adaptive development.  Large scale screening presents many challenges, foremost among them choice of instruments, sensitivity and specificity of screens, and procedures for addressing children who screen positive.

Our screening effort was focused on children in the first 5 years of life, and conducted in two primary care settings serving families on public assistance.  We used 3 instruments with some demonstrated validity: the Ages and Stages Questionnaire Version 3 (ASQ3), the Ages and Stages Questionnaire Social Emotional (ASQ-SE), and the Early Childhood Screening Assessment (ECSA).  The screening efforts were staffed by a combination of behavioral health consultants embedded in the pediatric primary care settings along with in-house staff at the settings.  Screening was attempted during well child visits.

For screening of this type to be effective, it should be universal (or as close to universal as possible).  In the current project, we achieved a screening completion rate of 75% (when appointment were kept by families), resulting in 3,238 screens in a year.  This completion rate was from a base of 6,276 scheduled appointments, where approximately 1/3 of those appointments were not kept by the family.  The remaining screens that were not completed primarily owing to staffing issues, lack of time, and parent refusal.

Screens are most efficient when they are both sensitive and specific.  When screens were completed in this project, 944 (58%) were positive for some problem on one of the instruments (either by quantitative cutoff or explicit expression of parental concern).  This compares with estimates that approximately 1/3 of children developing in poverty circumstances within this age range have frank developmental problems.  This apparent lack of specificity is consistent with triage follow-up assessments conducted a part of this project.  We are currently processing data on children who screened positive regarding their history of service utilization.

These results (and others) point to pragmatic challenges in implementing screening, identifying which children should be followed more intensively based on screening results, and development of resources to address these developmental and behavioral concerns in otherwise overburdened pediatric primary care settings.