Methods
A protocol for this qualitative systematic review and evidence synthesis was not registered with a database. However, systematic review methods were followed to identify all the relevant qualitative literature pertaining to the research question and are described below.
Search Strategy
A comprehensive search strategy for Embase was developed to ensure that all relevant literature was captured. A combination of text words and the following indexing terms (MeSH) was used: "papillomavirus", "wart virus", "vaccination", "immunization", "immunization programs", "wart virus vaccines", "qualitative", "interviews" and "focus groups" (Additional file 1 http://www.biomedcentral.com/1471-2458/14/700/additional). The search strategy was subsequently modified for other databases. The following databases were searched from inception to 9 March 2012: CINAHL; Embase; Medline; PsycINFO, and; ISI Web of Science and ISI Proceedings. All abstracts were saved using Endnote ×3 reference manager software.
Inclusion and Exclusion Criteria
Studies were eligible if qualitative research methods (interviews, focus groups, observations) or open-ended questions in questionnaires were used to explore views and behaviours related to decision-making of HPV vaccination of young women. The populations of interest were: young women; their parents and/or primary care givers; healthcare professionals involved in the delivery of the programme, and; other relevant stakeholders such as policy makers, community leaders, and teachers. Young women were defined as adolescent girls aged between nine and 18 years. Studies which included the views of women over the age of 18 years were included if the views of young women were reported separately.
No restriction was imposed by publication date in order to capture views of the HPV vaccine before its general availability (which might influence future delivery or uptake) as well as existing HPV vaccination programmes.
Studies not published in English were excluded. Conference abstracts, editorials, letters and books were included only if they presented original qualitative data. Primary studies with adults were excluded if views about HPV vaccination of young women were not reported in the results. Questionnaire studies reporting only closed questions were excluded. For this paper, with a focus on examining factors affecting decision-making in high-income countries, studies were categorised as high-income using the World Bank classifications.
Study Selection
Two reviewers (HF and SA) independently assessed the titles and abstracts from the literature searches and the relevance of studies retrieved as full text. Disagreements were resolved by discussion. The reference lists and bibliographies from relevant studies and reviews were hand-searched by one reviewer (HF) for additional primary studies not retrieved by the electronic search.
Quality Assessment
Currently, there is no consensus regarding the assessment of the quality of qualitative research and subsequent exclusion from systematic reviews. For this systematic review and evidence synthesis, each primary study was appraised using the Critical Appraisal Skills Programme criteria for evaluating qualitative research. Studies were not automatically excluded on the basis of overall 'low quality' if they contributed relevant qualitative information. However, the methodology and results of one study was presented in such a way that the findings were considered unreliable and this was excluded.
Data Extraction
Data pertaining to the methodology and context, including study and participant characteristics of each primary study, were extracted and entered into an excel spread sheet by one reviewer (HF).
Thematic Synthesis
Several methodologies for the synthesis of qualitative research exist. To analysis the data, the methodology was based on the methods of thematic synthesis reported by Thomas and Harden and using the Framework methods of qualitative data management. These methods suit studies with a priori aims and objectives designed to directly inform policy and practice. The overall purpose of the synthesis was to 'pool' the results from individual primary studies by initially separating the findings, interpreting and then combining them through the identification of key themes across the studies.
Thematic synthesis was undertaken by one reviewer (HF) with discussions held with the second reviewer (SA) as analysis progressed. Elements of the text reported in the 'results' section of each primary study represented the basic units of the review. The text from each primary study was extracted verbatim and entered into a spreadsheet. The data was 'charted' into the matrix for studies relating to: young women, their parents/carers, and professionals.
Familiarisation with the dataset included reading and rereading the textual data in these primary charts. Sections of text were coded, with multiple codes being allocated where appropriate. The primary charts were retained and revisited as required, but streamlined versions were produced as the process of summarising and synthesising the data progressed. In these subsequent charts, key terms and phrases were retained while repetition within studies and extraneous text were removed. During this process, differences or similarities were identified within emerging themes.
The Socio-ecological Model
During analysis the findings were considered in relation to the socio-ecological model which considers that behaviour is shaped by a complex interaction between factors operating at the following levels: (i) public policy; (ii) community; (iii) organisational; (iv) interpersonal, and; (v) intrapersonal. We used the socio-ecological model to provide a framework for understanding how decisions of stakeholders at different levels of the model may affect access of the HPV vaccine for young women.