HPV Vaccine Hesitancy

Findings From a Statewide Survey of Health Care Providers

Annie-Laurie McRee, DrPH; Melissa B. Gilkey, PhD; Amanda F. Dempsey, MD, PhD, MPH


J Pediatr Health Care. 2014;28(6):541-549. 

In This Article


We surveyed a statewide sample of physicians and nurse practitioners who provide preventive care to preteens and adolescents ages 11 to 17 years. We identified potential participants through publicly available lists provided by the Minnesota Boards of Medical Practice and of Nursing. From these lists, we sampled health care providers in pediatric and family medicine specialties. Because our sampling frame included many providers who potentially did not meet study criteria (e.g., neonatal specialists), the survey used a screener question to limit respondents to those who provided preventive care to patients ages 11 to 17 years.

In April 2013, we invited 3,923 health care providers to participate in the study. A total of 615 respondents provided informed consent and took our cross-sectional, online survey (adjusted response rate based on American Association for Public Opinion Research [AAPOR] formula 4: 28%; AAPOR, 2009). Among those who were confirmed to be eligible based on survey screening questions, the cooperation rate was 86% (i.e., the proportion of eligible providers contacted who completed the survey; AAPOR, 2009). Study data were collected and managed using REDCap (Research Electronic Data Capture) tools hosted at the University of Minnesota (Harris et al., 2009). Respondents received a survey invitation describing the study's purpose and providing a link to the survey; nonrespondents received up to three reminder e-mail messages. No incentives were provided for participation. The Institutional Review Board at the University of Minnesota approved the study.


We developed items based on the literature and our previous research with parents (Dempsey, Abraham, Dalton, & Ruffin, 2009; Gilkey et al., 2013; Hughes et al., 2011; Kempe et al., 2011; Opel et al., 2012; Vadaparampil et al., 2011). We conducted cognitive interviews with five health care providers to identify potential sources of response error and ensure the clarity of new survey items (Willis, 2005). We used findings from cognitive interviews to refine and improve survey items before fielding the full survey. Questions specified the 11- to 12-year-old age group to align with HPV vaccination guidelines from the Advisory Committee on Immunization Practices (CDC, 2010; Dunne et al., 2011; Markowitz et al., 2007).

The survey assessed providers' HPV vaccine recommendation practices with three parallel items asked separately about girls and boys. The first question asked how often providers recommended HPV vaccine as a part of routine care for 11- to 12-year-olds. We categorized responses for this analysis as "most of the time" (defined as "more than 75% of the time") versus less often to reflect Advisory Committee on Immunization Practices guidelines for routine administration (CDC, 2010; Dunne et al., 2011; Markowitz et al., 2007) and to address the skewed nature of the distributions. Based on our formative qualitative research, the survey also included two statements assessing the strength of providers' HPV vaccine recommendations.

After text instructing respondents to think about the times during the past year they recommended HPV vaccine to parents of 11- to 12-year-old patients, providers indicated how often they encountered different types of parental reactions (five items) and rated how much potential concerns contributed to parents' decisions to delay or refuse HPV vaccine for their child (12 items). The survey then assessed providers' responses to parental HPV vaccine hesitancy (six items), self-efficacy (i.e., confidence in their ability to address parents' concerns; four items), outcome expectations (i.e., expectations that their discussions with parents will lead to vaccination; three items; Bandura, 1986), and perceptions of potential strategies and tools for counseling HPV-vaccine hesitant parents (eight items).

Respondents also provided information about sociodemographic and practice characteristics, including age, sex, training specialty, year finished clinical training, practice type and location, and proportion of patients who are 11 to 17 years old.


Of the 615 total respondents, the analytic sample for the present analysis includes 575 respondents who answered questions about HPV vaccine recommendations and parental hesitancy. We used χ2 analyses to test for differences in recommendation practices by provider type (i.e., pediatrician, family medicine physician, or nurse practitioner). We assessed associations between routinely recommending HPV vaccine and providers' perceptions of parents' reactions, responses to hesitancy, self-efficacy to address parental concerns, and outcomes expectations using logistic regression models controlling for provider type. We conducted all analyses in Stata SE version 10.0 (Statacorp, College Station, TX) using two-tailed tests and a critical alpha of 0.05.