|Home | About | Journals | Submit | Contact Us | Français|
Colorectal cancer (CRC) poses a serious health problem worldwide. While screening is effective in reducing CRC mortality, participation in screening tests is generally suboptimal and social inequities in participation are frequently reported. The goal of this review is to synthesise factors that influence an individual’s decision to participate in CRC screening, and to explore how those factors vary by sex, ethnicity and socioeconomic status.
A primary search of Cumulative Index to Nursing and Allied Health Literature (CINAHL), MEDLINE, EMBASE, PsycINFO, and a secondary search of grey literature and articles taken from references of included articles (from inception to July 2013).
A systematic review and Meta-study synthesis of qualitative studies that address perceived benefits and barriers to participation in CRC screening tests among adults 50 years of age or older.
The two-staged Meta-study methodology by Paterson will be used to conduct this review. In stage 1, similarities/differences, patterns and themes will be identified across three levels of analysis while preserving the context of original studies. In stage 2, synthesis will extend beyond the analysis to generate new theory of the phenomenon through a process called Meta-synthesis.
This review offers to generate a framework to better understand benefits and barriers that affect decision-making to participate in CRC screening among different sectors of the population. This framework will be a relevant tool for policy makers in framing educational materials, for patient-centered communication, and for researchers interested in the science of equity. This review is registered in PROSPERO (registration number: CRD42013005025).
Colorectal cancer (CRC) poses a serious health problem worldwide. CRC is the second most common cause of cancer death in the USA,1 Canada,2 the UK,3 4 Germany,5 Australia6 and Japan.7 It is estimated that by 2013, 142 820 new CRC cases and 50 830 CRC deaths will occur in the USA,1 and 23 900 new CRC cases and 9200 CRC deaths will occur in Canada.2
Screening for CRC can reduce the burden of the disease. Screening tests for CRC include faecal occult blood testing (guaiac FOBT) and faecal immunochemical test (FIT), flexible sigmoidoscopy, colonoscopy, CT colonography (CTC) and faecal DNA testing. Several of these tests are effective in reducing the incidence of, and in some instances, the mortality from the disease. Three landmark randomised controlled trials (RCTs) demonstrated that biennial use of guaiac FOBT coupled with colonoscopy in persons who tested positive was associated with a reduction in CRC mortality by 15%.8–10
Screening for CRC is a complex process, and many publicly funded healthcare systems have implemented an organised, population-based approach for screening such as in the UK,11 most provinces in Canada,12 19 of the 27 European Union (EU) countries,13 Japan14 and Korea.15 Population-based organised screening programmes involve inviting a defined population at average risk for the disease (ie, people who do not have CRC, or strong family history of CRC, or medical conditions that put them at higher risk of developing CRC such as Crohn's disease or ulcerative colitis) to attend screening. The success of a high-quality organised, population-based CRC screening programme depends on adequate uptake as well as social equity in uptake.16 Early evaluation indicates an overall low participation and social inequity in participation. Participation in CRC screening tends to be lower among ethnic minorities,11 17–19 low socioeconomic status individuals11 20–22 and among men.20 22–24
While social inequities in uptake are well described in the literature,25 26 what is missing is a clear understanding of why CRC screening is or is not appealing to individuals, aspects of screening that are valued and those that are culturally acceptable. Qualitative studies are important sources for this information. To date, a wide range of qualitative studies have elicited views on the perceived benefits and barriers to participation in screening from a range of ethnic and socioeconomic groups in various countries. The in-depth analyses in these studies reveal the complexity of social factors that affect an individual’s decision to participate in screening. For example, studies have shown that difficulties in doing screening tests at home (ie, FOBT) and the perceived need for screening while having no symptoms of colorectal disease are the main barriers for participation across different population groups.27 28 In certain cultures, men perceive colonoscopy as embarrassing, invasive and an affront to their masculinity.22–24 29–37 Women, in general, believe that their experience with other cancer screening tests such as mammography encourages them to do CRC screening,38 and because they often assume the role of caregiver in a family, they value the importance of self-care and early detection in order to prevent personal and family suffering.22 Less education, consistently equated with poorer health literacy skills, is often cited as the main barrier for CRC screening among low SES individuals. Poor health literacy is associated with reduced ability to ‘obtain, process and understand health information’,22 and the likelihood of engaging in preventive health behaviours such as CRC screening.39–41 Other reported factors influencing participation in CRC screening among certain ethnic populations include maintaining a positive energy (qi) and spirit (jing shen), as well as the belief that moderation of exercise and diet were enough to control the ‘toxins’ and prevent CRC.19
Systematic reviews of quantitative studies have focused on investigating the efficacy of CRC screening tests,42 43 the determinants of CRC screening participation25 26 and the effectiveness of interventions to increase screening participation.26 44 45 However, no synthesis of qualitative studies exists to investigate why individuals undergo CRC screening or not, their perceptions of and experiences with CRC screening and which aspects of screening are valued and culturally acceptable. A well-designed synthesis of qualitative studies is needed to achieve a greater conceptual understanding of the perceived barriers and benefits associated with participation in CRC screening. This understanding is a necessary step to direct intervention designs to raise overall participation, reduce inequities in participation and eventually reduce mortality from CRC.
The Meta-study approach, a commonly used method to synthesise qualitative studies, was the most suitable approach to answer our research question. We considered other methods such as the Realist review (which seeks to understand what works for whom, under what circumstances and why) and Meta-ethnography (which aims to uncover a new theory to explain a range of findings), neither focuses on the experiences of people specifically nor considers the quality of included studies as part of the analysis.
The objectives of our study are to systematically review the literature for qualitative evidence that explores the factors that influence the decision of individuals aged 50 years or over at average risk for CRC to participate in CRC screening, and how those factors vary by sex, ethnicity and SES. Our secondary aim will be to generate a framework to better understand the perceived benefits and barriers that affect individual decision-making.
We will use the Meta-study methodology to conduct our review, which is a systematical, analytical and synthesis research method pioneered by Paterson et al.46 We selected this methodology because it was the most suitable to answer our research question. Meta-study is a multifaceted, systematical knowledge synthesis method aimed at better understanding how people construct knowledge.47 In the context of our study, this is related to better understanding the determinants of CRC screening test participation. More specifically, it is an interpretive qualitative research approach in the constructivist paradigm (ie, the role of the investigator is to understand how people construct knowledge about the phenomenon under study).48 The aims of Meta-study are to ‘analyse’ and ‘synthesise’ what has been reported in the literature—these are considered distinct. Analysis involves identifying commonalities, differences, patterns and themes in a body of qualitative research (ie, what is typically done in a qualitative systematic review). Synthesis extends beyond analysis to identify ‘truths’ about the phenomenon under study by considering how the primary researchers interpreted the data (ie, Meta-data), the design and quality of studies (Meta-method) and the theoretical frameworks or perspectives used in these research reports (Meta-theory). To answer our research questions, we need to go beyond the ‘analysis’ of existing literature, as CRC screening is complex, and currently, it is unknown why people do or do not undergo CRC screening. We hypothesise that there may be underlying factors involved in an individual’s perceptions and experiences well beyond CRC as a disease itself that influences their decision to undergo diagnostic testing (eg, cultural beliefs). Meta-study will allow us to extend beyond the typical ‘analysis’ phase because it considers the triangulation of the raw data (meta-data) and its quality (meta-method) as well as the theoretical underpinnings of this data (meta-theory). This level of ‘synthesis’ called ‘Meta-synthesis’ will lead to a new understanding of CRC and screening decisions (eg, colonoscopy) beyond what would be discovered in a qualitative systematical review (which tends to focus entirely on the primary research findings).
The proposed flow of our Meta-study methods is represented in figure 1. Our Meta-study will be guided by the reporting standards as outlined in the ENTREQ criteria (Enhancing Transparency in Reporting the Synthesis of Qualitative Research).49 This is a 21-item checklist grouped into five main domains: introduction, methods and methodology, literature search and selection, appraisal and synthesis of findings. The protocol has been registered in the PROSPERO database (registration number: CRD42013005025, available at http://www.crd.york.ac.uk/PROSPERO).
We developed our eligibility criteria from our research questions. The review will use the following PICOS (Population, Intervention, Context, Outcomes and Study design) elements: (1) Population: Adults aged 50 years or over referred for CRC screening; exclusion criteria are studies investigating participants previously diagnosed with CRC; a hereditary, personal or family history of CRC (eg, Familial Adenomatous Polyposis (AFP) and hereditary non-polyposis CRC (HNPCC)); and a history of inflammatory bowel disease (eg, ulcerative colitis and Crohn's disease); (2) Intervention: We will identify all articles investigating perceptions of CRC screening as well as those investigating CRC as a disease ; (3) Context: We will investigate any variations in perceptions by sex, ethnicity, SES and other factors influencing CRC screening behaviour; (4) Outcomes: Perceptions related to CRC as a disease, causes of CRC, benefits and barriers to CRC screening and any other contextual factors that motivate or influence people's decision to participate in CRC screening; (5) Study design: We will include all qualitative studies and mixed-methods studies with a qualitative component. We will exclude experimental, observational and any non-empirical studies (ie, not based on observation or experience, opinion-driven or no hypothesis testing) such as editorials, letters, commentaries and narrative reviews.
We will conduct a systematic search in the following electronic databases from inception to July 2013: MEDLINE, EMBASE, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycINFO and Social Science Abstracts (SSA). We will conduct a secondary search of the grey literature (unpublished) from sources such as Cancer Care Ontario and the National Health System Bowel Cancer Screening Programme. We will also search the reference lists of included articles and identify other articles through contact with experts in the field and linkages with our team members (eg, Cancer Care Ontario). There will be no language restrictions in our searches. We anticipate completing the review by April 2014.
Literature searching will be conducted by an experienced information specialist. The search strategy for the main database (MEDLINE) will be peer reviewed by another experienced information specialist using the PRESS checklist (ie, Peer Review of Electronic Search Strategies).50 The resulting retrieval yield will be limited to qualitative studies and mixed methods with a qualitative component using the optimised search strategy filter for qualitative studies of selected databases: MEDLINE,51 EMBASE,52 PsycINFO53 and CINAHL.54 The draft search strategy for MEDLINE is available in online supplementary appendix 2. For the other databases, the search strategies are available from the authors on request.
We will first perform a calibration exercise to ensure reliability of screening. Using the inclusion/exclusion criteria available in online supplementary appendix 1, two reviewers will independently screen a random sample of citations (25–50 citations) using our online Synthesi.SR Tool (proprietary online systematic review software developed for our Knowledge Synthesis Center at St. Michael's Hospital).55 We will calculate inter-rater agreement for study inclusion using per cent agreement, and repeat our pilot screening exercise until we reach at least 90% agreement at which point investigators will independently review titles and abstracts of potentially relevant articles in duplicate (level 1 screening). For level 2 screening, we will follow a similar calibration exercise as described for level 1 screening to identify full-text articles. Conflicts will be resolved through research team consensus for both levels of screening.
Two reviewers will abstract data independently using a standardised data collection form. The form will first be pilot tested on a random sample of 5–10 included studies and modified accordingly. Data abstraction will begin only if agreement is at least 95% among the two abstractors. We will extract data on study characteristics (eg, first author and citation) and qualitative study quality criteria according to the CASP tool (Critical Appraisal Skills Programme), which includes a 10-item checklist to assess the clarity of research aims, appropriateness of methodology and recruitment strategy, data collection, ethical considerations including the relationship between researcher and participants, the rigour of analysis, clear statement of findings and the value of the research.56 All data abstraction will be conducted using our online Synthesi.SR Tool, which provides a platform to resolve conflicts between reviewers directly in the system. Discrepancies will be reviewed and resolved by discussion among the team. The reporting of our review will be guided by the ENTREQ criteria.49
We will perform a two-staged synthesis of the data (ie, Analysis and Synthesis) with the goal of creating a new interpretation of the phenomenon under investigation (ie, a new understanding of CRC and screening decisions; figure 1).
We will identify the similarities and differences, patterns and themes across three levels of analysis.46 47 57 Level 1—Meta-data analysis: this will involve the interpretive analysis of research findings from primary studies to identify similarities and discrepancies among them using any one of several qualitative data analytical approaches (eg, line of argument; grounded theory; thematic analysis). The type of analysis method we select will be driven by the data that will emerge. In the context of our work, we anticipate that this will likely involve using thematical analysis to group themes (such as the benefits and barriers to CRC screening) according to sex, SES or other factors that emerge, and then noting the similarities and differences between them. Level 2—Meta-method: This level of analysis will examine how the research methods and procedures in primary studies were used to generate and interpret data and shape the findings. It will include a process of appraising each included study according to the CASP tool for quality assessment of qualitative studies.56 A third reviewer will be available to settle discrepancies between reviewers for applying the CASP criteria. Level 3—Meta-theory: This level of analysis examines the theories that underpin the study authors’ framing of their research questions, their criteria for inclusion and their conceptual framework for interpretation. It is the level at which the theoretical perspectives in qualitative reports can be interrogated to explain the phenomenon under study. We will review each report to identify the theoretical perspective used and the ‘schools of thought’ around CRC screening, and to determine how context may influence such perspectives.
In stage 2, synthesis will extend beyond the three levels of analysis to generate a new and expanded theory of the phenomenon through a process called Meta-synthesis. In contrast to the three-level analytic stage, Meta-synthesis is ‘a creative, dynamic and interactive process that defies codification’.46 It involves interpreting the influence of method and theory variation in the findings to produce a new understanding of the phenomenon. For example, we will determine these influences by documenting how each study performs their data analysis (eg, thematic analysis of semistructured interviews=Meta-data analysis); whether they used a theoretical framework to drive their study (eg, the Health Belief Model=Meta-theory); and to determine the study quality (eg, the CASP criteria=Meta-method). Once we collect this data from all studies, we will be able to triangulate this data from individual studies to reveal a new, collective understanding of CRC screening participation. This interpretation will be documented during data extractions. To reduce the potential of bias introduced from such an interpretive process, two investigators will independently perform this interpretation, which will be discussed and finalised with input of the entire research team. We will use findings informed by the three-level analysis to develop a framework that shows the perceived benefits and barriers of CRC screening participation according to sex, SES, cultural beliefs and other factors that may emerge.
We will use findings from our in-depth analysis of qualitative studies to generate a framework to better understand the benefits and barriers that affect decision-making to participate in CRC screening among different sectors of the population. We anticipate that this framework will be relevant for a wide range of knowledge users: policy makers will be able to use the framework as a tool to frame educational materials to address barriers to CRC screening; and physicians may use it as a tool in patient-centered communication or in group education sessions in order to engage culturally heterogeneous population into a discussion on CRC screening. This review also offers advancement in the science of equity by identifying the determinants of social inequities in CRC screening participation. Using the anticipated framework, researchers may also design novel interventions to address those inequities, which may lead to improved quality in practice and advancement in evidence-based decision-making. Furthermore, synthesis of available qualitative evidence of barriers to participation in CRC screening currently does not exist. Therefore, our findings may trigger other systematic reviews of gaps in information that we may identify. We will also advance the knowledge of conducting Meta-study reviews by rigorously executing its steps in the context of our research question and to document this process extensively in our final report.
Our study may also have some limitations. As with any qualitative studies, our work may be susceptible to threats to internal validity (ie, credibility), external validity (ie, transferability) and reliability (dependability).57 We will address potential threats to credibility by pilot testing the data abstraction forms and involving group team discussions throughout the interpretation of findings. The knowledge produced in our review may not be transferable to other people or settings. For example, findings may be limited to individuals from different ethnic minorities living in developed countries, which may limit the transferability of our findings to the overall ethnic population. However, we will abstract a detailed account of the population and setting of each included qualitative study to maximise the potential for transferability of our findings. To limit the potential of biases that may be introduced by investigators with respect to the dependability and conformability of our work, we will standardise procedures, methods and analysis strategies across all aspects of the review process.
We will ensure broad dissemination of this synthesis review to include publication in open access journals as well as conference presentations. We will also plan to hold a meeting with our key stakeholders (ie, clinicians, researchers, people with CRC and decision-makers) to discuss the findings, to generate key messages most relevant to each and to discuss the next steps including the development of educational materials that will address gaps in CRC screening participation.
Contributors: GHA and NB helped conceive the study; GHA, NB, MK and VV conceived the study design. GHA and MK helped draft the protocol. LP developed and executed the search strategy and edited the draft protocol. All authors helped editing the draft protocol, read and approved the final manuscript.
Funding: This research was supported through a Cancer Care Ontario research grant and Canadian Cancer Society Research Institute award (grant number 2011-700803).
Competing interests: NB holds the Cancer Care Ontario Health Services Research Chair.
Provenance and peer review: Not commissioned; externally peer reviewed.
Data sharing statement: Unpublished study data such as the search strategies for the other databases (EMBASE, CINAHL, PsycINFO, SSA) are available on request to the corresponding author.