More than 40 strains of human papillomavirus (HPV) are sexually transmitted and infect mucosal surfaces of the lower genital area . Around 15 of these strains, including HPV16 and HPV18, are known as ‘high-risk’ types because they are a necessary cause of cervical cancer . These high-risk types are common , asymptomatic, and cleared by most women in a few months; in some women, however, persistent infection(s) can lead to cervical pre-cancer and cancer .
It is generally accepted that, in the near future, cervical screening will be based on testing for infection with high-risk HPV types, rather than on conventional cytological smears [4–6]. Compared to smears, HPV testing (which can be conducted on residual smear samples) has higher negative predictive value and higher sensitivity for the detection of pre-cancer  making HPV-based screening likely to be effective [8, 9] and cost-effective [10, 11]. In addition, two prophylactic HPV vaccines have been developed . For both vaccines, the vaccination schedule involves three doses at intervals within a six-month period and is most effective when administered before HPV exposure (i.e., among the sexually naive). Organised vaccination, together with screening, could prevent most cervical cancers .
Ireland, which has a mixed public-private healthcare system, is at the forefront of this transformation in cervical cancer prevention. A national screening programme, CervicalCheck, was rolled-out in autumn 2008. The programme invites women aged 25 to 60 for a free smear in primary care every 3 to 5 years . Prior to this, a woman could obtain a smear from her general practitioner (GP) or ‘Well Woman’ clinic for around €50. HPV testing is available through some practices and Well Woman clinics. CervicalCheck is currently introducing HPV testing in the follow-up of women treated for pre-cancer and is considering other uses, for example as a primary screening tool. HPV vaccination was licensed in males and females aged 9 to 26 in 2006, and can be purchased in primary care for around €600. In autumn 2010, a national schools-based vaccination programme started, providing free vaccination to girls aged around 12 . These developments have been accompanied by changes in primary care, notably a move from single-doctor to multi-doctor practices supported by practice nurses. Traditionally, GPs were the primary smear takers, but practice nurses now play an important role in smear taking  and perform around one-half of the smears within CervicalCheck.
In addition to providing smear tests, GPs and practice nurses are likely to be key sources of information and advice for patients on HPV infection, vaccination, and testing. For women, their GPs’ attitude influences their own prevention behaviours [17, 18]. Moreover, healthcare professionals’ compliance with, and encouragement of, HPV vaccination is crucial in achieving high vaccination rates . Therefore, GPs’ and practice nurses’ clinical practices in relation to HPV will impact on the success of cervical cancer prevention strategies.
Little is known about what influences HPV-related clinical practice. Most research has focussed on practitioners’ knowledge, and while this is an important predictor of clinical behaviour, it is unlikely to be the sole influence [20, 21]. A 2004 US family doctors’ survey found that substantial proportions were unaware of information on HPV infection relevant for patient counselling . A 2007 survey of GPs in Ireland, using the same instrument (see Additional file 1), found lower knowledge levels than in the US survey , and important gaps in knowledge about HPV vaccination, consistent with findings from elsewhere [24–27]. More than 95% of GPs desired national guidelines or policy on HPV vaccination and testing. HPV infection knowledge predicted HPV vaccination intentions: GPs with higher knowledge scores were significantly more likely to be willing to vaccinate sexually naive girls aged under 16 than those with lower knowledge scores. A 2007 study among US family doctors, found the Theory of Planned Behaviour  variables intentions, subjective norms (perceptions about whether others approve of vaccination), and perceived behavioural control (perceptions about whether the decision to vaccinate is within the control of the doctor) influenced HPV vaccination behaviour . No studies have investigated determinants of practice nurses’ clinical behaviours in this field.
ATHENS (A Trial of HPV Education and Support), which is being conducted under the umbrella of the CERVIVA research consortium (http://www.cerviva.ie) aims to develop a theory-based intervention to support primary care practitioners in their practice in relation to HPV infection, vaccination, and testing. The current study is the first step in this intervention development process. The primary aims were to: identify HPV-related clinical behaviours that the intervention will target; clarify roles and responsibilities of GPs and practice nurses in these areas; and determine what influences these clinical behaviours. Because little is known about practice, or potentially relevant psychological theories, we used qualitative methods to generate data with richness and depth, and analysed this using the Theoretical Domains Framework (TDF) . As the TDF was originally developed to aid understanding of clinical behaviours around evidence-based guidelines, a secondary objective was to reflect the utility of the TDF in a way that may inform other researchers who are considering using it.
This article is one of a series documenting the development and use of the TDF to advance the science of implementation research. An overview of the articles contained in the series is provided in the introductory article .