Applications of implementation science frameworks, models and theories in disparities-focused cancer screening interventions: a scoping review protocol Authors: Serena Xiong, Lisa Klesges, Michelle Doering, Rebekah J Pratt Published: 2023 ABSTRACT BackgroundImplementation science (IS) frameworks, models and theories (FMTs) have gained popularity in guiding the implementation and evaluation of evidence-based interventions (EBIs) for cancer screening. However, there are significant research gaps in understanding their applications in cancer health disparities contexts. This paper outlines a scoping review protocol designed to explore the utilisation of IS FMTs in cancer screening EBIs to inform intervention designs and adaptations.Methods and analysisThis scoping review protocol adheres to Arksey and O’Malley’s five-step methodological framework for conducting scoping studies. Search strategies were conducted in five bibliographic databases: Ovid MEDLINE, PubMed, Scopus, Web of Science and EMBASE. The search was run on 22 June 2023 with an English language filter and a date limit of 2001-current. Two reviewers will independently screen studies for inclusion and exclusion criteria. A third reviewer will be consulted, where appropriate at any of the review stages, to achieve consensus or resolve conflicts. Data will be collected, managed and analysed using Covidence. A narrative synthesis, based on Popay et al’s methodology, will guide reporting and summarisation of results. The review will adhere to the PRISMA Extension for Scoping Reviews guidelines.Ethics and disseminationThis scoping review is a novel approach for examining a growing corpus of research literature on IS FMT applications used in cancer screening EBIs. As a secondary analysis, this scoping review does not require approval from an institutional review board. We anticipate the review will produce insightful information (eg, challenges, key areas for future directions) on the applications of IS TMFs in designing, deploying and testing EBIs for populations experiencing cancer screening disparities. We will disseminate the results through journals and conferences targeting IS and cancer prevention researchers and practitioners. STRENGTHS AND LIMITATIONS OF THIS STUDY This novel scoping review aims to characterise how implementation science frameworks, models and theories have been employed in cancer screening evidence-based interventions to inform and guide intervention designs and/or adaptations within cancer health disparities contexts. This novel scoping review aims to characterise how implementation science frameworks, models and theories have been employed in cancer screening evidence-based interventions to inform and guide intervention designs and/or adaptations within cancer health disparities contexts. A feasible method for identifying cancer screening interventions that have used implementation science frameworks, models and theories over the past 20 years is presented. A feasible method for identifying cancer screening interventions that have used implementation science frameworks, models and theories over the past 20 years is presented. A pragmatic technique for synthesising research that encompasses diverse methodological approaches, settings, study populations and cancer screening types is described. A pragmatic technique for synthesising research that encompasses diverse methodological approaches, settings, study populations and cancer screening types is described. Due to the heterogeneity and extensive scope of the studies included, the final data extraction framework will not be completed until the conclusion of the review. Due to the heterogeneity and extensive scope of the studies included, the final data extraction framework will not be completed until the conclusion of the review. The included studies within the review will be limited to English-only, peer-reviewed publications. The included studies within the review will be limited to English-only, peer-reviewed publications. Introduction Significant advances in research and healthcare services have been made to curb cancer mortality in the USA; however, population groups from disadvantaged racial/ethnic, sex, gender, age, geographical and socioeconomic backgrounds have minimally benefited from these advances.1 Consequently, cancer disparities between more-advantaged and less-advantaged population groups have continued to widen. According to the National Cancer Institute’s (NCI) Surveillance, Epidemiology and End Results Programme, non-Hispanic blacks had the highest cancer death rates among all racial/ethnic groups, with a rate of 166.8 deaths per 100 000 people, compared with 149.3 deaths per 100 000 people among non-Hispanic whites over a 20-year period (2000–2020).2 Many of these cancers are preventable, yet disadvantaged populations bear a disproportionate burden of cancer deaths. Reducing these cancer inequities will require researchers and healthcare practitioners to prioritise the design and implementation of evidence-based cancer preventions targeting health disparities.3 4 Research on cancer prevention has led to the development of numerous evidence-based interventions (EBIs) (eg, screening practices and programmes, vaccination recommendations).5 6 However, many cancer prevention EBIs are never fully implemented or fail to be scaled-up.7–9 Challenges associated with implementing these EBIs often include difficulties on the part of researchers and practitioners in understanding and identifying EBIs, limited planning strategies (eg, use of a preimplementation theory or framework) to enhance their delivery, and problems adapting existing EBIs for new settings and populations. Recently, researchers within the cancer care continuum have been calling for increased recognition and application of implementation science (IS) to illuminate and address some of these implementation challenges.1 Frameworks, models and theories (FMT) are becoming recognised as critical instruments to the mission of IS.10 When selected intently, an FMT can significantly improve an IS study. Models provide a systematic structure for developing, managing and evaluating implementation efforts, linking study aims, design, measures and analytical strategies.11 12 Frameworks can also help narrow the scope of a study by assisting with the focus of the research question and guiding the selection of constructs to measure. Framing a study within an IS FMT can also help explain why an EBI works or does not work in its chosen context. In this way, theoretical applications can help on the front-end to organise and understand implementation impacts and on the back-end to understand why or how implementation strategies succeed or fail.13 14 Current FMTs, however, include little guidance on how to adapt an intervention or identify implementation strategies for different disadvantaged groups to address health disparities.15 16 Adaptations, defined as intentional change(s) to the design or delivery of an intervention to facilitate effectiveness or contextual fit,17 may positively impact cancer prevention and implementation outcomes in populations affected by health disparities.3 18–20 For example, as shown in several pilot studies with the Somali community in Minnesota, culturally tailored cancer-prevention efforts can increase motivation to participate in cancer screening.21 22 Yet, little research exists on whether the specific tailoring or adaptations identified and implemented through IS FMTs in cancer prevention EBIs can lead to any significant reduction in cancer-related disparities. Discussions on how IS FMTs can address health inequities have only recently emerged within both national and global health contexts.23 Central to these conversations is how FMTs can be employed to appropriately identify and adapt evidence-based cancer screening practices to the context, culture and acceptance levels of key stakeholders (eg, implementing organisations, intended programme recipients).16 17 Since the awarding of the National Institutes of Health NCI supplements to promote dissemination and implementation of evidence-based cancer programmes and practices in 2001,24 no effort has been made to characterise how IS FMTs have been applied within cancer screening interventions. The aims of this scoping review are: (1) to characterise how IS FMTs have been employed in cancer screening EBIs within cancer health disparities contexts and (2) to describe how IS FMTs have informed and guided intervention designs and/or adaptations. For the scope of this review, we have defined a cancer health disparities context as a health disparity population, legally defined by the 2000 Minority Health and Health Disparities Research and Education Act and endorsed by the National Institute on Minority Health and Health Disparities as ‘a population is a health disparity population if there is a significant disparity in the overall rate of disease incidence, prevalence, morbidity, mortality or survival rates in the population as compared with the health status of the general population’ with a known or reported cancer health disparity (eg, difference in cancer incidence, prevalence mortality, screening rates and so forth).25 Methods and analysis Given the variability and growing body of literature in cancer screening EBIs, we decided to undertake a scoping review of published studies to best map out and characterise how IS TMFs have been applied in this cancer prevention domain. Unlike systematic reviews or meta-analyses that require narrow parameters (eg, questions, populations) and quality assessments, scoping reviews are conducive to examining general topics and most useful for comprehensively and systematically reviewing the literature for key concepts, theories, evidence and research gaps.26 27 To guide the design of our scoping study, we drew from Arksey and O’Malley’s five-step methodological framework.27 We have outlined and discussed each of the five stages below. Stage 1: identifying the research questions The objective of our scoping review is to examine how IS FMTs have been employed in cancer screening EBIs to target disparities. We were aware that FMTs can span a variety of disciplines and may not be specific to the field of IS alone. Per Arksey and O’Malley’s framework, we used an iterative approach to continually refine our FMT parameters and arrive at our research questions (see table 1). Given the health disparities focus of this scoping review, we also drew from the PRISMA-Equity 2012 Extension28 to include components of equity in our research questions. Table 1List of research questions and operational definitionsResearch questionsOperational definitionsWhat types of frameworks, models and theories (FMTS) are being used in cancer screening evidence-based interventions (EBIs) for populations experiencing cancer health disparities?Definitions and applications of FMTs used in implementation science are based on Nilsen’s taxonomy14Process modelsDeterminant frameworksClassic theoriesImplementation theoriesEvaluation frameworksDissemination and Implementation Models in Health Science Webtool.37Cancer screening refers to secondary prevention services focus on detecting and treating cancers in specific anatomical sites that have well-established screening methods recommended by clinical guidelines.38 The specific cancer sites of interest for this scoping review are:BreastCervicalColorectalLungEBIs are broadly defined as healthcare programmes, practices, processes, policies and guidelines proven with efficacy and effectiveness.39How are FMTs being used in cancer screening EBIs?Applications of FMTs are based on Nilsen’s taxonomy14 and as stated by the author(s)The level of FMT integration, defined as the extent to which a framework, model, or theory is integrated into the intervention/implementation process of a study, is captured with the following de novo scale:Low (defined as <20% of constructs applied).Moderate (defined as ≥20% to 50% of constructs applied).High (defined as ≥51% constructs applied).If an FMT was used to adapt a cancer screening EBI, the FRAME17 will be used to extract information on this process and outcome.What barriers and facilitators to applying FMTs are being described in the cancer screening literature?As stated by the author(s)How are cancer health disparities being described and integrated in published studies?Cancer health disparities are informed by the following rationale and objective elements of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses - Equity (PRISMA-Equity) extension checklist28:Population characteristics including gender, age, race/ethnicity, geographical location, socioeconomic statusAssumptions and rationale related to equity are statedIntention to address equity as the focus of study, the research question or within the analysis List of research questions and operational definitions Process models Process models Determinant frameworks Determinant frameworks Classic theories Classic theories Implementation theories Implementation theories Evaluation frameworks Evaluation frameworks Breast Breast Cervical Cervical Colorectal Colorectal Lung Lung Low (defined as <20% of constructs applied). Low (defined as <20% of constructs applied). Moderate (defined as ≥20% to 50% of constructs applied). Moderate (defined as ≥20% to 50% of constructs applied). High (defined as ≥51% constructs applied). High (defined as ≥51% constructs applied). Population characteristics including gender, age, race/ethnicity, geographical location, socioeconomic status Population characteristics including gender, age, race/ethnicity, geographical location, socioeconomic status Assumptions and rationale related to equity are stated Assumptions and rationale related to equity are stated Intention to address equity as the focus of study, the research question or within the analysis Intention to address equity as the focus of study, the research question or within the analysis Stage 2: identifying relevant studies To identify relevant studies for our scoping review, we deliberated and decided on the following criteria for eligibility, bibliographic databases to search and formulated a search strategy and key terms. Eligibility criteria The following inclusion criteria were used to guide the search strategy and will also be used when reviewing articles: Any type of study design (including randomised controlled trials (RCTs), non-RCTs, intervention and non-intervention studies, comparative-effectiveness research), qualitative, quantitative and mixed-methods). Any type of study design (including randomised controlled trials (RCTs), non-RCTs, intervention and non-intervention studies, comparative-effectiveness research), qualitative, quantitative and mixed-methods). Any study involving an EBI reflecting the following cancer screening types: breast, cervical, colon and/or lung. Any study involving an EBI reflecting the following cancer screening types: breast, cervical, colon and/or lung. A study conducted within the context of a health disparity population(s) with a reported or known cancer disparity(ies). A study conducted within the context of a health disparity population(s) with a reported or known cancer disparity(ies). Published in peer-reviewed journals with full-text available for sufficient information to assess eligibility. Published in peer-reviewed journals with full-text available for sufficient information to assess eligibility. English-only publications. English-only publications. Published between 2001 and current (2023). Published between 2001 and current (2023). Explicitly used an IS FMT (as referenced in Nilsen’s taxonomy or the D&I Models in Health Research and Practice Webtool). Explicitly used an IS FMT (as referenced in Nilsen’s taxonomy or the D&I Models in Health Research and Practice Webtool). Studies were excluded if they included pilot or formative studies leading to the development of an IS FMT, case studies and case series of cancer screening interventions, systematic reviews or meta-analyses of cancer screening interventions, studies that did not use an IS framework, and studies with no empirical data (involving primary data collection) or were research protocols. Databases Studies were identified through searches from five bibliographic databases: (1) Ovid MEDLINE, (2) PubMed, (3) Scopus, (4) Web of Science and (5) EMBASE. Search strategies and key terms The published literature was searched using strategies created by a medical librarian for implementation science frameworks for cancer screening and health disparities. The search strategies were established using a combination of standardised terms and keywords including, but not limited to, (colorectal cancer OR lung cancer OR breast cancer OR cervical cancer) AND (screening OR prevention OR detection OR smoking cessation OR mammogram OR pap test OR colonoscopy) AND (disparity OR low-to-middle-income country OR minority group OR inequity OR socioeconomic factors) AND (implementation science theory, model, or framework OR known IS TMFs OR evidence based medicine). The search was run on 22 June 2023 with an English language filter and a date limit of 2001–current in the databases Ovid MEDLINE, PubMed, Scopus, Web of Science, and EMBASE. A total of 9153 citations were imported into Covidence. A total of 4637 duplicates were assumed to be accurately identified and removed for a total of 4516 unique citations. Full electronic search strategies are provided in online supplemental table S1. Additional studies will also be identified by checking the bibliographies of included studies (eg, snowball searching with forward and backward citations). Stage 3: study selection All identified citations from the search strategies will be exported and managed in Covidence,29 wherein duplicates will be removed. Titles and abstracts of the remaining citation records will be screened for inclusion against the eligibility criteria. Full-text articles of records that appear to meet the eligibility criteria will be retrieved and screened for inclusion, and those that fulfil all the inclusion criteria will be included in the review. All the above steps will be conducted by two reviewers working independently at first and then meeting to compare the results and reach agreement. A pilot round will be conducted between the two reviewers with a subset (eg, 5–10%) of identified articles during the title and abstract screening stage. Following refinements, the review process will proceed as planned. If any discrepancies arise at any stage of the review, they will be arbitrated and resolved through consensus and consultation with a third reviewer. Kappa statistics will also be calculated to assess inter-rater reliability agreement for both inclusion and exclusion processes. Studies not meeting the inclusion criteria will be excluded. Once all studies have been thoroughly vetted and arbitrated, an additional pilot round will be conducted between two reviewers with 10% of included studies during the chart review (data extraction) phase. If any conflicts arise during charting, they will be arbitrated and resolved through consensus and consultation with a third reviewer. The final results will be reported in a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram and follow the PRISMA-Extension for Scoping Review (PRISMA-ScR) guidelines.30 Stage 4: charting the data Extraction of data on study characteristics and study findings will be organised into three domains: (1) study characteristics, (2) IS characteristics, (3) IS FMT-informed applications and/or adaptations and (4) cancer health disparities context. Study characteristics For each included study, the following characteristics will be collected: study design, study dates, analytical sample, type of cancer/cancer site, study location, setting of intervention and intervention description. Study design refers to the type of design employed in the study. Study dates refer to the year(s) in which data collection for the study took place (eg, 2009–2010). Analytical sample is defined as the sample size used to analyse cancer-related outcomes. Type of cancer/cancer site reflects the type of cancer(s)/cancer site(s) the screening intervention or study targeted. Study location refers to the geographic context (eg, country, state) in which the intervention took place. Setting of intervention refers to the place in which the intervention was deployed (eg, faith-based organisation, healthcare system). Study/intervention description reflects a short overview of the cancer screening EBI. For example, ‘a patient navigation colorectal cancer screening programme for African-American patients’. IS characteristics Given the focus on IS, we will extract the following information to elucidate the IS processes and outcomes of cancer screening EBIs identified in the search: (1) phase of intervention implementation, (2) implementation strategies and (3) implementation outcomes. Phase of intervention implementation refers to the temporal context in which the study was undertaken (eg, preimplementation) and will be coded with the framework from Handley et al.31 Implementation strategies defined by Proctor et al as ‘the methods or techniques used to enhance the adoption, implementation and sustainability of a clinical programme or practice’32 will be extracted and catalogued using the Expert Recommendation for Implementing Change Project inventory developed by Powell et al.33 Implementation outcomes (eg, acceptability, adoption) will be extracted using the definitions and schemes provided by Proctor and colleagues.34 Included studies will also be extracted for information on whether they used the Standards for Reporting Implementation Studies checklist.35 IS FMT applications To characterise the applications of IS FMTs (research questions 1–3, table 1), the following information will be extracted for all included studies: (1) name(s) of IS FMT used in the study; (2) FMT applications; (3) level of FMT integration and (4) barriers/facilitators to applying FMTs as stated by authors. All IS FMT names will be referenced and verified with Nilsen’s taxonomy or the D&I Models in Health Science Webtool. FMT applications will be assessed using the authors’ stated purpose(s) or objective(s) for using the IS FMTs. Coding of FMT applications will then be based on the following options: (1) inform research questions, aims and/or hypotheses; (2) guide intervention design; (3) guide intervention adaptation and (4) identification, sampling or recruitment of the focus population; (5) selection of variables (predictors, determinants and implementation outcomes); (6) measurement and analysis and (7) other (describe). For studies that used an IS FMT to guide intervention adaptation, defined as modifications or alterations made to the design or delivery of an EBI to improve its fit or effectiveness, information on the adaptation process and outcome will be extracted and codified with the expanded Framework for Modification and Adaptations.17 This framework provides a comprehensive approach for identifying rationales, the timing, contextual factors (eg, funding, policies) and actors driving the adaptations to an EBI. The level of FMT integration, defined as the extent to which a framework, model or theory, is integrated into the intervention/implementation process of a study, is captured with the following de novo scale: low (defined as <20% of constructs applied), moderate (defined as ≥20%–50% of constructs applied) and high (defined as ≥51% constructs applied). Barriers and/or facilitators to selecting and integrating IS FMTs in the identified cancer screening EBIs (research question 4, table 1) will also be captured and described based on the contextual clues and reflections provided by authors. Given the iterative nature of the scoping review process, this coding scheme will be expected to be changed in light of new evidence. Cancer health disparities context Information related to the cancer health disparities context in identified studies will draw from the rationale and objective elements of the PRISMA-Equity extension checklist.28 The data extracted on these elements will include (1) the population characteristics of the target population(s), such as gender, age, race/ethnicity, socioeconomic status; (2) the authors’ stated assumptions and rationale for addressing equity and (3) their intention to address equity as the focus of study, the research question or within the analysis. Information on these characteristics will be pulled and triangulated from various sections (eg, introduction, methods, limitations conclusion) within the publication of the included study, and narratively described in the scoping review. Given the varied places where equity intentions and assumptions may be described throughout a publication, the data extraction process for capturing the cancer health disparities context may be later refined based on earlier results. Stage 5: collating, summarising and reporting the results The study characteristics of all three domains will be collected and managed with an ad hoc template. This template will be created by the research team through consultations and will operate as an Excel spreadsheet. It should be noted that the template may undergo refinements in the future based on further consultations. Whenever necessary and possible, the authors of included studies will be contacted to obtain any missing data. The findings will be organised according to the three domains of data extraction elements. These data elements have been designed to address the key objectives of the scoping review. The reporting of all study characteristics will be presented in a narrative format using the methodology proposed by Popay et al,36 and where appropriate, with a numerical summary. For instance, we may report the percentage of studies conducted within various settings (eg, healthcare setting vs community-based setting). Due to the potential variations among studies conducted in low-income countries, lower-middle-income countries, upper-middle-income countries and high-income countries, the results will be categorised according to the World Bank’s analytical income classification groups to characterise differences at both the micro (within countries) and macro (across countries) levels.37 Whenever feasible, the findings will be presented in tables or concept maps to enhance their visual interpretation. The final conduct of the scoping review will adhere to the PRISMA-ScR guidelines.30 Patient and public involvement No patients’ or public input was sought in the development and conceptualisation of this scoping review. Ethics and dissemination As a secondary analysis, this scoping review does not require approval from an institutional review board. The results of the review will be shared through two channels. First, as publications in peer-reviewed journals. It is anticipated that there will be two separate publications—one focusing on all types of cancer screening and another specifically on cervical cancer. Additionally, the findings will be presented at national and international conferences. Supplementary Material Reviewer comments Author's manuscript