Validation of a measure of knowledge about human papillomavirus (HPV) using item response theory and classical test theory

Jo Waller,Remo Ostini,Laura A.V Marlow,Kirsten Mccaffery,Gregory Zimet

doi:10.1016/j.ypmed.2012.10.028

Abstract

ObjectivePublic understanding of HPV is important to ensure informed participation in cervical cancer prevention programmes. While many studies have measured HPV knowledge, none has developed a validated measure for use across countries. We aimed to develop and validate such a measure. MethodItems tapping knowledge of HPV, HPV testing and HPV vaccination were developed from previous literature and with expert consultation. The 29-item measure was administered via the internet to 2409 adults in the UK, US and Australia in 2011. Classical test theory and item response theory were used to establish the measure's psychometric properties. ResultsTotal scale reliability was very good (α=0.838), as was internal consistency for a 16-item general HPV knowledge subset (α=0.849). Subsets of HPV testing and vaccination items showed reasonable test–retest reliability (rtest–retest=0.62 and 0.69) but moderate internal consistency (α=0.52 and 0.56). Dimensionality analyses suggested that one item was not measuring the same construct as the remainder of the questionnaire. A 2-parameter logistic item response theory (IRT) model was fitted to the remaining 28 scale items. ConclusionsA structurally coherent set of items covering a range of important HPV knowledge was developed. Responses indicated a reliable questionnaire, which allowed the fitting of an IRT model.

Highlights

High-risk human papillomavirus (HPV) is the primary cause of cervical cancer (Bosch and Munoz, 2002)
The Root Mean Square Error of Approximation (RMSEA) and Standardised Root Mean Square Residual (SRMR) statistics provide stronger support for the unidimensionality assumption, including for the full 29 item set in the case of SRMR
The scaled score distribution of the general HPV knowledge subset of items had a mean of − 0.234 (s.d. = 0.981), while for the HPV testing subset of items the mean was 0.000 (s.d. = 0.729) and for the HPV vaccination subset it was − 0.021 (s.d. = 0.760). These results show that respondents scored best on HPV testing items and worst on general HPV knowledge

Summary

Introduction

High-risk human papillomavirus (HPV) is the primary cause of cervical cancer (Bosch and Munoz, 2002). HPV testing and vaccination are important elements for cervical cancer prevention. HPV DNA testing is increasingly used as part of cervical screening (Albrow et al, 2012; National Health and Medical Research Council Cervical Screening Guidelines Review Group, 2005; U.S.Preventive Services Task Force, 2012) and prophylactic vaccination programmes have been established in many countries (Arbyn et al, 2012). Low awareness and poor knowledge of HPV are concerning given the active role that people are increasingly expected to play in making decisions about their healthcare (Woolf et al, 2005). Studies have explored what women need to know about HPV from a clinical perspective (Tristram, 2006), and what women want to know about

Methods

Results

Conclusion