The effect of remote treatment on medical provider creative thinking and patient disclosure: protocol for the MED-CREATE trial.

This trial is in the early stage of construction, as version 1.0, dated January 6th, 2025. We anticipate to start recruitment after funding is secured, likely in October 2025. It is registered on ClinicalTrials.gov (Number NCT06784635 ). It also will be pre-registered on the Open Science Framework, often used in behavioral science research.

The COVID-19 pandemic triggered a telehealth revolution, significantly expanding available services and increasing their acceptance among patients and providers, with many advancements persisting post-pandemic [ 1 , 2 ]. A 2019 survey in the USA found that only 4.2% of respondents had engaged in video conferencing visits with their physicians [ 3 ], whereas 86% of visits were conducted over telehealth platforms in 2023 [ 4 ]. Preliminary comparisons between telehealth and in-person care have been favorable across many clinical settings, with high satisfaction rates generally reported by patients and clinicians [ 5 ]. In addition, a recent meta-analysis of 77 studies comparing telehealth to in- person care found that, for most indicators, there was no significant difference between outcomes [ 6 ]. However, little consideration has been given to possible trade offs, including the inhibition of creative thinking that has been linked to video conferencing [ 7 ]. High-quality experimental research in non-medical fields has shown that online interactions result in fewer ideas than in-person interactions. This suggests that elements of creativity—such as divergent thinking, elaboration, problem sensitivity, and innovation—are inhibited online compared to offline environments [ 7 ]. Additionally, some research suggests people are more willing to disclose personal information online than in person, including around sensitive matters that have direct health implications such as sexuality [ 8 ]. Given the importance of creativity and openness in clinical care and health outcomes, this potential tradeoff is important to explore when comparing online and in-person care. This study seeks to understand how disclosure and creativity interact in the context of video-enabled medical care. Patient disclosure is crucial in the patient-clinician relationship, enabling the collection of an accurate patient history, which is essential for accurate diagnosis and targeted treatment. It underpins the patient-centered approach now favored in diagnosis and treatment [ 9 ]. Patient-centered approaches emphasize the patient’s experiences and perspectives in collaboration with the provider. Conversely, physician-centered methods rely more on “objective sources” like documented medical histories, blood tests, imaging studies, and observed symptoms, with less emphasis on the patient’s personal experience [ 9 ]. Open interactions within a patient-centered framework strongly predict positive outcomes, according to several studies [ 10 – 12 ], and remote healthcare is no exception [ 13 ]. Studies investigating disclosure in online settings have yielded mixed results, with only limited research exploring the impact on health consultations. In non-medical settings, it remains unclear whether people are more likely to disclose information about themselves. A meta-analysis from 2012 involving 15 non-medical studies was inconclusive, yielding mixed results [ 14 ]. In health-related fields, two narrative reviews suggested that, broadly, people tend to provide more personal information online [ 15 , 16 ], although the results of individual studies have been mixed. In a Chinese study examining over 46,000 doctor-patient interactions, patients were more likely to disclose information online if they had previously visited the doctor offline [ 17 ]. A smaller online study found that the majority of university students were willing to self-disclose private health information to researchers, regardless of whether participants were made aware of privacy risks or whether researchers made themselves seem distrustful [ 18 ]. Conversely, a large Dutch study ( N = 2,251) found that patients were less likely to disclose information about intimate body parts in video consultations compared to non-intimate body parts [ 19 ], although no comparison was made to offline consultation. In addition, limited research has suggested that a wide variety of additional factors, including trust in health systems, individual experiences, social determinants of health, and culture, may influence disclosure tendencies by patients [ 20 ], although only the evidence on these factors is often contradictory. Such discrepancies suggest that the context of the interaction, such as the relationship between the individuals and the purpose of the communication, can influence disclosure tendencies. While much research has examined how telehealth affects patient behavior (see [ 6 , 19 ]), there is far less research on how the behavior of medical professionals is affected by the online environment. However, there is some research on how doctors adapted in-person protocols for online environments [ 21 ]. In a patient-centered approach, medical professionals are required to ask the best questions in a time-limited, resource-tight environment that presents several barriers to information flow [ 22 ], especially in emergency situations [ 23 ]. This requires a degree of creativity [ 24 ], which encompasses skills known to be crucial in medical practice, including critical thinking, cognitive flexibility, divergent thinking, problem sensitivity, innovation, and elaboration [ 25 – 28 ]. Moreover, managing patients necessitates a level of general creativity, involving aspects of emotional intelligence like recognizing and appropriately responding to emotions [ 29 ]. Such crucial creativity may not translate well to online consultations. A well designed, large randomized controlled trial of 1,490 engineers found that they tend to be far less creative online than in person [ 7 ]. This study compared idea generation in in-person meetings versus online meetings across five countries. Online participants generated fewer ideas overall (“quantity”) and fewer creative ideas (“quality”) compared to in-person participants, indicating that online interactions may hinder creativity. The authors propose that video conferencing, by narrowing a user’s visual field, can restrict cognitive focus [ 7 ]. Two other studies [ 30 , 31 ] suggest that the intensity of gaze during video calls (i.e., locked eye contact) can be psychologically taxing and increase cognitive load— something that has been shown to reduce creativity [ 32 ]. With the rise of telehealth, it is possible that doctors’ creative thinking and idea generation are suffering— with potential consequences for care quality. This randomized controlled trial will evaluate the quality of remote healthcare communication by recruiting final-year medical students and patients for online video telehealth or in-person sessions. Our primary objectives are to examine if creativity is reduced in medical consultations in online settings (compared to offline) and to examine if patient disclosure is affected by online versus offline interactions. Our secondary objective is to examine the intrapersonal factors that affect creativity and disclosure in both the acting patient and medical provider. In this study, the patients will be provided with a script based on a difficult-to-diagnose condition. Simultaneously, the medical student will be instructed not to hold off on asking questions that might be perceived as intrusive by patients, so long as they are pertinent to the medical condition, with some hints. It is hypothesized that the online medical students will come up with fewer diagnoses (H1a) and will be less likely to provide the correct diagnosis (H1b) compared to those working in person with a patient. We also hypothesize that participants (acting out the role of patients) will be more likely to provide more details about potentially embarrassing things in an online environment than offline (H2). We also hypothesize that medical students will be more likely to have more diagnoses (H3a) and have more correct diagnoses (H3b) when they have higher empathy, working memory, extraversion, and openness. Finally, we hypothesize that patients will be more willing generally to disclose embarrassing details when they have higher health literacy (H4). The results of this study may guide healthcare resource allocation and inform the appropriate use of telehealth for specific conditions, such as those requiring high diagnostic creativity or sensitive patient disclosures.

This is the first study to evaluate both creativity and disclosure in online versus offline medical environments. The MED-CREATE randomized controlled trial seeks to address a significant gap in the literature: how to understand the advantages and disadvantages of telehealth by considering the social-psychological implications of the medium as compared to in person treatment. The study’s impact may be significant as it can guide healthcare organizations on when and where to deploy vital, and more costly, in-person resources. In addition, by considering the covariates involved, we may be able to provide a nuanced understanding of how a variety of factors impact healthcare interactions. If the study shows that creativity is hampered in medical interactions, it may affect recommendations for healthcare systems. For example, if our study shows that doctors are less able to generate ideas in an online environment, patients with symptoms that are difficult to diagnose (e.g., general fatigue [ 40 , 44 – 46 ]) or with risky comorbidities (e.g., cardiac issues and age; [ 46 ]) may be redirected to in-person visits, despite the higher cost. Furthermore, the study may demonstrate the unique utility of telehealth for patient disclosure in certain conditions. For example, it may be possible to suggest more telehealth for sexual health checkups in health settings as it may be possible that people are more comfortable at disclosing this information remotely. Considering that the best available data show that many people— especially racial and sexual minorities [ 47 , 48 ]— delay or avoid sexual healthcare visits, our research may provide a foundation for promoting sexual telehealth. Overall, the proposed RCT may inform healthcare methods by better understanding how telehealth fits in our systems and compares with in person settings. Even a null result for patient disclosure and medical creativity would emphasize the value of telehealth.

This study will use a parallel, two-arm group trial design. Participants will be assigned to one of two conditions (online or offline) via a computerized random number generator with 1:1 block randomization, with participants allocated via REDCap’s system. Odd-numbered participants will be placed online, while even-numbered participants will be in an offline environment. It will be an open-label randomized controlled trial, with participants unaware of their group assignment until the study commences, with a superiority framework. Given the overt differences between the online and in-person conditions, the blinding of participants or investigators is not feasible. We will be recruiting two types of participants for the study. For the group of final-year medical students, participants will be recruited from the Cedars-Sinai medical system and the UCLA medical system via emails using an internal list. For our “patient” group, we will recruit using social media and the Cedars-Sinai’s recruitment system (All of Us hub). Our “patient” group, herein referred to as acting participants, is a methodological choice commonly used in medicine as a popular assessment method in clinical teaching. Research on medical students and residents has repeatedly shown no significant differences in clinical reasoning when working with real patients versus simulated patients, for the same clinical diagnosis [ 33 – 35 ]. For this project, using acting participants with scripts means that we can standardize the scenario in a way that is familiar to the medical student, leading to a reliable measure that can be extrapolated beyond this study. For the final-year medical student sample, we will have the following inclusion criteria: English speaking Be enrolled in an MD or DO degree Be in their final year of study English speaking Be enrolled in an MD or DO degree Be in their final year of study Medical students will be excluded from recruitment or analysis if they: Have previous experience as healthcare providers that is separate from their current medical training (e.g., being a nurse) Work in a technology sector Declare a conflict of interest Unable or unwilling to sign informed consent documents Have previous experience as healthcare providers that is separate from their current medical training (e.g., being a nurse) Work in a technology sector Declare a conflict of interest Unable or unwilling to sign informed consent documents For the acting participant sample, we will have the following inclusion criteria: Aged 18 or older Must state they have current healthcare coverage English speaking at native fluency level Aged 18 or older Must state they have current healthcare coverage English speaking at native fluency level Patients will be excluded if There is significant current or past psychiatric illness that may inhibit their ability to consent or participate (self-described) They are not willing or able to sign informed consent documents There is significant current or past psychiatric illness that may inhibit their ability to consent or participate (self-described) They are not willing or able to sign informed consent documents After screening, all participants will need to be able to travel to the main study location (Cedars-Sinai’s Medical Offices, Los Angeles, CACalifornia, USAUnited States), which will be the only setting for this study. All participants will be sent a recruitment flyer and link using the appropriate platform (i.e., via email for the medical students, and via social media (Instagram and Facebook ads) for acting participants for the general public [ 36 ]). Upon clicking the link, participants will be asked if they are interested in a study on how lighting and sound environments affect healthcare interactions (the first of two deceptive elements in this study). If interested, participants will be provided, via email with an overview of what would be required of them, including travelling to Cedars-Sinai Medical campus and coming into an office with another person for a psychological test and 15-min health-related discussion, and the amount they will be compensated in return ($30 for the medical student, $20 for the participant for the 30-min session). If the participant is interested, they will be given a consent form for e-signature (REDCAP) and a calendar to list availability for the next month, and enrolled by AB (the first author of this study). If medical student and acting participant calendars match, they will be offered an appointment to come in for the study. The consent form will directly outline all parts of the study (including the possibility for sensitive questions), although the purpose will be concealed from the participant. At this stage, both medical students and acting participants will be randomized as online or offline via a simple random assortment (selecting a randomly generated string of odd or even numbers via REDCap), but they must still come to campus. In all cases, all participants will be asked to fill in the consent form again in a digital format, which will note clearly that this is not a diagnostic interview, but a hypothetical case being acted out, and no medical advice is being given. After consent is acquired, all participants will fill out questionnaires on their age, socioeconomic status, gender, education level, health literacy via the All Aspects of Health Literacy Scale [ 37 ], working memory via the NIH Toolbox List Sorting Working Memory Test [ 38 ], and the personality traits of openness and extraversion via the International Personality Item Pool [ 39 ]. Table 1 outlines these scales. While all participants complete these scales, the researcher will configure the lighting via a portable lamp and play very quiet white noise (the second element of deception). At this point, if the medical student is assigned to the “odd” group, the acting participant will be asked to stay in the room and video conference with the other member of the dyad using the same laptop via Google Meet, an end-to-end encrypted video call service. If the medical student is assigned to the “even” group, the medical student will be asked to move to the other room containing the acting participant. At this point, the researcher will provide both members a document, with the acting participant being asked to act a “diagnosis” and bring up their illness to the medical student, and the medical student being told to ask relevant questions to help diagnose the condition, even if they may be perceived as intrusive. Besides pretending to have that diagnosis, the acting participants in this study will be asked to respond to less relevant questions truthfully; for example, if the acting participant has a script of endometriosis, and the doctor asks about height and weight, it is expected the acting participant will respond with their real height and weight. Notably, the cases will be matched to the participant (e.g., participants will only get the endometriosis diagnosis if they state they are female). Both parties will also be informed that there will be no physical examination. Toward the conclusion, the medical students will be asked to provide a list of potential diagnoses verbally to the participant. This conversation will be recorded via the computer in the room and will last up to 20 min. Table 1 Psychometric scales Scale name Number of items, item score range Scoring method NIH Toolbox List Sorting Working Memory Test [ 35 ] Number of items can range from 20 to 28 (depending on performance), with two blocks. Each trial is scored as correct (1) or (0) incorrect based on the participant’s ability to recall and reorder the stimuli accurately The total number of correct trials across both test blocks is summed to generate the List Sorting Total Correct score International Personality Item Pool—Openness subscale [ 36 ] 10 items describing character traits or behavioral tendencies. 5-point Likert scale ranging from “very inaccurate” to “very accurate” Summing the individual scores for all the items, after accounting for reverse scoring of some items International Personality Item Pool—Extraversion subscale [ 36 ] 10 items describing character traits or behavioral tendencies. 5-point Likert scale ranging from “very inaccurate” to “very accurate” Summing the individual scores for all the items, after accounting for reverse scoring of some items All Aspects of Health Literacy Scale [ 34 ] 47 items on functional health literacy, communicative health literacy, and critical health literacy. 5-point Likert scale ranging from “strongly disagree” to “strongly agree” The scores for the items within each subscale are summed to obtain a raw subscale score (accounting for reverse scoring of some items), with the final score being a sum of all averages of subscales Multidimensional Fatigue Inventory [ 38 ] 20 items describing domains in general fatigue, physical fatigue, reduced motivation, reduced activity, mental fatigue. 5-point Likert scale (higher = more fatigue), some items are reverse scored, from “strongly disagree” to “strongly agree” The scores for the items within each subscale are summed to obtain a raw subscale score (accounting for reverse scoring of some items), with the final score being a sum of all averages of subscale Basic Empathy Scale for Adults [ 37 ] 19 items, describing domains of emotional contagion, cognitive empathy, and emotional disconnection. 5-point Likert scale (higher = more empathy), from “strongly disagree” to “strongly agree.” Some items are reverse scored The scores for the items within each subscale are summed to obtain a raw subscale score (accounting for reverse scoring of some items), with the final score being a sum of all averages of subscales Psychometric scales Upon completion, both the medical student and acting participant will be given a final questionnaire measuring empathy (the same scale provided earlier in the study) [ 40 ], an adapted fatigue scale used in medical settings [ 41 ], and a disclosure document outlining the actual purpose of the study with a list of free or low-cost resources for psychological or medical distress. Immediately after this session, the audio will be transcribed via a private, local program installed on the same computer, via Whisper CCP (i.e., an artificial intelligence program that does not use the internet to process speech to text [ 42 ]). This method of transcription was approved by the Institutional Review Board of Cedars-Sinai Medical Center. Participants will be offered the ability to ask questions, withdraw their data, a transcript of their conversation, and will be compensated via a gift card of their choosing. Immediately after the participant leaves, the researcher will listen to the audio and check to see if the program made any mistakes in understanding and transcribing the audio (e.g., mistaking one word for another), correcting them as needed. Any identifying details provided by either participant will be deleted from the transcript, besides demographic information already provided in the REDCap survey. After these checks, the audio recording will be deleted. Data for the questionnaires will remain on REDCap, and the transcript will be uploaded to the participants’ REDCap. Qualitative data will be exported to Word documents with the randomly allocated string of digits that represents the pair of participants (as the participant identifier). A CSV file will be used after downloading the data from REDCap for analysis using RStudio, with all participants’ names and emails no longer available to researchers for confidentiality, in order to maintain anonymity. Through this method, verbal and written informed consent will be collected multiple times from participants (before agreeing to be in the study, again right before the study’s commencement, and the debrief process). The schedule of enrollment is available in Fig. 1 . Fig. 1 Schedule of enrollment, interventions, and assessment Schedule of enrollment, interventions, and assessment Post-trial care will not be provided beyond direction to low-cost or free resources outlined in the disclosure document, as there are no expected harms besides those normally encountered during daily life. We will seek to recruit 194 participant dyads ( N = 388). This sample size was calculated using G*Power, based on the effect sizes seen in a similar study for total ideas generated per dyad, to have 97 medical students per group (online versus offline), with a matching number of acting patient participants. In the original study, (see [ 7 ], laboratory experiment stimulus 1), they had a Cohen’s d of 0.405 (averaging 13.89 ideas for the online group ( n = 150; SD = 5.65) and 16.15 ideas for the offline group ( n = 150; SD = 5.51)). Therefore, our study mimics the effect size requirements (Cohen’s d = 0.405, or f = 0.2025), with 2 groups, 80% power, for a fixed effect one- way ANOVA. We may over-recruit by 20% if dropout or withdrawal rates are high in the first 2two weeks of the study running, (adjusting to a total N of 466); we chose 20% as this is the median rate of dropouts for clinical appointments worldwide [ 43 ]. More details on the statistical analysis are available in Tables 2 and 3 . Table 2 Statistical analysis plan Section Item Description Section 1: Administrative information Title and trial registration Statistical analysis plan for the MED-CREATE study—trial registration number to be provided upon registration Statistical analysis version Version 1.0 (August 23, 2024) Section 2: Introduction Background and rationale The MED-CREATE study aims to investigate the impact of online vs. offline interactions on medical students’ diagnostic accuracy and creativity, as well as patient disclosure tendencies. This is important to understand as healthcare increasingly shifts toward telemedicine and virtual consultations Objectives Primary objective: to compare the number of potential diagnoses generated, diagnostic accuracy, and patient disclosure between online and offline consultations Hypothesis 1a: medical students in the online condition will generate fewer potential diagnoses than those in the offline condition Hypothesis 1b: diagnostic accuracy will be poorer in the online condition than the offline condition Section 3: Study methods Hypothesis 2: patients will disclose more information about sensitive health topics in the online condition compared to the offline condition Trial design Two-arm, parallel group, single-blind randomized controlled trial with 1:1 allocation to online or offline consultation conditions Randomization Computerized random number generator; odd numbers assigned to online condition, even numbers to offline condition. No minimization or stratification Sample size Target sample size is 388 participants (194 pairs). See the study protocol for the full sample size calculation Framework Superiority framework for the number of potential diagnoses and diagnostic accuracy generated in the offline condition Statistical interim analyses and stopping guidance No interim analyses planned. No planned adjustment of the significance level due to interim analysis. No expectations to end trial early Timing of final analysis All outcomes will be analyzed collectively after the completion of data collection Timing of outcome assessments Outcomes will be assessed immediately after the consultation Section 4: Statistical principles Confidence intervals and P values Level of statistical significance: two-tailed alpha = 0.05: no adjustment for multiplicity as there are a limited number of primary outcomes. 95% confidence intervals will be reported for all relevant estimates. Full power analysis included in protocol based on most similar and recent study Adherence and protocol deviations Adherence and deviation: adherence is defined as completing the assigned consultation (online or offline) in its entirety. Adherence will be summarized by group (online vs. offline). Protocol deviations include any significant departures from the study procedures outlined in the protocol. All major protocol deviations will be summarized and their potential impact on the study results will be discussed Analysis populations All randomized participants will be included in the ITT analysis, regardless of adherence or protocol deviation. Participants who completed the consultation according to the protocol and did not have any major protocol deviations will be included in the PP analysis. No safety analysis is needed Section 5: Trial population Screening data Screening data: screening data will be summarized to describe the number of individuals screened, eligible, and randomized Eligibility Eligibility: see the study protocol for the full eligibility criteria Recruitment The CONSORT flow diagram will include the number of participants screened, assessed for eligibility, randomized, allocated to each intervention, received the allocated intervention, completed follow-up, and analyzed for each outcome Withdrawal/follow-up Withdrawal: withdrawal from the study at any of the consent timepoints or after the study’s completion. Time of withdrawal will be recorded. Reasons for withdrawal will be summarized. Withdrawal data will be presented in the CONSORT flow diagram Baseline patient characteristics Baseline characteristics to be summarized: age, socioeconomic status, gender, education level, health literacy (participants); gender, working memory, openness, extraversion (medical students); length of encounter, time of day (encounter). Baseline characteristics will be summarized using descriptive statistics (mean and standard deviation for continuous variables, frequencies and percentages for categorical variables) by group (online vs. offline) Section 6: Analysis Outcome definitions Outcomes: primary outcomes: number of potential diagnoses generated by medical student, accuracy of diagnosis (correct/incorrect), number of words used by the participant to describe their health condition in response to questions about sexual health and/or bowel movements. Number of potential diagnoses will be count based. Diagnostic accuracy will be binary (correct/incorrect). Number of words used by participants will be count based, only counting nouns, verbs, and adjectives relating to their health. No transformations planned except after checking the data, where appropriate transformations will be used based on the shape of the data Analysis methods Analysis method: for count based analysis, hierarchical Linear regression to compare primary outcomes between online and offline groups, adjusting for pre-specified covariates. For binary outcomes, logistic regression, using similar methods. Treatment effects will be presented as adjusted mean differences or odds ratios with 95% confidence intervals. Covariates will be included in the regression models as fixed effects. Assumptions to be checked for statistical methods include linearity via scatterplots and residual plots, normality, via histograms and Q-Q plots of residuals, homoscedasticity, via residual plots, and independence via Durbin-Watson test Table 3 SPIRIT checklist. SPIRIT 2013 checklist: recommended items to address in a clinical trial protocol and related documents* Section/item Description Yes/no, page number Notes Administrative information Title 1 Descriptive title identifying the study design, population, interventions, and, if applicable, trial acronym Yes, page 1 Trial registration 2a Trial identifier and registry name. If not yet registered, name of intended registry No, page 16 (Trial status) Registry will be OSF 2b All items from the World Health Organization Trial Registration Data Set Not applicable Not applicable for behavioral science Protocol version 3 Date and version identifier Yes, page 16 (Trial status) Funding 4 Sources and types of financial, material, and other support Yes, page 17 Support provided by Cedars-Sinai Medical Center Roles and responsibilities 5a Names, affiliations, and roles of protocol contributors Yes, title page and page 17 (Author contributions) 5b Name and contact information for the trial sponsor Yes, title page 5c Role of study sponsor and funders, if any, in study design; collection, management, analysis, and interpretation of data; writing of the report; and the decision to submit the report for publication, including whether they will have ultimate authority over any of these activities Yes, page 17 5 d Composition, roles, and responsibilities of the coordinating center, steering committee, endpoint adjudication committee, data management team, and other individuals or groups overseeing the trial, if applicable (see item 21a for data monitoring committee) Not applicable Data monitoring only applicable through normal IRB requirements Introduction Background and rationale 6a Description of research question and justification for undertaking the trial, including summary of relevant studies (published and unpublished) examining benefits and harms for each intervention Yes, pages 5–8 (Introduction) Harm not expected beyond as low risk 6b Explanation for choice of comparators Yes, pages 6–8 (Introduction—effect of telehealth section) Objectives 7 Specific objectives or hypotheses Yes, page 8 (Research gap, present study objectives, hypotheses) Trial design 8 Description of trial design including type of trial (e.g., parallel group, crossover, factorial, single group), allocation ratio, and framework (e.g., superiority, equivalence, noninferiority, exploratory) Yes, page 8 (Trial design) Methods: participants, interventions, and outcomes Study setting 9 Description of study settings (e.g., community clinic, academic hospital) and list of countries where data will be collected. Reference to where list of study sites can be obtained Yes, page 10 (Procedure) Eligibility criteria 10 Inclusion and exclusion criteria for participants. If applicable, eligibility criteria for study centers and individuals who will perform the interventions (e.g., surgeons, psychotherapists) Yes, page 9 (Participant characteristics for inclusion exclusion) Who will perform the intervention is not needed, as outside scope of study Interventions 11a Interventions for each group with sufficient detail to allow replication, including how and when they will be administered Yes, pages 10–11 (Procedure) 11b Criteria for discontinuing or modifying allocated interventions for a given trial participant (e.g., drug dose change in response to harms, participant request, or improving/worsening disease) Not applicable No drugs or risks are expected, but participants have the ability to withdraw 11c Strategies to improve adherence to intervention protocols, and any procedures for monitoring adherence (e.g., drug tablet return, laboratory tests) Not applicable No adherence required, as single time point study 11 d Relevant concomitant care and interventions that are permitted or prohibited during the trial Not applicable No care provided Outcomes 12 Primary, secondary, and other outcomes, including the specific measurement variable (e.g., systolic blood pressure), analysis metric (e.g., change from baseline, final value, time to event), method of aggregation (e.g., median, proportion), and time point for each outcome. Explanation of the clinical relevance of chosen efficacy and harm outcomes is strongly recommended Yes, Table 2 (Statistical analysis plan) Participant timeline 13 Time schedule of enrollment, interventions (including any run-ins and washouts), assessments, and visits for participants. A schematic diagram is highly recommended (see Figure) Not appliable Only one time point needed Sample size 14 Estimated number of participants needed to achieve study objectives and how it was determined, including clinical and statistical assumptions supporting any sample size calculations Yes, Table 2 (Statistical analysis plan) Recruitment 15 Strategies for achieving adequate participant enrollment to reach target sample size Yes, page 14 Will require further ethical review and amendment to protocol Methods: assignment of interventions (for controlled trials) Allocation:  Sequence generation 16a Method of generating the allocation sequence (e.g., computer-generated random numbers), and list of any factors for stratification. To reduce predictability of a random sequence, details of any planned restriction (e.g., blocking) should be provided in a separate document that is unavailable to those who enroll participants or assign interventions Yes, page 8  Allocation concealment mechanism 16b Mechanism of implementing the allocation sequence (e.g., central telephone; sequentially numbered, opaque, sealed envelopes), describing any steps to conceal the sequence until interventions are assigned Yes, page 8  Implementation 16c Who will generate the allocation sequence, who will enroll participants, and who will assign participants to interventions Yes, page 8 REDCap will be used for this, along with Excel Blinding (masking) 17a Who will be blinded after assignment to interventions (e.g., trial participants, care providers, outcome assessors, data analysts), and how Yes, page 10 (Procedure) 17b If blinded, circumstances under which unblinding is permissible, and procedure for revealing a participant’s allocated intervention during the trial Not applicable It will be single blind, but participants will be made aware at the end of the blinding Methods: data collection, management, and analysis Data collection methods 18a Plans for assessment and collection of outcome, baseline, and other trial data, including any related processes to promote data quality (e.g., duplicate measurements, training of assessors) and a description of study instruments (e.g., questionnaires, laboratory tests) along with their reliability and validity, if known. Reference to where data collection forms can be found, if not in the protocol Yes, Table 1 (Psychometric scales) 18b Plans to promote participant retention and complete follow-up, including list of any outcome data to be collected for participants who discontinue or deviate from intervention protocols Not applicable Participants will not be retained Data management 19 Plans for data entry, coding, security, and storage, including any related processes to promote data quality (e.g., double data entry; range checks for data values). Reference to where details of data management procedures can be found, if not in the protocol Yes, page 12 (Procedure, blinding, and materials) Statistical methods 20a Statistical methods for analyzing primary and secondary outcomes. Reference to where other details of the statistical analysis plan can be found, if not in the protocol Yes, Table 2 20b Methods for any additional analyses (e.g., subgroup and adjusted analyses) Yes, Table 2 20c Definition of analysis population relating to protocol non-adherence (e.g., as randomized analysis), and any statistical methods to handle missing data (e.g., multiple imputation) Not applicable Missing data will not be possible due to the content of the study Methods: monitoring Data monitoring 21a Composition of data monitoring committee (DMC); summary of its role and reporting structure; statement of whether it is independent from the sponsor and competing interests; and reference to where further details about its charter can be found, if not in the protocol. Alternatively, an explanation of why a DMC is not needed No IRB requires specific data management plan to be in place 21b Description of any interim analyses and stopping guidelines, including who will have access to these interim results and make the final decision to terminate the trial Yes, page 15 No specific need to terminate trial, but decision capacity shared between both authors Harms 22 Plans for collecting, assessing, reporting, and managing solicited and spontaneously reported adverse events and other unintended effects of trial interventions or trial conduct Yes, page 15 (Ethical aspects) Auditing 23 Frequency and procedures for auditing trial conduct, if any, and whether the process will be independent from investigators and the sponsor Yes, page 15 (Ethical aspects) Ethics and dissemination Research ethics approval 24 Plans for seeking research ethics committee/institutional review board (REC/IRB) approval Not applicable Already approved (page 15) Protocol amendments 25 Plans for communicating important protocol modifications (e.g., changes to eligibility criteria, outcomes, analyses) to relevant parties (e.g., investigators, REC/IRBs, trial participants, trial registries, journals, regulators) Yes, page 14 (Potential changes) Consent or assent 26a Who will obtain informed consent or assent from potential trial participants or authorized surrogates, and how (see item 32) Yes, page 10 (Procedure, blinding, and materials) 26b Additional consent provisions for collection and use of participant data and biological specimens in ancillary studies, if applicable Not applicable No ancillary studies planned Confidentiality 27 How personal information about potential and enrolled participants will be collected, shared, and maintained in order to protect confidentiality before, during, and after the trial Yes, page 12 (Procedure, blinding, and materials) Declaration of interests 28 Financial and other competing interests for principal investigators for the overall trial and each study site Yes, page 18 (Competing interests) Access to data 29 Statement of who will have access to the final trial dataset, and disclosure of contractual agreements that limit such access for investigators Yes, page 18 (Availability of data and materials) Ancillary and post-trial care 30 Provisions, if any, for ancillary and post-trial care, and for compensation to those who suffer harm from trial participation Yes, page 13 (Post-trial care) Dissemination policy 31a Plans for investigators and sponsor to communicate trial results to participants, healthcare professionals, the public, and other relevant groups (e.g., via publication, reporting in results databases, or other data sharing arrangements), including any publication restrictions Yes, page 15 (Dissemination) 31b Authorship eligibility guidelines and any intended use of professional writers Not applicable No authors anticipated to be added 31c Plans, if any, for granting public access to the full protocol, participant-level dataset, and statistical code Yes, page 18 (Availability of data and materials) Appendices Informed consent materials 32 Model consent form and other related documentation given to participants and authorized surrogates Yes, Appendix A Biological specimens 33 Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in the current trial and for future use in ancillary studies, if applicable Not applicable None collected Statistical analysis plan SPIRIT checklist. SPIRIT 2013 checklist: recommended items to address in a clinical trial protocol and related documents* The primary outcomes will be the number of possible diagnoses generated by the medical student, whether or not they get the correct diagnosis, and the number of words used by the participant to describe their health condition in response to the questions about their sexual health and/or bowel movements, with the averages being used to represent each group. The first two outcomes (number of diagnoses and number of correct diagnoses) are meant to partially replicate the methods of Brucks and Levav’s study [ 7 ] exploring creativity in online settings, while the third outcome is a novel measurement. We will compare these between the online and offline group using hierarchical linear regression, adjusting for pre-specified covariates for the participant, medical student, and encounter characteristics. The participant covariates will include age, socioeconomic status, gender, education level, and health literacy. The medical student covariates will include age, gender, cognitive measures of working memory, and openness and extraversion. The encounter covariates will include the length of the encounter and time of day (morning or afternoon). The regression model will estimate the difference in our three outcome variables between the two groups, along with a 95% confidence interval. We will assess the model’s assumptions (e.g., linearity, normality of residuals) through diagnostic plots and statistical tests. Table 2 explains how we will test our regression model via our statistical analysis plan; our secondary objective will also be met through examining the effect of these covariates on the dependent variables in the same model. All analyses will be performed using R and will adhere to a two-tailed alpha level of 0.05 for statistical significance. Due to the methods of the study, there will be no missing data (i.e., REDCap will not allow participants to not have all data inputted and submit their documentation). All analysis code will be publicly available, as will the data (excluding the transcripts). Two changes will be made if we are unable to reach the required sample size: We will modify the online component to recruit participants through Prolific, a mass recruiting platform, and shorten the study by excluding the empathy and fatigue scales. With a reduced participation burden, we will compensate participants at half the original amount. This will require an amendment for ethics. We would increase the amount paid for in person participation by 1.5x the original amount to increase participation in the offline setting. We will modify the online component to recruit participants through Prolific, a mass recruiting platform, and shorten the study by excluding the empathy and fatigue scales. With a reduced participation burden, we will compensate participants at half the original amount. This will require an amendment for ethics. We would increase the amount paid for in person participation by 1.5x the original amount to increase participation in the offline setting. If achieving the required sample size remains challenging, we will seek inclusion in the Psychological Science Accelerator to obtain a larger, more diverse sample. These changes will be communicated on the Open Science Framework page, with re-evaluation by the Institutional Review Board of Cedars-Sinai Medical Center. Termination of the trial will be decided through joint decision by the two authors AB and EA. This study has been accepted by the Institutional Review Board of Cedars-Sinai Medical Center (STUDY00003845). All participants will be required to provide written, informed consent twice (at time of enrollment and at time of study). All participants will be given the ability to withdraw their data (recording, responses, etc.) at the end of the study, with data deletion following immediately. All audio recordings will be deleted after transcription. Participants will be identified by random initials in publications, with only age and gender disclosed, omitting all other identifying details from publicly accessible data. No participant’s full transcript will be published. In accordance with institutional and other mandates (i.e., Health Insurance Portability and Accountability Act (HIPPA), NIH), transcripts and relevant forms will be held for 6 years in an encrypted drive stored securely at Cedar-Sinai Medical Center, before being destroyed. Interim access to the data will only be given to AB and EA until completion, with IRB access available under IRB guidelines. All adverse events will be reported to the Institutional Review Board of Cedars-Sinai Medical Center, which will also conduct yearly and concluding audits. Results from this trial will be disclosed in international peer- reviewed journals and oral presentations at local, national, and international scientific meetings. Neutral, negative, and positive findings will be included in all communication. Participants will be offered the ability to be reminded in 2 years for the study results.

Supplementary Material 1 Supplementary Material 1