Development and use of research vignettes to collect qualitative data from healthcare professionals: a scoping review.

OBJECTIVES: To clarify the definition of vignette-based methodology in qualitative research and to identify key elements underpinning its development and utilisation in qualitative empirical studies involving healthcare professionals.
DESIGN: Scoping review according to the Joanna Briggs Institute framework and Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines. DATA SOURCES: Electronic databases: Academic Search Complete, CINAHL Plus, MEDLINE, PsycINFO and SocINDEX (January 2000-December 2020). ELIGIBILITY CRITERIA: Empirical studies in English or French with a qualitative design including an explicit methodological description of the development and/or use of vignettes to collect qualitative data from healthcare professionals. Titles and abstracts were screened, and full text was reviewed by pairs of researchers according to inclusion/exclusion criteria. DATA EXTRACTION AND SYNTHESIS: Data extraction included study characteristics, definition, development and utilisation of a vignette, as well as strengths, limitations and recommendations from authors of the included articles. Systematic qualitative thematic analysis was performed, followed by data matrices to display the findings according to the scoping review questions.
RESULTS: Ten articles were included. An explicit definition of vignettes was provided in only half the studies. Variations of the development process (steps, expert consultation and pretesting), data collection and analysis demonstrate opportunities for improvement in rigour and transparency of the whole research process. Most studies failed to address quality criteria of the wider qualitative design and to discuss study limitations.
CONCLUSIONS: Vignette-based studies in qualitative research appear promising to deepen our understanding of sensitive and challenging situations lived by healthcare professionals. However, vignettes require conceptual clarification and robust methodological guidance so that researchers can systematically plan their study. Focusing on quality criteria of qualitative design can produce stronger evidence around measures that may help healthcare professionals reflect on and learn to cope with adversity. © Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

To our knowledge, this is the first scoping review to focus on methodological issues regarding the definition, development and utilisation of vignette-based methodology to collect qualitative data from healthcare professionals.

Our study provides a broad overview of how vignette-based methodology has been used in qualitative studies involving healthcare professionals over the last two decades.

The review process follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guideline universally recognised to improve the uptake of research findings.

Although our content analysis considers quality criteria, in line with recommendations for the conduct of scoping reviews, we do not systematically appraise included studies.

Relevant studies may have been excluded in our three-step screening process, as titles and abstracts do not always specify whether the vignette is used when conducting qualitative research.

Introduction

Vignettes are commonly referred to as short hypothetical accounts reflecting real-world situations. Vignettes are presented to knowledgeable individuals who are invited to respond.1 Generally speaking, vignettes allow participants to clarify and share their perceptions on sensitive topics such as dealing with adversity in challenging environments, discussing team functioning issues or moral dilemmas they face daily, and reflect on potential solutions. Vignette-based methodology in qualitative research appears useful to our research team, which is currently piloting an intervention to co-construct, implement and assess resilience at work among cancer teams, as a means of integrating the knowledge of cancer professionals on how to face adversity. The objective of the scoping review is to learn from prior use of vignette-based methodology in qualitative research in healthcare settings.

Team resilience at work refers to the capacity of team members to face and adapt to adverse situations.2 Cancer care offers a valuable clinical context to study team resilience at work because professionals face daily adversity with overlapping challenges such as delivering news of a new cancer diagnosis or disease progression, constant changes in therapeutic regimens, frequent staff turn-over and shortages, and increased administrative tasks.3–7 Cancer team members are exposed to mental health threats such as high stress, anxiety, compassion fatigue and loss of a sense of coherence8 associated with absenteeism, burnout or depression.4 5 9–12 While these negative effects of adversity have grown exponentially with each wave of the COVID-19 pandemic,13 14 solutions to manage and minimise these effects remain understudied. Cancer team members must manage and learn from difficult situations related to their practice context and the pandemic environment. The vignette-based methodology provides an opportunity to reflect and plan supportive interventions and offers an empirically based research approach that is well suited to this complex context.

Vignette-based methodology in qualitative research explores and interprets contextualised phenomena to identify influential factors and understand how participants perceive moral issues or sensitive experiences.15 It also enables reflexive learning from practice, stimulates exchange on professional responses to difficult situations and supports tailored actions to make sense of adversity. Vignette-based methodology is of interest in disciplines such as psychology, social science, education, medicine and nursing.16–20 It has been developed and used to collect data on perceptions, beliefs, attitudes and knowledge,17 19 from individuals or teams,19 21 through individual or group interviews or questionnaires.15 18 21 Commonly formatted as written narratives, vignettes can also be presented as audio segments, photographs or videos.18 21

Empirical studies use different definitions of the vignette and provide little detail about how it is developed and used to collect data.15 19 21 Such methodological inconsistencies raise questions about the quality criteria of this qualitative approach.17 Concerns have also been expressed around whether data collection approaches ensure an appropriate distance between the occurrence of sensitive events and the interview19 and around the need to mitigate the risk that participants provide socially desirable responses.15 Finally, our preliminary search for studies using vignette-based methodology to collect qualitative data from professionals in cancer care found only one study.22 These factors emphasise the need to arrive at a working definition of this approach to inform data collection in subsequent qualitative studies and provide the rationale for this scoping review.23 24

This study aims to clarify the definition of vignette-based methodology in qualitative research and to identify key elements underpinning its development and utilisation in empirical studies involving healthcare professionals.

Methods

This scoping review mobilises the Joanna Briggs Institute’s methodological guidelines,23 which build on the seminal works of Arksey and O’Malley25 and Levac et al.26 Scoping reviews examine the number, range and nature of studies relevant to a particular research question and are used to analyse and report available evidence.27 The present scoping review follows the steps described by Peters et al.23 The Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA-ScR) checklist criteria24 are followed to report results (online supplemental appendix 1). The protocol was registered prospectively with the Open Science Framework on 1 July 2020 (https://osf.io/muz4x/?view_only=5943aa0ffb6541d6979ebeedba7464cb).

Patient and public involvement

No patients or public involved in carrying out this scoping review.

Scoping review questions

The questions of the scoping review have a methodological focus: (1) how has vignette-based methodology in qualitative research been defined?; (2) what steps have been involved in developing vignettes to collect qualitative data in studies involving healthcare professionals?; and (3) how is vignette-based methodology used to collect qualitative data from healthcare professionals?

Planned approach

The Population/participants, Concept and Context (PCC) framework, with the addition of the type of evidence source (type of study and type of publication), is used to guide the selection of eligibility criteria and the search strategy.23 28 PCC generally allows a wide range of articles to be considered for inclusion. The concept of interest is the vignette as used in qualitative research. A preliminary search of qualitative vignette-based methodology development and utilisation with cancer team members found only one study. Therefore, the search was expanded to include qualitative studies as well as systematic and scoping reviews (type of evidence source) in healthcare contexts other than oncology (context), with healthcare professionals in both practice and educational settings (population/participants).

Eligibility criteria

Inclusion criteria were: (A) empirical studies with specific focus and/or statements about the development or utilisation of vignettes in qualitative studies involving healthcare professionals in clinical practice, training or continuing education; (B) qualitative study design (action research, intervention research with clinical or educational application and professional practice-based initiatives); (C) written in English or French; and (D) published between January 2000 and December 2020 in journals listed in electronic databases. The search was limited to 2000 due to the very small number of publications prior to that year using vignettes in qualitative research involving healthcare professionals. Exclusion criteria were: (A) absence of the word ‘vignette’ in title, in order to target studies with a clear focus on methodological development or use in qualitative research; (B) background articles or other articles that did not report outcomes from use of vignettes in qualitative data collection; (C) studies using vignette with quantitative or mixed methods design; (D) studies reported in grey literature; and (E) articles without an abstract.

Search strategy

Research team members including researchers and professionals from various disciplines (eg, nursing, psychology, economics, human resources management and medicine) were involved in search strategy preplanning. An academic librarian contributed to determining the databases, search terms, boolean operators and query modifiers (online supplemental appendix 2). A total of five peer-reviewed online databases were searched: Academic Search Complete, CINAHL Plus, MEDLINE, PsycINFO and SocINDEX. The search was supplemented by hand-searching reference lists.

Source of evidence screening and selection

Articles were uploaded to Rayyan, a cloud-based application for systematic reviews.29 Duplicates were removed before undertaking the three-step screening process30: title, abstract and full-text assessment. Two reviewers (DT and AT) independently completed each screening step.31 Disagreements on article selection and on reasons for exclusion were resolved by consensus through discussion between the two reviewers and two other team members (SL and EG). Reviewers selected and applied the highest reason for exclusion from a screening criteria priority list, which was agreed on ahead of time.

Data extraction and analysis

Data extraction was performed in two cycles, according to Peters et al’s recommendations on key information to extract.23 The first cycle aimed to describe study characteristics (eg, authors, country and year of publication, study phenomenon and setting). The second cycle was based on a thematic analysis for data condensation.32 The coding grid aligned with our review questions: vignette definition; vignette development (steps described, actors involved/developers, source and format of vignette content); vignette utilisation (study participants, delivery method, introduction items, vignette presentation and handling, interview process, design and strategy for data analysis); and strengths and limitations relating to vignette development or utilisation, advantages or disadvantages of using the vignette and recommendations reported by authors. The coding approach was defined by consensus between research team members (DT, AT, SL and EG). Data extraction was performed using QDA Miner (V.5.0.34).33

A thematic analysis on the development and utilisation of vignettes, as well as recommendations from authors that emerged from the reviewed articles, were synthesised in charting tables. Several research team meetings were carried out during the iterative data extraction and analysis process. Data matrices were used to display the findings according to the scoping review questions.

Results

Search results

The removal of duplicates and the addition of one record from hand-searching left 157 potentially eligible articles. Screening by title excluded 127 articles, while screening of abstracts excluded 14 more. Full-text assessment excluded an additional six articles. The main reasons for exclusion were wrong concept (not vignette-based methodology in qualitative research) and wrong population (not healthcare professionals). A total of 10 articles were eligible for inclusion in the review. Search results are presented in a flow diagram34 (figure 1).

Figure 1

PRISMA flow diagram of article selection process. Adapted from: Page et al.34 PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Characteristics of included studies

Included studies are published between 2002 and 2020 and involve healthcare professionals from four countries: Australia,35 Canada,22 36 Norway37 and the UK.38–43 Study settings include oncology, primary care, mental health, public health, hospital care, health and social work, health education and critical care. Various phenomena are investigated, such as quality of care related to professional practices, understanding of policy issues, appreciation of health services, perceptions towards patients and moral or ethical issues. These characteristics are included in tables in the next sections.

Vignette-based methodology in qualitative research

The first question in this review concerns how studies define the vignette-based methodology in qualitative research. While a definition is missing in two articles,40 41 four articles22 36 38 39 provide an original definition informed by one or more key references. For example, Morrison (p. 362)36 defines vignettes as ‘carefully designed short stories about a specific scenario presented to informants to prompt discussion related to their perceptions, beliefs, and attitudes’. The other four articles refer to key authors without giving an explicit definition.35 37 42 43

Vignettes are referred to as short stories about hypothetical characters in specified circumstances that participants are invited to respond to.35 36 38 42 43 Other elements specified in definitions include the form of the vignette (eg, text),39 the nature of the stories or scenarios (eg, simulations of real events, fictional or composite)38 43 or the aim of the vignette (eg, to elicit individuals’ perceptions, attitudes, beliefs and social norms).36 38

Methodological development of vignettes for qualitative research

The second question of interest pertains to the methodological steps involved in developing a vignette to collect qualitative data from healthcare professionals. Table 1 presents a description of the vignettes in each study, the extent to which development steps are reported, as well as the steps and actors involved in vignette development.

Table 1

Description of vignette development in included studies

Study	Vignette	Number of steps*	Development steps with actors involved
			Content(based on)	Format	Choice of approach	Interview questions	Preliminary versions	Anticipated responses	External validation/review	Pretest	Final version
Andrews et al, 202039UKPrimary care –self-monitoring of blood pressure	Six short sections on multiple points of care	M	R (S)	W	–	R	–	–	–	R, E	R
Cazale et al, 200622CanadaOncology –professional practices in cancer care	Clinical vignette, sequence of four events from the care coordination of a cancer patient	6	R (Li)	W	R	–	R, E	R	–	R, A	R
Holley and Gillard, 201838UKMental health –understandings of risk and recovery	Five sequential scenarios on issues of living in the community with serious mental illness	2	R, A (Li, S)	W	–	R	R	R	–	R, A	R
Jackson et al, 201535AustraliaPublic health – promotion of unhealthy foods and beverages	10 scenarios of marketing practices of a fictional multinational confectionery company	8	R (Li)	W	R	–	R	–	R, E	R, A	R
Johnson et al, 200540UKHospital and primary care –role of advice in diabetes foot care	Continuous story in six stages of a patient with diabetes-related foot complications	DD	R (Li)	W	R	R	R	–	–	R, E	R
Morrison, 201536CanadaOncology –support in cancer survivors’ work integration	Seven combinations of photographs and narratives, reflective of cancer survivors’ experiences of work integration	DD	R (S)	P, W	–	R	R	–	–	R	R
Østby and Bjørkly, 201137NorwayHealth and social work –ethical challenges in interactions	Four short, open-ended descriptions of interactions between people with intellectual disabilities and care staff	6	R (S)	W	–	R	R	–	R, A	–	R
Richman and Mercer, 200242UKPsychiatric hospital –discursive structures of nurses	12 short scenarios detailing case histories of a high-risk patient (six white/six black)	M	R (Li)	W	R	–	–	–	–	–	R
Spalding and Phillips, 200743UKHealth education –preoperative education practice	One snapshot, 20 portraits and one composite, within an action research to improve preoperative education	DD	R (S)	W	R	–	R	–	R, E	–	R
Thompson et al, 200341UKCritical care –adherence to advance directives	One clinical vignette of a fictitious patient who had signed an advance directive before developing dementia	M	R (–)	W	–	R	–	–	–	–	R

*, number if clearly stated; –, not reported; A, targeted audience; DD, diffusely discussed; E, experts; Li, literature, including knowledge from reviews, existing frameworks or guidelines; M, minimally or not discussed; P, photographs; R, researcher(s); S, empirical study conducted; W, written.

Vignettes are designed as stories,40 scenarios,35 38 42 43 clinical situations emerging along the cancer trajectory22 or descriptions of a plausible individual or social situation.36 37 39 41 Including 1–20 situations, they are presented in written narrative form in all studies but one, which combines narratives and photographs.36 Three studies use temporally sequenced vignettes.22 38 40 To emphasise the plausibility of the content, six articles mention the source of inspiration: real-life clinical situations or patient experiences,22 36 39 41 observational research43 or situations involving ethical challenges seen in field study.37

The steps used to develop the vignette are clearly described in four studies. In the other studies, authors are either vague about the steps36 40 43 or provide minimal to no information.39 41 42 Although the number of steps ranges from 2 to 8, with various degrees of specification, design and pretesting appear as the most common steps to arrive at the version of the research vignette delivered in interviews. Other steps involve establishing the vignette content and format and choosing a delivery approach (eg, individual or group interview). Drawn either from literature (eg, knowledge from reviews, existing frameworks or guidelines) or from empirical studies, the content is either developed by researchers, sometimes with input from clinical experts22 or exploratory focus groups of individuals similar to research participants.38

Strategies are described to improve the internal validity of vignettes (relevance, reliability, effectiveness, completeness, familiarity and intelligibility). Three studies stress the importance of reviewing vignette content, conducting a survey with respondents similar to the targeted audience37 or obtaining feedback from experts.35 43 Vignettes are pretested in six studies, through piloting with experts39 40 or individuals35 or through group discussion22 38; one study mentions testing the vignettes and interview protocol without providing further detail.36 Other strategies to improve internal validity include: use of a panel of experts,38–40 43 use of primary research data36–39 or framework22 to develop the content; removal of elements from the vignettes that may bias the interviews37; and selecting a small number of scenarios (up to four) to be included in the vignette.37

Strategies to increase generalisability include making the vignettes realistic36 37 43 and comparing pretest responses from experts with responses anticipated by the research team.22 Researchers22 35 37 38 40 43 also mention making changes to content, format or delivery method as needed throughout validation and/or pretesting steps to assure internal and external validity.

Utilisation of vignette-based methodology in qualitative research

The third question we explore in the review is how vignette-based methodology is used to collect qualitative data from healthcare professionals (table 2).

Table 2

Description of vignette-based methodology utilisation in included studies

Study	Participants	Delivery approach	Introduction	Presentation / Handling	Interview process	Design and data analysis
Andrews et al 202039UKPrimary care –self-monitoring of blood pressure	Physicians (n=14); nurses (n=7)Total (n=21)	Focus groups (n=5). 2–8 per group. 1 hour.	Not reported.	Each vignette read out by researcher.	Semistructured. Interview guide. One question on vignette with 2–5 follow-up questions on participants’ experiences.	Thematic analysis. Transcribed verbatim. Field notes. Validation by three researchers.
Cazale et al 200622CanadaOncology –professional practices in cancer care	Interdisciplinary teams of clinicians in oncology.Total (n=41)	Focus groups (n=5). 5–14 per group. 1 hour.	Study objectives. Ground rules.	Each event presented by expert consultant. Sequential.	Semistructured. One open-ended question per event on participants’ own actual practices. Low control/high process style of moderation.	Coding base: cancer programme guidelines. Transcribed verbatim. Field notes. Intercoder reliability assessment by two researchers.
Holley and Gillard, 201838UKMental health –understandings of risk and recovery	Psychiatrists, mental health professionals(n=8); service users (n=8)Total (n=16)	Individual interviews.	Participants’ demographics.	Each vignette presented by researcher. Sequential.	Interview guide. Open-ended questions (n=not reported) on participants’ thoughts about the vignettes and their own experiences in similar circumstances.	Thematic analysis. Transcribed verbatim.
Jackson et al 201535AustraliaPublic health – promotion of unhealthy foods and beverages	Public health professionals (n=10); marketing and industry professionals (n=11)Total (n=21)	Individual interviews. In person or by phone.	Ground rules.	Email prior to phone interview. Each scenario read by participant or researcher. One by one.	Open discussion on perceived challenges, threats and opportunities, drawing on professional background, opinions or experiences. Prompts to further explore threats or challenges.	Hermeneutic analysis. Transcribed verbatim. Field notes. Research journal.
Johnson et al 200540UKHospital and primary care – role of advice in diabetes foot care	Healthcare professionals, consultants, physicians and specialists (n=15);patients (n=15)Total (n=30)	Individual interviews.	Study objectives. Ground rules.	Each stage presented visually and verbally by researcher. Sequential.	Interview guide. 1–2 open-ended questions per sequence, on participants’ views about services to patients. Participant’s own issues discussed at the end.	Framework analysis with coding. Transcribed verbatim.
Morrison, 201536CanadaOncology –support in cancer survivors’ work integration	Oncologists (n=5);physicians (n=5)Total (n=10)	Individual interviews. 1–1.25 hours.	Participants’ demographics.	Stack of vignettes evidently placed. Each read and kept by participant until taken by researcher. One by one.	Semistructured. Interview guide. Open discussion on perspectives, beliefs, attitudes and behaviours.	Interpretive description. Transcribed verbatim.
Østby and Bjørkly, 201137NorwayHealth and social work –ethical challenges in interactions	Social educators Total (n=8)	Individual interviews.	Ground rules.	One by one.	Interview guide. 2 sets of 3 questions with three follow-up subquestions: first set on participant’s reflections and actions; second set on views of how others would have reflected on or behaved. Additional question to assess vignette familiarity and relevance.	Not reported.
Richman and Mercer, 200242UKPsychiatric hospital –discursive structures of nurses	Clinical nursesTotal (n=30)	Individual interviews. 0.75–2 hours.	Not reported.	Vignettes selected and read by participant.	Open discussion on participants’ own practice experiences, emotional reactions and larger cultural and media representations.	Not reported.
Spalding and Phillips, 200743UKHealth education –preoperative education practice	Healthcare professionals also presenters of education programme.Total (n=not reported)	Team meetings.	Not reported.	Each vignette read by participant.	Open discussion on participants’ perceptions, beliefs and meanings.	Not reported.
Thompson et al 200341UKCritical care –adherence to advance directives	Healthcare professionals and specialists from various disciplines.Total (n=46)	Individual interviews (n=12). Focus groups (n=6). 4–9 per group.	Not reported.	Critical care vignette shown by researcher.	One planned open-ended question, about the right thing to do.	Modified grounded theory. Coding base: topic guide. Transcribed verbatim. Independent coding validation by three researchers.

Studies employ convenience37 or purposive35 36 38 39 41 sampling to determine inclusion and exclusion criteria for participants. Sociodemographics (age, gender or sex and years of experience) are reported in three studies,37 39 41 while participants’ profession is reported in all studies.

Vignettes are delivered through individual interviews in seven studies.35–38 40–42 The number of individuals varies from 8 to 30. Four studies present the vignettes in group interviews22 39 41 or team meetings43 of 2–14 participants. Johnson et al40 consider that individual interviews are best suited to explore professionals’ personal views, for logistical reasons and to reduce the risk of inhibiting expression due to power differentials between participants. In contrast, Cazale et al22 use focus groups to observe the interaction between participants, which seems promising to generate data in their study aimed at assessing the quality of care provided by interdisciplinary teams. One study41 uses both individual and group interviews, without explicit justification.

Six studies report that researchers introduced study objectives to participants, explained ground rules such as confidentiality, the interview procedure and assured them there were no right or wrong answers. This is similar to other qualitative methods.

Various interviewing approaches are adopted in the studies: open discussion, semistructured or structured. Interview guides are used in five studies.36–40 All studies include questions about the participants’ perceptions, views or beliefs regarding their own experiences or practices. One study includes questions to elicit participants’ thoughts on whether the vignette content reflects their personal experience (plausibility).38 Another adds questions on how others may have interpreted or behaved in a similar situation, which helps verify that the vignettes describe real-life practice situations and thus contributes to establishing their validity.37

Some note that the method is generally well received by participants,35 36 despite two health professionals who ‘opined that the vignettes were unnecessary to facilitate the dialogue that could have been accomplished by direct questioning’ (p. 369).36 Certain issues are also reported regarding the quality of the answers elicited (eg, answers from own perspective instead of others’; answers to avoid disclosing confidential or problematic information; answers tailored to social desirability).35 37 38

Various qualitative design and data analysis approaches are employed, including thematic analysis of interview responses, hermeneutic analysis, framework analysis, interpretive description or modified grounded theory. Only three studies include information on reliability assessment using content validation by experts, pretest or interview modalities.22 39 41

Synthesis of recommendations from included studies

A synthesis of the recommendations on vignette development and utilisation is presented in table 3. These are based on analysis of the strengths and limitations reported in the 10 studies included in this scoping review.

Table 3

Synthesis of strengths (S), limitations (L) and authors’ recommendations in included studies

Study	Vignette development	Vignette utilisation
Andrews et al 202039UKPrimary care –self-monitoring of blood pressure	Primary data (eg, excerpts from interviews) to provide authenticity to the study materials (S).	Coding theme validation by multiple researchers (S). Participant heterogeneity for larger perspective (L).
Cazale et al 200622CanadaOncology –professional practices in cancer care	Explicit development process (S). Solid framework for development and analysis (S). Involvement of experts (S). Content in descriptive tone to avoid socially desirable responses (S). Avoidance of information overload in vignette (S).	Utilisation to support learning and reflexivity (S). Skilled facilitator such as external expert (S). Support from assistant facilitators (S). Triangulation using multiple data sources (L). Standardised data collection if multi-site study (L).
Holley and Gillard, 201838UKMental health –understandings of risk and recovery	Exploratory focus groups to identify content (primary data), for vignette validity (S). Respondent validity check through feedback focus groups with experts (S). Prompts on own experiences, as questions on vignette may attract abstract or idealised responses (S). Content based on sufficient and solid sources to allow validation of vignette (L). Clear sociodemographic aspects (gender, ethnicity, etc) in content and when sampling participants, to explore whether vignettes might elicit data that respond to issues of marginalisation (L). Clear definition of concepts used (L). Presentation of realistic information (L). Interview guides that allow to explore a full range of possible responses (L).	Vignette elicited data on the complexities of the participants’ roles while addressing their own responsibilities (S).
Jackson et al 2015AustraliaPublic health – promotion of unhealthy foods and beverages	Amount of scenarios and range of concepts (variables) to explore within time available (L). Scenarios that generate a response but are not too extreme (L).	Utilisation as natural set of parameters for interview discussions, while allowing deeper investigation (S). Consideration for how participants approach the vignettes (eg, real-life; microlevel or macrolevel) and how that may lead to socially desirable/guarded responses (S). Interviewer skills to refocus (S). Peer-debriefing with research team (S). Triangulation using various analysis methods (S). Prolonged engagement with data (S). Consistency of vignette utilisation (same variables) between research populations for data comparison (S).
Johnson et al 200540UKHospital and primary care –role of advice in diabetes foot care	Test with expert panel and pilot to increase internal validity (S). Wrap-up question at the end of the interview (S).	Consistency of vignette utilisation between research populations to allow data comparison (S). Recognition of difference between potential behaviour of fictitious character in vignette and actual experiences of the participant (S).
Morrison, 201536CanadaOncology –support in cancer survivors’ work integration	Content that provides a fair representation of the topic (reality, gravity) (S). Consideration for the time available for participation (S). Consideration for the interview questioning format: in third person to create safe distance; consistency in format used (L). Consideration for number of vignettes (eg, less than seven) (L).	Utilisation to invoke self-reflection (S). Reaching saturation (S). Interviewing skills (L). Consideration for busy participants (time, distractions) (L).
Østby and Bjørkly, 201137NorwayHealth and social work –ethical challenges in interactions	Removal of content that can lead to interpretations and choices (S). Validation procedure to increase internal validity (S). Questions and sub-questions designed to reduce socially desirable responses (S). Questions to improve validity: situation perceived as familiar; own stories about similar situations; ask why? (S). Triangulation (eg, with quantitative measures) for further validation (L).	Validated vignettes for enhanced reflections (S). Reach of saturation (S).
Richman and Mercer, 200242UKPsychiatric hospital –discursive structures of nurses	Decisions about: data for content (existing or constructed data), temporality (static or serial), degree of specialised information (specialised or everyday activities); aims of the project (analytical or prescriptive); medium (written, filmed or oral); role (to test or to generate hypothesis).	Utilisation as a prompt to reflect on personal experiences (S).
Spalding and Phillips, 200743UKHealth education – preoperative education practice	Primary data to develop vignettes that are meaningful, contextualised and reflect reality (S).	Utilisation to facilitate reflection within an action research cycle (S).
Thompson et al 200341UKCritical care –adherence to advance directives	None relating to development.	Effective stimulus for discussion (S). Utilisation to highlight the gap between knowledge and action (S). Caution about how vignette reflects the multifactorial arena of decision making in real world (L). Verification of understanding of terminology used (L).

Researchers in all the studies report that vignette-based methodology in qualitative research is an effective means of exploring sensitive or difficult topics and eliciting in-depth responses and reflexivity.

Eight authors’ recommendations emerge from our scoping review around the methodology for development of vignettes in qualitative research: (1) follow a rigorous stepwise development process22 42; (2) involve experts who are knowledgeable informants or a multidisciplinary team in refining content22 38; (3) use credible sources such as primary research data, frameworks or literature reviews to develop content22 38 39 43; (4) be mindful of participants’ availability when determining the number of sections or vignettes35 36; (5) avoid content that uses unclear terminology,38 lacks information (eg, not the full clinical picture),38 includes too many variables22 35 or leads to particular interpretations or choices22 37; (6) provide vignettes that are meaningful and allow participants to identify with and reflect on the story36 38 43; (7) use validation strategies and test the quality of the vignette37 40; and (8) pay attention to the delivery, including semistructured interview questions and form of probing36–38 (eg, a third person format can help create safe distance to explore difficult topics36; consistency in the format: mixing second and third person questions can lead participants to answer most questions based on their personal experience).36

Our scoping review further suggests a number of recommendations regarding the utilisation of vignette-based methodology: (1) use the vignette consistently with each participant or group of participants to allow systematic data collection22 35 40; (2) make sure the interviewer has the skills to conduct individual or group interviews22 35 36; (3) recognise and try to discourage socially desirable responses35; (4) be cautious about the extent to which it reflects real-world situations for the participants35 40 41; (5) add one facilitator and one observer during focus groups22; (6) reach saturation in data collection36 37; and (7) use validation strategies in data analysis (eg, intercoder reliability assessment; theme validation)39 and triangulation to reinforce the quality of results.22 35

Discussion

This scoping review contributes to clarify the definition of vignette-based methodology in qualitative research, details its development steps, describes its utilisation and assesses its strengths and limitations based on quality criteria for qualitative studies. It can inform planning of future research employing this qualitative approach. Ten studies are included that involve healthcare professionals in various settings.

Main findings

Our results suggest an expanded use of the vignette as a qualitative methodology. Vignette-based methodology is not commonly used in qualitative studies involving healthcare professionals, despite being recognised as a suitable approach for ‘reflecting-on’ and ‘reflecting-in’ practice.44 The methodology is well suited to intervention research, establishing partnership between knowledgeable actors from the field and researchers to define a problem and potential solutions.45

During the article-screening process, 112 out of 156 articles were excluded due to ‘wrong concept’ (71,7%); that is, they did not address or use vignette-based methodology in qualitative research (see figure 1). One contributing factor to the high exclusion rate is that many articles used the term ‘vignette’ without defining the term. Vignettes are used in the scientific literature in various ways (clinical case reports, training materials, evaluations of clinician knowledge, etc). Our review findings reveal the need to clearly state ‘what’ is vignette-based methodology in qualitative research and describe the logic of its use by researchers.

Vignettes can be used to describe a phenomenon in multiple contexts that are different from qualitative research. We acknowledge that variation may be appropriate across vignette utilisation. However, in qualitative studies, a number of basic principles are considered necessary to assure reliability of analysis: explicit description of the study context and procedures used in data collection and analysis to produce knowledge.32 Our scoping review shows that vignette-based qualitative research studies often fail to fully describe how these three principles are met. This points to a lack of engagement with standards for reporting qualitative research46 and compromises replicability and the utilisation of knowledge arising from vignette-based studies. Finally, standards for reporting qualitative research suggest that the title indicates that the study is qualitative or include a commonly used term that identifies the approach.47

In sum, an article title that states the research method and a clear definition of ‘vignette’ in the report contribute to aligning the research objectives, study design and methods. They allow readers and reviewers to understand the type of vignette study at hand and support the reliability, transferability and usefulness of results.48

Despite the efforts of authors to clarify the concept, less than half the studies included in our review provide an explicit definition. Based on our scoping review, the vignette-based methodology in qualitative research can be defined as evidence- and practice-informed short stories, scenarios, events or situations in specified circumstances, to which individuals or groups are invited to respond.1 22 36 39

Details of vignette development are only scarcely reported. Less than half of the studies explicitly report all steps in development. The range of development steps reflects the lack of standardised quality criteria for reporting vignette-based methodology in qualitative research. Greater transparency is needed to establish internal validity and enable study replication, notably around knowledgeable informant involvement in establishing vignette content and/or participating in validation steps.

Our results highlight that vignettes are delivered through individual interviews in most studies, but that some researchers opt for or add group interviews to meet their study objectives. The choice may depend on whether the study seeks to elicit personal views or interaction between participants. However, the choice of interview approach is not always explained.

Our synthesis of strengths, limitations and authors’ recommendations in included articles (see table 3) provides an overview of what vignette-based methodology adds to the studies. Some advantages highlighted in included articles are not specific to the vignette development and use. For example, it has been mentioned that it allows the interview to be structured, provides a systematic way of collecting data and facilitates saturation. Other contributions appear to be more specific, notably increasing acceptability to participants when the study phenomenon is sensitive, such as with ethical issues, practice gaps or recovery from challenging clinical situations. By creating a safe distance through use of a fictitious scenario, the method encourages respondents to engage in deeper reflection on sensitive topics that they may otherwise prefer to avoid. More marginally, some authors appreciate the potential flexibility of the vignette (eg, manipulation of certain characteristics).42 Some authors22 37 recommend using the vignette in combination with other methods to compensate for limitations. Additionally, Morrison considers that the vignette is a static approach that does not leave enough room for interactions.36 This point of view suggests that the vignette may not elicit authentic discussion among participants unless the interviewer has the skills to facilitate exchanges.

Our results raise the need to explicitly consider and report strategies to ensure rigour and transparency in both the development of the vignette and the quality criteria of the wider qualitative study design (credibility, dependability, confirmability and transferability).49 Even with well-designed vignette-based studies, limitations in external validity must be documented.

The vignette-based methodology in qualitative research has an added value in intervention research in which the definition of problems and solutions is carried out in partnership between healthcare professionals and researchers.50 After expert consultation and pretesting, a vignette content that allows an in-depth understanding of a complex and highly contextualised phenomenon where a multitude of factors can, alone or in combination, influence the practice in clinical settings. Vignette-based qualitative studies offer the possibility of reflecting on challenging topics and supporting evidence-based decision making and action in practice and in future research.

Strengths and limitations

Although strategies are employed to ensure the rigour of the review process, we recognise several limitations. This scoping review was conducted to inform qualitative data collection from healthcare professionals using a reflexive approach, which explains why quantitative studies were excluded. We recognise that there is considerable use of vignettes in quantitative research. Their purpose and therefore the quality criteria for their use are categorically different than for qualitative studies, in terms of both vignette development and utilisation. Stakeholders can better understand the complex world of health professionals if researchers move throughout complementary approach to better understand complex issues.51

The search strategy is limited to empirical studies retrieved from electronic databases after 2000 and excludes grey literature. It covers only a proportion of published literature using vignettes as a qualitative research approach. We are aware that various search terms (eg, vignette, scenario, case report and snapshot) carry meanings that may be used interchangeably. What we attempt is not a meta-level synthesis of vignette-based qualitative research, but the pooling of content from included studies in our scoping review.52 Because our initial interest is to learn from prior use of vignettes in research in healthcare settings, it is possible that included articles reflect a selection bias related to our methodological focus. The small number of eligible studies reduces the robustness of recommendations for the development and utilisation of vignette-based methodology in qualitative research. The number may reflect our decision to include only articles that feature ‘vignette’ in their title. Moreover, screening was challenging because studies provided little detail about how the eligibility of professional participants was determined or what qualitative approach was used, and mixed-methods was an exclusion criteria in our search strategy.

Despite these limitations, we consider that the evidence around the development steps and utilisation of vignettes that emerges from our scoping review helps deepen our understanding of the method and provides valuable recommendations for future research. While Peters et al23 suggest that information scientists, stakeholders and/or experts may be consulted to validate the interpretations of scoping reviews, this step appears unnecessary given the diversity of our research team and the small number of included articles.

Conclusion

This scoping review generates a summary of vignette-based methodology and offers guidance regarding the development and use of vignettes in qualitative research involving healthcare professionals, which can be applied in various settings including oncology. Future research may contribute to overcoming identified risks to quality by reporting: (1) an explicit definition of vignette-based methodology as for all qualitative study design; (2) details about vignette development steps (internal validity); (3) rich description of vignette utilisation (external validity); and (4) strengths and limitations based on quality criteria for qualitative studies.

It is expected that future research will more systematically plan and document the development and utilisation of vignette-based methodology and report the research process with sufficient detail to establish how the plausible content of the vignette is associated with study results. Future publications should take into account recommendations from the studies reported in this scoping review and integrate reporting on quality criteria.