An Immunization Information System and Provider Education Partnership: Results of a Pilot Study

Background:
The Physician Peer Education Project on Immunization (PPEPI) has provided immunization education sessions since 1996. Evaluations have shown statistically significant changes in provider knowledge and behaviors, both directly after the lecture and at six months. The project evaluation has not, however, included changes in immunization rates as an outcome variable.

Objectives:
1. To explore the feasibility of targeting providers for peer education;
2. To develop alternative uses of the immunization registry
a. identify practices in need of assistance;
b. assess the efficacy of education;
c. pair immunization rate reports with education

Methods:
Primary care providers with fourth dose DTaP rates lower than 75% were invited to participate in the study. Those who agreed to participate signed an informed consent document allowing evaluators to review their practice's DTaP#4 immunization rates in MCIR, received a $75 honorarium, and were randomized to intervention or control groups. Practices assigned to the intervention group were required to attend a one-time lecture. Investigators collected 4th dose DTaP rates in MCIR at baseline and 3, 6, 12, 18, and 24 months.

Results:
540 eligible providers were identified and invited to participate; non-responders were additionally contacted by telephone and fax. Sixty-four providers enrolled in the study (12% participation rate). Eight of the 33 practices (24%) in the intervention group didn't attend the education session. There was an attrition rate of 16% over the course of the study. The evaluation was underpowered to identify an intervention effect, however, fourth dose DTaP rates increased in both groups over the course of the study.

Conclusions:
Partnering PPEPI with MCIR data enabled us to target interventions to those practices most in need, and provide timely, accurate immunization rate assessments as an important follow-up to the intervention session. Additional variables will be added to the model and presented at the conference.