Physicians' Intention to Vaccinate: An Analysis of Pre- and Post-Licensure Survey Data

Background:From 2005-2006, new vaccines were licensed for herpes zoster (HZ), rotavirus (RV), and human papillomavirus (HPV). Pre- and post-vaccine licensure surveys about physician intentions and subsequent practices related to these vaccines were conducted.

Objectives:To determine: 1) if physicians’ intentions to vaccinate correlated with their practices after vaccine was available; 2) whether perceived barriers to vaccine delivery changed from pre- to post-vaccine licensure; and 3) if physicians responses from pre- to post-licensure shifted towards more/less strongly recommending the vaccine and perceiving a barrier as more/less significant.

Methods:Nationally representative networks of primary care pediatric (Peds), family medicine (FM), and general internal medicine (GIM) physicians were surveyed pre- and post- vaccine licensure by mail or internet.  Kendall’s tau and Bhapkar’s test were used to examine correlations and response shifts respectively.

Results:Responses for 284 physicians were analyzed for HZ (FM/GIM combined); 184 for RV (Peds only); and 152 for HPV (Peds only).  For HZ, RV, and HPV there was a positive and marginally significant correlation between physicians’ pre/post-licensure recommendations [τ range=.10-.37, (p value range= .06-.10)].  When recommendations changed for HZ, RV, and HPV for the 11-18 age group physicians generally shifted towards more strongly recommending the vaccine after licensure (p<.0001 for all).  However, HPV vaccine recommendations for 9-10 year olds shifted towards not recommending the vaccine (p<.001).  Thirteen barriers were analyzed across all three vaccines.  Generally, physicians’ perceptions of a given barrier pre- and post-licensure were positively correlated and marginally significant, [τ range=.10-.46, (p value range= .05-.08)].  When perceptions of barriers changed from the pre- to post-licensure, the shift was generally towards reporting that a given barrier was less important (p<.05 for all).

Conclusions: Surveys are useful tools to predict whether physicians will adopt new vaccines. Physicians tend to over-estimate barriers to implementation of new vaccines pre-licensure.