Visual Scribing: A Qualitative Research Tool in a Community Engagement Project for Birth Equity

Abstract

Visual scribing is a form of notetaking in which the visual scribe pays attention to live discussions and captures the themes and ideas that emerge through the combined use of pictures, diagrams, doodles, and text. In recent years, visual scribing has become increasingly used; however, as this field has emerged from practice and not academia, related literature and theory remain scarce. The act of visual scribing during research activities can be a practice that facilitates the documentation of a researcher’s individual observations, yet also serves to understand researchers’ own reasoning as observers. Additionally, when scribing products are shared with research participants, this helps increase trust and mutual understanding as knowledge is being co-created in an accessible medium (illustration). In this way, scribing becomes both a practice for, and product of, research. In this article we describe how visual scribing methods were used as a complementary approach for notetaking, data collection, and analysis in a community engagement project focused on birth equity in California. We found three main benefits of using visual scribing as part of a broader set of qualitative approaches: (1) triangulation, (2) expanded insight, and (3) participant engagement. Here we present the literature that informed our approach, lessons learned, and potential areas for future inquiry.

Keywords

visual scribing graphic recording mixed-methods creative research methods

Why Visual Scribing?

Visual scribing, also referred to as graphic recording, visual notetaking, or sketch noting, is a form of notetaking in which the visual scribe pays attention to live discussions, conversations, or meetings, and captures the themes and ideas that emerge through the combined use of pictures, diagrams, doodles, and text (Zheng et al., 2021). In recent years, visual scribing has become increasingly used among community-based organizations, multilateral organizations and development practitioners in workshop, meeting, and conference settings to synthesize discussions and brainstorming sessions. However, as this field has emerged from practice and not academia, relevant literature and theory are still developing (Bertoncin et al., 2020; Dean-Coffey, 2013).

Nonetheless, visual scribing has great potential to expand the field of qualitative inquiry. In this article we describe how visual scribing methods were used as a complementary approach for notetaking, data collection, and analysis in a community engagement project focused on birth equity in California. Given the equity thread that ran through all elements of project design (Batra et al., 2023a), the project team considered visual scribing to be an appropriate tool due to its ability to increase empowerment and cultural responsiveness as it creates shared experiences between stakeholders and researchers. Wallerstein et al. (2020) describe a collective empowerment process that allows engagement with communities for equity. In this process, several tools were developed and adapted to facilitate a collective engagement process including a Visioning Tool. We see visual scribing methods as a complementary or alternative approach for the visioning process that is central to community-based and equity-focused projects. Further, visual scribing can often help remove some of the biases that exist in language – images can help us move past words that carry other meanings and help increase understanding from different perspectives (Dean-Coffey, 2013).

My¹ qualitative methods journey began when I was beginning my master’s degree in public health. As a physician, my medical training had instilled a deep respect and appreciation for the “lab-based sciences” and positivist approaches. I came to my public health degree determined to specialize in quantitative methods and was especially interested in an advanced epidemiology class that was taught on Thursday afternoons. But alas, the class coincided with a mandatory course in the core curriculum – Principles of Social Science Research. I reluctantly put on hold my plans for advanced epidemiology and attended my mandatory class, which quite unexpectedly, changed my whole research trajectory and way of looking at the world around me.

One decade later, I find myself having left quantitative endeavors to others, and immersing myself in research approaches that focus on reflexivity, co-constructed knowledge, and interpretivist theories. A key factor that attracts me to qualitative research is that I see it as a creative practice. A strong qualitative analysis that tells an engaging story and highlights new insights necessitates creativity, inventiveness, originality, and not being afraid to think outside of the box. Since I can remember I have loved doodling, drawing, and other creative endeavors – to the point that when I was in high school my top two choices for university programs included art school or medical school. Thus, on account of my personal artistic inclinations, I jumped at the opportunity of exploring visual scribing as a qualitative method when the rest of the birth equity project team² suggested it as an option for data collection.

Here, we share our approach to visual scribing and analysis, the literature that informed our approach, lessons learned, and potential areas for future inquiry.

Bridging Science, Creativity, and Rigor

Many qualitative methodologists have come to agree that “the scientific and creative aspects of qualitative research cannot be separated” (Bailey et al., 1999, p. 169). The grounded theory approach that is often used in qualitative research bridges science, creativity, and rigor. Such work involves building new theories and ways of thinking rather than testing preexisting theories. To build new theories requires creativity and a capacity to see the data in different ways – allowing for intuition to guide part of the research process (Hunter et al., 2002). Creativity also emerges in qualitative research through the design of novel methodological approaches to answer specific questions, innovative ways to present and organize the data, and challenging traditional ways of thinking (Whittemore et al., 2001).

However, some researchers may balk at the term “creativity” as it is perceived to be unscientific. This stems from the historical evolution of ways of knowing and thinking. During the Enlightenment period, disciplinary knowledge separated scientific thinking which was based on reason and logic from rhetorical thinking which stemmed from creativity and intuition. This duality has been critiqued by feminist scholars, as this polarization favors certain types of reasoning and thinking over others (McDowell, 1992). Because of this, earlier qualitative methodologists emphasized the minimization of subjective circumstances in research encounters for the sake of objectivity and neutrality, even though it is not realistic to separate the context, circumstances, and lived experience of researchers and participants (Bailey et al., 1999).

The tension between creativity and rigor/validity in qualitative methods may be addressed by making two methodological improvements. To address concerns regarding rigor, researchers should improve their descriptions of qualitative analytical procedures. Qualitative analysis is often perceived as a black box where data enter, and insights come out. By carefully documenting all the steps and iterations of the analysis we can increase credibility, transferability, dependability, and confirmability. Second, we as a discipline should consider shifting the concept of validity in qualitative research from statistical inferences and generalizability to what are the strategies by which the researcher can move between and negotiate fieldwork and theory (Bailey et al., 1999).

To ground creativity within the scientific process, Amabile and Tighe’s four step process of creativity can be used as a guiding framework (Amabile & Tighe, 1993). The process includes gaining insight, achieving saturation, a period for incubation, and ends with the practice of verification. One notable feature of this process is that it draws certain parallels to the scientific method, and as such, is well suited for research efforts. The observation or drafting of an initial research question could be equated to gaining insight, researching, hypothesizing, and collecting data would be translated into saturation, data analysis is the incubation period, and reporting and validation activities are the final verification step.

Qualitative researchers who lean into creative methodologies can then draw from many different tools to tell a story (Janesick, 2001). One of these tools being arts-based research. In arts-based research methods the researcher can either be an observer of the process of making art, or they can be involved in the process as an artist who constructs part of the research as a form of data collection (Gerstenblatt, 2013). These artistic products can then be used as a reflective process, either to elicit information from people, or to summarize and conceptualize ideas expressed by research participants.

A scoping review of arts-based health research included eight studies in which participants created drawings to enhance analysis, however no studies were included in which researcher-developed drawings contributed to the analysis (Boydell et al., 2012). In recent years, arts-based methodologies have been increasingly used in qualitative research often with the aim to disseminate research in creative ways and reaching a broader audience, yet researcher-created drawings continue to be less commonly used (Ho et al., 2021). These visual methods most frequently engage with the images as the subject of analysis, and not as a way to conduct analysis (Jellema et al., 2022). For example, graphic elicitation is a frequently used qualitative approach in which research participants are asked to create drawings or other visual media according to different prompts and then the products are analyzed alongside other data collected (Copeland & Agosto, 2012). Other visual inquiry approaches such as making collages or concept mapping can help research participants articulate relationships between things and explain them to researchers (Butler-Kisber & Poldma, 2010). In other instances, visual notetaking has been assessed as a learning strategy where taking notes in the form of drawings instead of text improved learning and retention (Courneya & Cox, 2020).

Methodologists who advocate for the expanded use of visual materials to include researcher-created materials explain that qualitative researchers already often take written notes as part of their research practice or use audio transcriptions as a central part of research engagement. Drawings created by researchers could be part of this corpus, serving to provide information on the researcher’s relationship to the data, as a tool to connect the dots, and a means to disseminate research findings (Bertoncin et al., 2020; Jellema et al., 2022).

The act of visual scribing during research activities can facilitate individual observation, while illuminating our own reasoning as observers. The different elements used in scribing such as color, scale, position, shapes, and text makes connections visible and narrates what we did and saw (Bertoncin et al., 2020). Additionally, sharing co-developed scribing products with research participants where they had the opportunity provide feedback on the final product helps increase trust and mutual understanding as knowledge is being co-created through an accessible medium (Jellema et al., 2022). In this way, scribing becomes both a practice for – and product of – qualitative research.

In contrast to written notes, scribing offers a broad variety of ways to convey meaning and importance. For example, the choice of specific text words embedded in the drawings and the size of the words can signify level of importance (Dean-Coffey, 2013). The layout of the page, how space is used and the general composition can also carry meaning in terms of relationships between ideas (Ho et al., 2021). Other relevant features include the structuring of elements, the visual styling, dimensions, levels of conciseness, types of illustrations included, figures, personification, clustering, placement, and overall cohesion (Zheng et al., 2021). Visual scribing is a very involved process with multiple dimensions of decisions and strategies that the scribe navigates. These dimensions include: (1) capturing the idea, (2) emphasizing key concepts, (3) creating a narrative, (4) developing a visual identity, and (5) contending with the constraints of live drawing (Zheng et al., 2021). Given this dynamic environment, scribing requires active listening and constant, iterative analysis of information in real time. Because of this, it is likely to yield an expanded set of insights relative to those obtained through only traditional written notes.

Scribing is not equivalent to concept mapping or storyboarding. Concept mapping or mind mapping is a process that shows how ideas are related to each other and links relational concepts as they emerge (Butler-Kisber & Poldma, 2010). Storyboarding is splitting up an end product into sequential “scenes” to tell a story in a specific order (Dean-Coffey, 2013). Scribing, on the other hand, is non-linear and intended to support group discussion and collective thinking.

Visual Scribing in Action

Our Project

As part of a community engagement project for birth equity in California, our team chose visual scribing as a method for notetaking and analysis. The Birth-Centered Outcomes Research Engagement in Medi-Cal (B-CORE) project used a deliberative democracy approach to engage stakeholders who had lived experience in the Medi-Cal (California Medicaid) perinatal care system. The goal was to generate a community-driven agenda with recommendations to address inequities in maternal mortality and severe maternal morbidity among the Medi-Cal population (Batra et al., 2023b).

Deliberative democracy approaches seek to engage members of the public in a formal dialogue about certain key issues. The aim is to include participants’ opinions and values in a decision-making process, while ensuring that participants are well-informed on the topic at hand (Rychetnik et al., 2013). To accomplish this, deliberative democracy approaches have two main phases: an information phase, and a deliberation phase.

Stakeholders were recruited through purposive sampling and represented Medi-Cal beneficiaries who had experienced a birth in the last five years, perinatal care providers within the Medi-Cal system, Medi-Cal health plan administrators, and maternal health advocates. The 37 stakeholders then participated in virtual co-learning sessions over the course of one year, during which they attended monthly sessions which comprised the “information phase”. During the information phase sessions, stakeholders learned together about key topics related to maternal mortality and severe maternal morbidity in Medi-Cal.

Once the co-learning sessions were completed, stakeholders were divided into four separate panels for the deliberation phase. Each 90-minute deliberation panel was comprised of eight to ten participants and were guided by the same experienced facilitator, using the same structured protocol and discussion prompts. The structured deliberation protocol had three key activities: (1) framing of a key issue, (2) deliberation by panel on how to address the issue, (3) consensus on key recommendations.

Our Data Collection

During each deliberation panel, a member of the research team took written notes. These notes were aggregated and uploaded into Dedoose Version 9.0.54, a qualitative analysis software. We began with a round of inductive open coding using descriptive and in-vivo codes. In-vivo coding consists of attaching labels to excerpts of text by using words or phrases used by participants without having previously defined codes, while descriptive codes are those in which the coder assesses a text excerpt and attaches a code depending on the content of the text. We then refined initial open codes into a formal coding structure. In a second round of coding, we applied these codes to the written deliberation notes through a deductive thematic content approach. Thematic content analysis uses a set of defined codes or labels and attaches these to different excerpts of text in the documents being analyzed (Seale, 1999; Walker & Myrick, 2006). This systematic process allowed us to explore patterns in code application and nuance in how codes are applied to create themes that describe meaning expressed by individuals (Vaismoradi et al., 2013).

In addition to the written notes taken during the deliberation sessions, we used visual scribing as a complementary qualitative analysis approach. I (GA) took the lead on visual scribing efforts and used an iPad and Apple Pencil to digitally render the drawings. At the start of each session, we explained to participants that notes in the forms of doodles and drawings would be taken in real time to reflect the discussion that was happening on each of the key topics, and that the drawings would be shared during the half-way break and at the end of the session.

At the half-way point of each deliberation session, sharing visual scribing products in progress allowed participants to see a quick summary of what had already been discussed and prompted them to emphasize or clarify points or take the discussion in a new direction to address a gap they perceived in the visual summary. At the end of the session, the drawings provided an opportunity for stakeholders to quickly view what had been covered over the last 90 minutes, which elicited final thoughts and concluding statements. The visual scribing products conveyed that participants’ thoughts and recommendations were being captured and retained in an accessible manner to share these ideas with others. Participants expressed their enthusiasm about having their points visually depicted and illustrated.

During the deliberation sessions the pace and depth of the discussion around different topics varied, allowing me varying time frames to capture and add details to doodles and visual depictions. This led us to see presence or absence of graphic representations as a proxy for length of time discussing a particular issue or depth of discussion. For example, if a fleeting idea was shared and no other stakeholder continued the thought or added to the idea, the group would move to the next issue, which did not leave time for the scribe to fully develop a drawing. On the other hand, topics that were discussed at length and had more nuanced considerations provided more examples, words, and imagery for the scribe to interpret and draw. Some additional detail like color and re-writing words for increased clarity were added after the end of the deliberation session to improve the visual quality of the final product, but no additional graphic elements or ideas were added on to the page. It is important to note that the depth to which a particular topic is discussed and represented in the graphic representations is also subject to other variables such as the facilitator’s skills, the level of engagement of individual participants, the level of comfort with different topics on the discussion protocol, as well as the expertise of the visual scribe.

In Deliberation Session A (Figure 1), stakeholders explained that to increase equitable outcomes for pregnant people under Medi-Cal, coverage for services needed to be expanded to include home births, nutritional support, childbirth education classes, and complementary and alternative medicine (e.g., expansion of acupuncture coverage). Stakeholders also spent considerable time discussing what integrated care would look like, and recommended moving services into the community, having better care navigation, continuity of care, and data-informed decisions. On the topic of data, stakeholders in this discussion further recommended that data be further disaggregated by race and ethnicity, and that data findings be accessible to communities. Another consideration for reduction of racial disparities was emphasis on increasing avenues for culturally concordant care. Lastly, one suggestion included providing cash transfers to pregnant people with “no strings attached” to use as they best saw fit.

Figure 1.

Visual scribing product from deliberation session A.

Stakeholders in Deliberation Session B (Figure 2) noted that currently, commercial insurance provides higher reimbursement rates for services in comparison to Medi-Cal. To bridge the gap, stakeholders recommended increasing reimbursement for Medi-Cal perinatal care providers, and linking those enhanced rates to patient experience and quality of care. Medi-Cal eligibility during pregnancy was also a topic of conversation. Stakeholders said that Medi-Cal eligibility should be automatic for pregnant people, without having to meet other financial criteria. An expansion of Medi-Cal eligibility was also recommended for pre-conception care and a prolonged postpartum period. Anti-racism training was highlighted to address racial disparities. Lastly, in terms of data, the importance of data integration across systems in the state was noted, as well as translating data from patient experience into decision-making tools for patients.

Figure 2.

Visual scribing product from deliberation session B.

In Deliberation Session C (Figure 3), participants recommended having single patient identification numbers to facilitate data collection and sharing across health systems and settings. Echoing participants in deliberation session A, whose input they had not seen or heard, Session C deliberants also discussed the need for more granular data by race and ethnicity. They used specific examples such as the broad diversity that exists within the Asian American and Pacific Islander category, as well as other nuances to racial and ethnic identity such as generational status and language spoken at home. In addition to data collection, data sharing was viewed as an important element of the system, with recommendations to share maternal mortality data through social media platforms, and to engage community members in data interpretation. Stakeholders from this deliberation panel also recommended prioritizing racially concordant care, however, they provided more concrete pathways for this by recommending reductions in barriers to entry into the workforce and subsidizing education for health care professions.

Figure 3.

Visual scribing product from deliberation session C.

Lastly, in Deliberation Session D (Figure 4), stakeholders noted that patients are often unaware of the services for which they are eligible and recommended a “pregnancy booklet” explaining all Medi-Cal benefits and how to navigate the system to access said resources. This group also recommended anti-racism training for providers, also highlighting that anti-racism education needed to be implemented further upstream in elementary and high school settings to be truly effective. Stakeholders encouraged health systems to use the data they collect to develop data-driven insights, update these often, and share with communities. One participant shared a personal experience of discrimination based on insurance status, this prompted the rest of the group to question the need for providers to have access to insurance information in patient files and recommended that maybe this information be only shared at the administrative level, where it is necessary for billing purposes.

Figure 4.

Visual scribing product from deliberation session D.

Our Analysis

The image analysis stage did not include participation by the stakeholders involved in the deliberation process. However, we shared preliminary results with the B-CORE steering committee.

Non-linearity is an important consideration for image analysis. In the written language, the semiotic symbols are presented in a sequential order, while for images or scribing products, the visual signs are all present at the same time. Further, the relationship between one sign and another in written language is temporal, while for images it is spatial (Penn, 2000).

To analyze the visual scribing products that resulted from our four deliberation sessions, we followed Penn’s four steps to analyzing images (Penn, 2000). The approach consists of two broad areas: dissection and articulation.

When a person looks at an image, there is a constructive process that happens for them to interpret meaning. And this meaning will depend on the individual’s experience, cultural context, and idiosyncrasies. Because cultural knowledge is necessary to interpret an image, we want to make explicit through analysis the contextual knowledge needed to interpret the image.

The four steps to image analysis are as follow:

1. Select materials to analyze

2. Denotational inventory

3. Higher levels of signification

4. Present findings

The materials we selected for analysis were comprised of the four separate scribing products that resulted from each of the four stakeholder deliberation sessions.

For our denotational inventory, we proceeded to first annotate elements on the visual materials (see red annotations on Figure 5) and then produced a systematic list of the elements in the materials (Table 1).

Figure 5.

Denotational inventory through margin annotation.

Table 1.

Denotational Inventory in Listed Format.

Session	Elements
Session	Drawings	Text
A	✓ Umbrella	✓ Coverage
	✓ Book	✓ Postpartum
	✓ Apple	✓ Dietician
	✓ Prescription bottles	✓ Home birth
	✓ Acupuncture needles	✓ Meds
	✓ Mother and child embracing	✓ Alternative therapies
	✓ Bag of money	✓ Reimbursement
	✓ House	✓ Data and surveys
	✓ Graphs	✓ Integrated care
	✓ Clipboard and pencil	✓ Expanded languages
	✓ Hemi-brain	✓ Culturally concordant care
	✓ Birth control (intrauterine device, pills, injections)	✓ Continuity of care
	✓ Pieces of paper with Spanish and English headers	✓ Navigation systems
	✓ Stack of money, puppet strings and scissors	✓ Financial support
	✓ Compass	✓ Share data back with the community
	✓ Heart with EKG trace	✓ Community input
	✓ Hospital	✓ Patient experience
	✓ Provider with hand on patient’s shoulder	✓ Patient input
	✓ Provider darker skin tone	✓ Move services into the community
	✓ Pregnant people	✓ Move services into the community
B	✓ Person thinking about pregnancy	✓ Quality-based reimbursement
	✓ Lightbulb	✓ Quality, outcomes, value
	✓ Heart and blood pressure gauge	✓ Commercial
	✓ Brain	✓ Medi-Cal
	✓ Group of providers with different skin tones	✓ Reimbursement
	✓ Red blood cells and platelets	✓ Medi-Cal eligibility
	✓ Pregnant person talking on the phone	✓ Expanded post-partum care
	✓ Infographics/information booklet	✓ Anti-racist training
	✓ Laptop with graphs on screen	✓ Access to all services and specialties
	✓ Scales	✓ Communication
	✓ Stack of money and coins	✓ Easy ways to access provider
	✓ Smaller stack of coins and weight of other things (quality, value, outcomes)	✓ Translate data
	✓ Mother holding baby	✓ Patient-friendly information
	✓ Map of California with connected dots	✓ Patient experience data
	✓ Map of California with connected dots	✓ Data integration & sharing
C	✓ Pregnant person in yellow circle	✓ Single patient identification number
	✓ Group of pregnant people with different skin tones	✓ Single obstetrician
	✓ Graphs	✓ Improved data collection
	✓ Social media logos (Twitter, Facebook, Instagram)	✓ Data disaggregation
	✓ “3x” and stack of money and coins	✓ Data sharing
	✓ Provider with hand on patient’s shoulder (non-white skin tones)	✓ Community advisory board
	✓ Parent breastfeeding baby	✓ Social media
	✓ Person with very long, flowy hair	✓ Expanded care and coverage
		✓ Group support
		✓ Increase reimbursement rates
		✓ Diverse workforce
		✓ Racially concordant care
		✓ Home births
		✓ Lactation support
		✓ Coverage after birth
		✓ Genetic counseling
		✓ Mental health
		✓ Transportation
		✓ Birth control
		✓ Coverage after birth
D	✓ Pregnant person	✓ Don’t just collect data – use it
	✓ Checklist	✓ Analyze
	✓ Map of California with arrows across sectors	✓ Share
	✓ Gap between two pathways, wedge to close the gap	✓ Yearly anti-racist trainings
	✓ Book	✓ Anti-racist, experiential, and critical thinking education
	✓ Pregnant person with long, flowy hair	✓ Blind insurance information
	✓ “K-12”	✓ What is covered and what is not
	✓ Calendar with date marked with star	✓ Bridge the reimbursement gap
	✓ Clipboard with patient information redacted	✓ Medical provider – non-medical providers
	✓ Stacks of paper	✓ Automatic Medi-Cal eligibility
	✓ Graphs	✓ Expand coverage
		✓ Standard mental health care visit
		✓ Linkage to primary care
		✓ Postpartum visits
		✓ Reduce medical bias
		✓ Access to same services as private insurance

For higher levels of signification, we asked questions of each element listed in Table 1. We explored what each element conveyed and how they related to each other (Table 2). Through an iterative approach we condensed these meanings and interpretations into a thematic aggregation that we considered was a fair representation of the discussion across the four deliberation sessions.

Table 2.

Higher Levels of Signification Drawn From Denotational Inventory.

Elements		Interpretation
Drawings	Text	Interpretation
-Umbrella	-Coverage	Services that are embedded within a broader structure, coverage under an umbrella of services, protection under an umbrella, natural part of patient care, services that expand
-Long, flowy hair	-Access to all services and specialties
-Apple	-Classes for birth partners	Services that stakeholders believe should be available to all pregnant people
-Prescription bottles	-Acupuncture	✓ Home births
-Acupuncture needles	-Access to all services and specialties	✓ Expanded postpartum coverage
-Book	-Lactation support	✓ Nutrition support
-House	-Nutrition	✓ Mental health
-Mother and child	-Mental health	✓ Primary care
-Birth control	-Coverage after birth	✓ Prescriptions and medications
-Heart and EKG trace	-Transportation	✓ Alternative therapies
-Brain	-Birth control	✓ Postpartum support
-Blood cells	-Postpartum visits	✓ Breastfeeding support
	-Postpartum support	✓ Transportation
	-Linkage to primary care	✓ Classes and education
	-Access to same services as private insurance	✓ Birth control options
-Bag of money	-Reimbursement	There are two main things that need to be addressed within reimbursement for services. First, there is a gap in reimbursement at two levels: Between medical and non-medical providers, as well as between Medi-Cal and commercial payers. Second, there needs to be a shift in what is valued for reimbursement and weigh more heavily quality of care indicators to drive reimbursement rates
-Stack of money and coins	-Medi-Cal versus commercial
-Small amount of coins and added value of quality and outcomes	-Quality-based reimbursement
-Reimbursement scales	-Value
-“3x” stack of money and coins	-Quality
-Wedge filling gap between medical and non-medical provider reimbursement rates	-Outcomes
	-Increase reimbursement rates
	-Bridge the reimbursement gap
-Graphs – bar charts and pie charts with many different colors and categories	-Data and surveys	The multiple colors and categories on the charts reference the fact that stakeholders felt that data collection was not well disaggregated and further categories and information on patients needed to be collected and document – especially racial and ethnic subcategories
-Laptop with graphs	-Race/ethnicity	The stacks of paper signify bureaucracy or stagnation, where data are collected and then unused. Data needs to be both collected and used to inform action
-Stacks of paper	-Cause of death
-Clipboard and pencil	-Patient experience
	-Data disaggregation
	-AAPI
	-Language
	-Ethnic subcategories
	-Generation
-Provider with hand on patient’s shoulder	-Culturally concordant care	Racism in healthcare was a salient topic of conversation. Stakeholders recommended focusing on increasing opportunities for anti-racist training in the workplace and other educational settings, as well as increasing the diversity of the healthcare workforce to be able to provide culturally and racially concordant care
-Provider darker skin tone	-Anti-racist training for providers
-Light bulb	-Diverse workforce
-Group of providers with different skin tones	-Racially concordant care
-Provider with hand on patient’s shoulder (non-white skin tones)	-Anti-racist, experiential and critical thinking education
-Calendar with star	-Yearly anti-racist training for providers
-“K-12”	-Blinding insurance information
-Patient chart with redacted information	-Blinding insurance information
-Compass	-Continuity of care	The compass signifies finding the path or direction – in the context of perinatal care, this means support for patient navigation, making services easier to access when they are based out of the home or community-based organizations, and integrating care so that patients do not have to go to multiple different places and can establish high-quality, long-lasting relationships with providers
-Houses/homes	-Integrated care
-Forms in Spanish and English	-Move services into the community
	-Communication
	-Easy ways to access provider
	-Single obstetrician
	-Expanded languages
-California map with connected dots	-Data integration and sharing	Current systems across the state of California are not well integrated and data sharing across geographic regions needs to be improved to increase quality of care. One way to address this is with a single patient identification number to better track individuals across the state
-California maps with arrows between sectors	-Maintaining eligibility across counties & data sharing
-California maps with arrows between sectors	-Single patient identification number
-Books	-Share data back with the community	Books and pamphlets signify access to information. Social media icons also allude to broad access to information that is easily accessible to the public. Stakeholders emphasized that data insights should not only be shared with hospitals, providers, and Medi-Cal, but also with the community and individual patients
-Pamphlets	-Translate data
-Patient materials and infographics	-Patient-friendly decision-making tools
-Social media icons	-Data sharing
	-Social media
	-Medi-Cal pregnancy booklet: What is covered and what is not

When we compared across the four visuals and through the denotational inventory process, we were able to identify four categories of recommendations that were shared in all of the deliberative sessions: (1) Medi-Cal coverage, (2) Racism as a contributing factor to adverse maternal health outcomes, (3) Opportunities to better use and share maternal health data, and (4) Importance of patient experience to guide quality improvement.

To present the findings we decided to continue with a visual format and developed the diagram shown in Figure 6 as a summary visual for the B-CORE community engagement project recommendations.

Figure 6.

Visual scribing thematic aggregation: Pathways to decrease maternal mortality (MM) among Medi-Cal beneficiaries.

Scribing versus Written Notes

When we compared the recommendation categories obtained through the scribing visual analysis and through qualitative coding of written notes, we found a high level of agreement between the two approaches. The scribing approach yielded four categories of recommendations, all of which were mirrored in the traditional qualitative approach (Table 3). The qualitative coding of the written notes yielded an additional area of recommendation that did not emerge through the scribing approach.

Table 3.

Comparison of Recommendation Categories Between Scribing and Written Note Approaches.

Written notes category	Written Notes Recommendation^a	Scribing Category	Scribing Recommendation^a
Stakeholders recommended that Medi-Cal health plans expand coverage and the services/benefits offered to pregnant Medi-Cal beneficiaries	- Revise the current Medi-Cal eligibility process and criteria to expand access to perinatal health care	Coverage	-Linkage to primary care
	- Extend the period of Medi-Cal coverage for pregnancy to include extended preconception and postpartum windows		- Mental health
	- Increase awareness of existing covered benefits and expand the range of what is covered		- Preconception coverage
	- Better integrate full-scope mental health services into basic perinatal coverage		- Extended postpartum care
Stakeholders all agreed on anti-racism in perinatal care as a key priority and recommended new approaches to ensure accountability for racism and bias in the health care system contributing to MM/SMM	- Integrate sustainable evidence-based strategies to promote anti-racism	Equity & racism	-Community-based services
	-Assure a single equitable standard of care for all pregnant people		- Anti-racism training
			-Expanded languages for patient resources
			- Culturally concordant care
			- Workforce diversity
Stakeholders recommended new strategies for MM/SMM data collection, data use, and data sharing	- Disaggregate the data (e.g., into additional racial/ethnic subgroups, geographies, insurance types)	Data	- Share data across systems and counties
	- Use the data better (don’t just collect it)		- Share data with community
	- Share accessible data with all affected communities		- Racial & ethnic disaggregation
	- Share accessible data with all affected communities		-Collect & analyze patient experience data
Stakeholders recommended that patient experience be a guiding principle in measuring quality of care in Medi-Cal	- Make patient-centered care the norm and not the exception	Patient experience	- Continuity of care
	- Turn patient experience measures into key performance indicators of quality		- Systems navigation
			- Integrated care
			- Collect & analyze patient experience data
Stakeholders recommended that Medi-Cal reimbursement rates for perinatal care providers be increased	-Increase reimbursement rates for Medi-Cal providers to match or surpass what is paid by private insurance	Not present	Not present

^aRecommendations that emerged through both the written notes and visual scribing are bolded.

One type of analysis that we were able to conduct with the coded written notes that was not done with scribing materials was code co-occurrence. Code co-occurrence is when an excerpt is double coded, meaning more than one label/code was attached to the same text excerpt. This allowed us to assess how stakeholders talked about things and how they constructed levels of relatedness when they referred to different ideas and concepts. Through this, we gained an initial sense of the domains of understanding, recommendations, and knowledge shared by stakeholders during deliberation sessions. Codes and topics that overlapped during discussions were analyzed for commonalities, and four key domains were found to exist: (1) health system structure, (2) responsibility for anti-racism and equity, (3) data use, and (4) Medi-Cal reimbursement and payment conditions. For example, stakeholders noted that the health care delivery system and providers of perinatal care were responsible to enact recommendations regarding the provision of anti-racist perinatal care to Medi-Cal beneficiaries (Figure 7, “Domain 2”). Further, stakeholders recommended that clinics and hospitals adopt and monitor measures of patient experience, and that health plans link patient experience measure performance to payment (Figure 7, “Domain 4”). These shared domains helped us understand the general categories of how stakeholders thought of topics and how they related to each other.

Figure 7.

Stakeholders’ shared domains of understanding of maternal mortality-related topics.

The shared domains of understanding from Figure 7 were developed from analysis of the written notes, however, these domains of understanding also emerged through the visual scribing. Specifically, domains 2, 3, and 4 were prevalent across all four deliberation groups.

What We Learned

We conducted a community engagement birth equity project using deliberative democracy methods and employed multiple qualitative approaches to analyzing the resulting stakeholder recommendations. We reported aggregated recommendation themes for the B-CORE project that drew from the analysis of both written notes and drawn materials.

We found three main benefits of using visual scribing within a broader set of qualitative approaches. First, employing more than one analytical approach allowed us to triangulate and internally validate and confirm recommendation themes. Finding similar results using both thematic content analysis and visual aggregation gave us the assurance that the recommendations to reduce MM among Medi-Cal–insured patients were accurately identified and categorized.

Second, we were able to capture more nuance and a different type of perspective through visual scribing in comparison to traditional thematic content analysis of written materials. For example, one category of recommendations was present in the thematic content analysis but not in the visual aggregation diagram. This was the theme focused on Medi-Cal payment (increasing reimbursement rates to Medi-Cal perinatal care providers). We found that in all four deliberation groups reimbursement rates were mentioned, but in one of the groups it was only in passing and little discussion focused on that topic. This means that in traditional written notes it was captured that they mentioned reimbursement rates, however, with the visual scribing approach, it was such a minor mention that it did not yield enough time for a drawing to be produced, nor enough discussion was centered around it to generate ideas and metaphors on how to visually represent the issue. This could lead to the interpretation that while Medi-Cal reimbursement as a theme existed, stakeholders felt more strongly about other recommendations and had more specific feedback and ideas.

Last, we found that our project participants responded very positively to the visual scribing approach. Stakeholders during deliberation sessions used the mid-point break to review what had been discussed so far and used the opportunity to clarify ideas or circle back to issues they had forgotten to bring up. At the end of the session, there was a single comprehensive summary that captured the discussion in a manner that was easy to interpret and understand. These illustrations not only facilitated the discussion, they also are valuable as part of the dissemination because unlike a graphic based on the findings, they were shared with participants during and after the discussions and in their final form. As a result, they are a participatory product of the meeting and reflect both the engagement of the stakeholders and their views that these convey the discussions in a representative and accessible way. Not all participants would find abstracts or other graphics as accessible, nor would we expect to have the level of engagement in terms of their viewing and commenting on the images. As such these convey something that is more closely connected to their recommendations and more engaging to a diverse audience of readers, many of whom are positioned to act on the recommendations and or to take them to leadership in their own organizations.

We also recognize there are limitations in this approach. Given the limited academic literature on this type of method, there is no current blueprint on how to conduct visual scribing and how to apply it within a research framework. We document here our efforts for rigor and reproducibility in hopes to add to the literature on this topic and encourage other researchers to continue to build upon this work as well. Another limitation is the balance between insider and outsider status of the visual scribe. In this case, the visual scribe was not a professional artist or scribe, but an existing member of the research team who works in the field of maternal and child health. A professional scribe might have had more experience in capturing live discussions by drawing and might have drawn with more speed. However, we felt there was also value in the scribe having content expertise to better interpret the words and meaning of the deliberation discussions.

Areas for further research include exploring how different researchers incorporate visual scribing practices into their work, the differences yielded by visual and text approaches, impact on participant engagement and rapport, and more systematic descriptions on how to analyze visual data.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The B-CORE Birth Equity Project was partially funded by a Patient-Centered Outcomes Research Institute (PCORI) Eugene Washington PCORI Engagement Award (#20972).

ORCID iD

Gabriela Alvarado

Notes

References

Amabile

T.M.

Tighe

(1993). Questions of Creativity. In Brockman

(Ed.), Creativity (4). New York: Simon & Schuster.

Bailey

White

Pain

(1999). Evaluating qualitative research: Dealing with the tension between ‘science’ and ‘creativity’. Area, 31(2), 169–178. https://doi.org/10.1111/j.1475-4762.1999.tb00182.x

Batra

Alvarado

Bird

C. E.

(2023a). Birth-centered outcomes research engagement (B-CORE) in Medi-Cal: Community-generated recommendations to decrease maternal mortality and severe maternal morbidity. Womens Health Issues, 33(5), 474–480. https://www.rand.org/pubs/research_reports/RRA2630-1-v2.html

Batra

Alvarado

Bird

C. E.

(2023b). The “birth-centered outcomes research engagement (B-CORE) in medi-cal” project: Community-generated recommendations to decrease maternal mortality and severe maternal morbidity. Women’s Health Issues: Official Publication of the Jacobs Institute of Women’s Health, 33(5), 474–480. https://doi.org/10.1016/j.whi.2023.03.012

Bertoncin

Pase

Peterle

Quatrida

(2020). Graphic geography: Drawing territories at the Po delta (Italy). Cultural Geographies, 28(1), 19–39. https://doi.org/10.1177/1474474020931534

Boydell

K. M.

Gladstone

B. M.

Volpe

Allemang

Stasiulis

(2012). The production and dissemination of knowledge: A scoping review of arts-based health research. Forum for Qualitative Social Research, 13(1), 32. https://www.qualitative-research.net/index.php/fqs/article/view/1711/3329

Butler-Kisber

Poldma

(2010). The power of visual approaches in qualitative inquiry: The use of collage making and concept mapping in experiential research. Journal of Research Practice, 6(2), 1–16. https://jrp.icaap.org/index.php/jrp/article/view/197/196

Copeland

A. J.

Agosto

D. E.

(2012). Diagrams and relational maps: The use of graphic elicitation techniques with interviewing for data collection, analysis, and display. International Journal of Qualitative Methods, 11(5), 513–533. https://doi.org/10.1177/160940691201100501

Courneya

C. A.

Cox

S. M.

(2020). Visual note taking for medical students in the age of Instagram. Health Professions Education, 6(2), 126–135. https://doi.org/10.1016/j.hpe.2020.01.004

10.

Dean-Coffey

(2013). Graphic recording. New Directions for Evaluation, 2013(140), 47–67. https://doi.org/10.1002/ev.20073

11.

Gerstenblatt

(2013). Collage portraits as a method of analysis in qualitative research. International Journal of Qualitative Methods, 12(1), 294–309. https://doi.org/10.1177/160940691301200114

12.

E. S.

Wright

F. V.

Parsons

J. A.

(2021). Animated analysis: Drawing deeper analytical insights from qualitative data. International Journal of Qualitative Methods, 20(January-December), 160940692199049. https://doi.org/10.1177/1609406921990494

13.

Hunter

Lusardi

Zucker

Jacelon

Chandler

(2002). Making meaning: The creative component in qualitative research. Qualitative Health Research, 12(3), 388–398. https://doi.org/10.1177/104973202129119964

14.

Janesick

V. J.

(2001). Intuition and creativity: A pas de Deux for qualitative researchers. Qualitative Inquiry, 7(5), 531–540. https://doi.org/10.1177/107780040100700501

15.

Jellema

Annemans

Heylighen

(2022). Drawing the researcher into data: Drawing as an analytical tool in qualitative research. Qualitative Research, 23(5), 1398–1417. https://doi.org/10.1177/14687941221079530

16.

McDowell

(1992). Doing gender: Feminism, feminists and research methods in human geography. In McDowell

Sharpe

(Eds.), Space, gender, knowledge: Feminist readings (pp. 399–416). Routledge.

17.

Penn

(2000). Semiotic analysis of still images. In Bauer

M. W.

Gaskell

(Eds.), Qualitative researching with text, image, and sound. A practical handbook (pp. 227–245). Sage.

18.

Rychetnik

Carter

S. M.

Abelson

Thornton

Barratt

Entwistle

V. A.

Mackenzie

Salkeld

Glasziou

(2013). Enhancing citizen engagement in cancer screening through deliberative democracy. Journal of the National Cancer Institute, 105(6), 380–386. https://doi.org/10.1093/jnci/djs649

19.

Seale

(1999). Quality in Qualitative Research. Qualitative Inquiry, 5(4), 465–478. https://doi.org/10.1177/107780049900500402

20.

Vaismoradi

Turunen

Bondas

(2013). Content analysis and thematic analysis: Implications for conducting a qualitative descriptive study. Nursing & Health Sciences, 15(3), 398–405. https://doi.org/10.1111/nhs.12048

21.

Walker

Myrick

(2006). Grounded Theory: An Exploration of Process and Procedure. Qualitative Health Research, 16(4), 547–559. https://doi.org/10.1177/1049732305285972

22.

Wallerstein

Oetzel

J. G.

Sanchez-Youngman

Boursaw

Dickson

Kastelic

Koegel

Lucero

Magarati

Ortiz

Parker

Pena

Richmond

Duran

(2020). Engage for equity: A long-term study of community-based participatory research and community-engaged research practices and outcomes. Health Education & Behavior: The Official Publication of the Society for Public Health Education, 47(3), 380–390. https://doi.org/10.1177/1090198119897075

23.

Whittemore

Chase

S. K.

Mandle

C. L.

(2001). Validity in qualitative research. Qualitative Health Research, 11(4), 522–537. https://doi.org/10.1177/104973201129119299

24.

Zheng

Fernández Camporro

Romat

Henry Riche

Bach

Chevalier

Hinckley

Marquardt

(2021). Sketchnote components, design space dimensions, and strategies for effective visual note taking. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (pp. 1–15). Association for Computing Machinery.