Article
Chatbot versus flowchart: Are interactive decision support tools superior to static ones?
Search Medline for
Authors
Published: | September 24, 2021 |
---|
Outline
Text
Introduction: Laypersons with symptoms indicative of COVID-19 infection, commonly seek guidance on whether and where to seek medical care, and whether self-isolation is required of them. Numerous web-based decision support tools (DSTs) have been developed to assist in decision-making. Few studies already report benefits of such tools [1], [2], [3]. Though most of these tools’ underlying algorithms are simple and similar, reproducing the local health authority’s guidelines (eg, Center for Disease Control (CDC) [4]), their mode of presentation differs: some DSTs present the decision tree algorithm as a static flowchart [5], [6], [7], [8], [9], while others are designed as interactive chatbots [10], [11], [12], [13], guiding the user through the decision tree’s nodes step-by-step. Our study assesses whether interactive DSTs for laypersons with suspected COVID-19 disease are superior to static ones in terms of their quality of decision support.
Methods: We developed a static flowchart and an interactive prototype designed with InVision [14] as mock DSTs. Their underlying algorithm was identical and based on the CDC’s official guidelines. The layout mimicked a patient-facing, freely available DST [11]. We recruited adult US residents for an online survey in which participants evaluated seven fictitious descriptions of patients (case vignettes) constructed by us, some with and some without signs indicative of COVID-19. Participants were tasked with appraising the appropriate social and help-seeking behavior for each vignette. Participants randomly assigned to the control group had to solve the case vignettes without support while participants in the experimental groups received either of the two mock DST as support. The main outcome measures were perceived certainty in making decisions (measured with the Decisional Conflict Subscale [15]) and accuracy of decisions according to health authority recommendations. Additionally, the experimental groups were asked about their perceptions of the DSTs (i.a., self-reported trust). We used ANOVAs and subsequent t-tests to determine statistical significance.
Results: We recruited 196 participants. The mean accuracy of participants’ decisions was higher in the experimental group (mean (SD) of total number of correct decision (max=14): Static DST=11.45 (2.48); Interactive DST=11.71 (2.37)) than in the control group (10.17 (2.0)). Decisional certainty was significantly higher in the experimental groups, too. Differences for both measures between the two experimental groups were non-significant in post-hoc t-tests. Scores for user’s trust in the tool did not differ significantly between DST groups either.
Discussion: Benefits of the static mock DST were equal to that of the interactive mock DST when providing the same advice. As static flowcharts reveal their underlying decision algorithm more transparently, they might prove to be more suitable in guiding laypersons through the healthcare systems by communicating their reasoning and thereby empowering patients. Further research should investigate when decisions become too complex for static DST to be on par with interactive ones.
Conclusions: Our experiment shows that DSTs guiding laypersons in their decision-making on currently common decisions improve users’ accuracy and certainty in decision-making. When the decision space is limited, a static flowchart is potentially as suitable as an interactive tool in enhancing the decision quality of laypersons.
The authors declare that they have no competing interests.
The authors declare that a positive ethics committee vote has been obtained.
References
- 1.
- Lunn PD, Timmons S, Julienne H, Belton CA, Barjaková M, Lavin C et al. Using decision aids to support self-isolation during the COVID-19 pandemic. Psychol Health. 2021; 36(2):195–213.
- 2.
- Judson TJ, Odisho AY, Neinstein AB, Chao J, Williams A, Miller C, et al. Rapid Design and Implementation of an Integrated Patient Self-Triage and Self-Scheduling Tool for COVID-19. Journal of the American Medical Informatics Association: JAMIA. 2020 Jun 1;27(6):860-866. DOI: 10.1093/jamia/ocaa051
- 3.
- Hautz WE, Exadaktylos A, Sauter TC. Online forward triage during the COVID-19 outbreak. Emerg Med J. 2021; 38(2):106–8. DOI: 10.1136/emermed-2020-209792
- 4.
- Centers for Disease Control and Prevention. What to Do If You Are Sick. 2021 [cited 2021 May 5]. Available from: https://www.cdc.gov/coronavirus/2019-ncov/if-you-are-sick/steps-when-sick.html
- 5.
- Michigangov. Contact Tracing Flow Chart: For potential and actual exposures to COVID-19. 2020 [cited 2021 May 2]. Available from: https://www.michigan.gov/documents/coronavirus/Person_Exposure_Final_3-25-2020_684832_7.pdf
- 6.
- Jackson Free Press. COVID-19 Flowchart. 2020 [cited 2021 May 2]. Available from: https://www.jacksonfreepress.com/photos/2020/mar/14/36329/
- 7.
- North Dakota Department of Health. COVID-19 Pandemic: What should I do if I've been in close contact with someone…. 2020 [cited 2020 Oct 9]. Available from: https://www.kxnet.com/wp-content/uploads/sites/16/2020/04/EUjo2CyXsAQsskb-1.jpg
- 8.
- Centers for Disease Control and Prevention. Coronavirus Disease 2019 (COVID-19) Risk Assessment and Public Health Management Decision Making. 2020 [cited 2021 May 2]. Available from: https://dhr.idaho.gov/wp-content/uploads/2020/COVID-19/CDCDecisionMakingModel.pdf
- 9.
- Robert Koch Institut. COVID-19: Bin ich betroffen und was ist zu tun? Orientierungshilfe für Bürgerinnen und Bürger. 2020 [cited 2021 May 2]. Available from: https://edoc.rki.de/bitstream/handle/176904/6779.2/Orientierungshilfe_COVID_Buerger_20-05-06_doi_10.25646-6629-2_korr.pdf?sequence=1&isAllowed=y
- 10.
- Centers for Disease Control and Prevention. Coronavirus Self-Checker. 2020 [cited 2021 May 2]. Available from: https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html
- 11.
- The Johns Hopkins University; The Johns Hopkins Hospital; The Johns Hopkins Health System Corporation. Coronavirus Self-Checker and COVID-19 Vaccine FAQ. 2020 [cited 2021 May 2]. Available from: https://www.hopkinsmedicine.org/coronavirus/covid-19-self-checker.html
- 12.
- Apple Inc. COVID-19-Screening-Tool. 2020 [cited 2021 May 4]. Available from: https://covid19.apple.com/screening/
- 13.
- Ada Health GmbH. COVID-19 Screener. 2020 [cited 2021 May 2]. Available from: https://ada.com/de/covid-19-screener/#assessment
- 14.
- InVision. New York, USA: InVisionApp Inc. Available from: https://www.invisionapp.com/
- 15.
- O’Connor AM. Validation of a decisional conflict scale. Medical decision making: an international journal of the Society for Medical Decision Making. 1995; 15(1):25-30.