Background: During the COVID-19 pandemic, medical laypersons with symptoms indicative of a COVID-19 infection commonly sought guidance on whether and where to find medical care. Numerous web-based decision support tools (DSTs) have been developed, both by public and commercial stakeholders, to assist their decision making. Though most of the DSTs’ underlying algorithms are similar and simple decision trees, their mode of presentation differs: some DSTs present a static flowchart, while others are designed as a conversational agent, guiding the user through the decision tree’s nodes step-by-step in an interactive manner.
Objective: This study aims to investigate whether interactive DSTs provide greater decision support than noninteractive (ie, static) flowcharts.
Methods: We developed mock interfaces for 2 DSTs (1 static, 1 interactive), mimicking patient-facing, freely available DSTs for COVID-19-related self-assessment. Their underlying algorithm was identical and based on the Centers for Disease Control and Prevention’s guidelines. We recruited adult US residents online in November 2020. Participants appraised the appropriate social and care-seeking behavior for 7 fictitious descriptions of patients (case vignettes). Participants in the experimental groups received either the static or the interactive mock DST as support, while the control group appraised the case vignettes unsupported. We determined participants’ accuracy, decision certainty (after deciding), and mental effort to measure the quality of decision support. Participants’ ratings of the DSTs’ usefulness, ease of use, trust, and future intention to use the tools served as measures to analyze differences in participants’ perception of the tools. We used ANOVAs and t tests to assess statistical significance.
Results: Our survey yielded 196 responses. The mean number of correct assessments was higher in the intervention groups (interactive DST group: mean 11.71, SD 2.37; static DST group: mean 11.45, SD 2.48) than in the control group (mean 10.17, SD 2.00). Decisional certainty was significantly higher in the experimental groups (interactive DST group: mean 80.7%, SD 14.1%; static DST group: mean 80.5%, SD 15.8%) compared to the control group (mean 65.8%, SD 20.8%). The differences in these measures proved statistically significant in t tests comparing each intervention group with the control group (P<.001 for all 4 t tests). ANOVA detected no significant differences regarding mental effort between the 3 study groups. Differences between the 2 intervention groups were of small effect sizes and nonsignificant for all 3 measures of the quality of decision support and most measures of participants’ perception of the DSTs.
Conclusions: When the decision space is limited, as is the case in common COVID-19 self-assessment DSTs, static flowcharts might prove as beneficial in enhancing decision quality as interactive tools. Given that static flowcharts reveal the underlying decision algorithm more transparently and require less effort to develop, they might prove more efficient in providing guidance to the public. Further research should validate our findings on different use cases, elaborate on the trade-off between transparency and convenience in DSTs, and investigate whether subgroups of users benefit more with 1 type of user interface than the other.
Trial Registration: Deutsches Register Klinischer Studien DRKS00028136; https://tinyurl.com/4bcfausx (retrospectively registered)
In December 2019, the spread of the novel lung disease COVID-19 caused by SARS-CoV-2 with previously unknown etiology was detected, and it developed into a global pandemic within a few weeks [, ]. The disease courses of COVID-19 are heterogeneous. On the one hand, it is dangerous and can be lethal, even among previously healthy individuals. On the other hand, COVID-19 can present itself with unspecific and mild symptoms [ ]. Given the dynamic development of the disease and public health measures, laypersons often felt uncertain about appropriate behavior, especially when a COVID-19 infection was suspected but not proven. This situation resulted in considerable uncertainty both about what level of medical care was needed (care-seeking behavior) and whether isolation or quarantine was required (social behavior). As a result, capacity utilization in the health care system increased [ ].
The increased demand for medical advice could not be met through traditional routes (eg, telephone consultation, visit to the general practitioner, visit to the emergency room, local health authorities). As a consequence, various online services with information and self-assessments (ie, patient-facing clinical decision support tools [DSTs]) were developed. These DSTs are comparable to commonly known symptom checkers (SCs), which “are tools developed to provide support to laypersons. Users can enter their complaints and, with some SCs, demographic or health-related information (eg, age, gender, past medical history) to obtain advice on the urgency of their complaints (triage advice) and the most likely diagnosis” . The DSTs considered in the context of COVID-19 often require information about possible exposure to the virus, in addition to standard inputs related to gender, age, and major complaints. As output, the DSTs provide the probability of a COVID-19 infection rather than a diagnosis. Though most of the underlying algorithms are rule-based, simple, and similar decision trees, implementing official guidelines (eg, of the Centers for Disease Control and Prevention [CDC] [ ]), their mode of presentation differs: Some DSTs present the rule-based decision tree algorithm as a static flowchart [ - ], while others are designed as conversational agents (ie, similar to chatbots) with a graphical user interface, which guides the user through the decision tree’s nodes step-by-step in an interactive manner [ - ].
Our review of the literature suggests that there is no research, neither on SCs nor on DSTs for COVID-19, indicating whether and how the interactivity of a user interface influences the decision outcome and decision support experience. Currently available studies assessing DSTs for COVID-19 report no findings on the influence between user interaction and quality of decision support [, - ]. However, many publications on so-called patient decision aids (PtDAs) exist that have examined web-based and paper-based decision aids. In these studies, web-based PtDAs showed no difference from paper-based PtDAs in their effect on participants’ decision making [ ]. In terms of web-based PtDAs, interactive and static formats have not yet been compared to the best of our knowledge.
However, the comparison between web-based tools is particularly relevant during the COVID-19 pandemic, as laypersons’ decision making should ideally be supported at home to minimize the risks of unnecessary COVID-19 infections. The aim of this study is to assess whether static and interactive DSTs increase laypersons’ accuracy and confidence when making COVID-19-related decisions via the internet and whether 1 type of DST (static or interactive) is superior to the other.
We chose a between-subjects design with 1 three-level independent variable: participants had to solve fictional case vignettes while receiving different types of decision support (no support in the control group, a DST in the form of a static flowchart in the first intervention group, and an interactive DST in the second intervention group). Participants were randomly assigned to 1 of the 3 groups. Decision support quality was evaluated by analyzing the tools’ effects on decision making and the participants’ judgement of the tools with multiple dependent variables: Regarding effects on decision making, the number of correct appraisals (using official CDC recommendations from October 14, 2020, as the standard ), perceived certainty regarding the appraisal, and mental effort in making the appraisals were examined. Participants' judgments were collected on tool usefulness, ease of use, trust in the tools’ recommendations, and intention to use the tools in the future.
Ethics Approval and Consent to Participate
The Ethics Committee of the Department of Psychology and Ergonomics (IPA) at the Technische Universität Berlin approved this study (tracking number ROEB_01_20200715). Participants volunteered to take part in the survey and provided informed consent before participating.
Development of the Decision Support Tools
We designed 2 mock DSTs. The design was intended to mimic existing tools available on the internet. Features of various tools related to COVID-19 were analyzed. Relevant common features of these tools were the use of short, precise questions or statements to which the participants could respond in agreement (yes) or disagreement (no), and sometimes also with uncertainty (unsure). These questions guided participants through a decision tree’s nodes step-by-step until a recommendation could be provided. Input requirements and output content were mostly based on recommendations from public health agencies, such as the CDC. For the purpose of our study, the John Hopkins University Coronavirus Self-Checker was identified as a most suitable template because it is a decision tree–based interactive tool that requires a small amount of information at each step as input . This facilitated the construction of a second DST providing identical advice in the form of a static flowchart, outlining the entire decision tree at once. In both cases, the content was based on the CDC recommendations from October 14, 2020 [ ].
Considering these aspects, we developed 2 mock DSTs with identical content that differed only with respect to the interactivity of their user interface: a static flowchart where participants could follow the arrows of the appropriate path using their gaze (). This flowchart was developed using Microsoft PowerPoint [ ]. The other tool was implemented as a simple, interactive conversational agent, where participants clicked buttons to respond to questions, guiding them through the nodes of the decision tree step-by-step ( ). The individual screens of the interactive DST were also designed using Microsoft PowerPoint and then linked together using InVision [ ], enabling dynamic interaction. Mimicking the output of existing tools, 5 different recommendations were possible: “emergency care is required,” “self-care and physical distancing are sufficient,” “self-care is sufficient and quarantine necessary,” “self-care is sufficient and isolation is necessary,” and “nonemergency care and isolation are required.” We anticipated that laypersons would struggle comprehending the difference between “quarantine” and “isolation,” yet we decided to differentiate between them, as some COVID-19 DSTs use these terms without explaining them [ , ]. During data analyses, however, we rated answers as correct when participants confused these 2 terms (see the Data Analysis section). We developed mock tools rather than choosing existing DSTs from the internet in order to ensure comparability of the algorithm and design of the tools, as well as stable integration into the survey throughout the surveying period.
Preparing the Case Vignettes
We created 7 short patient descriptions (case vignettes). In this process, we extracted decision criteria identified by the CDC and its recommendations concerning help-seeking and social behavior in the case of a potential COVID-19 infection . Decision criteria included the presence of typical COVID-19 symptoms with the following expressions: 0, no symptoms; 1, primary symptoms (typical for a COVID-19 infection); and 2, secondary symptoms (can also occur in a COVID-19 infection but are less typical) (see ) [ ]. Another decision criterion represented the potential contact with the pathogen SARS-CoV-2 and the following expressions: 1, critical contact with a confirmed infected person has occurred; 2, contact may have occurred; or 3, critical contact can be excluded. The last important criterion is the presence or absence of risk factors classified by the Robert Koch Institute (RKI), Germany’s federal public health authority (eg, certain pre-existing conditions such as chronic lung disease; see ) [ ]. Through a pretest with 15 participants, in which we evaluated the response behavior of the participants as well as their free-text feedback, we revised the wording of the vignettes, improving intelligibility and removing errors spotted by the pretest participants. The vignettes included more information than necessary for the appraisal to better simulate a real-life decisional context and increase ecological validity. We also ensured that all information asked for in the developed DSTs was included in the vignettes and that the fictitious patients represented diverse age and gender groups. In addition, each possible outcome of the mock tools was covered at least once as a correct solution. and contain an outline of the 7 case vignettes.
All participants lived in the United States, were at least 18 years old, and had no professional medical background. Our investigation was limited to US residents, as recommendations of the developed DSTs were based on CDC recommendations and guidelines and thus applied to the US region.
Survey and Instruments
We created an online survey using UNIPARK (QuestBack GmbH, Germany) . To ensure differences between the 2 groups were not due to confounding variables, we surveyed variables we suspected to have an influence on the main outcome measures, namely sociodemographic factors, affinity toward interacting with technology, prior knowledge of COVID-19, prior use of a COVID-19 DST, and perceived threat of a COVID-19 infection. The sociodemographic variables were age, gender, US residency, and highest level of completed formal education. Affinity toward interacting with technology was surveyed with the Affinity for Technology Interaction Short Scale (ATI-S) [ ]. The perceived threat of COVID-19 was surveyed with an instrument developed by Kim and Park [ ]. Participants’ prior knowledge of COVID-19 necessary to appraise the adequate health-seeking and social behavior for a suspected COVID-19 patient was assessed with 5 self-developed multiple-choice questions (see ).
In the survey’s main part, participants were presented the 7 case vignettes in random order. In both the experimental groups and the control group, participants provided their personal appraisal of the adequate help-seeking behavior for each case vignette and, in a further question, of the adequate social behavior. In the 2 experimental groups, participants were also prompted to state which recommendations of the tool (supposedly) provided concerned help-seeking behavior and social behavior (seeand ).
As we anticipated participants commonly erring in determining the appropriate social behavior when required to differentiate between isolation and quarantine, we deemed both answers correct for our main analysis and conducted a second analysis without this adjustment ().
After each case vignette, participants were further asked to rate their mental effort required in making decisions concerning the respective fictitious case presentation using a 9-point category scale ranging from very, very low mental effort to very, very high mental effort (see) [ ]. This scale by Paas et al [ ] can be considered a “subjective, indirect measure of cognitive load” in making decisions regarding case vignettes. Since mental effort increases with higher perceived demands of a stimulus or task [ ], it was included here as an indicator of the quality of decision support. We assume that a good DST guides people to the right decision without requiring high cognitive load.
Following all 7 case vignettes, participants’ decisional uncertainty was assessed once using the O'Connor  Decisional Conflict Subscale. On 3 items with a 5-point scale, participants rated how confident they were in their decisions (see ). The scale was developed to evaluate health care consumer decision aids. We used this scale as an indicator of the quality of decision support, since we assume that a good DST helps medical laypersons to feel confident in their decisions. This assumption is supported by O'Connor's [ ] claim that “decision aids should reduce uncertainty and confusion in choosing a course of action.”
In addition to the effect of the tools on decision making and participants' perceptions of their decision making, we also asked directly about participant's perceptions of the tools in the 2 intervention groups. They were asked about the perceived usefulness and ease of use of the tools after having worked on all 7 cases. Perceived usefulness is defined as the individual's perception of the extent to which using the tool improves decision-making performance. Perceived ease of use implies that using the tool does not require any effort . Both constructs were measured using scales from the Davis Technology Acceptance Model [ ]. In addition, the intentions to use the DSTs in the future and the trust in the recommendations of the DSTs were measured as a subjective rating using 1 item each ( presents the exact phrasing of the items).
We recruited participants via the crowdsourcing platform Prolific.co . This platform is aimed at researchers and allows them to conveniently recruit participants for online surveys and experiments. Prolific.co provides a pool of participants that can be screened based on demographic data. Researchers can make their survey or experiment available on Prolific.co, and users registered on Prolific.co as prospective participants can choose whether to participate in studies that they are eligible for according to the set prescreening criteria. We chose Prolific.co as it is characterized by a diverse population that provides high-quality data [ ]. Each participant received a remuneration of US $2.00 for completing the survey. In addition, participants could earn a bonus of US $0.20 for each correct decision. To ensure that all participants answered the questions attentively, attention checks were added to the survey. Participants failing more than 1 of 3 attention checks were excluded. Furthermore, at the end of the survey, participants were asked to self-report whether they had participated in the survey attentively, honestly, and without external assistance, as suggested by Rouse [ ]. Participants were remunerated independently of their answer, but only data from participants confirming this question were included in the analysis. By selecting the weekend day and early afternoon PDTs, we attempted to recruit a population as diverse as possible, as suggested by Casey et al [ ]. The survey was released for participation on Prolific.co on 4 different days (November 21, 2020, at 1 PM Pacific Daylight Time [PDT]; November 22, 2020, at 11 AM PDT; November 28, 2020, at 12 PM PDT; and November 29, 2020, at 1 PM PDT). On each day, 50 participants were recruited within a few hours after release.
Data were cleaned and explored using R 4.0.2  and the tidyverse packages [ ]. After examining the central tendencies and distributions of the variables separately for the 3 groups and plotting them using the ggplot2 package [ ], we performed a one-way analysis of variance to compare the 3 conditions separately for the dependent measures accuracy, decisional certainty, and mental effort. If significant group differences were present, we followed up with Bonferroni-corrected post hoc pairwise t tests. For the dependent variables’ usefulness, ease of use, trust in tool recommendations, and future use intention, we conducted Welch two-sample t tests to compare the 2 intervention groups because the sample sizes of the groups were unequal [ ]. Effect sizes were quantified with Cohen d and calculated using the rstatix package [ ]. When results were not statistically significant, we performed a post hoc power analysis using the pwr package [ ] with corresponding group sizes, an α level of .05, and a power 1-β of .80.
In total, our survey was accessed 233 times during the 4 days it was available. Overall, 37 participations could not be used for data analysis because they dropped out of the questionnaire (n=12, 32%), exceeded the maximum survey completion time (n=10, 27%), failed attention checks (n=6, 16%), or did not meet eligibility criteria (n=9, 25%). The remaining participants all affirmed that they had paid close attention during the survey and answered honestly. This yielded a total of 196 (84.1%) participants, who assessed all 7 case vignettes and could be included in the analysis (seefor details). The mean time for completion of the survey was 22 minutes and 3 seconds. Across the 3 groups, the participant characteristics were similar; see .
|Characteristics||Total sample||Group 1: control group (no DST)||Group 2: static DST||Group 3: interactive DST|
|Sample size, N||196||66||62||68|
|Age (years), median (IQR)||30 (18)||30 (17.2)||26.5 (13.2)||33 (20.5)|
|Gender, n (%)|
|Female||94 (48)||31 (47)||27 (44)||36 (53)|
|Male||100 (51)||33 (50)||35 (56)||32 (47)|
|Other||2 (1)||2 (3)||0||0|
|Education, n (%)|
|Non-high-school graduate||4 (2)||1 (2)||1 (2)||2 (3)|
|High school graduate||34 (17)||9 (14)||18 (29)||7 (10)|
|Some college||66 (34)||22 (33)||20 (32)||24 (35)|
|Bachelor’s degree||49 (25)||20 (30)||12 (19)||17 (25)|
|Graduate degree or higher||43 (22)||14 (21)||11 (18)||18 (26)|
|Recentb experience with COVID-19 decisions, n (%)|
|Recently faced a triage decision||70 (35)||25 (38)||27 (44)||18 (26)|
|Recently faced a social behavior decision||101 (52)||35 (53)||32 (52)||34 (50)|
|Recently consulted a static DST to face the decision||56 (29)||15 (23)||21 (34)||20 (29)|
|Recently consulted an interactive DST to face the decision||53 (27)||21 (32)||15 (24)||17 (25)|
|Medical training, n (%)|
|No training||163 (83)||55 (83)||51 (82)||57 (84)|
|Basic first-aid training||33 (17)||11 (17)||11 (18)||11 (16)|
|Affinity for technology interaction on a scale of 1-6, median (IQR)c||4 (1.2)||4 (1.7)||4 (1)||4 (1)|
|Perceived threat of COVID-19 on a scale of 1-7, median (IQR)d||5 (2)||5.25 (1.6)||5 (2.1)||5 (2)|
|Prior knowledge of COVID-19 on a scale of 0-5, median (IQR)e||3 (2)||3 (2)||3 (1)||3 (1)|
aDST: decision support tool.
b“Recent” was defined as “in the past 6 months.”
cMeasured by the Wessel Affinity for Technology Interaction Short Scale (ATI-S).
d Measured by a subjective self-assessment on 2 items on a scale of 1-7 adapted from Kim and Park .
eMeasured by the number of correctly answered multiple-choice questions with reference to COVID-19.
Effects on Decision Making
Omnibus ANOVAs detected group differences with respect to decision accuracy (overall and separately for decisions about help-seeking behavior and social behavior) and perceived certainty but not for mental effort in decision making; see.
|Dependent variables||Group 1: without DST||Group 2: static DST||Group 3: interactive DST||Test statistics of group comparison|
|Accuracy, mean (SD)|
|Total correct decisions (min=0, max=14)b||10.17 (2.00)||11.45 (2.48)||11.71 (2.37)||8.59||<.001||0.08|
|Correct decisions on help-seeking behavior (min=0, max=7)||4.82 (0.96)||5.47 (1.39)||5.54 (1.40)||6.58||<.001||0.002|
|Correct decisions on social behavior (min=0, max=7)b||5.35 (1.41)||5.98 (1.26)||6.16 (1.18)||7.33||<.001||0.02|
|Decisional certainty: certainty score from 0 to 100c, mean (SD)||65.78 (20.78)||80.51 (15.89)||80.72 (14.08)||15.67||<.001||0.14|
|Mental effort on a scale from 1 to 9, mean (SD)||5.62 (1.57)||5.40 (1.68)||5.09 (1.78)||1.73||.18||N/Ad|
aDST: decision support tool.
bThe response options “quarantine” and “isolation” were not differentiated for this analysis, that is, they were both considered correct because layperson participants commonly confuse these terms. An analysis without this adjustment is provided inand shows the same trend.
cResponses were transformed into a certainty score between 0 and 100 (ie, 0=person feels extremely uncertain about the best choice and 100=person feels extremely certain about the best choice).
dN/A: not applicable.
Across all groups, participants’ appraisals were commonly correct. The average participant decided correctly in more than 10 (70%) of the 14 decisions. Overall decision accuracy was higher in the experimental groups receiving support from a DST than in the unsupported control group; seeand . Bonferroni-corrected pairwise Welch two-sample t tests indicated statistically significant differences with moderate effect sizes between control and experimental groups (control group vs static DST group: t117.35=–3.21, P<.001, Cohen d=–0.57, 95% CI –1.1 to –0.14; control group vs interactive DST group: t129.6=–4.06, P<.001, Cohen d=–0.70, 95% CI –1.31 to –0.21). The difference in decision accuracy between the 2 experimental groups showed a negligible effect size and was not significant (t125.58=–0.59, P=.55, Cohen d=–0.10, 95% CI –0.54 to 0.34). Based on a post hoc power analysis, we estimated that on a population level, the real difference in accuracy was less than Cohen d=0.5 (equivalent to 1.22 correct responses), with a power of 0.8.
We found similar results when analyzing participants’ accuracy for help-seeking behavior and social behavior separately; see. When the differences between the response options “isolation” and “quarantine” were considered, the pattern remained the same, but the gap in accuracy between experimental and control groups widened; see .
Participants were commonly certain in their decision making. Less than 15 (8%) of the 196 participants indicated a certainty of less than 50 on a scale from 0 to 100, in contrast to 32 (16%) indicating maximum certainty (100/100). Participants’ certainty in their decisions differed between the 3 groups: certainty in decision making was rated lower by participants without decision support than by those receiving decision support; seeand . Bonferroni-corrected pairwise Welch two-sample t tests marked these differences as statistically significant with large effect sizes (control group vs static DST group: t121.1=–4.51, P<.001, Cohen d=–0.79, 95% CI –1.22 to –0.37; control group vs interactive DST group: t115.67=–4.67, P<.001, Cohen d=–0.81, 95% CI –1.27 to –0.44).
Decision certainty in the 2 experimental groups was nearly identical, with mean certainty scores of 80.51 (static DST) and 80.72 (interactive DST). The inferential analysis indicated this difference to be of negligible effect size and nonsignificant (t123.68=–0.09, P=.92, Cohen d=0.01, 95% CI –0.47 to 0.43). Based on a post hoc power analysis, we estimated that on a population level, the real difference in decisional certainty between both experimental groups was less than Cohen d=0.5 (equivalent to 7.5 percentage points), with a power of 0.8.
The perception of how demanding the task was varied among participants. Although a third (63/196, 32.1%) stated that the tasks required low mental effort (ie, they indicated an average mental effort score below 4.5 on a scale from 1 to 9), 4 (40%) of 10 considered the tasks to be at least somewhat demanding (indicating scores above 5.5).
Participants rated the mental effort of decision making lowest in the interactive DST group (mean 5.06) and highest in the control group unsupported by a DST (mean 5.62); see. However, the differences between group means were not significant in the omnibus ANOVA; see . Based on a post hoc power analysis, we estimated that on a population level, the effect size of the difference was not greater than η²=0.047, with a power of 0.8.
Participants’ Perceptions of the Tools
We investigated participants’ perceptions of the mock tools with 4 metrics: perceived usefulness, perceived ease of use, trust, and intention to use the tools again in the future. On average, participants perceived the DSTs positively: their self-reported trust in the tools was generally high in both experimental groups as was the number of DST recommendations the participants followed (on average, in more than 12 [86%] of a total of 14 recommendations); see. Accordingly, almost all (126/130, 95%) participants judged the mock tools to be useful (ie, indicated a perceived usefulness score of above 3.0 on a scale from 1 to 5), with half of the participants scoring the usefulness at or above 4.75. Ease of use was rated similarly positive (see ), and their intention to use the tools in the future was high, on average. Regarding differences between the 2 DSTs, the mock users’ scores on perceived usefulness, trust, and future use intention were higher in the static than in the interactive DST group. Ease of use was perceived higher in the interactive DST group. However, all effect sizes of differences between the 2 DST groups were small (Cohen d≤0.35) and proved significant only for future use intention in a Welch two-sample t test; see .
Based on a post hoc power analysis, we estimated that on a population level, the real difference in these variables between both groups was less than Cohen d=0.5 (equivalent to a difference of 0.34 for perceived usefulness and perceived ease of use, 0.42 for self-reported trust, 0.96 for followed decisions, and 0.51 for future use intention), with a power of 0.8.
|Dependent variables||Group 2: static DST||Group 3: interactive DST||Test statistics of group comparison|
|tdf||P value||Cohen d|
|Perceived usefulness on a scale from 1 to 5, mean (SD)||4.56 (0.65)||4.35 (0.71)||t128=1.7||.09||0.30|
|Perceived ease of use on a scale from 1 to 5, mean (SD)||4.26 (0.71)||4.47 (0.58)||t117.92=–1.90||.06||–0.34|
|Trust, mean (SD)|
|Self-reported trust in the tools\' recommendation on a scale from 1 to 7||6.08 (0.84)||5.83 (0.85)||t127.15=1.74||.08||0.31|
|Decisions where the recommendation of the tools was followed on a scale from 0 to 14||12.73 (1.81)||12.5 (2.03)||t127.94=0.67||.75||0.12|
|Future use intention on a scale from 1 to 7, mean (SD)||6.23 (0.88)||5.87 (1.16)||t123.91=1.99||.05||0.35|
aDST: decision support tool.
In our study, DSTs increased laypersons’ accuracy and certainty in decision making. Potentially, DSTs can also reduce the mental effort in decision making, but this effect was statistically nonsignificant in our experiment. Thus, our experiment confirms the benefit of DSTs for COVID-19-related self-triage discussed in prior studies [, ].
Regarding the question of whether 1 mode of presentation of DSTs (ie, static or interactive presentation) is superior to the other, our experiment produced evidence that differences in measures of the quality of decision support and participants’ perception of the tools are small. Interactive DSTs are potentially more convenient to use. Users of the interactive DST rated the mental effort as lower and the perceived ease of use as higher than the users of the static DST. However, these differences were not significant and did not translate into a higher level of trust, greater perceived usefulness, or greater intention to use the tool in the future. On the contrary, users rated the static tool more favorably on these measures than the interactive DST. Overall, the effect sizes for differences in these measures were low and statistically nonsignificant.
To the best of our knowledge, our study is among the first to directly compare the effectiveness of different modes of presentation on the quality of support received from web-based, patient-facing clinical DSTs. Our results are in line with findings from similar research in a different field of application: In their 2018 meta-analysis on the use of PtDAs to support decisions concerning prostate cancer screening, Baptista et al  concluded that web-based decision aids reduce decisional conflict compared to no decision aid, but no more than static, printed-out decision aids. Comparability between Baptista et al [ ] and our study is limited, as web-based decision aids and their paper-based printed-out versions assessed by Baptista et al [ ] do not directly match the mock conversational agent and the static flowchart we presented digitally in our study. However, both results suggest that for a decision scenario with clear-cut options, the mode of presentation with a DST has little to no effect on how helpful it is for its users.
First, our results underline the benefits of making DSTs available to laypersons for decisions with clear-cut options, such as those encountered in the COVID-19 pandemic, as both decisional accuracy and trust in the decision increased when laypersons were supported by both mock DSTs. Second, the quality of decision support provided by the static flowchart and the interactive tool did not differ significantly, and post hoc sensitivity analyses on effect sizes indicated that we can rule out large effects. Thus, factors such as development effort might potentially weigh more than the format of interaction when deciding on how to present decision support to laypersons.
When complexity is low, as in the case of the COVID-19 decisions, the static version may provide a better overview and thus make the decision more transparent to its users and be quicker and more cost-efficient to develop and publish than an interactive conversational agent. In contrast, for decisions with higher complexity, full transparency might compromise ease of use of a static DST. Thus, for more complex decisions, interactive tools that guide the user step-by-step may be more user friendly. However, our results suggest that interactivity is not an effective means in itself to increase the usefulness of a DST: First, the higher convenience of using the interactive DST did not translate into greater trust or perceived usefulness, nor did it yield a greater increase in decision accuracy or certainty than static flowcharts. As the latter entail less development efforts, while increasing transparency, they might be the preferable mode of DSTs for public health officials to implement on their websites to provide guidance to the public on decisions of low complexity and limited decision space.
Our study also raises 2 topics for further research. First, which factors increase users’ trust in DSTs, and how are they weighed? From our results, we can only speculate that users prefer transparency over convenience. Second, would subgroups of the population prefer 1 type of interface over the other? Although our results showed only minor differences between sample averages of users of the static and interactive DSTs, potentially subgroups of the population benefit more from 1 type of interface than the other. For example, users with low technological affinity might prefer a static flowchart over a conversational agent.
The results of our study are mainly limited by concerns about external validity. That is, the interactive mock DST we developed does not fully represent the whole variety of interactive tools that are available on the internet. Although our DST mimics the interactive tool from Johns Hopkins University, other tools incorporate significantly more decision factors and possible outcomes. Second, the sampled study population’s composition of educational background is not representative of the adult US population. Our study sample included a higher proportion of highly educated persons, with only 4 (2%) of 196 participants being nonhighschool graduates as compared to 9.8% in the US adult population . In addition, the mock interactive tool we developed for this paper does not fully exhaust the potential of interactivity, as our participants only followed prompts to respond to binary questions by clicking either Yes or No buttons, while more interactive tools also require users to enter information manually (eg, age or current location). Finally, participants in this study did not use the tool for self-assessment but to appraise cases with fictitious patients. This means that they did not experience personal concerns, as would likely be the case with COVID-19 suspicion in a real-world use situation. To promote external validity, the recommendations of the DSTs we developed conform with all CDC guidelines and their interactive capacities are basic but mimic those of existing DSTs.
When the decision space is limited, a static flowchart potentially performs just as well as an interactive tool in enhancing the decision quality of laypersons with symptoms indicative of a COVID-19 infection. As static flowcharts reveal their underlying decision algorithm more transparently, they might prove to be more suitable in not only guiding laypersons through the health care system but also communicating the reasoning and thereby empowering patients. Further research should validate our findings on different use cases, elaborate on the trade-off between transparency and convenience in DSTs, and investigate whether subgroups of users benefit more from 1 type of user interface than the other by assessing interactions between outcome variables (eg, accuracy, mental effort, perceived usefulness) and participant characteristics (eg, age, eHealth literacy). We have made all data necessary to conduct these exploratory analyses and to reproduce our reported findings publicly available .
The authors express their gratitude to the participants and to Eike Richter for his advice on statistical methods. The project was funded by the home institutions of the last authors (MAF and FB). No external funding was required to conduct this study. The authors acknowledge support from the German Research Foundation (DFG) and the Open Access Publication Fund of Charité - Universitätsmedizin Berlin.
AR and MLS conceived the study. AR created the questionnaire, developed the mock decision support tool (DSTs), designed and conducted the analyses, and wrote the first draft of the paper. MLS oversaw the creation of the case vignette adaptations and aided in the design of the methods and manuscript development. MK assisted in statistical analyses and manuscript development. MAF and FB provided critical input and advised on the study and questionnaire design, analysis methods, and drafts of the paper. AR and MLS share the first authorship. MAF and FB share the last authorship. All authors accept full responsibility for the final version of the paper.
The lead authors affirm that this manuscript is an honest, accurate, and transparent account of the study being reported, that no important aspects of the study have been omitted, and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Conflicts of Interest
All authors have completed the International Committee of Medical Journal Editors uniform disclosure form and declare no support from any organization for the submitted work, no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years, and no other relationships or activities that could appear to have influenced the submitted work.
This randomized study was only retrospectively registered, given that the authors believed registration was unnecessary for this kind of trial with a theoretical case example and the particular outcomes which are measured. The editor granted an exception from ICMJE rules mandating prospective registration of randomized trials. However, readers are advised to carefully assess the validity of any potential explicit or implicit claims related to primary outcomes or effectiveness, as retrospective registration does not prevent authors from changing their outcome measures retrospectively.
Static decision support flowchart for laypersons with suspected COVID-19 infection developed as part of the scientific study.PDF File (Adobe PDF File), 391 KB
Interactive decision support tool for laypersons with suspected COVID-19 infection developed as part of the scientific study.PDF File (Adobe PDF File), 254 KB
List of COVID-19 symptoms according to the Robert Koch Institute.DOCX File , 54 KB
List of risk factors for COVID-19 according to the Robert Koch Institute.DOCX File , 54 KB
Overview of the case vignettes developed as part of the scientific study.PDF File (Adobe PDF File), 262 KB
Outline of all 7 case vignettes developed as part of the scientific study.DOCX File , 57 KB
Multiple-choice questions used to assess the participants' prior knowledge.PDF File (Adobe PDF File), 154 KB
Decisions on each case vignette in the control group without decision support.PDF File (Adobe PDF File), 194 KB
Decisions on each case vignette in the intervention groups with decision support, here illustrated by the flowchart example.PDF File (Adobe PDF File), 281 KB
Single item for mental effort.PDF File (Adobe PDF File), 52 KB
Decisional Conflict Subscale.PDF File (Adobe PDF File), 72 KB
Items measuring participants’ perspectives on the 2 DSTs with items on ease of use, usefulness, trust, and future intention to use the DSTs. DST: decision support tool.PDF File (Adobe PDF File), 223 KB
Boxplot for decision accuracy in regard to decision support type. Responses where participants confused "quarantine" and "isolation" were rated as wrong in this analysis.PDF File (Adobe PDF File), 90 KB
CONSORT-eHEALTH checklist (V 1.6.1).PDF File (Adobe PDF File), 9667 KB
- Sohrabi C, Alsafi Z, O'Neill N, Khan M, Kerwan A, Al-Jabir A, et al. World Health Organization declares global emergency: a review of the 2019 novel coronavirus (COVID-19). Int J Surg 2020 Apr 26;76:71-76 [FREE Full text] [CrossRef] [Medline]
- World Health Organization. Pandemie der Coronavirus-Krankheit (COVID-19). URL: https://www.euro.who.int/de/health-topics/health-emergencies/coronavirus-covid-19 [accessed 2022-03-07]
- Epidemiologischer Steckbrief zu SARS-CoV-2 und COVID-19. Demografische Faktoren, Symptome und Krankheitsverlauf. Robert-Koch-Institut. URL: https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Steckbrief.html [accessed 2021-05-05]
- Judson T, Odisho A, Neinstein A, 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. J Am Med Inform Assoc 2020;27(6):866. [CrossRef]
- Schmieding ML, Mörgeli R, Schmieding MAL, Feufel MA, Balzer F. Benchmarking triage capability of symptom checkers against that of medical laypersons: survey study. J Med Internet Res 2021 Mar 10;23(3):e24475. [CrossRef]
- What to Do if You Are Sick. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/coronavirus/2019-ncov/if-you-are-sick/steps-when-sick.html [accessed 2021-05-05]
- Coronavirus Disease 2019 (COVID-19) Risk Assessment and Public Health Management Decision Making. Centers for Disease Control and Prevention. URL: https://dhr.idaho.gov/wp-content/uploads/2020/COVID-19/CDCDecisionMakingModel.pdf [accessed 2020-10-14]
- COVID-19 Flowchart. Jackson Free Press. URL: https://www.jacksonfreepress.com/photos/2020/mar/14/36329/ [accessed 2021-05-02]
- North Dakota Department of Health. What To Do If You Are A Close Contact. URL: https://www.health.nd.gov/what-do-if-you-are-close-contact [accessed 2022-03-07]
- COVID-19: Bin ich betroffen und was ist zu tun?: Orientierungshilfe für Bürgerinnen und Bürger. Robert-Koch-Institut. URL: 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 [accessed 2022-03-07]
- Contact Tracing Flow Chart: For potential and actual exposures to COVID-19. State of Michigan. URL: https://www.michigan.gov/documents/coronavirus/Person_Exposure_Final_3-25-2020_684832_7.pdf [accessed 2021-05-02]
- Coronavirus-Entscheidungshilfe. ZEIT ONLINE GmbH. URL: https://www.zeit.de/wissen/gesundheit/2020-03/Coronavirus-Entscheidungshilfe-2020-03-31.pdf [accessed 2021-05-02]
- Ada Health GmBH. COVID-19 Screener. 2020. URL: https://ada.com/de/covid-19-screener/#assessment [accessed 2022-03-07]
- Coronavirus Self-Checker. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html [accessed 2021-05-02]
- CovApp: Haben Sie Symptome? CovApp zeigt Ihnen mögliche nächste Schritte. Charité – Universitätsmedizin Berlin. URL: https://covapp.charite.de/ [accessed 2022-03-07]
- Emory University School of Medicine. Coronavirus Checker. URL: https://c19check.com/start [accessed 2022-03-07]
- Together against Coronavirus: COVID-GUIDE. in4medicine AG. URL: https://covidguide.health/de/ [accessed 2021-05-02]
- Coronavirus Self-Checker and COVID-19 Vaccine FAQ. The Johns Hopkins University, The Johns Hopkins Hospital, and The Johns Hopkins Health System Corporation. URL: https://www.hopkinsmedicine.org/coronavirus/covid-19-self-checker.html [accessed 2021-05-02]
- Universitätsklinikum Giessen und Marburg GmbH. COVID-ONLINE. URL: https://covid-online.de/#/ [accessed 2022-03-07]
- Covid Assessment. The USC Gehr Family Center for Health Systems Sciences and Innovation. URL: https://covidassessment.org/acknowledgement [accessed 2021-05-02]
- COVID-19 Symptom Checker. WebMD LLC. URL: https://www.webmd.com/coronavirus/coronavirus-assessment/default.htm [accessed 2021-05-02]
- Kahnbach L, Lehr D, Brandenburger J, Mallwitz T, Jent S, Hannibal S, et al. Quality and adoption of COVID-19 tracing apps and recommendations for development: systematic interdisciplinary review of European apps. J Med Internet Res 2021 Jun 02;23(6):e27989 [FREE Full text] [CrossRef] [Medline]
- 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 Feb 19;36(2):195-213. [CrossRef] [Medline]
- Munsch N, Martin A, Gruarin S, Nateqi J, Abdarahmane I, Weingartner-Ortner R, et al. Diagnostic accuracy of web-based COVID-19 symptom checkers: comparison study. J Med Internet Res 2020 Oct 06;22(10):e21299 [FREE Full text] [CrossRef] [Medline]
- Perlman A, Vodonos Zilberg A, Bak P, Dreyfuss M, Leventer-Roberts M, Vurembrand Y, et al. Characteristics and symptoms of app users seeking COVID-19-related digital health information and remote services: retrospective cohort study. J Med Internet Res 2020 Oct 20;22(10):e23197 [FREE Full text] [CrossRef] [Medline]
- Zens M, Brammertz A, Herpich J, Südkamp N, Hinterseer M. App-based tracking of self-reported COVID-19 symptoms: analysis of questionnaire data. J Med Internet Res 2020 Sep 9;22(9):e21956 [FREE Full text] [CrossRef] [Medline]
- Baptista S, Teles Sampaio E, Heleno B, Azevedo LF, Martins C. Web-based versus usual care and other formats of decision aids to support prostate cancer screening decisions: systematic review and meta-analysis. J Med Internet Res 2018 Jun 26;20(6):e228 [FREE Full text] [CrossRef] [Medline]
- If You Are Sick or Caring for Someone. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/coronavirus/2019-ncov/if-you-are-sick/index.html [accessed 2021-05-02]
- Microsoft Corporation. Microsoft PowerPoint (Version 2019). URL: https://www.microsoft.com/de-de/microsoft-365/powerpoint [accessed 2022-03-07]
- InVision. InVisionApp Inc. URL: https://www.invisionapp.com/ [accessed 2020-08-15]
- Über das Online-Befragungstool UNIPARK. Trivian XI GmbH. 2021. URL: https://www.unipark.com/ [accessed 2020-07-05]
- Wessel D, Attig C, Franke T. ATI-S: an ultra-short scale for assessing Affinity for Technology Interaction in user studies. 2019 Presented at: Proceedings of Mensch und Computer 2019; 2019; Hamburg Germany. [CrossRef]
- Kim J, Park H. Development of a health information technology acceptance model using consumers' health behavior intention. J Med Internet Res 2012 Oct 01;14(5):e133 [FREE Full text] [CrossRef] [Medline]
- Paas F, Tuovinen JE, Tabbers H, Van Gerven PWM. Cognitive load measurement as a means to advance cognitive load theory. Educ Psychol 2003 Mar;38(1):63-71. [CrossRef]
- Kirschner P, Kirschner F. Mental effort. In: Seel NM, editor. Encyclopedia of the Sciences of Learning. Boston, MA: Springer; 2012.
- O'Connor AM. Validation of a Decisional Conflict Scale. Med Decis Making 2016 Jul 02;15(1):25-30. [CrossRef]
- Ammenwerth E. Technology acceptance models in health informatics: TAMUTAUT. In: Scott P, de Keizer N, Georgiou A, editors. Applied Interdisciplinary Theory in Health Informatics. Amsterdam, the Netherlands: IOS Press BV; 2019:64.
- Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Q 1989 Sep;13(3):319. [CrossRef]
- Online Participant Recruitment For Surveys and Market Research. Prolific Academic. URL: https://www.prolific.co/ [accessed 2021-04-09]
- Peer E, Brandimarte L, Samat S, Acquisti A. Beyond the Turk: alternative platforms for crowdsourcing behavioral research. J Exp Soc Psychol 2017 May;70:153-163. [CrossRef]
- Rouse SV. A reliability analysis of Mechanical Turk data. Comput Hum Behav 2015 Feb;43:304-307. [CrossRef]
- Casey LS, Chandler J, Levine AS, Proctor A, Strolovitch DZ. Intertemporal differences among MTurk workers: time-based sample variations and implications for online data collection. SAGE Open 2017 Jun 14;7(2):215824401771277. [CrossRef]
- R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2021.
- Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the Tidyverse. JOSS 2019 Nov;4(43):1686. [CrossRef]
- Wickham H. ggplot2: 3.3.5. Vienna, Austria: R Foundation for Statistical Computing; 2021.
- Delacre M, Lakens D, Leys C. Why psychologists should by default use Welch’s t-test instead of Student’s t-test. Int Rev Soc Psychol 2017 Apr 05;30(1):92. [CrossRef]
- Kassambara A. rstatix: Pipe-Friendly Framework for Basic Statistical Tests. URL: https://CRAN.R-project.org/package=rstatix [accessed 2022-03-07]
- Champely S. pwr: Basic Functions for Power Analysis. URL: https://cran.r-project.org/package=pwr [accessed 2022-03-07]
- United States Census Bureau. Educational Attainment in the United States. URL: https://www.census.gov/data/tables/2020/demo/educational-attainment/cps-detailed-tables.html [accessed 2022-03-07]
- Röbbelen A, Schmieding ML, Kopka M, Balzer F, Feufel MA. Data set supplementing “Interactive versus static decision support tools for COVID-19: An experimental comparison. Zenodo 2021 Nov 6. [CrossRef]
|CDC: Centers for Disease Control and Prevention|
|DST: decision support tool|
|PtDA: patient decision aid|
|SC: symptom checker|
Edited by T Sanchez; submitted 23.09.21; peer-reviewed by J Michel, A Hill; comments to author 01.10.21; revised version received 21.11.21; accepted 30.11.21; published 15.04.22Copyright
©Alice Röbbelen, Malte L Schmieding, Marvin Kopka, Felix Balzer, Markus A Feufel. Originally published in JMIR Public Health and Surveillance (https://publichealth.jmir.org), 15.04.2022.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on https://publichealth.jmir.org, as well as this copyright and license information must be included.