Development, Implementation and Evaluation of an Individual Decision Aid in Swedish Cancer Screening Programs

INTRODUCTION AND RATIONALE For cancer screening programs to be effective, a high willingness from seemingly healthy individuals to participate is a prerequisite. Participation rates differ and are relatively high in some population groups but at the same time low in other groups. It is well known that individuals with lower socioeconomic status, ethnic minorities and individuals with disabilities participate to a limited extent in cancer screening. A decision aid (DA) is set to deliver information about different health care options and to help individuals make their values visible connected to the options available. Using a DA in connection to cancer screening can increase participation based on knowledge rather than misconception or fear.

The present project is based on the theoretical framework of shared decision making (SDM), with the overarching goal to improve the quality of health care decisions. SDM in this project is based on the work of Charles and Gafni with three cornerstones important for the decision-making process: information/knowledge - to make an informed decision persons need information to gain knowledge about e.g., pros and cons; Values/preferences - persons process their options in relation to attitudes and behaviors; and Involvement - of both the individual and health care professionals.

To date, no DA for any of the screening programs exist in Sweden why this project is warranted. By designing the communication strategies in various ways, regarding information about screening options, and to help individuals construct, clarify and communicate personal values, the researchers intention is to support the decision-making process and to address and acknowledge the equity perspective.

METHODS

Phase 1

Design This study has an evaluative approach encompassing both a process-, an implementation and an outcome evaluation. Multiple methods will be used to answer the research questions including patient reported data, focus group discussions and individual interviews using the think aloud technique. The project is based on the framework from IPDAS and the proposed model development process for DAs, as presented by Coulter et al.

Procedure and outcomes

The procedure will follow the model development process for DAs, applying the following steps:

1. Define scope - a description of breast, cervix and bowel cancer including screening will be formulated, information will be gathered regarding treatment, screening tests, true positive true negative results, risks, and benefits with screening vs. non-screening and detection probabilities. The target audience is all individuals invited to cancer screening.
2. Steering group - a multidisciplinary group will be formed with stakeholders: lay people (aged 23-74), different socioeconomic status and ethnicity, from urban and rural areas, invited to and not invited to screening); clinical experts (oncologists, endoscopists, gastroenterologist, gynecologists, registered nurses, psychologists) and other experts (in psychometry and IT); collaboration partners (The Swedish Association of Local Authorities and Regions - SKR, Regional Cancer Centers - RCC, National Board of Health and Welfare). No disclosed declaration of interest.
3. Design - provision of the iDA in paper format, i.e., a version with the tentative content, including the research and evidence that the iDA is based on as well as a description of the prototype development. The design of the paper version is inspired by the work of Schwartz and colleagues and will be evaluated by lay persons and experts for its content and comprehensibility regarding format (information and descriptions in text and images) and language use.
4. Alpha testing - the web-based iDA will be evaluated for its content, comprehensibility, usability and feasibility, regarding format (information and descriptions in text, spoken animations, images, audio, animated videos, certain clickable words to provide additional information, and interactive questions to help individuals to clarify and express values); language use; setting (web-based) and timing (open for anyone, link will be provided with the invitation).
5. Beta testing - the iDA will be tested in a "real world setting" and evaluated regarding format (information and descriptions in text, spoken animations, images, audio, animated videos, certain clickable words to provide additional information, and interactive questions to help individuals to clarify and express values); language use; setting (web-based) and timing (open for anyone, link will be provided with the invitation).

Data collection During the steps 3-5 data will be collected using different qualitative methods. Data regarding content, format, language use, setting and timing will be generated from focus group discussion and individual interviews. An interview guide with opened-ended questions and optional probe questions will guide the data collection. Participants will be able to choose to participate in a focus group discussion or an individual interview.

During step 3, data regarding the questions, including interactive questions (values and preferences, knowledge, and lifestyle exercises) (step 4) will be generated from concurrent cognitive interviewing where the individual gives a verbal account of their thinking while responding to the interactive questions included in the iDA. The interviews will follow a "think aloud" protocol where individuals are asked to respond to the interactive questions and encouraged to think out loud and to verbalize their thoughts while doing so. This gives an understanding on the perception of each question, as the participants work themselves through the content. In addition, probing questions will be asked when changes in appearance, such as frowning or hesitation occurs and when, or if, the session leader needs further clarification.

Data management Demographic data will be pseudo anonymized, meaning that all personal names will be removed during transcription and all participants in the studies will be given a unique code. A code list will not be established since no personal data is handled. All data will be storage in a secured facility at the Division of Nursing, Karolinska Institutet according to regulations at the university.

Data analyses Data generated from focus group discussions and individual interviews will be analyzed using descriptive and manifest content analysis. Data generated from cognitive interviews applying the "think aloud" method will be compiled and organized according to responses.

Phase 2

Procedure and outcomes After the beta-testing the iDA will be public and accessible for all who are invited and/or interested in the cancer screening programs. To be able to evaluate and develop the iDA further, topics such as digital health literacy, knowledge, values and preferences, decisional conflict, concerns, and risk awareness will be studied with self-reported questions. In addition, demographic questions such as sex, age, educational level, employment will be included as well as questions regarding lifestyle and physical activity. Usability of the iDA including the chat function, helpline, and behavior flow, i.e., how individuals navigate in the iDA will also be scrutinized.

Data collection Data will be collected through self-reported questions included in the iDA.

Data management Questionnaire data will be stored in the web-platform owned by RCC. Extracted data will be pseudo anonymized, meaning that all proper names will be removed during transcription and all participants in the studies will be given a unique code. A code list will not be established since no personal number will be handled. All data will be storage in a secured facility at the Division of Nursing, Karolinska Institutet according to regulations at the university.

Data analyses Data will be analyzed and compared using relevant descriptive and inferential statistics.

First results from the alpha- and beta phases are published, please see reference list.