Implementation Research posted on January 27, 2010 09:52 Tracy M. King, Samar Muzaffar, Maureen George.. Johns Hopkins University, Baltimore, MD; Boston Medical Center, Boston, MA. BACKGROUND: Current recommendations call for periodic developmental screening of all children and prompt referral of children who fail such screening. Providers' willingness to adopt these recommendations, however, may be influenced by prior experiences, particularly in high-risk urban environments where the prevalence of delays is higher, and challenges in completing referrals are greater, than in other settings. OBJECTIVE: To describe providers' experiences with developmental screening and referral in a high-risk urban community. DESIGN/METHODS: Nine providers from seven clinics participated in semi-structured interviews addressing their experiences with developmental screening and referral. Interviews were recorded, transcribed, and analyzed by 2 independent readers to identify major themes. RESULTS: All providers reported "screening" children's development. Screening techniques varied, however, from parent-completed questionnaires to milestone checklists to clinician "gestalt". Providers using a standardized instrument often made adaptations or modifications to the instrument during use, such as omitting items (to reduce the time required for administration) or changing the interpretation of results (because parents' reports of their children's abilities were often perceived as inaccurate). Referral practices were also variable. Although most providers referred young children to the state early intervention program, some used a tiered approach, referring children with severe delays only to physician subspecialists, while others reevaluated children in the primary care setting prior to making referrals. Many providers felt that parents often failed to share providers' concerns about possible delays. No provider had an effective system for tracking referrals, other than asking parents whether their child had been evaluated. Most providers expressed frustration at the lack of feedback from early intervention programs regarding patients they had referred. CONCLUSIONS: In this urban, high-risk setting, providers reported many developmental screening and referral practices that depart from current standards of care. These findings highlight the need not only for intensive provider education but also for systems changes, within practices and among early intervention programs, to improve the effectiveness of developmental screening and referral. Note from Frances Page Glascoe (PEDS author): I had a complete "aha" reaction to this study, surmising that clinicians think they are screening when they use informal approache such as key items from the Denver-II or other measures, or non-validated questions to parents. This study really confirms the need to dispel that notion (and thus a good line of research inquiry would be to have providers compare their referral rates using informal approaches to rates when quality measures are deployed-such as those by Dr. Schonwald). Providers don't seem to get much feedback when they aren't doing a great job (e.g., parents leave the practice and don't say why). So establishing feedback mechanisms, self-study, quality improvement, and monitoring detection rates is essential!