Reducing Diagnostic Errors in Primary Care Pediatrics (Project RedDE)

Reducing Diagnostic Errors in Primary Care Pediatrics

The proposal will focus on 3 specific, high-risk, pediatric ambulatory diagnostic errors each representing a unique dimension of diagnostic assessment: evaluation of symptoms, evaluation of signs and follow-up of diagnostic tests. Adolescent depression (i.e. symptoms) affects nearly 10% of teenagers, is misdiagnosed in almost 75% of adolescents and causes significant morbidity. Pediatric elevated blood pressure (signs) is misdiagnosed in 74-87% of patients, often due to inaccurate application of blood pressure parameters that change based on age, gender and height. Actionable pediatric laboratory values (diagnostic tests) are potentially delayed up to 26% of the time in preliminary investigations and 7-65% in adults, leading to harm and malpractice claims.

The investigators propose to conduct a multisite, prospective, stepped wedge cluster randomized trial testing a quality improvement collaborative (QIC) intervention within the American Academy of Pediatrics' Quality Improvement Innovation Networks (QuIIN) to reduce the incidence of pediatric primary care diagnostic errors. QuIIN is a national network of over 300 primary care practices, ranging from tertiary care academic medical centers to single practitioner private practices, interested in and experienced with QICs. Because many processes are likely to be common across diagnostic errors in outpatient settings, a multifaceted intervention, such as a QIC, has a high likelihood of success and broad applicability across populations. Preparatory inquiries to QuIIN primary care providers suggest high interest in reducing these 3 diagnostic errors and provider agreement with randomization to evaluate diagnostic error interventions. Practices will be randomized to one of three groups, with each group collecting retrospective baseline data on one error above, and then intervening to reduce that error during the first eight months. Each group will concurrently collect control data on an error they are not intervening on during those eight months. Following those eight months, the groups will continue intervening on their first error, begin intervening on the error they were a control site for, and begin collecting data on the third error for which they will be a control site for. Finally, in the final eight months, all groups will intervene on all three errors. A second wave of practices will be recruited to join the groups after eight months and will only intervene on two of the three errors.

Objectives:

Primary

• To determine whether a QIC consisting of evidence-based best-practice methodologies, mini-root cause analyses, data sharing, and behavior change techniques, is associated with a reduction in 3 specific diagnostic error rates in a national group of pediatric primary care practices.

• Hypothesis 1: Implementation of a QIC will lead to a 40% reduction in missed diagnosis of adolescent depression.
• Hypothesis 2: Implementation of a QIC will lead to a 30% reduction in missed diagnosis of pediatric elevated blood pressure.
• Hypothesis 3: Implementation of a QIC will lead to a 45% reduction in delayed diagnosis of actionable laboratory results.

Secondary

• To determine if a QIC's effect changes for wave 1 versus wave 2 participants, or for the second versus the first error a practice intervenes on.
• To further investigate the epidemiology of three ambulatory pediatric diagnostic errors: missed diagnosis of adolescent depression, missed diagnosis of pediatric elevated blood pressure, and delayed diagnosis of actionable laboratory results.
• To evaluate patient outcomes related to these diagnoses including outcomes after positive depression screening, missed elevated blood pressure screening and delayed diagnosis of actionable laboratory values.

Inclusion Criteria:

• The investigators will include 30 primary care pediatric practices that are part of the American Academy of Pediatrics' QuIIN (Quality Improvement Innovation Networks) organization. The second wave will recruit 15 additional practices.
• Practices must have sufficient volumes of adolescent well child visits (17 per month) and all well child visits (30 per month), and be able to query their EHR systems in order to be included in the study