How Well Do Practices Incorporate QI Procedures for Vaccine Delivery? A Study From the South Carolina Pediatric Practice Research Network

Background: Vaccine delivery is a primary marker of quality in pediatric practices. Prompting providers using the quality improvement (QI) approach to immunize during sick or acute care visits is one intervention that may improve delivery. To be effective and sustainable, QI procedures should be tailored to the individual characteristics of the practice.

Objectives: Determine the extent to which pediatric practices in the South Carolina pediatric practice research network (SCPPRN) instituted QI procedures to improve vaccine rates of 19-35 month olds.

Methods: A QI training session was held at the beginning of a project aimed at improving vaccine delivery.  This QI session emphasized the use of the Plan/Do/Study/Act (PDSA) cycle in order to understand the issue, develop an intervention strategy, evaluate its effects, and act on the findings. Baseline data were obtained on children aged 18-30 months including all immunization data and visit data beginning at 6 months of age.  Each practice developed and adopted initial interventions (a prompt for giving immunizations at sick visits) based on their current procedures and then conducted self evaluations, by chart audits, to determine if their initial prompting intervention was effective.  Where appropriate, revisions of their prompting intervention were made.

Results: Six of the 9 SCPPRN practices participated in the 1 year study. Pre-intervention immunization rates for each SCPPRN practice were 78%, 68%, 62%, 57%, 41% and 54%.  There were varying levels of QI participation in each practice, with all of the practices developing initial interventions. Two practices incorporated the initial prompt and, based on self-assessments as part of the PDSA cycle, determined that a revision of the prompt was needed. These two practices also had the most involvement of the entire office and nursing staff. Subsequent audits demonstrated proper use of the prompt.  Three other practices maintained their primary prompt with some minor revisions throughout the study. The sixth practice intended to use a specially created alert in their electronic medical record (EMR); however, significant technical and personnel barriers prevented the production of the alert for their system. Other QI steps were attempted at this practice, but not well incorporated.

Conclusions: Adoption of QI procedures is variable in clinical practice.  Staff involvement appeared to be at least as effective as clinician involvement. Contrary to initial plans, the EMR appeared to have been a barrier to implementing QI in one practice.