Picture this: Evaluating the efficacy of genetic counseling visual aids

Abstract

Visual aids have been validated as effective tools for educating patients in a variety of medical settings. However, research exploring the efficacy and potential benefit of genetic counseling visual aids is lacking. To begin to address this gap, this study assessed participant knowledge of genetic counseling concepts after viewing either visual or non‐visual educational content. Participants were recruited from the general population using the crowdsourcing platform Mechanical Turk. Wilcoxon rank‐sum tests were carried out to evaluate differences in knowledge survey scores between the visual and non‐visual groups, and Poisson regression models were fitted to evaluate these differences across a variety of demographic backgrounds. The visual group had equal or higher scores than the non‐visual group across all analyses. The difference in group scores was statistically significant for autosomal recessive inheritance knowledge scores (p < 0.05). In addition, this difference was approaching significance for higher‐level knowledge scores (p = 0.05) and total knowledge scores in individuals who have not completed post‐secondary education (p = 0.05). These results indicate that visual aids improve knowledge of specific genetic counseling concepts such as inheritance patterns; the education of which is often integral to genetic counseling. These results also indicate that visual aids may facilitate a deeper understanding of genetic counseling concepts and may be particularly valuable for individuals with lower educational backgrounds. Together, the results of this study support the inclusion of visual aids in genetic counseling education to help improve patient understanding and the accessibility of genetic healthcare information.

What is known about this topic

Visual aids have been validated as effective tools in educating patients across a variety of medical specialties. However, research exploring the efficacy and potential benefit of genetic counseling visual aids is lacking.

What this paper adds to the topic

This paper evaluates the efficacy of genetic counseling visual aids in facilitating the understanding of genetic concepts as compared to written descriptions alone.

1. INTRODUCTION

Genetic counseling visual aids have long been used as tools to facilitate patient understanding of complex genetic concepts, with the first edition of genetic counseling visual aids published nearly 30 years ago by Greenwood Genetic Center (Potts et al., 1984). A recent study demonstrated that visual aids are commonly used by genetic counselors, particularly in prenatal and oncology settings (Porter et al., 2021). Despite this, research evaluating the efficacy of genetic counseling visual aids is lacking.

Visual aids such as depictions of medical procedures, videos, and comics have been validated as effective tools in educating patients in a variety of other medical settings to aid in obtaining informed consent (Furuno & Sasajima, 2015; Hodges‐Wills et al., 2021; Stewart et al., 2021; Tae et al., 2012). In addition, previous studies have demonstrated that visual aids such as graphs, pie charts, and icon arrays improve understanding of risk and reduce the influence of anecdotal reasoning on healthcare decision‐making (Fagerlin et al., 2005; Garcia‐Retamero & Cokely, 2017; Garcia‐Retamero & Galesic, 2010). Of note, previous studies have found that visual aids are particularly beneficial for the health education of individuals with low health literacy, who experience significantly poorer health outcomes and decreased use of medical services (Berkman et al., 2011; Mbanda et al., 2021; Shahid et al., 2022). Indeed, the American Medical Association recommends using tools such as visual aids to make health information more accessible and understandable to patients with limited health literacy (Weiss, 2007). As limited health literacy is associated with lower genetic knowledge, evaluating and developing genetic counseling‐specific visual aids is particularly important to help reduce barriers to understanding genetic information and improve disparities in genetic health outcomes (Kaphingst et al., 2016).

Only general visual elements, and not genetic counseling‐specific visual aids, have been broadly evaluated in genetic counseling settings. Individuals with low health literacy have more favorable views of visual components in genetic test reports (Dwyer et al., 2020). Educational resources with visual components (including interactive computer programs, audiovisual educational videos, and graphic comics) improve patient knowledge in prenatal genetic counseling settings (Björklund et al., 2012; Dugger et al., 2021; Mulla et al., 2018; Peters et al., 2017; Yee et al., 2014). Pictures, diagrams, and tables improve knowledge of hereditary breast and ovarian cancer in oncology genetic counseling (Tea et al., 2018). These studies demonstrate the value of a visual component in genetic counseling patient education. However, these investigations did not assess the efficacy of genetic counseling‐specific visual aids, such as Greenwood Genetic Center illustrations.

Only one previous study has evaluated the effect of Greenwood Genetic Center visual aids (Knyszek et al., 2012). This study found that knowledge survey scores did not significantly improve in participants who underwent prenatal genetic counseling with Greenwood Genetic Center visual aids as compared to those who underwent counseling without visual aids. However, the study was limited by a small sample size and only included images of chromosome and FISH analyses, and prenatal diagnostic procedures. This study, therefore, did not assess visual aids depicting fundamental genetic concepts that may be used by genetic counselors across many specialties.

More research is therefore needed to evaluate the efficacy of genetic counseling‐specific visual aids in educating patients. Facilitating and improving patient understanding of genetic concepts is imperative to genetic counseling. A greater understanding of genetic concepts can help patients better assess their circumstances, communicate genetic mechanisms and inheritance patterns to family members, provide informed consent for genetic testing, and make educated healthcare decisions. This study aims to determine whether genetic counseling visual aids improve understanding of genetic concepts as compared to written descriptions alone. This study also aims to help identify groups of individuals for whom visual aids are most helpful. We hypothesized that visual aids improve knowledge of genetic counseling concepts and are particularly beneficial for individuals with lower educational backgrounds and health literacy.

2. METHODS

2.1. Instrumentation

A survey was developed by the research team to gather participant demographic information and assess knowledge of genetic counseling concepts after viewing either visual or non‐visual educational content. A full copy of the survey can be found in the Appendix S1.

Survey questions were written to collect demographic information including age, gender, ethnicity, native language, primary language, highest level of completed education, previous genetic counseling, and previous genetics education. Four validated health literacy screening questions were used to measure participant health literacy (Chew et al., 2008; Haun et al., 2009). An “I prefer not to answer” response option was included for all demographic questions.

Visual and non‐visual educational pages were created for five genetic counseling concepts generalizable to many specialties. These concepts were as follows: (1) chromosomes, genes, and proteins, (2) trisomy, (3) autosomal recessive inheritance, (4) trinucleotide repeat disorders, and (5) mosaicism. The visual educational pages consisted of visual aids and written descriptions of the genetic counseling concepts, and the non‐visual educational pages consisted of the written descriptions alone. Visual aids were sourced from the most recent edition of Tribble, 2020, from Greenwood Genetic Center (Illustrations from Genetic Counseling Aids, 7th Edition, Copyright 2020, permission for use granted by Greenwood Genetic Center).

Multiple‐choice knowledge assessment questions were written for each genetic concept. These questions consisted of one recall question designed to assess basic understanding of that concept, and one or two higher‐level questions designed to assess deeper understanding and the ability to apply knowledge of that concept. Knowledge assessment questions were asked immediately following each respective educational page. An “I don't know” response option was included for each knowledge assessment question to reduce random selection of correct answers. In addition, survey directions explicitly instructed participants not to use the Internet to search for correct answers, which has been shown to significantly reduce participant cheating (Goodman et al., 2013).

This study was reviewed and granted an exemption by the University of Wisconsin‐Madison Institutional Review Board.

2.2. Participants

Participants for this study were recruited from the general population through Mechanical Turk or MTurk, an online crowdsourcing marketplace administered by Amazon. MTurk allows “requesters” to recruit “workers” to complete “Human Intelligence Tasks” (or “HITs”) in exchange for monetary compensation. This study required that workers had >100 approved HITs and 95% HIT approval ratings. These are validated standards for obtaining high‐quality data and greater participant attentiveness (Peer et al., 2014). To further strengthen data quality, study participants were restricted to those with “Masters Qualification,” a qualification granted to workers who have consistently demonstrated high‐quality work based on MTurk's internal statistical analyses.

2.3. Procedures

This survey was posted to MTurk under the general title of “Health Education Survey” to prevent self‐selection of participants with a strong interest in or experience with genetics. Participants who chose to complete this HIT were prompted to take the survey online through an anonymous Qualtrics link. Participants were ineligible to take the survey if they indicated that they were (1) unable to clearly see and interpret images and text on their device's screen, (2) a genetics professional, or (3) under the age of 18. Eligible participants were provided with a consent form detailing the study aims, procedures, benefits, risks, and protection of confidentiality. Participants acknowledged that they understood the presented information and wished to participate in this study prior to beginning the survey. A CAPTCHA verification and internal consistency attention‐check questions were used to reduce the threat of bots and random responses.

Once completed, participants were provided with a unique completion code to submit to MTurk for $1.80 of compensation. Compensation was not affected by the number of correct answers; a validated method of reducing participant cheating (Goodman et al., 2013). The survey was released in three separate batches spanning both weekdays and weekends over the course of 2.5 weeks. After each batch, response quality was deemed satisfactory following examination of internal consistency attention‐check questions and time spent taking the survey. To prevent multiple submissions by the same participant, a qualification was created and updated after each batch to exclude MTurk workers who completed previous batches of the survey.

2.4. Statistical analysis

Participant responses were reviewed and removed from analysis if (1) participants failed internal consistency attention‐check questions, or (2) participants spent <30% of the average total time taking the survey or viewing the educational pages.

Wilcoxon rank‐sum tests were used to evaluate differences in total, higher‐level, probability‐based, and concept‐specific knowledge survey scores between the visual and non‐visual groups. Sensitivity analyses were performed to assess whether the greater distribution of participants with lower health literacy levels in the non‐visual group impacted these results. Poisson regression models were fitted to evaluate the difference between visual and non‐visual total scores across a variety of demographic groups: highest level of completed education, previous genetics education, health literacy, and age. This difference could not be analyzed between English and non‐English native/primary speakers due to a limited number of participants with non‐English language backgrounds.

Results were considered statistically significant if analyses yielded a two‐sided p‐value <0.05. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).

3. RESULTS

3.1. Participants

A total of 156 participant responses were included in the study (see Table 1 for participant demographics). The number of participants randomized into each group was comparable, with 82 (52.6%) participants in the visual group and 74 (47.4%) participants in the non‐visual group. Fifty percent of participants identified as female and 50% identified as male. Age of participants ranged from 28 to 71 years of age, with an average age of 46.2 (standard deviation 11.1) and median (IQR) age of 43.0 (36.5–57.0). Age‐frequency distributions were comparable between the two groups. The majority (80.8%) of participants identified as White, 5.8% identified as Asian, 5.8% identified as Black or African American, 3.2% identified as Hispanic or Latinx, and 4.5% identified as biracial. 97.4% of participants reported English as their native language and 99.4% reported English as their primary language.

TABLE 1.

Participant demographics.

	Frequency (N)	Percent (%)
Group
Visual	82	52.6
Non‐visual	74	47.4
Gender
Female (cis or trans)	78	50.0
Male (cis or trans)	78	50.0
Age
28–39	53	34.0
40–49	47	30.1
50+	56	35.9
Ethnicity
Asian	9	5.8
Black/African American	9	5.8
Hispanic/Latinx	5	3.2
White	126	80.8
More than one ethnicity	7	4.5
Native language
English	152	97.4
Non‐English	3	1.9
Prefer not to answer	1	0.6
Primary language
English	155	99.4
Non‐English	1	0.6
Health literacy
Adequate	136	87.2
Inadequate or marginal	20	12.8
Highest level of completed education
Any amount of high school	42	26.9
Associate or bachelor's degree	94	60.3
Master's or doctorate degree	20	12.8
Highest level of previous genetics education
None	87	55.8
Grade or middle school	7	4.5
High school	40	25.6
Undergraduate university	18	11.5
Other, prefer not to answer, or conflicting responses	4	2.6

Note: Frequency and proportion of participants with various demographic backgrounds.

Based on health literacy screening responses, 87.2% of participants had adequate health literacy and 12.8% had inadequate or marginal health literacy. Notably, 15 of the 20 participants with inadequate or marginal health literacy were randomly assigned to the non‐visual group and only 5 to the visual group, limiting the health literacy analysis and biasing the sample against the hypothesized result. Additional sensitivity analyses excluding inadequate and marginal health literacy participants were performed for knowledge survey score comparisons to evaluate whether the skewed distribution of these participants affected results. No change in significance was found after removing the inadequate and marginal health literacy participants in any comparisons, and therefore this skewed distribution is thought to only have limited health literacy analyses.

The majority of participants had completed higher education, with 60.3% having completed an associate or bachelor's degree and 12.8% a master's or doctorate degree. The frequency distributions of education levels were comparable between the two groups. Over half of participants (55.8%) had not received any previous genetics education; 25.6% completed their highest level of genetics education in high school, 11.5% in university, and 4.5% in grade or middle school.

3.2. Total knowledge survey score comparison

Wilcoxon rank‐sum tests were used to evaluate differences in total, higher‐level, probability‐based, and concept‐specific knowledge survey scores between the visual and non‐visual groups (see Table 2). The mean and median total knowledge survey scores were greater in the visual group as compared to the non‐visual group. Of a possible 11 points, the visual group had a mean total score of 7.3 and median (IQR) total score of 8 (5–10), while the non‐visual group had a mean non‐visual total score of 6.7 and median (IQR) total score of 7 (5–9). However, the difference between the underlying distributions of the visual and non‐visual total knowledge scores was not statistically significant (p = 0.144).

TABLE 2.

Knowledge survey score comparisons between the visual and non‐visual groups.

	Chromosome, gene, and protein score (/2)	Trisomy score (/2)	Autosomal recessive inheritance score (/3)	Trinucleotide repeat score (/2)	Mosaicism score (/2)	Total score (/11)	Higher‐level score only (/6)	Probability score only (/2)
Visual
Mean	1.4	1.2	2.0	1.4	1.3	7.3	3.7	1.3
Median (IQR)	1 (1–2)	1 (1–2)	2 (1–3)	2 (1–2)	1 (1–2)	8 (5–10)	4 (2–5)	1 (1–2)
Non‐visual
Mean	1.2	1.2	1.6	1.3	1.3	6.7	3.1	1.1
Median (IQR)	1 (1–2)	1 (1–2)	2 (1–2)	1 (1–2)	1 (1–2)	7 (5–9)	3 (2–5)	1 (1–2)
p‐Value	0.381	0.904	0.030 a	0.352	0.810	0.144	0.051 b	0.299

Note: Mean and median knowledge survey scores for concept‐specific questions, total questions, higher‐level (more difficult, application‐based) questions, and probability‐based questions. Two‐sided p‐values evaluate the difference between knowledge survey scores of the visual and non‐visual groups using Wilcoxon rank‐sum tests.

^a The difference between visual and non‐visual knowledge scores is statistically significant (p < 0.05).

^b The difference between visual and non‐visual knowledge scores is marginally non‐significant (p = 0.05).

3.3. Higher‐level and probability‐based knowledge score comparisons

The visual group had greater mean and median scores on higher‐level (i.e., more difficult, application‐based) questions than the non‐visual group (see Figure 1). Of 6 possible higher‐level points, the visual group had a mean score of 3.7 and a median (IQR) score of 4 (2–5), while the non‐visual group had a mean score of 3.1 and median (IQR) score of 3 (2–5). The difference between the underlying distributions of the visual and non‐visual higher‐level scores was marginally non‐significant (p = 0.051). Given that the statistical power of this analysis was limited by factors such as the number of questions assessing knowledge, these results may indicate a true difference in higher‐level knowledge.

FIGURE 1.

Distribution of Wilcoxon Scores for Higher‐Level Questions in the Visual and Non‐Visual Groups. Distribution of Wilcoxon scores for higher‐level questions in the visual and non‐visual groups. Y‐axis represents Wilcoxon scores, lower and upper margins of boxes represent 25th and 75th quantiles, horizontal lines within boxes represent median, whispers represent minimum and maximum, and diamonds represent mean. ^†The difference between visual and non‐visual higher‐level knowledge scores is marginally non‐significant (p = 0.05).

The visual and non‐visual groups had comparable scores on probability‐based questions (see Table 2). These questions assessed knowledge of probabilistic outcomes of autosomal recessive inheritance and mosaicism. No statistically significant difference was found between the underlying distributions of the visual and non‐visual total probability scores (p = 0.299).

3.4. Concept‐specific knowledge score comparisons

For all five genetic concepts, the visual group had equal or higher mean and median scores as compared to the non‐visual group (see Figure 2). The difference in group scores was statistically significant for autosomal recessive inheritance knowledge scores (p = 0.030). Of three possible autosomal recessive inheritance points, the visual group had a mean score of 2.0 and a median (IQR) score of 2 (1–3), and the non‐visual group had a mean score of 1.6 and a median (IQR) score of 2 (1–2). All other concept‐specific knowledge survey score comparisons were not statistically significant.

FIGURE 2.

Distribution of Wilcoxon Scores for Autosomal Recessive Inheritance Questions in the Visual and Non‐Visual Groups. Distribution of Wilcoxon scores for autosomal recessive inheritance questions in the visual and non‐visual groups. Y‐axis represents Wilcoxon scores, lower and upper margins of boxes represent 25th and 75th quantiles, horizontal lines within boxes represent median, whispers represent minimum and maximum, and diamonds represent mean. *The difference between visual and non‐visual autosomal recessive knowledge scores is statistically significant (p < 0.05).

3.5. Demographic group comparisons

Poisson regression models were fitted to evaluate the difference between visual and non‐visual total scores across a variety of demographic backgrounds.

No statistically significant relationship was found between the difference in visual and non‐visual total knowledge survey scores and highest level of completed education (any high school compared to master's/doctorate degree p = 0.456, and associate/bachelor's degree compared to master's/doctorate degree p = 0.671). As shown in Figure 3, a possible difference between visual and non‐visual total knowledge scores was observed in participants whose highest level of education is any amount of high school (marginally non‐significant, p = 0.053). Within this level of highest education, the visual group had a mean total score of 7.3 and a median (IQR) total score of 9 (4–10), while the non‐visual group had a mean total score of 5.7 and a median (IQR) total score of 6 (4–7). No statistically significant difference between visual and non‐visual total scores was observed within higher levels of completed education (associate/bachelor's degree p = 0.876, master's/doctorate degree p = 0.581).

FIGURE 3.

Total Knowledge Survey Scores by Highest Level of Completed Education in the Visual and Non‐Visual Groups. Poisson regression fit of total knowledge survey scores by highest level of completed education. Values on the Y‐axis represent predicted total knowledge scores with 95% confidence intervals. Dark gray lines represent visual group scores, light gray lines represent non‐visual group scores, whiskers represent 95% confidence intervals, and circles represent predicted means. ^†The difference between visual and non‐visual total knowledge scores is marginally non‐significant in participants whose highest level of education is high school (p = 0.05).

No statistically significant relationship was observed between the difference in visual and non‐visual total knowledge survey scores and highest level of previous genetics education (p = 0.867 for no genetics education compared to genetics education in university, p = 0.973 for genetics education in grade/middle school compared to in university, and p = 0.397 for genetics education in high school compared to in university). No statistically significant difference between visual and non‐visual total scores was observed within specific levels of highest previous genetics education (no previous genetics education p = 0.188, genetics education in grade/middle school p = 0.627, genetics education in high school p = 0.835, and genetics education in university p = 0.374).

No statistically significant relationship was observed between the difference in visual and non‐visual total knowledge survey scores and level of health literacy (inadequate/marginal health literacy compared to adequate health literacy p = 0.924). While the visual group scored higher than the non‐visual group within each health literacy level, these differences were not statistically significant (inadequate/marginal literacy p = 0.618, adequate health literacy p = 0.232). Notably, health literacy analyses were limited by relatively few participants with lower levels of health literacy and a greater distribution of participants with inadequate or marginal health literacy in the non‐visual group than in the visual group (see Section 3.1).

No statistically significant relationship was observed between the difference in visual and non‐visual knowledge survey scores and participant age (p = 0.929).

4. DISCUSSION

4.1. Implications of results

While no statistically significant difference was observed in total knowledge survey scores between the visual and non‐visual groups, the visual group often had higher scores and never had lower scores than the non‐visual group across all comparisons. Moreover, specific comparisons reveal areas in which visual aids may be particularly valuable as well as areas for future research.

A significant difference was observed in the autosomal recessive inheritance knowledge scores between the visual and non‐visual groups, with visual group scores being higher. Knowledge and application of inheritance patterns allow patients to understand not only the genetic mechanisms of their personal conditions but also the risk of their conditions to current and future family members. That visual aids may serve as a useful tool to facilitate this process is a pertinent result for the field of genetic counseling, exemplified by the fact that education on inheritance is included in the definition of genetic counseling itself by the National Society of Genetic Counselors (National Society of Genetic Counselors' Definition Task Force et al., 2006). Moreover, numeracy and probabilistic reasoning skills (e.g., the ability to understand inheritance probabilities and genetic risks) are limited in the general population, underscoring the importance of developing genetic inheritance educational aids (Garcia‐Retamero et al., 2019; Rodríguez et al., 2013). Given that understanding of inheritance facilitates understanding of genetic risk, the results of this analysis are supported by previous findings that visual aids robustly improve understanding of risk in diverse populations (Garcia‐Retamero & Cokely, 2017). Worth noting is that inheritance visual aids may be especially beneficial when drawn live and personalized to a family's situation during a genetic counseling session (Gale et al., 2010). The results of this analysis therefore support a growing body of evidence that genetic counseling‐specific inheritance visual aids may improve patients' understanding of inheritance and, subsequently, understanding of risk information and implications of a genetic condition within their family context.

Notably, no statistically significant difference was observed in probability‐based knowledge scores between the visual and non‐visual groups. The visual aids used for these questions represent a single probability event, which we hypothesized may hinder understanding of genetic probabilities (e.g., by reinforcing the misunderstanding that 1‐in‐4 children of carrier parents will always have an autosomal recessive condition). These results suggest that visual representation of a single probability event does not negatively impact the understanding and application of genetic probability. Indeed, they may even be helpful should visual aids improve understanding of concepts necessary to accurately assess probabilistic outcomes, as was demonstrated for autosomal recessive inheritance.

The lack of statistically significant improvements in knowledge of concepts other than autosomal recessive inheritance suggests that the value of existing genetic counseling visual aids is concept specific. This may indicate that some concepts are better visually reinforced than others, and/or suggest a need to improve existing visual aids. Novel visual representations of genetic counseling information may facilitate greater patient understanding of concepts for which a significant difference was not observed. For example, one previous study has suggested that narrative character depictions (not used in this study) allow for greater self‐identification and, subsequently, greater understanding and retention of genetic counseling information (Dugger et al., 2021).

A possible difference (marginally non‐significant) was observed in the higher‐level (more difficult, application‐based) knowledge scores between the visual and non‐visual groups, with visual group scores being higher. This may indicate that visual aids help facilitate a deeper understanding of genetic concepts as compared to written descriptions alone. This is of great significance to genetic counseling education, as patients must be able to apply genetic knowledge to their own personal and familial circumstances to make informed healthcare, testing, and reproductive decisions.

A possible difference (marginally non‐significant) was also observed between visual and non‐visual total scores in participants whose highest level of education is any amount of high school, with the visual group scores being higher. This difference was non‐significant in participants with higher levels of education. These results may indicate that visual aids are especially beneficial to patients with lower educational backgrounds or who have not completed post‐secondary education. This supports the inclusion of visual aids in genetic counseling education to help make genetic information more accessible to patients with various educational backgrounds.

No statistically significant relationship was observed between the difference in visual and non‐visual total scores and health literacy levels. However, this analysis was significantly limited by a relatively small number of participants with lower levels of health literacy, and a randomly skewed distribution of these participants to the non‐visual group (see Section 4.2 below). Given previous studies' findings that individuals with lower health literacy often have limited genetic knowledge, have favorable views of visual aids, and benefit from visual aid inclusion in health education, genetic counseling visual aids may well prove to be particularly helpful to those with limited health literacy in more robust studies (Dwyer et al., 2020; Kaphingst et al., 2016; Mbanda et al., 2021).

4.2. Limitations and future studies

There were some notable limitations of this study that should be addressed in future research. This study used only two to three questions to assess understanding of each genetic counseling concept, limiting the statistical power of these analyses. In addition, this sample was composed of an older, whiter, more English‐speaking, and more educated population as compared to the general US population, limiting the generalizability of these results (U.S. Census Bureau, 2021). As previously noted, the vast majority of participants had adequate health literacy, with only 20 of the 156 participants having either inadequate or marginal health literacy. Future studies with larger and more diverse samples are therefore needed to examine the benefits of genetic counseling visual aids for children and young adults, non‐native English speakers, and individuals with limited health literacy.

Fifteen of the participants with inadequate or marginal health literacy were randomly assigned to the non‐visual group and only five to the visual group, significantly limiting the power of health literacy analyses. Given the possibility that individuals with lower health literacy levels may score lower on average on genetic knowledge questions than individuals with higher health literacy levels, it was possible that the greater distribution of individuals with lower health literacy levels in the non‐visual group could have biased other comparisons. To address this, sensitivity analyses excluding inadequate and marginal health literacy participants were performed for all knowledge survey score comparisons. No changes in the significance of results were found after removing the inadequate and marginal health literacy participants, and therefore this skewed distribution is not thought to have limited comparisons other than the health literacy analyses.

Future studies are also needed to characterize the types and sources of visual aids commonly used by genetic counselors in practice, the understanding of which may incite more research in this area. In addition, more research is needed to elucidate how genetic counseling visual aids facilitate understanding of genetic concepts (e.g., by clarifying a concept visually vs. by promoting greater attention or engagement), as this study only assessed the effect of visual aids on knowledge survey scores. Future studies should also directly evaluate the effect of visual aid inclusion in genetic counseling appointments rather than solely in educational materials as in this study. Finally, future studies should examine patient preferences and emotional responses to visual and non‐visual genetic counseling conditions, as positive and inclusive visual depictions of genetic mechanisms and conditions may provide additional psychosocial value to patients.

Data gathered from these future studies can provide insight into the development of more effective, inclusive, and accessible visual resources and materials to further bolster genetic counseling patient education. Co‐design, or the active involvement of relevant patient populations, is imperative to both the creation and evaluation of new genetic counseling visual aids and educational resources. Indeed, involving patients in a co‐creation process has been shown to yield more understandable patient‐facing materials (Dwyer et al., 2020). In particular, individuals with limited health literacy should be involved in the development of new visual aids, as their inclusion in this process is associated with greater benefits for these individuals (Mbanda et al., 2021).

5. CONCLUSION

The findings of this study support the inclusion of visual aids in genetic counseling educational materials and appointments. The visual group never had lower knowledge survey scores than the non‐visual group across this study's comparisons, indicating that visual aids do not impair comprehension. Moreover, this study's results suggest that visual aids improve knowledge of specific genetic counseling concepts, such as inheritance patterns, may facilitate a deeper understanding of genetic counseling concepts, and may be particularly beneficial for individuals with lower educational backgrounds. Future studies should be carried out to further evaluate genetic counseling visual aids in various settings and with diverse populations, and novel visual aids and resources should be developed with these findings and in collaboration with patient communities.

AUTHOR CONTRIBUTIONS

Author Viviane Pederson drafted this study's survey and manuscript and acquired and interpreted this study's data. Author Viviane Pederson confirms that she had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Authors Elizabeth M. Petty, Jennifer Rietzler, and Abigail Freeman made substantial contributions to this project's study design and critically revised its written materials. All the authors gave final approval of this version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

CONFLICT OF INTEREST STATEMENT

Author Viviane Pederson, Author Jennifer Rietzler, Author Abigail Freeman, and Author Elizabeth M. Petty declare that they have no conflict of interest.

ETHICS STATEMENT

Human Studies and Informed Consent: This study was reviewed and granted an exemption by the University of Wisconsin‐Madison Institutional Review Board. All applicable international, national, and institutional guidelines were followed. Informed consent was obtained from individuals who voluntarily completed the online survey and submitted their responses.

Animal Studies: No non‐human animal studies were performed by the authors for this paper.

Supporting information

Appendix S1

Pederson, V. , Rietzler, J. , Freeman, A. , & Petty, E. M. (2024). Picture this: Evaluating the efficacy of genetic counseling visual aids. Journal of Genetic Counseling, 33, 1365–1374. 10.1002/jgc4.1862

DATA AVAILABILITY STATEMENT

Data used to generate these analyses are not publicly available.