Limits...
Informing the Design of Direct-to-Consumer Interactive Personal Genomics Reports.

Shaer O, Nov O, Okerlund J, Balestra M, Stowell E, Ascher L, Bi J, Schlenker C, Ball M - J. Med. Internet Res. (2015)

Bottom Line: In particular, results showed that a new interactive bubble chart visualization designed for the study resulted in the highest comprehension scores, as well as the highest perceived comprehension scores.These scores were significantly higher than scores received using the industry standard tabular reports currently used for communicating personal genomic information.We discuss implications for designers and researchers.

View Article: PubMed Central - HTML - PubMed

Affiliation: Human-Computer Interaction Lab, Computer Science Department, Wellesley College, Wellesley, MA, United States.

ABSTRACT

Background: In recent years, people who sought direct-to-consumer genetic testing services have been increasingly confronted with an unprecedented amount of personal genomic information, which influences their decisions, emotional state, and well-being. However, these users of direct-to-consumer genetic services, who vary in their education and interests, frequently have little relevant experience or tools for understanding, reasoning about, and interacting with their personal genomic data. Online interactive techniques can play a central role in making personal genomic data useful for these users.

Objective: We sought to (1) identify the needs of diverse users as they make sense of their personal genomic data, (2) consequently develop effective interactive visualizations of genomic trait data to address these users' needs, and (3) evaluate the effectiveness of the developed visualizations in facilitating comprehension.

Methods: The first two user studies, conducted with 63 volunteers in the Personal Genome Project and with 36 personal genomic users who participated in a design workshop, respectively, employed surveys and interviews to identify the needs and expectations of diverse users. Building on the two initial studies, the third study was conducted with 730 Amazon Mechanical Turk users and employed a controlled experimental design to examine the effectiveness of different design interventions on user comprehension.

Results: The first two studies identified searching, comparing, sharing, and organizing data as fundamental to users' understanding of personal genomic data. The third study demonstrated that interactive and visual design interventions could improve the understandability of personal genomic reports for consumers. In particular, results showed that a new interactive bubble chart visualization designed for the study resulted in the highest comprehension scores, as well as the highest perceived comprehension scores. These scores were significantly higher than scores received using the industry standard tabular reports currently used for communicating personal genomic information.

Conclusions: Drawing on multiple research methods and populations, the findings of the studies reported in this paper offer deep understanding of users' needs and practices, and demonstrate that interactive online design interventions can improve the understandability of personal genomic reports for consumers. We discuss implications for designers and researchers.

No MeSH data available.


Two-level treemap prototype visualization of genetic variants. The top screen is the landing page for the visualization, whereas the bottom screen shows what happens when a higher-level rectangle is clicked on. Red represents pathogenic impact. Green represents protective impact.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4526936&req=5

figure2: Two-level treemap prototype visualization of genetic variants. The top screen is the landing page for the visualization, whereas the bottom screen shows what happens when a higher-level rectangle is clicked on. Red represents pathogenic impact. Green represents protective impact.

Mentions: Following a brief presentation that reviewed the goals of our research, 36 PGP volunteers—15 female (42%), aged 21 to 83 with an average age of 45 (SD 19)—were recruited to participate. We conducted in-depth, semistructured interviews with each participant. We asked users to explain their goals in engaging with personal genomic information, to share their information practices, and to show us how they use tools to learn from their data. We also asked participants to walk us through their workflow as they explore their personal GET-Evidence report (see Figure 1) [28]. Finally, to elicit ideas about new ways for visualizing and interacting with personal genomics, we presented users with a treemap visualization (see Figure 2) [29] of their own personal genomic data. Participants’ personal genomic data were retrieved from the PGP public database. We chose treemaps as a starting point for a discussion about new ways for presenting personal genomic data because they have been successfully applied to the visualization of gene ontologies [30]. Their application to personal genomics for use by consumers, however, is new. We asked users to compare the tabular report with the new visual report and to suggest further ideas that could improve their engagement with the data.


Informing the Design of Direct-to-Consumer Interactive Personal Genomics Reports.

Shaer O, Nov O, Okerlund J, Balestra M, Stowell E, Ascher L, Bi J, Schlenker C, Ball M - J. Med. Internet Res. (2015)

Two-level treemap prototype visualization of genetic variants. The top screen is the landing page for the visualization, whereas the bottom screen shows what happens when a higher-level rectangle is clicked on. Red represents pathogenic impact. Green represents protective impact.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4526936&req=5

figure2: Two-level treemap prototype visualization of genetic variants. The top screen is the landing page for the visualization, whereas the bottom screen shows what happens when a higher-level rectangle is clicked on. Red represents pathogenic impact. Green represents protective impact.
Mentions: Following a brief presentation that reviewed the goals of our research, 36 PGP volunteers—15 female (42%), aged 21 to 83 with an average age of 45 (SD 19)—were recruited to participate. We conducted in-depth, semistructured interviews with each participant. We asked users to explain their goals in engaging with personal genomic information, to share their information practices, and to show us how they use tools to learn from their data. We also asked participants to walk us through their workflow as they explore their personal GET-Evidence report (see Figure 1) [28]. Finally, to elicit ideas about new ways for visualizing and interacting with personal genomics, we presented users with a treemap visualization (see Figure 2) [29] of their own personal genomic data. Participants’ personal genomic data were retrieved from the PGP public database. We chose treemaps as a starting point for a discussion about new ways for presenting personal genomic data because they have been successfully applied to the visualization of gene ontologies [30]. Their application to personal genomics for use by consumers, however, is new. We asked users to compare the tabular report with the new visual report and to suggest further ideas that could improve their engagement with the data.

Bottom Line: In particular, results showed that a new interactive bubble chart visualization designed for the study resulted in the highest comprehension scores, as well as the highest perceived comprehension scores.These scores were significantly higher than scores received using the industry standard tabular reports currently used for communicating personal genomic information.We discuss implications for designers and researchers.

View Article: PubMed Central - HTML - PubMed

Affiliation: Human-Computer Interaction Lab, Computer Science Department, Wellesley College, Wellesley, MA, United States.

ABSTRACT

Background: In recent years, people who sought direct-to-consumer genetic testing services have been increasingly confronted with an unprecedented amount of personal genomic information, which influences their decisions, emotional state, and well-being. However, these users of direct-to-consumer genetic services, who vary in their education and interests, frequently have little relevant experience or tools for understanding, reasoning about, and interacting with their personal genomic data. Online interactive techniques can play a central role in making personal genomic data useful for these users.

Objective: We sought to (1) identify the needs of diverse users as they make sense of their personal genomic data, (2) consequently develop effective interactive visualizations of genomic trait data to address these users' needs, and (3) evaluate the effectiveness of the developed visualizations in facilitating comprehension.

Methods: The first two user studies, conducted with 63 volunteers in the Personal Genome Project and with 36 personal genomic users who participated in a design workshop, respectively, employed surveys and interviews to identify the needs and expectations of diverse users. Building on the two initial studies, the third study was conducted with 730 Amazon Mechanical Turk users and employed a controlled experimental design to examine the effectiveness of different design interventions on user comprehension.

Results: The first two studies identified searching, comparing, sharing, and organizing data as fundamental to users' understanding of personal genomic data. The third study demonstrated that interactive and visual design interventions could improve the understandability of personal genomic reports for consumers. In particular, results showed that a new interactive bubble chart visualization designed for the study resulted in the highest comprehension scores, as well as the highest perceived comprehension scores. These scores were significantly higher than scores received using the industry standard tabular reports currently used for communicating personal genomic information.

Conclusions: Drawing on multiple research methods and populations, the findings of the studies reported in this paper offer deep understanding of users' needs and practices, and demonstrate that interactive online design interventions can improve the understandability of personal genomic reports for consumers. We discuss implications for designers and researchers.

No MeSH data available.