Seasonal influenza vaccination reminders for children with high-risk conditions: a registry-based randomized trial.

Abstract

Children with chronic conditions have an increased risk of complications from influenza and have low influenza vaccination rates. To assess the feasibility and effectiveness of using a statewide immunization information system (IIS) for seasonal influenza vaccine reminders from local health departments (LHDs) targeting children with high-risk conditions. A randomized community intervention. The study was conducted in a population of 3618 children aged 24-60 months with a high-risk condition residing in three Michigan counties. Children were identified using a statewide IIS in October 2008. Children were randomized to intervention (reminder) or control (no reminder) groups. Reminders for seasonal influenza vaccination were mailed by LHDs in November 2008. Feasibility of notification (address validity, address deliverability) was assessed (November 2008-February 2009), and frequencies of notification feasibility measures were determined (analyses conducted in 2010). Effectiveness of notification (seasonal influenza vaccine receipt) was assessed using bivariate logistic regression. Among 3618 children with a high-risk condition, 2730 (75.5%) had not received a 2008-2009 influenza vaccination and were eligible at the time of notification. Among children assigned to the reminder group (n=1374), 42.6% had an address determined to be either invalid, undeliverable, or both. Among those with valid addresses (n=2001), a greater percentage of children with deliverable reminders received at least one influenza vaccination (30.8%) during the outcome observation period than did children assigned to no reminder (24.3%, OR=1.39, 95% CI=1.13, 1.72); children with an undeliverable reminder had an influenza vaccination rate (22.8%) similar to children assigned to no reminder. Receipt of a reminder was positively associated with seasonal influenza vaccination. However, more than 40% of children assigned to receive a reminder were determined to have an invalid or undeliverable address, emphasizing the need for increased quality of IIS contact information. This study is registered at www.ClinicalTrials.gov NCT01431183.